Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Group Practice of the Transcendental Meditation® and TM-Sidhi® Program and Reductions in Infant Mortality and Drug-Related Death: A Quasi-Experimental Analysis

Group Practice of the Transcendental Meditation® and TM-Sidhi® Program and Reductions in Infant... These two studies tested the prediction that the group practice of a procedure for the development of consciousness, the Transcendental Meditation and TM-Sidhi program, by a sufficiently large group of individuals would be sufficient to reduce collective stress in the larger population, reflected in two stress-related health indicators, infant mortality rate and drug- related fatality rate. Based on theoretical prediction and prior research, from January 2007 through 2010 (intervention period), this effect should have been measurable. Change in the rates of these two indicators during the intervention period were estimated from 2002 through 2010 data using a broken-trend (or segmented trend) intervention model with time series regression methods. Significant changes in trend for both the infant mortality rate and drug-related fatality rate were evident at the predicted time and in the predicted direction, controlling for preintervention trends, seasonality, and autocorrelation. The changes in trend were both statistically and practically significant, indicating an average annual decline of 3.12% in infant mortality rate and 7.61% in drug-related fatality rate. Diagnostic tests are satisfactory and indicate that it is unlikely that the statistical results are attributable to spurious regression. The mechanism for these collective effects is discussed in view of possible alternative hypotheses. Keywords behavioral sciences, alcohol, drugs, tobacco, sociology of health and illness, sociology, social sciences, collective behavior/ social movements, social change and modernization, medical sociology This article reports results of two studies on the quality of life We will first describe these procedures, then the theoreti- in the United States that extend a previous research program cal implications for the nature of consciousness implied by on consciousness and social well-being into the area of col- previous research on these procedures, and then discuss the lective stress and stress-related indicators of public health, indicators chosen for the present studies. namely, infant mortality rate (IMR) and drug-related fatality The Transcendental Meditation technique is described by rate (DFR). Previous research analyzed the impact of the its founder as a simple, experiential mental procedure of same intervention on homicide and violent crime (Dillbeck “turning the attention inwards towards the subtler levels of a & Cavanaugh, 2016) and motor vehicle fatalities and fatali- thought until the mind . . . arrives at the source of the thought” ties due to other accidents (Cavanaugh & Dillbeck, 2017). (Maharishi Mahesh Yogi, 1969, p. 470). “Source of thought” In a broader context, this article expands a substantive indicates a state of “pure consciousness” gained when the body of research on the collective effects of changes in con- mind settles down to a mode of inner silence in which the sciousness and quality of life. In these studies, change in the division of knower, knowing, and known is transcended and quality of societal consciousness is measured by group prac- awareness is open to itself alone (Roth, 2002). tice of the TM-Sidhi program, an advance practice of the Transcendental Meditation technique. Previous studies have Maharishi University of Management, Fairfield, IA, USA found that when practiced by a sufficient number of people, Corresponding Author: these subjective procedures are associated with an extended Michael C. Dillbeck, Institute of Science, Technology, and Public Policy, influence on social behavior in society, a finding that has Maharishi University of Management, Fairfield, IA 52557, USA. implications for a broader understanding of consciousness. Email: mdillbeck@mum.edu Creative Commons CC BY: This article is distributed under the terms of the Creative Commons Attribution 3.0 License (http://www.creativecommons.org/licenses/by/3.0/) which permits any use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.sagepub.com/en-us/nam/open-access-at-sage). 2 SAGE Open Meta-analysis of research on individuals practicing A number of published studies have used time series (TS) Transcendental Meditation has found that increased mental intervention analysis or transfer function (TF) analysis (Box silence during the practice, in comparison with sitting with & Jenkins, 1976) to evaluate effects at the city, state, or eyes closed, is associated physiologically with a unique state national/regional levels on indicators that reflect reduction of of restful alertness (higher basal galvanic skin resistance, social stress, such as decreased crime and violence, or lower breath rate, lower plasma lactate); this state of restful improvement in comprehensive quality of life indices. alertness, or reduced physiological stress, is also evident in For example, Dillbeck, Cavanaugh, Glenn, Orme- participants outside the practice period (lower respiration Johnson, and Mittlefehldt (1987) found that when the size of rate, lower heart rate, lower plasma lactate, and fewer sponta- a group of TM-Sidhi program experts who had assembled on neous skin resistance responses; Dillbeck & Orme-Johnson, courses exceeded the √1% of the populations of each course 1987). Alertness is indicated by increased electroencephalo- location (Delhi, India; Metro Manila, Philippines; Puerto gram (EEG) coherence during Transcendental Meditation Rico), TS intervention analysis showed reduced crime inde- practice; increased integration of brain functioning is also pendent of alternative explanations. Using TS intervention found longitudinally (Dillbeck & Bronson, 1981; Travis methods, similar results were found on monthly crime rates et al., 2010; Travis et al., 2009). Additional research, includ- in Merseyside, United Kingdom (Hatchard, Deans, ing reduced cortisol during the practice and more effective Cavanaugh, & Orme-Johnson, 1996). Two studies in cortisol response to stress, indicates that the physiological Washington, D.C., using TF methodology found predicted changes during the practice are counter to those found during variations between the size of a TM-Sidhi group in the city stress and support recovery from stress (Jevning, Wilson, & and decreased violent crime rate when the size of the group Davidson, 1978; MacLean et al., 1997; Walton et al., 2004). was over the predicted threshold (Dillbeck, Banus, Polanzi, There is a large body of research on the stress-reducing & Landrith, 1988; Hagelin et al., 1999). The second study effects of this procedure for health, particularly cardiovascu- (Hagelin et al., 1999) was a prospective study with predic- lar health (e.g., reviewed by Schneider & Carr, 2014), and on tions registered in advance. Both studies in Washington, psychological variables, which can be considered in the con- D.C., found no evidence for alternative hypotheses. text of theories of human development (e.g., Alexander et al., At both state and national levels, improvement was found 1990; Dillbeck & Alexander, 1989). using TS intervention analysis on a comprehensive index of In the 1960s, based on the understanding of conscious- quality of life integrating multiple monthly behavioral and ness contained in the Vedic tradition of India, Maharishi pre- health-related variables (e.g., homicide or crime, motor vehi- dicted that even a small proportion of the society experiencing cle fatalities, cigarette consumption) during months when the this state of pure consciousness would be sufficient to enliven size of TM-Sidhi group was larger than the required √1% in greater order and positivity in the collective consciousness of Rhode Island (Dillbeck et al., 1987) or in the United States or society, reflected in decreased stress, resulting in decreased the United States plus Canada (Dillbeck & Rainforth, 1996). negative behavioral trends such as crime and violence Similarly, reductions in an index of national violent fatalities (Maharishi Mahesh Yogi, 1977). (homicide, suicide, and motor vehicle fatalities) were found In 1976, Maharishi introduced the advanced TM-Sidhi using TS intervention analysis over a multiple-year period program, the purpose of which is described as accelerating during weeks when the size of a stable, large TM-Sidhi group the integration of the inner silence experienced during was sufficient to predict an effect for either the United States Transcendental Meditation with daily activity outside of or Canada (Assimakis & Dillbeck, 1995; Dillbeck, 1990). meditation (Maharishi Mahesh Yogi, 1986). In discussions A study in Israel, spanning multiple societal levels and with physical scientists in the light of the ancient Vedic addressing multiple outcome parameters for each, found that knowledge, he formalized the prediction that when a group over a 2-month period, daily quality of life indices for practices the TM-Sidhi program together, only the square Jerusalem, Israel as a whole, and the Israel–Lebanon conflict root of 1% (√1%) of the population—as compared with 1% (ongoing at that time) significantly improved during days practicing Transcendental Meditation individually—would when the size of a temporary group was sufficiently large to be required to create a calming and orderly influence in soci- yield theoretically predicted effects at the city, national, or ety (Maharishi Mahesh Yogi, 1986). The square root term is regional levels (Orme-Johnson, Alexander, Davies, Chandler, by analogy to coherent phenomena in physical systems in & Larimore, 1988). Comparable results were found using TS which the combined intensity of coherent elements can be intervention analysis and TF analysis. A number of alterna- proportional to their square (Hagelin, 1987). On that basis, it tive explanations and alternative analysis models were effec- should be possible to influence very large populations: for tively considered both in the original article and in subsequent example, the √1% of a city of 9 million is 300 individuals, responses to methodological questions (Orme-Johnson, and is near 1,800 for the current U.S. population. Alexander, & Davies, 1990; Orme-Johnson & Oates, 2009). The relatively small number of required TM-Sidhi pro- In a more detailed and comprehensive analysis of vio- gram group participants permits quasi-experiments using a lence due to the Lebanese conflict, Davies and Alexander time scale of analysis finer than that of annual data. (2005) looked at all seven intervention periods during a Dillbeck and Cavanaugh 3 2.25-year period (1983-1985) when there were TM-Sidhi American mothers much more severely than White mothers program groups of sufficient size (on temporary courses) to or mothers of other racial/ethnic groups, and preterm birth is predict an impact upon the conflict in Lebanon either from substantially higher among African American mothers within Lebanon, near to Lebanon, or even at some distance. (MacDorman & Mathews, 2011; Mathews, MacDorman, & Using daily event data from nine news sources that were Thoma, 2015). Preterm birth risk has been shown to be blind coded by an independent Lebanese coder using scales related to perceived stress, pregnancy-related anxiety, and derived independently for research on the conflict, TS inter- perception of racial discrimination (Dole et al., 2003). The vention analysis indicated a significant impact of each of the causes of IM are multiple; nevertheless, stress is one major seven temporary groups on reduced conflict. The analysis influence, as described above. The cascade of physiological/ controlled for temperature, holidays, and weekends, and the hormonal events in the mother’s body initiated by stress and findings were independent of alternative TS noise models. their interaction with the physiology of labor initiation are Multiple indicators of reduced conflict also replicated this described in detail by Gennaro and Hennessy (2003) and by effect when combining intervention periods. Latendresse (2009). In these studies, there is an implied underlying connection DD of all types also reflect to a great degree the immedi- between individuals that would permit such far-reaching ate effects of stress and are affected by the degree of con- effects of stress reduction beyond that which could be scious alertness or vigilance. The category of DD fatalities explained by behavioral interactions. That is, as described by includes all deaths for which drugs are the essential cause, the Vedic tradition of India (Radhakrishnan, 1953), pure con- irrespective of whether the drug was illicit or prescription, sciousness is proposed to have a field-like character, as including acute poisoning and also medical conditions aris- opposed to the isolated quality of individual consciousness. ing from chronic use. DD have increased substantially since On that basis, it is predicted that a calming or stress-reducing 1990, a period in which prescriptions for opioid analgesics influence will be experienced in the broader collective con- for pain management rose dramatically. Between 1999 and sciousness of society (Maharishi Mahesh Yogi, 1986). 2002, deaths due to opioid analgesic poisonings rose over In extending these studies into the health area, we assume 90%, becoming a larger source of drug poisoning deaths than that collective stress (or its reduction) can have an impact those due to heroin or cocaine (Paulozzi, Budnitz, & Xi, upon some stress-related health outcomes, not only upon 2006). An investigation in 2010 found that of all drug over- behavioral violence such as crime. Analyzing U.S. annual dose deaths, 57.7% involved pharmaceuticals, and of those, data using a state stress index that combined economic, fam- 75.2% involved opioids (either alone or in combination with ily, and community stressors (Linsky, Bachman, & Straus, other drugs; Jones, Mack, & Paulozzi, 2013). 1995; Linsky & Straus, 1986), the degree of social stress pre- Opioid analgesics have been increasingly prescribed as dicted violent crime rate and also maladaptive behaviors part of more aggressive pain management regimens. (Linsky & Straus, 1986). However, social stress so defined Chemically, the opioids influence the body in a similar way was less associated with mortality due to illness, illnesses as opium-derived opiates such as morphine and heroin whose morbidity at the individual level are exacerbated by (Volkow, 2014). For this reason, vigilance must be main- stress. One of the possible reasons for this is the unknown tained to avoid addiction to opioid analgesics. Chronic stress and variable time lag between serious morbidity and mortal- has been found to increase the vulnerability to addiction ity, not to mention between serious stress and morbidity (Sinha, 2008). (Linsky & Straus, 1986). By 2010, among those 35 to 54 years of age, poisoning The health indicators selected for this study are those for had become the most common type of accidental death, more which mortality is directly influenced by stress, rather than common than auto-related deaths (Harmon, 2010). Recent those indicative of disease states that develop over time and research has found that the death rate for White non-Hispan- whose time course may be inconsistent and hard to specify. ics aged 45 to 54 years actually rose from 1999 to 2013, par- Specifically, infant mortality (IM) and deaths due to drugs ticularly among the least educated, reversing a trend of (DD) were selected for investigation. The following is a previous decades (Case & Deaton, 2015). The authors rationale for each outcome variable. hypothesize that this increase in mortality could be driven by IM is both a fundamental indicator of national health as factors including deaths due to the increased availability of well as a stress-related phenomenon in the United States. opioid prescriptions and, in some cases where addiction had Compared with other developed countries, IM rates are high, occurred, transition to heroin when opioid restriction with the United States ranking 26th among Organisation for increased. They note that financial stress and insecurity may Economic Co-Operation and Development (OECD) coun- have contributed to these results, including wage stagnation tries (MacDorman, Mathews, Mohangoo, & Zeitlin, 2014). and diminished retirement prospects. Morbidity also Preterm birth is one of the major causes of IM; in 2005, increased in this group as indicated by self-reported level of 36.5% of U.S. infant deaths were due to preterm-related physiological and mental health as well as increased preva- causes, and 68.6% of all infant deaths occurred to preterm lence of chronic pain. Increased DD and alcohol poisoning infants (MacDorman & Mathews, 2008). IM affects African (as well as several other factors) also increased in other 4 SAGE Open 5-year age groups (30-34, 60-64) although not enough to (Dillbeck & Cavanaugh, 2016), and this period was subse- raise the overall mortality rate. quently extended to other variables for the sake of compari- It is clear that the rapid rise of DD in the past two decades son (Cavanaugh & Dillbeck, 2017). The binary intervention is a serious health concern in the American population as a variable (I ) was specified as 0 from July 2001 to December whole. At the same time, it seems likely that this variable is 2006, and 1 from January 2007 to December 2010. (Archival directly affected by perceived life stress and, in the case of data of the group size was not available continuously prior to opioid medication, is influenced by vigilance and self-aware- July 2001.) ness to avoid addiction. It is in this context that change in DD is a potential reactive indicator of changes in the quality of Dependent Variables collective consciousness. The two studies reported here evaluate the effects on U.S. The dependent variable data for both studies were obtained DD and IM associated with the rapid creation of a large from the National Center for Health Statistics of the Centers Transcendental Meditation and TM-Sidhi group sufficient in for Disease Control and Prevention (CDC). For Study 1, the size to create a hypothesized influence of reduced stress and dependent variable is the monthly U.S. IMR within 1 week of increased alertness in the collective consciousness of the birth per 10,000 live births. The IM monthly total was United States. The rapid establishment of this group, approxi- obtained from the data set “Underlying Causes of Death, mating in size a step function that could be appropriately mod- 1999-2013” through the CDC WONDER Online Database eled by a binary variable, offered a straightforward opportunity (CDC, National Center for Health Statistics, 2015a). The for quasi-experimental intervention research (Cavanaugh & number of live births each month was available also from Dillbeck, 2017; Dillbeck & Cavanaugh, 2016). CDC WONDER or the National Center for Health Statistics The hypothesis of the current research, consistent with VitalStats system as natality public-use data in different previous studies, is that there would be a significant impact source files depending upon the years, 1995 to 2002, 2003 to of the independent variable measured in terms of decreased 2006, and 2006 to 2013, posted in November 2005, March rates of IM and drug-related deaths. 