AJH 2000;13:184 190 Electronic Pill-Boxes in the Evaluation of Antihypertensive Treatment Compliance: Comparison of Once Daily Versus Twice Daily Regimen Michel Andrejak, Nathalie Genes, Laurent Vaur, Pascal Poncelet, Pierre Clerson, and Alain Carré The objective was to compare the compliance of hypertensive patients treated with captopril twice daily or trandolapril once daily. After a 2-week placebo period, hypertensive patients (diastolic BP 95 115 mm Hg) were randomly allocated to trandolapril 2 mg once daily or to captopril 25 mg twice daily for 6 months. and captopril were packed in electronic pill-boxes equipped with a microprocessor that recorded date and time of each opening (MEMS ). Patients compliance was assessed both by standard pillcount and by electronic monitoring. Blood pressure was measured using a validated semi-automatic device at the end of the placebo period and of the treatment period. One hundred sixty-two patients entered the study. Compliance data were evaluable for 133 patients (62 in the captopril group and 71 in the trandolapril group). Treatment groups were comparable at baseline except for age (P.046). Using electronic pill-box, overall compliance was 98.9% in the trandolapril group and 97.5% in the captopril group (P.002). The percentage of missed doses was 2.6% in the trandolapril group and 3.3% in the captopril group (P.06). The percentage of delayed doses was 1.8% in the trandolapril group and 11.7% in the captopril group (P.0001). The percentage of correct dosing periods, ie, a period with only one correct recorded opening, was 94.0% in the trandolapril group and 78.1% in the captopril group (P.0001). Results were unchanged when adjusted for age. At the end of the study, 41% of patients in the trandolapril group and 27% in the captopril group (NS) had their blood pressure normalized (systolic BP <140 and diastolic BP <90 mm Hg). In this 6-month study, the electronic pill-box allowed refined analysis of compliance of hypertensive patients. Patients compliance with once daily trandolapril was higher than with twice daily captopril. The between-group difference is mainly explained by an increase in delayed doses in the twice daily group. Am J Hypertens 2000;13:184 190 2000 American Journal of Hypertension, Ltd. KEY WORDS: Electronic pill-box, pill-count, treatment compliance, ACE inhibitor, treatment regimen. Received February 1, 1999. Accepted June 8, 1999. From the Service de Pharmacologie Clinique, Amiens, France (MA); Laboratoires Roussel Diamant, Medical Department, Paris, France (NG, LV); CMC Darcy, Henin-Beaumont, France (PP); Orgamétrie, Mouvaux, France (PC); and Hôpital Cardiologique, Lille, France (AC). Address correspondence and reprint requests to Dr. Nathalie Genès, Laboratoires Roussel Diamant, Tour HMR, 1 Terrasse Bellini 92910 Paris la Défense Cedex, France. 2000 by the American Journal of Hypertension, Ltd. 0895-7061/00/$20.00 Published by Elsevier Science, Inc. PII S0895-7061(99)00175-2
AJH FEBRUARY 2000 VOL. 13, NO. 2 COMPLIANCE AND TREATMENT REGIMEN 185 In asymptomatic chronic diseases such as hypertension, treatment compliance is frequently low and difficult to improve. 1 Low compliance has been associated with poorer clinical prognosis. Interruption of drug intake by the patient will bring about a period without effective drug activity. 2 In certain cases the sudden stopping of therapy might well lead to potentially serious withdrawal symptoms, including major cardiovascular complications. 3 6 On the other hand, excessive drug effects might occur when dosing is suddenly resumed. 2 Adequate measurement of compliance is an important challenge. The patient s self-report as well as pillcount are usually used, due to their costs and facility. The use of sophisticated techniques such as biologic assays is restricted to research. All these methods have shown their limitations. 7 9 The development of electronic monitors for medication bottles that register the date and the time of each opening constitutes a marked improvement. 10 Unlike the previously described methods, such devices allow information on the time interval between the doses of medication. Adherence to the instructions regarding treatment may thus be assessed. 11 The MACH 3 study is a part of the French MACH (Medication event monitoring system for the Assessment of the Compliance of Hypertensives) research program. The MACH 1 and MACH 2 studies aimed to determine factors affecting the short-term compliance of hypertensive patients treated with a once daily ACE inhibitor. 