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

Learn More →

Are COPD self-management mobile applications effective? A systematic review and meta-analysis

Are COPD self-management mobile applications effective? A systematic review and meta-analysis www.nature.com/npjpcrm REVIEW ARTICLE OPEN Are COPD self-management mobile applications effective? A systematic review and meta-analysis 1 2✉ 2 3 2 G. Shaw , M. E. Whelan , L. C. Armitage , N. Roberts and A. J. Farmer The burden of chronic obstructive pulmonary disease (COPD) to patients and health services is steadily increasing. Self- management supported by mobile device applications could improve outcomes for people with COPD. Our aim was to synthesize evidence on the effectiveness of mobile health applications compared with usual care. A systematic review was conducted to identify randomized controlled trials. Outcomes of interest included exacerbations, physical function, and Quality of Life (QoL). Where possible, outcome data were pooled for meta-analyses. Of 1709 citations returned, 13 were eligible trials. Number of exacerbations, quality of life, physical function, dyspnea, physical activity, and self-efficacy were reported. Evidence for effectiveness was inconsistent between studies, and the pooled effect size for physical function and QoL was not significant. There was notable variation in outcome measures used across trials. Developing a standardized outcome-reporting framework for digital health interventions in COPD self-management may help standardize future research. npj Primary Care Respiratory Medicine (2020) 30:11 ; https://doi.org/10.1038/s41533-020-0167-1 INTRODUCTION outcomes. Meta-analyses to date have pooled trials investigating 9 10 10 hospital admissions , physical activity , physical function , Chronic obstructive pulmonary disease (COPD) affects the 10 11 dyspnea , and exacerbations . However, reviews to date have functional capacity of the lungs, characterized by airflow limitation used varying eligibility criteria for inclusion, excluding tablet and is commonly progressive . One in 20 adults aged over 40 computers , excluding trials with any healthcare professional years old in the United Kingdom have diagnosed COPD and it is 12 9 input , excluding trials shorter than 1 month in duration , or only projected to be the fourth leading cause of global mortality by including trials reporting hospitalization or exacerbation 2030 . Despite the preventable and treatable nature of the 9,11 events . With technologies rapidly evolving, it is also important condition , it poses a high financial burden to the healthcare to identify the effective and less effective components of current systems globally. In England, the annual direct healthcare costs of interventions to help inform future interventions, so this review COPD were estimated to be £1.5 billion in 2011, with severe will provide a detailed description of each intervention. The aim of exacerbations costing £3726 per event . There are also substantial this systematic review was to build on existing reviews by indirect and intangible costs associated with COPD, which are synthesizing and appraising evidence on the effectiveness of much harder to quantify, but include time lost from work, impact 5 mobile applications (encompassing smartphones, tablet compu- on family, and additional social and care costs . ters, and accompanying devices such as wearable sensors) in Acute exacerbations of COPD are defined as acute events people with COPD. leading to the worsening of respiratory condition beyond normal daily variation . Increased frequency of exacerbations and ongoing, progressive development of the condition itself can RESULTS significantly impact QoL and increase the risk of mortality . Initial Study selection studies incorporating technology into self-management interven- The initial search identified 1709 citations; 738 duplicates were tions for COPD patients combined phone calls with weekly visits removed. After screening titles and abstracts, 933 papers were from health professionals, and indicated that this strategy could excluded. Thirty-eight trials were assessed using full texts and 11 result in fewer exacerbation-related hospital attendances . were deemed eligible for inclusion. After screening reference lists Increasing attention to the potential for self-management has of the included trials, two additional trials were identified, highlighted the role of digital health technologies. The capabilities resulting in a total of 13 trials for inclusion (Fig. 1). of mobile device technologies have substantially increased, and applications can facilitate access to and awareness of self- Study characteristics management strategies for patients living with long-term condi- 13–24 tions such as COPD. Study characteristics are reported in Table 1. All 13 trials were Studies exploring patient experience and acceptability of apps published after 2008, with most (12 of 13) published since 2011. have shown promise , suggesting that such technology may be Trials were conducted in a number of countries and settings; 17–20 able to complement current clinical care. However, the evidence however, most were in the Netherlands or the United 13,16,22,23 14,17,18,23,25 base to support this approach is currently unclear. Several Kingdom . Five trials included fewer than 50 systematic reviews have been conducted exploring applications participants and the largest number of participants was 343 . to support self-management of COPD, but questions remain Across all 13 trials, the total number of participants was 1447. regarding their potential to improve clinical and nonclinical Participants were generally aged ≥60 years, and the proportion of 1 2 3 Exeter College, University of Oxford, Oxford, UK. Nuffield Department of Primary Care Health Sciences, University of Oxford, Oxford, UK. Bodleian Health Care Libraries, University of Oxford, Oxford, UK. email: maxine.whelan@phc.ox.ac.uk Published in partnership with Primary Care Respiratory Society UK 1234567890():,; G Shaw et al. PRISMA 2009 flow diagram Records identified through Additional records identified database searching through other sources (n = 1709) (n = 0) Records after duplicates removed (n = 971) Records screened Records excluded (n = 971) (n = 933) Full-text articles assessed Full-text articles excluded, for eligibility with reasons (n = 38) (n = 25) Conference abstracts, n = 13 Wrong intervention, n = 8 Wrong study design, n = 3 Wrong comparator, n = 1 Studies included in qualitative synthesis (n = 13) Studies included in quantitative synthesis (meta-analysis) (n = 10) Fig. 1 PRISMA flowchart. The PRISMA flowchart reporting the number of papers identified, screened, and excluded. males and females was similar within trials. One study included smartphone application, with only the intervention group receiv- 14 13 male participants only and another only included one female ing alerts, and participants were not informed of their allocation . 14,18,23,24 participant. Baseline measures of lung function were identified in Similarly, four trials were considered at high risk of bias for 13–18,20,21,25 23 15,18,25 nine trials . Study duration varied from 2 weeks to the blinding of outcome assessments, three trials were 15,16,20,22 13,16,19–23 12 months . unclear, and the remaining six trials at low risk of bias. A description of the interventions is outlined in Supplementary 13,15,17,19,21,23–25 Table 1. Eight of the interventions were smart- Primary outcome phone-based, using custom applications whereby participants 13,14,16,22,25 Five trials reported the frequency of COPD exacerba- entered COPD symptom data and received custom or automated tions that led to clinical intervention (hospitalization or managed feedback based on their responses. Healthcare professional in the community). However, only one of these trials reported involvement through active monitoring of entered data, clinical pre-intervention and post-intervention exacerbation data. One advice, or intervention on deteriorating observations was noted in 16 trial presented patient self-reported exacerbations but only post- 14,16,18–20,22,24 13–15,17–21,23–25 seven trials . Eleven trials delivered the intervention data. A summary of the main findings of the included 16,22 intervention through a smartphone and two utilized a mobile trials can be seen in Table 2. 14,18,21–23 tablet device. Five trials provided participants with a monitoring device such as a pulse oximeter and a pedometer, Other outcomes which linked to the applications to provide additional data. Physical function. Physical function was reported in five trials 15,18,20,21,25 25 (Table 3) . One trial recorded the incremental shuttle- Risk of bias within studies walking test and showed the results that were neither statistically An overview of the results for the bias assessment is presented in significant nor indicated a clinically important difference between 15,18,20,21 Fig. 2. Random sequence generation was clearly carried out in 12 intervention and control groups. The other trials used the trials, with one trial unclear on random sequence generation . Six 6-minute walk test. Only one trial recorded a significant 14,15,19,20,24,25 trials were unclear on concealment of allocation. Risk difference between the groups in the post-intervention period. of selective reporting was considered low in 12 trials with the No difference between intervention and usual care was found for remaining trial classified as having a high risk of bias. Regarding the 6-minute walk test (mean difference, 8.38 m, 95% CI, −4.40 to 19–21,23 blinding of participants to intervention, four trials were 21.17, p = 0.20; Fig. 3). The I estimate was 52% that represents 14–18,22,24,25 considered at high risk of bias, eight trials did not moderate-to-substantial heterogeneity. provide sufficient information for assessment about the degree of participant blinding, and the remaining trial was considered at Quality of life (QoL). Twelve of the 13 trials reported QoL; two of 15,21 low risk. Halpin et al. (2011) was judged to be at low risk because these trials reported two different quality-of-life measures. both control and intervention participants had access to a Across all 12 trials, 14 quality-of-life measures were reported npj Primary Care Respiratory Medicine (2020) 11 Published in partnership with Primary Care Respiratory Society UK 1234567890():,; Included Eligibility Screening Identification G Shaw et al. Published in partnership with Primary Care Respiratory Society UK npj Primary Care Respiratory Medicine (2020) 11 Table 1. An overview of study characteristics of the included 13 trials, reported as mean (SD) or median (lower quartile–upper quartile), unless otherwise stated. Author (year) Setting Sample size Age Sex (M %) Lung function (using Primary Other outcomes Duration of Duration FEV % predicted) outcome intervention of study Intervention Control Intervention Control Intervention Control Intervention Control Liu (2008) – 24 24 71.4 (1.7) 72.8 (1.3) 100 100 45.2 (3.2) 46.0 (2.8) ISWT Exacerbation, QoL 9 months 9 months Halpin (2011) Primary care 40 39 68.5 (1.5) 70.2 (1.6) 74.4 73.7 48.0 (4.0) 53.0 (3.0) Exacerbations Application ability to 4 months 4 months predict increases in exacerbation, changes in health status Chau (2012) Outpatient clinic 22 18 73.5 (6.1) 72.2 (6.1) 95.5 100 38.0 (12.9) 37.7 – User satisfaction, 2 months 2 months (16.52) QoL, and hospital admissions Nguyen (2013) Community 43 41 68.5 (11.0) 69.3 (8.0) 58.1 58.5 53.3 (20.4) 49.4 (19.8) Dyspnea 6MWT, QoL (CRQ) 12 months 12 months Pinnock (2013) Primary care and 128 128 69.4 (8.8) 68.4 (8.4) 41.4 49.2 44.0 (18.8) 40.0 (17.0) Time to the Admissions, 12 months 12 months community first hospital exacerbations, self- admission due Efficacy, QoL, anxiety, to and depression exacerbation Tabak (2014-A) Primary and 12 12 64.1 (9.0) 62.8 (7.4) 50 50 50 36 Use of 6MWT, dyspnea, 9 months 9 months secondary care (33.3–61.5) (26–53.5) application fatigue, QoL, No. of and hospitalizations, and satisfaction activity level Tabak (2014-B) Secondary care 14 16 65.2 (9.0) 67.9 (5.7) 57.1 68.8 48.7 (16.7) 56.4 (10.6) Activity level Dyspnea, fatigue, 1 month 1 month and QoL van der Primary care 65 68 57.5 (7.0) 59.2 (7.5) 47.7 45.6 –– Activity level QoL, general and 4–6 months 9 months Weegen (2015) exercise self-efficacy Vorrink (2016) Physiotherapist 84 73 62.0 (9.0) 63.0 (8.0) 50 49.3 59.0 (20.0) 53.0 (15.0) Activity level 6MWT, QoL, and BMI 6 months 12 months Demeyer (2017) Community 171 172 66.0 (8.0) 67.0 (8.0) 65 63 55.0 (20.0) 57.0 (21.0) Activity level QoL and 6MWT 12 weeks 12 weeks Farmer (2017) Primary and 110 56 69.8 (9.1) 69.8 (10.6) 61.8 60.7 –– QoL Hospital admissions, 12 months 12 months secondary care, exacerbations community Orme (2018) Secondary care 12 11 – – –– –– Feasibility and Dyspnea, fatigue, 2 weeks 2 weeks acceptability anxiety, depression, QoL, and self-efficacy Wang (2018) Secondary care 32 32 66.4 (6.2) 67.1 (6.2) 65.6 71.9 –– Self- Dyspnea 3 months 3 months management G Shaw et al. Liu (2008) LOW UNCLEAR UNCLEAR UNCLEAR HIGH LOW Halpin (2011) LOW LOW LOW LOW LOW LOW Chau (2012) LOW UNCLEAR UNCLEAR HIGH LOW LOW Nguyen (2013) UNCLEAR UNCLEAR UNCLEAR UNCLEAR LOW LOW Pinnock (2013) LOW LOW UNCLEAR LOW HIGH LOW Tabak (2014-A) LOW LOW UNCLEAR UNCLEAR HIGH HIGH Tabak (2014-B) LOW LOW UNCLEAR HIGH HIGH LOW van der Weegen (2015) LOW UNCLEAR HIGH LOW LOW LOW Vorrink (2016) LOW UNCLEAR HIGH LOW HIGH LOW Demeyer (2017) LOW LOW HIGH LOW LOW LOW Farmer (2017) LOW LOW UNCLEAR LOW LOW LOW Orme (2018) LOW LOW HIGH HIGH HIGH LOW Wang (2018) LOW UNCLEAR UNCLEAR HIGH LOW LOW Fig. 2 Risk of bias assessment. An outline of the bias assessment findings for the 13 included trials. 