User login
Durability of Improvement Achieved in a Clinical Trial: Is Compliance an Issue?
METHODS: Nineteen previously untreated middle-aged men with hypertension had their office and ambulatory blood pressure recorded after 4 weeks of placebo treatment, 4 weeks of active treatment in a clinical trial, and 31 months of treatment in clinical practice. All recording was done by the same physician (IE).
RESULTS: Mean 24-hour blood pressure was 138/92 mm Hg after 4 weeks of placebo treatment, 128/85 mm Hg after 4 weeks of active treatment in the clinical trial, and 136/87 mm Hg after a mean of 31 months of treatment in clinical practice. The corresponding blood pressure values Ž140/90 mm Hg during the daytime were 47%, 24%, and 39%, and office blood pressures were 155/101, 145/93, and 150/91 mm Hg. Individual comparison revealed that 6 of the 19 patients had higher mean 24-hour blood pressure after several months of treatment in clinical practice than after 4 weeks of active treatment in the clinical trial.
CONCLUSIONS: In our study, the significantly reduced blood pressure in the clinical trial did not persist when followed up in clinical practice. At follow-up, one third of the patients had blood pressure values similar to those before active treatment. The reason for this is unclear, but inconsistent compliance may play a part in the lack of durability of the improvements. Our results indicate that effects seen in short-term clinical trials may not translate to long-term benefits in clinical practice.
Poor compliance with prescribed drug regimens attenuates the benefits of treatment, making compliance a key link between process and outcome in ambulatory care.1 In clinical trials, noncompliance, also called nonadherence, is one of the most important sources of the variance in drug response.2 Inconsistent compliance has the largest impact on statistical power, since it lowers the mean drug response and adds variance.3 Deficient compliance can also lead to overestimates of dosage requirements. The rate of compliance differs in clinical trials4,5 and in clinical practice.4,6,7 It has been calculated that depending on the method used to monitor compliance, only 20% to 80% of patients treated for hypertension can be considered good compliers.7
In our pilot study we examined the persistence of blood pressure effects 31 months after a clinical trial of 4 weeks of treatment with hypotensive agents.
Methods
Our study included 19 previously untreated middle-aged men with hypertension (mean age=52.1 years; range=40-64 years) recruited from a blood pressure screening project in a municipality in southern Sweden.8 They were followed up in 1991-1992 in clinical practice after having completed a trial comparing 50 mg atenolol with 20 mg enalapril, which were found to be equally effective in lowering blood pressure.9 Office and ambulatory blood pressures recorded after 4 weeks of placebo treatment and after 4 weeks of active treatment in the clinical trial were compared with office and ambulatory blood pressures recorded after several months of treatment in a clinical practice at the same health center. The same physician (IE) involved in the clinical trial followed up and treated the patients in her practice at the health center. Although some patients were treated with the same drugs as in the clinical trial, most had a different treatment ([Table 1]). Side effects were the most common reason for a patient to change treatment. In the clinical trial, ambulatory blood pressure was recorded every 15 minutes between 6 AM and 6 PM and every 30 min between 6 PM and 6 AM with the SpaceLabs 5200 equipment (Chatsworth, Calif). At the follow-up in clinical practice, ambulatory blood pressure was recorded every 20 minutes with SpaceLabs 90207 equipment (Redmond, Wash). Both these equipments have been validated and found to be accurate.10-12 The equipment was attached to the patient between 8 AM and 10 AM in the clinical trial and at other various times of the day, but mostly between 2 PM and 5 PM when followed up in clinical practice. Office blood pressure was recorded as the mean of 4 measurements obtained by the same physician in the same way (supine position after 5-10 minutes of rest with appropriate cuffs), twice when the ambulatory blood pressure equipment was attached and twice when it was removed from the patient. All patients worked full time during the clinical trial; in the follw-up in clinical practice 7 were retired and 1 had part-time work.
Statistical Analysis
Standard statistics such as mean, standard deviation, minimum, and maximum were used to summarize the data. Mean blood pressure values are presented together with 95% confidence intervals, which were calculated using normal approximation. Daytime was defined as the time between waking and going to bed.