2009, and January 2015, respectively (CDC, National Center for Health Statistics, 2015b). For Study 2, the dependent variable is the drug-related General Method fatality rate (DFR) per 1 million population. As noted above, this category of mortality includes any type of drug and any Intervention circumstance of death, as long as drugs are cited on the death To assess the predicted effect on these two fatality rates of the record as the essential cause of death. Data were obtained largest group of TM-Sidhi participants in North America, the from the National Center for Health Statistics through the analysis used a binary intervention variable based on the size CDC WONDER Online Database (CDC, National Center of this group. The location of the group was in Fairfield, Iowa, for Health Statistics, 2012). The computation of the rate per at Maharishi University of Management, where students, fac- 1 million population used U.S. Census counts for April 2000 ulty, staff, and community members assemble each day to and April 2010, with monthly linear interpolation and exten- practice the Transcendental Meditation and TM-Sidhi pro- sion to December 2010. gram together before and after school or work. The meditation halls on campus record the morning and evening daily totals. Data Analysis To expand the size of the TM-Sidhi program group from under 1,000 to a number sufficient to create a predicted posi- To analyze the results of the quasi-experiment, we use inter- tive influence for the whole United States (approximately vention analysis, or interrupted TS analysis, of data for 2002 1,725 needed for the 297 million population at that time, to 2010 (Chelimsky, Shadish, & Orwin, 1997; Cook & according to the √1% formula), a special course was held at Campbell, 1979; Shadish, Cook, & Campbell, 2002) to test the University beginning in July 2006 for visitors from the the hypothesis that a significant reduction in trend for IMR United States or around the world. To further increase the and DFR occurred beginning with the onset of the interven- number of participants, starting in November 2006, a large tion period in January 2007. TS regression analysis is used to group of several hundred visiting Indian experts in the estimate a broken-trend intervention model (Perron, 1989; TM-Sidhi program were hosted nearby in their own facili- Rappoport & Reichlin, 1989) to test for the hypothesized ties. After that, the total number of TM-Sidhi participants trend shifts in IMR and DFR. The intervention model in each began to exceed the predicted threshold in January 2007 and case includes a preintervention linear trend with an exoge- remained above or near that level throughout the interven- nous structural break in the trend function at the theoretically tion period of the study, 2007 through 2010. An intervention predicted date of December 2006. The intervention compo- period of that length was adopted because when the first data nent is modeled as a binary (0-1) step function that triggers a collection began to evaluate this intervention, 2010 was the shift in the linear trend function with the onset of the inter- most recent data available for homicide and violent crime vention period. Dillbeck and Cavanaugh 5 Figure 1. Plots of monthly IMR and size of the TM-Sidhi group (GROUP). Note. In Panel (a), the plot of IMR for August 2002 to December 2010 displays irregular variation plus weak monthly seasonal variation around a flat or slightly declining preintervention trend that shifts to a more steeply declining trend in the intervention period starting in January 2007 (see vertical line in the plot). During the intervention period, actual IMR declines more rapidly than predicted by its prior trend. In Panel (b), the plot of the monthly average daily size of the TM-Sidhi group rises rapidly starting in July 2006 until in January 2007, it exceeds the predicted critical threshold of 1,725, the √1% of the U.S. population at that time. The average size of GROUP is 591 participants for the 53 preintervention months and 1,792 for the 48 months of the intervention. Panel (b) adapted from “The Contribution of Proposed Field Effects of Consciousness to the Prevention of US Accidental Fatalities: Theory and Empirical Tests,” by K. L. Cavanaugh and M. C. Dillbeck, 2017, Journal of Consciousness Studies, 24(1-2), p. 66. Copyright 2017 by Imprint Academic, Ltd. IMR = infant mortality rate. preintervention TS behavior (see Note 8). The sample is Study 1: Results of Analysis of Monthly August 2002 through December 2010, with effective sample IMR size N = 101. This sample was chosen to give the largest pos- Plot of Monthly IMR sible effective sample (equivalent for intervention and sta- tionarity tests) after allowing for both first differencing and Panel (a) of Figure 1 displays the plot of IMR and the IMR forecast (dotted line) for 2007 to 2010 based on IMR’s for 12 lags of first-differenced dependent variables required 6 SAGE Open for diagnostic testing of the statistical assumption of declining trend during the intervention period 2007 to 2010. stationarity. The absolute value of the effect size measure is the square root In the preintervention period, the IMR trend is relatively of Cohen’s f for a regression variable (or set of variables), flat or very slightly negative with irregular variation and with 0.59, 0.39, and 0.14 considered large, medium, and small weak seasonal variation around the trend. Beginning in effects, respectively (Cohen, 1988). Panel (a) of Figure 2 January 2007 (see vertical line in the plot), IMR displays a graphically compares the preintervention trend slope for IMR shift to a more rapidly declining trend that continues through with that for 2007 to 2010 and Panel (b) displays the estimated the end of the sample period. During the intervention period, trend shift parameter (−0.02292) with 95% confidence interval actual IMR declines more rapidly than predicted by its prior = [−0.0128, −0.0330]. Although we have a clear a priori direc- trend and faster than the IMR forecast. tional hypothesis for the shift in trend, to be conservative, two- Panel (b) of Figure 1 shows the plot of the GROUP series. tailed tests are reported for the estimated trend shift (and all As noted above, the GROUP plot approximates a step func- other parameters). For reasons discussed below, the results tion, with an average of 591 participants for the 53-month reported in Table 1 are based on t ratios that remain valid in the baseline period compared with 1,792 for the 48-month inter- presence of heteroscedasticity and autocorrelation of the vention. In January 2007, the GROUP size for the first time regression residuals (Newey & West, 1987). The trend shift in the sample period rose above the theoretically predicted remains highly significant using conventional OLS standard critical threshold of 1,725, the √1% of the U.S. population. errors, t(87) = −3.32, p = .001, but the preintervention trend is not significantly different from zero, t(87) = −1.88, p = .064. Relative to the preintervention trend, the estimated trend Regression Results for IMR shift implies a total reduction of 1.100 in IMR during the To test the hypothesis of a decrease in the trend slope for 2007 through 2010 intervention period. This is a reduction IMR, we estimate the following broken-trend intervention of 12.47% (or 3.12% per year) compared with the mean pre- model that incorporates a shift of linear trend beginning with intervention monthly rate of 8.821 fatalities per 10,000 live the onset of the intervention (January 2007): births. The decline in IMR translates to a total of 992 averted infant deaths for 2007 to 2010, deaths projected to occur had IMR =+ ββ tD +− () ββ TS ++ D ε . tt 01 21 jt jt ∑ (1) the preintervention trend continued unchanged through 2007 j 8 to 2010. Thus, the estimated shift in trend for IMR during In Equation 1, β is the regression intercept, t is a linear time 2007 through 2010 has the predicted negative sign and is trend (t = 1, 2, 3, . . ., N), and β is the preintervention trend both statistically and practically significant. Statistical analy- slope for IMR. The variable DT models the shift in trend due sis (see Table 1) indicates that these results cannot be to the intervention with DT = (t − t )I , where t is the time explained by seasonal variation, autocorrelation, or preexist- t B t B of the hypothesized break in the linear trend function ing trends in IMR. (December 2006) and I is a binary (0-1) indicator variable (step function) that takes the value zero for the preinterven- Regression Diagnostics for Analysis of IMR tion period and 1.0 for the intervention period (t > t ). The regression coefficient (β − β ) for DT gives the change in Table 1 reports diagnostic tests for evaluating the adequacy 2 1 t trend slope for IMR from the preintervention value (β ) to the of the estimated intervention model. The null hypothesis of slope in the intervention period (β ). The hypothesis of a serially uncorrelated (white noise) regression residuals at negative shift in trend for IMR during the intervention lags 1 to 6, and 1 to 12 is rejected by the Breusch–Godfrey implies (β − β ) < 0. Lagrange Multiplier (LM) test (Godfrey, 1978). Only one 2 1 The summation term in Equation 1 is a deterministic sea- autocorrelation at lags 1 to 36 is statistically significant. The sonal component to control for the monthly seasonal varia- largest individual autocorrelations are at lag 1 (−0.208), tion in IMR. The seasonal component consists of 11 binary which is just significant at the 5% level, and lags 4 (−0.196) (0-1) seasonal dummy variables D (with monthly index j = and 16 (−0.224), which are nearly significant. Because of 1, 2, . . ., 11 where January is denoted by j = 1; Granger & this modest serial correlation, Table 1 reports t ratios based Newbold, 1986). The seasonal regression coefficient for on SEs that are valid (consistent) in the presence of serial each month is given by S . Finally, ε is an independent and correlation and heteroscedasticity of possibly unknown form jt t identically distributed, serially uncorrelated normal error (Newey & West, 1987). with mean zero and variance σ . Although some mild autocorrelation is present, the autocor- Table 1 reports the ordinary least squares (OLS) regression relations are all small, and thus the residual series appears to be results for Equation 1. The estimated preintervention trend for clearly stationary. A TS is said to be covariance stationary (or IMR is small but negative and significant. The shift in trend has weakly stationary) if its mean, variance, and autocorrelations negative sign, as hypothesized, and is highly significant, t(87) (or, equivalently, its autocovariances) are invariant with respect −5 = −4.50, p = 2.1 × 10 , effect size f = −0.482, a medium to to a change in time origin (Enders, 2010). Empirical support for large effect, indicating an acceleration of the preintervention stationarity of the IMR residual series is provided by a Dillbeck and Cavanaugh 7 Table 1. OLS Regression Analysis of Monthly U.S. IMR. a b Parameter Estimate SE t ratio β 9.360 0.235 39.87*** −3 −3 β −6.875 × 10 2.129 × 10 −3.23** −2 −3 β − β −2.292 × 10 5.091 × 10 −4.50*** 2 1 S −0.233 0.288 −0.81 1t S −0.185 0.219 −0.84 2t S −0.192 0.265 −0.73 3t S −0.245 0.303 −0.81 4t S 0.141 0.236 0.60 5t S −0.403 0.219 −1.84 6t S −0.297 0.261 −1.13 7t S −0.672 0.275 −2.44* 8t S −0.875 0.269 −3.25** 9t S −0.371 0.276 −1.34 10t S 0.178 0.307 0.58 11t F statistic: F(13, 87) = 11.65*** Mean of IMR = 8.386 SE of regression = 0.501 SE of IMR = 0.774 Sum of squared residuals = 21.864 Log-likelihood = −66.033 2 2 R = .635; Adjusted R = .581 BIC = 1.947; AIC = 1.585 Diagnostics Serial correlation test Heteroscedasticity test Lags 1-6: F(6, 81) = 3.370 (p = .005) F(15, 85) = 1.234 (p = .263) Lags 1-12: F(12, 75) = 1.948 (p = .042) Test for ARCH Normality test: χ (2) = 1.260 (p = .533) Lags 1-6: F(6, 89) = 6.803 (p = .570) Robinson test : t(61) = −8.177 (p < .001) Test of functional form Perron unit root test: τ = −6.612 (p < .01) F(2, 85) = 1.607 (p = .207) Note. Sample is August 2002 to December 2010, N = 101. OLS = ordinary least squares; IMR = infant mortality rate; BIC = Bayesian information criterion; AIC = Akaike information criterion; ARCH = autoregressive conditional heteroscedasticity. Newey–West SEs and t ratios (Newey & West, 1987). df = 87. Frequency domain test for nonstationarity (Robinson, 1995). *p < .05. **p < .01. ***p < .001. Figure 2. Comparison of mortality trends during the preintervention and intervention periods. Note. Panel (a) shows that the declining linear trend for IMR in the preintervention period (August 2002–December 2006) displays a shift to more rapidly declining trend during the intervention period (January 2007–December 2010). The IMR trend slope is the rate of change per month in monthly infant deaths within 1 week of birth per 10,000 live births. Panel (b) displays the statistically significant trend shift for IMR during 2007 through 2010 with 95% confidence interval. IMR = infant mortality rate. 8 SAGE Open frequency domain test based on the periodogram of regression Study 2: Results of Analysis of DFR residuals (Baum & Wiggins, 2000; Robinson, 1995). The Plot of DFR Robinson test evaluates stationarity based on an empirical esti- mate of the order of integration of a TS, I(d), where d is the The plot of DFR in Figure 3 displays monthly seasonal fluc- number of times the series must be differenced to induce sta- tuations and a rising preintervention trend followed by a tionarity. For example, in the case of a nonstationary random flattening of the trend slope in 2007 through 2010. During walk, the differencing parameter d = 1, and the series may be 2007 through 2010, the forecast of DFR (dotted line) based transformed to I(0), or stationarity, by first differencing. The on preintervention data continues DFR’s preintervention Robinson test rejects the null hypothesis (p < .001) that the IMR upward trend, rising substantially above observed DFR (see residual series is nonstationary (see Table 1), supporting the Note 18). conclusion that IMR is broken-trend stationary, exhibiting sta- tionary fluctuations around a broken trend. Regression Results for DFR Stationarity of the regression residuals is required for valid statistical inferences regarding the estimated regression We estimate the following broken-trend intervention model parameters in the intervention model (Banerjee, Dolado, to test the hypothesis of a shift to a reduced trend slope for Galbraith, & Hendry, 1993; Enders, 2010; Granger & DFR during the intervention period: Newbold, 1986). Broken-trend stationarity for IMR implies DFRD =+ ββ tD +− () ββ T ++ β FR tt 01 21 31 t− that the OLS regression estimates in Table 1 have standard distributions and thus the observed significant change in SD + ε . (2) jt jt trend is unlikely to be the result of “spurious regression” j (Banerjee et al., 1993; Granger & Newbold, 1986). Spurious where lag 1 of DFR is included (with regression coefficient regressions can result when time trends are fitted to nonsta- β ) to model its first-order autoregressive dynamics. All tionary variables that contain a random walk component other terms are defined as in Equation 1. (“stochastic trend”). A key signature of spurious regressions The OLS regression estimates of the model coefficients is highly autocorrelated, nonstationary regression residuals. for Equation 2 are reported in Table 2. The DFR preinterven- The latter violate the distributional assumptions underlying tion trend is positive and significant. The change in trend has statistical inference for TS regression. negative sign and is highly significant, consistent with theo- Table 1 reports a formal test for broken-trend stationarity retical prediction. of a TS with a known (exogenous) structural break (Perron, In Table 2, the regression estimate of β − β , the change 2 1 1989, 2006). The Perron unit root test rejects the null in trend, gives the initial intervention impact. The significant hypothesis (p < .01) that the IMR TS is a nonstationary ran- first-order autoregressive dynamics of DFR around the trend dom walk with drift (Perron, 1989; Zivot & Andrews, implies a gradual adjustment of the preintervention trend to 1992). The alternative hypothesis is that IMR is broken- its equilibrium, or long-run, value after onset of the interven- trend stationary, displaying stationary fluctuations around a tion. If (1 − β ) ≠ 0, the total change in trend is given by κ = linear trend with a known, one-time break in the trend func- (β − β ) / (1 − β ). A formal test rejects the hypothesis (1 − 2 1 3 tion in December 2006. β ) = 0 (p < .01) in favor of the alternative that (|β | < 1), 3 3 The results of all other diagnostic tests for model ade- indicating that the estimated model is dynamically stable and quacy are satisfactory. The null hypothesis of no autoregres- thus that the trend will converge to its equilibrium value κ in sive conditional heteroscedasticity (ARCH) for the regression the long run (Banerjee, Dolado, & Mestre, 1998; Doornik & residuals is not rejected by the LM test for ARCH (Engle, Hendry, 2013; Enders, 2010). For the DFR analysis κ = −10 1982). Likewise, White’s general test for heteroscedasticity −0.0578, t(86) = −7.12, p = 3.1 × 10 , effect size f = −0.449, (White, 1980) fails to reject the null hypothesis that the a medium to large effect, and 95% confidence interval = regression errors are homoscedastic or, if heteroscedasticity [−0.0417, −0.0739], where κ is the coefficient on DT in