25 14,15,17,20,24 (Table 4). Only one trial reported the SF-12 measure, reporting a Dyspnea. Five trials reported data relating to dyspnea 14,15,20 significant difference between intervention and control post- (Supplementary Table 3). Three trials used the dyspnea 15,19 intervention. Two trials used the SF-36 measure, but these did component of the Chronic Respiratory Disease Questionnaire 21 17,24 not identify statistically significant differences. One trial reported measure, while the other two trials used the modified Medical the individual mental, functional, and symptom domains of the Research Council dyspnea scale. Only one trial reported a Chronic COPD Questionnaire. There was a significant difference statistically significant difference between groups. between the intervention and control groups in the Functional 14,17,18,20,23 CCQ measure post intervention but not in other domains. Two Fatigue. Five trials reported data concerning fatigue 17,18 14,20 trials recorded the total CCQ score, but the results were not (Supplementary Table 4). Two trials reported the fatigue 17,18 significant. The Chronic Respiratory Disease Questionnaire was component of the CRQ, two trials reported the Multidimen- 15 14,20 23 reported in full by one trial , and partially by two trials (only sional Fatigue Inventory, and one trial used the Functional reporting the emotion and mastery domains). These three trials Assessment of Chronic Illness Therapy measure. None of these reported non-significant results for these domains. Three trials reported significant improvements in the intervention arm 13,16,22 trials reported the St. George’s Respiratory Questionnaire compared with control. 21,23 and two trials reported the COPD Assessment Test measure of 17–21 QoL, but none of them showed significant differences between Physical activity. Five trials reported device-based levels of intervention and control groups. The 12 trials reporting QoL were physical activity (Supplementary Table 5). Four trials recorded assessed for inclusion for the meta-analysis, but trials that did not physical activity using accelerometers, while the remaining trial report a total or summative score were excluded, resulting in a used pedometers. Only one trial reported a statistically total of eight eligible trials (Fig. 4). The trial by Nguyen et al. (2013) significant difference in physical activity outcomes between reported two total scores reflecting QoL (Chronic Respiratory groups in the post-intervention period. Two of these five trials Disease Questionnaire and SF-36); the disease-specific scale also provided self-reported levels of physical activity, using the (Chronic Respiratory Disease Questionnaire) was included in the Moderate Physical Activity questionnaire and the Baecke meta-analysis. No difference in QoL was found between mobile Physical Activity Questionnaire . Both trials reported non- device application intervention and usual care (standardized mean significant changes from baseline. difference, −0.4 points; 95% CI, −0.86 to 0.05, p = 0.08). The I 15,16,19,23 estimate was 83% that represents considerable heterogeneity. The Self-efficacy. Four trials reported patient self-efficacy minimal clinically important differences for the back-translated (Supplementary Table 6). The employed measures focused on 15 23 19 16,19 standardized mean differences are presented in Supplementary dyspnea , falls , exercise , and self-efficacy more generally . Table 2. No trials recorded statistically significant findings. npj Primary Care Respiratory Medicine (2020) 11 Published in partnership with Primary Care Respiratory Society UK Random sequence generation Allocation concealment Blinding of participants and personnel Blinding of outcome assessment Incomplete outcome data addressed Selective reporting G Shaw et al. Table 2. A summary of the main findings for exacerbations. Author (year) sample size Type of exacerbation reported Group allocation Intervention Control Liu (2008) Intervention N = 24, Control N = 24 Managed in the community, N; p = NS F: N =4F: N = 26 Leading to hospitalization, N; p = NS F: N =2F: N = 22 Halpin (2011) Intervention N = 40, Control N = 39 Clinical exacerbation frequency, mean (SD); p = NS F: 0.95 (1.71) F: 1.17 (1.81) Chau (2012) Intervention N = 22, Control N = 18 Managed in the community, N; p = NS F: N =7F: N = 3 Leading to hospitalization, mean (SD) or N; p = NS B: 2.41 (1.57) B: 2.89 (2.32) F: N = 7 F: N = 3 Pinnock (2013) Intervention N = 8, Control N = 8 Leading to hospitalization, mean (SD); p = NS F: 1.5 (2.3) F: 1.3 (1.8) Managed in the community, mean (SD); p= NS F: 15 (12.7) F: 12.8 (11.8) Farmer (2017) Intervention N = 110, Control N = 56 Unspecified, median (IQR); p = NS F: 1 (0–2) F: 1 (0–3) B baseline, F follow-up, IQR interquartile range, N number of participants, NS nonsignificant, SD standard deviation. Table 3. A summary of the main findings for physical function. Author (year) sample size Type of physical function assessment reported Group allocation Intervention Control Liu (2008) Intervention N = 24, Control N = 24 ISWT (m), mean (SD); p = NS B: 255.8 (200.9) B: 262.9 (88.8) F: 306.7 (103.9) F: 237.8 (60.7) Nguyen (2013) Intervention N = 43, Control N = 41 6MWT (m), mean (SD); p = NS B: 400.5 (100.0) B: 398.0 (99.7) F: 431.3 (124.4) F: 406.6 (125.0) Tabak (2014-A) Intervention N = 12, Control N = 12 6MWT (m), mean (SD); p = NS B: 409.5 (102.2) B: 300.1 (116.4) F: 412 (134.1) F: 312.4 (152.4) Vorrink (2016) Intervention N = 84, Control N = 73 6MWT (m), mean (SD) or median (IQR); p= NS B: 456 (128.3) B: 461 (73.3) C: 0.8 (−8.8 to 10.3) C: 4 (−2.4 to 10.3) Demeyer (2017) Intervention N = 171, Control N = 172 6MWT (m), mean (SD); p = 0.009 B: 444 (106) B: 450 (106) F: 457 (108) F: 449 (118) 6MWT 6-minute walk test, B baseline, C change, F follow-up, IQR interquartile range, m meters, NS nonsignificant, SD standard deviation. 16,23 Anxiety and depression. Two trials reported anxiety and durations. However, the present review was prospectively depression, using the Hospital Anxiety and Depression Scale registered on a database of systematic reviews and included (HADS), and no statistically significant differences were observed. trials published in any language in several databases from inception. A sensitive search strategy was developed, and screening of citations was performed independently, minimizing DISCUSSION the risk of bias at review level. The review was inclusive of a broad This systematic review provided a comprehensive description and range of outcome measures, contributing to its comprehensive summarized the findings of mobile device application interven- nature. tions for COPD self-management. The interventions identified Although exacerbations can negatively impact QoL and were heterogeneous in nature, including the components (such as increase mortality , only five of the included trials reported the inclusion of periphery devices), the degree and frequency of exacerbations. Only one of these trials reported pre-intervention involvement of healthcare professionals, and frequency of and post-intervention exacerbation frequency , and exacerba- participant-performed measurements and data entry. It remains tions were reported using a wide range of metrics, including those unclear whether mobile device applications are more effective at exacerbations managed in the community and leading to preventing exacerbations when compared with usual care. hospitalization. An 80% reduction in likelihood of having an As only published trials were eligible for inclusion, there is exacerbation has been demonstrated previously in a meta-analysis potential for publication bias within the review. Also, the risk comparing a smartphone intervention with usual care . However, assessment bias tool was challenging to implement because the meta-analysis showed moderate heterogeneity in this blinding of participants in digital health interventions where the healthcare professional contact, in part possible because of the comparator is usual care may not be feasible to implement. In small sample size of the three trials pooled. It is unclear if addition, our ability to pool further outcome measures using reporting the number of contacts with healthcare professionals is meta-analysis was limited, given the variety of outcome measures a suitable outcome measure to represent COPD exacerbations; used across the trials. There are also limitations to interpreting digital interventions can offer an alternate means of contacting a summary estimates from pooled data, particularly when the healthcare professional, impacting the accuracy of assessing design of the studies, scales used to assess effectiveness, and exacerbation frequency in this way. With prevention and manage- interventions tested are heterogeneous and use varying follow-up ment of exacerbations being a key feature of COPD care, and an Published in partnership with Primary Care Respiratory Society UK npj Primary Care Respiratory Medicine (2020) 11 G Shaw et al. Fig. 3 Physical function forest plot. Forest plot of the effect of mobile device applications on physical function. Table 4. A summary of the main findings for quality of life. Author (year) sample size Form of QoL assessment reported Group allocation Intervention Control Liu (2008) Intervention N = 24, SF-12 PCS, mean (SD); p < 0.001 B: 38.7 (8.82) B: 40.1 (6.37) Control N = 24 F: 47.9 (7.35) F: 30.9 (10.78) Halpin (2011) Intervention N = 40, SGRQ, mean (SD); p = NS B: 52.4 (16.44) B: 53.6 (14.99) Control N = 39 F: 49.7 (15.18) F: 51.5 (14.99) Chau (2012) Intervention N = 22, CRQ (Emotion), mean (SD); p = NS B: 4.84 (1.47) B: 5.24 (1.42) Control N = 18 F: 4.92 (1.40) F: 5.61 (1.17) CRQ (Mastery), mean (SD); p= NS B: 4.60 (1.43) B: 4.94 (1.16) F: 4.61 (1.62) F: 4.88 (1.27) Nguyen (2013) Intervention N = 43, CRQ (Total), mean (SD); p = NS B: 96.4 (19.91) B: 96.2 (19.76) Control N = 41 F: 104.8 (23.92) F: 98.4 (24.34) Pinnock (2013) Intervention N = 8, SGRQ, mean (SD); p = NS B: 68.6 (16.6) B: 68.0 (15.2) Control N = 8 F: 68.2 (16.3) F: 67.3 (17.3) Tabak (2014-A) Intervention N = 12, CCQ (Total), mean (SD); p = NS B: 2.0 (0.90) B: 2.7 (0.94) Control N = 12 F: 1.8 (0.83) F: 2.3 (0.90) Tabak (2014-B) Intervention N = 14, CCQ (Total), mean (SD); p = NS B: 2.0 (0.8) B: 1.8 (1.0) Control N = 16 F: 1.7 (0.5) F: 1.8 (0.6) van der Weegen (2015) Intervention SF-36 (Physical), mean (SD); p = NS B: 42.5 (11.1) B: 45.8 (9.4) N = 65, Control N = 68 F: 44.1 (9.5) F: 45.8 (9.5) SF-36 (Mental), mean (SD); p = NS B: 48.2 (10.3) B: 50.1 (9.5) F: 48.3 (11.7) F: 50.3 (8.3) Vorrink (2016) Intervention N = 84, CRQ (Emotion), mean (SD) or median (IQR); p = NS B: 5.0 (1.1) B: 4.8 (1.2) Control N = 73 C: 0.09 (−0.07 to 0.24) C: 0.19 (−0.31 to 0.11) CRQ (Mastery), mean (SD) or median (IQR); p = NS B: 5.4 (1.1) B: 5.3 (1.1) C: −0.1 (−0.31 to 0.11) C: −0.23 (−0.39 to −0.06) Demeyer (2017) Intervention N = 171, CCQ (Mental), median (IQR); p = NS B: 1 (0–2.5) B: 1 (0–2) Control N = 172 F: 1 (0–2.5) F: 1 (0–2) CCQ (Functional), median (IQR); p= 0.026 B: 1.5 (1–2.75) B: 1.5 (0.75– 2.75) F: 1.5 (1–2.75) F: 1.75 (0.75– 2.75) CCQ (Symptoms), median (IQR); p = NS B: 1.75 (1.25– 2.5) B: 1.75 (1.5–2.75) F: 1.75 (1.25– 2.5) F: 2 (1.25– 2.75) Farmer (2017) Intervention N = 110, SGRQ, mean (SD); p = NS B: 56.4 (19.7) B: 55.5 (16.2) Control N = 56 F: 56.9 (19.5) F: 56.8 (20.9) Orme (2018) Intervention N = 12, CAT, mean (SD); p = NS B: 22.6 (4.4) B: 24.5 (9.7) Control N = 11 F: 21.6 (5) F: 23.8 (11.1) B baseline, C change, CCQ Clinical COPD Questionnaire, CRQ Chronic Respiratory Disease Questionnaire, F follow-up, SGRQ St. George's Respiratory Questionnaire, SF short form, CAT COPD Assessment Test, IQR interquartile range, SD standard deviation, NS nonsignificant. 