Results
Mean (standard deviation [SD]) 24-hour blood pressure was 138/92 (14/5) mm Hg after 4 weeks of placebo treatment, 128/85 (14/7) mm Hg after 4 weeks of active treatment, and 136/87 (18/9) mm Hg after 31(19) months of treatment in clinical practice. Confidence intervals are given in [Table 2]. The corresponding blood pressure values Ž140/90 (SD) mm Hg during the daytime were 47% (25%), 24% (22%), and 39% (29%), and office blood pressures were 155/101 (16/4), 145/93 (17/6), and 150/91 (23/9) mm Hg. Individual comparison revealed that 6 patients had a higher mean 24-hour blood pressure after several months of treatment in clinical practice than after 4 weeks of active treatment in the clinical trial. Of these 6 patients, 3 had the same treatment (atenolol 50 mg) as in the clinical trial, 1 had changed from enalapril to atenolol, 1 from enalapril to nifedipine, and 1 from atenolol to diltiazem.
Discussion
We examined the persistence of blood pressure effects 31 months after a clinical trial of hypotensive agents. The results showed that mean ambulatory blood pressure was lower after 4 weeks of treatment in the clinical trial than after several months of treatment in clinical practice. Although 4 weeks of treatment with either 50 mg atenolol or 20 mg enalapril in the clinical trial reduced ambulatory blood pressure significantly, these lower blood pressure values did not persist after a longer period of treatment. Individual analysis revealed that 6 patients had higher blood pressure values at follow-up in clinical practice than in the clinical trial. In our clinical trial, all patients except 1 (95%) had reduced their ambulatory blood pressure during active treatment.
In any case of poor response to therapy, it is important to suspect defective compliance. It has been estimated that 16% to 50% of patients with newly diagnosed hypertension discontinue their antihypertensive medications within the first year.6 Electronic monitoring methods have revealed that many patients omit prescribed doses irrespective of disease, prognosis, or symptoms.1 Subjects at particular risk of poor compliance with antihypertensive drugs seem to be middle-aged men still active in work and without previous cardiovascular disease.6 In our study, however, blood pressure was lower during the clinical trial, when all patients were still active in work, than at the follow-up when some patients were retired. In fact, 2 of the patients whose blood pressure was not controlled at the follow-up were retired at that time.
It might be argued that changed antihypertensive treatment could be the reason for the higher blood pressure at follow-up in our study. This, however, seems unlikely, since 3 of the 6 nonresponders at follow-up had the same treatment (50 mg atenolol). Progression of the disease cannot be excluded, but it seems unlikely in 2 of the 6 nonresponders who were followed up after 3 and 10 months. The reason for the almost 100% response rate in the clinical trial did not appear be caused by an extraordinarily enthusiastic physician, since the same physician treated the patients in her clinical practice. Perhaps a patient taking part in a clinical trial feels better cared for and attended to than a patient in clinical practice, when the time between visits has been prolonged. It is known that many people have difficulty following self-administered medical treatment. There are many trials for individual drug efficacies, but only a handful of rigorous trials of adherence interventions.14 And many of the interventions for long-term medications are exceedingly complex and labor intensive, including care provided at the work site, special pill containers, counseling, reminders, self-monitoring, support groups, feedback, and reinforcement.15,16 Most studies also lack follow-up after the intervention had been discontinued.14
Being aware of the importance of noncompliance, manufacturers have tried to make things easier for the patient by producing, for example, sustained-release preparations and blister packages that indicate each day of the week. However, the reason for noncompliance may not always be forgetfulness, and lately a change in the terminology from compliance to concordance has been proposed in the hope that this might lead to a new view of the patient’s role.17 The price of compliance is said to be dependency and thus belongs to an older world.18 However, since there have been great advances in medical therapeutics during the past 2 decades, adherence to medication today may be more important for the outcome than it used to be. People today are more exposed to various kinds of information that may influence them. Information technology not only makes people more informed but sometimes also more irresolute. Alarm reports, more or less true, about adverse effect of medicines are not rare and can make patients discontinue their medication without consulting their physicians.
Conclusions
In our study the significantly reduced blood pressure in the clinical trial did not persist when followed up in clinical practice. At follow-up one third of the patients had blood pressure values similar to those before active treatment. Effects seen in short-term clinical trials, therefore, may not translate to long-term benefits in clinical practice. Noncompliance with treatment may be a factor in these results. Since this is a pilot study, larger studies on this topic would be of value to confirm the results.
1. Urquhart J. Patient non-compliance with drug regimens: measurement, clinical correlates, economic impact. Eur Heart J 1996;17 (Suppl A):S8-15.