12,13 The aim of the present study was to compare the compliance of hypertensive patients treated either with captopril twice daily or trandolapril once daily. PATIENTS AND METHODS Patients Patients aged 18 or over with mild to moderate essential hypertension were eligible for this study. After a 2-week single-blind placebo run-in period, the patients with an office sphygmomanometer diastolic blood pressure (DBP) between 95 and 115 mm Hg were included in the study. The main exclusion criteria were: any serious chronic disease, known hypersensitivity or contraindication to ACE inhibitors, impaired renal function (serum creatinine 130 mol/l), dyskalaemia, and concomitant antihypertensive drug intake. All patients gave their written informed consent. The protocol was approved by the Comité Consultatif de Protection des Personnes se prêtant à la Recherche Biomédicale of Amiens, France (Ethics Committee). Study Design This was a multicenter, nonblinded, randomized, parallel-group study. At the screening visit, eligibility criteria were assessed, written informed consent was obtained, and all present antihypertensive drug therapy was discontinued. After a 2-week placebo run-in period, eligible patients were randomly allocated to either trandolapril, 2 mg once daily, or captopril, 25 mg twice daily, for 6 months according to the parallel design. The patients were told to take their medication at 8:00 am for the trandolapril group, whereas for the captopril group, the patients were told to take their medication at 8:00 am and at 8:00 pm. In nonresponder patients (diastolic BP 90 mm Hg or diastolic BP decrease 10 mm Hg) on the 60th day or on the 120th day, furosemide (20 40 mg) was added. Measurement of Compliance Assessed by MEMS A Medication Event Monitoring System (MEMS, Aprex Corporation, Fremont, CA) was used to assess compliance patterns during the active treatment phase of the study. and captopril were packed in standard pill bottles fitted with a specialized cap containing a microprocessor, which registers date, time, and duration of each opening of the cap. Patients were made aware of its presence and purpose. Patients were given instructions in how to use the special cap on their medication bottles. They were instructed to open the pill bottle only to take their pill. Each patient had the same electronic cap throughout the study and the containers were only changed at Day 60 and Day 120. For each patient, all calculations were derived from the date and the time of each opening. Each opening recorded by the pill-box was considered a single dose intake. We defined for captopril patients two 12-h dosing periods: 2:00 am 1:59 pm and 2:00 pm 1:59 am, whereas for trandolapril patients a single 24-h dosing period (2:00 am 1:59 am) was defined. Four quantitative indices were used to assess the individual compliance: overall compliance, missed doses, delayed doses, and correct dosing periods. Overall compliance was defined by the number of pill-box openings. Missed doses were defined by the absence of recorded openings during the defined dosing periods. Delayed doses were defined by the opening of the pill bottle more than 26 h after the preceding opening for the patients receiving the once daily regimen (trandolapril). For patients receiving the twice daily regimen (captopril), an interval of more than 13 h defined a delayed dose. A dosing period was seen to be correct when including one single correct recorded opening (ie, without a missed dose, a delayed dose, or an extra dose). So dosing periods were judged correct when a single opening occurred within 13 h (captopril) or 26 h (trandolapril) after the preceding dose. Overall compliance, missed doses, delayed doses, and correct dosing periods were expressed as
186 ANDREJAK ET AL AJH FEBRUARY 2000 VOL. 13, NO. 2 percentages of prescribed doses (treatment duration number of daily doses). Measurement of Compliance Assessed by Pill-Count Standard pill-count was also used to assess compliance. At each visit, the number of handed-in pills in the containers was counted. Compliance was defined by the number of pills taken (dispensed pills minus returned pills) in relation to the theoretical number of prescribed doses and was expressed as a percentage. Blood Pressure Measurements Sphygmomanometer office blood pressure was recorded at the end of the placebo run-in period to check that inclusion criteria had been met. Physicians were asked to perform, using a mercury sphygmomanometer, three consecutive measurements while the patients were in the sitting position after a 5-min rest. The systolic blood pressure (SBP) was defined using phase I of the Korotkoff s sound and DBP by phase V. The average of the three readings was defined as office SBP and DBP. Automatic office blood pressure was performed by the physician at the end of the placebo run-in period, at each follow-up visit (Day 60, Day 120), and at the end of the active treatment period (Day 180). Three consecutive measurements were performed with the patient in the sitting position after a 5-min rest using thea& DUA751 device (A & D Engineering, Milipitas, CA). This device is a printer-equipped, semi-automatic, digitized one and uses the oscillometric method. It was previously validated by comparison with a random-zero mercury sphygmomanometer. 