Table 3. Summary of interventions and effect on clinical outcomes (n = 47).

Author	Condition	Participants blinded to adherence monitoring	Type of EAM	Method of adherence feedback (Direct to patient or Indirect via health professional)	Immediate real-time feedback to participant about adherence	Method of adherence measurement (for both groups unless otherwise stated)	Effect on adherence	Size of effect	Clinical effect	Clinical measure	Main Clinical Outcome
Andrade et al., 2005 [36]	HIV	Not stated	Integrated	Both	No	EM, Adherence questionnaire	0, + Mean (±SE) adherence scores did not differ between groups (80 ± 2.3% vs 65 ±2.7%)	+15% (overall), +20% (memory-impaired subgroup)	+/-	Viral load, CD4	At week 24, plasma HIV RNA load was undetectable for 34% in the intervention group and 38% of the control (P = 0.49). Overall: Greater reduction in viral load, no difference in CD4
Artinian et al., 2003 [69]	Congestive Heart Failure (HF)	Not stated	Integrated	Direct	No	EM (intervention only), Pill count (both groups)	Not compared; only reported that the monitor group had 94% adherence	Not available	+/-	HF self-care scale, 6 min walk, Minnesota Living with HF Questionnaire for QOL, NYHA FC	No significant group by time interaction (F1, 16 = 0.02, P = -0.902), suggesting that there was no significant difference in the amount of HF self-care improvement between groups. Overall: No effect on clinical measures but had better physical QOL in the monitor group between baseline and follow-up.
Brath et al., 2013 [62]	Hypertension and dyslipidaemia	Not stated	Simple EAM	Direct	No	EM (intervention phase); pill count (control phase)	0 Medians in both intervention and control groups reported as 1 (1–1 for interquartile range), with no significant differences other than for metformin (p = 0.04)	Not available	++	BP Control, HbA1c, fasting blood glucose conc., blood cholesterol concentration	BP improved from 133/75 to 127/70 (p = 0.02/0.0003) from beginning to end of crossover phase; Total cholesterol improved from median (lower-upper quartile) = 166 (147–183) to 155 (141–167); p = 0.02. Overall: Improved BP, total cholesterol. Nil significant changes in glucose, body weight, or LDL or HDL cholesterol
Burgess et al., 2010 [27]	Asthma	No	Simple EAM	Indirect	No	EM	++ Mean (SD) adherence in intervention 79.0 (13.1)% vs control 57.9 (25.3)%; P<0.01) (data from author).	+21%	++	Reliever use, lung function, exacerbations, asthma symptoms	Intervention = 3 exacerbations vs control = 1 (p = 0.4); change in FEV1 from baseline = 13.8 (intervention) vs 9.8 (control) (p = 0.9); number of people using reliever medication ≥3 times per week = 2 (intervention) vs 0 (control). Overall: Improved asthma control based on changes in FEV1 and number of exacerbations but not significant between groups
Chan et al., 2015 [28]	Asthma	Yes	Simple EAM	Direct	Yes	EM	++ Intervention mean (SD) adherence: 78.5 (18.7)%, control 35.0 (23.2)% (data from author)	+54%	++	Asthma Morbidity Questionnaire, ACT	Mean (SD) change from baseline: Intervention +3.86 (5.8) vs control +2.62 (5.9). Overall: Improved asthma control based on asthma questionnaires
Charles et al., 2007 [29]	Asthma	Yes	Simple EAM	Direct	Yes	EM	++ Intervention mean (SD) adherence: 88 (16)% vs control 66(27)%	+22%	0	ACQ	Change from baseline = 0.5 in both groups; no SD given Overall: No effect on clinical
Christensen et al., 2010 [14]	Hypertension	Not stated	Simple EAM	Direct	No	EM, Self-report	0 Self reported compliance was higher in the intervention group than control (5.5% difference) in the 1st 6 months, but the opposite was seen in the crossover (-2.1%).	+6% then -2% with crossover	0	BP control	BP systolic change at 6 months: intervention group before cross over = -28.8 vs. control group -28.3 mmHg (p = 0.801) BP diastolic change at 6 months: intervention -13.4 vs control -13.6 mmHg (p = 0.808) Overall: No effect on clinical
