BMJ
Confidential: For Review Only
Randomis ed controlled trial of different systolic blood pressure targets for people with a history of stroke or transient ischaemic attack: the PAST-BP (Prevention After Stroke – Blood Pressure) study.
Journal: BMJ
Manuscript ID BMJ.2015.029605
Article Type: Research
BMJ Journal: BMJ
Date Submitted by the Author: 26-Sep-2015
Complete List of Authors: Mant, Jonathan; University of Cambridge, General Practice and Primary Care Research Unit McManus, Richard; University of Oxford, Dept of Primary Care Health Sciences Roalfe, Andrea; University of Birmingham, Primary Care Clinical Sciences Fletcher, Kate; University of Birmingham, Primary Care Clinical Sciences Taylor, Clare; University of Birmingham, Martin, Una; University of Birmingham, Dept of Clinical Pharmacology Virdee, Satnam; University of Birmingham, Greenfield, Sheila; University of Birmingham, Primary Care and General Practice Hobbs, FD Ricahrd; University of Oxford, Department of Primary Care Health Science
stroke, hypertension, randomised controlled trial, blood pressure lowering, Keywords: primary care
https://mc.manuscriptcentral.com/bmj Page 1 of 35 BMJ
1 2 3 Randomised controlled trial of different systolic blood pressure targets for 4 5 people with a history of stroke or transient ischaemic attack: the PAST-BP 6 (Prevention After Stroke – Blood Pressure) study. 7 8 Confidential: For Review Only 9 10 11 12 13 14 15 16 17 Jonathan Mant, Professor of Primary Care Research (1) 18 19 Richard J McManus, (2) NIHR Professor of Primary Care Research 20 21 Andrea Roalfe, Senior Lecturer in Medical Statistics (3) 22 23 Kate Fletcher, (3) Non-clinical Director Primary Care Clinical Research Trials Unit (PC-CRTU) 24 25 Clare J Taylor, (3) General Practitioner and NIHR Doctoral Research Fellow 26 27 Una Martin, (4) Professor of Clinical Pharmacology 28 29 Satnam Virdee, (3) Research Associate 30 31 Sheila Greenfield, Professor of Medical Sociology (3) 32 33 FD Richard Hobbs, Professor of Primary Care Health Sciences (2) 34 35 36 37 (1) Primary Care Unit, Department of Public Health & Primary Care, Strangeways Research 38 Laboratory, University of Cambridge, Wort’s Causeway, Cambridge CB1 8RN 39 40 41 (2) Nuffield Department of Primary Care Health Sciences, NIHR School for Primary Care 42 Research, University of Oxford, Oxford, OX2 6GG. 43 44 45 (3) Primary Care Clinical Sciences, University of Birmingham, Birmingham, B15 2TT. 46 47 (4) School of Clinical & Experimental Medicine, University of Birmingham, Birmingham B15 2TT. 48 49 50 51 Correspondence to: 52 53 Jonathan Mant 54 55 [email protected] 56 57 01223 330325 58 59 60 https://mc.manuscriptcentral.com/bmj BMJ Page 2 of 35
1 2 3 Abstract 4 5 6 Objectives: Blood pressure lowering is effective at reducing risk of stroke recurrence in people who 7 8 Confidential:have had a cerebrovascular event, but it is Foruncertain how Review low blood pressure should Only be lowered in 9 10 this population. We assessed whether using intensive blood pressure targets would lead to lower 11 12 blood pressure in a community population of people with prevalent cerebrovascular disease. 13 14 15 Design: Open label randomised controlled trial. 16 17 Setting: 99 General Practices in England, with participants recruited 2009-2011. 18 19 20 Participants: People with a history of stroke or transient ischaemic attack whose systolic blood 21 22 pressure was ≥ 125 mmHg. 23 24 25 Interventions: Intensive systolic blood pressure target (130mmHg or 10mmHg reduction from 26 27 baseline if this was < 140 mmHg) or a standard target (140mmHg). Apart from the different target, 28 29 patients in both arms were actively managed in the same way with regular reviews by the primary 30 31 care team. 32 33 34 Main outcome measure : Change in systolic blood pressure between baseline and twelve months. 35 36 37 Results : 529 patients, mean age 72, were enrolled, 266 to the intensive target arm and 263 to the 38 39 standard target arm, of whom 379 were included in the primary analysis (182, 68% intensive arm; 40 41 197, 75% standard arm). 84 patients withdrew from the study during the follow up period (52 42 43 intensive arm; 32 standard arm). Mean systolic blood pressure dropped by 16.1mmHg to 44 45 127.4mmHg in the intensive target arm and by 12.8mmHg to 129.4mmmHg in the standard arm 46 47 (difference between groups 2.9 mmHg, 95% confidence interval (0.2 to 5.7); p = 0.03). 48 49 50 Conclusions: Aiming for a 130mmHg or lower target for systolic blood pressure in people with 51 52 cerebrovascular disease in primary care rather than a 140mmHg target leads to a small additional 53 54 reduction in blood pressure. Active management of systolic blood pressure in this population using 55 56 57 a 140mmHg target leads to a clinically important reduction in blood pressure. 58 59 60 https://mc.manuscriptcentral.com/bmj Page 3 of 35 BMJ
1 2 3 Trial Registration : ISRCTN29062286. 4 5 6 7 8 Confidential: For Review Only 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 https://mc.manuscriptcentral.com/bmj BMJ Page 4 of 35