28–30 increasing interest in predicting the onset of exacerbations , apparent lack of impact may be from the small size of the future trials are recommended to consider this when reporting studies, with 8 of the 13 trials reporting a sample size of fewer 13–15,17,18,23,25,31 exacerbations to more accurately quantify the impact of digital than 100 participants . In addition, the extent to interventions on this important clinical outcome. which the measures used in these studies were sensitive to The trials identified in this systematic review do not yet provide change is unclear. strong evidence for implementing mobile digital health interven- Hanlon et al. conducted a metareview of telehealth trials across tions for COPD. Only four trials reported clinical differences multiple health conditions, including COPD, diabetes, cancer, and between the intervention and control groups, and these heart failure . Their findings suggest that the evidence base is differences were in a range of outcomes, including physical more developed in diabetes and heart failure and more intensive 19,21,25,31 function, QoL, physical activity, and dypsnea . This and multifaceted interventions associated with greater npj Primary Care Respiratory Medicine (2020) 11 Published in partnership with Primary Care Respiratory Society UK G Shaw et al. Fig. 4 QoL forest plot. Forest plot of the effect of mobile device applications on quality of life. improvements in asthma, diabetes, and heart failure. Building on to standardize the outcomes used in this area of research are encouraged to increase the comparability of future trials. published reviews focused on COPD, our findings also report on QoL, self-efficacy, fatigue, anxiety, and depression, as well as exacerbations, physical function, and physical activity. In addition, METHODS we provide an in-depth description of the interventions within the Registration included trials. The results from our pooled data meta-analysis do not identify a The review was registered on the International Prospective statistically significant effect on measures of physical function or Register of Systematic Reviews (PROSPERO reference number: QoL. Previous meta-analyses have identified no differences in CRD42019124232). 10 10 physical function (using the 6-minute walk test) , dyspnea , and average days of hospitalization , but have noted that the Eligibility criteria intervention arm was favored for physical activity and a lower Randomized controlled trials of adults with a clinical diagnosis of risk of hospital admissions . COPD were included where the intervention group received a Looking beyond the effectiveness of the intervention for clinical mobile device application to support their COPD self- outcomes, it is possible that there are efficiency and organiza- management. A mobile device application was defined as a tional benefits of digital and telehealth care compared with more contained program that served a specific function relating to traditional models of care. None of the studies included in this COPD and personal health on a portable, electronic device review reported service outcomes. (including smartphones and tablet computers). This definition is in The trial interventions identified in our review focused on 11,12 line with previous systematic reviews on the topic . For the varying components of COPD self-management, including mon- purpose of inclusion, self-management was defined as patient itoring symptoms, encouraging lifestyle changes (such as management of their personal symptoms and medication regimes increases in physical activity or exercise), and hosting educational related to the condition, as well as coping with the emotional and material concerning COPD. Some of the trials explored ease of use, 35,36 lifestyle impacts of the condition . Studies were eligible where feasibility, and accessibility of the technologies. Aligning with this the comparator group received usual care only. Outcomes heterogeneity is the variety of outcome measures used to assess included but were not restricted to exacerbations, QoL, physical the effectiveness of the intervention. This review highlights the function, physical activity, and dyspnea. number of outcome measures used and variation in which the tool was used for data collection between studies. Information sources and search Our findings and the challenges encountered in synthesizing Medline, EMBASE, Cochrane Library, CINAHL, and the Science the evidence from these trials highlight the importance of Citation Index were searched from inception to 12th April 2019 developing a minimum and standardized set of clinically following the methods recommended by the Preferred Reporting important core-outcome measures to allow comparison of trials Items for Systematic Reviews and Meta-Analyses guidelines . Full involving people with COPD. This would be in line with minimum search strategies are included in Supplementary Methods. The reporting guidelines for other areas of clinical speciality, including search algorithm focused on keywords relating to ‘COPD’, ‘mobile rheumatology . In practice, the use of mobile device applications phone application’, and ‘self-management’ and included inter- to support self-management may have some negative effects. For ventions with or without healthcare professional input. example, a patient might be falsely reassured if they feel their data were being monitored by a healthcare professional. On the other Study selection hand, the data can supplement routine care with information about variation in symptoms and clinical markers of the condition. The resulting citations were imported into the web-based From a policy perspective, the economic cost of telehealth for Covidence systematic review software (Veritas Health Innovation, chronic disease is high (£92,000/QALY), which restricts its Melbourne, Australia). Screening of titles and abstracts was implementation in the majority of healthcare settings . completed by two authors independently (G.S. and M.W.). In the In conclusion, this systematic review demonstrates that there event of disagreement, two further reviewers (L.A. and A.F.) are a number of trials being conducted in this area of COPD. decided their eligibility. Subsequently, full-text screening was However, there is insufficient evidence to date to suggest that conducted by two authors independently (G.S. and M.W.). Any mobile device applications are effective for the self-management disagreements were resolved following discussion with the other of COPD over usual care. This may in part be due to a limited reviewers (L.A. and A.F.). The reference lists of the included trials ability for data to be pooled, owing to marked variation in were also screened to identify any additional potentially eligible methodology and reporting of outcome measures. Future efforts trials. Published in partnership with Primary Care Respiratory Society UK npj Primary Care Respiratory Medicine (2020) 11 G Shaw et al. Data collection process was used to help control for differences between trials, and to limit the impact of heterogeneity. Trials were weighted by sample Extraction forms were used to capture the following data: lead size, and 95% confidence intervals were reported around point author, year, country, trial setting, sample size, age, sex, lung estimates. Measures were selected for inclusion if they were function, primary and secondary outcomes, duration of interven- reported by at least three trials to align with the recent Cochrane tion and study, as well as the main findings. Data extraction was review . For continuous data with consistent units of measure- completed independently by two authors (G.S. and M.W.), and any ments (such as the 6-minute walk performance in meters), the disagreements were resolved through discussion. When data were mean difference in change between baseline and follow-up not directly identifiable within text or tables, authors were measurements was calculated. In instances where continuous data contacted or Microsoft Paint (Microsoft, Washington, USA) was were inconsistent between trials (i.e., multiple questionnaires with used to extract values from graphs. The graphical summaries were varying scales used to measure QoL), the standardized mean captured by screenshot and copy-pasted into the software. No differences between timepoints were calculated. Back-translation correction for rotation was required. Horizontal lines were inserted of the standardized mean difference for each scale was conducted across from the center of the datapoints of interest to the point of to the original scale, to present a mean difference for each QoL intersection on the y-axis. The y-axis was segmented into smaller instrument to give information of the clinical significance of this increments, marked by adding small lines to the axis, until a value difference. Where change in standard deviation was not reported could be extracted to 1 decimal place. The values were extracted by individual trials, the standard deviation for changes from from the original y-axis scale, meaning the x and y positions were baseline was imputed by calculating a correlation coefficient from not translated. Two authors (G.S. and M.W.) independently looked trials reporting a change in standard deviation. If the data were at the graphs to identify the value of interest. In the event any not reported, authors were contacted to access this information. disagreements were identified, G.S. and M.W. reassessed the The I statistic was used to estimate heterogeneity. Cochrane graphs and agreed on a value. recommendations for interpreting the I statistic are as follows: We subsequently replicated the data extraction using web plot 30–60% may represent moderate heterogeneity, 50–90% may digitizer software (Automeris version 3.9, https://automeris.io/ represent substantial heterogeneity, and 75–100% may represent WebPlotDigitizer/). The graphical summaries were captured by considerable heterogeneity . No funnel plot was produced as it is screenshot and saved as a PNG file before being uploaded to the not recommended for meta-analyses with fewer than 10 trials . web-based plot digitizer software. No correction for rotation was required. Once uploaded, two anchoring points were assigned to each axis: the highest and lowest value on the y-axis and baseline Reporting summary and follow-up for the x-axis. Values reflecting these anchoring Further information on research design is available in the Nature points were declared. The datapoints were selected using the Research Reporting Summary linked to this article. center of each point to 14 decimal places, and the acquired data were recorded in the form of coordinates that aligned with the scales in the original graphs. DATA AVAILABILITY All data generated or analyzed during this study are included in this published article Risk of bias assessment and Supplementary Material files. The included trials were assessed for potential bias at study level using the Cochrane risk of bias tool . Two authors (G.S. and M.W.) CODE AVAILABILITY independently completed the assessment of bias, and any No custom code or mathematical algorithm were used. disagreements were resolved through discussion with the other reviewers (L.A. and A.F.). Received: 17 July 2019; Accepted: 28 February 2020; Synthesis of data The results were converted to mean (standard deviation) when possible; otherwise data were reported as median (lower to upper quartile). A pragmatic decision was made to include outcome REFERENCES measures reported by four or more trials in the main table and 1. World Health Organisation. The Global Burden of Disease. (2008) https://www.who. those reported less frequently in the text. Where the duration of int/healthinfo/global_burden_disease/GBD_report_2004update_full.pdf?ua=1. intervention period and study duration differed, data were 2. Mathers, C. D. & Loncar, D. Projections of global mortality and burden of disease extracted for the end of the observation period. Outcomes were from 2002 to 2030. PLoS Med. 3, e442 (2006). grouped together where different measures were used, for 3. Global Initiative for Chronic Obstructive Lung Disease. Global Strategy for the example, where different scales for QoL measurement were used. Diagnosis, Management and Prevention of Chronic Obstructive Pulmonary Disease. 