de Bruin et al., 2010 [37]	HIV	No	Simple EAM	Both	Yes	EM	++, + Effect on timing adherence was significant (F(1, 129) = 14.11, p < 0.001, mean difference = 7.40% [3.50–11.30%])	+7% (overall), +15% (<95% baseline adherence)	++	Viral load, CD4	Logistic regression showed that the intervention group had a higher chance of being undetectable at postintervention than the control group (p <0.05, OR [95% CI] = 3.32 [1.13–9.80]). Overall: Greater reduction in viral load
De Geest et al., 2006 [56]	Renal transplant	No	Simple EAM	Indirect	No	EM	0 Chance of non-adherence reduced more in the intervention vs control group but difference was not significant (p = 0.31).	Not available	Not measured	-	Not measured
Dobbels et al., 2017 [64]	Heart, liver and lung transplant	No	Simple EAM	Both	Yes	EM, Self-report	0 Intervention group had a 16% higher dosing adherence post-intervention (95.1% intervention vs 79.1% control group; p<0.001), resulting in odds of adherence being 5 times higher in the intervention than in the control (odds ratio 5.17, 95% confidence interval 2.86–9.38).	Not available as reported as OR.	+	Event-free survival	5-year clinical event-free survival was 82.5% (intervention) vs 72.5% (control) relative risk 0.64, 95% CI 0.38–1.08; log-rank test p = 0.18) Overall: Higher rates of event-free survival (trend)
Duncan et al., 2013 [30]	Asthma	Yes	Simple EAM	Indirect	No	EM	++ Teamwork intervention group had significantly higher medication adherence rates—at 20 weeks, mean (SD) adherence: Intervention 81.0 (24.9), Asthma education group 33.6 (27.4), Control 37.0 (32.3).	+44%	++	Parental conflict, functional severity index for asthma, spirometry	Intervention improved mean (SD) scores from 0.7 (0.7) to 0.5 (0.5); Education group from 1.4 (1.1) to 0.9 (0.8), and control from 1.2 (0.8) to 1.3 (1.0). Overall: Improved asthma control as measured by functional severity index
Elixhauser et al., 1990 [63]	Bipolar affective disorder	No	Simple EAM	Indirect	No	Self-report, Prescription refill, Levels	0 No difference in adherence from any measure was detected at any visit time point–the percentage of prescription refills was higher (82% vs 69%) in the monitored group between visit 1 and 2. These differences disappear after both groups received feedback (between visits 2 and 3) on adherence (74% intervention vs 81% control). Overall prescription refill rate (76% intervention vs 73% control). In the monitored group, adherence was 83.2% with monitoring alone, then with feedback (and monitoring), it declined to 76.0%. No SD data for the percentages of prescriptions refilled.	+3%	--	Symptom Questionnaire	Monitored group patients had higher adjusted scores for anxiety (P = 0.03), symptoms of depression (P = 0.02), and somatism (P = 0.03). No significant differences between groups’ hostility scale scores were detected. Overall: Worsened anxiety, depression and somatic complaints
Erickson et al., 2005 [45]	Hypertension	Not stated	Integrated	Direct	No	EM (intervention only), Adherence questionnaire (Morisky self report scores; both groups, used to compare adherence between groups)	0 The change in mean (SD) adherence between the 2 groups was not significant (0.13 (0.4) (control) vs 0.34 (0.6) (intervention); p = 0.20) though a significant increase in self-reported adherence within the intervention group compared to baseline was seen. This may be affected by the lower baseline adherence in the intervention group (4.62 (0.6) (baseline) 4.96 (0.1) (post-intervention); p = 0.02).	Not available	+	BP control	Mean (SD) change in Systolic BP: 3.4 (18.8) (control) vs –7.6(6.9) (intervention); p = 0.07 Mean (SD) change in Diastolic BP: 0.2(7.1) (control) vs –4.4(10.4) (intervention); p = 0.13 Overall: Greater BP lowering (non-significant favouring intervention group)