1 2 3 Introduction 4 5 6 Stroke accounts for about 10% of deaths internationally, and for over 4% of direct health care costs 7 1 8 Confidential:in developed countries. If other resources, For such as lost Reviewproductivity, benefits payments Only and 9 10 informal care costs are taken into account, the total costs double – for example in the United 11 12 Kingdom annual care costs are around £4.4 billion, but total costs are £9 billion per annum. 2 Over 13 14 20% of strokes are recurrent events, 3 and if one also takes into account prior history of transient 15 16 ischaemic attack (TIA), this figure rises to about 30%. 1 Therefore, secondary prevention has a major 17 18 19 potential role to play in reducing both morbidity and costs of stroke care. Hypertension is a key risk 20 21 factor for stroke. A 20 mm Hg difference in usual systolic blood pressure is associated with a 60% 22 23 lower risk of death from stroke in someone aged 50 to 70, and a 50% lower risk in someone aged 70 24 25 to 79. 4 26 27 28 The PROGRESS trial demonstrated that treatment to lower blood pressure in people who have had a 29 30 stroke or TIA reduces risk of further stroke. 5 However, there is debate over how to apply this 31 32 evidence in clinical practice.6 7 In particular, there is uncertainty over how intensively to lower blood 33 34 pressure in people who have had a stroke or TIA. 8 A post hoc observational analysis of the PROFESS 35 36 trial found that people with recent ischaemic stroke whose systolic blood pressure was less than 37 38 130mmHg had a higher risk of vascular events. 9 Conversely, in PROGRESS participants whose 39 40 41 baseline systolic blood pressure was 120-140mmHg who were randomised to combination therapy 42 10 43 had significantly reduced stroke risk. The SPS3 trial of different blood pressure targets in younger 44 45 (mean age 63) patients with recent lacunar stroke found a non-significant 19% reduction in risk of 46 47 stroke after one year in people treated with a systolic blood pressure target of less than 130 mmHg 48 49 as compared to a 130-149mmHg target. 11 Recent guidelines have drawn different conclusions from 50 51 the evidence base, with the European guidelines recommending a target systolic blood pressure of 52
53 12 13 54 140mmHg (or higher) and British guidelines a target of 130mmHg. 55 56 57 58 59 60 https://mc.manuscriptcentral.com/bmj Page 5 of 35 BMJ
1 2 3 In view of these controversies, the Prevention After Stroke- Blood Pressure (PAST-BP) study 4 5 compared two different targets for blood pressure lowering after stroke or TIA in people recruited 6 7 from a prevalent primary care population. The aim was to determine whether setting a more 8 Confidential: For Review Only 9 intensive target in primary care would lead to a lower blood pressure, as a prelude to a trial powered 10 11 12 to detect whether such a strategy would lead to a reduction in stroke recurrence. 13 14 15 16 17 Methods 18 19 20 Participants
21 14 22 The methods used in PAST-BP have been reported in detail elsewhere. PAST-BP was an individually 23 24 randomised trial in which participants were allocated either to an intensive blood pressure target 25 26 (<130mmHg or a 10mmHg reduction if baseline pressure <140mmHg) or a standard target (<140 27 28 mmHg). Patients were recruited from 106 General Practices (of whom 99 contributed at least one 29 30 patient) in England during 2009-2011. Patients were considered for inclusion if they were on the 31 32 practice TIA/stroke register. They were excluded if: their baseline systolic blood pressure was less 33 34 35 than 125 mmHg; they were already on 3 or more antihypertensives; they had >20mmHg postural 36 37 change in systolic blood pressure on standing; they were already being treated to a 130mmHg 38 39 systolic blood pressure target; they were unable to provide informed consent; or there was 40 41 insufficient corroborative evidence that they had had a stroke or TIA. Potentially eligible participants 42 43 were identified using a search of the General Practice clinical computer system. A general 44 45 practitioner reviewed this list to exclude patients for whom a study invitation would be 46 47 48 inappropriate. The remainder were sent a letter inviting them to attend a study clinic appointment 49 50 held at their General Practice by a research nurse, where written informed consent was obtained. 51 52 Ethical approval was provided by the Warwickshire Research Ethics Committee (reference 53 54 08/H1211/121). 55 56 57 58 59 60 https://mc.manuscriptcentral.com/bmj BMJ Page 6 of 35
1 2 3 Randomisation and masking 4 5 Randomisation was performed by the central study team at the University of Birmingham and was 6 7 minimised on the basis of age, sex, diabetes mellitus, atrial fibrillation, baseline systolic blood 8 Confidential: For Review Only 9 pressure and general practice. Treatment allocation was ascertained by the research nurse either by 10 11 12 telephone or online. 