the is present, it is unrelated to the regressors. The null hypoth- long-run equation for DFR (Hendry, 1995). In this context, esis that the functional form of the regression model is cor- “long run” denotes the mathematically expected equilibrium rectly specified is not rejected by Ramsey’s (1969) Regression value and does not necessarily imply a long period of time Error Specification Test (RESET). The omnibus Doornik– (Hendry, 1995). The estimated cumulative lag weights for Hansen test for normality of the regression errors (Doornik Equation 2 indicate that 50.2%, 75.25%, 87.6%, and 93.8% & Hansen, 2008) fails to reject the null hypothesis that the of the adjustment in trend after the onset of the intervention errors are drawn from a normal distribution. No regression is completed within 1, 2, 3, and 4 months, respectively. Panel residuals exceed 3.5 standard errors, indicating the absence (a) of Figure 4 compares the equilibrium trend slopes for the of extreme outliers. Thus, the diagnostic tests for model ade- preintervention and 2007 through 2010 periods and Panel (b) quacy support statistical conclusion validity for the statistical shows the estimated trend shift with 95% confidence inferences reported in Table 1. interval. Dillbeck and Cavanaugh 9 Figure 3. Plot of monthly DFR. Note. The plot of the monthly rate of drug-related accidental fatalities for August 2002 to December 2010 displays monthly seasonal variation and a rising trend in the preintervention period followed by a flattening of the trend during the intervention period 2007 through 2010 (see vertical line in the plot). During 2007 through 2010, the forecast of DFR (dotted line) continues its preintervention upward trend, rising substantially above observed DFR. DFR = drug-related fatality rate. Relative to the preintervention trend, the equilibrium tests for adequacy of the broken-trend regression model sup- trend shift for 2007 to 2010 implies a total cumulative decline port statistical conclusion validity for the statistical infer- in DFR of 2.7723 monthly fatalities per million population ences reported in Table 2. during the intervention period. This is a reduction of 30.42% (or 7.61% annually) compared with the mean prein- Discussion tervention monthly rate of 9.112 fatalities per million people. The decline in DFR translates to a projected total of 26,425 The results of the two studies indicate that, as hypothesized, averted drug-related fatalities for 2007 to 2010. Thus, the there was a significant and substantial reduction of the trends estimated shift in trend for DFR is negative, as predicted, and of IMR and DFR during the intervention period 2007 through both practically and statistically significant. The statistical 2010, beginning in January 2007 with the onset of the inter- findings reported in Table 2 suggest that these results cannot vention. The null hypothesis of no effect of the intervention be explained by autocorrelation, seasonal variation, or preex- on mortality trends was strongly rejected for both mortality isting trends in DFR. rate series. Diagnostic tests for the broken-trend intervention analysis indicate the appropriateness of the statistical assumptions underlying the analyses. These results are thus Regression Diagnostics for DFR consistent with the conclusion that the intervention may have All diagnostic tests reported in Table 2 for adequacy of the contributed to the observed trend shifts for both mortality estimated intervention model are satisfactory. That no regres- rates. sion residuals exceed 3.5 standard errors indicates the The use of an intervention at the national level in a pro- absence of extreme outliers. The LM tests for autocorrelation spective quasi-experimental design is inherently stronger as of residuals at lags 1 to 6 and lags 1 to 12 are not significant, a design than an ex post analysis of archival data on the rela- and no autocorrelations at lags 1 to 36 are individually sig- tionship of interdependent factors using variants of multiple nificant, indicating that the regression residuals appear to be regression analysis (Glass, 1997). Nevertheless, it is neces- serially uncorrelated, stationary white noise. The frequency sary to consider possible threats to internal validity, namely, domain test (Baum & Wiggins, 2000; Robinson, 1995) alternative possible causes of the reduction of the stress- rejects the null hypothesis (p < .001) that the residual series related mortality variables investigated in these studies. is nonstationary (see Table 2). The latter test, plus the A major societal impact that occurred during the interven- absence of significant residual autocorrelation, lend further tion period was the economic recession of 2008-2009 associ- support to the conclusion of the Perron test (see Table 2) that ated with the subprime mortgage crisis. However, the onset DFR is likely broken-trend stationary. Thus, the diagnostic and duration of this recession (December 2007, with peak 10 SAGE Open Table 2. OLS Regression Analysis of Monthly U.S. DFR. a b Parameter Estimate SE t ratio β 3.484 0.606 5.76*** −2 −3 β 2.998 × 10 6.266 × 10 4.78*** −2 −3 β − β −2.897 × 10 6.968 × 10 −4.16*** 2 1 −2 β 0.498 9.560 × 10 5.21*** S 0.276 0.146 1.89 1t S −0.417 0.160 −2.61** 2t S 0.507 0.146 3.47*** 3t S −0.336 0.170 −1.97 4t S 0.305 0.150 2.03* 5t S −0.433 0.165 −2.63** 6t S 0.189 0.147 1.29 7t S −0.223 0.154 −1.45 8t S −0.529 0.145 −3.64*** 9t S −0.091 0.139 −0.65 10t S −0.422 0.141 −2.99** 11t F statistic: F(14, 86) = 102.30** Mean of DFR = 9.856 SE of regression = 0.296 SE of DFR = 1.152 Sum of squared residuals = 7.523 Log-likelihood = −12.159 2 2 R = .943; Adjusted R = .934 BIC = 0.926; AIC = 0.538 Diagnostics Serial correlation test Heteroscedasticity test Lags 1-6: F(6, 80) = 0.918 (p = .487) F(19, 81) = 1.359 (p = .172) Lags 1-12: F(12, 74) = 0.823 (p = .627) Test for ARCH Normality test: χ (2) = 0.951 (p = .622) Lags 1-6: F(6, 89) = 0.883 (p = .511) Robinson test : t(61) = −6.175 (p < .001) Test of functional form Perron unit root test: τ = −6.477 (p < .01) F(2, 84) = 1.990 (p = .143) Note. Sample is August 2002 to December 2010, N = 101. OLS = ordinary least squares; DFR = drug-related fatality rate; BIC = Bayesian information criterion; AIC = Akaike information criterion; ARCH = autoregressive conditional heteroscedasticity. OLS SEs and t ratios. df = 86. Frequency domain test for nonstationarity (Robinson, 1995). *p < .05. **p < .01. ***p < .001. unemployment in October 2009) does not precisely track the explanation will require almost simultaneous implementa- intervention period. More importantly, the economic stresses tion of a treatment or medical technology or program that associated with the recession (e.g., unemployment) would be would influence both the variables of these studies and dis- predicted, if anything, to increase rather than decrease the proportionately affect two different subpopulations of the stress-related health indicators examined in these studies. United States (higher prevalence for African American One specific factor to consider in the case of drug-related women in the case of IM and for middle-age Whites in the deaths is that one might speculate that increased public and case of DD). We are unable to think of any such plausible professional awareness of the hazards of opioid analgesics in alternative explanation that would satisfy these two require- recent years might have contributed to a reduction in their ments. In the absence of plausible alternative explanations, prescription during the intervention period. As outlined in the significant intervention impacts on both mortality vari- the introduction, increased used of opioid drugs is consid- ables found in this article suggest the possibility of some ered to contribute substantially to the rise of DD. However, underlying connection between individuals who are not data indicate that opioid sales increased throughout the base- apparently behaviorally connected. As outlined in the intro- line and intervention period of this research, reaching their duction to this article, such a connection independent of peak in 2011 (Volkow, 2014). Thus, the opioids were becom- behavioral interaction is described as the collective con- ing more available rather than less available through the sciousness of society. More specifically, the Vedic tradition intervention period. of India identifies collective consciousness as having its Because of the national scope of measurement of these basis in an underlying level of pure consciousness, which is studies, and the specificity of the intervention period and predicted to be influenced by the intervention examined associated changes in mortality rates, a viable alternative here, group practice of the advanced TM-Sidhi program. Dillbeck and Cavanaugh 11 Figure 4. Comparison of DFR trends during the preintervention and intervention periods. Note. Panel (a) shows that the positive linear trend for DFR in the preintervention period shifts to a flatter trend during the intervention period 2007 through 2010 (equilibrium trend slopes). The DFR slope is the rate of change per month of monthly drug-related fatalities per 1 million population. Panel (b) displays the statistically significant negative trend shift for DFR during the intervention with 95% confidence interval. DFR = drug-related fatality rate. Acknowledgments An alternative hypothesis affecting generalizability of the results is whether there is selection bias in the use of The authors appreciate the cooperation of the Invincible America members in the intervention group who were already Assembly Office of Maharishi University of Management for pro- experts in the Transcendental Meditation and TM-Sidhi viding group meditation data used in this study. The second author acknowledges access to the use of a university laptop and software program and may therefore be a special group in terms of for this research. consciousness training. From a theoretical perspective, it is only necessary that the members of the intervention group be trained in the advanced TM-Sidhi program to stimulate Declaration of Conflicting Interests the hypothesized field of pure consciousness. It is under- The authors declared no potential conflicts of interest with stood that, as a practical necessity, the individuals who par- respect to the research, authorship, and/or publication of this ticipated in this project would be those with some prior article: The authors have no financial relationship with the expertise/training to volunteer their time. But, in principle, foundation that teaches Transcendental Meditation in the for this effect to be created, any group with basic stability, United States (Maharishi Foundation USA). The first author such as groups of students or the military, could be is a Trustee and research professor and the second author is employed once trained in the TM-Sidhi program. In some a retired faculty member of the university sponsoring the countries, such groups are being created and future research project evaluated in this research, but both currently receive will be necessary to confirm their effects. no financial compensation from the university for their The result of these studies, if further replicated, provide research or other activities, which are donated as contrib- government leaders with a potential means to relieve stress uted services. on a large scale in society that is independent of structural social change and benefits public health. Although social Funding stress as a mediating variable was not directly measured, The authors received no financial support for the research and/or the consistency of the present results with those reported authorship of this article. previously for reduced homicide and violent crime (Dillbeck & Cavanaugh, 2016) lends support to a hypothe- Notes sis of such a mediating influence. In light of these results, it 1. Transcendental Meditation and TM-Sidhi are service marks is suggested that when TM-Sidhi groups are established registered in the U.S. Patent and Trademark Office, used by with government support (most easily, with existing groups, Maharishi Foundation, USA, under sublicense. such as military personnel), the potential impact of such 2. Crime measures were daily Indian Penal Code crimes for New TM-Sidhi groups be evaluated on stress-related public Delhi, weekly Philippine Crime Index totals for Metro Manila, health parameters. and monthly FBI Part 1 crimes for Puerto Rico. 3. From late July to late October 2006, a second group was formed Authors’ Note in Washington, D.C., to add to the effect of daily group prac- tice of the TM-Sidhi program. Because the two groups were Raw data may be obtained from the first author. 12 SAGE Open separate and distant, the predicted effect of these two groups 11. The Robinson periodogram test evaluates the null hypothesis 2 2 () nn + was added independently, that is, square root of , that the differencing parameter d takes a nonstationary value ver- 1 2 where n refers to the Iowa group and n to the Washington sus the alternative of stationarity. In the theoretical framework 1 2 group.The Washington group was much smaller than the Iowa underlying this test, both integer and fractional values of d are group, so the effect was inconsequential for the total number permitted, allowing for the possibility of fractional integration, for those 3 months, although it was included for the sake of including possible “long memory” behavior for values of d > 0. completeness. The Robinson test rejects the null hypothesis that the estimated 4. Scholarship funding to Maharishi University of Management difference parameter falls in the nonstationary region d ≥ 0.5 in and its related organizations was provided by the Howard and favor of the alternative d < 0.5 (see Table 1), thus supporting Alice Settle Foundation. stationarity of the IMR residual series. The sample estimate of 5. Data for year 2010 are compiled from the Multiple Cause of d = −0.2695 indicates stationary “short memory” or “antiper- Death File 2010, Series 20 No. 2P, 2012. Data for year 2009 sistence,” presumably reflecting the predominance of negative are compiled from the Multiple Cause of Death File 2009, autocorrelations among the largest residual autocorrelations. Series 20 No. 2O, 2012. Data for year 2008 are compiled from 12. As discussed in Dillbeck and Cavanaugh (2016), standard the Multiple Cause of Death File 2008, Series 20 No. 2N, approaches to testing for nonstationarity (unit root tests) such 2011. Data for year 2007 are compiled from Multiple Cause of as the augmented Dickey–Fuller test (Dickey & Fuller, 1979; Death File 2007, Series 20 No. 2M, 2010. Data for years 2005 Said & Dickey, 1984) and others have been shown to be biased to 2006 are compiled from Multiple Cause of Death File 2005- in favor of concluding that a broken-trend stationary time 2006, Series 20, No. 2L, 2009. Data for years 1999 to 2004 are series (TS) is nonstationary. compiled from the Multiple Cause of Death File 1999-2004, 13. Following Zivot and Andrews (1992), we test for nonsta- Series 20, No. 2J, 2007. tionarity using the “change in growth” broken-trend model 6. The effect size f is given by the t ratio (or square root of the based on Equation 15 in Perron (1989, p. 1381), which is F statistic) for the regression coefficient divided by the square Model B in Zivot and Andrews (1992, p. 253). The test root of the residual degrees of freedom. The unsquared effect statistic has a nonstandard distribution tabulated by Perron size metric is reported, rather than squared, because unsquared (1989, Table V.B). The critical value for the test depends measures better indicate the relative magnitude of effects across on the parameter λ= tN /, the time of the break divided variables (Darlington, 1990). The guidelines for large, medium, by total sample size. For the two variables analyzed in this and small values of f mentioned above are the square root of article, λ = 0.51. those given by Cohen for f (0.35, 0.15, and 0.02, respectively). 14. Because of possible monthly seasonality displayed by IMR, a 7. The estimated change in trend slope of −0.0229146 (before maximum of 12 lags of first-differenced IMR were considered rounding) gives the estimated reduction in the mathematically for inclusion in the regression used to calculate the Perron test expected value of the monthly infant mortality rate (IMR) for statistic. Because superfluous regressors inflate SEs for the each month of the intervention period relative to the preinter- estimated regression coefficients (Greene, 2011), lags that vention trend. Multiplying by 48 months gives −1.100 fatali- were not significant at the 20% level (lags 6-10 and 12) were ties per 10,000 live births, the expected value of the cumulative deleted from the Perron test regression, thus substantially total change in IMR for 2007 to 2010. increasing the power of the Perron test. 8. The preintervention data were used to generate monthly fore- 15. Omitting the seasonal coefficients to save space, the estimated casts of IMR for 2007 to 2010 based on ordinary least squares long-run equation is given by DFR = 6.945 + 0.0598t − 0.0578 (OLS) estimates of Equation 1 after deleting the trend-change DT , where the SEs for the regression coefficients are 0.3154, variable DT from the model. The cumulative total number of 0.00433, and 0.00811, respectively. fatalities averted for 2007 to 2010 (N ) was calculated by 16. The SE for the estimate of κ reported by PcGive 14 software IM () ββ − summing the difference between forecast and actual IM for is calculated from the SEs of β and in Equation each month: 2 by numerical differentiation using a nonlinear algorithm (Bårdsen, 1989; Doornik & Hendry, 2013). As shown by Inder   () forecast IMR − actual IMR × kk (1993), t tests for κ based on critical values from the standard N = ,, k = 12,,  48,,   IM b /, 10 000   normal distribution have both good power and actual p values  k  close to their nominal level. where N is the estimate of total prevented IM and b is the 17. The long-run, or equilibrium, change in slope −0.0577561 IM k number of U.S. live births for month k. The summation is over (before rounding) gives the estimated reduction in the math- kt =− () t k, the month of the intervention, with for tt > , ematically expected value of drug-related fatalities per mil- B B where t is the month of the trend break. lion population (DFR) for each month of the intervention 9. Following the recommendation of Kiviet (1986) and Harvey period relative to the preintervention trend. Multiplying (1990) for smaller samples, the Breusch–Godfrey and all other by 48 months gives the expected value of the cumulative Lagrange Multiplier (LM) tests reported in Table 2 are given change for 2007 to 2010: −2.7723 monthly fatalities per mil- in their F statistic form rather than chi-square. lion population. 