Forni Ogna et al., 2016 [50]	Percutaneous coronary intervention with stent	No	Simple EAM	Indirect	No	EM, self-report	++ Medication adherence was higher and less variable in the intervention group: median (min–max) Intervention 101 (94–102)% vs. 99 (83–101) % in the usual care group; P < 0.001. Median (min–max) correct adherence 99 (93–100) % (intervention) vs. 98 (80–100) % (usual care); P < 0.001.	+2%	0	Platelet reactivity index–vasodilator-stimulated phosphoprotein phosphorylation–platelet reactivity index (VASP-PRI)	Baseline mean VASP-PRI was 48.3 ± 18.8%. No significant difference between groups was observed (Mean (SD): 47.0 (15.8) control vs. 47.3 (19.2) intervention; p = 0.761. Overall: No effect on clinical
Foster et al., 2014 [11]	Asthma	Yes	Simple EAM	Both	Yes	EM	++ Digital group mean (SD) adherence = 71 (34.8)% vs. control = 46 (32.5)%; p = 0.0003	+27%	+	ACT, exacerbations	Mean (SD) ACT in intervention = 4.8 (4.53) vs control 3.6 (4.37); mean change overall 4.5 (4.9); p<0.0001). Severe exacerbations were experienced by 11% of the patients in the intervention group and 28% of the patients in control group (P = 0.013) Overall: Improved asthma control and improved exacerbations
Frick et al., 2001 [38]	HIV	No	Simple EAM	Direct	No	EM, Pill count, Self-report	++ Intervention group significantly more likely to have good adherence (defined as ≥95%) than those in the control group (82% vs 36%). Median rate of daily adherence was 100% in the intervention vs. 93% in the control (P<0.001); median rates of hourly adherence were 97% vs 87% in the intervention vs. control groups respectively (P<0.001).	+46%	Not measured	-	Not measured
Gregoriano et al. 2019 [35]	COPD, Asthma	No	Simple EAM	Both	No	EM	++ Mean (SD) percentage of days in target range of 80–100% adherence: 81.6 (14.2)% intervention vs 60.1 (30.3)% control; p<0.001 for puff inhalers; 89.6 (9.8)% intervention 80.2 (21.3)% control; p = 0.01 for dry powder inhalers	+22% (puff inhalers); 9% (dry powder inhalers)	+	Exacerbations	Longer average time to the next exacerbation was observed in the intervention compared to the control group (102 days [95% CI, 76 to 128] vs. 86 days [95% CI, 66 to 106], p = 0.19). Intervention had no effect on time to first exacerbation (HR 0.65, 95% CI 0.21 to 2.07, p = 0.24), but showed a trend toward a 39% decreased frequency of exacerbations (RR = 0.61, 95% CI 0.35 to 1.03, p = 0.07). No significant differences in QoL were found between the two groups. Overall: Decreased frequency of exacerbations; longer average time to next exacerbation (trend only; not significant)
Hardstaff et al., 2003 [57]	Renal transplant	No	Simple EAM	Indirect	No	EM	0 No difference in adherence was noted between the groups, either before or after feedback (26% improved in feedback group vs 20% in control; 26% improved post-feedback vs 8% in control).	Not available	Not measured	-	Not measured
Henriksson et al. 2016 [58]	Renal transplant	Not stated	Simple EAM	Direct	Yes	EM (only in intervention group) (adherence not assessed in control group)	0 Adherence 97.8% in the intervention group. No comparison data as medication adherence in the control group was not assessed (group did not use EAM).	Not available	+	Rejection rate	Non significant positive intervention effect: Biopsy-verified rejection was three times more common among controls (13 vs. 4 patients; p = 0.054). Overall: Lower rejection rates (trend)
Hermann et al., 2011 [68]	Glaucoma	Yes	Simple EAM	No feedback	No feedback	EM	0 No difference in mean (SD) adherence between intervention and control groups (70 (17)% intervention vs 77 (6)% control for twice daily dosing; 65 (14)% intervention vs 62 (9)% control for three times daily dosing; p = 0.24).	-7% (BD dosing), +3% (TDS)	Not measured	-	Not measured