13 14 Neither participants nor clinicians were blinded to treatment allocation. The primary outcome 15 16 measure (blood pressure) was obtained using automated sphygmomanometers and measured by a 17 18 19 research nurse who was not otherwise involved in the patient’s care. 20 21 Procedures 22 23 Patients randomised to the intensive arm were given a target systolic blood pressure of 130mmHg, 24 25 26 or a target reduction of 10mmHg if their baseline blood pressure was between 125 and 140 mmHg. 27 28 The target in the standard arm was 140 mmHg irrespective of baseline blood pressure. Apart from 29 30 the different blood pressure targets, the management of blood pressure was the same in both 31 32 groups, and was carried out by a practice nurse (to monitor blood pressure) and a general 33 34 practitioner (responsible for modifying blood pressure treatment). Patients whose systolic blood 35 36 pressure at baseline was above target (everyone in the intensive arm, and those patients in the 37 38 standard arm whose blood pressure was greater than 140 mmHg) had their antihypertensive 39 40 41 therapy reviewed by their General Practitioner. A practice nurse would see all patients at three 42 43 month intervals (if their blood pressure was below target when previously measured) or at a one 44 45 month interval (if previous blood pressure was above target), and refer to the general practitioner if 46 47 the blood pressure was above target. No down-titration of therapy was performed if blood pressure 48 49 was below target. General practitioners were provided with treatment protocols that reflected the 50 51 national guidelines for blood pressure lowering in operation at the time of the trial.15 In both arms of 52 53 54 the trial, the general practitioners had access to a computer based algorithm that actively suggested 55 56 drugs and dosage if the participant was above target. Follow up ceased if the participant had a 57 58 major cardiovascular event. 59 60 https://mc.manuscriptcentral.com/bmj Page 7 of 35 BMJ
1 2 3 The primary outcome was change in systolic blood pressure between baseline and one year. 4 5 Participants had blood pressure measured by a research nurse (separate from the practice nurse 6 7 measurement described above) at baseline, six and twelve months. Blood pressure was measured 8 Confidential: For Review Only 9 using a British Hypertension Society validated automated electronic monitor supplied and validated 10
11 16 12 for the study. Blood pressure was measured in a standardised way, with the patient seated for five 13 14 minutes and then six measurements taken at minute intervals. The primary outcome was the 15 16 average of the second and third measurements. 17 18 19 Secondary measures of blood pressure included diastolic blood pressure at six and twelve months, 20 21 systolic blood pressure at six months, and proportion achieving target blood pressures at twelve 22 23 months. For the systolic blood pressure we also calculated the means of readings 2 to 6 and 5 to 6 to 24 25 look for any differential effects with regard to habituation to blood pressure measurement. 26 27 28 Clinical events were identified through review of the general practice record at twelve months. 29 30 These comprised: major cardiovascular events (composite of fatal and nonfatal stroke, myocardial 31 32 infarction, fatal coronary heart disease or other cardiovascular death), emergency hospital 33 34 admissions and deaths. Participants were flagged for mortality at the NHS Central Register. Side 35 36 effects were assessed through the use of standard questionnaires. 14 37 38 39 Statistical analysis 40 41 We estimated that a sample size of 305 patients in each group would detect a 5 mmHg difference in 42 43 systolic blood pressure between groups with 90% power at a significant level of 5% assuming a 44 45 standard deviation of 17.5 mmHg, 10% loss to follow up, 5% mortality and 10% major vascular 46
47 5 7 48 events. For the primary analysis, mixed models were used, adjusting for baseline blood pressure, 49 50 age group (<80 years, ≥80 years), gender, diabetes mellitus, atrial fibrillation and practice (as a 51 52 random effect). The principal analysis was a complete case analysis. We also explored the potential 53 54 effects of missing values by the use of three approaches: multiple imputation, group mean and by 55 56 last available value. Subgroup analyses were pre-specified for diabetes mellitus, atrial fibrillation 57 58 59 60 https://mc.manuscriptcentral.com/bmj BMJ Page 8 of 35
1 2 3 and age group. In addition, we performed a sub-group analysis by baseline systolic blood pressure 4 5 (<140mmHg, ≥140mmHg). The number of consultations, treatment changes and side effects were 6 7 compared using generalised mixed modelling, adjusting for the same variables as the primary 8 Confidential: For Review Only 9 outcome. For clinical events, we calculated hazard ratios and their 95% confidence interval using Cox 10 11 12 proportional hazards modelling adjusting for the same covariates mentioned previously. We 13 14 checked the proportional hazard assumption with Schoenfeld residual plots and by including 15 16 interaction terms in the model (for each term by time). For all clinical event analyses, patients were 17 18 censored at the time of the first event relevant to that analysis. Thus, if a patient had more than one 19 20 emergency hospital admission, only the first one would be counted. Analysis was undertaken using 21 22 SAS 9.2 and Stata 12. 23 24 25 26 27 28 Results 29 30 Figure 1 shows the trial profile. 529 patients from 99 general practices (range 1 – 16 per practice) 31 32 entered the trial. 