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

Learn More →

The effect of a primary care-based Medical Weight Loss Program on weight loss and anthropomorphic metrics

The effect of a primary care-based Medical Weight Loss Program on weight loss and anthropomorphic... Abstract Background Diet and lifestyle intervention programs have been shown to be effective in decreasing obesity/overweight and many associated comorbidities in specialty research settings. There is very little information however as to the efficacy of such programs conducted in usual/typical primary care practices. We analysed effectiveness of the Medical Weight Loss Program (MWLP) designed to specifically address overweight/obesity in the setting of an urban academic primary care practice. Objective To determine whether participation in the MWLP within a general primary care setting can result in weight loss. Methods A retrospective medical chart review of patients treated in MWLP and a control group of patients with obesity receiving regular care in the general primary care setting. From the practice database (1 April 2015–31 March 2016), 209 patients (≥18 years old) who participated in the MWLP were identified; 265 controls were selected from the remaining population based on the presence of the obesity-related diagnoses. Results MWLP patients lost on average 2.35 ± 5.88 kg in 6 months compared to their baseline weight (P < 0.0001). In contrast, the control group demonstrated a trend of gaining on average 0.37 ± 6.03 kg. Having three or more visits with the MWLP provider within 6 months after program initiation was the most important factor associated with successful loss of at least 5% of the baseline weight. Weight loss also correlated with a decrease in abdominal girth. Conclusion MWLP integrated into the general primary care practice may potentially be an effective model for managing obesity and related morbidities. Lifestyle modification/health behaviour change, Medical Weight Loss Program, obesity, obesity management, prevention, primary care Key Messages Overweight/obesity and comorbid conditions are a major health problem. It is feasible to integrate obesity management into a primary care practice. Weight loss can be achieved in a primary care-based Medical Weight Loss Program. Introduction Obesity rates have continued to rise over the past decades, with a prevalence of obesity in 2015–16 approaching 40% of the US adult population (1). Obesity is one of the biggest drivers of preventable chronic diseases and health care costs in the USA (2). Furthermore, indirect costs of obesity due to lost productivity and absenteeism run into the billions of dollars (2–4). With little evidence of imminent remission, it has become increasingly important to formulate effective means of screening, assessing and treating obesity. Primary care providers must be included in these approaches as they see the vast majority of patients who present for ambulatory care (5). A vast proportion of these patients suffer from obesity and related conditions (6,7). While surgical options have been shown to provide good evidence of disease remittance (8–10), many patients with obesity do not qualify for, nor desire surgery. These patients must therefore be managed medically. The US Preventive Task Force recommends screening all adults for obesity (11) and found adequate evidence that intensive, multicomponent behavioural interventions can lead to weight loss, improved glucose tolerance and reduction in risk factors for cardiovascular disease (11). For patients with body mass index (BMI) >30 kg/m2, clinicians are encouraged to offer or refer patients for intensive, multicomponent behavioural interventions (11). There is overwhelming evidence that diet and lifestyle intervention programs can be effective in decreasing obesity and many associated comorbidities in specialty research settings (12–14). The Diabetes Prevention Program and Look Ahead studies were two landmark randomized clinical trials that provided strong evidence of this approach (12–14). However, these large and highly resource-intensive multi-centre studies were conducted over the course of several years, according to rigorous research protocols with dedicated research personnel. Subsequent research directed at investigating the translatability of elements of these programs into real-world clinical settings showed varying degrees of success (15–17). In this study, we assessed the effectiveness of the Medical Weight Loss Program (MWLP) based on lifestyle changes in a large urban primary care practice that serves a predominantly socioeconomically, disadvantaged minority population. Methods Study design We conducted a retrospective medical chart review comparing patients treated in MWLP to the control group of patients with obesity who received usual care in the same primary care practice. Description of MWLP Setting The program was offered at the clinical offices of the University Family Medicine (UFM) on the campus of the University of Maryland Medical Systems (UMMS). The UFM is a large urban, academic safety-net primary care practice that serves a primarily minority population with about 50 000 patient visits per year. The practice is located in Baltimore City where an adult obesity rate of 36.6% surpasses the state average (18). Clinical intervention MWLP was under the direction of the physician certified by the American Board of Obesity Medicine and a physician assistant certified in diabetes education. Both professionals had extensive training in nutrition, especially as it pertains to overweight and obesity. Interactions with the Obesity Medicine Specialist were based on the principle of motivational interviewing, which has been shown to facilitate behaviour change (19–21). The intake appointment included a detailed review of medical history and information from the comprehensive questionnaire provided to patients prior to scheduling a visit. The questionnaire encompassed dietary and lifestyle issues known to impact weight based on recommendations from the Obesity Medicine Association. This information was used in developing individualized treatment plans, including diet and exercise counselling, support for self-management and pharmacotherapy for select patients. The plan also integrated information about socioeconomic status and domestic situation. At the initial visit, standard vital signs were obtained. Anthropomorphic measurements were obtained using a standard tape measure. Valhalla Scientific Body Composition Scale was used to collect data on weight, percent body fat and lean tissue, resting energy expenditure and target weight. This scale uses electrical impedance to differentiate between different types of body tissue and is commonly used in clinical practice (22,23). For all patients, metabolic data were ordered if not available from the past 3–6 months (Comprehensive Metabolic Panel, Lipids, Glycohemoglobin, Thyroid Stimulating Hormone, CBC, micronutrients levels). Medications were also reviewed, and weight-positive drugs were replaced with alternate weight-neutral or weight-negative drugs where possible. A calorie deficit of ~500–1000 calories per day was recommended (24), which sometimes included meal replacement products. Patients were asked to keep a food journal. If indicated, patients were referred to other specialists such as Sleep Medicine, Cardiology, Physical Therapy, Psychology/Behavioural Medicine or Bariatric Surgery. Patients were also prescribed an exercise program customized to their physical ability. Based on the evidence that the group process can be effective in promoting weight loss, particularly in some minority populations (25,26), weekly group meetings attendance was encouraged but not mandated. Follow-up visits included weight measurement, monitoring of body composition, diet and exercise review, and adjustment of eating plan. Since the adjunctive use of anti-obesity medications (AOMs) has proved effective in reducing weight compared to placebo (27–29), suitable candidates for drug therapy were prescribed AOMs as appropriate. Prescribing guidelines were strictly followed. Study population MWLP group Patients accessed the program either by referral from specialty or primary care providers or self-referral. Informational material about the program was available within the UMMS network as well as through community activities such as health fairs. All patients with overweight and obesity who requested medical weight management services were accepted except patients who had undergone bariatric surgery within 6 months of program enrolment. Records for adult UFM patients (≥18 years old) with invoices between 1 April 2015 and 31 March 2016 were retrieved using IBM Cognos Analytics software (Armonk, NY). MWLP patients were identified based on their record of participation in the program. Medical charts for these patients were reviewed to determine the date of the program enrolment (time frame July 2012–August 2016). Patients who had at least two weight measurements for at least 6 months were included. All available measurements within a 12-month time frame were used regardless of whether they were obtained as a part of MWLP or unrelated clinical visits. Pregnant women and patients with cancer were excluded. Data for co-morbidities (diabetes mellitus type 2, insulin resistance, hypertension, dyslipidemia, sleep apnoea, history of gastric bypass surgery, malabsorption, depression, anxiety, bipolar disorder) were identified by ICD10 codes. Demographic data (age, sex, race, ethnicity), anthropomorphic and vital measurements, number of visits with the MWLP provider and calorie restriction recommendations were obtained from medical charts. Control group Control patients with obesity were identified from the remaining records based on the presence of the obesity-related ICD10 (the International Classification of Diseases, 10th Revision, Clinical Modification) codes (overweight, obese with or without complications, unusual weight gain). Patients were selected using the same inclusion/exclusion criteria as the MWLP group. Additional inclusion criterion for the control group was the same period for observations as for MWLP patients (time frame July 2012–August 2016). Outcomes The major outcome measure was relative weight change at 6 months in the MWLP group compared to the control patients. Secondary outcomes for MWLP patients were 5% weight loss was achieved at any point in 12 months and the correlation between 6-month changes in weight and anthropomorphic measurements. Statistical analysis Data were analysed using SAS statistical package version 9.3 (SAS Institute, Cary, NC). For MWLP patients, the baseline was the measurement at the first program visit. For the control group, the baseline was the first recorded vital measurement in the chart within the study time frame. Weight change at 6 months was estimated by linear, quadratic or higher order polynomial regression analysis using weight measurements flanking a 6-months’ time point (R2 > 0.95). For the continuous variables, differences between the treatment and control groups were assessed using a two-tailed Student’s t-test. Associations between categorical variables were examined using chi-square or Fisher’s exact tests. Correlations between changes in weight and anthropomorphic metrics (waist circumference) were assessed using the Pearson correlation coefficient. The effects of MWLP intervention, demographic