Joost et al., 2014 [51]	Renal transplant	No	Simple EAM	Indirect	No	EM, Pill count, Self-report, Q	++ Mean adherence (91% (95% CI, 90.52–91.94) intervention vs 75% (95% CI, 74.57–76.09) control, p = 0.014)	+16%	0	Transplant function (eGFR), transplant rejections, SF36 (HRQOL), HADS-D for depression/ anxiety	Two patients had rejection in the control and one in the intervention group (p = 0.54). Transplant function improved in both groups and were similar at study end (Intervention 49 (14.3) mL/min vs control 46 (15.4) mL/min; p = 0.446. Overall: No effect on clinical
Kozuki et al., 2006 [59]	Psychotic disorder	Not stated	Simple EAM	Indirect	No	EM, Drug levels, Pill count	++ Adherence rates of the intervention group slightly increased compared to a decline in the control group (P = 0.026) (rates at 12 weeks were 87.9% vs 68.4% intervention vs control).	+20%	+	Positive and Negative Symptom Scale (PANSS)	PANSS positive scores: Change from 20.7 (2.6) to 20.0 (3.9)(intervention) vs. 21.2 (4.3) to 23.0 (3.5)(control). PANSS negative scores: Change from 18.2 (4.1) to 17.9 (2.9)(intervention) vs. 18.7 (3.3) to 19.7 (4.0) (control). Neither change statistically significant between the two groups. Overall: Improved PANSS (trend, not significant)
Matteson-Kome et al., 2014 [26]	Inflammatory bowel disease	Not stated	Simple EAM	Indirect	No	EM	0 MEMS change scores for the intervention group increased (MEMS difference mean = −0.07; SD = 0.03) and the MEMS change scores for the control group decreased slightly (MEMS difference mean = 0.01, SD = 0.06). Between the two groups, the change scores were not statistically significant (P = 0.14; CI: -0.19–0.045).	Not available	Not measured	-	Not measured
McKenney et al., 1992 [46]	Hypertension	Yes	Simple EAM	Direct	Yes	Pill count	++ Intervention group had a higher average adherence (95.1% vs 78%; p = 0.0002); less variation in adherence (62–111% vs 16–110%); lower % of non-adherent patients taking ≤80% of doses (8% vs 50%).	+17%	++	BP control	Intervention had greater mean (SD) decreases in systolic BP (-7.64 (17.24) mmHg, p = 0.006 vs -2.79 (13.82) mmHg, p = 0.13) and diastolic BP (-8.78 (8.93), p<0.0001 mmHg vs -0.24 (7.50) mmHg, p = 0.43). Overall: Greater reduction in BP
Mehta et al., 2019 [49]	Hypertension	Not stated	Simple EAM	Direct	No	EM, Self-report (intervention group only–adherence not measured in control group)	n/a Only medication adherence reported for intervention group (70.8%); not measured in control	n/a	0	BP	Intervention mean (SD) change in systolic BP: -4.3 (21.5) vs control -4.7 (23.4); p = 0.94 Intervention mean (SD) change in diastolic BP: +6.5 (15.2) vs control +4.0 (12.6); p = 0.44 Overall: No effect on clinical
Morton et al., 2017 [32]	Asthma	No	Simple EAM	Both	Yes	EM	++ Intervention group mean (SD) = 70 (22.8)% vs control = 49 (26)%; p<0.001	+21%	++	ACQ, Oral steroid use, hospital admissions, FEV%	Mean (SD) ACQ: Intervention 1.14 (1.37) vs control (0.95 (1.37). Overall: Decrease oral steroid use and hospital admissions, but no effect on ACQ or FEV%
Murray et al., 2007 [52]	Congestive Heart Failure	Not stated	Simple EAM	Indirect	No	EM, Prescription refill, Self-report	++, 0 on f/u Overall taking adherence was 67.9% and 78.8% in the usual care and intervention groups, respectively (difference, 10.9 percentage points [95% CI, 5.0 to 16.7 percentage points]); but these effects disappeared in the 3-month post-intervention phase in which taking adherence was 66.7% and 70.6%, respectively (difference, 3.9 percentage points [CI, -5.9 to 6.5 percentage points])	+11%	++	HF exacerbations requiring ED/hospital admission, costs, disease specific quality of life using the CHF questionnaire	Emergency department visits and hospital admissions were overall 19.4% less (for all causes of admission, mean (SD) 2.94 (4.69) intervention vs 3.65 (6.26) control group; risk ratio 0.82, 95% CI 0.70–0.95). QoL improved from BL to 6- and 12-months to a greater extent in the intervention group (0.28 vs 0.21 at 6m (P = 0.52); 0.39 vs 0.24 at 12m (P = 0.21)) Overall: Reduced ED and hospital visits, greater improvement in QOL and patient satisfaction