84 patients withdrew from the trial in the twelve months following randomisation 33 34 35 (52, 20% in the intensive target arm and 32, 12% in the standard target arm, p =0.02). Primary 36 37 outcome data were available for 379 participants at one year follow up (182, 68% in the intensive 38 39 target arm and 197, 75% in the standard target arm). All patients were followed up for clinical 40 41 events and deaths . Table 1 shows baseline patient characteristics. About a quarter of participants 42 43 were on no blood pressure lowering treatment at randomisation (76 in intensive arm; 63 in standard 44 45 arm). For half of participants, the index event was a TIA. Just under 20% of participants reported at 46 47 48 least moderate disability (modified Rankin score of three or more). There were no important 49 50 differences in characteristics between participants who did and did not have blood pressure 51 52 recorded at twelve months. 53 54 55 The intensive target arm was associated with significantly more consultations with the general 56 57 practitioner and practice nurse for blood pressure control than the standard target arm (median 58 59 60 https://mc.manuscriptcentral.com/bmj Page 9 of 35 BMJ
1 2 3 visits 2 versus 1, p < 0.0001 and 3 versus 2, p = 0.002 respectively). This higher consultation rate led 4 5 to more intensifications of blood pressure treatment (458 versus 278, p < 0.0001), and more changes 6 7 due to side effects (77 versus 30, p < 0.0001). However, patients were also less likely to have their 8 Confidential: For Review Only 9 blood pressure treatment increased after review by the general practitioner when the blood 10 11 12 pressure was above target in the intensive arm (109 versus 57, p = 0.005) (table 2). At the end of the 13 14 study, the number of antihypertensive drugs that patients were on in both arms had increased by a 15 16 similar amount (mean number of antihypertensive drugs 2.1 in intensive arm and 1.9 in standard 17 18 arm, p = 0.13). 19 20 21 Treatment to a more intensive target was associated with a significantly greater reduction in systolic 22 23 blood pressure at twelve months (primary outcome) (table 3). Systolic blood pressure was reduced 24 25 by 16mmHg in the intensive target arm and by 13mmHg in the standard target arm. This difference 26 27 persisted if it was calculated using the mean of the 5 th and 6 th reading: -3.2 mmHg, 95%CI -5.8 to - 28 29 0.64) or the mean of the 2 nd to 6 th reading: -3.3mmHg, 95% CI -5.8 to -0.67) (see web appendix table 30 31 32 i). Taking account of the missing values had different impact depending upon the method used (see 33 34 web appendix table ii). Using multiple imputation the effect size was -3.2mmHg, 95% CI -5.7 to - 35 36 0.65, using the group mean it was -2.3 mmHg, 95% CI -4.3 to -0.32 and using the last value carried 37 38 forward -1.8mmHg, 95% CI -4.2 to 0.57. Blood pressure target at one year was achieved in 93 39 40 (51.1%) patients in the intensive arm. Proportions achieving a systolic blood pressure of less than 41 42 140 mmHg were similar in the two arms (150/182, 82.4% versus 161/197, 81.7%, p = 0.59). There 43 44 was no evidence of a significant difference in effectiveness of using an intensive blood pressure 45 46 47 target in any patient sub-group (figure 2). 48 49 There was one major cardiovascular event in the intensive target arm (a non-fatal myocardial 50 51 infarction), and five in the standard care arm (3 strokes; 1 non-fatal myocardial infarction and 1 52 53 54 cardiovascular death) (HR 0.19, 95% CI 0.02 to 1.87, p = 0.16). There were two deaths in the 55 56 intensive target arm and one in the standard target arm. Risk of emergency admission was 12.8% 57 58 per annum in the intensive target arm and 7.8% per annum in the standard target arm (HR 1.56, 59 60 https://mc.manuscriptcentral.com/bmj BMJ Page 10 of 35
1 2 3 95%CI 0.84 to 2.93, p = 0.16). Two admissions in each arm were related to falls. Apart from TIA 4 5 (responsible for five admissions in the standard target arm and three admissions in the intensive 6 7 target arm) and stroke, no single diagnosis accounted for more than two admissions. Table 4 shows 8 Confidential: For Review Only 9 the commonest symptoms at twelve months by treatment allocation. There were no significant 10 11 12 differences between the two groups. 13 14 15 16 17 Discussion 18 19 20 Statement of principal findings 21 22 We found that aiming for a target systolic blood pressure of 130 mmHg or lower in a primary care 23 24 population with prevalent cerebrovascular disease led to a lower systolic blood pressure than if a 25 26 140 mmHg target was aimed for, but the difference was small – about 3mmHg and was associated 27 28 with increased workload – an extra consultation each for GPs and nurses per year. The intensive 29 30 target arm was not associated with more side effects as measured at follow up, but there were more 31 32 changes to treatment because of side effects during the trial. More people withdrew consent for 33 34 35 the trial from the intensive target arm, and this might have reflected unwillingness to persevere with 36 37 the increased medication regime. Perhaps the most important finding was the greater than 38 39 10mmHg reductions in mean systolic blood pressure in both arms of the study, so that over 80% of 40 41 participants in each arm had achieved a blood pressure of < 140mmHg by the end of the trial, as 42 43 compared to less than 50% at baseline. 44 45 46 47 48 49 Strengths and weaknesses of the study 50 51 Blood pressure at twelve months was not available for 28% of patients randomised. This reflected a 52 53 high number of patient withdrawals from the study, with some differential loss to follow up in the 54 55 56 intensive target arm. However, if missing values were imputed using multiple imputation – the most 57 58 robust method -the difference in achieved blood pressure between arms at one year was very 59 60 https://mc.manuscriptcentral.com/bmj Page 11 of 35 BMJ