characteristics and comorbidities on either 6-month weight loss or anthropomorphic metrics were assessed using multivariable linear regression models. For the loss of 5% of baseline weight, time to achieve target weight loss was estimated by regression analysis using dates flanking the target date. Time-to-event analysis was performed on imputed time values using Kaplan–Meier’s survival curves and Cox’s proportional hazard modelling with observation time censored at 12 months after acceptance to the program. Patients who had not achieved a 5% weight loss in 12 months were censored at their maximum observation time. Covariates included in the multivariable regression models were identified based on preliminary bivariate analyses looking for potential effect modifiers and confounders. Statistical significance was established at two-sided α = 0.05. Results Participants Overall, 8392 records from 3291 individuals with active charts between 1 April 2015 and 31 March 2016 were retrieved, and 236 MWLP patients and 391 controls were identified. Data for 209 MWLP patients and 265 controls who had at least 2 weight measurements for at least 6 months were used for the analysis (Supplementary Figure 1). Control patients excluded from the analysis were younger compared to those who were included (37.5 ± 12.1 and 41.3 ± 14.0 years respectively, P = 0.0067) and had lower prevalence of diabetes (10% and 24% respectively, P = 0.0128) and sleep apnoea (1.5% and 6% respectively, P = 0.0431). MWLP patients excluded from the analyses did not differ significantly in demographic and clinical characteristics from those who were included except that they had lower AOM prescription rate (26% and 52% respectively, P = 0.0103). Characteristics of the study population are given in Table 1. Compared to the controls, the MWLP group had higher initial weight and BMI, had more women, Black patients, and patients with diabetes, sleep apnoea, depression and bipolar disorder (Table 1). The median number of weight measurements per patient within the first 6 months was 4 (range 2–23), with median time between observations 0.8 months (25th–75th percentile 0.4–1.3 months). Among MWLP patients, 52% were prescribed AOM. The number of visits with the MWLP provider within 6 months after acceptance to the program varied from 1 to 10 (median 3 visits). For those who had two or more visits, the median time between visits was 1.1 months. A calorie and carb-restricted diet with 1300–1400 calories per day was recommended to 51% of patients; 9% of patients were placed on a more restricted diet (1100–1200 cal/day). Table 1. Baseline characteristics of 474 patients seen in the UFM practice between July 2012 and August 2016 Characteristics . . MWLP, N = 209 . Control, N = 265 . P* . Age at baseline, years Mean ± SD 42.3 ± 11.2 41.3±14.0 0.3590T Median 40 40 Q1–Q3 34–50 30–52 Range 18–77 18–83 Weight at baseline, kg Mean ± SD 121.0 ± 28.3 110.2±27.6 <0.0001 T Median 116.3 104.9 Q1–Q3 101.0–135.6 88.5–126.1 Range 65.2–221.8 65.3–231.8 BMI at baseline, kg/m2 Mean ± SD 44.4 ± 9.7 40.1±9.6 <0.0001 T Median 42.9 37.9 Q1–Q3 37.3–49.4 33.5 – 45.4 Range 26.7–86.6 25.3–93.4 Age categories at baseline, N (%) 18–34 years 59 (28%) 93 (35%) 0.0742 35–54 years 111 (53%) 113 (43%) 55+ years 39 (19%) 58 (22%) Sex, N (%) Female 189 (90%) 212 (80%) 0.0018 Male 20 (10%) 53 (20%) Race, N (%) Black 179 (86%) 202 (76%) <0.0001 White 12 (6%) 52 (20%) Other 18 (8%) 11 (4%) BMI category at baseline, N (%) 25–29.9 kg/m2 4 (2%) 27 (10%) <0.0001 30–39.9 kg/m2 64 (31%) 129 (49%) 40–49.9 kg/m2 93 (44%) 72 (27%) 50+ kg/m2 48 (23%) 37 (14%) Obesity, N (%) Overweight 4 (2%) 15 (6%) <0.0001 Obese 95 (46%) 136 (51%) Morbidly obese 71 (34%) 101 (38%) Abnormal weight gain 5 (2%) 13 (5%) No related ICD10 code 34 (16%) 0 (0%) Diabetes, N (%) No related ICD10 code 111 (53%) 181 (71%) 0.0003 Prediabetes 42 (20%) 20 (8%) DM II without complications 39 (19%) 44 (14%) DM II with complications 17 (8%) 20 (7%) Insulin resistance (Yes), N (%) 10 (5%) 5 (2%) 0.0735 Dyslipidemia (Yes), N (%) 23 (11%) 22 (8%) 0.3189 Sleep apnoea (Yes), N (%) 20 (10%) 4 (2%) <0.0001 Gastric bypass surgery (Yes), N (%) 6 (3%) 1 (0.4%) 0.0475 F Malabsorption (Yes), N (%) 7 (3%) 0 (0%) 0.0031 F Hypertension (Yes), N (%) 93 (45%) 102 (38%) 0.1870 Depression (Yes), N (%) 32 (15%) 16 (6%) 0.0009 Bipolar disorder (Yes), N (%) 7 (3%) 1 (0.4%) 0.0243 F Anxiety (Yes), N (%) 6 (3%) 8 (3%) 1.0000 Anti-obesity medication (Yes), N (%) 109 (52%) 3 (0.8%) <0.0001 Characteristics . . MWLP, N = 209 . Control, N = 265 . P* . Age at baseline, years Mean ± SD 42.3 ± 11.2 41.3±14.0 0.3590T Median 40 40 Q1–Q3 34–50 30–52 Range 18–77 18–83 Weight at baseline, kg Mean ± SD 121.0 ± 28.3 110.2±27.6 <0.0001 T Median 116.3 104.9 Q1–Q3 101.0–135.6 88.5–126.1 Range 65.2–221.8 65.3–231.8 BMI at baseline, kg/m2 Mean ± SD 44.4 ± 9.7 40.1±9.6 <0.0001 T Median 42.9 37.9 Q1–Q3 37.3–49.4 33.5 – 45.4 Range 26.7–86.6 25.3–93.4 Age categories at baseline, N (%) 18–34 years 59 (28%) 93 (35%) 0.0742 35–54 years 111 (53%) 113 (43%) 55+ years 39 (19%) 58 (22%) Sex, N (%) Female 189 (90%) 212 (80%) 0.0018 Male 20 (10%) 53 (20%) Race, N (%) Black 179 (86%) 202 (76%) <0.0001 White 12 (6%) 52 (20%) Other 18 (8%) 11 (4%) BMI category at baseline, N (%) 25–29.9 kg/m2 4 (2%) 27 (10%) <0.0001 30–39.9 kg/m2 64 (31%) 129 (49%) 40–49.9 kg/m2 93 (44%) 72 (27%) 50+ kg/m2 48 (23%) 37 (14%) Obesity, N (%) Overweight 4 (2%) 15 (6%) <0.0001 Obese 95 (46%) 136 (51%) Morbidly obese 71 (34%) 101 (38%) Abnormal weight gain 5 (2%) 13 (5%) No related ICD10 code 34 (16%) 0 (0%) Diabetes, N (%) No related ICD10 code 111 (53%) 181 (71%) 0.0003 Prediabetes 42 (20%) 20 (8%) DM II without complications 39 (19%) 44 (14%) DM II with complications 17 (8%) 20 (7%) Insulin resistance (Yes), N (%) 10 (5%) 5 (2%) 0.0735 Dyslipidemia (Yes), N (%) 23 (11%) 22 (8%) 0.3189 Sleep apnoea (Yes), N (%) 20 (10%) 4 (2%) <0.0001 Gastric bypass surgery (Yes), N (%) 6 (3%) 1 (0.4%) 0.0475 F Malabsorption (Yes), N (%) 7 (3%) 0 (0%) 0.0031 F Hypertension (Yes), N (%) 93 (45%) 102 (38%) 0.1870 Depression (Yes), N (%) 32 (15%) 16 (6%) 0.0009 Bipolar disorder (Yes), N (%) 7 (3%) 1 (0.4%) 0.0243 F Anxiety (Yes), N (%) 6 (3%) 8 (3%) 1.0000 Anti-obesity medication (Yes), N (%) 109 (52%) 3 (0.8%) <0.0001 Data for 209 MWLP and 265 control patients with a baseline observation time between July 2012 and August 2016 are shown. DM, diabetes mellitus type II; ICD10, The International Classification of Diseases, 10th Revision; UFM, University of Maryland Family Medicine. P-values <0.05 are shown in bold. Q1–Q3: 25th and 75th percentiles respectively. *Chi-square test, unless indicated otherwise: T: Student’s t-test (two-sided), F: Fisher’s exact test (two-sided). Open in new tab Table 1. Baseline characteristics of 474 patients seen in the UFM practice between July 2012 and August 2016 Characteristics . . MWLP, N = 209 . Control, N = 265 . P* . Age at baseline, years Mean ± SD 42.3 ± 11.2 41.3±14.0 0.3590T Median 40 40 Q1–Q3 34–50 30–52 Range 18–77 18–83 Weight at baseline, kg Mean ± SD 121.0 ± 28.3 110.2±27.6 <0.0001 T Median 116.3 104.9 Q1–Q3 101.0–135.6 88.5–126.1 Range 65.2–221.8 65.3–231.8 BMI at baseline, kg/m2 Mean ± SD 44.4 ± 9.7 40.1±9.6 <0.0001 T Median 42.9 37.9 Q1–Q3 37.3–49.4 33.5 – 45.4 Range 26.7–86.6 25.3–93.4 Age categories at baseline, N (%) 18–34 years 59 (28%) 93 (35%) 0.0742 35–54 years 111 (53%) 113 (43%) 55+ years 39 (19%) 58 (22%) Sex, N (%) Female 189 (90%) 212 (80%) 0.0018 Male 20 (10%) 53 (20%) Race, N (%) Black 179 (86%) 202 (76%) <0.0001 White 12 (6%) 52 (20%) Other 18 (8%) 11 (4%) BMI category at baseline, N (%) 25–29.9 kg/m2 4 (2%) 27 (10%) <0.0001 30–39.9 kg/m2 64 (31%) 129 (49%) 40–49.9 kg/m2 93 (44%) 72 (27%) 50+ kg/m2 48 (23%) 37 (14%) Obesity, N (%) Overweight 4 (2%) 15 (6%) <0.0001 Obese 95 (46%) 136 (51%) Morbidly obese 71 (34%) 101 (38%) Abnormal weight gain 5 (2%) 13 (5%) No related ICD10 code 34 (16%) 0 (0%) Diabetes, N (%) No related ICD10 code 111 (53%) 181 (71%) 0.0003 Prediabetes 42 (20%) 20 (8%) DM II without complications 39 (19%) 44 (14%) DM II with complications 17 (8%) 20 (7%) Insulin resistance (Yes), N (%) 10 (5%) 5 (2%) 0.0735 Dyslipidemia (Yes), N (%) 23 (11%) 22 (8%) 0.3189 Sleep apnoea (Yes), N (%) 20 (10%) 4 (2%) <0.0001 Gastric bypass surgery (Yes), N (%) 6 (3%) 1 (0.4%) 0.0475 F Malabsorption (Yes), N (%) 7 (3%) 0 (0%) 0.0031 F Hypertension (Yes), N (%) 93 (45%) 102 (38%) 0.1870 Depression (Yes), N (%) 32 (15%) 16 (6%) 0.0009 Bipolar disorder (Yes), N (%) 7 (3%) 1 (0.4%) 0.0243 F Anxiety (Yes), N (%) 6 (3%) 8 (3%) 1.0000 Anti-obesity medication (Yes), N (%) 109 (52%) 3 (0.8%) <0.0001 Characteristics . . MWLP, N = 209 . Control, N = 265 . P* . Age at baseline, years Mean ± SD 42.3 ± 11.2 41.3±14.0 0.3590T Median 40 40 Q1–Q3 34–50 30–52 Range 18–77 18–83 Weight at baseline, kg Mean ± SD 121.0 ± 28.3 110.2±27.6 <0.0001 T Median 116.3 104.9 Q1–Q3 101.0–135.6 88.5–126.1 Range 65.2–221.8 65.3–231.8 BMI at baseline, kg/m2 Mean ± SD 44.4 ± 9.7 40.1±9.6 <0.0001 T Median 42.9 37.9 Q1–Q3 37.3–49.4 33.5 – 45.4 Range 26.7–86.6 25.3–93.4 Age categories at baseline, N (%) 18–34 years 59 (28%) 93 (35%) 0.0742 35–54 years 111 (53%) 113 (43%) 55+ years 39 (19%) 58 (22%) Sex, N (%) Female 189 (90%) 212 (80%) 0.0018 Male 20 (10%) 53 (20%) Race, N (%) Black 179 (86%) 202 (76%) <0.0001 White 12 (6%) 52 (20%) Other 18 (8%) 11 (4%) BMI category at baseline, N (%) 25–29.9 kg/m2 4 (2%) 27 (10%) <0.0001 30–39.9 kg/m2 64 (31%) 129 (49%) 40–49.9 kg/m2 93 (44%) 72 (27%) 50+ kg/m2 48 (23%) 37 (14%) Obesity, N (%) Overweight 4 (2%) 15 (6%) <0.0001 Obese 95 (46%) 136 (51%) Morbidly obese 71 (34%) 101 (38%) Abnormal weight gain 5 (2%) 13 (5%) No related ICD10 code 34 (16%) 0 (0%) Diabetes, N (%) No related ICD10 code 111 (53%) 181 (71%) 0.0003 Prediabetes 42 (20%) 20 (8%) DM II without complications 39 (19%) 44 (14%) DM II with complications 17 (8%) 20 (7%) Insulin resistance (Yes), N (%) 10 (5%) 5 (2%) 0.0735 Dyslipidemia (Yes), N (%) 23 (11%) 22 (8%) 0.3189 Sleep apnoea (Yes), N (%) 20 (10%) 4 (2%) <0.0001 Gastric bypass surgery (Yes), N (%) 6 (3%) 1 (0.4%) 0.0475 F Malabsorption (Yes), N (%) 7 (3%) 0 (0%) 0.0031 F Hypertension (Yes), N (%) 93 (45%) 102 (38%) 0.1870 Depression (Yes), N (%) 32 (15%) 16 (6%) 0.0009 Bipolar disorder (Yes), N (%) 7 (3%) 1 (0.4%) 0.0243 F Anxiety (Yes), N (%) 6 (3%) 8 (3%) 1.0000 Anti-obesity medication (Yes), N (%) 109 (52%) 3 (0.8%) <0.0001 Data for 209 MWLP and 265 control patients with a baseline observation time between July 2012 and August 2016 are shown. DM, diabetes mellitus type II; ICD10, The International Classification of Diseases, 10th Revision; UFM, University of Maryland Family Medicine. P-values <0.05 are shown in bold. Q1–Q3: 25th and 75th percentiles respectively. *Chi-square test, unless indicated otherwise: T: Student’s t-test (two-sided), F: Fisher’s exact test (two-sided). Open in new tab Weight change at 6 months To characterize the impact of MWLP on weight loss, we examined whether weight change over the period of 6 months was different between groups of comparison. MWLP patients lost on average 2.35 ± 5.88 kg in 6 months (P < 0.0001) compared to their baseline weight. In contrast, the control group demonstrated a trend of gaining on average 0.37 ± 6.03 kg, although the change did not differ significantly from zero (P = 0.3140). In the bivariate analysis, a relative difference in the 6-month weight change between MWLP and control groups was 2.73 ± 5.96 kg (P < 