2. Harter JG, Peck CC. Chronobiology: suggestions for integrating it into drug development. Ann NY Sci 1991;618:563-71.
3. Urquhart J. Compliance and clinical trials. Lancet 1991;337:1224-25.
4. Lasagna L, Hutt PB. Health care, research and regulatory impact of noncompliance. In: Cramer JA, Spilker B, eds. Patient compliance in medical practice and clinical trials. New York, NY: Raven Press; 1991;393-403.
5. Efron B, Feldman D. Compliance as an explanatory variable in clinical trials. J Am St Assoc 1991;86:9-17.
6. Flack JM, Novikov SV, Ferrario CM. Benefits of adherence to anti-hypertensive drug therapy. Eur Heart J 1996;17(Suppl A):S16-20
7. Costa FV. Compliance with antihypertensive treatment. Clin Exp Hypertens 1996;18:463-72.
8. Enström I, Thulin T, Lindholm LH. How good are standardized blood pressure recordings for diagnosing hypertension? A comparison between office and ambulatory blood pressure. J Hypertens 1991;9:561-66.
9. Enström I, Thulin T, Lindholm LH. A plea for more comprehensive blood pressure measurements when evaluating drug treatment of hypertension. J Hypertens 1988;6:959-64.
10. Graettinger WF, Lipson JL, Cheung DG, Weber MA. Validation of portable non-invasive blood pressure monitoring devices: comparisons with intra-arterial and sphygmomanometer measurements. Am Heart J 1988;116:1155-60.
11. Modesti PA, Gensini GF, Conti C, Neri Serneri CG. Clinical evaluation of an automatic blood pressure monitoring device. J Clin Hypertens 1987;3:631-34.
12. O’Brien E, Mee F, Atkin N, O’Malley K. Accuracy of the SpaceeLabs 90207 determined by the British Hypertension Society Protocol. J Hypertens 1991;9:573-74.
13. Sackett DL, Snow JC. The magnitude of adherence and nonadherence. In: Haynes RB, Taylor DW, Sackett DL, eds. Compliance in health care. Baltimore, Md: Johns Hopkins University Press; 1979;11-22.
14. Haynes RB, McKibbon KA, Kanani R. Systematic review of randomised trials of intervention to assist patients to follow prescriptions for medications. Lancet 1996;348:383-86.
15. Logan AG, Milne BJ, Achber C, Campbell WP, Haynes RB. Work site treatment of hypertension by specially trained nurses. Lancet 1979;ii:1175-78.
16. Logan AS, Milne BJ, Achber C, Campbell WP, Haynes RB. Cost-effectiveness of a worksite hypertension program. Hypertension 1981;3:211-18.
17. Mullen PD. Compliance becomes concordance. Making a change in terminology produce a change in behaviour. BMJ 1997;314:691-92.
18. Marinker M. Personal paper: writing prescriptions is easy. BMJ 1997;314:747-48.
METHODS: Nineteen previously untreated middle-aged men with hypertension had their office and ambulatory blood pressure recorded after 4 weeks of placebo treatment, 4 weeks of active treatment in a clinical trial, and 31 months of treatment in clinical practice. All recording was done by the same physician (IE).
RESULTS: Mean 24-hour blood pressure was 138/92 mm Hg after 4 weeks of placebo treatment, 128/85 mm Hg after 4 weeks of active treatment in the clinical trial, and 136/87 mm Hg after a mean of 31 months of treatment in clinical practice. The corresponding blood pressure values Ž140/90 mm Hg during the daytime were 47%, 24%, and 39%, and office blood pressures were 155/101, 145/93, and 150/91 mm Hg. Individual comparison revealed that 6 of the 19 patients had higher mean 24-hour blood pressure after several months of treatment in clinical practice than after 4 weeks of active treatment in the clinical trial.
CONCLUSIONS: In our study, the significantly reduced blood pressure in the clinical trial did not persist when followed up in clinical practice. At follow-up, one third of the patients had blood pressure values similar to those before active treatment. The reason for this is unclear, but inconsistent compliance may play a part in the lack of durability of the improvements. Our results indicate that effects seen in short-term clinical trials may not translate to long-term benefits in clinical practice.