14 The average of the three readings was used to define the automatic SBP and DBP. These automatic blood pressure data were used to assess antihypertensive efficiency. Data Processing and Statistical Analysis At the end of the study, data retrieval and analysis for MEMS was done through specific Aprex software. Compliance data were then entered into a SAS data set (SAS Institute, Cary, NC). Quantitative data were expressed in terms of means SD and were compared using Student s t test (intergroup comparisons). Qualitative variables were expressed as percentages and were compared using a 2 test. As compliance data did not follow a normal distribution, they were expressed by their median and 10th and 90th percentiles and were compared using the Wilcoxon test. The statistical significance threshold was chosen as P.05. RESULTS Baseline Characteristics of Patients One hundred and sixty-two patients entered the active treatment period. Seventy-eight patients were randomized to the captopril group and 84 to the trandolapril group. TABLE 1. DEMOGRAPHICS AND CHARACTERISTICS OF PATIENTS Age (years) 55 12 59 12.046 Male (%) 44 47.97 BMI (kg/m 2 ) 28 5 28 4.26 Hypertension duration (years) 6 6 5 6.21 SBP (mm Hg) 163 18 161 18.65 DBP (mm Hg) 100 11 99 9.99 Twenty-nine patients (16 in the captopril group and 13 in the trandolapril group) were excluded from the compliance analysis due to withdrawals for adverse events (10 patients), lack of pill-box data (8 patients), withdrawals for patients convenience (3 patients), or major protocol violation (8 patients). Analysis was thus performed on 133 patients (62 in the captopril group and 71 in the trandolapril group). The patients demographic and clinical characteristics are indicated in Table 1. All demographic characteristics (except for age) and initial blood pressure data were comparable in both treatment groups. The patients in the captopril group were slightly older than the patients in the trandolapril group (P.046). Overall Compliance The average treatment duration was 183 11 days in the trandolapril group and 182 12 days in the captopril group (P.70). The overall compliance assessed by pill-count during the 6-month period was 98.9% in the trandolapril group and 97.6% in the captopril group. This difference was statistically significant (P.049). The same difference was observed when assessed by electronic pill-boxes: 98.9% in the trandolapril group and 97.5% in the captopril group (P.002). Although statistically significant, these differences between the two groups were not clinically relevant: the overall compliance was over 80% in the majority of patients. Whatever the method, the difference in overall compliance between the two groups increased with time as shown in Table 2. No difference was observed during the first 2 months. The results became significant during the next 2 months. This statistical significance increased during the last 2 months. However, a significant time treatment interaction was only observed for electronic pill-box data (P.02). Assessment of Daily Compliance by Electronic Pill- Boxes Missed Doses The distribution of missed doses is presented in Figure 1. In the trandolapril group, the patients missed 2.6% of doses (5 doses) during the 6-month treatment and 15.4% of patients forgot more than 10% of doses. In the captopril group,
AJH FEBRUARY 2000 VOL. 13, NO. 2 COMPLIANCE AND TREATMENT REGIMEN 187 TABLE 2. COMPARISON OF OVERALL COMPLIANCE OVER TIME, ASSESSED BY PILL COUNT OR ELECTRONIC PILL-BOX Overall Compliance (%), Pill-Count Overall Compliance (%), Electronic Pill-Box Day 0 Day 180 98.9 (89.0 101.5) 97.6 (87.4 100.3).049 98.9 (89.1 101.1) 97.5 (83.3 100.0).002 Day 0 Day 60 98.5 (89.7 100.0) 98.3 (90.9 102.9).99 98.5 (90.2 101.7) 99.1 (88.3 100.9).51 Day 60 Day 120 98.5 (86.5 101.6) 97.5 (78.6 100.8).04 100.0 (87.5 101.8) 97.7 (83.9 100.0).01 Day 120 Day 180 100.0 (89.7 102.6) 97.3 (85.3 100.7).01 100.0 (90.5 103.2) 96.7 (75.8 99.3).0005 Time treatment interaction.10.02 Data are expressed as median (10th and 90th percentiles). the patients missed 3.3% of doses (12 doses) and 22.5% of patients forgot more than 10% of doses. The between-group comparison is shown in Table 3: during the 6-month treatment, the difference was close to significance (P.06). However, a time treatment interaction was found (P.03). The difference between the two groups became highly significant the last 2 months (P.003). Delayed Doses The distribution of delayed doses is presented in Figure 2. In the trandolapril group, the patients had 1.8% of the doses delayed (3 doses) and 8.4% of patients delayed more than 10% of doses. In the captopril group, the patients had 11.7% of the doses delayed (41 doses) and 59.7% of patients delayed more than 10% of doses. As shown in Table 3, the between-group difference was highly significant throughout the whole study (P.0001). Correct Dosing Period The percentage of correct dosing periods during the study was 94.0% in the trandolapril group and 78.1% in the captopril group (P.0001). This difference was observed throughout the study (Table 3). These results remained unchanged when adjusted for age. FIGURE 1. Distribution of patients according to the percentage of missed doses during the 6-month period.