2018 report (2018). The total scores from the QoL measurement tools were extracted 4. McLean, S. et al. Projecting the COPD population and costs in England and when these were reported; otherwise individual component Scotland: 2011 to 2030. Sci. Rep. 6,1–10 (2016). scores were extracted. Similarly, exacerbations that were treated 5. Foundation, B. L. Estimating the Economic Burden Of Respiratory Illness in the UK blf in the community were grouped, to include self-reported Estimating the Economic Burden of Respiratory Illness in the UK, British Lung exacerbations (where a participant may have initiated a rescue Foundation (2014). pack), alongside exacerbations that were managed by primary 6. Ståhl, E. et al. Health-related quality of life is related to COPD disease severity. care teams. Measures of physical activity were included in the Health Qual. Life Outcomes 3, 56 (2005). summary table if these were objectively measured; self-report of 7. Bourbeau, J. et al. Reduction of hospital utilization in patients with chronic physical activity was not included. obstructive pulmonary disease: a disease-specific self-management intervention. Arch. Intern. Med. 163, 585–591 (2003). 8. Williams, V., Price, J., Hardinge, M., Tarassenko, L. & Farmer, A. Using a mobile Synthesis of results health application to support self-management in COPD: a qualitative study. Br. J. Meta-analysis was carried out using Review Manager (Review Gen. Pract. 64, e392–e400 (2014). 9. Yang, F., Wang, Y., Yang, C., Hu, H. & Xiong, Z. Mobile health applications in self- Manager [RevMan] version 5.3, Cochrane Collaboration, Copenha- management of patients with chronic obstructive pulmonary disease: a sys- gen, Denmark). A difference-in-difference random effect analysis npj Primary Care Respiratory Medicine (2020) 11 Published in partnership with Primary Care Respiratory Society UK G Shaw et al. tematic review and meta-analysis of their efficacy. BMC Pulm. Med. 18, 147 (2018). 34. Henderson, C. et al. Cost effectiveness of telehealth for patients with long term 10. Lundell, S., Holmner, Å., Rehn, B., Nyberg, A. & Wadell, K. Telehealthcare in COPD: conditions (Whole Systems Demonstrator telehealth questionnaire study): nested a systematic review and meta-analysis on physical outcomes and dyspnea. Respir. economic evaluation in a pragmatic, cluster randomised controlled trial. BMJ 22, Med. 109,11–26 (2015). 346 (2013). 11. Alwashmi, M. et al. The effect of smartphone interventions on patients with 35. Barlow, J., Wright, C., Sheasby, J., Turner, A. & Hainsworth, J. Self-management chronic obstructive pulmonary disease exacerbations: a systematic review and approaches for people with chronic conditions: a review. Patient Educ. Couns. 48, meta-analysis. JMIR mHealth uHealth 4, e105 (2016). 177–187 (2002). 12. McCabe, C., McCann, M. & Brady, A. M. Computer and mobile technology inter- 36. Glasgow, R. E., Davis, C. L., Funnell, M. M. & Beck, A. Implementing practical ventions for self-management in chronic obstructive pulmonary disease. interventions to support chronic illness self-management. Jt. Comm. J. Qual. Saf. Cochrane Database Syst. Rev. 5, CD011425 (2017). 29, 563–574 (2003). 13. MG Halpin, D. et al. A randomised controlled trial of the effect of automated 37. Moher, D., Liberati, A., Tetzlaff, J. & Altman, D. G., PRISMA Group. Preferred interactive calling combined with a health risk forecast on frequency and severity reporting items for systematic reviews and meta-analyses: the PRISMA statement. of exacerbations of COPD assessed clinically and using EXACT PRO. Prim. Care PLoS Med. 6, e1000097 (2009). Respir. J. 20, 324–331 (2011). 38. Higgins, J. P. T., Thomas, J., Chandler, J., Cumpston, M., Li, T., Page, M. J. & Welch, 14. Chau, J. P.-C. et al. A feasibility study to investigate the acceptability and potential V. A. (eds) Cochrane Handbook for Systematic Reviews of Interventions version effectiveness of a telecare service for older people with chronic obstructive 6.0 (updated July 2019). Cochrane. (2019). Available from www.training.cochrane. pulmonary disease. Int. J. Med. Inform. 81, 674–682 (2012). org/handbook. 15. Nguyen, H. Q. et al. Internet-based dyspnea self-management support for patients 39. Cochrane Collaboration. 9.5.2 Identifying and Measuring Heterogeneity. https:// with chronic obstructive pulmonary disease. J. Pain Symptom Manag. 46,43–55 (2013). handbook-5-1.cochrane.org/chapter_9/9_5_2_identifying_and_measuring_ 16. Pinnock, H. et al. Effectiveness of telemonitoring integrated into existing clinical heterogeneity.htm. Accessed 30 Oct 2019. services on hospital admission for exacerbation of chronic obstructive pulmonary 40. Sterne, J. A. C. et al. Recommendations for examining and interpreting funnel plot disease: researcher blind, multicentre, randomised controlled trial. BMJ 347, asymmetry in meta-analyses of randomised controlled trials. BMJ 343, d4002 f6070 (2013). (2011). 17. Tabak, M., Vollenbroek-Hutten, M. M., van der Valk, P. D., van der Palen, J. & Hermens, H. J. A telerehabilitation intervention for patients with chronic obstructive pulmonary disease: a randomized controlled pilot trial. Clin. Rehabil. 28,582–591 (2014). ACKNOWLEDGEMENTS 18. Tabak, M., Brusse-Keizer, M., van der Valk, P., Hermens, H. & Vollenbroek-Hutten, The authors would like to thank Ms. Ran Xu for her support in translating one of M. A telehealth program for self-management of COPD exacerbations and pro- the papers. A.J.F. is a NIHR Senior Investigator. This project is supported by the motion of an active lifestyle: a pilot randomized controlled trial. Int. J. Chron. NIHR Oxford Biomedical Research Centre. The views expressed are those of the Obstruct. Pulmon. Dis. 9, 935–944 (2014). authors and not necessarily those of the NIHR or the Department of Health and 19. van der Weegen, S. et al. It’s life! Mobile and web-based monitoring and feedback Social Care. tool embedded in primary care increases physical activity: a cluster randomized controlled trial. J. Med. Internet Res. 17, e184 (2015). AUTHOR CONTRIBUTIONS 20. Vorrink, S. N. W., Kort, H. S. M., Troosters, T., Zanen, P. & Lammers, J.-W. J. Efficacy of an mHealth intervention to stimulate physical activity in COPD patients after G.S. made substantial contributions to the conception of the work, acquisition, pulmonary rehabilitation. Eur. Respir. J. 48, 1019–1029 (2016). analysis, and interpretation of data for the work, drafted the work, approved the 21. Demeyer, H. et al. Physical activity is increased by a 12-week semiautomated final version to be published, and agrees to be accountable for all aspects of the telecoaching programme in patients with COPD: a multicentre randomised work. M.E.W. made substantial contributions to the acquisition, analysis, and controlled trial. Thorax https://doi.org/10.1136/thoraxjnl-2016-209026 (2015). interpretation of data for the work, as well as revised the work critically for 22. Farmer, A. et al. Self-management support using a digital health system com- important intellectual content, approved the final version to be published, and pared with usual care for chronic obstructive pulmonary disease: randomized agrees to be accountable for all aspects of the work. L.C.A. made substantial controlled trial. J. Med. Internet Res. 19, e144 (2017). contributions to the acquisition, analysis, and interpretation of data for the work, 23. Orme, M. W. et al. Findings of the chronic obstructive pulmonary disease-sitting as well as revised the work critically for important intellectual content, approved and exacerbations trial (COPD-SEAT) in reducing sedentary time using wearable the final version to be published, and agrees to be accountable for all aspects of and mobile technologies with educational support: randomized controlled fea- the work. A.J.F. made substantial contributions to the conception of the work and sibility trial. JMIR mHealth uHealth 6, e84 (2018). interpretation of data, revised the work critically for important intellectual content, 24. Wang, T. & Yang, Q. Effect study on improving self-management of patients with approved the final version to be published, and agrees to be accountable for all chronic obstructive pulmonary disease based on APP. Chin. Nurs. Res. 32, aspects of the work. N.R. made substantial contributions to the acquisition of data 3121–3124 (2018). for the work, as well as revised the work critically for important intellectual content, approved the final version to be published, and agrees to be accountable 25. Liu, W.-T. et al. Efficacy of a cell phone-based exercise programme for COPD. Eur. for all aspects of the work. Respir. J. 32, 651–659 (2008). 26. Seemungal, T. A. R. et al. Effect of exacerbation on quality of life in patients with chronic obstructive pulmonary disease. Am. J. Respir. Crit. Care Med. 157, 1418–1422 (1998). COMPETING INTERESTS 27. Soler-Cataluña, J. J. et al. Severe acute exacerbations and mortality in patients The authors declare no competing interests. with chronic obstructive pulmonary disease. Thorax 60, 925–931 (2005). 28. Miłkowska-Dymanowska, J., Białas, A. J., Obrębski, W., Górski, P. & Piotrowski, W. J. A pilot study of daily telemonitoring to predict acute exacerbation in chronic ADDITIONAL INFORMATION obstructive pulmonary disease. Int. J. Med. Inform. 116,46–51 (2018). Supplementary information is available for this paper at https://doi.org/10.1038/ 29. Shah, S. A. et al. Personalized alerts for patients with COPD using pulse oximetry s41533-020-0167-1. and symptom scores. In 2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 3164–3167 (IEEE, 2014). Correspondence and requests for materials should be addressed to M.E.W. 30. Fernandez-Granero, M. A., Sanchez-Morillo, D. & Leon-Jimenez, A. An artificial intelligence approach to early predict symptom-based exacerbations of COPD. Reprints and permission information is available at http://www.nature.com/ Biotechnol. Biotechnol. Equip. 32, 778–784 (2018). reprints 31. Zhang, Q. et al. Disease knowledge level is a noteworthy risk factor of anxiety and depression in patients with chronic obstructive pulmonary disease: a cross- Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims sectional study. BMC Pulm. Med. 14, 92 (2014). in published maps and institutional affiliations. 32. Hanlon, P. et al. Telehealth interventions to support self-management of long-term conditions: a systematic metareview of diabetes, heart failure, asthma, chronic obstructive pulmonary disease, and cancer. J. Med. Internet Res. 19, e172 (2017). 33. Tugwell, P. et al. OMERACT: an international initiative to improve outcome measurement in rheumatology. Trials 8, 38 (2007). Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, Published in partnership with Primary Care Respiratory Society UK npj Primary Care Respiratory Medicine (2020) 11 G Shaw et al. adaptation, distribution and reproduction in any medium or format, as long as you give regulation or exceeds the permitted use, you will need to obtain permission directly appropriate credit to the original author(s) and the source, provide a link to the Creative from the copyright holder. To view a copy of this license, visit http://creativecommons. Commons license, and indicate if changes were made. The images or other third party org/licenses/by/4.0/. material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the © The Author(s) 2020 article’s Creative Commons license and your intended use is not permitted by statutory npj Primary Care Respiratory Medicine (2020) 11 Published in partnership with Primary Care Respiratory Society UK http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png npj Primary Care Respiratory Medicine Springer Journals

Are COPD self-management mobile applications effective? A systematic review and meta-analysis

Loading next page...