10. The bandwidth of four Newey and West (1987) lags was 18. Using preintervention data, the OLS estimate of Equation 2 (with selected automatically by PcGive 14 using the integer part of trend shift variable DT omitted) was used to generate monthly 2/9 4(N / 100) . Experimentation with a wide range of alternative ex ante (dynamic) forecasts of DFR for 2007 to 2010 (Doornik & bandwidths indicated that the conclusions of the trend shift Hendry, 2013). The cumulative total number of averted drug-related analysis are not sensitive to bandwidth. deaths during 2007 through 2010 (N ) was estimated as follows: DD Dillbeck and Cavanaugh 13 Centers for Disease Control and Prevention, National Center for   () forecast DFRa − ctual DFR kk Nk = ,, = 12,,  48,   DD Health Statistics. (2012). Underlying cause of death, 1999- USPOP     2010 on CDC WONDER online database, released 2012. where summation is over month k of the intervention and Retrieved from http://wonder.cdc.gov/ucd-icd10.html USPOP is the monthly U.S. population estimate (in millions). k Centers for Disease Control and Prevention, National Center for 19. The Robinson test rejects the null hypothesis that the differ- Health Statistics. (2015a). Underlying cause of death 1999- encing parameter d falls in the nonstationary region d ≥ 0.5 2013 on CDC WONDER online database, released 2015. in favor of the stationary alternative d < 0.5. The sample esti- Retrieved from http://wonder.cdc.gov/ucd-icd10.html mate of the differencing parameter is d = −0.0232. The null Centers for Disease Control and Prevention, National Center for hypothesis of stationarity, d = 0, also was not rejected, t(61) = Health Statistics. (2015b). Natality public use data 2003-2006, −0.2995, p = .766. All frequency domain tests were estimated 2007-2013 on CDC WONDER online database, released 2012. in Stata 14 software employing add-on module roblpr (Baum Retrieved from http://wonder.cdc.gov/natality-v2006.html and & Wiggins, 2000; StataCorp, 2015). The number of periodo- http://wonder.cdc.gov/natality-current.html gram ordinates used in the tests for DFR and IMR was the Chelimsky, E., Shadish, W. R., & Orwin, R. G. (1997). Twenty-one 0.9 default value m = N (m = 63), where N = 101 is the effective years and counting: The interrupted time series comes of age. sample size. The test results were not sensitive to alternative In E. Chelimsky & R. Shadish (Eds.), Evaluation for the 21st values of m. century (pp. 443-465). London, England: Sage. 20. As in the case of the Perron test for IMR, lags 1 to 12 of first- Cohen, J. (1988). Statistical power analysis for the behavioral sci- differenced DFR that were not significant at the 20% level ences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum. (lags 1 and 4–10) were deleted from the regression equation Cook, T. D., & Campbell, D. T. (1979). Quasi-experimentation: for the Perron test, reducing SEs for the remaining regressors Design and analysis issues for field settings. Boston, MA: and increasing the power of the test. Houghton Mifflin. Darlington, R. B. (1990). Regression and linear models. New York, NY: McGraw-Hill. References Davies, J. L., & Alexander, C. N. (2005). Alleviating political vio- Alexander, C. N., Davies, J. L., Dixon, C., Dillbeck, M. C., Druker, lence through reducing collective tension: Impact assessment S., Oetzel, R. M., . . . Orme-Johnson, D. W. (1990). Growth analyses of the Lebanon war. Journal of Social Behavior and of higher states of consciousness: Maharishi’s Vedic psychol- Personality, 17, 285-338. ogy of human development. In C. N. Alexander & E. J. Langer Dickey, D., & Fuller, W. A. (1979). Distribution of the estimates (Eds.), Higher stages of human development: Perspectives on for autoregressive time series with a unit root. Journal of the adult growth (pp. 286-340). New York, NY: Oxford University American Statistical Association, 74, 427-431. Press. Dillbeck, M. C. (1990). Test of a field hypothesis of conscious- Assimakis, P. D., & Dillbeck, M. C. (1995). Time series analy- ness and social change: Time series analysis of participation sis of improved quality of life in Canada: Social change, in the TM-Sidhi program and reduction of violent death in the collective consciousness, and the TM-Sidhi program. U.S. Social Indicators Research, 22, 399-418. doi:10.1007/ Psychological Reports, 76, 1171-1193. doi:10.2466/ BF00303834 pr0.1995.76.3c.1171 Dillbeck, M. C., & Alexander, C. N. (1989). Higher states of con- Banerjee, A., Dolado, J. J., Galbraith, J. W., & Hendry, D. F. (1993). sciousness: Maharishi Mahesh Yogi’s Vedic psychology of Co-integration, error correction and the econometric analysis human development. The Journal of Mind and Behavior, 10, of non-stationary data. Oxford, UK: Oxford University Press. 307-334. Banerjee, A., Dolado, J. J., & Mestre, R. (1998). Error-correction Dillbeck, M. C., Banus, C. B., Polanzi, C., & Landrith, G. S., III. mechanism tests for cointegration in a single equation frame- (1988). Test of a field model of consciousness and social work. Journal of Time Series Analysis, 19, 267-283. change: Transcendental Meditation and TM-Sidhi program and Bårdsen, G. (1989). The estimation of long-run coefficients decreased urban crime. The Journal of Mind and Behavior, 9, from error-correction models. Oxford Bulletin of Economics 457-486. and Statistics, 51, 345-350. doi:10.1111/j.1468-0084.1989. Dillbeck, M. C., & Bronson, E. C. (1981). Short-term longitudinal mp51003008.x effects of the Transcendental Meditation technique on EEG Baum, C. F., & Wiggins, V. (2000). Tests for long memory in a power and coherence. International Journal of Neuroscience, time series. Stata Technical Bulletin, 57, 39-44. 14, 147-151. doi:10.3109/00207458108985827 Box, G. E. P., & Jenkins, G. M. (1976). Time series analysis: Dillbeck, M. C., & Cavanaugh, K. L. (2016). Societal violence Forecasting and control. San Francisco, CA: Holden-Day. and collective consciousness: Reduction of U.S. homi- Case, A., & Deaton, A. (2015). Rising morbidity and mortality in cide and urban violent crime rates. SAGE Open, 6(2), 1-16. midlife among white non-Hispanic Americans in the 21st cen- doi:10.1177/2158244016637891 tury. Proceedings of the National Academy of Sciences of the Dillbeck, M. C., Cavanaugh, K. L., Glenn, T., Orme-Johnson, D. W., United States of America, 112(49), 15078-15083. doi:10.1073/ & Mittlefehldt, V. (1987). Effects of Transcendental Meditation pnas.1518393112 and the TM-Sidhi program on quality of life indicators: Cavanaugh, K. L., & Dillbeck, M. C. (2017). The contribution of Consciousness as a field. The Journal of Mind and Behavior, proposed field effects of consciousness to the prevention of 8, 67-104. US accidental fatalities: Theory and empirical tests. Journal of Dillbeck, M. C., & Orme-Johnson, D. W. (1987). Physiological Consciousness Studies, 24, 53-86. differences between Transcendental Meditation and rest. 14 SAGE Open American Psychologist, 42, 879-881. doi:10.1037/0003- Hendry, D. F. (1995). Dynamic econometrics. Oxford, UK: Oxford 066X.42.9.879 University Press. Dillbeck, M. C., & Rainforth, M. V. (1996). Impact assessment Inder, B. (1993). Estimating long-run relationships in economics: A analysis of behavioral quality of life indices: Effects of group comparison of different approaches. Journal of Econometrics, practice of the Transcendental Meditation and TM-Sidhi pro- 57, 53-68. doi:10.1016/0304-4076(93)90058-D gram. Proceedings of the American Statistical Association, Jevning, R., Wilson, A. F., & Davidson, J. M. (1978). Social Statistics Section, 38-43. Adrenocortical activity during meditation. Hormones and Dole, N., Savitz, D. A., Hertz-Picciotto, I., Siega-Riz, A. M., Behavior, 10, 54-60. McMahon, M. J., & Buekens, P. (2003). Maternal stress and Jones, C. M., Mack, K. A., & Paulozzi, L. J. (2013). Pharmaceutical preterm birth. American Journal of Epidemiology, 157, 14-24. overdose deaths, United States, 2010. Journal of the American doi:10.1093/aje/kwf176 Medical Association, 309, 657-659. Doornik, J. A., & Hansen, H. (2008). An omnibus test for univariate Kiviet, J. F. (1986). On the rigour of some misspecification tests for and multivariate normality. Oxford Bulletin of Economics and modelling dynamic relationships. Review of Economic Studies, Statistics, 70, 927-939. doi:10.1111/j.1468-0084.2008.00537.x 53, 241-261. doi:10.2307/2297649 Doornik, J. A., & Hendry, D. F. (2013). Empirical economet- Latendresse, G. (2009). The interaction between chronic stress ric modelling–PcGive 14: Volume 1. London, England: and pregnancy: Preterm birth from a biobehavioral perspec- Timberlake Consultants Press. tive. Journal of Midwifery & Women’s Health, 54, 8-17. Enders, W. (2010). Applied econometric time series (3rd ed.). doi:10.1016./j.jmwh.2008.08.001 Hoboken, NJ: John Wiley. Linsky, A. S., Bachman, R., & Straus, M. A. (1995). Stress, culture, Engle, R. F. (1982). Autoregressive conditional heteroscedasticity & aggression. New Haven, CT: Yale University Press. with estimates of the variance of United Kingdom inflation. Linsky, A. S., & Straus, M. A. (1986). Social stress in the United Econometrica, 50, 987-1007. States. Dover, MA: Auburn House Publishing. Gennaro, S., & Hennessy, M. D. (2003). Psychological and MacDorman, M. F., & Mathews, T. J. (2008). Recent trends in physiological stress: Impact on preterm birth. Journal of infant mortality in the United States (NCHS Data Brief, No. 9). Obstetric, Gynecologic & Neonatal Nursing, 32, 668-675. Hyattsville, MD: National Center for Health Statistics. doi:10.1177/0884217503257484 MacDorman, M. F., & Mathews, T. J. (2011). Understanding racial Glass, G. V. (1997). Interrupted time series quasi-experiments. In and ethnic disparities in U.S. infant mortality rates (NCHS R. M. Jaeger (Ed.), Complementary methods for research in Data Brief, No. 74). Hyattsville, MD: National Center for education (2nd ed., pp. 589-608). Washington, DC: American Health Statistics. Educational Research Association. MacDorman, M. F., Mathews, T. J., Mohangoo, A. D., & Zeitlin, Godfrey, L. G. (1978). Testing for higher order serial correla- J. (2014, September 4). International comparisons of infant tion in regression equations when the regressors include mortality and related factors: United States and Europe, lagged dependent variables. Econometrica, 46, 1303-1313. 2010 (National Vital Statistics Reports, Vol. 63, Number 5). doi:10.2307/1913830 Hyattsville, MD: National Center for Health Statistics. Granger, C. W. J., & Newbold, P. (1986). Forecasting economic MacLean, C. R. K., Walton, K. G., Wenneberg, S. R., Levitsky, D. time series (2nd ed.). Orlando, FL: Academic Press. K., Mandarino, J. P., Waziri, R., . . . Schneider, R. H. (1997). Greene, W. H. (2011). Econometric analysis (7th ed.). Upper Effects of the Transcendental Meditation program on adap- Saddle River, NJ: Prentice Hall. tive mechanisms: Changes in hormone levels and responses to Hagelin, J. S. (1987). Is consciousness the unified field? A field stress after 4 months of practice. Psychoneuroendocrinology, theorist’s perspective. Modern Science and Vedic Science, 1, 22, 277-295. 28-87. Maharishi Mahesh Yogi. (1969). Maharishi Mahesh Yogi on the Hagelin, J. S., Rainforth, M. V., Orme-Johnson, D. W., Cavanaugh, Bhagavad-Gita: A new translation and commentary, chapters K. L., Alexander, C. N., Shatkin, S. F., . . . Ross, E. (1999). 1 to 6. Baltimore, MD: Penguin. Effects of group practice of the Transcendental Meditation Maharishi Mahesh Yogi. (1977). Creating an ideal society. program on preventing violent crime in Washington, DC: Rheinweiler, Germany: Maharishi European Research Results of the National Demonstration Project, June- University Press. July 1993. Social Indicators Research, 47, 153-201. Maharishi Mahesh Yogi. (1986). Life supported by natural law. doi:10.1023/A:1006978911496 Washington, DC: Age of Enlightenment Press. Harmon, K. (2010, April 6). Prescription drug deaths increase dra- Mathews, T. J., MacDorman, M. F., & Thoma, M. E. (2015, August matically. Scientific American. Retrieved from http://www.sci- 6). Infant mortality statistics from the 2013 period linked birth/ entificamerican.com/article/prescription-drug-deaths/ infant death data set (National Vital Statistics Reports, Vol. Harvey, A. (1990). The econometric analysis of time series (2nd 64, Number 9). Hyattsville, MD: National Center for Health ed.). Cambridge, MA: MIT Press. Statistics. Hatchard, G. D., Deans, A. J., Cavanaugh, K. L., & Orme-Johnson, Newey, W. K., & West, K. (1987). A simple positive semi-definite, D. W. (1996). The Maharishi effect: A model for social heteroscedasticity and autocorrelation consistent covariance improvement. Time series analysis of a phase transition to matrix. Econometrica, 55, 703-708. reduced crime in Merseyside metropolitan area. Psychology, Orme-Johnson, D. W., Alexander, C. N., & Davies, J. L. (1990). The Crime & Law, 2, 165-174. doi:10.1080/10683169608409775 effects of the Maharishi Technology of the Unified Field: Reply Dillbeck and Cavanaugh 15 to a methodological critique. Journal of Conflict Resolution, 34, StataCorp. (2015). Stata Statistical Software: Release 14. College 756-768. doi:10.1177/0022002790034004009 Station, TX: Author. Orme-Johnson, D. W., Alexander, C. N., Davies, J. L., Chandler, Travis, F., Haaga, D. A. F., Hagelin, J., Tanner, M., Arenander, A., H. M., & Larimore, W. E. (1988). International peace project Nidich, S., . . . Schneider, R. H. (2010). A self-referential default in the Middle East: The effects of the Maharishi Technology of brain state: Patterns of coherence, power, and eLORETA sources the Unified Field. Journal of Conflict Resolution, 32, 776-812. during eye-closed rest and Transcendental Meditation practice. doi:10.1177/0022002788032004009 Cognitive Processing, 11, 21-30. doi:10.1007/s10339-009-0343-2 Orme-Johnson, D. W., & Oates, R. M. (2009). A field-theoretic Travis, F., Haaga, D. A. F., Hagelin, J., Tanner, M., Nidich, S., Gaylord- view of consciousness: Reply to critics. Journal of Scientific King, C., . . . Schneider, R. H. (2009). Effects of Transcendental Exploration, 23, 139-166. Meditation practice on brain functioning and stress reactivity in Paulozzi, L. J., Budnitz, D. S., & Xi, Y. (2006). Increasing deaths from college students. International Journal of Psychophysiology, 71, opioid analgesics in the United States. Pharmacoepidemiology 170-176. doi:10.1016/j.ijpsycho.2008.09.007 & Drug Safety, 15, 618-625. doi:10.1002/pds.1276 Volkow, N. D. (2014, May 4). Testimony to Congress—America’s Perron, P. (1989). The great crash, the oil price shock, and the unit addiction to opioids: Heroin and prescription drug abuse. root hypothesis. Econometrica, 57, 1361-1401. doi:10.2307/ National Institute on Drug Abuse of the National Institutes of 1913712 Health. Retrieved from http://www.drugabuse.gov/about-nida/ Perron, P. (2006). Dealing with structural breaks. In T. C. Mills legislative-activities/testimony-to-congress/2015/americas- & K. Patterson (Eds.), Palgrave handbook of econometrics, addiction-to-opioids-heroin-prescription-drug-abuse Vol. 1: Econometric theory (pp. 278-352). Basingstoke, UK: Walton, K. G., Fields, J. Z., Levitsky, D. K., Harris, D. A., Pugh, Palgrave Macmillan. N. D., & Schneider, R. H. (2004). Lowering cortisol and Radhakrishnan, S. (1953). Principal Upanishads. New York, NY: CVD risk in postmenopausal women: A pilot study using the Harper. Transcendental Meditation program. Annals of the New York Ramsey, J. B. (1969). Tests for specification errors in classical Academy of Sciences, 1032, 211-215. linear least squares regression analysis. Journal of the Royal White, H. (1980). A heteroskedastic-consistent covariance matrix Statistical Society: Series B—Methodological, 31, 350-371. estimator and a direct test for heteroskedasticity. Econometrica, Rappoport, P., & Reichlin, L. (1989). Segmented trends and non- 48, 817-838. doi:10.2307/1912934 stationary time series. Economic Journal, 99, 168-177. Zivot, E., & Andrews, K. (1992). Further evidence on the great Robinson, P. M. (1995). Log-periodogram regression of time series crash, the oil price shock, and the unit root hypothesis. Journal with long range dependence. Annals of Statistics, 23, 1048-1072. of Business & Economic Statistics, 10, 251-270. doi:10.1080/0 Roth, R. (2002). Maharishi Mahesh Yogi’s Transcendental 7350015.1992.10509904 Meditation. Washington, DC: Primus. Said, S. E., & Dickey, D. (1984). Testing for unit roots in autoregres- Author Biographies sive moving-average models with unknown order. Biometrika, Michael C. Dillbeck, PhD, is a research scientist at the Institute of 71, 599-607. doi:10.1093/biomet/71.3.599 Science, Technology and Public Policy and research professor and Schneider, R. H., & Carr, T. (2014). Transcendental Meditation in the Trustee at Maharishi University of Management. His research spans prevention and treatment of cardiovascular disease and pathophysi- the physiological, psychological, and sociological effects on the ological mechanisms: An evidence-based review. Advances in Transcendental Meditation and TM-Sidhi programs. Integrative Medicine, 1, 107-112. doi:10.1016/j.aimed.2014.08.003 Kenneth L. Cavanaugh, PhD, is a senior research scientist at the Shadish, W. R., Cook, T. D., & Campbell, D. T. (2002). Institute of Science, Technology and Public Policy and a professor Experimental and quasi-experimental designs for generalized of Applied Statistics, Emeritus at Maharishi University of causal inference. Boston, MA: Houghton Mifflin. Management. His currrent principal research interest is empirical Sinha, R. (2008). Chronic stress, drug use, and vulnerability to study of the hypothesized effects of the group practice of the addiction. Annals of the New York Academy of Sciences, 1141, TM-Sidhi program on societal quality of life. 105-130. doi:10.1196/annals.141.030 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png SAGE Open SAGE

Group Practice of the Transcendental Meditation® and TM-Sidhi® Program and Reductions in Infant Mortality and Drug-Related Death: A Quasi-Experimental Analysis

SAGE Open , Volume 7 (1): 1 – Mar 14, 2017

Loading next page...
