Targets and self monitoring in hypertension: randomised controlled tri
http://www.100md.com
《英国医生杂志》
1 Department of Primary Care and General Practice, University of Birmingham, Birmingham B15 2TT, 2 Health Economics Facility, University of Birmingham, Health Services Management Centre, B15 2RT, 3 Psychology, School of Life and Health Sciences, Birmingham B4 7ET
Correspondence to: R McManus r.j.mcmanus@bham.ac.uk
Objectives To assess whether blood pressure control in primary care could be improved with the use of patient held targets and self monitoring in a practice setting, and to assess the impact of these on health behaviours, anxiety, prescribed antihypertensive drugs, patients' preferences, and costs.
Design Randomised controlled trial.
Setting Eight general practices in south Birmingham.
Participants 441 people receiving treatment in primary care for hypertension but not controlled below the target of < 140/85 mm Hg.
Interventions Patients in the intervention group received treatment targets along with facilities to measure their own blood pressure at their general practice; they were also asked to visit their general practitioner or practice nurse if their blood pressure was repeatedly above the target level. Patients in the control group received usual care (blood pressure monitoring by their practice).
Main outcome measures Primary outcome: change in systolic blood pressure at six months and one year in both intervention and control groups. Secondary outcomes: change in health behaviours, anxiety, prescribed antihypertensive drugs, patients' preferences of method of blood pressure monitoring, and costs.
Results 400 (91%) patients attended follow up at one year. Systolic blood pressure in the intervention group had significantly reduced after six months (mean difference 4.3 mm Hg (95% confidence interval 0.8 mm Hg to 7.9 mm Hg)) but not after one year (mean difference 2.7 mm Hg (– 1.2 mm Hg to 6.6 mm Hg)). No overall difference was found in diastolic blood pressure, anxiety, health behaviours, or number of prescribed drugs. Patients who self monitored lost more weight than controls (as evidenced by a drop in body mass index), rated self monitoring above monitoring by a doctor or nurse, and consulted less often. Overall, self monitoring did not cost significantly more than usual care (￡251 ($437; 364 euros) (95% confidence interval ￡233 to ￡275) versus ￡240 (￡217 to ￡263).
Conclusions Practice based self monitoring resulted in small but significant improvements of blood pressure at six months, which were not sustained after a year. Self monitoring was well received by patients, anxiety did not increase, and there was no appreciable additional cost. Practice based self monitoring is feasible and results in blood pressure control that is similar to that in usual care.
Hypertension is a key risk factor for cardiovascular disease, the leading cause of death worldwide.1 Use of antihypertensive drugs leads to a significant reduction in both stroke and coronary heart disease risk and is cost effective, especially for individuals at highest risk of cardiovascular events.2 3 This potential benefit from drug treatment is reflected in recent hypertension guidelines, which recommend treatment targets of 140/85 mm Hg or below.4 5 However, international community based surveys show that in many countries only a minority of people treated for hypertension are controlled to these levels.6
Most patients with hypertension receive treatment in primary care and so it is here that potential reasons for this shortfall must be sought. These may include clinical inertia and excessive workloads on the part of physicians7 8; unmet information needs and poor adherence to antihypertensive treatment by patients9 10; and differing thresholds at which health professionals and patients would choose to start treatment.11
Self monitoring of blood pressure has the potential to improve blood pressure control at modest cost: a recent systematic review of home monitoring found a lowering of systolic blood pressure of about 4 mm Hg.12 However, most included trials were inadequately powered with short (six months or less) follow-up, none was performed in the United Kingdom, and only one study evaluated costs.13 Community based self monitoring, where patients have access to blood pressure measurement facilities without the need to see a health professional, has the potential to provide the benefits of home monitoring (reduction of the white coat effect, more readings, and improved control) without requiring the purchase of a blood pressure machine for all patients.14 15
We aimed to assess whether blood pressure control in primary care could be improved with the use of patient held targets and self monitoring in a practice setting and to assess the impact of these on health behaviours, anxiety, prescribed antihypertensive treatment, patients' preferences, and costs.
Methods
Participants
We recruited participants from eight primary care practices in Birmingham through the Midlands Research Practice Consortium between September 2001 and April 2002. The practices were selected so that there would be two from each quartile of the Townsend score (a measure of socioeconomic deprivation).16 A computer search identified patients aged 35-75 receiving treatment for hypertension. Those with at least one blood pressure reading greater than 140/85 mm Hg in the previous year were invited to attend a study clinic run by the principal investigator (RJMcM). If their blood pressure at this clinic was in the range 140/85 mm Hg to 200/100 mm Hg they were eligible for the study. All patients provided written informed consent before entry to the study.
