Twenty-nine trials with a total of 12,835 participants were included. The included studies evaluated: electronic vial caps (1 study, n=70), calendar packaging (5 studies, n=796), patient cards (2 studies, n=303), dosing regimen (4 studies, n=8,288), physician education (2 studies, n=301), patient education (5 studies, n=1,512), disease management (1 study, n=39), worksite care (2 studies, n=651), counselling (5 studies, n=524), and self-monitoring (2 studies, n=351).
Electronic vial cap (1 study): the odds of a patient being compliant in the electronic vial cap group were about six times higher that those in the control group; both groups had a similar effect on diastolic blood-pressure (no exact data reported).
Calendar packaging (5 studies): the summary OR for pill counts was 1.7 (95% CI: 1.4, 2.0, p-value not reported) for 3 trials. The remaining 2 trials either did not show a statistically-significant result or did not report adherence.
Patient cards (2 studies): both trials resulted in statistically-significant positive effects with a combined OR of 2.9 (95% CI: 2.7, 3.1, p-value not reported). One study reported data for blood-pressure, but the result was non significant.
Dosing regimen (4 studies): adherence results could only be calculated for one of the studies (OR 1.0, 95% CI: 0.7, 1.44, p-value not reported). The effect on blood-pressure was not statistically significant. The authors felt that the data do not provide support for the effectiveness of a simplified dosage regimen.
Reminders (3 studies): the results were statistically significant in the 2 studies that used mixed interventions (no exact data reported). The authors stated that there was insufficient evidence to demonstrate the effectiveness of reminders alone.
Physician education (2 studies): one study showed that the odds of a patient of a physician being compliant in the experimental group was 3.2 times higher than for those in the control group (no exact data reported), whereas the results of the other study were not statistically significant. Both studies showed statistically-significant results for blood-pressure (no exact data reported).
Patient education (5 studies): there were conflicting results of trials of patient education in hypertension (no exact data reported).
Worksite care (3 studies): in each of these trials, the summary statistics suggested a small statistically-significant positive effect on adherence and an improvement in blood-pressure was reported (no exact data reported).
Counselling (7 studies): adherence could be calculated for 6 of the 7 studies. Two studies showed statistically-significant positive results (exact data not reported). However, the results were conflicting and, overall, did not demonstrate that counselling per se was effective.
Self-monitoring (5 studies): adherence could only be calculated for 3 of the 5 studies. Only one of these showed a statistically-significant effect for both adherence and blood-pressure (no exact data reported).