Poor compliance with prescribed drug regimens attenuates the benefits of treatment, making compliance a key link between process and outcome in ambulatory care.1 In clinical trials, noncompliance, also called nonadherence, is one of the most important sources of the variance in drug response.2 Inconsistent compliance has the largest impact on statistical power, since it lowers the mean drug response and adds variance.3 Deficient compliance can also lead to overestimates of dosage requirements. The rate of compliance differs in clinical trials4,5 and in clinical practice.4,6,7 It has been calculated that depending on the method used to monitor compliance, only 20% to 80% of patients treated for hypertension can be considered good compliers.7
In our pilot study we examined the persistence of blood pressure effects 31 months after a clinical trial of 4 weeks of treatment with hypotensive agents.
Methods
Our study included 19 previously untreated middle-aged men with hypertension (mean age=52.1 years; range=40-64 years) recruited from a blood pressure screening project in a municipality in southern Sweden.8 They were followed up in 1991-1992 in clinical practice after having completed a trial comparing 50 mg atenolol with 20 mg enalapril, which were found to be equally effective in lowering blood pressure.9 Office and ambulatory blood pressures recorded after 4 weeks of placebo treatment and after 4 weeks of active treatment in the clinical trial were compared with office and ambulatory blood pressures recorded after several months of treatment in a clinical practice at the same health center. The same physician (IE) involved in the clinical trial followed up and treated the patients in her practice at the health center. Although some patients were treated with the same drugs as in the clinical trial, most had a different treatment ([Table 1]). Side effects were the most common reason for a patient to change treatment. In the clinical trial, ambulatory blood pressure was recorded every 15 minutes between 6 AM and 6 PM and every 30 min between 6 PM and 6 AM with the SpaceLabs 5200 equipment (Chatsworth, Calif). At the follow-up in clinical practice, ambulatory blood pressure was recorded every 20 minutes with SpaceLabs 90207 equipment (Redmond, Wash). Both these equipments have been validated and found to be accurate.10-12 The equipment was attached to the patient between 8 AM and 10 AM in the clinical trial and at other various times of the day, but mostly between 2 PM and 5 PM when followed up in clinical practice. Office blood pressure was recorded as the mean of 4 measurements obtained by the same physician in the same way (supine position after 5-10 minutes of rest with appropriate cuffs), twice when the ambulatory blood pressure equipment was attached and twice when it was removed from the patient. All patients worked full time during the clinical trial; in the follw-up in clinical practice 7 were retired and 1 had part-time work.
Statistical Analysis
Standard statistics such as mean, standard deviation, minimum, and maximum were used to summarize the data. Mean blood pressure values are presented together with 95% confidence intervals, which were calculated using normal approximation. Daytime was defined as the time between waking and going to bed.
Results
Mean (standard deviation [SD]) 24-hour blood pressure was 138/92 (14/5) mm Hg after 4 weeks of placebo treatment, 128/85 (14/7) mm Hg after 4 weeks of active treatment, and 136/87 (18/9) mm Hg after 31(19) months of treatment in clinical practice. Confidence intervals are given in [Table 2]. The corresponding blood pressure values Ž140/90 (SD) mm Hg during the daytime were 47% (25%), 24% (22%), and 39% (29%), and office blood pressures were 155/101 (16/4), 145/93 (17/6), and 150/91 (23/9) mm Hg. Individual comparison revealed that 6 patients had a higher mean 24-hour blood pressure after several months of treatment in clinical practice than after 4 weeks of active treatment in the clinical trial. Of these 6 patients, 3 had the same treatment (atenolol 50 mg) as in the clinical trial, 1 had changed from enalapril to atenolol, 1 from enalapril to nifedipine, and 1 from atenolol to diltiazem.
Discussion
We examined the persistence of blood pressure effects 31 months after a clinical trial of hypotensive agents. The results showed that mean ambulatory blood pressure was lower after 4 weeks of treatment in the clinical trial than after several months of treatment in clinical practice. Although 4 weeks of treatment with either 50 mg atenolol or 20 mg enalapril in the clinical trial reduced ambulatory blood pressure significantly, these lower blood pressure values did not persist after a longer period of treatment. Individual analysis revealed that 6 patients had higher blood pressure values at follow-up in clinical practice than in the clinical trial. In our clinical trial, all patients except 1 (95%) had reduced their ambulatory blood pressure during active treatment.