Assignment and blinding
Randomisation was stratified by practice and diabetic status; we used a random number generator to produce a series of blocks of random size.17 Allocations were transferred to opaque envelopes that were held centrally and opened by the trial secretary in response to a telephone call. In one practice, owing to an error in the blocking process, three people with diabetes were allocated to the control rather than intervention group. The study was unblinded.
Intervention
Patients randomised to intervention were asked to attend their practice every month to measure their own blood pressure using validated electronic blood pressure machines (Omron M5-I18). They did not need an appointment, and they did the monitoring in the waiting room or in a side room, depending on facilities at the practice. They were given about 10 minutes of instruction at baseline on how to use the electronic sphygmomanometers and a five minute refresher session at a six month follow-up. Reception staff were also trained to provide support to patients as required.
The patients in the intervention group each received a record card showing the blood pressure target they should aim for (the British Hypertension Society's treatment targets—at that time—of 140/85 mmHg (140/80 mm Hg for those with diabetes)).4 19 The cards had space for recording monthly blood pressure readings as well as advice that patients should attend their general practitioner or practice nurse if they recorded blood pressures above target on successive months—or earlier in the case of very high readings. Central telephone support was available.
Frequency of monitoring of patients in the control group was at the discretion of the general practitioner. All control patients received an information sheet on self help measures to lower blood pressure based on a fact sheet published by the British Hypertension Society (www.bhsoc.org). Responsibility for changes to treatment remained with the general practitioner for both the intervention and control group.
Data collection
All baseline data and measurements were recorded before randomisation. Patients were followed up by a researcher (RJMcM or RAO) in their own surgeries six months and one year after randomisation. All primary and secondary end point data were collected at each follow-up visit.
At follow-up sessions, blood pressure was measured with a validated automated sphygmomanometer (Omron 705CP) in the left arm three times at five minute intervals after 10 minutes rest, with the patient seated. We used the mean of the three readings in the analysis. We measured anxiety using the short form of the Spielberger state anxiety inventory20 and used previously validated questionnaires for alcohol consumption21 and exercise22 to measure lifestyle factors. We assessed addition of salt to food by using questions drawn from a validated questionnaire.23 We calculated body mass index using electronic scales (Seca 880, Vogel and Halke, Hamburg) and a portable height meter. At each follow up visit we collected data on prescribed antihypertensive drugs from the practice computer systems. We asked participants to rank four methods of blood pressure measurement according to preference—namely, by a doctor, by a nurse, self measurement at the surgery, and self measurement at home.
Analysis
The primary outcome was change in systolic blood pressure between baseline and the two follow-up sessions. A power calculation suggested that if 434 patients were followed up, it would be possible to detect a 5 mm Hg difference with a power of 90% and significance level of 5%, assuming a standard deviation of 16 mm Hg. We analysed the data with the statistical software SPSS (version 10) on an "intention to treat" basis using the "complete case" method with sensitivity analysis around assumptions for missing values.24 We used the GLM (general linear model) repeated measures technique to examine within subject differences in systolic blood pressure between baseline and the two follow-up sessions. We adjusted the primary analysis for practice (nested within intervention) and diabetic status by including these as fixed effects in the model along with any baseline differences with potential effects on outcome.
We also used the GLM repeated measures technique for the secondary outcomes—to analyse changes in diastolic blood pressure, anxiety, body mass index, and patients' preferences. Changes over time between intervention and control with respect to exercise (three times or more a week), alcohol (> 21 or > 14 units a week for men and woman respectively), smoking (yes or no), and addition of salt to food were assessed using logistic regression, taking into account repeated measures and adjusting for practice, diabetic status, and sex.
Costing
During the trial we collected data on the number of consultations for hypertension (with both general practitioners and practice nurses and defined as consultations with Read codes for hypertension or where blood pressure was recorded or where free text explicitly mentioning hypertension management was included), drug treatment, referrals for hypertension, and intervention costs (for equipment and training). Equipment costs were discounted at 3.5% (a year) over a five year life.25 We assumed that consultations had standard UK costs (￡16 ($28; 23) for a general practitioner and ￡8 for a practice nurse)26; we took treatment costs from the British National Formulary27; and we priced referrals at the rate for a general medical outpatient clinic (￡104).28 Cost data were bootstrapped (1000 samples) to estimate confidence intervals.
Results
In all, 441 people were randomised, of whom 400 (91%) attended follow-up at both six and 12 months (complete cases). The figure shows flow through the trial, and table 1 shows baseline details of complete cases. In all, 74% of participants in the intervention group measured their own blood pressure at least eight times over 12 months (prespecified as the definition of adequate compliance), and the median number of self measurements over a year was 11. Owing to the baseline difference in sex and the potential effect on outcome, sex was included in the model as prespecified in the analysis plan.