1 2 3 similar to that observed. Only 4% of patients on general practice stroke/TIA registers participated in 4 5 the trial. Participants had a low prevalence of disability for a prevalent cerebrovascular disease 6 7 population, were younger than typical patients in primary care with a history of cerebrovascular 8 Confidential: For Review Only 9 7 disease and over-represented people with a history of TIA only. It is likely therefore that the more 10 11 12 intensive target would have been even harder to achieve if the trial population was more 13 14 representative of people with prevalent cerebrovascular disease. The outcome measure was 15 16 unblinded, but obtained using an automated sphygmomanometer by a nurse not directly involved in 17 18 the participant’s care, so systematic recording bias is unlikely. 19 20 21 The standard target arm in PAST-BP was actively managed, with support of a computer based 22 23 algorithm that suggested medication changes rather than simply receiving ‘usual care’. If we had 24 25 used a more passive management strategy in the comparison group, we may have achieved a 26 27 greater separation in systolic blood pressure between arms. In another blood pressure lowering 28 29 study of patients with increased cardiovascular risk undertaken by our group in the same timeframe, 30 31 32 the standard care control arm dropped by 6mmHg from a similar baseline compared to 13mmHg in
33 17 34 the current study. We used an active control as we wanted to ascertain the impact of setting 35 36 different blood pressure targets, and to avoid confounding that would be introduced by having 37 38 different management strategies in the two arms. 39 40 41 42 43 Comparison with other studies and interpretation 44 45 46 The change in mean blood pressure that we observed in the intensive target arm was very similar to 47 48 that observed in the <130mmHg target arm of the SPS3 trial, with both PAST-BP and SPS3 achieving 49 50 a mean systolic blood pressure in the intensive arm of 127 mmHg after one year. 11 However, the 51 52 comparison arms had different achieved blood pressures (PAST-BP 129 mmHg versus SPS3 138 53 54 55 mmHg). This reflects the more conservative target in the higher target arm of SPS3 (140-159mmHg 56 57 58 59 60 https://mc.manuscriptcentral.com/bmj BMJ Page 12 of 35
1 2 3 as opposed to <140mmHg), and that antihypertensive therapy was reduced if blood pressure fell 4 5 below target. 6 7 8 Confidential:Most of the observed reduction in blood pressureFor is likely Review to have been mediated byOnly increased use 9 10 of antihypertensive drugs, which on average went up from 1 to 2 drugs per person over the year of 11 12 the study in both arms of the trial. Alternative explanations are that there was habituation to blood 13 14 pressure measurement leading to reduced white coat effect, or that there was regression dilution 15 16 bias. However, in a blood pressure monitoring trial in a similar post-stroke population with similar 17 18 19 mean baseline systolic blood pressure, no fall in blood pressure was observed in the control group 20 18 21 over a twelve month period, and in the SPS3 trial (also with similar mean baseline systolic blood 22 23 pressure to PAST-BP) there was a fall of just 4 mmHg in the 140 mmHg target arm over the study 24 25 period. 11 This suggests that the fall of 13 mmHg observed in the standard target arm of PAST-BP is 26 27 unlikely to be primarily due to effects of regression dilution or habituation to measurement. 28 29 30 Only 51% of patients in the intensive target arm of PAST-BP achieved their target blood pressure. 31 32 Both patient wishes and general practitioner decision making led to treatment not being intensified 33 34 when blood pressure was above target (table 2), and 10% of the intensive arm withdrew from the 35 36 trial because they did not want their blood pressure medication increased. Although reported side 37 38 effects and symptoms were similar in the two arms, and serious adverse events were infrequent 39 40 41 (two admissions for falls in each arm), significantly more changes to treatment needed to be made 42 43 because of side effects in the intensive target arm. 44 45
46 47 48 Implications 49 50 51 Using a systolic blood pressure target of < 130 mmHg or lower for people with prevalent 52 53 cerebrovascular disease in primary care will lead to lower blood pressure than an actively managed < 54 55 140 mmHg target, but the difference in achieved blood pressure is small. Clinically important 56 57 reductions in blood pressure can be achieved with active management to a < 140 mmHg target – 58 59 60 https://mc.manuscriptcentral.com/bmj Page 13 of 35 BMJ