0.0001) (Fig. 1). Figure 1. Open in new tabDownload slide Comparison of the 6-month weight change for 474 patients seen in the UFM practice. Data for 209 MWLP and 265 control patients with a baseline observation time between July 2012 and August 2016 are shown. Data are presented as box plots. Horizontal lines within the boxes are medians, diamonds mark mean, box frames mark 25th and 75th percentiles, whiskers represent 5th and 95th percentiles, dots are outliers. P-value on the graph is for the two-sided t-test. MWLP, Medical Weight Loss Program; UFM, University of Maryland Family Medicine. Figure 1. Open in new tabDownload slide Comparison of the 6-month weight change for 474 patients seen in the UFM practice. Data for 209 MWLP and 265 control patients with a baseline observation time between July 2012 and August 2016 are shown. Data are presented as box plots. Horizontal lines within the boxes are medians, diamonds mark mean, box frames mark 25th and 75th percentiles, whiskers represent 5th and 95th percentiles, dots are outliers. P-value on the graph is for the two-sided t-test. MWLP, Medical Weight Loss Program; UFM, University of Maryland Family Medicine. A multivariable linear regression model showed that, among demographic and clinical characteristics (listed in Table 1), only the intervention (MWLP) and BMI at baseline were significantly associated with a 6-month weight change. After adjusting for the baseline BMI, the difference in 6-month weight loss between MWLP and control groups was on average 2.23 kg (95% CI 1.12–3.34 kg; P < 0.0001). A significant weight loss was observed in MWLP patients with baseline BMI 30 kg/m2 or greater while in the control group a significant weight gain was observed in the group with BMI 25–29.9 kg/m2 (Table 2). Table 2. Results of the multivariable linear regression model estimating magnitude of weight change in 474 patients seen in the UFM practice between July 2012 and August 2016 BMI categories, kg/m2 . Intervention groups . Absolute change, kg . . . N . Mean . 95% CI . P . 25–29.9 Control 27 +2.5 +0.5 to +4.6 0.0168 MWLP 4 +0.3 −2.0 to +2.6 0.7930 30–39.9 Control 129 +0.7 −0.2 to +1.6 0.1127 MWLP 64 −1.5 −2.6 to −0.4 0.0086 40–49.9 Control 72 −0.5 −1.6 to +0.6 0.3345 MWLP 93 −2.8 −3.8 to −1.7 <0.0001 50+ Control 37 −0.7 −2.1 to +0.7 0.3355 MWLP 48 −2.9 −4.3 to −1.6 <0.0001 BMI categories, kg/m2 . Intervention groups . Absolute change, kg . . . N . Mean . 95% CI . P . 25–29.9 Control 27 +2.5 +0.5 to +4.6 0.0168 MWLP 4 +0.3 −2.0 to +2.6 0.7930 30–39.9 Control 129 +0.7 −0.2 to +1.6 0.1127 MWLP 64 −1.5 −2.6 to −0.4 0.0086 40–49.9 Control 72 −0.5 −1.6 to +0.6 0.3345 MWLP 93 −2.8 −3.8 to −1.7 <0.0001 50+ Control 37 −0.7 −2.1 to +0.7 0.3355 MWLP 48 −2.9 −4.3 to −1.6 <0.0001 Data for 209 MWLP and 265 control patients with a baseline observation time between July 2012 and August 2016 are shown. P-values are for the two-sided t-test; P-values <0.05 are shown in bold. MWLP, Medical Weight Loss Program; BMI, body mass index; CI, confidence interval; UFM, University of Maryland Family Medicine. Open in new tab Table 2. Results of the multivariable linear regression model estimating magnitude of weight change in 474 patients seen in the UFM practice between July 2012 and August 2016 BMI categories, kg/m2 . Intervention groups . Absolute change, kg . . . N . Mean . 95% CI . P . 25–29.9 Control 27 +2.5 +0.5 to +4.6 0.0168 MWLP 4 +0.3 −2.0 to +2.6 0.7930 30–39.9 Control 129 +0.7 −0.2 to +1.6 0.1127 MWLP 64 −1.5 −2.6 to −0.4 0.0086 40–49.9 Control 72 −0.5 −1.6 to +0.6 0.3345 MWLP 93 −2.8 −3.8 to −1.7 <0.0001 50+ Control 37 −0.7 −2.1 to +0.7 0.3355 MWLP 48 −2.9 −4.3 to −1.6 <0.0001 BMI categories, kg/m2 . Intervention groups . Absolute change, kg . . . N . Mean . 95% CI . P . 25–29.9 Control 27 +2.5 +0.5 to +4.6 0.0168 MWLP 4 +0.3 −2.0 to +2.6 0.7930 30–39.9 Control 129 +0.7 −0.2 to +1.6 0.1127 MWLP 64 −1.5 −2.6 to −0.4 0.0086 40–49.9 Control 72 −0.5 −1.6 to +0.6 0.3345 MWLP 93 −2.8 −3.8 to −1.7 <0.0001 50+ Control 37 −0.7 −2.1 to +0.7 0.3355 MWLP 48 −2.9 −4.3 to −1.6 <0.0001 Data for 209 MWLP and 265 control patients with a baseline observation time between July 2012 and August 2016 are shown. P-values are for the two-sided t-test; P-values <0.05 are shown in bold. MWLP, Medical Weight Loss Program; BMI, body mass index; CI, confidence interval; UFM, University of Maryland Family Medicine. Open in new tab Factors associated with a 5% loss of the baseline weight in MWLP patients Kaplan–Meier’s time-to-event analysis showed that 82 patients (39%) achieved a 5% loss of their baseline weight during 12 months of observation. The number of visits with the MWLP provider within 6 months after initiation of the program was the most important factor associated with successful weight loss (Fig. 2, log-rank test P < 0.0001). For patients who had three or more visits, the median time for losing 5% weight was 5.9 months (95% CI 3.9–7.1). Only 24% of patients who had less than three visits achieved 5% weight loss (25th quantile 11.1 months, 95% CI 6.4–12.6). Age and BMI at baseline were also significantly associated with the rate of weight loss (data not shown). Figure 2. Open in new tabDownload slide Association between weight loss and the number of visits in patients participated in the UFM MWLP. Data for 209 MWLP patients with a baseline observation time between July 2012 and August 2016 are shown. Time to achieve 5% weight loss within 12 months after program acceptance was estimated by regression analysis. Kaplan–Meier’s curves are shown for patients who had three or more visits within 6 months after program acceptance (solid line) compared to those who had one or two visits (dash line). Grey-shaded areas define upper and lower 95% confidence intervals. P-value on the graph is for a log-rank test. MWLP, Medical Weight Loss Program; UFM, University of Maryland Family Medicine. Figure 2. Open in new tabDownload slide Association between weight loss and the number of visits in patients participated in the UFM MWLP. Data for 209 MWLP patients with a baseline observation time between July 2012 and August 2016 are shown. Time to achieve 5% weight loss within 12 months after program acceptance was estimated by regression analysis. Kaplan–Meier’s curves are shown for patients who had three or more visits within 6 months after program acceptance (solid line) compared to those who had one or two visits (dash line). Grey-shaded areas define upper and lower 95% confidence intervals. P-value on the graph is for a log-rank test. MWLP, Medical Weight Loss Program; UFM, University of Maryland Family Medicine. Multivariable Cox’s proportional hazard modelling showed that, controlling for baseline age and BMI, the rate of weight loss for patients who had three or more visits was 4.4 times higher compared to patients who had less than three visits [Hazard Ratio (HR) = 4.4; 95% CI 2.8–7.0; P < 0.0001]. Controlling for the baseline age and number of MWLP visits, it was twice harder to lose 5% weight for patients with BMI 40–49.9 kg/m2 compared to patients with BMI <40 kg/m2 (HR = 0.5; 95% CI 0.3–0.8; P = 0.0093). There was no statistically significant difference in the rate of weight loss between patients with BMI 40–49.9 kg/m2 and BMI ≥50 kg/m2 (data not shown). It was easier for younger patients to lose weight, with a chance (hazard) of losing 5% weight decreasing by 2.5% per 1-year increase in baseline age (HR = 0.975; 95% CI 0.955–0.995; P = 0.0071). The effects of AOM prescription or initial calorie restriction recommendations were not significantly associated with a chance of losing 5% weight (data not shown). Correlation between anthropomorphic measurements and weight change at 6 months We also determined whether 6-month weight changes in MWLP patients were accompanied by the corresponding changes in waist circumference. In the univariate analysis, there was a significant decrease in the waist (−4.4 ± 8.9 cm, P < 0.0001) circumferences at 6 months that correlated with weight changes [Pearson correlation coefficient (ρ) = 0.47, P < 0.0001, Fig. 3]. In the multivariable linear regression model, a 6-month weight change, baseline weight and baseline waist circumference were significantly associated with a 6-month change in waist circumference (P < 0.0001 for all factors, data not shown). Adjusted for the baseline factors, waist circumference decreased on average by 0.64 cm per 1 kg of weight loss (95% CI 0.46–0.82 cm, P < 0.0001). Figure 3. Open in new tabDownload slide Association between changes in weight and waist circumference in patients participated in the UFM MWLP. Among 209 MWLP patients with baseline observation time between July 2012 and August 2016, 107 patients who had at least two anthropomorphic measurements during at least 6 months were identified. After removing two outliers, data for 105 MWLP patients were analysed using a linear regression model. Thick line represents a regression trend; grey-shaded area defines upper and lower 95% confidence intervals, and dash lines represent upper and lower prediction intervals. Numbers on the graph for a Pearson’s correlation coefficient (ρ) with a corresponding P-value. MWLP, Medical Weight Loss Program; UFM, University of Maryland Family Medicine. Figure 3. Open in new tabDownload slide Association between changes in weight and waist circumference in patients participated in the UFM MWLP. Among 209 MWLP patients with baseline observation time between July 2012 and August 2016, 107 patients who had at least two anthropomorphic measurements during at least 6 months were identified. After removing two outliers, data for 105 MWLP patients were analysed using a linear regression model. Thick line represents a regression trend; grey-shaded area defines upper and lower 95% confidence intervals, and dash lines represent upper and lower prediction intervals. Numbers on the graph for a Pearson’s correlation coefficient (ρ) with a corresponding P-value. MWLP, Medical Weight Loss Program; UFM, University of Maryland Family Medicine. Discussion Our study represents the first attempt of describing the MWLP established within a primary care practice and evaluating its effectiveness. While there were several studies that examined weight loss programs (12–14), our study is unique because it was conducted in a typical primary care practice that serves primarily low socioeconomic status patients. While MWLP did not have all the elements of an ideal multi-disciplinary weight loss program, successful weight loss was achieved by engaging patients with trained dedicated personnel who incorporated the existing resources of a general primary care practice into MWLP. On average, MWLP patients with obesity lost a significant amount of weight during the first 6 months. Moreover, a pronounced weight gain was observed among overweight (BMI 25–30 kg/m2) control group. This trend is concerning and consistent with the ever-increasing rates of obesity. This may be indicative that more efforts should be made to engage the overweight population in weight loss programs, especially in primary care. Patients attending the program were mostly urban, African American females with severe obesity and comorbid conditions. Our results demonstrated that weight loss is attainable for these patients despite the challenges inherent in this demographic group. Patients in the MWLP group also had a significantly higher baseline BMI and more comorbid conditions compared to the control group. These conditions