 
/lp/sage/group-practice-of-the-transcendental-meditation-and-tm-sidhi-program-IYSNNDrqWH

References (87)

Publisher
SAGE
Copyright
Copyright © 2022 by SAGE Publications Inc, unless otherwise noted. Manuscript content on this site is licensed under Creative Commons Licenses.
ISSN
2158-2440
eISSN
2158-2440
DOI
10.1177/2158244017697164
Publisher site
See Article on Publisher Site

Abstract

These two studies tested the prediction that the group practice of a procedure for the development of consciousness, the Transcendental Meditation and TM-Sidhi program, by a sufficiently large group of individuals would be sufficient to reduce collective stress in the larger population, reflected in two stress-related health indicators, infant mortality rate and drug- related fatality rate. Based on theoretical prediction and prior research, from January 2007 through 2010 (intervention period), this effect should have been measurable. Change in the rates of these two indicators during the intervention period were estimated from 2002 through 2010 data using a broken-trend (or segmented trend) intervention model with time series regression methods. Significant changes in trend for both the infant mortality rate and drug-related fatality rate were evident at the predicted time and in the predicted direction, controlling for preintervention trends, seasonality, and autocorrelation. The changes in trend were both statistically and practically significant, indicating an average annual decline of 3.12% in infant mortality rate and 7.61% in drug-related fatality rate. Diagnostic tests are satisfactory and indicate that it is unlikely that the statistical results are attributable to spurious regression. The mechanism for these collective effects is discussed in view of possible alternative hypotheses. Keywords behavioral sciences, alcohol, drugs, tobacco, sociology of health and illness, sociology, social sciences, collective behavior/ social movements, social change and modernization, medical sociology This article reports results of two studies on the quality of life We will first describe these procedures, then the theoreti- in the United States that extend a previous research program cal implications for the nature of consciousness implied by on consciousness and social well-being into the area of col- previous research on these procedures, and then discuss the lective stress and stress-related indicators of public health, indicators chosen for the present studies. namely, infant mortality rate (IMR) and drug-related fatality The Transcendental Meditation technique is described by rate (DFR). Previous research analyzed the impact of the its founder as a simple, experiential mental procedure of same intervention on homicide and violent crime (Dillbeck “turning the attention inwards towards the subtler levels of a & Cavanaugh, 2016) and motor vehicle fatalities and fatali- thought until the mind . . . arrives at the source of the thought” ties due to other accidents (Cavanaugh & Dillbeck, 2017). (Maharishi Mahesh Yogi, 1969, p. 470). “Source of thought” In a broader context, this article expands a substantive indicates a state of “pure consciousness” gained when the body of research on the collective effects of changes in con- mind settles down to a mode of inner silence in which the sciousness and quality of life. In these studies, change in the division of knower, knowing, and known is transcended and quality of societal consciousness is measured by group prac- awareness is open to itself alone (Roth, 2002). tice of the TM-Sidhi program, an advance practice of the Transcendental Meditation technique. Previous studies have Maharishi University of Management, Fairfield, IA, USA found that when practiced by a sufficient number of people, Corresponding Author: these subjective procedures are associated with an extended Michael C. Dillbeck, Institute of Science, Technology, and Public Policy, influence on social behavior in society, a finding that has Maharishi University of Management, Fairfield, IA 52557, USA. implications for a broader understanding of consciousness. Email: mdillbeck@mum.edu Creative Commons CC BY: This article is distributed under the terms of the Creative Commons Attribution 3.0 License (http://www.creativecommons.org/licenses/by/3.0/) which permits any use, reproduction and distribution of the work without further permission provided the original work is attributed as specified on the SAGE and Open Access pages (https://us.sagepub.com/en-us/nam/open-access-at-sage). 2 SAGE Open Meta-analysis of research on individuals practicing A number of published studies have used time series (TS) Transcendental Meditation has found that increased mental intervention analysis or transfer function (TF) analysis (Box silence during the practice, in comparison with sitting with & Jenkins, 1976) to evaluate effects at the city, state, or eyes closed, is associated physiologically with a unique state national/regional levels on indicators that reflect reduction of of restful alertness (higher basal galvanic skin resistance, social stress, such as decreased crime and violence, or lower breath rate, lower plasma lactate); this state of restful improvement in comprehensive quality of life indices. alertness, or reduced physiological stress, is also evident in For example, Dillbeck, Cavanaugh, Glenn, Orme- participants outside the practice period (lower respiration Johnson, and Mittlefehldt (1987) found that when the size of rate, lower heart rate, lower plasma lactate, and fewer sponta- a group of TM-Sidhi program experts who had assembled on neous skin resistance responses; Dillbeck & Orme-Johnson, courses exceeded the √1% of the populations of each course 1987). Alertness is indicated by increased electroencephalo- location (Delhi, India; Metro Manila, Philippines; Puerto gram (EEG) coherence during Transcendental Meditation Rico), TS intervention analysis showed reduced crime inde- practice; increased integration of brain functioning is also pendent of alternative explanations. Using TS intervention found longitudinally (Dillbeck & Bronson, 1981; Travis methods, similar results were found on monthly crime rates et al., 2010; Travis et al., 2009). Additional research, includ- in Merseyside, United Kingdom (Hatchard, Deans, ing reduced cortisol during the practice and more effective Cavanaugh, & Orme-Johnson, 1996). Two studies in cortisol response to stress, indicates that the physiological Washington, D.C., using TF methodology found predicted changes during the practice are counter to those found during variations between the size of a TM-Sidhi group in the city stress and support recovery from stress (Jevning, Wilson, & and decreased violent crime rate when the size of the group Davidson, 1978; MacLean et al., 1997; Walton et al., 2004). was over the predicted threshold (Dillbeck, Banus, Polanzi, There is a large body of research on the stress-reducing & Landrith, 1988; Hagelin et al., 1999). The second study effects of this procedure for health, particularly cardiovascu- (Hagelin et al., 1999) was a prospective study with predic- lar health (e.g., reviewed by Schneider & Carr, 2014), and on tions registered in advance. Both studies in Washington, psychological variables, which can be considered in the con- D.C., found no evidence for alternative hypotheses. text of theories of human development (e.g., Alexander et al., At both state and national levels, improvement was found 1990; Dillbeck & Alexander, 1989). using TS intervention analysis on a comprehensive index of In the 1960s, based on the understanding of conscious- quality of life integrating multiple monthly behavioral and ness contained in the Vedic tradition of India, Maharishi pre- health-related variables (e.g., homicide or crime, motor vehi- dicted that even a small proportion of the society experiencing cle fatalities, cigarette consumption) during months when the this state of pure consciousness would be sufficient to enliven size of TM-Sidhi group was larger than the required √1% in greater order and positivity in the collective consciousness of Rhode Island (Dillbeck et al., 1987) or in the United States or society, reflected in decreased stress, resulting in decreased the United States plus Canada (Dillbeck & Rainforth, 1996). negative behavioral trends such as crime and violence Similarly, reductions in an index of national violent fatalities (Maharishi Mahesh Yogi, 1977). (homicide, suicide, and motor vehicle fatalities) were found In 1976, Maharishi introduced the advanced TM-Sidhi using TS intervention analysis over a multiple-year period program, the purpose of which is described as accelerating during weeks when the size of a stable, large TM-Sidhi group the integration of the inner silence experienced during was sufficient to predict an effect for either the United States Transcendental Meditation with daily activity outside of or Canada (Assimakis & Dillbeck, 1995; Dillbeck, 1990). meditation (Maharishi Mahesh Yogi, 1986). In discussions A study in Israel, spanning multiple societal levels and with physical scientists in the light of the ancient Vedic addressing multiple outcome parameters for each, found that knowledge, he formalized the prediction that when a group over a 2-month period, daily quality of life indices for practices the TM-Sidhi program together, only the square Jerusalem, Israel as a whole, and the Israel–Lebanon conflict root of 1% (√1%) of the population—as compared with 1% (ongoing at that time) significantly improved during days practicing Transcendental Meditation individually—would when the size of a temporary group was sufficiently large to be required to create a calming and orderly influence in soci- yield theoretically predicted effects at the city, national, or ety (Maharishi Mahesh Yogi, 1986). The square root term is regional levels (Orme-Johnson, Alexander, Davies, Chandler, by analogy to coherent phenomena in physical systems in & Larimore, 1988). Comparable results were found using TS which the combined intensity of coherent elements can be intervention analysis and TF analysis. A number of alterna- proportional to their square (Hagelin, 1987). On that basis, it tive explanations and alternative analysis models were effec- should be possible to influence very large populations: for tively considered both in the original article and in subsequent example, the √1% of a city of 9 million is 300 individuals, responses to methodological questions (Orme-Johnson, and is near 1,800 for the current U.S. population. Alexander, & Davies, 1990; Orme-Johnson & Oates, 2009). The relatively small number of required TM-Sidhi pro- In a more detailed and comprehensive analysis of vio- gram group participants permits quasi-experiments using a lence due to the Lebanese conflict, Davies and Alexander time scale of analysis finer than that of annual data. (2005) looked at all seven intervention periods during a Dillbeck and Cavanaugh 3 2.25-year period (1983-1985) when there were TM-Sidhi American mothers much more severely than White mothers program groups of sufficient size (on temporary courses) to or mothers of other racial/ethnic groups, and preterm birth is predict an impact upon the conflict in Lebanon either from substantially higher among African American mothers within Lebanon, near to Lebanon, or even at some distance. (MacDorman & Mathews, 2011; Mathews, MacDorman, & Using daily event data from nine news sources that were Thoma, 2015). Preterm birth risk has been shown to be blind coded by an independent Lebanese coder using scales related to perceived stress, pregnancy-related anxiety, and derived independently for research on the conflict, TS inter- perception of racial discrimination (Dole et al., 2003). The vention analysis indicated a significant impact of each of the causes of IM are multiple; nevertheless, stress is one major seven temporary groups on reduced conflict. The analysis influence, as described above. The cascade of physiological/ controlled for temperature, holidays, and weekends, and the hormonal events in the mother’s body initiated by stress and findings were independent of alternative TS noise models. their interaction with the physiology of labor initiation are Multiple indicators of reduced conflict also replicated this described in detail by Gennaro and Hennessy (2003) and by effect when combining intervention periods. Latendresse (2009). In these studies, there is an implied underlying connection DD of all types also reflect to a great degree the immedi- between individuals that would permit such far-reaching ate effects of stress and are affected by the degree of con- effects of stress reduction beyond that which could be scious alertness or vigilance. The category of DD fatalities explained by behavioral interactions. That is, as described by includes all deaths for which drugs are the essential cause, the Vedic tradition of India (Radhakrishnan, 1953), pure con- irrespective of whether the drug was illicit or prescription, sciousness is proposed to have a field-like character, as including acute poisoning and also medical conditions aris- opposed to the isolated quality of individual consciousness. ing from chronic use. DD have increased substantially since On that basis, it is predicted that a calming or stress-reducing 1990, a period in which prescriptions for opioid analgesics influence will be experienced in the broader collective con- for pain management rose dramatically. Between 1999 and sciousness of society (Maharishi Mahesh Yogi, 1986). 2002, deaths due to opioid analgesic poisonings rose over In extending these studies into the health area, we assume 90%, becoming a larger source of drug poisoning deaths than that collective stress (or its reduction) can have an impact those due to heroin or cocaine (Paulozzi, Budnitz, & Xi, upon some stress-related health outcomes, not only upon 2006). An investigation in 2010 found that of all drug over- behavioral violence such as crime. Analyzing U.S. annual dose deaths, 57.7% involved pharmaceuticals, and of those, data using a state stress index that combined economic, fam- 75.2% involved opioids (either alone or in combination with ily, and community stressors (Linsky, Bachman, & Straus, other drugs; Jones, Mack, & Paulozzi, 2013). 1995; Linsky & Straus, 1986), the degree of social stress pre- Opioid analgesics have been increasingly prescribed as dicted violent crime rate and also maladaptive behaviors part of more aggressive pain management regimens. (Linsky & Straus, 1986). However, social stress so defined Chemically, the opioids influence the body in a similar way was less associated with mortality due to illness, illnesses as opium-derived opiates such as morphine and heroin whose morbidity at the individual level are exacerbated by (Volkow, 2014). For this reason, vigilance must be main- stress. One of the possible reasons for this is the unknown tained to avoid addiction to opioid analgesics. Chronic stress and variable time lag between serious morbidity and mortal- has been found to increase the vulnerability to addiction ity, not to mention between serious stress and morbidity (Sinha, 2008). (Linsky & Straus, 1986). By 2010, among those 35 to 54 years of age, poisoning The health indicators selected for this study are those for had become the most common type of accidental death, more which mortality is directly influenced by stress, rather than common than auto-related deaths (Harmon, 2010). Recent those indicative of disease states that develop over time and research has found that the death rate for White non-Hispan- whose time course may be inconsistent and hard to specify. ics aged 45 to 54 years actually rose from 1999 to 2013, par- Specifically, infant mortality (IM) and deaths due to drugs ticularly among the least educated, reversing a trend of (DD) were selected for investigation. The following is a previous decades (Case & Deaton, 2015). The authors rationale for each outcome variable. hypothesize that this increase in mortality could be driven by IM is both a fundamental indicator of national health as factors including deaths due to the increased availability of well as a stress-related phenomenon in the United States. opioid prescriptions and, in some cases where addiction had Compared with other developed countries, IM rates are high, occurred, transition to heroin when opioid restriction with the United States ranking 26th among Organisation for increased. They note that financial stress and insecurity may Economic Co-Operation and Development (OECD) coun- have contributed to these results, including wage stagnation tries (MacDorman, Mathews, Mohangoo, & Zeitlin, 2014). and diminished retirement prospects. Morbidity also Preterm birth is one of the major causes of IM; in 2005, increased in this group as indicated by self-reported level of 36.5% of U.S. infant deaths were due to preterm-related physiological and mental health as well as increased preva- causes, and 68.6% of all infant deaths occurred to preterm lence of chronic pain. Increased DD and alcohol poisoning infants (MacDorman & Mathews, 2008). IM affects African (as well as several other factors) also increased in other 4 SAGE Open 5-year age groups (30-34, 60-64) although not enough to (Dillbeck & Cavanaugh, 2016), and this period was subse- raise the overall mortality rate. quently extended to other variables for the sake of compari- It is clear that the rapid rise of DD in the past two decades son (Cavanaugh & Dillbeck, 2017). The binary intervention is a serious health concern in the American population as a variable (I ) was specified as 0 from July 2001 to December whole. At the same time, it seems likely that this variable is 2006, and 1 from January 2007 to December 2010. (Archival directly affected by perceived life stress and, in the case of data of the group size was not available continuously prior to opioid medication, is influenced by vigilance and self-aware- July 2001.) ness to avoid addiction. It is in this context that change in DD is a potential reactive indicator of changes in the quality of Dependent Variables collective consciousness. The two studies reported here evaluate the effects on U.S. The dependent variable data for both studies were obtained DD and IM associated with the rapid creation of a large from the National Center for Health Statistics of the Centers Transcendental Meditation and TM-Sidhi group sufficient in for Disease Control and Prevention (CDC). For Study 1, the size to create a hypothesized influence of reduced stress and dependent variable is the monthly U.S. IMR within 1 week of increased alertness in the collective consciousness of the birth per 10,000 live births. The IM monthly total was United States. The rapid establishment of this group, approxi- obtained from the data set “Underlying Causes of Death, mating in size a step function that could be appropriately mod- 1999-2013” through the CDC WONDER Online Database eled by a binary variable, offered a straightforward opportunity (CDC, National Center for Health Statistics, 2015a). The for quasi-experimental intervention research (Cavanaugh & number of live births each month was available also from Dillbeck, 2017; Dillbeck & Cavanaugh, 2016). CDC WONDER or the National Center for Health Statistics The hypothesis of the current research, consistent with VitalStats system as natality public-use data in different previous studies, is that there would be a significant impact source files depending upon the years, 1995 to 2002, 2003 to of the independent variable measured in terms of decreased 2006, and 2006 to 2013, posted in November 2005, March rates of IM and drug-related deaths. 