Table 2 Characteristics of RCTs of prescribable apps

From: Prescribable mHealth apps identified from an overview of systematic reviews

Study ID (Pilot?) Population and length of follow up Intervention Comparator Outcome Availability and Prescribability
Bonoto 2017 systematic review. Five relevant RCTs out of 13.
Berndt16 2014, Pilot Youth with T1DM (8–18 year olds, n = 68, 4 weeks) Treatment with support of telemedicine system Mobil Diab (mDiab) Conventional therapy only Between-group differences in results were not reported, though both groups saw reduction in HbA1c. Intervention group had higher 'diabetes self-efficacy' score. Overall usability and acceptance were rated as 'good' (41%). • Available in Germany in three versions: free stand-alone version mDiab Lite, full version mDiab €4.99, mDiab Pro version offers multi user license and connection to central web platform. • Lack of effectiveness is the barrier to prescribe this app.
Charpentier17 2011 Adults with T1DM (n = 180, HbA1c > 8%, 6 months) Group 1: quarterly visits + paper logbook Group 2: quarterly visits + DiaBeo app Group 3: biweekly teleconsultation + DiaBeo app (bolus calculator with validated algorithms) At 6 months G3 patients had 0.91%, and G2 had 0.67% reduction in HbA1c compared to G1 (control) (p < 0.001). Also, patients in G1 and G2 spent 5 h more than G3 for office consultations. • Available in France and reimbursed by the government. • Should it become available outside France and ongoing 700 patient trial results pending, it could be a prescribable app.
Drion18 2015 Adults with T1DM (n = 63, 3 months) DBEES app Standard paper diary No between-group differences were found in quality of life, HbA1c, daily frequency of self-measurement of blood glucose. • Available worldwide free of charge. • Lack of effectiveness is the barrier to prescribe this app.
Holmen19 2014 Adults with T2DM (n = 151, mean age 55.9–58.6, HbA1c > 7.1%, 1 year) Group 1: TAU + Few Touch app Group 2:TAU + Few Touch app + nurse counselling for the first 4 months Group 3: TAU with GP HbA1c level decreased in all groups, but did not differ between groups after 1 year. Those aged ≥ 63 years used the app more than their younger counterparts (OR 2.7; 95% CI 1.02–7.12; p = .045). • Available in Norway, Czech and USA under the name Diabetes Dagboka (Diary). • Lack of effectiveness is the barrier to prescribe this app.
Kirwan20 2013a Adults with T1DM (n = 72, HbA1c > 7.5%, 6 months, +9 months follow up) TAU + Glucose Buddy app + weekly SMS support Usual care (3 monthly visit to healthcare provider) There was a significant between group difference in HbA1c reduction (p < 0.001). At 9 months, the mean HbA1c reduction for intervention group was 1.10% (SD 0.74). The control group HbA1c increased slightly (mean 0.07, SD 0.99). • Available worldwide free of charge. • It is prescribable for improving glycaemic control.
Firth 2017 systematic review. Two relevant RCTs out of 9.
Pham21 2016, Pilot Adults with moderate anxiety (n = 63, 4 weeks) Flowy app Waitlist with weekly psychoeducation emails There were no between-group differences in reductions in anxiety, panic and hyperventilation. • Available worldwide free of charge. • Lack of effectiveness is the barrier to prescribe this app.
Roepke22 2015 Adults with depression (n = 283, 4 weeks) Group 1: SuperBetter app using CBT and positive psychotherapy (CBT-PPT SB) Group 2: General SuperBetter app Group 3: Waitlist control group Group 1 and 2 achieved greater reductions in CES-D questionnaire scores than control by posttest (Cohen's d = 0.67) and by follow-up (d = 1.05). • Available worldwide free of charge. • It could be prescribable. This trial suffered from high attrition (80%). Larger trials with longer follow ups are needed.
Flores-Mateo 2015 systematic review. Five relevant RCTs out of 9.