increase the difficulty of achieving a negative energy balance and thus of losing weight. Despite this, MWLP proved effective for these patients to lose weight. This would suggest that persons with obesity and related co-morbidities should be targeted for primary care-based lifestyle interventions for weight loss. Moreover, our data show that it was easier for younger MWLP patients as well as those with lower BMI (<40 kg/m2) to lose weight. With the previously noted tendency of patients to gain weight over time, targeting younger patients who are overweight or have a BMI of less than 40 kg/m2 could be a worthwhile strategy. Our study identified no effect of AOM on the likelihood of weight loss. However, the data were limited as there was only information on the prescription of the medications, not on actual usage. The number of visits with the MWLP provider was the most significant factor associated with the successful loss of 5% of initial weight. This might be due to the higher personal motivation of the patients who continued with the program as well as the effect of continued reinforcement of diet and lifestyle changes, support and accountability provided by the physician. This could be explored in future studies. For patients showing weight loss over a 6 months period, there was a correlating significant decrease in abdominal girth. It is well established that BMI and waist circumference are two factors that correlate with cardiovascular risk and can even be useful in risk stratification in certain patients (30–32). Thus, MWLP may play an important role in decreasing obesity-related cardiovascular risks in this population. Our study had several limitations. This was a retrospective review of the clinical records. Not all patients were observed exactly at 6 months, and analyses were conducted on imputed values. There was no defined baseline time for the control group. The results may be also limited in generalizability since our patient population consisted mainly of urban African American women, many of whom are negatively impacted by social determinants of health and represent a segment of the population in which obesity rates continue to rise (1). However, the study is important as it provides information on the effectiveness of obesity treatment in a real-world primary care practice where a vast majority of patients suffering from obesity and comorbid conditions are seen (6,7). More information is needed about the treatment of obesity in primary care settings, which rarely have the time, personnel and financial resources compared to the research setting in clinical trials. Our study aimed to determine whether obesity management can be effective in primary care and to provide data on a population that is often underrepresented in scientific studies. The results indicate that our model may be an effective approach that is worth developing. Conclusions Weight loss is achievable in a primary care setting. To address the obesity crisis, ways must be found to engage primary care providers and practices in translating known research into workable programs. Declaration Funding: The study was funded by departmental resources. Ethical approval: The exempt status of the study was confirmed by the University of Maryland Baltimore Institutional Review Board (approved 19 December 2016; reference number HR00073199). Conflict of interest: none. References 1. Hales CM , Carroll MD, Fryar CD, Ogden CL. Prevalence of Obesity Among Adults and Youth: United States, 2015–2016 . NCHS data brief, no 288. Hyattsville, MD : National Center for Health Statistics , 2017 . 2. Waters H , Graf M. America’s Obesity Crisis: The Health and Economic Costs of Excess Weight | Milken Institute . https://milkeninstitute.org/reports/americas-obesity-crisis-health-and-economic-costs-excess-weight (accessed on 18 April 2020 ). Published 2018. 3. Andreyeva T , Luedicke J, Wang YC. State-level estimates of obesity-attributable costs of absenteeism . J Occup Environ Med 2014 ; 56 : 1120 – 7 . Google Scholar Crossref Search ADS PubMed WorldCat 4. Hammond RA , Levine R. The economic impact of obesity in the United States . Diabetes Metab Syndr Obes 2010 ; 3 : 285 – 95 . Google Scholar Crossref Search ADS PubMed WorldCat 5. Rui P , Okeyode T. National Ambulatory Medical Care Survey: 2016 National Summary Tables . https://www.cdc.gov/nchs/data/ahcd/namcs_summary/2016_namcs_web_tables.pdf (accessed on 12 April 2020 ). 6. Ornstein SM , Jenkins RG, Litvin CB, Wessell AM, Nietert PJ. Preventive services delivery in patients with chronic illnesses: parallel opportunities rather than competing obligations . Ann Fam Med 2013 ; 11 : 344 – 9 . Google Scholar Crossref Search ADS PubMed WorldCat 7. Petterson S , McNellis R, Klink K, Meyers D, Bazemore A. The State of Primary Care in the United States: A Chartbook of Facts and Statistics . Robert Graham Center , January 2018. https://www.graham-center.org/content/dam/rgc/documents/publications-reports/reports/PrimaryCareChartbook.pdf (accessed on 12 April 2020 ). Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 8. Brethauer SA , Aminian A, Romero-Talamás Het al. . Can diabetes be surgically cured? Long-term metabolic effects of bariatric surgery in obese patients with type 2 diabetes mellitus . Ann Surg 2013 ; 258 : 628 – 36 ; discussion 636–7. Google Scholar Crossref Search ADS PubMed WorldCat 9. Schauer PR , Bhatt DL, Kashyap SR. Bariatric surgery or intensive medical therapy for diabetes after 5 years . N Engl J Med 2017 ; 376 : 1997 . Google Scholar Crossref Search ADS PubMed WorldCat 10. Colquitt JL , Pickett K, Loveman E, Frampton GK. Surgery for weight loss in adults . Cochrane Database Syst Rev 2014 ; ( 8 ): CD003641 . doi:10.1002/14651858.CD003641.pub4 Google Scholar OpenURL Placeholder Text WorldCat 11. Moyer VA , U.S. Preventive Services Task Force. Screening for and management of obesity in adults: U.S. Preventive Services Task Force recommendation statement . Ann Intern Med 2012 ; 157 ( 5 ): 373 – 8 , doi:10.7326/0003-4819-157-5-201209040-00475. Google Scholar PubMed OpenURL Placeholder Text WorldCat 12. Diabetes Prevention Program Research Group , Knowler WC, Fowler SEet al. . 10-year follow-up of diabetes incidence and weight loss in the Diabetes Prevention Program Outcomes Study . Lancet Lond Engl 2009 ; 374 ( 9702 ): 1677 – 86 . Google Scholar Crossref Search ADS WorldCat 13. Unick JL , Beavers D, Jakicic JMet al. .; Look AHEAD Research Group. Effectiveness of lifestyle interventions for individuals with severe obesity and type 2 diabetes: results from the Look AHEAD trial . Diabetes Care 2011 ; 34 : 2152 – 7 . Google Scholar Crossref Search ADS PubMed WorldCat 14. Look AHEAD Research Group . Eight-year weight losses with an intensive lifestyle intervention: the look AHEAD study . Obesity (Silver Spring) 2014 ; 22 ( 1 ): 5 – 13 . Crossref Search ADS PubMed WorldCat 15. Wadden TA , Volger S, Sarwer DBet al. . A two-year randomized trial of obesity treatment in primary care practice . N Engl J Med 2011 ; 365 : 1969 – 79 . Google Scholar Crossref Search ADS PubMed WorldCat 16. Carvajal R , Wadden TA, Tsai AG, Peck K, Moran CH. Managing obesity in primary care practice: a narrative review . Ann N Y Acad Sci 2013 ; 1281 : 191 – 206 . Google Scholar Crossref Search ADS PubMed WorldCat 17. Appel LJ , Clark JM, Yeh HCet al. . Comparative effectiveness of weight-loss interventions in clinical practice . N Engl J Med 2011 ; 365 : 1959 – 68 . Google Scholar Crossref Search ADS PubMed WorldCat 18. Welcome to MD-IBIS—Maryland’s Public Health Data Resource . https://ibis.health.maryland.gov/ (accessed on 18 April 2020 ). 19. Martins RK , McNeil DW. Review of motivational interviewing in promoting health behaviors . Clin Psychol Rev 2009 ; 29 : 283 – 93 . Google Scholar Crossref Search ADS PubMed WorldCat 20. Handbook of Motivational Counseling: Concepts, Approaches, and Assessment | Wiley . Wiley.com . https://www.wiley.com/en-us/Handbook+of+Motivational+Counseling%3A+Concepts%2C+Approaches%2C+and+Assessment-p-9780470092583 (accessed 18 April 2020 ). 21. West DS , DiLillo V, Bursac Z, Gore SA, Greene PG. Motivational interviewing improves weight loss in women with type 2 diabetes . Diabetes Care 2007 ; 30 : 1081 – 7 . Google Scholar Crossref Search ADS PubMed WorldCat 22. Peterson J , Repovich W, Parascand C. Accuracy of consumer grade bioelectrical impedance analysis devices compared to air displacement plethysmography . Int J Exerc Sci 2011 ; 4 : 176 – 83 . https://digitalcommons.wku.edu/ijes/vol4/iss3/2 Google Scholar OpenURL Placeholder Text WorldCat 23. Nichols J , Going S, Loftin M, Stewart D, Nowicki E, Pickrel J. Comparison of two bioelectrical impedance analysis instruments for determining body composition in adolescent girls . Int J Body Compos Res 2006 ; 4 : 153 – 60 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 24. (US) NOEIEP on the Identification, Evaluation, and Treatment of Obesity in Adults . Clinical Guidelines on the Identification, Evaluation, and Treatment of Overweight and Obesity in Adults . Bethesda, MD : National Heart, Lung, and Blood Institute , 1998 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 25. Renjilian DA , Perri MG, Nezu AM, McKelvey WF, Shermer RL, Anton SD. Individual versus group therapy for obesity: effects of matching participants to their treatment preferences . J Consult Clin Psychol 2001 ; 69 : 717 – 21 . Google Scholar Crossref Search ADS PubMed WorldCat 26. Hall DL , Lattie EG, McCalla JR, Saab PG. Translation of the diabetes prevention program to ethnic communities in the United States . J Immigr Minor Health 2016 ; 18 : 479 – 89 . Google Scholar Crossref Search ADS PubMed WorldCat 27. Khera R , Murad MH, Chandar AKet al. . Association of pharmacological treatments for obesity with weight loss and adverse events: a systematic review and meta-analysis . JAMA 2016 ; 315 : 2424 – 34 . Google Scholar Crossref Search ADS PubMed WorldCat 28. Bray GA , Frühbeck G, Ryan DH, Wilding JP. Management of obesity . Lancet 2016 ; 387 : 1947 – 56 . Google Scholar Crossref Search ADS PubMed WorldCat 29. LeBlanc ES , Patnode CD, Webber EM, Redmond N, Rushkin M, O’Connor EA. Behavioral and pharmacotherapy weight loss interventions to prevent obesity-related morbidity and mortality in adults: updated evidence report and systematic review for the US preventive services task force . JAMA 2018 ; 320 : 1172 – 91 . Google Scholar Crossref Search ADS PubMed WorldCat 30. Bombelli M , Facchetti R, Sega Ret al. . Impact of body mass index and waist circumference on the long-term risk of diabetes mellitus, hypertension, and cardiac organ damage . Hypertension 2011 ; 58 ( 6 ): 1029 – 35 . Google Scholar Crossref Search ADS PubMed WorldCat 31. Sobiczewski W , Wirtwein M, Jarosz D, Gruchala M. Superiority of waist circumference and body mass index in cardiovascular risk assessment in hypertensive patients with coronary heart disease . Blood Press 2015 ; 24 : 90 – 5 . Google Scholar Crossref Search ADS PubMed WorldCat 32. Nazare JA , Smith J, Borel ALet al. .; INSPIRE ME IAA Investigators. Usefulness of measuring both body mass index and waist circumference for the estimation of visceral adiposity and related cardiometabolic risk profile (from the INSPIRE ME IAA study) . Am J Cardiol 2015 ; 115 : 307 – 15 . Google Scholar Crossref Search ADS PubMed WorldCat © The Author(s) 2020. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Family Practice Oxford University Press