In any case of poor response to therapy, it is important to suspect defective compliance. It has been estimated that 16% to 50% of patients with newly diagnosed hypertension discontinue their antihypertensive medications within the first year.6 Electronic monitoring methods have revealed that many patients omit prescribed doses irrespective of disease, prognosis, or symptoms.1 Subjects at particular risk of poor compliance with antihypertensive drugs seem to be middle-aged men still active in work and without previous cardiovascular disease.6 In our study, however, blood pressure was lower during the clinical trial, when all patients were still active in work, than at the follow-up when some patients were retired. In fact, 2 of the patients whose blood pressure was not controlled at the follow-up were retired at that time.
It might be argued that changed antihypertensive treatment could be the reason for the higher blood pressure at follow-up in our study. This, however, seems unlikely, since 3 of the 6 nonresponders at follow-up had the same treatment (50 mg atenolol). Progression of the disease cannot be excluded, but it seems unlikely in 2 of the 6 nonresponders who were followed up after 3 and 10 months. The reason for the almost 100% response rate in the clinical trial did not appear be caused by an extraordinarily enthusiastic physician, since the same physician treated the patients in her clinical practice. Perhaps a patient taking part in a clinical trial feels better cared for and attended to than a patient in clinical practice, when the time between visits has been prolonged. It is known that many people have difficulty following self-administered medical treatment. There are many trials for individual drug efficacies, but only a handful of rigorous trials of adherence interventions.14 And many of the interventions for long-term medications are exceedingly complex and labor intensive, including care provided at the work site, special pill containers, counseling, reminders, self-monitoring, support groups, feedback, and reinforcement.15,16 Most studies also lack follow-up after the intervention had been discontinued.14
Being aware of the importance of noncompliance, manufacturers have tried to make things easier for the patient by producing, for example, sustained-release preparations and blister packages that indicate each day of the week. However, the reason for noncompliance may not always be forgetfulness, and lately a change in the terminology from compliance to concordance has been proposed in the hope that this might lead to a new view of the patient’s role.17 The price of compliance is said to be dependency and thus belongs to an older world.18 However, since there have been great advances in medical therapeutics during the past 2 decades, adherence to medication today may be more important for the outcome than it used to be. People today are more exposed to various kinds of information that may influence them. Information technology not only makes people more informed but sometimes also more irresolute. Alarm reports, more or less true, about adverse effect of medicines are not rare and can make patients discontinue their medication without consulting their physicians.
Conclusions
In our study the significantly reduced blood pressure in the clinical trial did not persist when followed up in clinical practice. At follow-up one third of the patients had blood pressure values similar to those before active treatment. Effects seen in short-term clinical trials, therefore, may not translate to long-term benefits in clinical practice. Noncompliance with treatment may be a factor in these results. Since this is a pilot study, larger studies on this topic would be of value to confirm the results.
METHODS: Nineteen previously untreated middle-aged men with hypertension had their office and ambulatory blood pressure recorded after 4 weeks of placebo treatment, 4 weeks of active treatment in a clinical trial, and 31 months of treatment in clinical practice. All recording was done by the same physician (IE).
RESULTS: Mean 24-hour blood pressure was 138/92 mm Hg after 4 weeks of placebo treatment, 128/85 mm Hg after 4 weeks of active treatment in the clinical trial, and 136/87 mm Hg after a mean of 31 months of treatment in clinical practice. The corresponding blood pressure values Ž140/90 mm Hg during the daytime were 47%, 24%, and 39%, and office blood pressures were 155/101, 145/93, and 150/91 mm Hg. Individual comparison revealed that 6 of the 19 patients had higher mean 24-hour blood pressure after several months of treatment in clinical practice than after 4 weeks of active treatment in the clinical trial.
CONCLUSIONS: In our study, the significantly reduced blood pressure in the clinical trial did not persist when followed up in clinical practice. At follow-up, one third of the patients had blood pressure values similar to those before active treatment. The reason for this is unclear, but inconsistent compliance may play a part in the lack of durability of the improvements. Our results indicate that effects seen in short-term clinical trials may not translate to long-term benefits in clinical practice.