1 2 3 13mmHg equates to more than 40% stroke risk reduction and more than 20% CHD risk reduction.19 4 5 Active management of blood pressure after stroke/TIA therefore appears more important than the 6 7 target that is set. The difficulty in achieving lower targets, the increased workload and extra 8 Confidential: For Review Only 9 medication changes required because of side effects suggest that primary care should focus on 10 11 12 achieving the more conservative <140mmHg target in this population. Given this conclusion, and 13 14 the relatively small difference in blood pressure achieved between arms, we did not feel that a study 15 16 powered to detect a difference in cardiovascular end-points using an intensive target in primary care 17 18 was warranted. 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 https://mc.manuscriptcentral.com/bmj BMJ Page 14 of 35
1 2 3 Panel: What this pape r adds 4 5 What is already know on this subject 6 • Lowering blood pressure after stroke is associated with lower risk of stroke recurrence, but 7 8 Confidential:there is uncertainty over what the targetFor blood preReviewssure should be Only 9 10 • One trial in people with recent lacunar stroke found that a systolic blood pressure target of < 11 130mmHg was associated with a non-significant reduction in stroke compared to a target of 12 13 130-149mmHg 14 • No trials have been carried out in primary care settings of different blood pressure targets 15 16 after stroke 17 18 19 What this study adds 20 21 • Patients set a target of 115 - 130mmHg achieved lower systolic blood pressures than those 22 23 set a target of < 140mmHg, but the difference was small (3mmHg) in the context of the 24 reduction in blood pressure observed in both arms (13mmHg and 16mmHg). 25 26 • Active management of blood pressure after stroke/TIA is more important than the target 27 28 that is set. 29 30 31 32 33 34 35 36 Competing Interest Declaration 37 38 All authors have completed the Unified Competing Interest form at 39 40 www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare 41 42 43 (1) no support from any organisation for the submitted work other than NIHR; (2) JM has received 44 45 funding in the form of provision of pills from Ferrer; RMcM has received funding in the form of 46 47 equipment from Omron and Lloyds Pharmacy (3) that spouses, partners, or children have no 48 49 financial relationships that may be relevant to the submitted work; and (4) no authors have any 50 51 other non-financial interests relevant to the submitted work. 52 53 54 55 56 57 58 59 60 https://mc.manuscriptcentral.com/bmj Page 15 of 35 BMJ
1 2 3 Details of Contributors 4 5 6 JM, RMcM, SG & FDRH had the original idea and gained the funding. KF, JM, RMcM, CT, UM, SV, SG 7 8 Confidential:& FDRH developed the protocol. AR conducted For the primary Review data analysis. KF & SV wereOnly responsible 9 10 for the data collection. JM wrote the first draft of the paper. All authors subsequently refined the 11 12 manuscript and approved the final version. JM is the study guarantor. JM affirms that the 13 14 manuscript is an honest, accurate, and transparent account of the study being reported; and that no 15 16 important aspects of the study have been omitted. There are no discrepancies from our original 17 18 19 plans for this study. 20 21 22 23 24 Ethical Approval 25 26 Ethical approval was provided by the Warwickshire Research Ethics Committee (reference 27 28 29 08/H1211/121). All participants gave informed consent before taking part in the study. 30 31 32 33 34 Funding 35 36 37 This report is independent research funded by the National Institute for Health Research (Stroke 38 39 Prevention in Primary Care, Programme Grant for Applied Research, RP-PG-0606-1153), and by an 40 41 NIHR Professorship (Prof McManus). FDRH is part funded as Director of the National Institute for 42 43 Health Research (NIHR) School for Primary Care Research (SPCR), Theme Leader of the NIHR Oxford 44 45 Biomedical Research Centre (BRC), and Director of the NIHR Collaboration for Leadership in Applied 46 47 Health Research and Care (CLAHRC) Oxford. CJT is funded by NIHR Doctoral Research Fellowship. 48 49 50 The views expressed in this publication are those of the author(s) and not necessarily those of the 51 52 NHS, the NIHR or the Department of Health. The study sponsor was the University of Birmingham. 53 54 The study funder and sponsor had no role in the study design, collection, analysis or interpretation 55 56 57 58 59 60 https://mc.manuscriptcentral.com/bmj BMJ Page 16 of 35