The effect of a primary care-based Medical Weight Loss Program on weight loss and anthropomorphic metrics

Download PDF
7 pages

Loading next page...
 
/lp/oxford-university-press/the-effect-of-a-primary-care-based-medical-weight-loss-program-on-LkknuB1lUB

References (30)

Publisher
Oxford University Press
Copyright
© The Author(s) 2020. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
ISSN
0263-2136
eISSN
1460-2229
DOI
10.1093/fampra/cmaa050
Publisher site
See Article on Publisher Site

Abstract

Abstract Background Diet and lifestyle intervention programs have been shown to be effective in decreasing obesity/overweight and many associated comorbidities in specialty research settings. There is very little information however as to the efficacy of such programs conducted in usual/typical primary care practices. We analysed effectiveness of the Medical Weight Loss Program (MWLP) designed to specifically address overweight/obesity in the setting of an urban academic primary care practice. Objective To determine whether participation in the MWLP within a general primary care setting can result in weight loss. Methods A retrospective medical chart review of patients treated in MWLP and a control group of patients with obesity receiving regular care in the general primary care setting. From the practice database (1 April 2015–31 March 2016), 209 patients (≥18 years old) who participated in the MWLP were identified; 265 controls were selected from the remaining population based on the presence of the obesity-related diagnoses. Results MWLP patients lost on average 2.35 ± 5.88 kg in 6 months compared to their baseline weight (P < 0.0001). In contrast, the control group demonstrated a trend of gaining on average 0.37 ± 6.03 kg. Having three or more visits with the MWLP provider within 6 months after program initiation was the most important factor associated with successful loss of at least 5% of the baseline weight. Weight loss also correlated with a decrease in abdominal girth. Conclusion MWLP integrated into the general primary care practice may potentially be an effective model for managing obesity and related morbidities. Lifestyle modification/health behaviour change, Medical Weight Loss Program, obesity, obesity management, prevention, primary care Key Messages Overweight/obesity and comorbid conditions are a major health problem. It is feasible to integrate obesity management into a primary care practice. Weight loss can be achieved in a primary care-based Medical Weight Loss Program. Introduction Obesity rates have continued to rise over the past decades, with a prevalence of obesity in 2015–16 approaching 40% of the US adult population (1). Obesity is one of the biggest drivers of preventable chronic diseases and health care costs in the USA (2). Furthermore, indirect costs of obesity due to lost productivity and absenteeism run into the billions of dollars (2–4). With little evidence of imminent remission, it has become increasingly important to formulate effective means of screening, assessing and treating obesity. Primary care providers must be included in these approaches as they see the vast majority of patients who present for ambulatory care (5). A vast proportion of these patients suffer from obesity and related conditions (6,7). While surgical options have been shown to provide good evidence of disease remittance (8–10), many patients with obesity do not qualify for, nor desire surgery. These patients must therefore be managed medically. The US Preventive Task Force recommends screening all adults for obesity (11) and found adequate evidence that intensive, multicomponent behavioural interventions can lead to weight loss, improved glucose tolerance and reduction in risk factors for cardiovascular disease (11). For patients with body mass index (BMI) >30 kg/m2, clinicians are encouraged to offer or refer patients for intensive, multicomponent behavioural interventions (11). There is overwhelming evidence that diet and lifestyle intervention programs can be effective in decreasing obesity and many associated comorbidities in specialty research settings (12–14). The Diabetes Prevention Program and Look Ahead studies were two landmark randomized clinical trials that provided strong evidence of this approach (12–14). However, these large and highly resource-intensive multi-centre studies were conducted over the course of several years, according to rigorous research protocols with dedicated research personnel. Subsequent research directed at investigating the translatability of elements of these programs into real-world clinical settings showed varying degrees of success (15–17). In this study, we assessed the effectiveness of the Medical Weight Loss Program (MWLP) based on lifestyle changes in a large urban primary care practice that serves a predominantly socioeconomically, disadvantaged minority population. Methods Study design We conducted a retrospective medical chart review comparing patients treated in MWLP to the control group of patients with obesity who received usual care in the same primary care practice. Description of MWLP Setting The program was offered at the clinical offices of the University Family Medicine (UFM) on the campus of the University of Maryland Medical Systems (UMMS). The UFM is a large urban, academic safety-net primary care practice that serves a primarily minority population with about 50 000 patient visits per year. The practice is located in Baltimore City where an adult obesity rate of 36.6% surpasses the state average (18). Clinical intervention MWLP was under the direction of the physician certified by the American Board of Obesity Medicine and a physician assistant certified in diabetes education. Both professionals had extensive training in nutrition, especially as it pertains to overweight and obesity. Interactions with the Obesity Medicine Specialist were based on the principle of motivational interviewing, which has been shown to facilitate behaviour change (19–21). The intake appointment included a detailed review of medical history and information from the comprehensive questionnaire provided to patients prior to scheduling a visit. The questionnaire encompassed dietary and lifestyle issues known to impact weight based on recommendations from the Obesity Medicine Association. This information was used in developing individualized treatment plans, including diet and exercise counselling, support for self-management and pharmacotherapy for select patients. The plan also integrated information about socioeconomic status and domestic situation. At the initial visit, standard vital signs were obtained. Anthropomorphic measurements were obtained using a standard tape measure. Valhalla Scientific Body Composition Scale was used to collect data on weight, percent body fat and lean tissue, resting energy expenditure and target weight. This scale uses electrical impedance to differentiate between different types of body tissue and is commonly used in clinical practice (22,23). For all patients, metabolic data were ordered if not available from the past 3–6 months (Comprehensive Metabolic Panel, Lipids, Glycohemoglobin, Thyroid Stimulating Hormone, CBC, micronutrients levels). Medications were also reviewed, and weight-positive drugs were replaced with alternate weight-neutral or weight-negative drugs where possible. A calorie deficit of ~500–1000 calories per day was recommended (24), which sometimes included meal replacement products. Patients were asked to keep a food journal. If indicated, patients were referred to other specialists such as Sleep Medicine, Cardiology, Physical Therapy, Psychology/Behavioural Medicine or Bariatric Surgery. Patients were also prescribed an exercise program customized to their physical ability. Based on the evidence that the group process can be effective in promoting weight loss, particularly in some minority populations (25,26), weekly group meetings attendance was encouraged but not mandated. Follow-up visits included weight measurement, monitoring of body composition, diet and exercise review, and adjustment of eating plan. Since the adjunctive use of anti-obesity medications (AOMs) has proved effective in reducing weight compared to placebo (27–29), suitable candidates for drug therapy were prescribed AOMs as appropriate. Prescribing guidelines were strictly followed. Study population MWLP group Patients accessed the program either by referral from specialty or primary care providers or self-referral. Informational material about the program was available within the UMMS network as well as through community activities such as health fairs. All patients with overweight and obesity who requested medical weight management services were accepted except patients who had undergone bariatric surgery within 6 months of program enrolment. Records for adult UFM patients (≥18 years old) with invoices between 1 April 2015 and 31 March 2016 were retrieved using IBM Cognos Analytics software (Armonk, NY). MWLP patients were identified based on their record of participation in the program. Medical charts for these patients were reviewed to determine the date of the program enrolment (time frame July 2012–August 2016). Patients who had at least two weight measurements for at least 6 months were included. All available measurements within a 12-month time frame were used regardless of whether they were obtained as a part of MWLP or unrelated clinical visits. Pregnant women and patients with cancer were excluded. Data for co-morbidities (diabetes mellitus type 2, insulin resistance, hypertension, dyslipidemia, sleep apnoea, history of gastric bypass surgery, malabsorption, depression, anxiety, bipolar disorder) were identified by ICD10 codes. Demographic data (age, sex, race, ethnicity), anthropomorphic and vital measurements, number of visits with the MWLP provider and calorie restriction recommendations were obtained from medical charts. Control group Control patients with obesity were identified from the remaining records based on the presence of the obesity-related ICD10 (the International Classification of Diseases, 10th Revision, Clinical Modification) codes (overweight, obese with or without complications, unusual weight gain). Patients were selected using the same inclusion/exclusion criteria as the MWLP group. Additional inclusion criterion for the control group was the same period for observations as for MWLP patients (time frame July 2012–August 2016). Outcomes The major outcome measure was relative weight change at 6 months in the MWLP group compared to the control patients. Secondary outcomes for MWLP patients were 5% weight loss was achieved at any point in 12 months and the correlation between 6-month changes in weight and anthropomorphic measurements. Statistical analysis Data were analysed using SAS statistical package version 9.3 (SAS Institute, Cary, NC). For MWLP patients, the baseline was the measurement at the first program visit. For the control group, the baseline was the first recorded vital measurement in the chart within the study time frame. Weight change at 6 months was estimated by linear, quadratic or higher order polynomial regression analysis using weight measurements flanking a 6-months’ time point (R2 > 0.95). For the continuous variables, differences between the treatment and control groups were assessed using a two-tailed Student’s t-test. Associations between categorical variables were examined using chi-square or Fisher’s exact tests. Correlations between changes in weight and