2009, and January 2015, respectively (CDC, National Center for Health Statistics, 2015b). For Study 2, the dependent variable is the drug-related General Method fatality rate (DFR) per 1 million population. As noted above, this category of mortality includes any type of drug and any Intervention circumstance of death, as long as drugs are cited on the death To assess the predicted effect on these two fatality rates of the record as the essential cause of death. Data were obtained largest group of TM-Sidhi participants in North America, the from the National Center for Health Statistics through the analysis used a binary intervention variable based on the size CDC WONDER Online Database (CDC, National Center of this group. The location of the group was in Fairfield, Iowa, for Health Statistics, 2012). The computation of the rate per at Maharishi University of Management, where students, fac- 1 million population used U.S. Census counts for April 2000 ulty, staff, and community members assemble each day to and April 2010, with monthly linear interpolation and exten- practice the Transcendental Meditation and TM-Sidhi pro- sion to December 2010. gram together before and after school or work. The meditation halls on campus record the morning and evening daily totals. Data Analysis To expand the size of the TM-Sidhi program group from under 1,000 to a number sufficient to create a predicted posi- To analyze the results of the quasi-experiment, we use inter- tive influence for the whole United States (approximately vention analysis, or interrupted TS analysis, of data for 2002 1,725 needed for the 297 million population at that time, to 2010 (Chelimsky, Shadish, & Orwin, 1997; Cook & according to the √1% formula), a special course was held at Campbell, 1979; Shadish, Cook, & Campbell, 2002) to test the University beginning in July 2006 for visitors from the the hypothesis that a significant reduction in trend for IMR United States or around the world. To further increase the and DFR occurred beginning with the onset of the interven- number of participants, starting in November 2006, a large tion period in January 2007. TS regression analysis is used to group of several hundred visiting Indian experts in the estimate a broken-trend intervention model (Perron, 1989; TM-Sidhi program were hosted nearby in their own facili- Rappoport & Reichlin, 1989) to test for the hypothesized ties. After that, the total number of TM-Sidhi participants trend shifts in IMR and DFR. The intervention model in each began to exceed the predicted threshold in January 2007 and case includes a preintervention linear trend with an exoge- remained above or near that level throughout the interven- nous structural break in the trend function at the theoretically tion period of the study, 2007 through 2010. An intervention predicted date of December 2006. The intervention compo- period of that length was adopted because when the first data nent is modeled as a binary (0-1) step function that triggers a collection began to evaluate this intervention, 2010 was the shift in the linear trend function with the onset of the inter- most recent data available for homicide and violent crime vention period. Dillbeck and Cavanaugh 5 Figure 1. Plots of monthly IMR and size of the TM-Sidhi group (GROUP). Note. In Panel (a), the plot of IMR for August 2002 to December 2010 displays irregular variation plus weak monthly seasonal variation around a flat or slightly declining preintervention trend that shifts to a more steeply declining trend in the intervention period starting in January 2007 (see vertical line in the plot). During the intervention period, actual IMR declines more rapidly than predicted by its prior trend. In Panel (b), the plot of the monthly average daily size of the TM-Sidhi group rises rapidly starting in July 2006 until in January 2007, it exceeds the predicted critical threshold of 1,725, the √1% of the U.S. population at that time. The average size of GROUP is 591 participants for the 53 preintervention months and 1,792 for the 48 months of the intervention. Panel (b) adapted from “The Contribution of Proposed Field Effects of Consciousness to the Prevention of US Accidental Fatalities: Theory and Empirical Tests,” by K. L. Cavanaugh and M. C. Dillbeck, 2017, Journal of Consciousness Studies, 24(1-2), p. 66. Copyright 2017 by Imprint Academic, Ltd. IMR = infant mortality rate. preintervention TS behavior (see Note 8). The sample is Study 1: Results of Analysis of Monthly August 2002 through December 2010, with effective sample IMR size N = 101. This sample was chosen to give the largest pos- Plot of Monthly IMR sible effective sample (equivalent for intervention and sta- tionarity tests) after allowing for both first differencing and Panel (a) of Figure 1 displays the plot of IMR and the IMR forecast (dotted line) for 2007 to 2010 based on IMR’s for 12 lags of first-differenced dependent variables required 6 SAGE Open for diagnostic testing of the statistical assumption of declining trend during the intervention period 2007 to 2010. stationarity. The absolute value of the effect size measure is the square root In the preintervention period, the IMR trend is relatively of Cohen’s f for a regression variable (or set of variables), flat or very slightly negative with irregular variation and with 0.59, 0.39, and 0.14 considered large, medium, and small weak seasonal variation around the trend. Beginning in effects, respectively (Cohen, 1988). Panel (a) of Figure 2 January 2007 (see vertical line in the plot), IMR displays a graphically compares the preintervention trend slope for IMR shift to a more rapidly declining trend that continues through with that for 2007 to 2010 and Panel (b) displays the estimated the end of the sample period. During the intervention period, trend shift parameter (−0.02292) with 95% confidence interval actual IMR declines more rapidly than predicted by its prior = [−0.0128, −0.0330]. Although we have a clear a priori direc- trend and faster than the IMR forecast. tional hypothesis for the shift in trend, to be conservative, two- Panel (b) of Figure 1 shows the plot of the GROUP series. tailed tests are reported for the estimated trend shift (and all As noted above, the GROUP plot approximates a step func- other parameters). For reasons discussed below, the results tion, with an average of 591 participants for the 53-month reported in Table 1 are based on t ratios that remain valid in the baseline period compared with 1,792 for the 48-month inter- presence of heteroscedasticity and autocorrelation of the vention. In January 2007, the GROUP size for the first time regression residuals (Newey & West, 1987). The trend shift in the sample period rose above the theoretically predicted remains highly significant using conventional OLS standard critical threshold of 1,725, the √1% of the U.S. population. errors, t(87) = −3.32, p = .001, but the preintervention trend is not significantly different from zero, t(87) = −1.88, p = .064. Relative to the preintervention trend, the estimated trend Regression Results for IMR shift implies a total reduction of 1.100 in IMR during the To test the hypothesis of a decrease in the trend slope for 2007 through 2010 intervention period. This is a reduction IMR, we estimate the following broken-trend intervention of 12.47% (or 3.12% per year) compared with the mean pre- model that incorporates a shift of linear trend beginning with intervention monthly rate of 8.821 fatalities per 10,000 live the onset of the intervention (January 2007): births. The decline in IMR translates to a total of 992 averted infant deaths for 2007 to 2010, deaths projected to occur had IMR =+ ββ tD +− () ββ TS ++ D ε . tt 01 21 jt jt ∑ (1) the preintervention trend continued unchanged through 2007 j 8 to 2010. Thus, the estimated shift in trend for IMR during In Equation 1, β is the regression intercept, t is a linear time 2007 through 2010 has the predicted negative sign and is trend (t = 1, 2, 3, . . ., N), and β is the preintervention trend both statistically and practically significant. Statistical analy- slope for IMR. The variable DT models the shift in trend due sis (see Table 1) indicates that these results cannot be to the intervention with DT = (t − t )I , where t is the time explained by seasonal variation, autocorrelation, or preexist- t B t B of the hypothesized break in the linear trend function ing trends in IMR. (December 2006) and I is a binary (0-1) indicator variable (step function) that takes the value zero for the preinterven- Regression Diagnostics for Analysis of IMR tion period and 1.0 for the intervention period (t > t ). The regression coefficient (β − β ) for DT gives the change in Table 1 reports diagnostic tests for evaluating the adequacy 2 1 t trend slope for IMR from the preintervention value (β ) to the of the estimated intervention model. The null hypothesis of slope in the intervention period (β ). The hypothesis of a serially uncorrelated (white noise) regression residuals at negative shift in trend for IMR during the intervention lags 1 to 6, and 1 to 12 is rejected by the Breusch–Godfrey implies (β − β ) < 0. Lagrange Multiplier (LM) test (Godfrey, 1978). Only one 2 1 The summation term in Equation 1 is a deterministic sea- autocorrelation at lags 1 to 36 is statistically significant. The sonal component to control for the monthly seasonal varia- largest individual autocorrelations are at lag 1 (−0.208), tion in IMR. The seasonal component consists of 11 binary which is just significant at the 5% level, and lags 4 (−0.196) (0-1) seasonal dummy variables D (with monthly index j = and 16 (−0.224), which are nearly significant. Because of 1, 2, . . ., 11 where January is denoted by j = 1; Granger & this modest serial correlation, Table 1 reports t ratios based Newbold, 1986). The seasonal regression coefficient for on SEs that are valid (consistent) in the presence of serial each month is given by S . Finally, ε is an independent and correlation and heteroscedasticity of possibly unknown form jt t identically distributed, serially uncorrelated normal error (Newey & West, 1987). with mean zero and variance σ . Although some mild autocorrelation is present, the autocor- Table 1 reports the ordinary least squares (OLS) regression relations are all small, and thus the residual series appears to be results for Equation 1. The estimated preintervention trend for clearly stationary. A TS is said to be covariance stationary (or IMR is small but negative and significant. The shift in trend has weakly stationary) if its mean, variance, and autocorrelations negative sign, as hypothesized, and is highly significant, t(87) (or, equivalently, its autocovariances) are invariant with respect −5 = −4.50, p = 2.1 × 10 , effect size f = −0.482, a medium to to a change in time origin (Enders, 2010). Empirical support for large effect, indicating an acceleration of the preintervention stationarity of the IMR residual series is provided by a Dillbeck and Cavanaugh 7 Table 1. OLS Regression Analysis of Monthly U.S. IMR. a b Parameter Estimate SE t ratio β 9.360 0.235 39.87*** −3 −3 β −6.875 × 10 2.129 × 10 −3.23** −2 −3 β − β −2.292 × 10 5.091 × 10 −4.50*** 2 1 S −0.233 0.288 −0.81 1t S −0.185 0.219 −0.84 2t S −0.192 0.265 −0.73 3t S −0.245 0.303 −0.81 4t S 0.141 0.236 0.60 5t S −0.403 0.219 −1.84 6t S −0.297 0.261 −1.13 7t S −0.672 0.275 −2.44* 8t S −0.875 0.269 −3.25** 9t S −0.371 0.276 −1.34 10t S 0.178 0.307 0.58 11t F statistic: F(13, 87) = 11.65*** Mean of IMR = 8.386 SE of regression = 0.501 SE of IMR = 0.774 Sum of squared residuals = 21.864 Log-likelihood = −66.033 2 2 R = .635; Adjusted R = .581 BIC = 1.947; AIC = 1.585 Diagnostics Serial correlation test Heteroscedasticity test Lags 1-6: F(6, 81) = 3.370 (p = .005) F(15, 85) = 1.234 (p = .263) Lags 1-12: F(12, 75) = 1.948 (p = .042) Test for ARCH Normality test: χ (2) = 1.260 (p = .533) Lags 1-6: F(6, 89) = 6.803 (p = .570) Robinson test : t(61) = −8.177 (p < .001) Test of functional form Perron unit root test: τ = −6.612 (p < .01) F(2, 85) = 1.607 (p = .207) Note. Sample is August 2002 to December 2010, N = 101. OLS = ordinary least squares; IMR = infant mortality rate; BIC = Bayesian information criterion; AIC = Akaike information criterion; ARCH = autoregressive conditional heteroscedasticity. Newey–West SEs and t ratios (Newey & West, 1987). df = 87. Frequency domain test for nonstationarity (Robinson, 1995). *p < .05. **p < .01. ***p < .001. Figure 2. Comparison of mortality trends during the preintervention and intervention periods. Note. Panel (a) shows that the declining linear trend for IMR in the preintervention period (August 2002–December 2006) displays a shift to more rapidly declining trend during the intervention period (January 2007–December 2010). The IMR trend slope is the rate of change per month in monthly infant deaths within 1 week of birth per 10,000 live births. Panel (b) displays the statistically significant trend shift for IMR during 2007 through 2010 with 95% confidence interval. IMR = infant mortality rate. 8 SAGE Open frequency domain test based on the periodogram of regression Study 2: Results of Analysis of DFR residuals (Baum & Wiggins, 2000; Robinson, 1995). The Plot of DFR Robinson test evaluates stationarity based on an empirical esti- mate of the order of integration of a TS, I(d), where d is the The plot of DFR in Figure 3 displays monthly seasonal fluc- number of times the series must be differenced to induce sta- tuations and a rising preintervention trend followed by a tionarity. For example, in the case of a nonstationary random flattening of the trend slope in 2007 through 2010. During walk, the differencing parameter d = 1, and the series may be 2007 through 2010, the forecast of DFR (dotted line) based transformed to I(0), or stationarity, by first differencing. The on preintervention data continues DFR’s preintervention Robinson test rejects the null hypothesis (p < .001) that the IMR upward trend, rising substantially above observed DFR (see residual series is nonstationary (see Table 1), supporting the Note 18). conclusion that IMR is broken-trend stationary, exhibiting sta- tionary fluctuations around a broken trend. Regression Results for DFR Stationarity of the regression residuals is required for valid statistical inferences regarding the estimated regression We estimate the following broken-trend intervention model parameters in the intervention model (Banerjee, Dolado, to test the hypothesis of a shift to a reduced trend slope for Galbraith, & Hendry, 1993; Enders, 2010; Granger & DFR during the intervention period: Newbold, 1986). Broken-trend stationarity for IMR implies DFRD =+ ββ tD +− () ββ T ++ β FR tt 01 21 31 t− that the OLS regression estimates in Table 1 have standard distributions and thus the observed significant change in SD + ε . (2) jt jt trend is unlikely to be the result of “spurious regression” j (Banerjee et al., 1993; Granger & Newbold, 1986). Spurious where lag 1 of DFR is included (with regression coefficient regressions can result when time trends are fitted to nonsta- β ) to model its first-order autoregressive dynamics. All tionary variables that contain a random walk component other terms are defined as in Equation 1. (“stochastic trend”). A key signature of spurious regressions The OLS regression estimates of the model coefficients is highly autocorrelated, nonstationary regression residuals. for Equation 2 are reported in Table 2. The DFR preinterven- The latter violate the distributional assumptions underlying tion trend is positive and significant. The change in trend has statistical inference for TS regression. negative sign and is highly significant, consistent with theo- Table 1 reports a formal test for broken-trend stationarity retical prediction. of a TS with a known (exogenous) structural break (Perron, In Table 2, the regression estimate of β − β , the change 2 1 1989, 2006). The Perron unit root test rejects the null in trend, gives the initial intervention impact. The significant hypothesis (p < .01) that the IMR TS is a nonstationary ran- first-order autoregressive dynamics of DFR around the trend dom walk with drift (Perron, 1989; Zivot & Andrews, implies a gradual adjustment of the preintervention trend to 1992). The alternative hypothesis is that IMR is broken- its equilibrium, or long-run, value after onset of the interven- trend stationary, displaying stationary fluctuations around a tion. If (1 − β ) ≠ 0, the total change in trend is given by κ = linear trend with a known, one-time break in the trend func- (β − β ) / (1 − β ). A formal test rejects the hypothesis (1 − 2 1 3 tion in December 2006. β ) = 0 (p < .01) in favor of the alternative that (|β | < 1), 3 3 The results of all other diagnostic tests for model ade- indicating that the estimated model is dynamically stable and quacy are satisfactory. The null hypothesis of no autoregres- thus that the trend will converge to its equilibrium value κ in sive conditional heteroscedasticity (ARCH) for the regression the long run (Banerjee, Dolado, & Mestre, 1998; Doornik & residuals is not rejected by the LM test for ARCH (Engle, Hendry, 2013; Enders, 2010). For the DFR analysis κ = −10 1982). Likewise, White’s general test for heteroscedasticity −0.0578, t(86) = −7.12, p = 3.1 × 10 , effect size f = −0.449, (White, 1980) fails to reject the null hypothesis that the a medium to large effect, and 95% confidence interval = regression errors are