1 2 3 of data, in the writing of the report, or in the decision to submit to publication. The researchers are 4 5 independent of the funders. 6 7 8 Confidential: For Review Only 9 10 Data sharing 11 12 13 No additional data available. 14 15 16 17 18 All authors have full access to all of the data (including statistical reports and tables) in the study and 19 20 can take responsibility for the integrity of the data and the accuracy of the data analysis. 21 22 23 24 25 26 Worldwide licence 27 28 The Corresponding Author has the right to grant on behalf of all authors and does grant on behalf of 29 30 31 all authors, a worldwide licence 32 33 (http://www.bmj.com/sites/default/files/BMJ%20Author%20Licence%20March%202013.doc) to the 34 35 Publishers and its licensees in perpetuity, in all forms, formats and media (whether known now or 36 37 created in the future), to i) publish, reproduce, distribute, display and store the Contribution, ii) 38 39 translate the Contribution into other languages, create adaptations, reprints, include within 40 41 collections and create summaries, extracts and/or, abstracts of the Contribution and convert or 42 43 44 allow conversion into any format including without limitation audio, iii) create any other derivative 45 46 work(s) based in whole or part on the on the Contribution, iv) to exploit all subsidiary rights to 47 48 exploit all subsidiary rights that currently exist or as may exist in the future in the Contribution, v) 49 50 the inclusion of electronic links from the Contribution to third party material where-ever it may be 51 52 located; and, vi) licence any third party to do any or all of the above. 53 54 55 56 57 58 59 60 https://mc.manuscriptcentral.com/bmj Page 17 of 35 BMJ
1 2 3 4 5 6 7 8 Confidential: For Review Only 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 https://mc.manuscriptcentral.com/bmj BMJ Page 18 of 35
1 2 3 References 4 5 6 1 Rothwell PM, Algra A, Amarenco P. Medical treatment in acute and long term secondary prevention after 7 transient ischaemic attack and ischaemic stroke. Lancet 2011; 377:1681-92. 8 Confidential:2 Saka O, McGuire A, Wolfe C. Cost of stroke in theFor United Kingdom. Review Age Ageing 2009; 38:27-32. Only 9 3 Rothwell PM, Coull AJ, Giles MF, Howard SC, Silver LE, Bull LM et al. Change in stroke incidence, mortality, 10 case fatality, severity and risk factors in Oxfordshire, UK from 1981 to 2004 (Oxford Vascular Study). Lancet 11 2004; 363:1925-33. 12 4 Prospective Studies Collaboration. Age-specific relevance of usual blood pressure to vascular mortality: a 13 meta-analysis of individual data for one million adults in 61 prospective studies. Lancet 2002; 360:1903-13. 14 5 PROGRESS Collaborative Group. Randomised trial of a perindopril-based blood pressure lowering regimen 15 among 6105 individuals with previous stroke or transient ischaemic attack. Lancet 2001; 358: 1033-41. 6 16 Wennberg R, Zimmermann C. The PROGRESS trial three years later: time for a balanced report of 17 effectiveness. BMJ 2004; 329:968-971. 7 18 Mant J, McManus RJ, Hare R. Applicability to primary care of national clinical guidelines on blood pressure 19 lowering for people with stroke: cross sectional study. BMJ 2006; 332:635-7. 8 20 Zanchetti A, Grassi G, Mancia G. When should antihypertensive drug treatment be initiated and to what 21 levels should systolic blood pressure be lowered? A critical reappraisal. Journal of Hypertension 2009; 27:923- 22 934. 9 23 Ovbiagele B, Diener H-C, Yusuf S, Martin RH, Cotton D, Vinisko R et al for the PROFESS Investigators. Level of systolic blood pressure within the normal range and risk of recurrent stroke. JAMA 2011; 306:2137-2144. 24 10 25 Arima H, Chalmers J, Woodward M, Anderson C, Rodgers A, Davis S et al for the PROGRESS Collaborative 26 Group. Lower target blood pressures are safe and effective for the prevention of recurrent stroke: the PROGRESS trial. Journal of Hypertension 2006; 24: 1201-1208. 27 11 28 The SPS3 Study Group. Blood pressure targets in patients with recent lacunar stroke: the SPS3 randomised trial. Lancet 2013. Published online May 29 th 2013, http://dx.doi.org/10.1016/S0140-6736(13)60852-1. 29 12 30 The Task Force for the management of arterial hypertension of the European Society of Hypertension (ESH) 31 and of the European Society of Cardiology (ESC). 2013 ESH/ESC Guidelines for the management of arterial hypertension. European Heart Journal 2013; doi:10.1093/eurheartj/eht151. 32 13 th 33 Intercollegiate Stroke Working Party. National Clinical Guidelines for Stroke, 4 Edition. London: Royal 34 College of Physicians 2012. 14 Fletcher K, Mant J, McManus R, Campbell S, Betts J, Taylor C et al. Protocol for PAST BP: a randomised 35 controlled trial of different blood pressure targets for people with a history of stroke or transient ischaemic 36 attack (TIA) in primary care. Cardiovascular Disorders 2010; 10:37 http://www.biomedcentral.com/1471- 37 2261/10/37. 38 15 National Collaborating Centre for Chronic Conditions. Management of hypertension in adults in primary 39 care: partial update. London: Royal College of Physicians 2006. 40 16 Mattu GS, Heran BS,Wright JM. Overall accuracy of the BpTRU—an automated electronic blood pressure 41 device. Blood Press Monit . 2004;9 (1):47-52. 42 17 McManus RJ, Mant J, Haque MS, Bray EP, Bryan S, Greenfield SM, et al. Effect of self-monitoring and 43 medication self-titration on systolic blood pressure in hypertensive patients at high risk of cardiovascular 44 disease: The TASMIN-SR randomized clinical trial JAMA 2014; 312:799-808 45 18 Kerry SM, Markus HS, Khong TK, Cloud GC, Tulloch J, Coster D et al. Home blood pressure monitoring with 46 nurse led telephone support among patients with hypertension and a history of stroke: a community based 47 randomised controlled trial. CMAJ 2012. DOI:10.1503/cmah.120832. 48 19 Law MR, Morris JK, Wald NJ. Use of blood pressure lowering drugs in the prevention of cardiovascular 49 disease: meta-analysis of 147 randomised trials in the context of expectations from prospective 50 epidemiological studies. BMJ 2009; 338: b1665 51 52 53 54 55 56 57 58 59 60 https://mc.manuscriptcentral.com/bmj Page 19 of 35 BMJ