188 ANDREJAK ET AL AJH FEBRUARY 2000 VOL. 13, NO. 2 TABLE 3. ASSESSMENT OF DAILY COMPLIANCE BY ELECTRONIC PILL-BOX Missed Doses (%) Delayed Doses (%) Correct Dosing Periods (%) Day 0 Day 180 2.6 (0 14.1) 3.3 (0.8 18.6).06 1.8 (0 8.2) 11.7 (5 31.3).0001 94.0 (75.4 99.0) 78.1 (52.2 92.5).0001 Day 0 Day 60 1.7 (0 11.8) 1.8 (0 12.5).6 1.7 (0 8.2) 11.7 (2.7 37.0).0001 92.4 (77.8 100) 80.6 (52.9 95.8).0001 Day 60 Day 120 1.7 (0 12.5) 2.8 (0 17.1).055 1.7 (0 7.9) 13.7 (5.1 32.1).0001 94.5 (73.0 100) 77.8 (51.5 92.2).0001 Day 120 Day 180 2.1 (0 12.5) 4.3 (1.4 24.2).003 1.7 (0 7.1) 11.4 (2.0 36.0).0001 93.3 (72.2 100) 77.5 (51.4 93.7).0001 Time treatment interaction.03.17.031 Data are expressed as median (10th and 90th percentiles). Antihypertensive Efficiency An addition of furosemide was needed for 40% of patients in the captopril group as well as for 46% of patients in the trandolapril group (NS). At the end of the 6-month period, the SBP decrease was 14.2 21 mm Hg in the captopril group as opposed to 17.4 17.7 mm Hg in the trandolapril group (P.22). The DBP decrease was respectively 7.7 12.6 mm Hg and 10.8 12.5 mm Hg (P.14). In the trandolapril group 58% of patients were responders (DBP 90 mm Hg or DBP fall 10 mm Hg), while 45% of patients responded to captopril (P.15). The percentage of normalized patients in each treatment group is shown in Table 4. At the end of the study, 41% of patients in the trandolapril group and 27% in the captopril group (NS) had their blood pressure normalized (SBP 140 and DBP 90 mm Hg). Safety The percentage of patients with adverse events was not statistically different in the two groups: 6.7% in the captopril group and 2.2% in the trandolapril group. DISCUSSION In the present study, we assessed patterns of compliance in hypertensive patients treated over 6 months. The overall compliance during that 6-month study assessed either by pill-count or by electronic pill-box was over 95%. Similar results were found in short- FIGURE 2. Distribution of patients according to the percentage of delayed doses during the 6-month period.