anthropomorphic metrics (waist circumference) were assessed using the Pearson correlation coefficient. The effects of MWLP intervention, demographic characteristics and comorbidities on either 6-month weight loss or anthropomorphic metrics were assessed using multivariable linear regression models. For the loss of 5% of baseline weight, time to achieve target weight loss was estimated by regression analysis using dates flanking the target date. Time-to-event analysis was performed on imputed time values using Kaplan–Meier’s survival curves and Cox’s proportional hazard modelling with observation time censored at 12 months after acceptance to the program. Patients who had not achieved a 5% weight loss in 12 months were censored at their maximum observation time. Covariates included in the multivariable regression models were identified based on preliminary bivariate analyses looking for potential effect modifiers and confounders. Statistical significance was established at two-sided α = 0.05. Results Participants Overall, 8392 records from 3291 individuals with active charts between 1 April 2015 and 31 March 2016 were retrieved, and 236 MWLP patients and 391 controls were identified. Data for 209 MWLP patients and 265 controls who had at least 2 weight measurements for at least 6 months were used for the analysis (Supplementary Figure 1). Control patients excluded from the analysis were younger compared to those who were included (37.5 ± 12.1 and 41.3 ± 14.0 years respectively, P = 0.0067) and had lower prevalence of diabetes (10% and 24% respectively, P = 0.0128) and sleep apnoea (1.5% and 6% respectively, P = 0.0431). MWLP patients excluded from the analyses did not differ significantly in demographic and clinical characteristics from those who were included except that they had lower AOM prescription rate (26% and 52% respectively, P = 0.0103). Characteristics of the study population are given in Table 1. Compared to the controls, the MWLP group had higher initial weight and BMI, had more women, Black patients, and patients with diabetes, sleep apnoea, depression and bipolar disorder (Table 1). The median number of weight measurements per patient within the first 6 months was 4 (range 2–23), with median time between observations 0.8 months (25th–75th percentile 0.4–1.3 months). Among MWLP patients, 52% were prescribed AOM. The number of visits with the MWLP provider within 6 months after acceptance to the program varied from 1 to 10 (median 3 visits). For those who had two or more visits, the median time between visits was 1.1 months. A calorie and carb-restricted diet with 1300–1400 calories per day was recommended to 51% of patients; 9% of patients were placed on a more restricted diet (1100–1200 cal/day). Table 1. Baseline characteristics of 474 patients seen in the UFM practice between July 2012 and August 2016 Characteristics . . MWLP, N = 209 . Control, N = 265 . P* . Age at baseline, years Mean ± SD 42.3 ± 11.2 41.3±14.0 0.3590T Median 40 40 Q1–Q3 34–50 30–52 Range 18–77 18–83 Weight at baseline, kg Mean ± SD 121.0 ± 28.3 110.2±27.6 <0.0001 T Median 116.3 104.9 Q1–Q3 101.0–135.6 88.5–126.1 Range 65.2–221.8 65.3–231.8 BMI at baseline, kg/m2 Mean ± SD 44.4 ± 9.7 40.1±9.6 <0.0001 T Median 42.9 37.9 Q1–Q3 37.3–49.4 33.5 – 45.4 Range 26.7–86.6 25.3–93.4 Age categories at baseline, N (%) 18–34 years 59 (28%) 93 (35%) 0.0742 35–54 years 111 (53%) 113 (43%) 55+ years 39 (19%) 58 (22%) Sex, N (%) Female 189 (90%) 212 (80%) 0.0018 Male 20 (10%) 53 (20%) Race, N (%) Black 179 (86%) 202 (76%) <0.0001 White 12 (6%) 52 (20%) Other 18 (8%) 11 (4%) BMI category at baseline, N (%) 25–29.9 kg/m2 4 (2%) 27 (10%) <0.0001 30–39.9 kg/m2 64 (31%) 129 (49%) 40–49.9 kg/m2 93 (44%) 72 (27%) 50+ kg/m2 48 (23%) 37 (14%) Obesity, N (%) Overweight 4 (2%) 15 (6%) <0.0001 Obese 95 (46%) 136 (51%) Morbidly obese 71 (34%) 101 (38%) Abnormal weight gain 5 (2%) 13 (5%) No related ICD10 code 34 (16%) 0 (0%) Diabetes, N (%) No related ICD10 code 111 (53%) 181 (71%) 0.0003 Prediabetes 42 (20%) 20 (8%) DM II without complications 39 (19%) 44 (14%) DM II with complications 17 (8%) 20 (7%) Insulin resistance (Yes), N (%) 10 (5%) 5 (2%) 0.0735 Dyslipidemia (Yes), N (%) 23 (11%) 22 (8%) 0.3189 Sleep apnoea (Yes), N (%) 20 (10%) 4 (2%) <0.0001 Gastric bypass surgery (Yes), N (%) 6 (3%) 1 (0.4%) 0.0475 F Malabsorption (Yes), N (%) 7 (3%) 0 (0%) 0.0031 F Hypertension (Yes), N (%) 93 (45%) 102 (38%) 0.1870 Depression (Yes), N (%) 32 (15%) 16 (6%) 0.0009 Bipolar disorder (Yes), N (%) 7 (3%) 1 (0.4%) 0.0243 F Anxiety (Yes), N (%) 6 (3%) 8 (3%) 1.0000 Anti-obesity medication (Yes), N (%) 109 (52%) 3 (0.8%) <0.0001 Characteristics . . MWLP, N = 209 . Control, N = 265 . P* . Age at baseline, years Mean ± SD 42.3 ± 11.2 41.3±14.0 0.3590T Median 40 40 Q1–Q3 34–50 30–52 Range 18–77 18–83 Weight at baseline, kg Mean ± SD 121.0 ± 28.3 110.2±27.6 <0.0001 T Median 116.3 104.9 Q1–Q3 101.0–135.6 88.5–126.1 Range 65.2–221.8 65.3–231.8 BMI at baseline, kg/m2 Mean ± SD 44.4 ± 9.7 40.1±9.6 <0.0001 T Median 42.9 37.9 Q1–Q3 37.3–49.4 33.5 – 45.4 Range 26.7–86.6 25.3–93.4 Age categories at baseline, N (%) 18–34 years 59 (28%) 93 (35%) 0.0742 35–54 years 111 (53%) 113 (43%) 55+ years 39 (19%) 58 (22%) Sex, N (%) Female 189 (90%) 212 (80%) 0.0018 Male 20 (10%) 53 (20%) Race, N (%) Black 179 (86%) 202 (76%) <0.0001 White 12 (6%) 52 (20%) Other 18 (8%) 11 (4%) BMI category at baseline, N (%) 25–29.9 kg/m2 4 (2%) 27 (10%) <0.0001 30–39.9 kg/m2 64 (31%) 129 (49%) 40–49.9 kg/m2 93 (44%) 72 (27%) 50+ kg/m2 48 (23%) 37 (14%) Obesity, N (%) Overweight 4 (2%) 15 (6%) <0.0001 Obese 95 (46%) 136 (51%) Morbidly obese 71 (34%) 101 (38%) Abnormal weight gain 5 (2%) 13 (5%) No related ICD10 code 34 (16%) 0 (0%) Diabetes, N (%) No related ICD10 code 111 (53%) 181 (71%) 0.0003 Prediabetes 42 (20%) 20 (8%) DM II without complications 39 (19%) 44 (14%) DM II with complications 17 (8%) 20 (7%) Insulin resistance (Yes), N (%) 10 (5%) 5 (2%) 0.0735 Dyslipidemia (Yes), N (%) 23 (11%) 22 (8%) 0.3189 Sleep apnoea (Yes), N (%) 20 (10%) 4 (2%) <0.0001 Gastric bypass surgery (Yes), N (%) 6 (3%) 1 (0.4%) 0.0475 F Malabsorption (Yes), N (%) 7 (3%) 0 (0%) 0.0031 F Hypertension (Yes), N (%) 93 (45%) 102 (38%) 0.1870 Depression (Yes), N (%) 32 (15%) 16 (6%) 0.0009 Bipolar disorder (Yes), N (%) 7 (3%) 1 (0.4%) 0.0243 F Anxiety (Yes), N (%) 6 (3%) 8 (3%) 1.0000 Anti-obesity medication (Yes), N (%) 109 (52%) 3 (0.8%) <0.0001 Data for 209 MWLP and 265 control patients with a baseline observation time between July 2012 and August 2016 are shown. DM, diabetes mellitus type II; ICD10, The International Classification of Diseases, 10th Revision; UFM, University of Maryland Family Medicine. P-values <0.05 are shown in bold. Q1–Q3: 25th and 75th percentiles respectively. *Chi-square test, unless indicated otherwise: T: Student’s t-test (two-sided), F: Fisher’s exact test (two-sided). Open in new tab Table 1. Baseline characteristics of 474 patients seen in the UFM practice between July 2012 and August 2016 Characteristics . . MWLP, N = 209 . Control, N = 265 . P* . Age at baseline, years Mean ± SD 42.3 ± 11.2 41.3±14.0 0.3590T Median 40 40 Q1–Q3 34–50 30–52 Range 18–77 18–83 Weight at baseline, kg Mean ± SD 121.0 ± 28.3 110.2±27.6 <0.0001 T Median 116.3 104.9 Q1–Q3 101.0–135.6 88.5–126.1 Range 65.2–221.8 65.3–231.8 BMI at baseline, kg/m2 Mean ± SD 44.4 ± 9.7 40.1±9.6 <0.0001 T Median 42.9 37.9 Q1–Q3 37.3–49.4 33.5 – 45.4 Range 26.7–86.6 25.3–93.4 Age categories at baseline, N (%) 18–34 years 59 (28%) 93 (35%) 0.0742 35–54 years 111 (53%) 113 (43%) 55+ years 39 (19%) 58 (22%) Sex, N (%) Female 189 (90%) 212 (80%) 0.0018 Male 20 (10%) 53 (20%) Race, N (%) Black 179 (86%) 202 (76%) <0.0001 White 12 (6%) 52 (20%) Other 18 (8%) 11 (4%) BMI category at baseline, N (%) 25–29.9 kg/m2 4 (2%) 27 (10%) <0.0001 30–39.9 kg/m2 64 (31%) 129 (49%) 40–49.9 kg/m2 93 (44%) 72 (27%) 50+ kg/m2 48 (23%) 37 (14%) Obesity, N (%) Overweight 4 (2%) 15 (6%) <0.0001 Obese 95 (46%) 136 (51%) Morbidly obese 71 (34%) 101 (38%) Abnormal weight gain 5 (2%) 13 (5%) No related ICD10 code 34 (16%) 0 (0%) Diabetes, N (%) No related ICD10 code 111 (53%) 181 (71%) 0.0003 Prediabetes 42 (20%) 20 (8%) DM II without complications 39 (19%) 44 (14%) DM II with complications 17 (8%) 20 (7%) Insulin resistance (Yes), N (%) 10 (5%) 5 (2%) 0.0735 Dyslipidemia (Yes), N (%) 23 (11%) 22 (8%) 0.3189 Sleep apnoea (Yes), N (%) 20 (10%) 4 (2%) <0.0001 Gastric bypass surgery (Yes), N (%) 6 (3%) 1 (0.4%) 0.0475 F Malabsorption (Yes), N (%) 7 (3%) 0 (0%) 0.0031 F Hypertension (Yes), N (%) 93 (45%) 102 (38%) 0.1870 Depression (Yes), N (%) 32 (15%) 16 (6%) 0.0009 Bipolar disorder (Yes), N (%) 7 (3%) 1 (0.4%) 0.0243 F Anxiety (Yes), N (%) 6 (3%) 8 (3%) 1.0000 Anti-obesity medication (Yes), N (%) 109 (52%) 3 (0.8%) <0.0001 Characteristics . . MWLP, N = 209 . Control, N = 265 . P* . Age at baseline, years Mean ± SD 42.3 ± 11.2 41.3±14.0 0.3590T Median 40 40 Q1–Q3 34–50 30–52 Range 18–77 18–83 Weight at baseline, kg Mean ± SD 121.0 ± 28.3 110.2±27.6 <0.0001 T Median 116.3 104.9 Q1–Q3 101.0–135.6 88.5–126.1 Range 65.2–221.8 65.3–231.8 BMI at baseline, kg/m2 Mean ± SD 44.4 ± 9.7 40.1±9.6 <0.0001 T Median 42.9 37.9 Q1–Q3 37.3–49.4 33.5 – 45.4 Range 26.7–86.6 25.3–93.4 Age categories at baseline, N (%) 18–34 years 59 (28%) 93 (35%) 0.0742 35–54 years 111 (53%) 113 (43%) 55+ years 39 (19%) 58 (22%) Sex, N (%) Female 189 (90%) 212 (80%) 0.0018 Male 20 (10%) 53 (20%) Race, N (%) Black 179 (86%) 202 (76%) <0.0001 White 12 (6%) 52 (20%) Other 18 (8%) 11 (4%) BMI category at baseline, N (%) 25–29.9 kg/m2 4 (2%) 27 (10%) <0.0001 30–39.9 kg/m2 64 (31%) 129 (49%) 40–49.9 kg/m2 93 (44%) 72 (27%) 50+ kg/m2 48 (23%) 37 (14%) Obesity, N (%) Overweight 4 (2%) 15 (6%) <0.0001 Obese 95 (46%) 136 (51%) Morbidly obese 71 (34%) 101 (38%) Abnormal weight gain 5 (2%) 13 (5%) No related ICD10 code 34 (16%) 0 (0%) Diabetes, N (%) No related ICD10 code 111 (53%) 181 (71%) 0.0003 Prediabetes 42 (20%) 20 (8%) DM II without complications 39 (19%) 44 (14%) DM II with complications 17 (8%) 20 (7%) Insulin resistance (Yes), N (%) 10 (5%) 5 (2%) 0.0735 Dyslipidemia (Yes), N (%) 23 (11%) 22 (8%) 0.3189 Sleep apnoea (Yes), N (%) 20 (10%) 4 (2%) <0.0001 Gastric bypass surgery (Yes), N (%) 6 (3%) 1 (0.4%) 0.0475 F Malabsorption (Yes), N (%) 7 (3%) 0 (0%) 0.0031 F Hypertension (Yes), N (%) 93 (45%) 102 (38%) 0.1870 Depression (Yes), N (%) 32 (15%) 16 (6%) 0.0009 Bipolar disorder (Yes), N (%) 7 (3%) 1 (0.4%) 0.0243 F Anxiety (Yes), N (%) 6 (3%) 8 (3%) 1.0000 Anti-obesity medication (Yes), N (%) 109 (52%) 3 (0.8%) <0.0001 Data for 209 MWLP and 265 control patients with a baseline observation time between July 2012 and August 2016 are shown. DM, diabetes mellitus type II; ICD10, The International Classification of Diseases, 10th Revision; UFM, University of Maryland Family Medicine. P-values <0.05 are shown in bold. Q1–Q3: 25th and 75th percentiles respectively. *Chi-square test, unless indicated otherwise: T: Student’s t-test (two-sided), F: Fisher’s exact test (two-sided). Open in new tab Weight change at 6 months To characterize the impact of MWLP on weight loss, we examined whether weight change over the period of 6 months was different between groups of comparison. MWLP patients lost on average 2.35 ± 5.88 kg in 6 months (P < 0.0001) compared to their baseline weight. In contrast, the control group demonstrated a trend of gaining on average 0.37 ± 6.03 kg, although the change did not differ significantly from zero (P = 0.3140). In the bivariate analysis, a relative difference in the 6-month