Nides et al., 1993 [65]	COPD	Yes	Simple EAM	Indirect	No	EM, Self-report, Canister Weights	++ Mean (SD) % adherence: intervention 88.8 (9.6)% vs control 68.8 (25.7)%; p<0.0001	+20%	Not measured	-	Not measured
Okeke et al., 2009 [71]	Glaucoma	No	Simple EAM	Direct	No	EM, Self-report	++ Mean (SD) intervention group adherence 73 (22) vs. control 51 (30)%.	+22%	0	Intraocular pressure	Intraocular pressure did not change significantly from baseline in either groups (p = 0.81). Overall: No effect on clinical
Onyirimba et al., 2003 [31]	Asthma	No	Simple EAM	Indirect	No	EM	++ Mean weekly inhaled steroid adherence over the first week was not significantly different (61 ± 9% vs 51 ±5% treatment and control respectively). By second week, adherence was 81 ± 7% (treatment group) vs 47 ± 7% (control) (P = 0.003). Adherence remained >70% in the treatment group for the entire trial but decreased in the control group to below 30%.	+34%	0	FEV1, Asthma QOL Questionnaire	FEV1 did not change significantly from baseline in either group (+0.04 ± 0.11L and +0.16 ± 11L for treatment and control groups, P = 0.44). QOL scores improved significantly from BL in both groups (1.13 ± 0.31 units and 0.76 ± 0.33 tx vs control groups; P<0.05) but there was no group difference in the degree of improvement (P = 0.43). Overall: No effect on clinical
Reddy et al., 2016 [60]	Coronary artery disease	Not stated	Simple EAM	Both	No	EM	++ Control group adherence rate: 0.67(0.60,0.75); Individual feedback (0.89 (0.81,0.97); Feedback with partner (0.86(0.80,0.92))	+20%	0	LDL levels	Change in LDL at 6 months: Control -9.14 (-19.47,1.18) vs. individual feedback group -5.01 (-15.49,5.48), p = 0.85 vs. feedback with partner -4.60 (-13.55, 4.35), p = 0.79. Overall: No effect on clinical
Rigsby et al., 2000 [40]	HIV	Not stated	Simple EAM	Indirect	No	EM, Drug levels	++, 0 on f/u There was a significant increase over time in adherence for the intervention group with cash reinforcement (P = 0.0005) but not the cue dose training group alone (P = 0.79). Adherence declined in the 8 weeks after discontinuation of training and reinforcement to near-baseline levels. The improvement in adherence in the cash-reinforced group was accompanied by a significant loss of these gains in the follow-up period.	Not available	0	Viral load	There was no differences in viral load change between each of the groups (+0.64 in the reinforcement group, -0.29 in the cue dose training group, +0.34 control). Overall: No effect on clinical
Rosen et al., 2004 [53]	Diabetes	No	Simple EAM	Both	Yes	EM, Self-report	++ Mean improvement in adherence of 15% in the intervention group to metformin (~60% increased to 80% but control group remained at ~60%; P = 0.017); improvements noted for other antihyperglycemic medications too but as only 17 patients were on a second medication, this was not significant (P = 0.50). No difference between group when self-report used.	+15%	0	HbA1c	There was little change in either group’s HbA1c from baseline and HbA1c did not differ between groups. Overall: No effect on clinical