Poor compliance with prescribed drug regimens attenuates the benefits of treatment, making compliance a key link between process and outcome in ambulatory care.1 In clinical trials, noncompliance, also called nonadherence, is one of the most important sources of the variance in drug response.2 Inconsistent compliance has the largest impact on statistical power, since it lowers the mean drug response and adds variance.3 Deficient compliance can also lead to overestimates of dosage requirements. The rate of compliance differs in clinical trials4,5 and in clinical practice.4,6,7 It has been calculated that depending on the method used to monitor compliance, only 20% to 80% of patients treated for hypertension can be considered good compliers.7
In our pilot study we examined the persistence of blood pressure effects 31 months after a clinical trial of 4 weeks of treatment with hypotensive agents.
Methods
Our study included 19 previously untreated middle-aged men with hypertension (mean age=52.1 years; range=40-64 years) recruited from a blood pressure screening project in a municipality in southern Sweden.8 They were followed up in 1991-1992 in clinical practice after having completed a trial comparing 50 mg atenolol with 20 mg enalapril, which were found to be equally effective in lowering blood pressure.9 Office and ambulatory blood pressures recorded after 4 weeks of placebo treatment and after 4 weeks of active treatment in the clinical trial were compared with office and ambulatory blood pressures recorded after several months of treatment in a clinical practice at the same health center. The same physician (IE) involved in the clinical trial followed up and treated the patients in her practice at the health center. Although some patients were treated with the same drugs as in the clinical trial, most had a different treatment ([Table 1]). Side effects were the most common reason for a patient to change treatment. In the clinical trial, ambulatory blood pressure was recorded every 15 minutes between 6 AM and 6 PM and every 30 min between 6 PM and 6 AM with the SpaceLabs 5200 equipment (Chatsworth, Calif). At the follow-up in clinical practice, ambulatory blood pressure was recorded every 20 minutes with SpaceLabs 90207 equipment (Redmond, Wash). Both these equipments have been validated and found to be accurate.10-12 The equipment was attached to the patient between 8 AM and 10 AM in the clinical trial and at other various times of the day, but mostly between 2 PM and 5 PM when followed up in clinical practice. Office blood pressure was recorded as the mean of 4 measurements obtained by the same physician in the same way (supine position after 5-10 minutes of rest with appropriate cuffs), twice when the ambulatory blood pressure equipment was attached and twice when it was removed from the patient. All patients worked full time during the clinical trial; in the follw-up in clinical practice 7 were retired and 1 had part-time work.
Statistical Analysis
Standard statistics such as mean, standard deviation, minimum, and maximum were used to summarize the data. Mean blood pressure values are presented together with 95% confidence intervals, which were calculated using normal approximation. Daytime was defined as the time between waking and going to bed.
Results
Mean (standard deviation [SD]) 24-hour blood pressure was 138/92 (14/5) mm Hg after 4 weeks of placebo treatment, 128/85 (14/7) mm Hg after 4 weeks of active treatment, and 136/87 (18/9) mm Hg after 31(19) months of treatment in clinical practice. Confidence intervals are given in [Table 2]. The corresponding blood pressure values Ž140/90 (SD) mm Hg during the daytime were 47% (25%), 24% (22%), and 39% (29%), and office blood pressures were 155/101 (16/4), 145/93 (17/6), and 150/91 (23/9) mm Hg. Individual comparison revealed that 6 patients had a higher mean 24-hour blood pressure after several months of treatment in clinical practice than after 4 weeks of active treatment in the clinical trial. Of these 6 patients, 3 had the same treatment (atenolol 50 mg) as in the clinical trial, 1 had changed from enalapril to atenolol, 1 from enalapril to nifedipine, and 1 from atenolol to diltiazem.
Discussion
We examined the persistence of blood pressure effects 31 months after a clinical trial of hypotensive agents. The results showed that mean ambulatory blood pressure was lower after 4 weeks of treatment in the clinical trial than after several months of treatment in clinical practice. Although 4 weeks of treatment with either 50 mg atenolol or 20 mg enalapril in the clinical trial reduced ambulatory blood pressure significantly, these lower blood pressure values did not persist after a longer period of treatment. Individual analysis revealed that 6 patients had higher blood pressure values at follow-up in clinical practice than in the clinical trial. In our clinical trial, all patients except 1 (95%) had reduced their ambulatory blood pressure during active treatment.