AJH FEBRUARY 2000 VOL. 13, NO. 2 COMPLIANCE AND TREATMENT REGIMEN 189 TABLE 4. EVALUATION OF BLOOD PRESSURE NORMALIZATION, N 71 (% of patients), N 62 (% of patients) P SBP 140 mm Hg 47.9 40.3 NS DBP 90 mm Hg 59.2 38.7.02 SBP 140 mmh and DBP 90 mm Hg 40.9 24.4 NS term studies where compliance ranged 82% 95% after 4 10 weeks of treatment. 11,12,15,16 However, long-term compliance has been reported to be lower. After a 6-month follow-up, Sackett et al 17 reported that only 57% of patients took more than 80% of the prescribed doses. Following a 12-month study, Guerrero et al 18 found that compliance declined with time. However, Leenen et al reported a high degree of compliance in patients treated during 20 weeks by either amlodipine or diltiazem. 19 The nearly optimal compliance found in our study can be explained by the following facts: 1. The fact that patients had the electronic pill-box fully explained as a means of following their treatment may have encouraged them to be more compliant. It has been previously reported that the patients compliance increased from 78% 91% when information concerning data recording by the pill dispenser was provided. 10 2. The participation in a randomized well-controlled clinical trial may have led the investigators to select the most compliant patients. Consequently, the rate of compliance that we found was probably overestimated. 3. The withdrawals for adverse events or for the patients convenience were excluded from the analysis. Those patients were, however, the more susceptible to poor compliance. The result of the analysis revealed a significant difference in compliance with therapy between the two treatment groups that clearly favored the trandolapril group. Seeing that a) captopril and trandolapril both belong to the same therapeutic class, ie, ACE inhibitors; b) they were shown to have a similar tolerance in this study; and c) the withdrawals were excluded from the analysis, this difference is therefore most likely to be related to the daily regimen. Although crude, pill-count exhibited a significant between-group difference after 2 months of treatment. However, results were more consistent when using parameters derived from the Medication Event Monitoring System. When assessed by the correct dosing period method, compliance was 78.1% in the twice daily group, as opposed to 94.0% in the once daily group (P.0001). It is logical to believe that compliance is related to the number of prescribed daily doses. Indeed it was easily demonstrated that compliance improved from a three or four times daily regimen to a once daily regimen. 20 22 However, in those studies, no difference in compliance was found between once and twice daily regimens. Three recent studies using MEMS have shown a lower compliance for the twice daily regimen compared with the once daily. 15,19,23 Detry et al, 15 using a parameter close to our correct dosing period, showed that compliance with amlodipine once daily was superior to slowrelease nifedipine twice daily (85% v 64%, P.001). In a similar manner, the percentage of days with correct number of doses taken was significantly higher in patients receiving amlodipine once daily (90%) when compared with patients receiving diltiazem twice daily (82%, P.01). 19 Determining of the number of correct dosing periods depends on the number of missed doses as well as the delayed ones. Obviously the weight of the number of delayed doses plays a leading part in this determination. The number of delayed doses is markedly increased in the twice daily regimen (11.7% v 1.8%, P.0001). The definition of a delayed dose varies from one study to another. The chosen threshold may be expressed either in hours (eg, 26 h for once daily dosages) or in the percentage of the theoretical interval of drug intake (eg, 25%). The latter is readily used as it allows a single definition whatever the drug regimen. 24 However, it might appear unfair, seeing that the delay allocated for each drug intake for a twice daily regimen is half that of a once daily regimen. In the present study, a dose was defined as delayed when the box opening occurred with a delay of 8.5% of the theoretical interval of the drug intake (13 h for a twice a day regimen and 26 h for a once a day regimen). The number of missed doses is marginally increased in the twice daily regimen (3.3% v 2.6%, P.06). However, the betweengroup difference increased with time. At the end of the study, the percentage of missed doses was twice that in the captopril group (4.3% v 2.1%, P.0003). Increasing compliance using once daily antihypertensive drugs may lead to a better blood pressure control. 2 In the SHEP pilot study, it has been shown that compliance 80% was associated with a greater probability of achieving target blood pressure reduction. 25 Mallion et al 11 found a significant correlation between single daily compliance assessed by electronic pill-box and blood pressure control. In our study, we failed to demonstrate a relationship between compliance and blood pressure decrease. However, this could be explained by the discrepancy between the single blood pressure evaluation and the daily compliance evaluation throughout a 6-month period. In addition, we did not find a difference in