Rosen et al., 2007 [39]	HIV	Not stated	Simple EAM	Indirect	No	EM, Self-report, Drug levels	++ Mean MEMS-measured adherence to the reinforced medication increased from 61% to 76% in the 16-week treatment phase and was higher than the supportive counselling group (59% to 44%) (P = 0.01). This difference wast lost in the 16 weeks of follow-up—no difference between the groups (P = 0.07). Those receiving the intervention were more likely to achieve 95% adherence in weeks 1–16 than the control (P = 0.02).	+32%	++	Viral load, CD4	Proportion of patients with viral load <400 HIV-RNA/mL: Intervention 13/18 (72%) vs control 10/21 (48%); p = 0.12 At follow up: intervention 9/14 (64%); control 12/17 (71%); p = 0.71 Overall: Reduced viral load, but improvements lost on follow-up
Ruppar, 2010 [47]	Hypertension	Not stated	Simple EAM	Both	Yes	EM	++, 0 on f/u Treatment group had better antihypertensive medication adherence than the control group (median adherence improved 15.4% with the treatment group whereas the control group had -5.6% in adherence, P = 0.003). Change during the intervention from 75.5 to 96.4% (int) vs 34.1 to 16.40% (control); at week 20 (after 12 weeks follow-up) the intervention adherence = 94.3% vs 40% in control, p = 0.206.	+80%*	++	BP control	The intervention group’s median systolic BP lowered to 130 mm Hg (interquartile range (IQR), 17 mm Hg) vs. control where the median SBP increased to to 152 mm Hg (IQR, 61 mm Hg); p = 0.008). Overall: Greater reduction in BP
Russell et al., 2011 [54]	Renal transplant	Not stated	Simple EAM	Indirect	No	EM	++ Mean medication adherence score for intervention group higher than control (P = 0.03) with a large intervention effect size (Cohen’s d = 1.4; r = 0.6).	Not available	Not measured	-	Not measured
Sabin et al., 2010 [41]	HIV	No	Simple EAM	Indirect	No	EM, Self-report	++ Mean (SD) adherence at month 12 mean adherence had increased to 96.4 (3.4)% in the intervention group but was unchanged in the control at 84.1 (21.4)% (P = 0.003).	+12%	+/-	Viral load, CD4	A higher proportion of subjects experienced an increase in CD4 (71.0%) than the control (48.4%)(P = 0.07). Mean CD4 count at month 12 rose by 90.0 (171.6) cells/microlitre in the intervention group but declined by 8.8 (152.6) cells/microlitre in the control (P = 0.02). Proportion of subjects with HIV RNA <400copies/mL at month 12 did not differ significantly: 27/31 (87.1%) intervention vs 31/33 (93.9%); p = 0.3518. Overall: Improved CD4 count but not viral load
Smith et al., 2003 [42]	HIV	No	Simple EAM	Indirect	No	EM	++ Individuals in the intervention group significantly more likely to take 80% of more of their doses each week than individuals in the control group (OR = 7.8, 95% CI 2.2–28.1). Average weekly adherence was higher in the intervention group at all time points. Average adherence increased over time and by 12 weeks average weekly adherence in the intervention group was 96% vs 37% in the control group.	+59%	+	Viral load, CD 4	41% (9/22) of individuals in the Intervention group had at least one viral load of 400 copies or less vs. 24% of individuals (5/21) in the control group; p = 0.27. Overall: Reduced viral load (Trend, not significant)
Sulaiman et al. 2018 [33]	Asthma	Not stated	Simple EAM	Indirect	No	EM	++ Mean (SD) adherence: 73 (24) % (intervention) vs 63 (26) % (control), p<0.01	+10%	0	Asthma control	Data on peak flow, ACT, quality of life and adherence combined to produce asthma control measure. Reported no difference between groups Overall: No effect on clinical
Sutton et al., 2014 [61]	Diabetes	No	Simple EAM	No feedback	No feedback	EM (intervention only), Adherence questionnaire (both groups; used to compare adherence between groups)	0 Self-reported mean (SD) adherence using adherence questionnaire: intervention 24.2 (1.1) vs control 23.8 (1.9); p = 0.11.	Not available	+	HbA1c	Intervention mean (SD): 8.22 (1.30)% vs control: 8.39 (1.16); p = 0.25. Overall: Improved HbA1c (trend, not significant)