Table 1 Baseline characteristics of 400 complete cases (unadjusted). Values are numbers (percentages) of participants, unless stated otherwise
Primary outcome (systolic blood pressure)
The GLM repeated measures—taking into account intervention group, practice (nested within intervention), sex, and diabetic status—showed a significant difference in systolic blood pressure over time between intervention and control (F (2, 762) = 3.7; P = 0.02). Further analysis of this change in systolic blood pressure showed a significant difference between groups in favour of the intervention between baseline and six month follow-up (mean difference in change 4.3 mm Hg (95% confidence interval 0.8 mm Hg to 7.9 mm Hg); F (1, 381) = 8.2; P = 0.004) but not between six months and one year (– 1.6 mm Hg (– 5.3 mm Hg to 2.2 mm Hg); F (1, 381) = 1.0; P = 0.33) (table 2). The intervention group experienced a greater fall in blood pressure in the first six months, but from a higher baseline. Mean systolic blood pressure at six and 12 months was similar in the two groups.
Table 2 Mean systolic and diastolic blood pressure (mm Hg)
Flow of participants through the trial
Secondary outcomes
Diastolic pressure did not change significantly over time between the groups (F (1.9, 737.9) = 0.08; P = 0.91) (fractional degrees of freedom result from adjustments made when Mauchly's test for sphericity was significant) (table 2). The groups did not differ over time for anxiety, smoking, exercise, salt intake, or number of prescribed drugs (tables 3 and 4). Body mass index reduced significantly more over time in the intervention group than in the control group (F (1.6, 590.9) = 6.010; P = 0.005). Most of the reduction occurred in the first six months of the study: with the repeated measures comparisons, the change between baseline and six months is significant (F (1, 370) = 10.3; P = 0.001) whereas that between six months and one year is not (F (1, 370) = 0.58; P = 0.45). Reported alcohol intake reduced significantly in the intervention group compared with the control group in the first six months but not thereafter (P = 0.03 and P = 0.56 respectively).
Table 3 Secondary outcomes: anxiety, body mass index, and number of prescribed medications
Table 4 Secondary outcomes: health behaviour. Values are numbers (percentages) of participants, unless stated otherwise
Table 5 shows patients' preferences at the end of the study for method of blood pressure measurement. The ranking was significantly different between the intervention and control groups (F (2.1, 820.5) = 37.4; P < 0.001). Patients in the intervention group ranked home measurement highest, followed by self measurement in the surgery. Those in the control group ranked measurement by a doctor highest, followed by nurse measurement.
Table 5 Patients' ranking of methods of blood pressure measurement at end of study*
Sensitivity analysis
Missing values analysis on the primary outcome of systolic blood pressure found that for substitution of missing values by either the mean of the series or by the last recorded value for an individual, the change in systolic blood pressure in the first six months remained significant (P = 0.03 in both cases), but the overall difference between the groups over time was no longer significant (P = 0.11 (substitution by mean of series) and P = 0.10 (substitution by last recorded value)).
Cost effectiveness
Table 6 shows the cost and effectiveness results. The intervention group consulted less frequently than the control group, and the drug costs did not differ significantly between the two groups. The reduction in consultation rate observed in the intervention group reflected fewer consultations in which blood pressure monitoring alone took place. Intervention costs (￡26.80 per patient) were dominated by the cost of general practitioners' time in training staff and patients (￡25.40), with discounted equipment costs relatively trivial (￡1.39). The incremental cost effectiveness ratio for practice based self monitoring—that is, the cost per additional 1 mm Hg reduction in blood pressure—was ￡5.10, with confidence intervals that crossed zero (table 5).
Table 6 Costs and effects of self monitoring compared with usual care (adjusted effects). Values are per patient per year (95% confidence interval)
Discussion
Mathers CD, Bernard C, Iburg CM, Innoue M, Fat DM, Shibuya K, et al. Global burden of disease in 2002: data sources, methods and results. Geneva: World Health Organization, 2003.
Psaty BM, Lumley T, Furberg CD, Schellenbaum G, Pahor M, Alderman MH, et al. Health outcomes associated with various antihypertensive therapies used as first-line agents: a network meta-analysis. JAMA 2003;289: 2534-44.
Montgomery AA, Fahey T, Ben Shlomo Y, Harding J. The influence of absolute cardiovascular risk, patient utilities, and costs on the decision to treat hypertension: a Markov decision analysis. J Hypertens 2003;21: 1753-9.
Williams B, Poulter NR, Brown MJ, Davis M, McInnes GT, Potter JF, et al. Guidelines for management of hypertension: report of the fourth working party of the British Hypertension Society, 2004-BHS IV. J Hum Hypertens 2004;18: 139-85.
Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA, Izzo JL Jr, et al. The Seventh report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure: the JNC 7 report. JAMA 2003;289: 2560-72.
Wolf-Maier K, Cooper RS, Kramer H, Banegas JR, Giampaoli S, Joffres MR, et al. Hypertension treatment and control in five European countries, Canada, and the United States. Hypertension 2004;43: 10-7.
Phillips LS, Branch WT, Cook CB, Doyle JP, El Kebbi IM, Gallina DL, et al. Clinical inertia. Ann Intern Med 2001;135: 825-34.
Marshall T, Rouse A. Resource implications and health benefits of primary prevention strategies for cardiovascular disease in people aged 30 to 74: mathematical modelling study. BMJ 2002;325: 197.
Jones JK, Gorkin L, Lian JF, Staffa JA, Fletcher AP. Discontinuation of and changes in treatment after start of new courses of antihypertensive drugs: a study of a United Kingdom population. BMJ 1995;311: 293-5.
Ogden J, Ambrose L, Khadra A, Manthri S, Symons L, Vass A, et al. A questionnaire study of GPs' and patients' beliefs about the different components of patient centredness. Patient Education and Counseling 2002;47: 223-7.
Steel N. Thresholds for taking antihypertensive drugs in different professional and lay groups: questionnaire survey. BMJ 2000;320: 1446-7.
Cappuccio FP, Kerry SM, Forbes L, Donald A. Blood pressure control by home monitoring: meta-analysis of randomised trials. BMJ 2004;329: 145-51.
Soghikian K, Casper SM, Fireman BH, Hunkeler EM, Hurley LB, Tekawa IS, et al. Home blood pressure monitoring. Effect on use of medical services and medical care costs. Med Care 1992;30: 855-65.
Little P, Barnett J, Barnsley L, Marjoram J, Fitzgerald-Barron A, Mant D. Comparison of agreement between different measures of blood pressure in primary care and daytime ambulatory blood pressure. BMJ 2002;325: 254-7.
Hamilton W, Round A, Goodchild R, Baker C. Do community based self-reading sphygmomanometers improve detection of hypertension? A feasibility study. J Public Health Med 2003;25: 125-30.
Townsend P, Phillimore P, Beattie A. Health and deprivation. Inequality and the north. London: Croom Helm, 1988.
Schulz KF, Grimes DA. Generation of allocation sequences in randomised trials: chance, not choice. Lancet 2002;359: 515-9.
El Assad MA, Topouchian JA, Asmar RG. Evaluation of two devices for self-measurement of blood pressure according to the international protocol: the Omron M5-I and the Omron 705IT. Blood Pressure Monitoring 2003;8: 127-33.
Ramsay L, Williams B, Johnston G, MacGregor G, Poston L, Potter J, et al. Guidelines for management of hypertension: report of the third working party of the British Hypertension Society. J Hum Hypertens 1999;13: 569-92.
Marteau TM, Bekker H. The development of a six-item short-form of the state scale of the Spielberger state-trait anxiety inventory (STAI). Br J Clin Psychol 1992;31(part 3): 301-6.
Mant J, Murphy M, Rose P, Vessey M. The accuracy of general practitioner records of smoking and alcohol use: comparison with patient questionnaires. J Public Health Med 2000;22: 198-201.
Godin G, Shephard RJ. A simple method to assess exercise behavior in the community. Can J Appl Sport Sci 1985;10: 141-6.
Little P, Barnett J, Margetts B, Kinmonth AL, Gabbay J, Thompson R, et al. The validity of dietary assessment in general practice. J Epidemiol Community Health 1999;53: 165-72.
Hollis S, Campbell F. What is meant by intention to treat analysis? Survey of published randomised controlled trials. BMJ 1999;319: 670-4.
HM Treasury. The Green Book: appraisal and evaluation in central government. London: Stationery Office, 2003.
Netten A, Curtis L. Unit costs of health and social care 2002. www.pssru.ac.uk/pdf/uc2002/Unit%20Costs%202002.pdf (accessed 8 Aug 2005).
Joint Formulary Committee. British national formulary 42. London: BMA, Royal Pharmaceutical Society of Great Britain, 2001.
Department of Health. National reference costs 2002. www.dh.gov.uk/PolicyAndGuidance/OrganisationPolicy/ FinanceAndPlanning/NHSReferenceCosts/fs/en (accessed 8 Aug 2005).
Kernick DP, Reinhold DM, Netten A. What does it cost the patient to see the doctor? Br J Gen Pract 2000;50: 401-3.
