Hypertension in the elderly: attitudes of British patients and general practitioners

Transcription

1 Journal of Human Hypertension (1998) 12, Stockton Press. All rights reserved /98 $ ORIGINAL ARTICLE Hypertension in the elderly: attitudes of British patients and general practitioners M Cranney, E Warren and T Walley Prescribing Research Group, Department of Pharmacology & Therapeutics, University of Liverpool, 70 Pembroke Place, Liverpool L69 3BX, UK The perceptions of patients and GPs of the risk of stroke in treated and untreated elderly hypertensives, and their attitudes towards anti-hypertensive therapy were examined. To explore attitudes of patients to the management of hypertension a qualitative approach was used, employing semi-structured interviews, with subsequent thematic analysis of the transcriptions. A questionnaire study of GPs attitudes to the same subject was also conducted. The elderly (n 75) greatly overestimate the risks of hypertension and the benefits of treatment. Most would accept anti-hypertensive therapy despite being informed of the true risks, citing confidence in their doctor as the major determinant in their decision. GPs (n 121) were well informed of the risks and benefits, but less than half adhere to current guidelines. GPs should be aware how much the elderly overestimate the risks of hypertension and the benefits of its treatment. When considering treating hypertension in this group, patient contributions in the treatment decisionmaking process should be actively encouraged, especially as many elderly hold a deferential attitude towards their doctor. Patients should be informed of the risks of their disease and the benefits of treatment in terms they understand. The use of visual aids helps patients to grasp the difficult concepts of risk and benefit. Keywords: hypertension; elderly; general practitioners; attitudes; primary care Introduction Treating hypertension in the elderly prevents strokes and heart attacks. 1 3 Guidelines have been produced to facilitate the implementation of this message. 4 Unfortunately many studies show that this condition is currently not well managed in British general practice. 5 Effectively, guidelines are frequently ignored despite strong, consistent and convincing evidence that treating healthy elderly patients with hypertension can prevent disabling consequences of coronary and cerebrovascular disease. 2 Although the benefits to populations of patients are well defined by the existing evidence, patients and GPs may be more influenced in their decisions to accept treatment or to treat respectively by the benefits of treatment to the individual patient. General practitioners and patients have moved towards sharing medical decisions. 6 The patient brings knowledge of their own subjective aims and values, while the GP brings knowledge of the risks and benefits of various treatment options, both parties contributing to an effective consultation. 7,8 This requires that doctors themselves are well informed of the risks and benefits, and communicate these to the patient. A recent report 9 has identified that discrepancies between the doctor s and the patient s perception of risk may be a significant factor in non- Correspondence: Mike Cranney, 17 Villiers Crescent, Eccleston, Merseyside WA10 5HP, UK Received 6 March 1998; revised and accepted 23 April 1998 compliance. However, there has been little research done in this important area and it is uncertain what influence accurate information has on the patient s decision to accept treatment. To identify some of the barriers preventing implementation of guidelines for the management of hypertension we explored how the well elderly and GPs perceived the risks of stroke in untreated hypertension and the benefits of treatment, and whether knowledge of the true risks and benefits altered their attitudes to treatment. Patients and methods The study involved two groups, patients and GPs. The elderly subjects of this study were healthy patients between the ages of 65 and 79 (defined as those who had not received any medication in the previous 12 months). They were all drawn from the practice of one of the authors (MC), this practice is situated in a suburb of Liverpool and comprises predominantly social class three and four patients, it has a higher than average elderly population with 19.5% 65 years; 9.5% 75 years and 5% 80 years. All patients in the specified age range were invited to participate. The invitation to participate was made by letter and sent on university headed notepaper. A research assistant at the university who was unknown to the patients of the practice signed the letter and a stamped addressed envelope was enclosed to facilitate reply. A second and third letter was devised in the event of a poor response. A target of patients was aimed for. Patients