weight change between MWLP and control groups was 2.73 ± 5.96 kg (P < 0.0001) (Fig. 1). Figure 1. Open in new tabDownload slide Comparison of the 6-month weight change for 474 patients seen in the UFM practice. Data for 209 MWLP and 265 control patients with a baseline observation time between July 2012 and August 2016 are shown. Data are presented as box plots. Horizontal lines within the boxes are medians, diamonds mark mean, box frames mark 25th and 75th percentiles, whiskers represent 5th and 95th percentiles, dots are outliers. P-value on the graph is for the two-sided t-test. MWLP, Medical Weight Loss Program; UFM, University of Maryland Family Medicine. Figure 1. Open in new tabDownload slide Comparison of the 6-month weight change for 474 patients seen in the UFM practice. Data for 209 MWLP and 265 control patients with a baseline observation time between July 2012 and August 2016 are shown. Data are presented as box plots. Horizontal lines within the boxes are medians, diamonds mark mean, box frames mark 25th and 75th percentiles, whiskers represent 5th and 95th percentiles, dots are outliers. P-value on the graph is for the two-sided t-test. MWLP, Medical Weight Loss Program; UFM, University of Maryland Family Medicine. A multivariable linear regression model showed that, among demographic and clinical characteristics (listed in Table 1), only the intervention (MWLP) and BMI at baseline were significantly associated with a 6-month weight change. After adjusting for the baseline BMI, the difference in 6-month weight loss between MWLP and control groups was on average 2.23 kg (95% CI 1.12–3.34 kg; P < 0.0001). A significant weight loss was observed in MWLP patients with baseline BMI 30 kg/m2 or greater while in the control group a significant weight gain was observed in the group with BMI 25–29.9 kg/m2 (Table 2). Table 2. Results of the multivariable linear regression model estimating magnitude of weight change in 474 patients seen in the UFM practice between July 2012 and August 2016 BMI categories, kg/m2 . Intervention groups . Absolute change, kg . . . N . Mean . 95% CI . P . 25–29.9 Control 27 +2.5 +0.5 to +4.6 0.0168 MWLP 4 +0.3 −2.0 to +2.6 0.7930 30–39.9 Control 129 +0.7 −0.2 to +1.6 0.1127 MWLP 64 −1.5 −2.6 to −0.4 0.0086 40–49.9 Control 72 −0.5 −1.6 to +0.6 0.3345 MWLP 93 −2.8 −3.8 to −1.7 <0.0001 50+ Control 37 −0.7 −2.1 to +0.7 0.3355 MWLP 48 −2.9 −4.3 to −1.6 <0.0001 BMI categories, kg/m2 . Intervention groups . Absolute change, kg . . . N . Mean . 95% CI . P . 25–29.9 Control 27 +2.5 +0.5 to +4.6 0.0168 MWLP 4 +0.3 −2.0 to +2.6 0.7930 30–39.9 Control 129 +0.7 −0.2 to +1.6 0.1127 MWLP 64 −1.5 −2.6 to −0.4 0.0086 40–49.9 Control 72 −0.5 −1.6 to +0.6 0.3345 MWLP 93 −2.8 −3.8 to −1.7 <0.0001 50+ Control 37 −0.7 −2.1 to +0.7 0.3355 MWLP 48 −2.9 −4.3 to −1.6 <0.0001 Data for 209 MWLP and 265 control patients with a baseline observation time between July 2012 and August 2016 are shown. P-values are for the two-sided t-test; P-values <0.05 are shown in bold. MWLP, Medical Weight Loss Program; BMI, body mass index; CI, confidence interval; UFM, University of Maryland Family Medicine. Open in new tab Table 2. Results of the multivariable linear regression model estimating magnitude of weight change in 474 patients seen in the UFM practice between July 2012 and August 2016 BMI categories, kg/m2 . Intervention groups . Absolute change, kg . . . N . Mean . 95% CI . P . 25–29.9 Control 27 +2.5 +0.5 to +4.6 0.0168 MWLP 4 +0.3 −2.0 to +2.6 0.7930 30–39.9 Control 129 +0.7 −0.2 to +1.6 0.1127 MWLP 64 −1.5 −2.6 to −0.4 0.0086 40–49.9 Control 72 −0.5 −1.6 to +0.6 0.3345 MWLP 93 −2.8 −3.8 to −1.7 <0.0001 50+ Control 37 −0.7 −2.1 to +0.7 0.3355 MWLP 48 −2.9 −4.3 to −1.6 <0.0001 BMI categories, kg/m2 . Intervention groups . Absolute change, kg . . . N . Mean . 95% CI . P . 25–29.9 Control 27 +2.5 +0.5 to +4.6 0.0168 MWLP 4 +0.3 −2.0 to +2.6 0.7930 30–39.9 Control 129 +0.7 −0.2 to +1.6 0.1127 MWLP 64 −1.5 −2.6 to −0.4 0.0086 40–49.9 Control 72 −0.5 −1.6 to +0.6 0.3345 MWLP 93 −2.8 −3.8 to −1.7 <0.0001 50+ Control 37 −0.7 −2.1 to +0.7 0.3355 MWLP 48 −2.9 −4.3 to −1.6 <0.0001 Data for 209 MWLP and 265 control patients with a baseline observation time between July 2012 and August 2016 are shown. P-values are for the two-sided t-test; P-values <0.05 are shown in bold. MWLP, Medical Weight Loss Program; BMI, body mass index; CI, confidence interval; UFM, University of Maryland Family Medicine. Open in new tab Factors associated with a 5% loss of the baseline weight in MWLP patients Kaplan–Meier’s time-to-event analysis showed that 82 patients (39%) achieved a 5% loss of their baseline weight during 12 months of observation. The number of visits with the MWLP provider within 6 months after initiation of the program was the most important factor associated with successful weight loss (Fig. 2, log-rank test P < 0.0001). For patients who had three or more visits, the median time for losing 5% weight was 5.9 months (95% CI 3.9–7.1). Only 24% of patients who had less than three visits achieved 5% weight loss (25th quantile 11.1 months, 95% CI 6.4–12.6). Age and BMI at baseline were also significantly associated with the rate of weight loss (data not shown). Figure 2. Open in new tabDownload slide Association between weight loss and the number of visits in patients participated in the UFM MWLP. Data for 209 MWLP patients with a baseline observation time between July 2012 and August 2016 are shown. Time to achieve 5% weight loss within 12 months after program acceptance was estimated by regression analysis. Kaplan–Meier’s curves are shown for patients who had three or more visits within 6 months after program acceptance (solid line) compared to those who had one or two visits (dash line). Grey-shaded areas define upper and lower 95% confidence intervals. P-value on the graph is for a log-rank test. MWLP, Medical Weight Loss Program; UFM, University of Maryland Family Medicine. Figure 2. Open in new tabDownload slide Association between weight loss and the number of visits in patients participated in the UFM MWLP. Data for 209 MWLP patients with a baseline observation time between July 2012 and August 2016 are shown. Time to achieve 5% weight loss within 12 months after program acceptance was estimated by regression analysis. Kaplan–Meier’s curves are shown for patients who had three or more visits within 6 months after program acceptance (solid line) compared to those who had one or two visits (dash line). Grey-shaded areas define upper and lower 95% confidence intervals. P-value on the graph is for a log-rank test. MWLP, Medical Weight Loss Program; UFM, University of Maryland Family Medicine. Multivariable Cox’s proportional hazard modelling showed that, controlling for baseline age and BMI, the rate of weight loss for patients who had three or more visits was 4.4 times higher compared to patients who had less than three visits [Hazard Ratio (HR) = 4.4; 95% CI 2.8–7.0; P < 0.0001]. Controlling for the baseline age and number of MWLP visits, it was twice harder to lose 5% weight for patients with BMI 40–49.9 kg/m2 compared to patients with BMI <40 kg/m2 (HR = 0.5; 95% CI 0.3–0.8; P = 0.0093). There was no statistically significant difference in the rate of weight loss between patients with BMI 40–49.9 kg/m2 and BMI ≥50 kg/m2 (data not shown). It was easier for younger patients to lose weight, with a chance (hazard) of losing 5% weight decreasing by 2.5% per 1-year increase in baseline age (HR = 0.975; 95% CI 0.955–0.995; P = 0.0071). The effects of AOM prescription or initial calorie restriction recommendations were not significantly associated with a chance of losing 5% weight (data not shown). Correlation between anthropomorphic measurements and weight change at 6 months We also determined whether 6-month weight changes in MWLP patients were accompanied by the corresponding changes in waist circumference. In the univariate analysis, there was a significant decrease in the waist (−4.4 ± 8.9 cm, P < 0.0001) circumferences at 6 months that correlated with weight changes [Pearson correlation coefficient (ρ) = 0.47, P < 0.0001, Fig. 3]. In the multivariable linear regression model, a 6-month weight change, baseline weight and baseline waist circumference were significantly associated with a 6-month change in waist circumference (P < 0.0001 for all factors, data not shown). Adjusted for the baseline factors, waist circumference decreased on average by 0.64 cm per 1 kg of weight loss (95% CI 0.46–0.82 cm, P < 0.0001). Figure 3. Open in new tabDownload slide Association between changes in weight and waist circumference in patients participated in the UFM MWLP. Among 209 MWLP patients with baseline observation time between July 2012 and August 2016, 107 patients who had at least two anthropomorphic measurements during at least 6 months were identified. After removing two outliers, data for 105 MWLP patients were analysed using a linear regression model. Thick line represents a regression trend; grey-shaded area defines upper and lower 95% confidence intervals, and dash lines represent upper and lower prediction intervals. Numbers on the graph for a Pearson’s correlation coefficient (ρ) with a corresponding P-value. MWLP, Medical Weight Loss Program; UFM, University of Maryland Family Medicine. Figure 3. Open in new tabDownload slide Association between changes in weight and waist circumference in patients participated in the UFM MWLP. Among 209 MWLP patients with baseline observation time between July 2012 and August 2016, 107 patients who had at least two anthropomorphic measurements during at least 6 months were identified. After removing two outliers, data for 105 MWLP patients were analysed using a linear regression model. Thick line represents a regression trend; grey-shaded area defines upper and lower 95% confidence intervals, and dash lines represent upper and lower prediction intervals. Numbers on the graph for a Pearson’s correlation coefficient (ρ) with a corresponding P-value. MWLP, Medical Weight Loss Program; UFM, University of Maryland Family Medicine. Discussion Our study represents the first attempt of describing the MWLP established within a primary care practice and evaluating its effectiveness. While there were several studies that examined weight loss programs (12–14), our study is unique because it was conducted in a typical primary care practice that serves primarily low socioeconomic status patients. While MWLP did not have all the elements of an ideal multi-disciplinary weight loss program, successful weight loss was achieved by engaging patients with trained dedicated personnel who incorporated the existing resources of a general primary care practice into MWLP. On average, MWLP patients with obesity lost a significant amount of weight during the first 6 months. Moreover, a pronounced weight gain was observed among overweight (BMI 25–30 kg/m2) control group. This trend is concerning and consistent with the ever-increasing rates of obesity. This may be indicative that more efforts should be made to engage the overweight population in weight loss programs, especially in primary care. Patients attending the program were mostly urban, African American females with severe obesity and comorbid conditions. Our results demonstrated that weight loss is attainable for these patients despite the challenges inherent in this demographic group. Patients in the MWLP group also had a significantly higher baseline BMI and more comorbid conditions compared to the control group. These conditions