Little P, Barnett J, Barnsley L, Marjoram J, Fitzgerald-Barron A, Mant D. Comparison of acceptability of and preferences for different methods of measuring blood pressure in primary care. BMJ 2002;325: 258-9.(R J McManus, clinical senior lecturer1, )
Correspondence to: R McManus r.j.mcmanus@bham.ac.uk
Objectives To assess whether blood pressure control in primary care could be improved with the use of patient held targets and self monitoring in a practice setting, and to assess the impact of these on health behaviours, anxiety, prescribed antihypertensive drugs, patients' preferences, and costs.
Design Randomised controlled trial.
Setting Eight general practices in south Birmingham.
Participants 441 people receiving treatment in primary care for hypertension but not controlled below the target of < 140/85 mm Hg.
Interventions Patients in the intervention group received treatment targets along with facilities to measure their own blood pressure at their general practice; they were also asked to visit their general practitioner or practice nurse if their blood pressure was repeatedly above the target level. Patients in the control group received usual care (blood pressure monitoring by their practice).
Main outcome measures Primary outcome: change in systolic blood pressure at six months and one year in both intervention and control groups. Secondary outcomes: change in health behaviours, anxiety, prescribed antihypertensive drugs, patients' preferences of method of blood pressure monitoring, and costs.
Results 400 (91%) patients attended follow up at one year. Systolic blood pressure in the intervention group had significantly reduced after six months (mean difference 4.3 mm Hg (95% confidence interval 0.8 mm Hg to 7.9 mm Hg)) but not after one year (mean difference 2.7 mm Hg (– 1.2 mm Hg to 6.6 mm Hg)). No overall difference was found in diastolic blood pressure, anxiety, health behaviours, or number of prescribed drugs. Patients who self monitored lost more weight than controls (as evidenced by a drop in body mass index), rated self monitoring above monitoring by a doctor or nurse, and consulted less often. Overall, self monitoring did not cost significantly more than usual care (￡251 ($437; 364 euros) (95% confidence interval ￡233 to ￡275) versus ￡240 (￡217 to ￡263).
Conclusions Practice based self monitoring resulted in small but significant improvements of blood pressure at six months, which were not sustained after a year. Self monitoring was well received by patients, anxiety did not increase, and there was no appreciable additional cost. Practice based self monitoring is feasible and results in blood pressure control that is similar to that in usual care.
Hypertension is a key risk factor for cardiovascular disease, the leading cause of death worldwide.1 Use of antihypertensive drugs leads to a significant reduction in both stroke and coronary heart disease risk and is cost effective, especially for individuals at highest risk of cardiovascular events.2 3 This potential benefit from drug treatment is reflected in recent hypertension guidelines, which recommend treatment targets of 140/85 mm Hg or below.4 5 However, international community based surveys show that in many countries only a minority of people treated for hypertension are controlled to these levels.6
Most patients with hypertension receive treatment in primary care and so it is here that potential reasons for this shortfall must be sought. These may include clinical inertia and excessive workloads on the part of physicians7 8; unmet information needs and poor adherence to antihypertensive treatment by patients9 10; and differing thresholds at which health professionals and patients would choose to start treatment.11
Self monitoring of blood pressure has the potential to improve blood pressure control at modest cost: a recent systematic review of home monitoring found a lowering of systolic blood pressure of about 4 mm Hg.12 However, most included trials were inadequately powered with short (six months or less) follow-up, none was performed in the United Kingdom, and only one study evaluated costs.13 Community based self monitoring, where patients have access to blood pressure measurement facilities without the need to see a health professional, has the potential to provide the benefits of home monitoring (reduction of the white coat effect, more readings, and improved control) without requiring the purchase of a blood pressure machine for all patients.14 15
We aimed to assess whether blood pressure control in primary care could be improved with the use of patient held targets and self monitoring in a practice setting and to assess the impact of these on health behaviours, anxiety, prescribed antihypertensive treatment, patients' preferences, and costs.
Methods
Participants
We recruited participants from eight primary care practices in Birmingham through the Midlands Research Practice Consortium between September 2001 and April 2002. The practices were selected so that there would be two from each quartile of the Townsend score (a measure of socioeconomic deprivation).16 A computer search identified patients aged 35-75 receiving treatment for hypertension. Those with at least one blood pressure reading greater than 140/85 mm Hg in the previous year were invited to attend a study clinic run by the principal investigator (RJMcM). If their blood pressure at this clinic was in the range 140/85 mm Hg to 200/100 mm Hg they were eligible for the study. All patients provided written informed consent before entry to the study.
Assignment and blinding
Randomisation was stratified by practice and diabetic status; we used a random number generator to produce a series of blocks of random size.17 Allocations were transferred to opaque envelopes that were held centrally and opened by the trial secretary in response to a telephone call. In one practice, owing to an error in the blocking process, three people with diabetes were allocated to the control rather than intervention group. The study was unblinded.