Allen27 2013a, Pilot Obese adults (BMI > 28 kg/m2, n = 68, 6 months) Group 1: Intensive counselling Group 2: Intensive counselling + Lose It! app Group 3: less intensive counselling + Lose It! app Group 4: Lose It!app alone At 6 months, there was no statistically significant weight loss between the groups (mean weight loss in G1 was −2.5 kg, in G2 −5.4 kg, in G3 −3.3 kg and in G4 −1.8 kg, respectively.) • Available worldwide. Basic version is free. • Lack of effectiveness is the barrier to prescribe this app on its own. Could be helpful as a support to weight loss counselling.
Carter28 2013a, Pilot Overweight adults (BMI > 27 kg/m2, n = 128, 6 months) MyMealMate app Self-monitoring slimming website OR calorie counting book by Weight Loss Resources company. At 6 months, there was statistically significant difference in mean weight loss between app group (−4.6 kg) and website group (−1.3 kg) (p = 0.004), but not between app and diary group (−2.9 kg) (p = 0.12). • Available worldwide free of charge. • It is prescribable for weight loss.
Glynn38 2014b Rural primary care patients (mean BMI 28.2 kg/m2, n = 90, 8 weeks) Accupedo-Pro app and physical activity goal of 10,000 steps/day Information on benefits of exercise and physical activity goal of walking for 30 min per day There was a difference in mean improvement of 2017 (95% CI 265 – 3768, p = 0.009) steps per day between the groups, favouring the intervention. No significant changes were observed for secondary outcomes of blood pressure, weight and BMI. • Available worldwide free of charge. • It is prescribable. An increase of over 1000 steps per day is clinically meaningful and, if continued, expected to result in long-term health benefits. Longer trials are needed to measure such effects.
Laing29 2014 Adult primary care patients (BMI > 25 kg/m2, n = 212, 6 months) MyFitnessPal app Usual care + 'any activities you would like to lose weight' At 6 months, there was no significant between group difference in weight change (−0.67 lb, 95% CI −3.3–2.11 lb, p = 0.63) and in self-reported behaviours around physical activity, diet and self-efficacy in weight loss. • Available worldwide free of charge. • Lack of effectiveness is the barrier to prescribe this app on its own.
Turner-McGrievy30 2011a Overweight adults (BMI 32.6 kg/m2, n = 96, 6 months) Weight loss podcast + Fat Secret calorie counter app and Twitter support group Podcast only (same as intervention, twice a week for 3 months, and 2 minipodcasts a week for months 3–6) Overall the two groups lost exactly the same amount of weight (−2.7 kg) and there was no significant difference in percentage weight loss between the groups (3.5% in intervention vs. 3.8% in control). • Available worldwide free of charge. • Podcast was designed by study team and proven effective in their 2009 RCT. However, the app addition did not make any difference in the results. Lack of evidence to prescribe this app.
Payne 2015 systematic review. Seven relevant RCTs out of 14.
Gajecki24 2014 University students with risky alcohol consumption (mean age 24.7, n = 1929, 7 weeks) Group 1: smartphone app Promillekoll Group 2: web-based app PartyPlanner that calculates blood alcohol concentration (BAC) No intervention Per-protocol analyses revealed only one significant time-by-group interaction, where Group 1 participants increased the frequency of their drinking occasions compared to controls (mean at baseline 2.24/wk, mean at FU 2.36/wk, p = 0.001). • Available in Sweden. • This study showed youth drinking behaviour needs to be explored further and apps need to provide more in-depth information than just the BAC. Such apps are not prescribable as they are.
Gustafson25 2014 Alcohol dependent adults leaving residential programs (mean age 38.3, n = 349, Intervention 8 months + follow up 4 months) TAU + Addiction-Comprehensive Health Enhancement Support System (A-CHESS) app Treatment as usual (TAU) (no continuing care) A-CHESS group reported fewer risky drinking days than the control group, with a mean of 1.39 vs 2.75 ds/previous 30 ds (mean difference, 1.37; 95%CI 0.46–2.27; p = 0.003). • Available in the USA through the agency involved with the study. • Should this app be made widely available, it can be prescribed to help with continuing care for people leaving residential programs and generally for people with alcohol dependency.
Watts23 2013, Pilot Adults with mild and moderate depression (n = 35, 8 weeks, +3 month follow up) The Sadness Program CBT lessons adapted for Get Happy app Same content on a website (previously proven effective) The results indicate that reductions in PHQ-9, the BDI-II and K-10 pre- to post-intervention and pre- to follow up, were significant, regardless of experimental group. • Available in Australia as Managing depression for AUD 59.99. • It is prescribable. The price could be a barrier for widespread use. It needs to be tested in a larger trial.