2 540 were asked for their co-operation with research into the views of patients towards blood pressure and its treatment and were asked if they were prepared to allow a research assistant from Liverpool University to visit them at home. The GP subjects were attending two, week long, refresher courses in April The demographic characteristics of the participating GPs are summarised in Table 1. Two similar questionnaires were devised, after piloting, to explore the pre-existing understanding and attitudes of the elderly and GPs towards hypertension (Appendix 1); opinions were sought from the elderly and the GPs before and after seeing the actual risks and benefits as defined by a major trial. We presented data and sought responses in terms of absolute risk, ie, strokes prevented per 100 patients treated. The patients questionnaire was administered by two trained research assistants using semistructured interviews in the patients home. All of the interviews were audiotaped and transcribed. The GPs completed their written questionnaires individually under supervision. The pilot stages had revealed how difficult it was to explore patients perceptions of risk. We therefore piloted several potential visual aids before deciding on that illustrated in Appendix 1, which was found the most understandable by the patients. These were then used with the questionnaire to explore the concept of risk and other factors that might influence decisions to accept treatment. Some open-ended questions were used so as not to limit the scope of the patients answers. 10 In providing a rigorous procedure for the analysis of the qualitative data, two of the authors (MC and EW) independently conducted thematic analysis of the patients responses using an agreed coding framework. The results of the separate analyses were found to be concordant, thus establishing interrater reliability. 11 In view of the qualitative nature of these findings, we report emergent themes and support them with some examples of direct quotations from patients. To determine whether GPs adhered to currently accepted guidelines, the GPs were also asked about their management of a hypothetical patient who, by any of these guidelines, should have received drug treatment (see Appendix 1). Table 1 Characteristics of GP respondents (n = 121) Years since qualification: Range: 6 to 43 years. Mean: 20 years. Median: 18 years. Postgraduate qualifications: None: 47 (39%) MRCGP: 39 (32%) Other Diploma: 35 (29%) Both of above: 27 (22%) Partner in a training practice: Yes: 40 (33%) No: 81 (66%) Sex of GP: Male: 90 (74%) Female: 31 (26%) Partnership status: Single handed: 15 (12%) In a partnership: 106 (88%) Results Of the 149 elderly invited to participate, 75 agreed (age range: 66 92; mean age: 73). As the target number of patients had been achieved with the first invitation letter, no further invitations were sent. All of the 121 GPs on the courses returned valid completed questionnaires. Perceptions of risk and benefits The perception of the elderly of the risk of stroke was high with over 40% of untreated hypertensive patients expected to have a stroke within 5 years. Similarly, the elderly perceived the benefits of treatment to be correspondingly great, in thinking there was more than a 50% relative reduction of stroke risk. GPs were more accurate in their predictions, giving an absolute risk of 11% and a relative reduction of 50% (Figure 1). There was no correlation between number of years qualified and perception of risk. Patients attitudes to treatment Almost all the elderly initially said they would be willing to take anti-hypertensive medication if prescribed by their doctor with only 3.9% being uncertain or declining treatment. However, after being shown the actual risks and benefits, a quarter became uncertain or would decline (Table 2). Five main themes emerged from the patients initial statements. The frequency of these responses is shown in Table 3. (i) Confidence in doctor This was the strongest theme emerging from patient s responses, both before and after receiving an explanation of the real risks of stroke. Many felt that GPs held the expertise and it was therefore pointless seeking advice and not following it. Also, they felt that GPs would not prescribe tablets unless they were of benefit. Pt no: 34 Yes I certainly would take them, because I believe in the doctor and whatever the doctor tells me I would do it. (ii) Control of blood pressure An awareness of the importance of keeping blood pressure controlled was another common opinion expressed by patients. Following an explanation of the real risks of stroke, fewer patients made reference to blood pressure reduction, however, of those who did, the predominant feeling was that it was still necessary to take anti-hypertensive tablets as prevention was far better than cure. Pt. no: 4 It s a difficult one isn t it? I think I still would take them. You are on the safe side I think, you re doing something to try and stop it or keep it level or whatever. (iii) Vague benefits Some patients referred to vague benefits of taking tablets for blood pressure reduction, demonstrating a belief that treatment