Intervention
Patients randomised to intervention were asked to attend their practice every month to measure their own blood pressure using validated electronic blood pressure machines (Omron M5-I18). They did not need an appointment, and they did the monitoring in the waiting room or in a side room, depending on facilities at the practice. They were given about 10 minutes of instruction at baseline on how to use the electronic sphygmomanometers and a five minute refresher session at a six month follow-up. Reception staff were also trained to provide support to patients as required.
The patients in the intervention group each received a record card showing the blood pressure target they should aim for (the British Hypertension Society's treatment targets—at that time—of 140/85 mmHg (140/80 mm Hg for those with diabetes)).4 19 The cards had space for recording monthly blood pressure readings as well as advice that patients should attend their general practitioner or practice nurse if they recorded blood pressures above target on successive months—or earlier in the case of very high readings. Central telephone support was available.
Frequency of monitoring of patients in the control group was at the discretion of the general practitioner. All control patients received an information sheet on self help measures to lower blood pressure based on a fact sheet published by the British Hypertension Society (www.bhsoc.org). Responsibility for changes to treatment remained with the general practitioner for both the intervention and control group.
Data collection
All baseline data and measurements were recorded before randomisation. Patients were followed up by a researcher (RJMcM or RAO) in their own surgeries six months and one year after randomisation. All primary and secondary end point data were collected at each follow-up visit.
At follow-up sessions, blood pressure was measured with a validated automated sphygmomanometer (Omron 705CP) in the left arm three times at five minute intervals after 10 minutes rest, with the patient seated. We used the mean of the three readings in the analysis. We measured anxiety using the short form of the Spielberger state anxiety inventory20 and used previously validated questionnaires for alcohol consumption21 and exercise22 to measure lifestyle factors. We assessed addition of salt to food by using questions drawn from a validated questionnaire.23 We calculated body mass index using electronic scales (Seca 880, Vogel and Halke, Hamburg) and a portable height meter. At each follow up visit we collected data on prescribed antihypertensive drugs from the practice computer systems. We asked participants to rank four methods of blood pressure measurement according to preference—namely, by a doctor, by a nurse, self measurement at the surgery, and self measurement at home.
Analysis
The primary outcome was change in systolic blood pressure between baseline and the two follow-up sessions. A power calculation suggested that if 434 patients were followed up, it would be possible to detect a 5 mm Hg difference with a power of 90% and significance level of 5%, assuming a standard deviation of 16 mm Hg. We analysed the data with the statistical software SPSS (version 10) on an "intention to treat" basis using the "complete case" method with sensitivity analysis around assumptions for missing values.24 We used the GLM (general linear model) repeated measures technique to examine within subject differences in systolic blood pressure between baseline and the two follow-up sessions. We adjusted the primary analysis for practice (nested within intervention) and diabetic status by including these as fixed effects in the model along with any baseline differences with potential effects on outcome.
We also used the GLM repeated measures technique for the secondary outcomes—to analyse changes in diastolic blood pressure, anxiety, body mass index, and patients' preferences. Changes over time between intervention and control with respect to exercise (three times or more a week), alcohol (> 21 or > 14 units a week for men and woman respectively), smoking (yes or no), and addition of salt to food were assessed using logistic regression, taking into account repeated measures and adjusting for practice, diabetic status, and sex.
Costing
During the trial we collected data on the number of consultations for hypertension (with both general practitioners and practice nurses and defined as consultations with Read codes for hypertension or where blood pressure was recorded or where free text explicitly mentioning hypertension management was included), drug treatment, referrals for hypertension, and intervention costs (for equipment and training). Equipment costs were discounted at 3.5% (a year) over a five year life.25 We assumed that consultations had standard UK costs (￡16 ($28; 23) for a general practitioner and ￡8 for a practice nurse)26; we took treatment costs from the British National Formulary27; and we priced referrals at the rate for a general medical outpatient clinic (￡104).28 Cost data were bootstrapped (1000 samples) to estimate confidence intervals.
Results
In all, 441 people were randomised, of whom 400 (91%) attended follow-up at both six and 12 months (complete cases). The figure shows flow through the trial, and table 1 shows baseline details of complete cases. In all, 74% of participants in the intervention group measured their own blood pressure at least eight times over 12 months (prespecified as the definition of adequate compliance), and the median number of self measurements over a year was 11. Owing to the baseline difference in sex and the potential effect on outcome, sex was included in the model as prespecified in the analysis plan.