homoscedastic or, if heteroscedasticity [−0.0417, −0.0739], where κ is the coefficient on DT in the is present, it is unrelated to the regressors. The null hypoth- long-run equation for DFR (Hendry, 1995). In this context, esis that the functional form of the regression model is cor- “long run” denotes the mathematically expected equilibrium rectly specified is not rejected by Ramsey’s (1969) Regression value and does not necessarily imply a long period of time Error Specification Test (RESET). The omnibus Doornik– (Hendry, 1995). The estimated cumulative lag weights for Hansen test for normality of the regression errors (Doornik Equation 2 indicate that 50.2%, 75.25%, 87.6%, and 93.8% & Hansen, 2008) fails to reject the null hypothesis that the of the adjustment in trend after the onset of the intervention errors are drawn from a normal distribution. No regression is completed within 1, 2, 3, and 4 months, respectively. Panel residuals exceed 3.5 standard errors, indicating the absence (a) of Figure 4 compares the equilibrium trend slopes for the of extreme outliers. Thus, the diagnostic tests for model ade- preintervention and 2007 through 2010 periods and Panel (b) quacy support statistical conclusion validity for the statistical shows the estimated trend shift with 95% confidence inferences reported in Table 1. interval. Dillbeck and Cavanaugh 9 Figure 3. Plot of monthly DFR. Note. The plot of the monthly rate of drug-related accidental fatalities for August 2002 to December 2010 displays monthly seasonal variation and a rising trend in the preintervention period followed by a flattening of the trend during the intervention period 2007 through 2010 (see vertical line in the plot). During 2007 through 2010, the forecast of DFR (dotted line) continues its preintervention upward trend, rising substantially above observed DFR. DFR = drug-related fatality rate. Relative to the preintervention trend, the equilibrium tests for adequacy of the broken-trend regression model sup- trend shift for 2007 to 2010 implies a total cumulative decline port statistical conclusion validity for the statistical infer- in DFR of 2.7723 monthly fatalities per million population ences reported in Table 2. during the intervention period. This is a reduction of 30.42% (or 7.61% annually) compared with the mean prein- Discussion tervention monthly rate of 9.112 fatalities per million people. The decline in DFR translates to a projected total of 26,425 The results of the two studies indicate that, as hypothesized, averted drug-related fatalities for 2007 to 2010. Thus, the there was a significant and substantial reduction of the trends estimated shift in trend for DFR is negative, as predicted, and of IMR and DFR during the intervention period 2007 through both practically and statistically significant. The statistical 2010, beginning in January 2007 with the onset of the inter- findings reported in Table 2 suggest that these results cannot vention. The null hypothesis of no effect of the intervention be explained by autocorrelation, seasonal variation, or preex- on mortality trends was strongly rejected for both mortality isting trends in DFR. rate series. Diagnostic tests for the broken-trend intervention analysis indicate the appropriateness of the statistical assumptions underlying the analyses. These results are thus Regression Diagnostics for DFR consistent with the conclusion that the intervention may have All diagnostic tests reported in Table 2 for adequacy of the contributed to the observed trend shifts for both mortality estimated intervention model are satisfactory. That no regres- rates. sion residuals exceed 3.5 standard errors indicates the The use of an intervention at the national level in a pro- absence of extreme outliers. The LM tests for autocorrelation spective quasi-experimental design is inherently stronger as of residuals at lags 1 to 6 and lags 1 to 12 are not significant, a design than an ex post analysis of archival data on the rela- and no autocorrelations at lags 1 to 36 are individually sig- tionship of interdependent factors using variants of multiple nificant, indicating that the regression residuals appear to be regression analysis (Glass, 1997). Nevertheless, it is neces- serially uncorrelated, stationary white noise. The frequency sary to consider possible threats to internal validity, namely, domain test (Baum & Wiggins, 2000; Robinson, 1995) alternative possible causes of the reduction of the stress- rejects the null hypothesis (p < .001) that the residual series related mortality variables investigated in these studies. is nonstationary (see Table 2). The latter test, plus the A major societal impact that occurred during the interven- absence of significant residual autocorrelation, lend further tion period was the economic recession of 2008-2009 associ- support to the conclusion of the Perron test (see Table 2) that ated with the subprime mortgage crisis. However, the onset DFR is likely broken-trend stationary. Thus, the diagnostic and duration of this recession (December 2007, with peak 10 SAGE Open Table 2. OLS Regression Analysis of Monthly U.S. DFR. a b Parameter Estimate SE t ratio β 3.484 0.606 5.76*** −2 −3 β 2.998 × 10 6.266 × 10 4.78*** −2 −3 β − β −2.897 × 10 6.968 × 10 −4.16*** 2 1 −2 β 0.498 9.560 × 10 5.21*** S 0.276 0.146 1.89 1t S −0.417 0.160 −2.61** 2t S 0.507 0.146 3.47*** 3t S −0.336 0.170 −1.97 4t S 0.305 0.150 2.03* 5t S −0.433 0.165 −2.63** 6t S 0.189 0.147 1.29 7t S −0.223 0.154 −1.45 8t S −0.529 0.145 −3.64*** 9t S −0.091 0.139 −0.65 10t S −0.422 0.141 −2.99** 11t F statistic: F(14, 86) = 102.30** Mean of DFR = 9.856 SE of regression = 0.296 SE of DFR = 1.152 Sum of squared residuals = 7.523 Log-likelihood = −12.159 2 2 R = .943; Adjusted R = .934 BIC = 0.926; AIC = 0.538 Diagnostics Serial correlation test Heteroscedasticity test Lags 1-6: F(6, 80) = 0.918 (p = .487) F(19, 81) = 1.359 (p = .172) Lags 1-12: F(12, 74) = 0.823 (p = .627) Test for ARCH Normality test: χ (2) = 0.951 (p = .622) Lags 1-6: F(6, 89) = 0.883 (p = .511) Robinson test : t(61) = −6.175 (p < .001) Test of functional form Perron unit root test: τ = −6.477 (p < .01) F(2, 84) = 1.990 (p = .143) Note. Sample is August 2002 to December 2010, N = 101. OLS = ordinary least squares; DFR = drug-related fatality rate; BIC = Bayesian information criterion; AIC = Akaike information criterion; ARCH = autoregressive conditional heteroscedasticity. OLS SEs and t ratios. df = 86. Frequency domain test for nonstationarity (Robinson, 1995). *p < .05. **p < .01. ***p < .001. unemployment in October 2009) does not precisely track the explanation will require almost simultaneous implementa- intervention period. More importantly, the economic stresses tion of a treatment or medical technology or program that associated with the recession (e.g., unemployment) would be would influence both the variables of these studies and dis- predicted, if anything, to increase rather than decrease the proportionately affect two different subpopulations of the stress-related health indicators examined in these studies. United States (higher prevalence for African American One specific factor to consider in the case of drug-related women in the case of IM and for middle-age Whites in the deaths is that one might speculate that increased public and case of DD). We are unable to think of any such plausible professional awareness of the hazards of opioid analgesics in alternative explanation that would satisfy these two require- recent years might have contributed to a reduction in their ments. In the absence of plausible alternative explanations, prescription during the intervention period. As outlined in the significant intervention impacts on both mortality vari- the introduction, increased used of opioid drugs is consid- ables found in this article suggest the possibility of some ered to contribute substantially to the rise of DD. However, underlying connection between individuals who are not data indicate that opioid sales increased throughout the base- apparently behaviorally connected. As outlined in the intro- line and intervention period of this research, reaching their duction to this article, such a connection independent of peak in 2011 (Volkow, 2014). Thus, the opioids were becom- behavioral interaction is described as the collective con- ing more available rather than less available through the sciousness of society. More specifically, the Vedic tradition intervention period. of India identifies collective consciousness as having its Because of the national scope of measurement of these basis in an underlying level of pure consciousness, which is studies, and the specificity of the intervention period and predicted to be influenced by the intervention examined associated changes in mortality rates, a viable alternative here, group practice of the advanced TM-Sidhi program. Dillbeck and Cavanaugh 11 Figure 4. Comparison of DFR trends during the preintervention and intervention periods. Note. Panel (a) shows that the positive linear trend for DFR in the preintervention period shifts to a flatter trend during the intervention period 2007 through 2010 (equilibrium trend slopes). The DFR slope is the rate of change per month of monthly drug-related fatalities per 1 million population. Panel (b) displays the statistically significant negative trend shift for DFR during the intervention with 95% confidence interval. DFR = drug-related fatality rate. Acknowledgments An alternative hypothesis affecting generalizability of the results is whether there is selection bias in the use of The authors appreciate the cooperation of the Invincible America members in the intervention group who were already Assembly Office of Maharishi University of Management for pro- experts in the Transcendental Meditation and TM-Sidhi viding group meditation data used in this study. The second author acknowledges access to the use of a university laptop and software program and may therefore be a special group in terms of for this research. consciousness training. From a theoretical perspective, it is only necessary that the members of the intervention group be trained in the advanced TM-Sidhi program to stimulate Declaration of Conflicting Interests the hypothesized field of pure consciousness. It is under- The authors declared no potential conflicts of interest with stood that, as a practical necessity, the individuals who par- respect to the research, authorship, and/or publication of this ticipated in this project would be those with some prior article: The authors have no financial relationship with the expertise/training to volunteer their time. But, in principle, foundation that teaches Transcendental Meditation in the for this effect to be created, any group with basic stability, United States (Maharishi Foundation USA). The first author such as groups of students or the military, could be is a Trustee and research professor and the second author is employed once trained in the TM-Sidhi program. In some a retired faculty member of the university sponsoring the countries, such groups are being created and future research project evaluated in this research, but both currently receive will be necessary to confirm their effects. no financial compensation from the university for their The result of these studies, if further replicated, provide research or other activities, which are donated as contrib- government leaders with a potential means to relieve stress uted services. on a large scale in society that is independent of structural social change and benefits public health. Although social Funding stress as a mediating variable was not directly measured, The authors received no financial support for the research and/or the consistency of the present results with those reported authorship of this article. previously for reduced homicide and violent crime (Dillbeck & Cavanaugh, 2016) lends support to a hypothe- Notes sis of such a mediating influence. In light of these results, it 1. Transcendental Meditation and TM-Sidhi are service marks is suggested that when TM-Sidhi groups are established registered in the U.S. Patent and Trademark Office, used by with government support (most easily, with existing groups, Maharishi Foundation, USA, under sublicense. such as military personnel), the potential impact of such 2. Crime measures were daily Indian Penal Code crimes for New TM-Sidhi groups be evaluated on stress-related public Delhi, weekly Philippine Crime Index totals for Metro Manila, health parameters. and monthly FBI Part 1 crimes for Puerto Rico. 3. From late July to late October 2006, a second group was formed Authors’ Note in Washington, D.C., to add to the effect of daily group prac- tice of the TM-Sidhi program. Because the two groups were Raw data may be obtained from the first author. 12 SAGE Open separate and distant, the predicted effect of these two groups 11. The Robinson periodogram test evaluates the null hypothesis 2 2 () nn + was added independently, that is, square root of , that the differencing parameter d takes a nonstationary value ver- 1 2 where n refers to the Iowa group and n to the Washington sus the alternative of stationarity. In the theoretical framework 1 2 group.The Washington group was much smaller than the Iowa underlying this test, both integer and fractional values of d are group, so the effect was inconsequential for the total number permitted, allowing for the possibility of fractional integration, for those 3 months, although it was included for the sake of including possible “long memory” behavior for values of d > 0. completeness. The Robinson test rejects the null hypothesis that the estimated 4. Scholarship funding to Maharishi University of Management difference parameter falls in the nonstationary region d ≥ 0.5 in and its related organizations was provided by the Howard and favor of the alternative d < 0.5 (see Table 1), thus supporting Alice Settle Foundation. stationarity of the IMR residual series. The sample estimate of 5. Data for year 2010 are compiled from the Multiple Cause of d = −0.2695 indicates stationary “short memory” or “antiper- Death File 2010, Series 20 No. 2P, 2012. Data for year 2009 sistence,” presumably reflecting the predominance of negative are compiled from the Multiple Cause of Death File 2009, autocorrelations among the largest residual autocorrelations. Series 20 No. 2O, 2012. Data for year 2008 are compiled from 12. As discussed in Dillbeck and Cavanaugh (2016), standard the Multiple Cause of Death File 2008, Series 20 No. 2N, approaches to testing for nonstationarity (unit root tests) such 2011. Data for year 2007 are compiled from Multiple Cause of as the augmented Dickey–Fuller test (Dickey & Fuller, 1979; Death File 2007, Series 20 No. 2M, 2010. Data for years 2005 Said & Dickey, 1984) and others have been shown to be biased to 2006 are compiled from Multiple Cause of Death File 2005- in favor of concluding that a broken-trend stationary time 2006, Series 20, No. 2L, 2009. Data for years 1999 to 2004 are series (TS) is nonstationary. compiled from the Multiple Cause of Death File 1999-2004, 13. Following Zivot and Andrews (1992), we test for nonsta- Series 20, No. 2J, 2007. tionarity using the “change in growth” broken-trend model 6. The effect size f is given by the t ratio (or square root of the based on Equation 15 in Perron (1989, p. 1381), which is F statistic) for the regression coefficient divided by the square Model B in Zivot and Andrews (1992, p. 253). The test root of the residual degrees of freedom. The unsquared effect statistic has a nonstandard distribution tabulated by Perron size metric is reported, rather than squared, because unsquared (1989, Table V.B). The critical value for the test depends measures better indicate the relative magnitude of effects across on the parameter λ= tN /, the time of the break divided variables (Darlington, 1990). The guidelines for large, medium, by total sample size. For the two variables analyzed in this and small values of f mentioned above are the square root of article, λ = 0.51. those given by Cohen for f (0.35, 0.15, and 0.02, respectively). 14. Because of possible monthly seasonality displayed by IMR, a 7. The estimated change in trend slope of −0.0229146 (before maximum of 12 lags of first-differenced IMR were considered rounding) gives the estimated reduction in the mathematically for inclusion in the regression used to calculate the Perron test expected value of the monthly infant mortality rate (IMR) for statistic. Because superfluous regressors inflate SEs for the each month of the intervention period relative to the preinter- estimated regression coefficients (Greene, 2011), lags that vention trend. Multiplying by 48 months gives −1.100 fatali- were not significant at the 20% level (lags 6-10 and 12) were ties per 10,000 live births, the expected value of the cumulative deleted from the Perron test regression, thus substantially total change in IMR for 2007 to 2010. increasing the power of the Perron test. 8. The preintervention data were used to generate monthly fore- 15. Omitting the seasonal coefficients to save space, the estimated casts of IMR for 2007 to 2010 based on ordinary least squares long-run equation is given by DFR = 6.945 + 0.0598t − 0.0578 (OLS) estimates of Equation 1 after deleting the trend-change DT , where the SEs for the regression coefficients are 0.3154, variable DT from the model. The cumulative total number of 0.00433, and 0.00811, respectively. fatalities averted for 2007 to 2010 (N ) was calculated by 16. The SE for the estimate of κ reported by PcGive 14 software IM () ββ − summing the difference between forecast and actual IM for is calculated from the SEs of β and in Equation each month: 2 by numerical differentiation using a nonlinear algorithm (Bårdsen, 1989; Doornik & Hendry, 2013). As shown by Inder   () forecast IMR − actual IMR × kk (1993), t tests for κ based on critical values from the standard N = ,, k = 12,,  48,,   IM b /, 10 000   normal distribution have both good power and actual p values  k  close to their nominal level. where N is the estimate of total prevented IM and b is the 17. The long-run, or equilibrium, change in slope −0.0577561 IM k number of U.S. live births for month k. The summation is over (before rounding) gives the estimated reduction in the math- kt =− () t k, the month of the intervention, with for tt > , ematically expected value of drug-related fatalities per mil- B B where t is the month of the trend break. lion population (DFR) for each month of the intervention 9. Following the recommendation of Kiviet (1986) and Harvey period relative to the preintervention trend. Multiplying (1990) for smaller samples, the Breusch–Godfrey and all other by 48 months gives the expected value of the cumulative Lagrange Multiplier (LM) tests reported in Table 2 are given change for 2007 to 2010: −2.7723 monthly fatalities per mil- in their F statistic form rather than chi-square. lion population. 