increase the difficulty of achieving a negative energy balance and thus of losing weight. Despite this, MWLP proved effective for these patients to lose weight. This would suggest that persons with obesity and related co-morbidities should be targeted for primary care-based lifestyle interventions for weight loss. Moreover, our data show that it was easier for younger MWLP patients as well as those with lower BMI (<40 kg/m2) to lose weight. With the previously noted tendency of patients to gain weight over time, targeting younger patients who are overweight or have a BMI of less than 40 kg/m2 could be a worthwhile strategy. Our study identified no effect of AOM on the likelihood of weight loss. However, the data were limited as there was only information on the prescription of the medications, not on actual usage. The number of visits with the MWLP provider was the most significant factor associated with the successful loss of 5% of initial weight. This might be due to the higher personal motivation of the patients who continued with the program as well as the effect of continued reinforcement of diet and lifestyle changes, support and accountability provided by the physician. This could be explored in future studies. For patients showing weight loss over a 6 months period, there was a correlating significant decrease in abdominal girth. It is well established that BMI and waist circumference are two factors that correlate with cardiovascular risk and can even be useful in risk stratification in certain patients (30–32). Thus, MWLP may play an important role in decreasing obesity-related cardiovascular risks in this population. Our study had several limitations. This was a retrospective review of the clinical records. Not all patients were observed exactly at 6 months, and analyses were conducted on imputed values. There was no defined baseline time for the control group. The results may be also limited in generalizability since our patient population consisted mainly of urban African American women, many of whom are negatively impacted by social determinants of health and represent a segment of the population in which obesity rates continue to rise (1). However, the study is important as it provides information on the effectiveness of obesity treatment in a real-world primary care practice where a vast majority of patients suffering from obesity and comorbid conditions are seen (6,7). More information is needed about the treatment of obesity in primary care settings, which rarely have the time, personnel and financial resources compared to the research setting in clinical trials. Our study aimed to determine whether obesity management can be effective in primary care and to provide data on a population that is often underrepresented in scientific studies. The results indicate that our model may be an effective approach that is worth developing. Conclusions Weight loss is achievable in a primary care setting. To address the obesity crisis, ways must be found to engage primary care providers and practices in translating known research into workable programs. Declaration Funding: The study was funded by departmental resources. Ethical approval: The exempt status of the study was confirmed by the University of Maryland Baltimore Institutional Review Board (approved 19 December 2016; reference number HR00073199). Conflict of interest: none. References 1. Hales CM , Carroll MD, Fryar CD, Ogden CL. Prevalence of Obesity Among Adults and Youth: United States, 2015–2016 . NCHS data brief, no 288. Hyattsville, MD : National Center for Health Statistics , 2017 . 2. Waters H , Graf M. America’s Obesity Crisis: The Health and Economic Costs of Excess Weight | Milken Institute . https://milkeninstitute.org/reports/americas-obesity-crisis-health-and-economic-costs-excess-weight (accessed on 18 April 2020 ). Published 2018. 3. Andreyeva T , Luedicke J, Wang YC. State-level estimates of obesity-attributable costs of absenteeism . J Occup Environ Med 2014 ; 56 : 1120 – 7 . Google Scholar Crossref Search ADS PubMed WorldCat 4. Hammond RA , Levine R. The economic impact of obesity in the United States . Diabetes Metab Syndr Obes 2010 ; 3 : 285 – 95 . Google Scholar Crossref Search ADS PubMed WorldCat 5. Rui P , Okeyode T. National Ambulatory Medical Care Survey: 2016 National Summary Tables . https://www.cdc.gov/nchs/data/ahcd/namcs_summary/2016_namcs_web_tables.pdf (accessed on 12 April 2020 ). 6. Ornstein SM , Jenkins RG, Litvin CB, Wessell AM, Nietert PJ. Preventive services delivery in patients with chronic illnesses: parallel opportunities rather than competing obligations . Ann Fam Med 2013 ; 11 : 344 – 9 . Google Scholar Crossref Search ADS PubMed WorldCat 7. Petterson S , McNellis R, Klink K, Meyers D, Bazemore A. The State of Primary Care in the United States: A Chartbook of Facts and Statistics . Robert Graham Center , January 2018. https://www.graham-center.org/content/dam/rgc/documents/publications-reports/reports/PrimaryCareChartbook.pdf (accessed on 12 April 2020 ). Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 8. Brethauer SA , Aminian A, Romero-Talamás Het al. . Can diabetes be surgically cured? Long-term metabolic effects of bariatric surgery in obese patients with type 2 diabetes mellitus . Ann Surg 2013 ; 258 : 628 – 36 ; discussion 636–7. Google Scholar Crossref Search ADS PubMed WorldCat 9. Schauer PR , Bhatt DL, Kashyap SR. Bariatric surgery or intensive medical therapy for diabetes after 5 years . N Engl J Med 2017 ; 376 : 1997 . Google Scholar Crossref Search ADS PubMed WorldCat 10. Colquitt JL , Pickett K, Loveman E, Frampton GK. Surgery for weight loss in adults . Cochrane Database Syst Rev 2014 ; ( 8 ): CD003641 . doi:10.1002/14651858.CD003641.pub4 Google Scholar OpenURL Placeholder Text WorldCat 11. Moyer VA , U.S. Preventive Services Task Force. Screening for and management of obesity in adults: U.S. Preventive Services Task Force recommendation statement . Ann Intern Med 2012 ; 157 ( 5 ): 373 – 8 , doi:10.7326/0003-4819-157-5-201209040-00475. Google Scholar PubMed OpenURL Placeholder Text WorldCat 12. Diabetes Prevention Program Research Group , Knowler WC, Fowler SEet al. . 10-year follow-up of diabetes incidence and weight loss in the Diabetes Prevention Program Outcomes Study . Lancet Lond Engl 2009 ; 374 ( 9702 ): 1677 – 86 . Google Scholar Crossref Search ADS WorldCat 13. Unick JL , Beavers D, Jakicic JMet al. .; Look AHEAD Research Group. Effectiveness of lifestyle interventions for individuals with severe obesity and type 2 diabetes: results from the Look AHEAD trial . Diabetes Care 2011 ; 34 : 2152 – 7 . Google Scholar Crossref Search ADS PubMed WorldCat 14. Look AHEAD Research Group . Eight-year weight losses with an intensive lifestyle intervention: the look AHEAD study . Obesity (Silver Spring) 2014 ; 22 ( 1 ): 5 – 13 . Crossref Search ADS PubMed WorldCat 15. Wadden TA , Volger S, Sarwer DBet al. . A two-year randomized trial of obesity treatment in primary care practice . N Engl J Med 2011 ; 365 : 1969 – 79 . Google Scholar Crossref Search ADS PubMed WorldCat 16. Carvajal R , Wadden TA, Tsai AG, Peck K, Moran CH. Managing obesity in primary care practice: a narrative review . Ann N Y Acad Sci 2013 ; 1281 : 191 – 206 . Google Scholar Crossref Search ADS PubMed WorldCat 17. Appel LJ , Clark JM, Yeh HCet al. . Comparative effectiveness of weight-loss interventions in clinical practice . N Engl J Med 2011 ; 365 : 1959 – 68 . Google Scholar Crossref Search ADS PubMed WorldCat 18. Welcome to MD-IBIS—Maryland’s Public Health Data Resource . https://ibis.health.maryland.gov/ (accessed on 18 April 2020 ). 19. Martins RK , McNeil DW. Review of motivational interviewing in promoting health behaviors . Clin Psychol Rev 2009 ; 29 : 283 – 93 . Google Scholar Crossref Search ADS PubMed WorldCat 20. Handbook of Motivational Counseling: Concepts, Approaches, and Assessment | Wiley . Wiley.com . https://www.wiley.com/en-us/Handbook+of+Motivational+Counseling%3A+Concepts%2C+Approaches%2C+and+Assessment-p-9780470092583 (accessed 18 April 2020 ). 21. West DS , DiLillo V, Bursac Z, Gore SA, Greene PG. Motivational interviewing improves weight loss in women with type 2 diabetes . Diabetes Care 2007 ; 30 : 1081 – 7 . Google Scholar Crossref Search ADS PubMed WorldCat 22. Peterson J , Repovich W, Parascand C. Accuracy of consumer grade bioelectrical impedance analysis devices compared to air displacement plethysmography . Int J Exerc Sci 2011 ; 4 : 176 – 83 . https://digitalcommons.wku.edu/ijes/vol4/iss3/2 Google Scholar OpenURL Placeholder Text WorldCat 23. Nichols J , Going S, Loftin M, Stewart D, Nowicki E, Pickrel J. Comparison of two bioelectrical impedance analysis instruments for determining body composition in adolescent girls . Int J Body Compos Res 2006 ; 4 : 153 – 60 . Google Scholar PubMed OpenURL Placeholder Text WorldCat 24. (US) NOEIEP on the Identification, Evaluation, and Treatment of Obesity in Adults . Clinical Guidelines on the Identification, Evaluation, and Treatment of Overweight and Obesity in Adults . Bethesda, MD : National Heart, Lung, and Blood Institute , 1998 . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC 25. Renjilian DA , Perri MG, Nezu AM, McKelvey WF, Shermer RL, Anton SD. Individual versus group therapy for obesity: effects of matching participants to their treatment preferences . J Consult Clin Psychol 2001 ; 69 : 717 – 21 . Google Scholar Crossref Search ADS PubMed WorldCat 26. Hall DL , Lattie EG, McCalla JR, Saab PG. Translation of the diabetes prevention program to ethnic communities in the United States . J Immigr Minor Health 2016 ; 18 : 479 – 89 . Google Scholar Crossref Search ADS PubMed WorldCat 27. Khera R , Murad MH, Chandar AKet al. . Association of pharmacological treatments for obesity with weight loss and adverse events: a systematic review and meta-analysis . JAMA 2016 ; 315 : 2424 – 34 . Google Scholar Crossref Search ADS PubMed WorldCat 28. Bray GA , Frühbeck G, Ryan DH, Wilding JP. Management of obesity . Lancet 2016 ; 387 : 1947 – 56 . Google Scholar Crossref Search ADS PubMed WorldCat 29. LeBlanc ES , Patnode CD, Webber EM, Redmond N, Rushkin M, O’Connor EA. Behavioral and pharmacotherapy weight loss interventions to prevent obesity-related morbidity and mortality in adults: updated evidence report and systematic review for the US preventive services task force . JAMA 2018 ; 320 : 1172 – 91 . Google Scholar Crossref Search ADS PubMed WorldCat 30. Bombelli M , Facchetti R, Sega Ret al. . Impact of body mass index and waist circumference on the long-term risk of diabetes mellitus, hypertension, and cardiac organ damage . Hypertension 2011 ; 58 ( 6 ): 1029 – 35 . Google Scholar Crossref Search ADS PubMed WorldCat 31. Sobiczewski W , Wirtwein M, Jarosz D, Gruchala M. Superiority of waist circumference and body mass index in cardiovascular risk assessment in hypertensive patients with coronary heart disease . Blood Press 2015 ; 24 : 90 – 5 . Google Scholar Crossref Search ADS PubMed WorldCat 32. Nazare JA , Smith J, Borel ALet al. .; INSPIRE ME IAA Investigators. Usefulness of measuring both body mass index and waist circumference for the estimation of visceral adiposity and related cardiometabolic risk profile (from the INSPIRE ME IAA study) . Am J Cardiol 2015 ; 115 : 307 – 15 . Google Scholar Crossref Search ADS PubMed WorldCat © The Author(s) 2020. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)

Journal

Family Practice – Oxford University Press

Published: Feb 4, 2021

There are no references for this article.