 
/lp/springer-journals/are-copd-self-management-mobile-applications-effective-a-systematic-55vNMmtMBm

References (51)

Publisher
Springer Journals
Copyright
Copyright © The Author(s) 2020
eISSN
2055-1010
DOI
10.1038/s41533-020-0167-1
Publisher site
See Article on Publisher Site

Abstract

www.nature.com/npjpcrm REVIEW ARTICLE OPEN Are COPD self-management mobile applications effective? A systematic review and meta-analysis 1 2✉ 2 3 2 G. Shaw , M. E. Whelan , L. C. Armitage , N. Roberts and A. J. Farmer The burden of chronic obstructive pulmonary disease (COPD) to patients and health services is steadily increasing. Self- management supported by mobile device applications could improve outcomes for people with COPD. Our aim was to synthesize evidence on the effectiveness of mobile health applications compared with usual care. A systematic review was conducted to identify randomized controlled trials. Outcomes of interest included exacerbations, physical function, and Quality of Life (QoL). Where possible, outcome data were pooled for meta-analyses. Of 1709 citations returned, 13 were eligible trials. Number of exacerbations, quality of life, physical function, dyspnea, physical activity, and self-efficacy were reported. Evidence for effectiveness was inconsistent between studies, and the pooled effect size for physical function and QoL was not significant. There was notable variation in outcome measures used across trials. Developing a standardized outcome-reporting framework for digital health interventions in COPD self-management may help standardize future research. npj Primary Care Respiratory Medicine (2020) 30:11 ; https://doi.org/10.1038/s41533-020-0167-1 INTRODUCTION outcomes. Meta-analyses to date have pooled trials investigating 9 10 10 hospital admissions , physical activity , physical function , Chronic obstructive pulmonary disease (COPD) affects the 10 11 dyspnea , and exacerbations . However, reviews to date have functional capacity of the lungs, characterized by airflow limitation used varying eligibility criteria for inclusion, excluding tablet and is commonly progressive . One in 20 adults aged over 40 computers , excluding trials with any healthcare professional years old in the United Kingdom have diagnosed COPD and it is 12 9 input , excluding trials shorter than 1 month in duration , or only projected to be the fourth leading cause of global mortality by including trials reporting hospitalization or exacerbation 2030 . Despite the preventable and treatable nature of the 9,11 events . With technologies rapidly evolving, it is also important condition , it poses a high financial burden to the healthcare to identify the effective and less effective components of current systems globally. In England, the annual direct healthcare costs of interventions to help inform future interventions, so this review COPD were estimated to be £1.5 billion in 2011, with severe will provide a detailed description of each intervention. The aim of exacerbations costing £3726 per event . There are also substantial this systematic review was to build on existing reviews by indirect and intangible costs associated with COPD, which are synthesizing and appraising evidence on the effectiveness of much harder to quantify, but include time lost from work, impact 5 mobile applications (encompassing smartphones, tablet compu- on family, and additional social and care costs . ters, and accompanying devices such as wearable sensors) in Acute exacerbations of COPD are defined as acute events people with COPD. leading to the worsening of respiratory condition beyond normal daily variation . Increased frequency of exacerbations and ongoing, progressive development of the condition itself can RESULTS significantly impact QoL and increase the risk of mortality . Initial Study selection studies incorporating technology into self-management interven- The initial search identified 1709 citations; 738 duplicates were tions for COPD patients combined phone calls with weekly visits removed. After screening titles and abstracts, 933 papers were from health professionals, and indicated that this strategy could excluded. Thirty-eight trials were assessed using full texts and 11 result in fewer exacerbation-related hospital attendances . were deemed eligible for inclusion. After screening reference lists Increasing attention to the potential for self-management has of the included trials, two additional trials were identified, highlighted the role of digital health technologies. The capabilities resulting in a total of 13 trials for inclusion (Fig. 1). of mobile device technologies have substantially increased, and applications can facilitate access to and awareness of self- Study characteristics management strategies for patients living with long-term condi- 13–24 tions such as COPD. Study characteristics are reported in Table 1. All 13 trials were Studies exploring patient experience and acceptability of apps published after 2008, with most (12 of 13) published since 2011. have shown promise , suggesting that such technology may be Trials were conducted in a number of countries and settings; 17–20 able to complement current clinical care. However, the evidence however, most were in the Netherlands or the United 13,16,22,23 14,17,18,23,25 base to support this approach is currently unclear. Several Kingdom . Five trials included fewer than 50 systematic reviews have been conducted exploring applications participants and the largest number of participants was 343 . to support self-management of COPD, but questions remain Across all 13 trials, the total number of participants was 1447. regarding their potential to improve clinical and nonclinical Participants were generally aged ≥60 years, and the proportion of 1 2 3 Exeter College, University of Oxford, Oxford, UK. Nuffield Department of Primary Care Health Sciences, University of Oxford, Oxford, UK. Bodleian Health Care Libraries, University of Oxford, Oxford, UK. email: maxine.whelan@phc.ox.ac.uk Published in partnership with Primary Care Respiratory Society UK 1234567890():,; G Shaw et al. PRISMA 2009 flow diagram Records identified through Additional records identified database searching through other sources (n = 1709) (n = 0) Records after duplicates removed (n = 971) Records screened Records excluded (n = 971) (n = 933) Full-text articles assessed Full-text articles excluded, for eligibility with reasons (n = 38) (n = 25) Conference abstracts, n = 13 Wrong intervention, n = 8 Wrong study design, n = 3 Wrong comparator, n = 1 Studies included in qualitative synthesis (n = 13) Studies included in quantitative synthesis (meta-analysis) (n = 10) Fig. 1 PRISMA flowchart. The PRISMA flowchart reporting the number of papers identified, screened, and excluded. males and females was similar within trials. One study included smartphone application, with only the intervention group receiv- 14 13 male participants only and another only included one female ing alerts, and participants were not informed of their allocation . 14,18,23,24 participant. Baseline measures of lung function were identified in Similarly, four trials were considered at high risk of bias for 13–18,20,21,25 23 15,18,25 nine trials . Study duration varied from 2 weeks to the blinding of outcome assessments, three trials were 15,16,20,22 13,16,19–23 12 months . unclear, and the remaining six trials at low risk of bias. A description of the interventions is outlined in Supplementary 13,15,17,19,21,23–25 Table 1. Eight of the interventions were smart- Primary outcome phone-based, using custom applications whereby participants 13,14,16,22,25 Five trials reported the frequency of COPD exacerba- entered COPD symptom data and received custom or automated tions that led to clinical intervention (hospitalization or managed feedback based on their responses. Healthcare professional in the community). However, only one of these trials reported involvement through active monitoring of entered data, clinical pre-intervention and post-intervention exacerbation data. One advice, or intervention on deteriorating observations was noted in 16 trial presented patient self-reported exacerbations but only post- 14,16,18–20,22,24 13–15,17–21,23–25 seven trials . Eleven trials delivered the intervention data. A summary of the main findings of the included 16,22 intervention through a smartphone and two utilized a mobile trials can be seen in Table 2. 14,18,21–23 tablet device. Five trials provided participants with a monitoring device such as a pulse oximeter and a pedometer, Other outcomes which linked to the applications to provide additional data. Physical function. Physical function was reported in five trials 15,18,20,21,25 25 (Table 3) . One trial recorded the incremental shuttle- Risk of bias within studies walking test and showed the results that were neither statistically An overview of the results for the bias assessment is presented in significant nor indicated a clinically important difference between 15,18,20,21 Fig. 2. Random sequence generation was clearly carried out in 12 intervention and control groups. The other trials used the trials, with one trial unclear on random sequence generation . Six 6-minute walk test. Only one trial recorded a significant 14,15,19,20,24,25 trials were unclear on concealment of allocation. Risk difference between the groups in the post-intervention period. of selective reporting was considered low in 12 trials with the No difference between intervention and usual care was found for remaining trial classified as having a high risk of bias. Regarding the 6-minute walk test (mean difference, 8.38 m, 95% CI, −4.40 to 19–21,23 blinding of participants to intervention, four trials were 21.17, p = 0.20; Fig. 3). The I estimate was 52% that represents 14–18,22,24,25 considered at high risk of bias, eight trials did not moderate-to-substantial heterogeneity. provide sufficient information for assessment about the degree of participant blinding, and the remaining trial was considered at Quality of life (QoL). Twelve of the 13 trials reported QoL; two of 15,21 low risk. Halpin et al. (2011) was judged to be at low risk because these trials reported two different quality-of-life measures. both control and intervention participants had access to a Across all 12 trials, 14 quality-of-life measures were reported npj Primary Care Respiratory Medicine (2020) 11 Published in partnership with Primary Care Respiratory Society UK 1234567890():,; Included Eligibility Screening Identification G Shaw et al. Published in partnership with Primary Care Respiratory Society UK npj Primary Care Respiratory Medicine (2020) 11 Table 1. An overview of study characteristics of the included 13 trials, reported as mean (SD) or median (lower quartile–upper quartile), unless otherwise stated. Author (year) Setting Sample size Age Sex (M %) Lung function (using Primary Other outcomes Duration of Duration FEV % predicted) outcome intervention of study Intervention Control Intervention Control Intervention Control Intervention Control Liu (2008) – 24 24 71.4 (1.7) 72.8 (1.3) 100 100 45.2 (3.2) 46.0 (2.8) ISWT Exacerbation, QoL 9 months 9 months Halpin (2011) Primary care 40 39 68.5 (1.5) 70.2 (1.6) 74.4 73.7 48.0 (4.0) 53.0 (3.0) Exacerbations Application ability to 4 months 4 months predict increases in exacerbation, changes in health status Chau (2012) Outpatient clinic 22 18 73.5 (6.1) 72.2 (6.1) 95.5 100 38.0 (12.9) 37.7 – User satisfaction, 2 months 2 months (16.52) QoL, and hospital admissions Nguyen (2013) Community 43 41 68.5 (11.0) 69.3 (8.0) 58.1 58.5 53.3 (20.4) 49.4 (19.8) Dyspnea 6MWT, QoL (CRQ) 12 months 12 months Pinnock (2013) Primary care and 128 128 69.4 (8.8) 68.4 (8.4) 41.4 49.2 44.0 (18.8) 40.0 (17.0) Time to the Admissions, 12 months 12 months community first hospital exacerbations, self- admission due Efficacy, QoL, anxiety, to and depression exacerbation Tabak (2014-A) Primary and 12 12 64.1 (9.0) 62.8 (7.4) 50 50 50 36 Use of 6MWT, dyspnea, 9 months 9 months secondary care (33.3–61.5) (26–53.5) application fatigue, QoL, No. of and hospitalizations, and satisfaction activity level Tabak (2014-B) Secondary care 14 16 65.2 (9.0) 67.9 (5.7) 57.1 68.8 48.7 (16.7) 56.4 (10.6) Activity level Dyspnea, fatigue, 1 month 1 month and QoL van der Primary care 65 68 57.5 (7.0) 59.2 (7.5) 47.7 45.6 –– Activity level QoL, general and 4–6 months 9 months Weegen (2015) exercise self-efficacy Vorrink (2016) Physiotherapist 84 73 62.0 (9.0) 63.0 (8.0) 50 49.3 59.0 (20.0) 53.0 (15.0) Activity level 6MWT, QoL, and BMI 6 months 12 months Demeyer (2017) Community 171 172 66.0 (8.0) 67.0 (8.0) 65 63 55.0 (20.0) 57.0 (21.0) Activity level QoL and 6MWT 12 weeks 12 weeks Farmer (2017) Primary and 110 56 69.8 (9.1) 69.8 (10.6) 61.8 60.7 –– QoL Hospital admissions, 12 months 12 months secondary care, exacerbations community Orme (2018) Secondary care 12 11 – – –– –– Feasibility and Dyspnea, fatigue, 2 weeks 2 weeks acceptability anxiety, depression, QoL, and self-efficacy Wang (2018) Secondary care 32 32 66.4 (6.2) 67.1 (6.2) 65.6 71.9 –– Self- Dyspnea 3 months 3 months management G Shaw et al. Liu (2008) LOW UNCLEAR UNCLEAR UNCLEAR HIGH LOW Halpin (2011) LOW LOW LOW LOW LOW LOW Chau (2012) LOW UNCLEAR UNCLEAR HIGH LOW LOW Nguyen (2013) UNCLEAR UNCLEAR UNCLEAR UNCLEAR LOW LOW Pinnock (2013) LOW LOW UNCLEAR LOW HIGH LOW Tabak (2014-A) LOW LOW UNCLEAR UNCLEAR HIGH HIGH Tabak (2014-B) LOW LOW UNCLEAR HIGH HIGH LOW van der Weegen (2015) LOW UNCLEAR HIGH LOW LOW LOW Vorrink (2016) LOW UNCLEAR HIGH LOW HIGH LOW Demeyer (2017) LOW LOW HIGH LOW LOW LOW Farmer (2017) LOW LOW UNCLEAR LOW LOW LOW Orme (2018) LOW LOW HIGH HIGH HIGH LOW Wang (2018) LOW UNCLEAR UNCLEAR HIGH LOW LOW Fig. 2 Risk of bias assessment. An outline of the bias assessment findings for the 13 included trials. 