In any case of poor response to therapy, it is important to suspect defective compliance. It has been estimated that 16% to 50% of patients with newly diagnosed hypertension discontinue their antihypertensive medications within the first year.6 Electronic monitoring methods have revealed that many patients omit prescribed doses irrespective of disease, prognosis, or symptoms.1 Subjects at particular risk of poor compliance with antihypertensive drugs seem to be middle-aged men still active in work and without previous cardiovascular disease.6 In our study, however, blood pressure was lower during the clinical trial, when all patients were still active in work, than at the follow-up when some patients were retired. In fact, 2 of the patients whose blood pressure was not controlled at the follow-up were retired at that time.
It might be argued that changed antihypertensive treatment could be the reason for the higher blood pressure at follow-up in our study. This, however, seems unlikely, since 3 of the 6 nonresponders at follow-up had the same treatment (50 mg atenolol). Progression of the disease cannot be excluded, but it seems unlikely in 2 of the 6 nonresponders who were followed up after 3 and 10 months. The reason for the almost 100% response rate in the clinical trial did not appear be caused by an extraordinarily enthusiastic physician, since the same physician treated the patients in her clinical practice. Perhaps a patient taking part in a clinical trial feels better cared for and attended to than a patient in clinical practice, when the time between visits has been prolonged. It is known that many people have difficulty following self-administered medical treatment. There are many trials for individual drug efficacies, but only a handful of rigorous trials of adherence interventions.14 And many of the interventions for long-term medications are exceedingly complex and labor intensive, including care provided at the work site, special pill containers, counseling, reminders, self-monitoring, support groups, feedback, and reinforcement.15,16 Most studies also lack follow-up after the intervention had been discontinued.14
Being aware of the importance of noncompliance, manufacturers have tried to make things easier for the patient by producing, for example, sustained-release preparations and blister packages that indicate each day of the week. However, the reason for noncompliance may not always be forgetfulness, and lately a change in the terminology from compliance to concordance has been proposed in the hope that this might lead to a new view of the patient’s role.17 The price of compliance is said to be dependency and thus belongs to an older world.18 However, since there have been great advances in medical therapeutics during the past 2 decades, adherence to medication today may be more important for the outcome than it used to be. People today are more exposed to various kinds of information that may influence them. Information technology not only makes people more informed but sometimes also more irresolute. Alarm reports, more or less true, about adverse effect of medicines are not rare and can make patients discontinue their medication without consulting their physicians.
Conclusions
In our study the significantly reduced blood pressure in the clinical trial did not persist when followed up in clinical practice. At follow-up one third of the patients had blood pressure values similar to those before active treatment. Effects seen in short-term clinical trials, therefore, may not translate to long-term benefits in clinical practice. Noncompliance with treatment may be a factor in these results. Since this is a pilot study, larger studies on this topic would be of value to confirm the results.
1. Urquhart J. Patient non-compliance with drug regimens: measurement, clinical correlates, economic impact. Eur Heart J 1996;17 (Suppl A):S8-15.
2. Harter JG, Peck CC. Chronobiology: suggestions for integrating it into drug development. Ann NY Sci 1991;618:563-71.
3. Urquhart J. Compliance and clinical trials. Lancet 1991;337:1224-25.
4. Lasagna L, Hutt PB. Health care, research and regulatory impact of noncompliance. In: Cramer JA, Spilker B, eds. Patient compliance in medical practice and clinical trials. New York, NY: Raven Press; 1991;393-403.
5. Efron B, Feldman D. Compliance as an explanatory variable in clinical trials. J Am St Assoc 1991;86:9-17.
6. Flack JM, Novikov SV, Ferrario CM. Benefits of adherence to anti-hypertensive drug therapy. Eur Heart J 1996;17(Suppl A):S16-20
7. Costa FV. Compliance with antihypertensive treatment. Clin Exp Hypertens 1996;18:463-72.
8. Enström I, Thulin T, Lindholm LH. How good are standardized blood pressure recordings for diagnosing hypertension? A comparison between office and ambulatory blood pressure. J Hypertens 1991;9:561-66.
9. Enström I, Thulin T, Lindholm LH. A plea for more comprehensive blood pressure measurements when evaluating drug treatment of hypertension. J Hypertens 1988;6:959-64.
10. Graettinger WF, Lipson JL, Cheung DG, Weber MA. Validation of portable non-invasive blood pressure monitoring devices: comparisons with intra-arterial and sphygmomanometer measurements. Am Heart J 1988;116:1155-60.
11. Modesti PA, Gensini GF, Conti C, Neri Serneri CG. Clinical evaluation of an automatic blood pressure monitoring device. J Clin Hypertens 1987;3:631-34.