Schoeppe 2016 systematic review. Eight relevant RCTs out of 20.
Cowdery32 2015 Adults (n = 40, 12 weeks) Choice of either Zombies! Run or The Walk + MOVES app MOVES (activity monitoring app) There were no significant between-group differences in physical activity, enjoyment of exercise and motivation to exercise. • All apps are available worldwide free of charge. • Lack of effectiveness is the barrier to prescribe these apps.
Dirieto33 2015 Insufficiently active healthy young people (14–17 years old, n = 51, 8 weeks) Group 1: Immersive exergame app Zombies! Run Group 2: non-immersive app Get Running No intervention There were no significant between-group differences in cardiorespiratory fitness (1-mile walk/run test) and self-reported physical activity levels and its predictors. • Both apps are freely available. • Lack of effectiveness is the barrier to prescribe this app.
Mummah34 2016, Pilot Overweight adults (BMI 32 kg/m2, n = 17, 12 weeks) Vegethon app (to monitor vegetable consumption, set goals, get feedback and comparison) Waitlist Intention to treat analysis at the end of 12 weeks showed the between group vegetable consumption difference was 7.4 servings a day (95% CI 1.4–13.5; p = 0.02). • Available worldwide free of charge. • This study was done on select participants of a 12-month weight loss program. It is prescribable; however, larger trial is needed.
Silviera35 2013, VanHet Reve36 2014 Autonomous-living seniors (mean age 75, n = 44, 12 weeks) Group 1: ActiveLifestyle tablet app for strength-balance exercises, social motivation version Group 2: ActiveLifestyle tablet app, individual motivation version Group 3: brochure-based intervention Between group comparison showed moderate improvement for gait velocity and cadence in tablet groups. Social motivation strategies proved more effective than individual strategies in stimulating the participants. • Available in Italy free of charge. • The study measured surrogate outcomes for falls prevention. Should it become available worldwide, it is prescribable to seniors and other people who need to improve their balance and gait.
Walsh39 2016, Pilot Young adults (17–26 years old, n = 55, 5 weeks) Aim for 10,000 steps/d using Accupedo-Pro pedometer app with feedback and goal setting functions Information on daily recommended physical activity and told to walk for 30 min/day Between group differences revealed intervention group increased daily steps significantly (2393, about a mile) more than those in the control group (1101; t53 = 2.07, p = 0.043). • Available worldwide free of charge. • This app is prescribable to those who are interested in improving their physical activity levels by walking.
Wharton37 2014, Pilot Adults with BMI 25–40 kg/m2, (n = 57, 8 weeks) Group 1: record food intake using Lose-It! app Group 2: record food intake using phone’s memo function Group 3: record using paper-and-pencil method There was no between group differences in weight loss, BMI, and Healthy Eating Index at the end of study. The app group lost a slightly less weight (−3.5 lb) than the other groups (G2 lost −6.5 lb and G3 lost −4.4 lb). • Available worldwide. Basic version is free. • This app could be prescribed as food intake recording tool. Larger and longer trials are needed to establish advantage of using this app in weight loss interventions.
Simblett 2016 systematic review. One relevant RCT out of 39.
Miner26 2016, Pilot Adults with PTSD symptoms (n = 49, 2 months) PTSD Coach app Waitlist (crossed over after 1 month) There was no statistically significant between group differences in PTSD scores according to the PTSD Checklist—Civilian (PCL-C). The app was deemed acceptable and the intervention feasible. • Available worldwide free of charge. • Lack of effectiveness is the barrier to prescribe this app. Larger and longer trials are needed.
  1. TAU treatment as usual, CES-D Center for Epidemiological Studies Depression Questionnaire, BMI body mass index, MET metabolic equivalent of task, PHQ-9 Patient Health Questionnaire 9, BDI-II Beck’s Depression Inventory Second edition, K-10 The Kessler 10-item Psychological Distress scale, PTSD posttraumatic stress disorder
  2. aThese studies are also included in Payne 2015 systematic review
  3. b These studies are also included in Schoeppe 2016 systematic review