3 541 Figure 1 Views of elderly patients and GPs: perceived risk of stroke with and without treatment for hypertension; compared with actual results from the MRC trial. 1 = Perceived risk of stroke held by elderly patients, for untreated hypertensiion in their age group. 2 = Perceived risk of stroke held by GPs, for untreated elderly hypertensives. 3 = Actual risk of stroke in untreated elderly hypertensives found in MRC trial. 4 = Perceived risk of stroke held by elderly patients, for treated hypertensives in their age group. 5 = Perceived risk of stroke held by GPs, for treated elderly hypertensives. 6 = Actual risk of stroke in treated elderly hypertensives found in MRC trial. Mean scores with 95% confidence intervals % confidence intervals mean Table 2 Willingness of elderly to accept drug treatment for hypertension Table 3 Reasons given by patients to explain their treatment decisions (n = 75) Yes No Don t know Reasons identified Number of patients giving a particular reason (%) Decision to accept drug treatment for hypertension, prior to seeing trial data. (96%) (2.6%) (1.3%) Before seeing trial data After seeing trial data Decision to accept drug treatment for hypertension, after seeing trial data. (72%) (13.3%) (14.6%) must be helping them in some way; however, this was very much related to patients trust in their GPs advice. Following an explanation of the real risks of stroke, perhaps paradoxically, more patients expressed a belief in the vague benefits of treatment. In addition to reasons already mentioned, some patients now felt that, although the risks were less than imagined, they would still rather not take the chance that they could be one of the few people who could suffer a stroke as a result of high blood pressure. (iv) Prevention of stroke Some patients specifically stated that they would accept treatment, as it was felt that taking tablets to reduce blood pressure, although no guarantee, could give them some control and help protect against having a stroke. Pt. No 8 Yes I would. To keep you safe from having a stroke of course. (i) Explanations given for accepting treatment Confidence in doctor 38 (51%) 30 (40%) Control of blood pressure 15 (20%) 6 (8%) Vague benefits 13 (17%) 18 (24%) Prevention of stroke 8 (10%) 4 (5%) Side effects 3 (4%) 0 Any reduction worthwhile 0 12 (16%) Try and see 0 5 (7%) (ii) Explanations given for not accepting treatment Side effects 2 (3%) 15 (20%) Benefits not as great as imagined 0 6 (8%) More information required 0 6 (8%) Totals* *Totals greater than sample size as some patients gave more than one reason. (v) Side effects Before we informed patients of the real risks of stroke, five (7%) mentioned side effects, and for two of the elderly, these were considered sufficiently significant to dissuade them from accepting therapy. However, after seeing the real risks of stroke, 15 patients (20%) gave side effects as the main reason why they would not accept treatment. This group felt that it was not worth enduring

4 542 the unpleasant side effects of the tablets for the apparently small reduction in the risk of stroke. After the explanation of the real risks of stroke in untreated and treated hypertension, confidence in the doctor was still the most important theme encouraging the elderly to accept treatment (see Table 3). For those who would continue to accept treatment, two new themes emerged: the first being that any reduction in the risk of stroke, however limited, was useful; the second, was that treatment would be cautiously taken, but reconsidered if any adverse effects developed. For those patients who changed their minds about accepting treatment, after receiving the same information of the real risks of stroke, two new themes of equal strength were identified from their responses. Some patients reported that they would be unwilling to accept treatment since the benefits were not as great as they had imagined initially. Other patients stated that they would consider accepting hypertensive treatment only after receiving further information from their doctor regarding such issues as immediate and long term side effects of medication and other possible restrictions that treatment might place upon their lifestyle. GP attitudes to treatment In their management of the hypothetical patient (see Appendix 1), only 51 (42%) of the GPs would have treated the patient based on their pre-existing knowledge. Fifty-three (44%) said they would not treat the patient, with a further 17 (14%) uncertain. There was no correlation between intention-to-treat and perception of risk, nor between intention-totreat and perceived benefits of treatment. The reasons given for treating or not treating initially are shown in Table 4. After seeing the actual risks and benefits reported by the MRC trial, 12 only 9% of the GPs changed their minds. Six per