25 14,15,17,20,24 (Table 4). Only one trial reported the SF-12 measure, reporting a Dyspnea. Five trials reported data relating to dyspnea 14,15,20 significant difference between intervention and control post- (Supplementary Table 3). Three trials used the dyspnea 15,19 intervention. Two trials used the SF-36 measure, but these did component of the Chronic Respiratory Disease Questionnaire 21 17,24 not identify statistically significant differences. One trial reported measure, while the other two trials used the modified Medical the individual mental, functional, and symptom domains of the Research Council dyspnea scale. Only one trial reported a Chronic COPD Questionnaire. There was a significant difference statistically significant difference between groups. between the intervention and control groups in the Functional 14,17,18,20,23 CCQ measure post intervention but not in other domains. Two Fatigue. Five trials reported data concerning fatigue 17,18 14,20 trials recorded the total CCQ score, but the results were not (Supplementary Table 4). Two trials reported the fatigue 17,18 significant. The Chronic Respiratory Disease Questionnaire was component of the CRQ, two trials reported the Multidimen- 15 14,20 23 reported in full by one trial , and partially by two trials (only sional Fatigue Inventory, and one trial used the Functional reporting the emotion and mastery domains). These three trials Assessment of Chronic Illness Therapy measure. None of these reported non-significant results for these domains. Three trials reported significant improvements in the intervention arm 13,16,22 trials reported the St. George’s Respiratory Questionnaire compared with control. 21,23 and two trials reported the COPD Assessment Test measure of 17–21 QoL, but none of them showed significant differences between Physical activity. Five trials reported device-based levels of intervention and control groups. The 12 trials reporting QoL were physical activity (Supplementary Table 5). Four trials recorded assessed for inclusion for the meta-analysis, but trials that did not physical activity using accelerometers, while the remaining trial report a total or summative score were excluded, resulting in a used pedometers. Only one trial reported a statistically total of eight eligible trials (Fig. 4). The trial by Nguyen et al. (2013) significant difference in physical activity outcomes between reported two total scores reflecting QoL (Chronic Respiratory groups in the post-intervention period. Two of these five trials Disease Questionnaire and SF-36); the disease-specific scale also provided self-reported levels of physical activity, using the (Chronic Respiratory Disease Questionnaire) was included in the Moderate Physical Activity questionnaire and the Baecke meta-analysis. No difference in QoL was found between mobile Physical Activity Questionnaire . Both trials reported non- device application intervention and usual care (standardized mean significant changes from baseline. difference, −0.4 points; 95% CI, −0.86 to 0.05, p = 0.08). The I 15,16,19,23 estimate was 83% that represents considerable heterogeneity. The Self-efficacy. Four trials reported patient self-efficacy minimal clinically important differences for the back-translated (Supplementary Table 6). The employed measures focused on 15 23 19 16,19 standardized mean differences are presented in Supplementary dyspnea , falls , exercise , and self-efficacy more generally . Table 2. No trials recorded statistically significant findings. npj Primary Care Respiratory Medicine (2020) 11 Published in partnership with Primary Care Respiratory Society UK Random sequence generation Allocation concealment Blinding of participants and personnel Blinding of outcome assessment Incomplete outcome data addressed Selective reporting G Shaw et al. Table 2. A summary of the main findings for exacerbations. Author (year) sample size Type of exacerbation reported Group allocation Intervention Control Liu (2008) Intervention N = 24, Control N = 24 Managed in the community, N; p = NS F: N =4F: N = 26 Leading to hospitalization, N; p = NS F: N =2F: N = 22 Halpin (2011) Intervention N = 40, Control N = 39 Clinical exacerbation frequency, mean (SD); p = NS F: 0.95 (1.71) F: 1.17 (1.81) Chau (2012) Intervention N = 22, Control N = 18 Managed in the community, N; p = NS F: N =7F: N = 3 Leading to hospitalization, mean (SD) or N; p = NS B: 2.41 (1.57) B: 2.89 (2.32) F: N = 7 F: N = 3 Pinnock (2013) Intervention N = 8, Control N = 8 Leading to hospitalization, mean (SD); p = NS F: 1.5 (2.3) F: 1.3 (1.8) Managed in the community, mean (SD); p= NS F: 15 (12.7) F: 12.8 (11.8) Farmer (2017) Intervention N = 110, Control N = 56 Unspecified, median (IQR); p = NS F: 1 (0–2) F: 1 (0–3) B baseline, F follow-up, IQR interquartile range, N number of participants, NS nonsignificant, SD standard deviation. Table 3. A summary of the main findings for physical function. Author (year) sample size Type of physical function assessment reported Group allocation Intervention Control Liu (2008) Intervention N = 24, Control N = 24 ISWT (m), mean (SD); p = NS B: 255.8 (200.9) B: 262.9 (88.8) F: 306.7 (103.9) F: 237.8 (60.7) Nguyen (2013) Intervention N = 43, Control N = 41 6MWT (m), mean (SD); p = NS B: 400.5 (100.0) B: 398.0 (99.7) F: 431.3 (124.4) F: 406.6 (125.0) Tabak (2014-A) Intervention N = 12, Control N = 12 6MWT (m), mean (SD); p = NS B: 409.5 (102.2) B: 300.1 (116.4) F: 412 (134.1) F: 312.4 (152.4) Vorrink (2016) Intervention N = 84, Control N = 73 6MWT (m), mean (SD) or median (IQR); p= NS B: 456 (128.3) B: 461 (73.3) C: 0.8 (−8.8 to 10.3) C: 4 (−2.4 to 10.3) Demeyer (2017) Intervention N = 171, Control N = 172 6MWT (m), mean (SD); p = 0.009 B: 444 (106) B: 450 (106) F: 457 (108) F: 449 (118) 6MWT 6-minute walk test, B baseline, C change, F follow-up, IQR interquartile range, m meters, NS nonsignificant, SD standard deviation. 16,23 Anxiety and depression. Two trials reported anxiety and durations. However, the present review was prospectively depression, using the Hospital Anxiety and Depression Scale registered on a database of systematic reviews and included (HADS), and no statistically significant differences were observed. trials published in any language in several databases from inception. A sensitive search strategy was developed, and screening of citations was performed independently, minimizing DISCUSSION the risk of bias at review level. The review was inclusive of a broad This systematic review provided a comprehensive description and range of outcome measures, contributing to its comprehensive summarized the findings of mobile device application interven- nature. tions for COPD self-management. The interventions identified Although exacerbations can negatively impact QoL and were heterogeneous in nature, including the components (such as increase mortality , only five of the included trials reported the inclusion of periphery devices), the degree and frequency of exacerbations. Only one of these trials reported pre-intervention involvement of healthcare professionals, and frequency of and post-intervention exacerbation frequency , and exacerba- participant-performed measurements and data entry. It remains tions were reported using a wide range of metrics, including those unclear whether mobile device applications are more effective at exacerbations managed in the community and leading to preventing exacerbations when compared with usual care. hospitalization. An 80% reduction in likelihood of having an As only published trials were eligible for inclusion, there is exacerbation has been demonstrated previously in a meta-analysis potential for publication bias within the review. Also, the risk comparing a smartphone intervention with usual care . However, assessment bias tool was challenging to implement because the meta-analysis showed moderate heterogeneity in this blinding of participants in digital health interventions where the healthcare professional contact, in part possible because of the comparator is usual care may not be feasible to implement. In small sample size of the three trials pooled. It is unclear if addition, our ability to pool further outcome measures using reporting the number of contacts with healthcare professionals is meta-analysis was limited, given the variety of outcome measures a suitable outcome measure to represent COPD exacerbations; used across the trials. There are also limitations to interpreting digital interventions can offer an alternate means of contacting a summary estimates from pooled data, particularly when the healthcare professional, impacting the accuracy of assessing design of the studies, scales used to assess effectiveness, and exacerbation frequency in this way. With prevention and manage- interventions tested are heterogeneous and use varying follow-up ment of exacerbations being a key feature of COPD care, and an Published in partnership with Primary Care Respiratory Society UK npj Primary Care Respiratory Medicine (2020) 11 G Shaw et al. Fig. 3 Physical function forest plot. Forest plot of the effect of mobile device applications on physical function. Table 4. A summary of the main findings for quality of life. Author (year) sample size Form of QoL assessment reported Group allocation Intervention Control Liu (2008) Intervention N = 24, SF-12 PCS, mean (SD); p < 0.001 B: 38.7 (8.82) B: 40.1 (6.37) Control N = 24 F: 47.9 (7.35) F: 30.9 (10.78) Halpin (2011) Intervention N = 40, SGRQ, mean (SD); p = NS B: 52.4 (16.44) B: 53.6 (14.99) Control N = 39 F: 49.7 (15.18) F: 51.5 (14.99) Chau (2012) Intervention N = 22, CRQ (Emotion), mean (SD); p = NS B: 4.84 (1.47) B: 5.24 (1.42) Control N = 18 F: 4.92 (1.40) F: 5.61 (1.17) CRQ (Mastery), mean (SD); p= NS B: 4.60 (1.43) B: 4.94 (1.16) F: 4.61 (1.62) F: 4.88 (1.27) Nguyen (2013) Intervention N = 43, CRQ (Total), mean (SD); p = NS B: 96.4 (19.91) B: 96.2 (19.76) Control N = 41 F: 104.8 (23.92) F: 98.4 (24.34) Pinnock (2013) Intervention N = 8, SGRQ, mean (SD); p = NS B: 68.6 (16.6) B: 68.0 (15.2) Control N = 8 F: 68.2 (16.3) F: 67.3 (17.3) Tabak (2014-A) Intervention N = 12, CCQ (Total), mean (SD); p = NS B: 2.0 (0.90) B: 2.7 (0.94) Control N = 12 F: 1.8 (0.83) F: 2.3 (0.90) Tabak (2014-B) Intervention N = 14, CCQ (Total), mean (SD); p = NS B: 2.0 (0.8) B: 1.8 (1.0) Control N = 16 F: 1.7 (0.5) F: 1.8 (0.6) van der Weegen (2015) Intervention SF-36 (Physical), mean (SD); p = NS B: 42.5 (11.1) B: 45.8 (9.4) N = 65, Control N = 68 F: 44.1 (9.5) F: 45.8 (9.5) SF-36 (Mental), mean (SD); p = NS B: 48.2 (10.3) B: 50.1 (9.5) F: 48.3 (11.7) F: 50.3 (8.3) Vorrink (2016) Intervention N = 84, CRQ (Emotion), mean (SD) or median (IQR); p = NS B: 5.0 (1.1) B: 4.8 (1.2) Control N = 73 C: 0.09 (−0.07 to 0.24) C: 0.19 (−0.31 to 0.11) CRQ (Mastery), mean (SD) or median (IQR); p = NS B: 5.4 (1.1) B: 5.3 (1.1) C: −0.1 (−0.31 to 0.11) C: −0.23 (−0.39 to −0.06) Demeyer (2017) Intervention N = 171, CCQ (Mental), median (IQR); p = NS B: 1 (0–2.5) B: 1 (0–2) Control N = 172 F: 1 (0–2.5) F: 1 (0–2) CCQ (Functional), median (IQR); p= 0.026 B: 1.5 (1–2.75) B: 1.5 (0.75– 2.75) F: 1.5 (1–2.75) F: 1.75 (0.75– 2.75) CCQ (Symptoms), median (IQR); p = NS B: 1.75 (1.25– 2.5) B: 1.75 (1.5–2.75) F: 1.75 (1.25– 2.5) F: 2 (1.25– 2.75) Farmer (2017) Intervention N = 110, SGRQ, mean (SD); p = NS B: 56.4 (19.7) B: 55.5 (16.2) Control N = 56 F: 56.9 (19.5) F: 56.8 (20.9) Orme (2018) Intervention N = 12, CAT, mean (SD); p = NS B: 22.6 (4.4) B: 24.5 (9.7) Control N = 11 F: 21.6 (5) F: 23.8 (11.1) B baseline, C change, CCQ Clinical COPD Questionnaire, CRQ Chronic Respiratory Disease Questionnaire, F follow-up, SGRQ St. George's Respiratory Questionnaire, SF short form, CAT COPD Assessment Test, IQR interquartile range, SD standard deviation, NS nonsignificant. 