190 ANDREJAK ET AL AJH FEBRUARY 2000 VOL. 13, NO. 2 blood pressure control between the two groups, whereas their 6-month compliance significantly differed. This might be explained by an increase in compliance in both groups during the few days preceding the visits, as the compliance to drug therapy is known to be better during the days before a consultation. 16 CONCLUSION In this 6-month study, the electronic pill-box allowed refined analysis of compliance of hypertensive patients. Using this device, the present study has shown that patient compliance with once daily trandolapril was higher than with twice daily captopril. The between-group difference is mainly explained by an increase in delayed doses in the twice daily group. REFERENCES 1. Sackett DL, Gibson ES, Taylor DW, et al: Randomised clinical trial of strategies for improving medication compliance in primary hypertension. Lancet 1975;1: 1205 1207. 2. Meredith PA: Therapeutic implications of drug holidays. Eur Heart J 1996;17(suppl A):21 24. 3. Houston MC: Beta-adrenergic blocker withdrawal syndromes in hypertension and other cardiovascular diseases. Am Heart J 1988;116:515 522. 4. Psaty JM, Koepsell TD, Wagner EH, et al: The relative risk of incident coronary heart disease associated with recently stopping the use of beta-blockers. JAMA 1990; 263:1653 1657. 5. Rangno RE, Langlois S: Comparison of withdrawal phenomena after propanolol, metoprolol and pindolol. Am Heart J 1982;104:473 478. 6. Weber MA: Discontinuation syndrome following cessation of treatment with clonidine and other antihypertensive agents. J Cardiovasc Pharmacol 1980; 2(suppl 1):573 589. 7. Pullar T, Kumar S, Tindal H, et al: Time to stop counting the tablets? Clin Pharmacol Ther 1989;46:163 166. 8. Rudd P, Byyny RL, Zachary V, et al: Pill count measures of compliance in a drug trial: variability and suitability. Am J Hypertens 1988;1:309 312. 9. Stewart M: The validity of an interview to assess a patient s drug taking. Am J Prev Med 1987;3:95 100. 10. Kruse W, Weber E: Dynamics of drug regimen compliance: its assessment by microprocessor-base monitoring. Eur J Clin Pharmacol 1990;38:561 565. 11. Mallion JM, Meilhac B, Tremel F, et al: Use of a microprocessor-equipped tablet box in monitoring compliance with antihypertensive treatment. J Cardiovasc Pharmacol 1992;19(suppl 2):S41 S48. 12. Mallion JM, Dutrey-Dupagne C, Vaur L, et al: Benefits of electronic pill-boxes in evaluating treatment compliance of patients with mild to moderate hypertension. J Hypertens 1996;14:137 144. 13. Vaur L, Vaisse B, Genès N, et al: Use of electronic pill-boxes to assess risk of poor treatment compliance: results of a large scale trial. Am J Hypertens 1999;12: 374 380. 14. Jamieson MJ, Webster J, Witte K, et al: An evaluation of thea&dua-751 semi-automated cuff-oscillometric sphygmomanometer. J Hypertens 1990;8:377 381. 15. Detry JMR, Black P, De Backer G, et al: The Belgian Collaborative Study Group: Patient compliance and therapeutic coverage: comparison of amlodipine and slow release nifedipine in the treatment of hypertension. Eur J Clin Pharmacol 1995;47:477 481. 16. Rudd P, Ahmed S, Zachary AS, et al: Improved compliance measures: applications in an ambulatory hypertensive trial. Clin Pharmacol Ther 1990;48:676 685. 17. Sackett DL, Haynes RB, Gibson ES, et al: Patient compliance with antihypertensive regimens. Patients Council. Health Educ 1978;1:18 21. 18. Guerrero D, Rudd P, Bryant-Kosling C, et al: Antihypertensive medication-taking. Investigation of a simple regimen. Am J Hypertens 1993;6:586 592. 19. Leenen FHH, Wilson TW, Bolli P, et al: Patterns of compliance with once versus twice daily antihypertensive drug therapy in primary care: a randomized clinical trial using electronic monitoring. Can J Cardiol 1997;13:914 920. 20. Cramer JA, Mattson RH, Prevey ML, et al: How often is medication taken as prescribed? A novel assessment technique. JAMA 1979;261:3273 3277. 21. Eisen SA, Miller DK, Woodward RS, et al: The effect of prescribed daily dose frequency on patient medication compliance. Arch Intern Med 1990;150:1881 1884. 22. Fujii J, Seki A: Compliance and compliance-improving strategies in hypertension: the Japanese experience. J Hypertens 1985;3(suppl 1):19 22. 23. Waeber B, Erne P, Saxenhofer H, etal: Useofdrug with more than a twenty-four-hour duration of action. J Hypertens 1994;12(suppl 3):S67 S71. 24. Lee JY, Kuseck JW, Greene PG, et al: Assessing medication adherence by pill count and electronic monitoring in the African American study of kidney disease and hypertension (AASK) pilot study. Am J Hypertens 1996;9:719 725. 25. Black DM, Brand RJ, Greenlick M, et al: Compliance to treatment for hypertension in elderly patients: the SHEP pilot study. Systolic Hypertension in the Elderly Program. J Gerontol 1987;42:552 557.