cent more now opted not to treat, citing lack of benefit as the reason. Three per cent more opted to treat, citing the reduction of risk of stroke as the reason. Table 4 Explanations given by GPs for their initial treatment decisions Reasons identified No. of mentions* (a) Would treat: n = 51 (42%) (i) Reduced risk of stroke 37 (ii) Reduced myocardial infarction 16 (iii) Reduced morbidity or mortality 10 (b) Would not treat: n = 53 (44%) (i) Treatment not worthwhile 29 (ii) Wish to consider other risk factors first 26 (iii) Wish to take more blood pressure 25 readings first (iv) Fear of potential side effects 12 (c) No explanation offered: n = 17 (14%) *Some respondents gave more than one explanation. Discussion The elderly greatly overestimated the risks of stroke from hypertension and the benefits of treatment. When shown the actual risks and benefits, a quarter changed their minds and would consider not accepting treatment. Active patient participation in decision making improves control in conditions such as hypertension. 13 Furthermore, a recent report 14 has suggested that compliance by the elderly could be improved by carefully explaining to them the need for medication and detailing possible side effects. We wanted to know whether providing such information, in a readily understandable way, would affect the decision of the elderly to accept treatment. We found that the visual aids used to convey the concept of risk greatly facilitated patient understanding of this abstract concept. Surprisingly, we found that despite knowing they had greatly overestimated the benefits, the majority of older people would still accept treatment. Most expressed continuing confidence in the authority of the doctor as the main reason for such concordance. GPs should be aware of this deferential attitude, and resist the temptation to practise paternalistic medicine. Even those patients who have little desire to be involved in the decision-making process, often wish for further information on their condition. 15 Although a wide variation occurred in the assessment by GPs of stroke risk in untreated and treated hypertension (1 60% vs 0 30%), the mean scores were reassuringly close to published assessments of stroke risk in non-treatment and treatment groups. 3 However, there was an interesting dichotomy of treatment decisions amongst GPs. The scenario of an individual patient was more likely to yield a positive intention to treat, than simply asking, Do you treat elderly hypertensives. 16 However, despite this and the evident awareness of risks of untreated hypertension and benefits of treatment, almost half of the GPs stated they would not treat a 70-year-old man with a blood pressure level that would certainly merit treatment according to all of the major guidelines. Given their general awareness of the absolute benefits, it is not surprising that very few GPs (only 9%) changed their treatment decisions after seeing the data from the MRC trial, and in no consistent manner in contrast to the elderly. Although more recent studies 1 3 give slightly different stroke rates, we presented data from the MRC trial because this was published in 1992 and was more likely to be known by the participating GPs. GPs have higher thresholds 17 for defining and treating hypertension in the elderly than is recommended by various guidelines. 4,18 21 Despite the guidelines, GPs disagreed on whether the hypothetical patient should be treated, with almost half not wishing to treat. This negative attitude of some GPs towards treating the elderly hypertensive, previously noted also by others, 22,23 is clearly not due to ignorance of the true absolute benefits of treating hypertension. The major reason, mentioned by over half of the non-treating GPs, seems to have been that they were not convinced that the benefits made treatment worthwhile. The least common reason

5 offered was concern about adverse effects of drugs (in contrast to an earlier study where side effects were the most common reason given for not prescribing). 24 Many of the GPs in our study wished to consider other risk factors before initiating treatment. Recently it has been suggested that GPs should make a multi-factorial assessment of all risk factors before commencing treatment, 25 a view supported by the New Zealand guidelines. 19 However, in the elderly the presence of a systolic pressure at or over 160 mm hg, or a diastolic pressure at or over 90 mm hg, indicates a substantial and potentially reversible risk of cardiovascular events, even in the absence of other risk factors. 5,26,27 Therefore, treating the elderly on the basis of hypertension alone, irrespective of other risk factors, is still appropriate. 28 Our sample of elderly patients was drawn from a list known to one GP and only half accepted the invitation to participate. This low response rate may have produced a biased picture. Furthermore our sample consisted entirely of healthy older patients who have better relationships with their GP than ill older patients, who are more critical of the care they receive. 