28–30 increasing interest in predicting the onset of exacerbations , apparent lack of impact may be from the small size of the future trials are recommended to consider this when reporting studies, with 8 of the 13 trials reporting a sample size of fewer 13–15,17,18,23,25,31 exacerbations to more accurately quantify the impact of digital than 100 participants . In addition, the extent to interventions on this important clinical outcome. which the measures used in these studies were sensitive to The trials identified in this systematic review do not yet provide change is unclear. strong evidence for implementing mobile digital health interven- Hanlon et al. conducted a metareview of telehealth trials across tions for COPD. Only four trials reported clinical differences multiple health conditions, including COPD, diabetes, cancer, and between the intervention and control groups, and these heart failure . Their findings suggest that the evidence base is differences were in a range of outcomes, including physical more developed in diabetes and heart failure and more intensive 19,21,25,31 function, QoL, physical activity, and dypsnea . This and multifaceted interventions associated with greater npj Primary Care Respiratory Medicine (2020) 11 Published in partnership with Primary Care Respiratory Society UK G Shaw et al. Fig. 4 QoL forest plot. Forest plot of the effect of mobile device applications on quality of life. improvements in asthma, diabetes, and heart failure. Building on to standardize the outcomes used in this area of research are encouraged to increase the comparability of future trials. published reviews focused on COPD, our findings also report on QoL, self-efficacy, fatigue, anxiety, and depression, as well as exacerbations, physical function, and physical activity. In addition, METHODS we provide an in-depth description of the interventions within the Registration included trials. The results from our pooled data meta-analysis do not identify a The review was registered on the International Prospective statistically significant effect on measures of physical function or Register of Systematic Reviews (PROSPERO reference number: QoL. Previous meta-analyses have identified no differences in CRD42019124232). 10 10 physical function (using the 6-minute walk test) , dyspnea , and average days of hospitalization , but have noted that the Eligibility criteria intervention arm was favored for physical activity and a lower Randomized controlled trials of adults with a clinical diagnosis of risk of hospital admissions . COPD were included where the intervention group received a Looking beyond the effectiveness of the intervention for clinical mobile device application to support their COPD self- outcomes, it is possible that there are efficiency and organiza- management. A mobile device application was defined as a tional benefits of digital and telehealth care compared with more contained program that served a specific function relating to traditional models of care. None of the studies included in this COPD and personal health on a portable, electronic device review reported service outcomes. (including smartphones and tablet computers). This definition is in The trial interventions identified in our review focused on 11,12 line with previous systematic reviews on the topic . For the varying components of COPD self-management, including mon- purpose of inclusion, self-management was defined as patient itoring symptoms, encouraging lifestyle changes (such as management of their personal symptoms and medication regimes increases in physical activity or exercise), and hosting educational related to the condition, as well as coping with the emotional and material concerning COPD. Some of the trials explored ease of use, 35,36 lifestyle impacts of the condition . Studies were eligible where feasibility, and accessibility of the technologies. Aligning with this the comparator group received usual care only. Outcomes heterogeneity is the variety of outcome measures used to assess included but were not restricted to exacerbations, QoL, physical the effectiveness of the intervention. This review highlights the function, physical activity, and dyspnea. number of outcome measures used and variation in which the tool was used for data collection between studies. Information sources and search Our findings and the challenges encountered in synthesizing Medline, EMBASE, Cochrane Library, CINAHL, and the Science the evidence from these trials highlight the importance of Citation Index were searched from inception to 12th April 2019 developing a minimum and standardized set of clinically following the methods recommended by the Preferred Reporting important core-outcome measures to allow comparison of trials Items for Systematic Reviews and Meta-Analyses guidelines . Full involving people with COPD. This would be in line with minimum search strategies are included in Supplementary Methods. The reporting guidelines for other areas of clinical speciality, including search algorithm focused on keywords relating to ‘COPD’, ‘mobile rheumatology . In practice, the use of mobile device applications phone application’, and ‘self-management’ and included inter- to support self-management may have some negative effects. For ventions with or without healthcare professional input. example, a patient might be falsely reassured if they feel their data were being monitored by a healthcare professional. On the other Study selection hand, the data can supplement routine care with information about variation in symptoms and clinical markers of the condition. The resulting citations were imported into the web-based From a policy perspective, the economic cost of telehealth for Covidence systematic review software (Veritas Health Innovation, chronic disease is high (£92,000/QALY), which restricts its Melbourne, Australia). Screening of titles and abstracts was implementation in the majority of healthcare settings . completed by two authors independently (G.S. and M.W.). In the In conclusion, this systematic review demonstrates that there event of disagreement, two further reviewers (L.A. and A.F.) are a number of trials being conducted in this area of COPD. decided their eligibility. Subsequently, full-text screening was However, there is insufficient evidence to date to suggest that conducted by two authors independently (G.S. and M.W.). Any mobile device applications are effective for the self-management disagreements were resolved following discussion with the other of COPD over usual care. This may in part be due to a limited reviewers (L.A. and A.F.). The reference lists of the included trials ability for data to be pooled, owing to marked variation in were also screened to identify any additional potentially eligible methodology and reporting of outcome measures. Future efforts trials. Published in partnership with Primary Care Respiratory Society UK npj Primary Care Respiratory Medicine (2020) 11 G Shaw et al. Data collection process was used to help control for differences between trials, and to limit the impact of heterogeneity. Trials were weighted by sample Extraction forms were used to capture the following data: lead size, and 95% confidence intervals were reported around point author, year, country, trial setting, sample size, age, sex, lung estimates. Measures were selected for inclusion if they were function, primary and secondary outcomes, duration of interven- reported by at least three trials to align with the recent Cochrane tion and study, as well as the main findings. Data extraction was review . For continuous data with consistent units of measure- completed independently by two authors (G.S. and M.W.), and any ments (such as the 6-minute walk performance in meters), the disagreements were resolved through discussion. When data were mean difference in change between baseline and follow-up not directly identifiable within text or tables, authors were measurements was calculated. In instances where continuous data contacted or Microsoft Paint (Microsoft, Washington, USA) was were inconsistent between trials (i.e., multiple questionnaires with used to extract values from graphs. The graphical summaries were varying scales used to measure QoL), the standardized mean captured by screenshot and copy-pasted into the software. No differences between timepoints were calculated. Back-translation correction for rotation was required. Horizontal lines were inserted of the standardized mean difference for each scale was conducted across from the center of the datapoints of interest to the point of to the original scale, to present a mean difference for each QoL intersection on the y-axis. The y-axis was segmented into smaller instrument to give information of the clinical significance of this increments, marked by adding small lines to the axis, until a value difference. Where change in standard deviation was not reported could be extracted to 1 decimal place. The values were extracted by individual trials, the standard deviation for changes from from the original y-axis scale, meaning the x and y positions were baseline was imputed by calculating a correlation coefficient from not translated. Two authors (G.S. and M.W.) independently looked trials reporting a change in standard deviation. If the data were at the graphs to identify the value of interest. In the event any not reported, authors were contacted to access this information. disagreements were identified, G.S. and M.W. reassessed the The I statistic was used to estimate heterogeneity. Cochrane graphs and agreed on a value. recommendations for interpreting the I statistic are as follows: We subsequently replicated the data extraction using web plot 30–60% may represent moderate heterogeneity, 50–90% may digitizer software (Automeris version 3.9, https://automeris.io/ represent substantial heterogeneity, and 75–100% may represent WebPlotDigitizer/). The graphical summaries were captured by considerable heterogeneity . No funnel plot was produced as it is screenshot and saved as a PNG file before being uploaded to the not recommended for meta-analyses with fewer than 10 trials . web-based plot digitizer software. No correction for rotation was required. Once uploaded, two anchoring points were assigned to each axis: the highest and lowest value on the y-axis and baseline Reporting summary and follow-up for the x-axis. Values reflecting these anchoring Further information on research design is available in the Nature points were declared. The datapoints were selected using the Research Reporting Summary linked to this article. center of each point to 14 decimal places, and the acquired data were recorded in the form of coordinates that aligned with the scales in the original graphs. DATA AVAILABILITY All data generated or analyzed during this study are included in this published article Risk of bias assessment and Supplementary Material files. The included trials were assessed for potential bias at study level using the Cochrane risk of bias tool . Two authors (G.S. and M.W.) CODE AVAILABILITY independently completed the assessment of bias, and any No custom code or mathematical algorithm were used. disagreements were resolved through discussion with the other reviewers (L.A. and A.F.). Received: 17 July 2019; Accepted: 28 February 2020; Synthesis of data The results were converted to mean (standard deviation) when possible; otherwise data were reported as median (lower to upper quartile). A pragmatic decision was made to include outcome REFERENCES measures reported by four or more trials in the main table and 1. World Health Organisation. The Global Burden of Disease. (2008) https://www.who. those reported less frequently in the text. Where the duration of int/healthinfo/global_burden_disease/GBD_report_2004update_full.pdf?ua=1. intervention period and study duration differed, data were 2. Mathers, C. D. & Loncar, D. Projections of global mortality and burden of disease extracted for the end of the observation period. Outcomes were from 2002 to 2030. PLoS Med. 3, e442 (2006). grouped together where different measures were used, for 3. Global Initiative for Chronic Obstructive Lung Disease. Global Strategy for the example, where different scales for QoL measurement were used. Diagnosis, Management and Prevention of Chronic Obstructive Pulmonary Disease. 