10. The bandwidth of four Newey and West (1987) lags was 18. Using preintervention data, the OLS estimate of Equation 2 (with selected automatically by PcGive 14 using the integer part of trend shift variable DT omitted) was used to generate monthly 2/9 4(N / 100) . Experimentation with a wide range of alternative ex ante (dynamic) forecasts of DFR for 2007 to 2010 (Doornik & bandwidths indicated that the conclusions of the trend shift Hendry, 2013). The cumulative total number of averted drug-related analysis are not sensitive to bandwidth. deaths during 2007 through 2010 (N ) was estimated as follows: DD Dillbeck and Cavanaugh 13 Centers for Disease Control and Prevention, National Center for   () forecast DFRa − ctual DFR kk Nk = ,, = 12,,  48,   DD Health Statistics. (2012). Underlying cause of death, 1999- USPOP     2010 on CDC WONDER online database, released 2012. where summation is over month k of the intervention and Retrieved from http://wonder.cdc.gov/ucd-icd10.html USPOP is the monthly U.S. population estimate (in millions). k Centers for Disease Control and Prevention, National Center for 19. The Robinson test rejects the null hypothesis that the differ- Health Statistics. (2015a). Underlying cause of death 1999- encing parameter d falls in the nonstationary region d ≥ 0.5 2013 on CDC WONDER online database, released 2015. in favor of the stationary alternative d < 0.5. The sample esti- Retrieved from http://wonder.cdc.gov/ucd-icd10.html mate of the differencing parameter is d = −0.0232. The null Centers for Disease Control and Prevention, National Center for hypothesis of stationarity, d = 0, also was not rejected, t(61) = Health Statistics. (2015b). Natality public use data 2003-2006, −0.2995, p = .766. All frequency domain tests were estimated 2007-2013 on CDC WONDER online database, released 2012. in Stata 14 software employing add-on module roblpr (Baum Retrieved from http://wonder.cdc.gov/natality-v2006.html and & Wiggins, 2000; StataCorp, 2015). The number of periodo- http://wonder.cdc.gov/natality-current.html gram ordinates used in the tests for DFR and IMR was the Chelimsky, E., Shadish, W. R., & Orwin, R. G. (1997). Twenty-one 0.9 default value m = N (m = 63), where N = 101 is the effective years and counting: The interrupted time series comes of age. sample size. The test results were not sensitive to alternative In E. Chelimsky & R. Shadish (Eds.), Evaluation for the 21st values of m. century (pp. 443-465). London, England: Sage. 20. As in the case of the Perron test for IMR, lags 1 to 12 of first- Cohen, J. (1988). Statistical power analysis for the behavioral sci- differenced DFR that were not significant at the 20% level ences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum. (lags 1 and 4–10) were deleted from the regression equation Cook, T. D., & Campbell, D. T. (1979). Quasi-experimentation: for the Perron test, reducing SEs for the remaining regressors Design and analysis issues for field settings. Boston, MA: and increasing the power of the test. Houghton Mifflin. Darlington, R. B. (1990). Regression and linear models. New York, NY: McGraw-Hill. References Davies, J. L., & Alexander, C. N. (2005). Alleviating political vio- Alexander, C. N., Davies, J. L., Dixon, C., Dillbeck, M. C., Druker, lence through reducing collective tension: Impact assessment S., Oetzel, R. M., . . . Orme-Johnson, D. W. (1990). Growth analyses of the Lebanon war. Journal of Social Behavior and of higher states of consciousness: Maharishi’s Vedic psychol- Personality, 17, 285-338. ogy of human development. In C. N. Alexander & E. J. Langer Dickey, D., & Fuller, W. A. (1979). Distribution of the estimates (Eds.), Higher stages of human development: Perspectives on for autoregressive time series with a unit root. Journal of the adult growth (pp. 286-340). New York, NY: Oxford University American Statistical Association, 74, 427-431. Press. Dillbeck, M. C. (1990). Test of a field hypothesis of conscious- Assimakis, P. D., & Dillbeck, M. C. (1995). Time series analy- ness and social change: Time series analysis of participation sis of improved quality of life in Canada: Social change, in the TM-Sidhi program and reduction of violent death in the collective consciousness, and the TM-Sidhi program. U.S. Social Indicators Research, 22, 399-418. doi:10.1007/ Psychological Reports, 76, 1171-1193. doi:10.2466/ BF00303834 pr0.1995.76.3c.1171 Dillbeck, M. C., & Alexander, C. N. (1989). Higher states of con- Banerjee, A., Dolado, J. J., Galbraith, J. W., & Hendry, D. F. (1993). sciousness: Maharishi Mahesh Yogi’s Vedic psychology of Co-integration, error correction and the econometric analysis human development. The Journal of Mind and Behavior, 10, of non-stationary data. Oxford, UK: Oxford University Press. 307-334. Banerjee, A., Dolado, J. J., & Mestre, R. (1998). Error-correction Dillbeck, M. C., Banus, C. B., Polanzi, C., & Landrith, G. S., III. mechanism tests for cointegration in a single equation frame- (1988). Test of a field model of consciousness and social work. Journal of Time Series Analysis, 19, 267-283. change: Transcendental Meditation and TM-Sidhi program and Bårdsen, G. (1989). The estimation of long-run coefficients decreased urban crime. The Journal of Mind and Behavior, 9, from error-correction models. Oxford Bulletin of Economics 457-486. and Statistics, 51, 345-350. doi:10.1111/j.1468-0084.1989. Dillbeck, M. C., & Bronson, E. C. (1981). Short-term longitudinal mp51003008.x effects of the Transcendental Meditation technique on EEG Baum, C. F., & Wiggins, V. (2000). Tests for long memory in a power and coherence. International Journal of Neuroscience, time series. Stata Technical Bulletin, 57, 39-44. 14, 147-151. doi:10.3109/00207458108985827 Box, G. E. P., & Jenkins, G. M. (1976). Time series analysis: Dillbeck, M. C., & Cavanaugh, K. L. (2016). Societal violence Forecasting and control. San Francisco, CA: Holden-Day. and collective consciousness: Reduction of U.S. homi- Case, A., & Deaton, A. (2015). Rising morbidity and mortality in cide and urban violent crime rates. SAGE Open, 6(2), 1-16. midlife among white non-Hispanic Americans in the 21st cen- doi:10.1177/2158244016637891 tury. Proceedings of the National Academy of Sciences of the Dillbeck, M. C., Cavanaugh, K. L., Glenn, T., Orme-Johnson, D. W., United States of America, 112(49), 15078-15083. doi:10.1073/ & Mittlefehldt, V. (1987). Effects of Transcendental Meditation pnas.1518393112 and the TM-Sidhi program on quality of life indicators: Cavanaugh, K. L., & Dillbeck, M. C. (2017). The contribution of Consciousness as a field. The Journal of Mind and Behavior, proposed field effects of consciousness to the prevention of 8, 67-104. US accidental fatalities: Theory and empirical tests. Journal of Dillbeck, M. C., & Orme-Johnson, D. W. (1987). Physiological Consciousness Studies, 24, 53-86. differences between Transcendental Meditation and rest. 14 SAGE Open American Psychologist, 42, 879-881. doi:10.1037/0003- Hendry, D. F. (1995). Dynamic econometrics. Oxford, UK: Oxford 066X.42.9.879 University Press. Dillbeck, M. C., & Rainforth, M. V. (1996). Impact assessment Inder, B. (1993). Estimating long-run relationships in economics: A analysis of behavioral quality of life indices: Effects of group comparison of different approaches. Journal of Econometrics, practice of the Transcendental Meditation and TM-Sidhi pro- 57, 53-68. doi:10.1016/0304-4076(93)90058-D gram. Proceedings of the American Statistical Association, Jevning, R., Wilson, A. F., & Davidson, J. M. (1978). Social Statistics Section, 38-43. Adrenocortical activity during meditation. Hormones and Dole, N., Savitz, D. A., Hertz-Picciotto, I., Siega-Riz, A. M., Behavior, 10, 54-60. McMahon, M. J., & Buekens, P. (2003). Maternal stress and Jones, C. M., Mack, K. A., & Paulozzi, L. J. (2013). Pharmaceutical preterm birth. American Journal of Epidemiology, 157, 14-24. overdose deaths, United States, 2010. Journal of the American doi:10.1093/aje/kwf176 Medical Association, 309, 657-659. Doornik, J. A., & Hansen, H. (2008). An omnibus test for univariate Kiviet, J. F. (1986). On the rigour of some misspecification tests for and multivariate normality. Oxford Bulletin of Economics and modelling dynamic relationships. Review of Economic Studies, Statistics, 70, 927-939. doi:10.1111/j.1468-0084.2008.00537.x 53, 241-261. doi:10.2307/2297649 Doornik, J. A., & Hendry, D. F. (2013). Empirical economet- Latendresse, G. (2009). The interaction between chronic stress ric modelling–PcGive 14: Volume 1. London, England: and pregnancy: Preterm birth from a biobehavioral perspec- Timberlake Consultants Press. tive. Journal of Midwifery & Women’s Health, 54, 8-17. Enders, W. (2010). Applied econometric time series (3rd ed.). doi:10.1016./j.jmwh.2008.08.001 Hoboken, NJ: John Wiley. Linsky, A. S., Bachman, R., & Straus, M. A. (1995). Stress, culture, Engle, R. F. (1982). Autoregressive conditional heteroscedasticity & aggression. New Haven, CT: Yale University Press. with estimates of the variance of United Kingdom inflation. Linsky, A. S., & Straus, M. A. (1986). Social stress in the United Econometrica, 50, 987-1007. States. Dover, MA: Auburn House Publishing. Gennaro, S., & Hennessy, M. D. (2003). Psychological and MacDorman, M. F., & Mathews, T. J. (2008). Recent trends in physiological stress: Impact on preterm birth. Journal of infant mortality in the United States (NCHS Data Brief, No. 9). Obstetric, Gynecologic & Neonatal Nursing, 32, 668-675. Hyattsville, MD: National Center for Health Statistics. doi:10.1177/0884217503257484 MacDorman, M. F., & Mathews, T. J. (2011). Understanding racial Glass, G. V. (1997). Interrupted time series quasi-experiments. In and ethnic disparities in U.S. infant mortality rates (NCHS R. M. Jaeger (Ed.), Complementary methods for research in Data Brief, No. 74). Hyattsville, MD: National Center for education (2nd ed., pp. 589-608). Washington, DC: American Health Statistics. Educational Research Association. MacDorman, M. F., Mathews, T. J., Mohangoo, A. D., & Zeitlin, Godfrey, L. G. (1978). Testing for higher order serial correla- J. (2014, September 4). International comparisons of infant tion in regression equations when the regressors include mortality and related factors: United States and Europe, lagged dependent variables. Econometrica, 46, 1303-1313. 2010 (National Vital Statistics Reports, Vol. 63, Number 5). doi:10.2307/1913830 Hyattsville, MD: National Center for Health Statistics. Granger, C. W. J., & Newbold, P. (1986). Forecasting economic MacLean, C. R. K., Walton, K. G., Wenneberg, S. R., Levitsky, D. time series (2nd ed.). Orlando, FL: Academic Press. K., Mandarino, J. P., Waziri, R., . . . Schneider, R. H. (1997). Greene, W. H. (2011). Econometric analysis (7th ed.). Upper Effects of the Transcendental Meditation program on adap- Saddle River, NJ: Prentice Hall. tive mechanisms: Changes in hormone levels and responses to Hagelin, J. S. (1987). Is consciousness the unified field? A field stress after 4 months of practice. Psychoneuroendocrinology, theorist’s perspective. Modern Science and Vedic Science, 1, 22, 277-295. 28-87. Maharishi Mahesh Yogi. (1969). Maharishi Mahesh Yogi on the Hagelin, J. S., Rainforth, M. V., Orme-Johnson, D. W., Cavanaugh, Bhagavad-Gita: A new translation and commentary, chapters K. L., Alexander, C. N., Shatkin, S. F., . . . Ross, E. (1999). 1 to 6. Baltimore, MD: Penguin. Effects of group practice of the Transcendental Meditation Maharishi Mahesh Yogi. (1977). Creating an ideal society. program on preventing violent crime in Washington, DC: Rheinweiler, Germany: Maharishi European Research Results of the National Demonstration Project, June- University Press. July 1993. Social Indicators Research, 47, 153-201. Maharishi Mahesh Yogi. (1986). Life supported by natural law. doi:10.1023/A:1006978911496 Washington, DC: Age of Enlightenment Press. Harmon, K. (2010, April 6). Prescription drug deaths increase dra- Mathews, T. J., MacDorman, M. F., & Thoma, M. E. (2015, August matically. Scientific American. Retrieved from http://www.sci- 6). Infant mortality statistics from the 2013 period linked birth/ entificamerican.com/article/prescription-drug-deaths/ infant death data set (National Vital Statistics Reports, Vol. Harvey, A. (1990). The econometric analysis of time series (2nd 64, Number 9). Hyattsville, MD: National Center for Health ed.). Cambridge, MA: MIT Press. Statistics. Hatchard, G. D., Deans, A. J., Cavanaugh, K. L., & Orme-Johnson, Newey, W. K., & West, K. (1987). A simple positive semi-definite, D. W. (1996). The Maharishi effect: A model for social heteroscedasticity and autocorrelation consistent covariance improvement. Time series analysis of a phase transition to matrix. Econometrica, 55, 703-708. reduced crime in Merseyside metropolitan area. Psychology, Orme-Johnson, D. W., Alexander, C. N., & Davies, J. L. (1990). The Crime & Law, 2, 165-174. doi:10.1080/10683169608409775 effects of the Maharishi Technology of the Unified Field: Reply Dillbeck and Cavanaugh 15 to a methodological critique. Journal of Conflict Resolution, 34, StataCorp. (2015). Stata Statistical Software: Release 14. College 756-768. doi:10.1177/0022002790034004009 Station, TX: Author. Orme-Johnson, D. W., Alexander, C. N., Davies, J. L., Chandler, Travis, F., Haaga, D. A. F., Hagelin, J., Tanner, M., Arenander, A., H. M., & Larimore, W. E. (1988). International peace project Nidich, S., . . . Schneider, R. H. (2010). A self-referential default in the Middle East: The effects of the Maharishi Technology of brain state: Patterns of coherence, power, and eLORETA sources the Unified Field. Journal of Conflict Resolution, 32, 776-812. during eye-closed rest and Transcendental Meditation practice. doi:10.1177/0022002788032004009 Cognitive Processing, 11, 21-30. doi:10.1007/s10339-009-0343-2 Orme-Johnson, D. W., & Oates, R. M. (2009). A field-theoretic Travis, F., Haaga, D. A. F., Hagelin, J., Tanner, M., Nidich, S., Gaylord- view of consciousness: Reply to critics. Journal of Scientific King, C., . . . Schneider, R. H. (2009). Effects of Transcendental Exploration, 23, 139-166. Meditation practice on brain functioning and stress reactivity in Paulozzi, L. J., Budnitz, D. S., & Xi, Y. (2006). Increasing deaths from college students. International Journal of Psychophysiology, 71, opioid analgesics in the United States. Pharmacoepidemiology 170-176. doi:10.1016/j.ijpsycho.2008.09.007 & Drug Safety, 15, 618-625. doi:10.1002/pds.1276 Volkow, N. D. (2014, May 4). Testimony to Congress—America’s Perron, P. (1989). The great crash, the oil price shock, and the unit addiction to opioids: Heroin and prescription drug abuse. root hypothesis. Econometrica, 57, 1361-1401. doi:10.2307/ National Institute on Drug Abuse of the National Institutes of 1913712 Health. Retrieved from http://www.drugabuse.gov/about-nida/ Perron, P. (2006). Dealing with structural breaks. In T. C. Mills legislative-activities/testimony-to-congress/2015/americas- & K. Patterson (Eds.), Palgrave handbook of econometrics, addiction-to-opioids-heroin-prescription-drug-abuse Vol. 1: Econometric theory (pp. 278-352). Basingstoke, UK: Walton, K. G., Fields, J. Z., Levitsky, D. K., Harris, D. A., Pugh, Palgrave Macmillan. N. D., & Schneider, R. H. (2004). Lowering cortisol and Radhakrishnan, S. (1953). Principal Upanishads. New York, NY: CVD risk in postmenopausal women: A pilot study using the Harper. Transcendental Meditation program. Annals of the New York Ramsey, J. B. (1969). Tests for specification errors in classical Academy of Sciences, 1032, 211-215. linear least squares regression analysis. Journal of the Royal White, H. (1980). A heteroskedastic-consistent covariance matrix Statistical Society: Series B—Methodological, 31, 350-371. estimator and a direct test for heteroskedasticity. Econometrica, Rappoport, P., & Reichlin, L. (1989). Segmented trends and non- 48, 817-838. doi:10.2307/1912934 stationary time series. Economic Journal, 99, 168-177. Zivot, E., & Andrews, K. (1992). Further evidence on the great Robinson, P. M. (1995). Log-periodogram regression of time series crash, the oil price shock, and the unit root hypothesis. Journal with long range dependence. Annals of Statistics, 23, 1048-1072. of Business & Economic Statistics, 10, 251-270. doi:10.1080/0 Roth, R. (2002). Maharishi Mahesh Yogi’s Transcendental 7350015.1992.10509904 Meditation. Washington, DC: Primus. Said, S. E., & Dickey, D. (1984). Testing for unit roots in autoregres- Author Biographies sive moving-average models with unknown order. Biometrika, Michael C. Dillbeck, PhD, is a research scientist at the Institute of 71, 599-607. doi:10.1093/biomet/71.3.599 Science, Technology and Public Policy and research professor and Schneider, R. H., & Carr, T. (2014). Transcendental Meditation in the Trustee at Maharishi University of Management. His research spans prevention and treatment of cardiovascular disease and pathophysi- the physiological, psychological, and sociological effects on the ological mechanisms: An evidence-based review. Advances in Transcendental Meditation and TM-Sidhi programs. Integrative Medicine, 1, 107-112. doi:10.1016/j.aimed.2014.08.003 Kenneth L. Cavanaugh, PhD, is a senior research scientist at the Shadish, W. R., Cook, T. D., & Campbell, D. T. (2002). Institute of Science, Technology and Public Policy and a professor Experimental and quasi-experimental designs for generalized of Applied Statistics, Emeritus at Maharishi University of causal inference. Boston, MA: Houghton Mifflin. Management. His currrent principal research interest is empirical Sinha, R. (2008). Chronic stress, drug use, and vulnerability to study of the hypothesized effects of the group practice of the addiction. Annals of the New York Academy of Sciences, 1141, TM-Sidhi program on societal quality of life. 105-130. doi:10.1196/annals.141.030

Journal

SAGE OpenSAGE

Published: Mar 14, 2017

Keywords: behavioral sciences; alcohol; drugs; tobacco; sociology of health and illness; sociology; social sciences; collective behavior/social movements; social change and modernization; medical sociology

There are no references for this article.