29 This additional bias could have affected the results and might explain the high level of confidence in the opinion of their GP. Our sample was not representative of the whole population. However, it is recognised that qualitative data collection is too time consuming and expensive to permit use of a probability sample, as a result statistical representativeness is not normally sought in qualitative research. 11 The generalisability of qualitative research is therefore accepted on a conceptual rather than a numerical level. 30,31 Furthermore, our sample size of 75 was larger than most qualitative studies, which usually do not interview more than 50 or 60 people. 10 Indeed, it was decided that a second invitation need not be sent to the non-responders as a sufficiently large enough sample had been obtained for the purposes of qualitative assessment. We had decided on a target population of participants partly for these considerations and partly because of a limitation in human and financial resources. We used independent interviewers, unknown by the patients, to obtain more truthful opinions, 32 and hopefully obviated the bias of allegiance to their current GP when giving their responses. Some GPs said they would not treat the hypothetical patient in the given scenario because they would first like to collect more blood pressure readings. Although this reflects a laudable adherence to guidelines, they had been informed when completing their questionnaires that the figure of 170/94 mm Hg was sustained, nevertheless these 25 GPs may have misunderstood the scenario presented, and this was a flaw in the questionnaire used. In summary, our results demonstrate the deferential attitudes elderly patients hold towards their GPs. We need to be aware of this and should endeavour to empower patients to contribute to decisions about treatment, encouraging a true partnership between patient and prescriber. 9 It should also be recognised that the elderly greatly overestimate the benefits of treatment for hypertension. It appears that a large percentage of GPs are unconvinced by the published guidelines for this condition. Future guideline development should therefore aim to be simpler rather than more complex and should contain a summary of outcomes of treatment. This data could be expressed in a readily accessible way, 33 sensitive to the effects of different methods of data presentation upon treatment decisions. 34 Finally, the use of visual aids can help patients co-operate with their doctors in making decisions. Acknowledgements We would like to thank all of the patients and GPs who co-operated in this study, two anonymous referees for their helpful comments and the North West Regional Health Authority for funding this project. References 1 Insua JT, Sacks H, Lau TS. Drug treatment of hypertension in the elderly: a meta-analysis. Ann Intern Med 1994; 121: Mulrow CD, Cornell JA, Herrera CR. Hypertension in the elderly. JAMA 1994; 272: Pearce KA, Furberg CD, Rushing J. Does antihypertensive treatment of the elderly prevent cardiovascular events or prolong life? A meta-analysis of hypertensive treatment trials. Arch Fam Med 1995; 4 Nov: Sever P, Beevers G, Bulpitt CJ. Management guidelines in essential hypertension: report of the second working party of the British Hypertension Society. Br Med J 1993; 306: Aylett MJ et al. Blood pressure control of drug treated hypertension in 18 general practices. J Hum Hypertens 1996; 10: Brock D, Wartman S. When competent patients make irrational choices. N Engl J Med 1990; 322(22): Forrow L, Wartman SA, Brock DW. Science, ethics, and the making of clinical decisions. JAMA 1988; 259: Middleton JF. The exceptional potential of the consultation revisited. J Roy Coll Gen Pract 1989; 39: The Royal Pharmaceutical Society of Great Britain Working Party. From compliance to concordance: achieving shared goals in medicine taking. Royal Pharmaceutical Society of Great Britain, London, Patton MQ. How to use qualitative methods in evaluation. Sage: London, 1987, pp Mays N, Pope C. Rigour and qualitative research. Br Med J 1995; 311: MRC Working Party. Medical Research Council trial of treatment in older adults: principal results. Br Med J 1992, 304: Kaplan SH, Greenfield S, Ware JE. Assessing the effects of physician-patient interactions in the outcomes of chronic disease. Med Care 1989; 27 (Suppl): S110 S The Royal College of Physicians of London Working Party. Medication for older people. second edition. The Royal College of Physicians of London, Nease RF, Blair-Brooks W. Patient desire for information and decision making in health care decisions. J Gen Int Med 1995; 10: Redelmeier DA, Tversky A. Discrepancy between medical decisions for individual patients and for groups. N Engl J Med 1990; 322(16):