Table 1 Baseline characteristics of 400 complete cases (unadjusted). Values are numbers (percentages) of participants, unless stated otherwise
Primary outcome (systolic blood pressure)
The GLM repeated measures—taking into account intervention group, practice (nested within intervention), sex, and diabetic status—showed a significant difference in systolic blood pressure over time between intervention and control (F (2, 762) = 3.7; P = 0.02). Further analysis of this change in systolic blood pressure showed a significant difference between groups in favour of the intervention between baseline and six month follow-up (mean difference in change 4.3 mm Hg (95% confidence interval 0.8 mm Hg to 7.9 mm Hg); F (1, 381) = 8.2; P = 0.004) but not between six months and one year (– 1.6 mm Hg (– 5.3 mm Hg to 2.2 mm Hg); F (1, 381) = 1.0; P = 0.33) (table 2). The intervention group experienced a greater fall in blood pressure in the first six months, but from a higher baseline. Mean systolic blood pressure at six and 12 months was similar in the two groups.
Table 2 Mean systolic and diastolic blood pressure (mm Hg)
Flow of participants through the trial
Secondary outcomes
Diastolic pressure did not change significantly over time between the groups (F (1.9, 737.9) = 0.08; P = 0.91) (fractional degrees of freedom result from adjustments made when Mauchly's test for sphericity was significant) (table 2). The groups did not differ over time for anxiety, smoking, exercise, salt intake, or number of prescribed drugs (tables 3 and 4). Body mass index reduced significantly more over time in the intervention group than in the control group (F (1.6, 590.9) = 6.010; P = 0.005). Most of the reduction occurred in the first six months of the study: with the repeated measures comparisons, the change between baseline and six months is significant (F (1, 370) = 10.3; P = 0.001) whereas that between six months and one year is not (F (1, 370) = 0.58; P = 0.45). Reported alcohol intake reduced significantly in the intervention group compared with the control group in the first six months but not thereafter (P = 0.03 and P = 0.56 respectively).
Table 3 Secondary outcomes: anxiety, body mass index, and number of prescribed medications
Table 4 Secondary outcomes: health behaviour. Values are numbers (percentages) of participants, unless stated otherwise
Table 5 shows patients' preferences at the end of the study for method of blood pressure measurement. The ranking was significantly different between the intervention and control groups (F (2.1, 820.5) = 37.4; P < 0.001). Patients in the intervention group ranked home measurement highest, followed by self measurement in the surgery. Those in the control group ranked measurement by a doctor highest, followed by nurse measurement.
Table 5 Patients' ranking of methods of blood pressure measurement at end of study*
Sensitivity analysis
Missing values analysis on the primary outcome of systolic blood pressure found that for substitution of missing values by either the mean of the series or by the last recorded value for an individual, the change in systolic blood pressure in the first six months remained significant (P = 0.03 in both cases), but the overall difference between the groups over time was no longer significant (P = 0.11 (substitution by mean of series) and P = 0.10 (substitution by last recorded value)).
Cost effectiveness
Table 6 shows the cost and effectiveness results. The intervention group consulted less frequently than the control group, and the drug costs did not differ significantly between the two groups. The reduction in consultation rate observed in the intervention group reflected fewer consultations in which blood pressure monitoring alone took place. Intervention costs (￡26.80 per patient) were dominated by the cost of general practitioners' time in training staff and patients (￡25.40), with discounted equipment costs relatively trivial (￡1.39). The incremental cost effectiveness ratio for practice based self monitoring—that is, the cost per additional 1 mm Hg reduction in blood pressure—was ￡5.10, with confidence intervals that crossed zero (table 5).
Table 6 Costs and effects of self monitoring compared with usual care (adjusted effects). Values are per patient per year (95% confidence interval)
Discussion
Mathers CD, Bernard C, Iburg CM, Innoue M, Fat DM, Shibuya K, et al. Global burden of disease in 2002: data sources, methods and results. Geneva: World Health Organization, 2003.
Psaty BM, Lumley T, Furberg CD, Schellenbaum G, Pahor M, Alderman MH, et al. Health outcomes associated with various antihypertensive therapies used as first-line agents: a network meta-analysis. JAMA 2003;289: 2534-44.
Montgomery AA, Fahey T, Ben Shlomo Y, Harding J. The influence of absolute cardiovascular risk, patient utilities, and costs on the decision to treat hypertension: a Markov decision analysis. J Hypertens 2003;21: 1753-9.
Williams B, Poulter NR, Brown MJ, Davis M, McInnes GT, Potter JF, et al. Guidelines for management of hypertension: report of the fourth working party of the British Hypertension Society, 2004-BHS IV. J Hum Hypertens 2004;18: 139-85.
Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA, Izzo JL Jr, et al. The Seventh report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure: the JNC 7 report. JAMA 2003;289: 2560-72.