Tashkin et al., 1991 [66]	COPD	Yes	Simple EAM	Indirect	No	EM, Self-report, Canister Weights	++ In the control group 87% self-reported inhaler use at least twice a day, but only 52% had used it two or more times daily as per the chronology EM. In the feedback group, 89% self-reported adherence, but only 78% had used the inhaler two or more times daily from the chronology EM. The proportion of feedback participant with satisfactory compliance was significantly greater than the proportion of compliant uninformed participants (P<0.0001).	+26%	Not measured	-	Not measured
van Onzenoort et al., 2012 [48]	Hypertension	No	Simple EAM	No feedback	No feedback	EM, Pill count	0 Based on pill counts—median adherence did not differ between the two groups (96.1% (88.8–98.4 interquartile range) (intervention) vs 94.2% (control); P = 0.97).	+2%	0	BP control	Mean (SD) difference in systolic BP: intervention 23 (23) mmHg vs 22 (19)mmHg, p = 0.42 Mean (SD) difference in diastolic BP: intervention 13(13) mmHg vs 12 (11) mmHg, p = 0.62 Overall: No effect on clinical
Vasbinder et al. 2017 [34]	Asthma	No	Simple EAM	Direct	Yes	EM	++ Adherence in the intervention group was 69.3% (95% CI 65.5–73.4%) and 57.3% (95% CI 52.8–61.7%) in the control group. The overall difference was statistically significant: 12.0% (95% CI 6.7–17.7%)	+12%	0	Asthma control, QOL, exacerbations	No significant change in c-ACT (-1.07 (95% confidence interval -3.51–0.56),p = 0.203; quality of life as per PAQLQ (-.-6 (= 0.41–0.15,p = 0.659); and exacerbations per year (= 0.14 (-0.61–0.25,p = 0.432) Overall: No effect on clinical
Velligan et al., 2013 [70]	Schizophrenia	Not stated	Integrated	Both	No	EM, Pill count	++ Mean (SD) adherence 91 (17.46)% for intervention vs. 72 (17.26)% for the control group.	+19%	0	Brief Psychiatric Rating Scale (BPRS), ED/ hosp use, Social and Occupational Functioning Scale (SOFAS) for global function	For SOFAS scores and BPRS, results of mixed-effects regression models yielded no significant main effects or interactions (all P values > .09). For ED use, 13 out of 47 patients (27.7%) in the intervention group, and 16 out of 47 (34%) in control had contact with hospital or emergency psychiatric services; p = 0.77. Overall: No effect on clinical
Wilson et al., 2010 [43]	HIV	Not stated	Simple EAM	Indirect	No	EM, Self-report	0 There was a trend toward an intervention effect; adherence was 2.0% (95% CI -1.95 to 5.9) higher in the intervention group but not significant (P = 0.32).	+2%	0	Viral load	No significant differences in viral load–data not reported Overall: No effect on clinical
Wu et al., 2006 [44]	HIV	Not stated	Integrated	Direct	No	EM	++ Adherence was improved with 77% vs 57% (see Andrade et al.)	+20%	-	QOL scores, instrumental activities of daily living (IADLs), Centers for Epidemiologic Studies Depression Scale for depression, Medical Outcomes Study HIV Health Survey (MOS-HIV) for QOL	MOS-HIV scale–quality of life scores: -10.61 (intervention) vs. 8.05 (control); p = 0.06 IALDs: 1.48 (intervention) vs. -1.79 (control); p = 0.02 (higher scores, worse health) Depression scale: -1.54 (intervention) vs. -7.45 (control); p = 0.03 (higher scores, worse health) Overall: Worsened QOL (trend, not significant)
Yeh et al., 2017 [55]	Multiple sclerosis	Not stated	Simple EAM	Indirect	No	EM, Prescription refill, Self-report	0 Intervention had worse adherence than control using MEMS at 3 months Cohen’s d (-0.34) and 6 months (-0.7).	n/a	Not measured	-	Not measured

^#++ (significant improvement), + (trend in improvements), +/- (improvement in some parameters but not others), 0 (no effect),—(trend towards worsening), — (significant worsening)

ACQ = Asthma Control Questionnaire; ACT = Asthma Control Test; BP = blood pressure; CI = confidence interval; COPD = chronic obstructive pulmonary disease; EAM = electronic adherence monitoring; EM = electronic monitoring; HF = Heart Failure; HIV = Human Immunodeficiency virus; QoL = Quality of Life