6 Dickerson JEC, Garratt CJ, Brown MJ. Management of hypertension in general practice: agreements with and variations from the British Hypertension Society guidelines. J Hum Hypertens 1995; 9: Beard K et al. Management of elderly patients with sustained hypertension. Br Med J 1992; 304: Jackson R, Barham P, Bills J. Management of raised blood pressure in New Zealand: a discussion document. Br Med J 1993; 307: Subcommittee of WHO/ISH Mild Hypertension Liaison Committee. Summary of 1993 World Health Organisation-International Society of Hypertension guidelines for the management of mild hypertension. Br Med J 1993; 307: Bennet NE. Hypertension in the elderly. Lancet 1994; 344: Fotherby MD, Harper GD, Potter JF. General practitioners management of hypertension in elderly patients. Br Med J 1992; 305: Fahey T, Silagy C. General practitioners knowledge of and attitudes to the management of hypertension in elderly patients. BJGP 1994; 44: Ekpo EB, Shah IU, Fernando MU, White AD. Isolated systolic hypertension in the elderly: survey of practitioners attitude and management. Gerontology 1993; 39: Fahey TP, Peters TJ. A general practice-based study examining the absolute risk of cardiovascular disease in treated hypertensive patients. BJGP 1996; 46: Larson MG. Assessment of cardiovascular risk factors in the elderly: the Framingham Heart study. Statistics in Medicine 1995; 14: Simon JA. Treating hypertension: the evidence from clinical trials. Br Med J 1996; 313: Barton S, Cranney M, Hatcher J, Walley T. The risk of cardiovascular disease in hypertensive patients. BJGP 1997; 47: Cartwright A. Medicine taking by people aged 65 or more. Br Med Bull 1990; 46: Fitzpatrick R, Boulton M. Qualitative methods for assessing health care. Quality in health care 1994; 3: Green J, Britten N. Qualitative research and evidence based medicine. Br Med J 1998; 316: Britten N. Qualitative interviews in medical research. Br Med J 1995; 311: Robson, J. Information needed to decide about cardiovascular treatment in primary care. Br Med J 1997; 314: Cranney M, Walley T. Same information, different decisions: the influence of evidence on the management of hypertension in the elderly. BJGP 1996; 46: Appendix 1 Questionnaire administered to healthy elderly patients You may know that high blood pressure can, in time, lead to a stroke. I am trying to find out how risky you think high blood pressure is in relation to stroke. Question 1) Here are 100 people aged over 65 years. with high blood pressure. How many of them do you think will have a stroke in the next five years because of their high blood pressure? Visual aid 1: (This displays a representation of 100 people, the respondent was asked to draw a line through the number they thought might have a stroke.) Question 2) Here are another 100 people who also have high blood pressure. However they take tablets to control their blood pressure. How many of them do you think will have a stroke in the next five years? (Visual aid 1: as above) Question 3) If you (or someone you knew) were told by your doctor you had high blood pressure and were prescribed tablets for it, would you take the tablets or encourage someone else to take them? If yes, why? If not, why not? If unsure, what else would you want to know to make up your mind? Question 4) (The correct values for questions 1 and 2 are now shown using visual aids 2 and 3) {ref MRC trial} 12 Given what you now know about the risks of high blood pressure and the difference that treatment makes, would you still take blood pressure tablets if your doctor prescribed them for you, and would you encourage someone else to take them? Visual Aid 2 Represents the risk of stroke in 100 Visual Aid 3 Represents the risk of stroke in 100 treated untreated elderly hypertensives over five years. (5.26) elderly hypertensives over five years. (3.57)

7 GP Questionnaire 545 Question 1. Here are 100 people aged over 65 years with high blood pressure ( 162/92). None of them take any tablets for their blood pressure. Assuming none have any other risk factors, how many of them do you think will have a stroke in the next five years? (Visual aid 1: as above.) Question 2. Here are another 100 people aged over 65 years who also have high blood pressure ( 162/92), and no other risk factors, however they take tablets which effectively controls their blood pressure. How many of them do you think will have a stroke in the next five years? (Visual aid 1: as above.) Question 3. If you had a male patient aged 70 years with a BP 170/94 would you prescribe tablets for him to lower his BP? (yes/no/don t know) Please enter your reasons. Question 4. (The correct values for questions 1 and 2 are now shown using visual aids 2 and 3: as above). Given what you now know about the risks of high blood pressure and the difference that treatment makes, would your prescribing habits remain the same as in question 3 or have you changed your viewpoint? (i) remain the same: (ii) change: (please explain your decision)