2018 report (2018). The total scores from the QoL measurement tools were extracted 4. McLean, S. et al. Projecting the COPD population and costs in England and when these were reported; otherwise individual component Scotland: 2011 to 2030. Sci. Rep. 6,1–10 (2016). scores were extracted. Similarly, exacerbations that were treated 5. Foundation, B. L. Estimating the Economic Burden Of Respiratory Illness in the UK blf in the community were grouped, to include self-reported Estimating the Economic Burden of Respiratory Illness in the UK, British Lung exacerbations (where a participant may have initiated a rescue Foundation (2014). pack), alongside exacerbations that were managed by primary 6. Ståhl, E. et al. Health-related quality of life is related to COPD disease severity. care teams. Measures of physical activity were included in the Health Qual. Life Outcomes 3, 56 (2005). summary table if these were objectively measured; self-report of 7. Bourbeau, J. et al. Reduction of hospital utilization in patients with chronic physical activity was not included. obstructive pulmonary disease: a disease-specific self-management intervention. Arch. Intern. Med. 163, 585–591 (2003). 8. Williams, V., Price, J., Hardinge, M., Tarassenko, L. & Farmer, A. Using a mobile Synthesis of results health application to support self-management in COPD: a qualitative study. Br. J. Meta-analysis was carried out using Review Manager (Review Gen. Pract. 64, e392–e400 (2014). 9. Yang, F., Wang, Y., Yang, C., Hu, H. & Xiong, Z. Mobile health applications in self- Manager [RevMan] version 5.3, Cochrane Collaboration, Copenha- management of patients with chronic obstructive pulmonary disease: a sys- gen, Denmark). A difference-in-difference random effect analysis npj Primary Care Respiratory Medicine (2020) 11 Published in partnership with Primary Care Respiratory Society UK G Shaw et al. tematic review and meta-analysis of their efficacy. BMC Pulm. Med. 18, 147 (2018). 34. Henderson, C. et al. Cost effectiveness of telehealth for patients with long term 10. Lundell, S., Holmner, Å., Rehn, B., Nyberg, A. & Wadell, K. Telehealthcare in COPD: conditions (Whole Systems Demonstrator telehealth questionnaire study): nested a systematic review and meta-analysis on physical outcomes and dyspnea. Respir. economic evaluation in a pragmatic, cluster randomised controlled trial. BMJ 22, Med. 109,11–26 (2015). 346 (2013). 11. Alwashmi, M. et al. The effect of smartphone interventions on patients with 35. Barlow, J., Wright, C., Sheasby, J., Turner, A. & Hainsworth, J. Self-management chronic obstructive pulmonary disease exacerbations: a systematic review and approaches for people with chronic conditions: a review. Patient Educ. Couns. 48, meta-analysis. JMIR mHealth uHealth 4, e105 (2016). 177–187 (2002). 12. McCabe, C., McCann, M. & Brady, A. M. Computer and mobile technology inter- 36. Glasgow, R. E., Davis, C. L., Funnell, M. M. & Beck, A. Implementing practical ventions for self-management in chronic obstructive pulmonary disease. interventions to support chronic illness self-management. Jt. Comm. J. Qual. Saf. Cochrane Database Syst. Rev. 5, CD011425 (2017). 29, 563–574 (2003). 13. MG Halpin, D. et al. A randomised controlled trial of the effect of automated 37. Moher, D., Liberati, A., Tetzlaff, J. & Altman, D. G., PRISMA Group. Preferred interactive calling combined with a health risk forecast on frequency and severity reporting items for systematic reviews and meta-analyses: the PRISMA statement. of exacerbations of COPD assessed clinically and using EXACT PRO. Prim. Care PLoS Med. 6, e1000097 (2009). Respir. J. 20, 324–331 (2011). 38. Higgins, J. P. T., Thomas, J., Chandler, J., Cumpston, M., Li, T., Page, M. J. & Welch, 14. Chau, J. P.-C. et al. A feasibility study to investigate the acceptability and potential V. A. (eds) Cochrane Handbook for Systematic Reviews of Interventions version effectiveness of a telecare service for older people with chronic obstructive 6.0 (updated July 2019). Cochrane. (2019). Available from www.training.cochrane. pulmonary disease. Int. J. Med. Inform. 81, 674–682 (2012). org/handbook. 15. Nguyen, H. Q. et al. Internet-based dyspnea self-management support for patients 39. Cochrane Collaboration. 9.5.2 Identifying and Measuring Heterogeneity. https:// with chronic obstructive pulmonary disease. J. Pain Symptom Manag. 46,43–55 (2013). handbook-5-1.cochrane.org/chapter_9/9_5_2_identifying_and_measuring_ 16. Pinnock, H. et al. Effectiveness of telemonitoring integrated into existing clinical heterogeneity.htm. Accessed 30 Oct 2019. services on hospital admission for exacerbation of chronic obstructive pulmonary 40. Sterne, J. A. C. et al. Recommendations for examining and interpreting funnel plot disease: researcher blind, multicentre, randomised controlled trial. BMJ 347, asymmetry in meta-analyses of randomised controlled trials. BMJ 343, d4002 f6070 (2013). (2011). 17. Tabak, M., Vollenbroek-Hutten, M. M., van der Valk, P. D., van der Palen, J. & Hermens, H. J. A telerehabilitation intervention for patients with chronic obstructive pulmonary disease: a randomized controlled pilot trial. Clin. Rehabil. 28,582–591 (2014). ACKNOWLEDGEMENTS 18. Tabak, M., Brusse-Keizer, M., van der Valk, P., Hermens, H. & Vollenbroek-Hutten, The authors would like to thank Ms. Ran Xu for her support in translating one of M. A telehealth program for self-management of COPD exacerbations and pro- the papers. A.J.F. is a NIHR Senior Investigator. This project is supported by the motion of an active lifestyle: a pilot randomized controlled trial. Int. J. Chron. NIHR Oxford Biomedical Research Centre. The views expressed are those of the Obstruct. Pulmon. Dis. 9, 935–944 (2014). authors and not necessarily those of the NIHR or the Department of Health and 19. van der Weegen, S. et al. It’s life! Mobile and web-based monitoring and feedback Social Care. tool embedded in primary care increases physical activity: a cluster randomized controlled trial. J. Med. Internet Res. 17, e184 (2015). AUTHOR CONTRIBUTIONS 20. Vorrink, S. N. W., Kort, H. S. M., Troosters, T., Zanen, P. & Lammers, J.-W. J. Efficacy of an mHealth intervention to stimulate physical activity in COPD patients after G.S. made substantial contributions to the conception of the work, acquisition, pulmonary rehabilitation. Eur. Respir. J. 48, 1019–1029 (2016). analysis, and interpretation of data for the work, drafted the work, approved the 21. Demeyer, H. et al. Physical activity is increased by a 12-week semiautomated final version to be published, and agrees to be accountable for all aspects of the telecoaching programme in patients with COPD: a multicentre randomised work. M.E.W. made substantial contributions to the acquisition, analysis, and controlled trial. Thorax https://doi.org/10.1136/thoraxjnl-2016-209026 (2015). interpretation of data for the work, as well as revised the work critically for 22. Farmer, A. et al. Self-management support using a digital health system com- important intellectual content, approved the final version to be published, and pared with usual care for chronic obstructive pulmonary disease: randomized agrees to be accountable for all aspects of the work. L.C.A. made substantial controlled trial. J. Med. Internet Res. 19, e144 (2017). contributions to the acquisition, analysis, and interpretation of data for the work, 23. Orme, M. W. et al. Findings of the chronic obstructive pulmonary disease-sitting as well as revised the work critically for important intellectual content, approved and exacerbations trial (COPD-SEAT) in reducing sedentary time using wearable the final version to be published, and agrees to be accountable for all aspects of and mobile technologies with educational support: randomized controlled fea- the work. A.J.F. made substantial contributions to the conception of the work and sibility trial. JMIR mHealth uHealth 6, e84 (2018). interpretation of data, revised the work critically for important intellectual content, 24. Wang, T. & Yang, Q. Effect study on improving self-management of patients with approved the final version to be published, and agrees to be accountable for all chronic obstructive pulmonary disease based on APP. Chin. Nurs. Res. 32, aspects of the work. N.R. made substantial contributions to the acquisition of data 3121–3124 (2018). for the work, as well as revised the work critically for important intellectual content, approved the final version to be published, and agrees to be accountable 25. Liu, W.-T. et al. Efficacy of a cell phone-based exercise programme for COPD. Eur. for all aspects of the work. Respir. J. 32, 651–659 (2008). 26. Seemungal, T. A. R. et al. Effect of exacerbation on quality of life in patients with chronic obstructive pulmonary disease. Am. J. Respir. Crit. Care Med. 157, 1418–1422 (1998). COMPETING INTERESTS 27. Soler-Cataluña, J. J. et al. Severe acute exacerbations and mortality in patients The authors declare no competing interests. with chronic obstructive pulmonary disease. Thorax 60, 925–931 (2005). 28. Miłkowska-Dymanowska, J., Białas, A. J., Obrębski, W., Górski, P. & Piotrowski, W. J. A pilot study of daily telemonitoring to predict acute exacerbation in chronic ADDITIONAL INFORMATION obstructive pulmonary disease. Int. J. Med. Inform. 116,46–51 (2018). Supplementary information is available for this paper at https://doi.org/10.1038/ 29. Shah, S. A. et al. Personalized alerts for patients with COPD using pulse oximetry s41533-020-0167-1. and symptom scores. In 2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 3164–3167 (IEEE, 2014). Correspondence and requests for materials should be addressed to M.E.W. 30. Fernandez-Granero, M. A., Sanchez-Morillo, D. & Leon-Jimenez, A. An artificial intelligence approach to early predict symptom-based exacerbations of COPD. Reprints and permission information is available at http://www.nature.com/ Biotechnol. Biotechnol. Equip. 32, 778–784 (2018). reprints 31. Zhang, Q. et al. Disease knowledge level is a noteworthy risk factor of anxiety and depression in patients with chronic obstructive pulmonary disease: a cross- Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims sectional study. BMC Pulm. Med. 14, 92 (2014). in published maps and institutional affiliations. 32. Hanlon, P. et al. Telehealth interventions to support self-management of long-term conditions: a systematic metareview of diabetes, heart failure, asthma, chronic obstructive pulmonary disease, and cancer. J. Med. Internet Res. 19, e172 (2017). 33. Tugwell, P. et al. OMERACT: an international initiative to improve outcome measurement in rheumatology. Trials 8, 38 (2007). Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, Published in partnership with Primary Care Respiratory Society UK npj Primary Care Respiratory Medicine (2020) 11 G Shaw et al. adaptation, distribution and reproduction in any medium or format, as long as you give regulation or exceeds the permitted use, you will need to obtain permission directly appropriate credit to the original author(s) and the source, provide a link to the Creative from the copyright holder. To view a copy of this license, visit http://creativecommons. Commons license, and indicate if changes were made. The images or other third party org/licenses/by/4.0/. material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the © The Author(s) 2020 article’s Creative Commons license and your intended use is not permitted by statutory npj Primary Care Respiratory Medicine (2020) 11 Published in partnership with Primary Care Respiratory Society UK

Journal

npj Primary Care Respiratory MedicineSpringer Journals

Published: Apr 1, 2020

There are no references for this article.