12. O’Brien E, Mee F, Atkin N, O’Malley K. Accuracy of the SpaceeLabs 90207 determined by the British Hypertension Society Protocol. J Hypertens 1991;9:573-74.
13. Sackett DL, Snow JC. The magnitude of adherence and nonadherence. In: Haynes RB, Taylor DW, Sackett DL, eds. Compliance in health care. Baltimore, Md: Johns Hopkins University Press; 1979;11-22.
14. Haynes RB, McKibbon KA, Kanani R. Systematic review of randomised trials of intervention to assist patients to follow prescriptions for medications. Lancet 1996;348:383-86.
15. Logan AG, Milne BJ, Achber C, Campbell WP, Haynes RB. Work site treatment of hypertension by specially trained nurses. Lancet 1979;ii:1175-78.
16. Logan AS, Milne BJ, Achber C, Campbell WP, Haynes RB. Cost-effectiveness of a worksite hypertension program. Hypertension 1981;3:211-18.
17. Mullen PD. Compliance becomes concordance. Making a change in terminology produce a change in behaviour. BMJ 1997;314:691-92.
18. Marinker M. Personal paper: writing prescriptions is easy. BMJ 1997;314:747-48.
1. Urquhart J. Patient non-compliance with drug regimens: measurement, clinical correlates, economic impact. Eur Heart J 1996;17 (Suppl A):S8-15.
2. Harter JG, Peck CC. Chronobiology: suggestions for integrating it into drug development. Ann NY Sci 1991;618:563-71.
3. Urquhart J. Compliance and clinical trials. Lancet 1991;337:1224-25.
4. Lasagna L, Hutt PB. Health care, research and regulatory impact of noncompliance. In: Cramer JA, Spilker B, eds. Patient compliance in medical practice and clinical trials. New York, NY: Raven Press; 1991;393-403.
5. Efron B, Feldman D. Compliance as an explanatory variable in clinical trials. J Am St Assoc 1991;86:9-17.
6. Flack JM, Novikov SV, Ferrario CM. Benefits of adherence to anti-hypertensive drug therapy. Eur Heart J 1996;17(Suppl A):S16-20
7. Costa FV. Compliance with antihypertensive treatment. Clin Exp Hypertens 1996;18:463-72.
8. Enström I, Thulin T, Lindholm LH. How good are standardized blood pressure recordings for diagnosing hypertension? A comparison between office and ambulatory blood pressure. J Hypertens 1991;9:561-66.
9. Enström I, Thulin T, Lindholm LH. A plea for more comprehensive blood pressure measurements when evaluating drug treatment of hypertension. J Hypertens 1988;6:959-64.
10. Graettinger WF, Lipson JL, Cheung DG, Weber MA. Validation of portable non-invasive blood pressure monitoring devices: comparisons with intra-arterial and sphygmomanometer measurements. Am Heart J 1988;116:1155-60.
11. Modesti PA, Gensini GF, Conti C, Neri Serneri CG. Clinical evaluation of an automatic blood pressure monitoring device. J Clin Hypertens 1987;3:631-34.
12. O’Brien E, Mee F, Atkin N, O’Malley K. Accuracy of the SpaceeLabs 90207 determined by the British Hypertension Society Protocol. J Hypertens 1991;9:573-74.
13. Sackett DL, Snow JC. The magnitude of adherence and nonadherence. In: Haynes RB, Taylor DW, Sackett DL, eds. Compliance in health care. Baltimore, Md: Johns Hopkins University Press; 1979;11-22.
14. Haynes RB, McKibbon KA, Kanani R. Systematic review of randomised trials of intervention to assist patients to follow prescriptions for medications. Lancet 1996;348:383-86.
15. Logan AG, Milne BJ, Achber C, Campbell WP, Haynes RB. Work site treatment of hypertension by specially trained nurses. Lancet 1979;ii:1175-78.
16. Logan AS, Milne BJ, Achber C, Campbell WP, Haynes RB. Cost-effectiveness of a worksite hypertension program. Hypertension 1981;3:211-18.
17. Mullen PD. Compliance becomes concordance. Making a change in terminology produce a change in behaviour. BMJ 1997;314:691-92.
18. Marinker M. Personal paper: writing prescriptions is easy. BMJ 1997;314:747-48.