Wolf-Maier K, Cooper RS, Kramer H, Banegas JR, Giampaoli S, Joffres MR, et al. Hypertension treatment and control in five European countries, Canada, and the United States. Hypertension 2004;43: 10-7.
Phillips LS, Branch WT, Cook CB, Doyle JP, El Kebbi IM, Gallina DL, et al. Clinical inertia. Ann Intern Med 2001;135: 825-34.
Marshall T, Rouse A. Resource implications and health benefits of primary prevention strategies for cardiovascular disease in people aged 30 to 74: mathematical modelling study. BMJ 2002;325: 197.
Jones JK, Gorkin L, Lian JF, Staffa JA, Fletcher AP. Discontinuation of and changes in treatment after start of new courses of antihypertensive drugs: a study of a United Kingdom population. BMJ 1995;311: 293-5.
Ogden J, Ambrose L, Khadra A, Manthri S, Symons L, Vass A, et al. A questionnaire study of GPs' and patients' beliefs about the different components of patient centredness. Patient Education and Counseling 2002;47: 223-7.
Steel N. Thresholds for taking antihypertensive drugs in different professional and lay groups: questionnaire survey. BMJ 2000;320: 1446-7.
Cappuccio FP, Kerry SM, Forbes L, Donald A. Blood pressure control by home monitoring: meta-analysis of randomised trials. BMJ 2004;329: 145-51.
Soghikian K, Casper SM, Fireman BH, Hunkeler EM, Hurley LB, Tekawa IS, et al. Home blood pressure monitoring. Effect on use of medical services and medical care costs. Med Care 1992;30: 855-65.
Little P, Barnett J, Barnsley L, Marjoram J, Fitzgerald-Barron A, Mant D. Comparison of agreement between different measures of blood pressure in primary care and daytime ambulatory blood pressure. BMJ 2002;325: 254-7.
Hamilton W, Round A, Goodchild R, Baker C. Do community based self-reading sphygmomanometers improve detection of hypertension? A feasibility study. J Public Health Med 2003;25: 125-30.
Townsend P, Phillimore P, Beattie A. Health and deprivation. Inequality and the north. London: Croom Helm, 1988.
Schulz KF, Grimes DA. Generation of allocation sequences in randomised trials: chance, not choice. Lancet 2002;359: 515-9.
El Assad MA, Topouchian JA, Asmar RG. Evaluation of two devices for self-measurement of blood pressure according to the international protocol: the Omron M5-I and the Omron 705IT. Blood Pressure Monitoring 2003;8: 127-33.
Ramsay L, Williams B, Johnston G, MacGregor G, Poston L, Potter J, et al. Guidelines for management of hypertension: report of the third working party of the British Hypertension Society. J Hum Hypertens 1999;13: 569-92.
Marteau TM, Bekker H. The development of a six-item short-form of the state scale of the Spielberger state-trait anxiety inventory (STAI). Br J Clin Psychol 1992;31(part 3): 301-6.
Mant J, Murphy M, Rose P, Vessey M. The accuracy of general practitioner records of smoking and alcohol use: comparison with patient questionnaires. J Public Health Med 2000;22: 198-201.
Godin G, Shephard RJ. A simple method to assess exercise behavior in the community. Can J Appl Sport Sci 1985;10: 141-6.
Little P, Barnett J, Margetts B, Kinmonth AL, Gabbay J, Thompson R, et al. The validity of dietary assessment in general practice. J Epidemiol Community Health 1999;53: 165-72.
Hollis S, Campbell F. What is meant by intention to treat analysis? Survey of published randomised controlled trials. BMJ 1999;319: 670-4.
HM Treasury. The Green Book: appraisal and evaluation in central government. London: Stationery Office, 2003.
Netten A, Curtis L. Unit costs of health and social care 2002. www.pssru.ac.uk/pdf/uc2002/Unit%20Costs%202002.pdf (accessed 8 Aug 2005).
Joint Formulary Committee. British national formulary 42. London: BMA, Royal Pharmaceutical Society of Great Britain, 2001.
Department of Health. National reference costs 2002. www.dh.gov.uk/PolicyAndGuidance/OrganisationPolicy/ FinanceAndPlanning/NHSReferenceCosts/fs/en (accessed 8 Aug 2005).
Kernick DP, Reinhold DM, Netten A. What does it cost the patient to see the doctor? Br J Gen Pract 2000;50: 401-3.
Little P, Barnett J, Barnsley L, Marjoram J, Fitzgerald-Barron A, Mant D. Comparison of acceptability of and preferences for different methods of measuring blood pressure in primary care. BMJ 2002;325: 258-9.(R J McManus, clinical senior lecturer1, )