1 2 3 4 5 LEGEND FOR FIGURE 1 6 7 (Figure provided in separate file) 8 Confidential: For Review Only 9 Figure 1: Trial profile 10 11 ‡ Reasons given: patient was housebound or in a nursing home (957, 33%); would be unable to 12 provide consent (338, 12%); co-morbidity (216, 7%); blood pressure too low (199, 7%); at risk of 13 14 falling (164, 6%); insufficient evidence of stroke/TIA (98, 3%); already being treated to 130 mmHg 15 target (71, 2%); other patient factors (69, 2%); patient choice (54, 2%); terminally ill (48, 2%); 16 deceased or left practice (41, 1%); participating in another trial (9). In 618 (21%) cases, no reason 17 was given. 18 19 †blood pressure < 125mmHg 447; lack of corrobora ve evidence of stroke/TIA 60; on 3 or more 20 21 antihypertensives 51; orthostatic hypotension 22; already being treated to lower BP target 4; unable 22 to provide informed consent 2. 23 24 SBP: Systolic Blood Pressure 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 https://mc.manuscriptcentral.com/bmj BMJ Page 20 of 35
1 2 3 All participants Participants with systolic blood pressure 4 recorded at 12 months 5 6 Intensive target Standard target Intensive target Standard target 7 8 Confidential: n=266 For Review n=263 n=182Only n=197 9 10 Age (years) 71.9 (9.1) 71.7 (9.4) 72.6 (8.3) 71.9 (9.5) 11 12 Men 157 (59.0) 156 (59.3) 104 (57.2) 125 (63.5) 13 14 White ethnicity 260 (97.7) 259 (98.5) 180 (98.8) 194 (98.5) 15 16 Current smoker 25 (9.4) 33 (12.6) 15 (8.3) 27 (13.9) 17 18 Systolic blood pressure 142.9 (14.0) 142.2 (13.4) 143.5 (13.5) 142.2 (12.9) 19 20 <140mmHg 128 (48.1) 129 (49.1) 79 (43.4) 98 (49.8) 21 22 >=140mmHg 138 (51.9) 134 (50.9) 103 (56.6) 99 (50.3) 23 Diastolic blood pressure 79.9 (10.0) 80.4 (9.8) 78.8 (9.3) 80.7 (10.1) 24 25 Diabetes mellitus 26 (9.8) 25 (9.5) 19 (10.4) 21 (10.7) 26 27 Atrial Fibrillation 28 (10.5) 27 (10.3) 21 (11.5) 22 (11.2) 28 29 Coronary heart disease 41 (15.4) 46 (17.5) 28 (15.4) 35 (17.8) 30 31 Chronic kidney disease 26 (9.8) 30 (11.4) 19 (10.4) 23 (11.7) 32 33 Heart failure 2 (0.8) 7 (2.7) 1 (0.6) 6 (3.1) 34 35 Peripheral vascular disease 11 (4.1) 11 (4.2) 7 (3.9) 6 (3.1) 36 37 Stroke 130 (48.9) 122 (46.4) 85 (46.7) 95 (48.2) 38 39 TIA only 135 (50.8) 141 (53.6) 97 (53.3) 102 (51.8) 40 41 Number of antihypertensive drugs 1.0 (0.8) 1.1 (0.8) 1.1 (0.8) 1.1 (0.8) 42 43 Number of other drugs 4.5 (2.5) 4.6 (2.6) 4.5 (2.5) 4.6 (2.6) 44 45 Total number of drugs 5.6 (2.8) 5.7 (2.7) 5.6 (2.7) 5.7 (2.7) 46 47 Modified Rankin scale† 48 49 0 or 1 135 (50.8) 125 (47.5) 98 (53.8) 84 (42.6) 50 51 2 65 (24.4) 69 (26.2) 42 (23.1) 57 (28.9) 52 53 3 or 4 47 (17.7) 51 (19.4) 29 (15.9) 42 (21.3) 54 55 Table 1: Baseline characteristics 56 57 Data are mean (SD) or number (%); †Data missing for 19 pa ents in intensive arm and 18 in standard arm (all 58 participants) and for 13 patients in intensive arm and 14 in standard arm (participants with 12 month systolic 59 blood pressure). 60 https://mc.manuscriptcentral.com/bmj Page 21 of 35 BMJ
1 2 3 4 5 Intensive target Standard target 6 (n=109) (n=57) 7 8 Confidential:Other blood pressure readings (e.g. home readings) For taken into Review account 17 Only 20 9 10 Patient did not want treatment intensified 22 13 11 12 Decision taken to re-measure blood pressure at future time 19 12 13 14 Symptoms attributed to blood pressure medication 24 5 15 16 Blood pressure only just above target 14 2 17 18 Patient had not been taking pills 9 5 19 20 Blood pressure reading attributed to patient anxiety 3 8 21 22 Changes to drug therapy already made 4 2 23 24 Postural hypotension 3 2 25 26 Awaiting specialist advice/ test results 5 - 27 28 Intercurrent illness 3 - 29 30 Patient too old for further increases in therapy 1 2 31 32 Change in lifestyle advocated rather than change in medication - 1 33 34 35 Table 2: Reasons given by general practitioner for not increasing blood pressure medication after 36 patient referred by practice nurse with blood pressure above target 37 38 39 A reason was given for 164 of 166 non-intensification decisions. Numbers add up to more than 164 as in some 40 cases two reasons were given. 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 https://mc.manuscriptcentral.com/bmj BMJ Page 22 of 35
1 2 3 4 5 Confidential: Mean blood pressure (mm Hg)For Mean difference Review from baseline Effect size (mm Hg,Only 95% CI)† 6 (mm Hg) 7 Baseline 6 months 12 months 6 months 12 months 6 months 12 months 8 9 Systolic blood pressure 10 11 Intensive target‡ 143.5 (13.5) 125.7 (14.5) 127.4 (14.8) -17.3 (16.7) -16.1 (15.0) -4.12 (-6.84 to -1.40) -2.94 (-5.68 to -0.21) 12 13 Standard target* 142.2 (12.9) 129.3 (14.6) 129.4 (14.8) -12.7 (16.7) -12.8 (17.2) .. .. 14 15 Diastolic blood pressure 16 17 Intensive target‡ 78.8 (9.3) 73.1 (10.3) 72.0 (9.0) -6.5 (10.7) -6.8 (9.1) -1.14 (-2.86 to 0.58) -1.63 (-3.10 to -0.15) 18 19 Standard target* 80.7 (10.1) 74.6 (9.8) 74.4 (8.9) -6.1 (9.7) -6.3 (9.4) .. .. 20 21 22 23 Table 3: Systolic and diastolic blood pressure in intensive target and standard target groups 24 25 Data are mean (standard deviation) 26 27 †Adjusted for baseline blood pressure, age group (<80, ≥80), gender, diabetes mellitus, atrial fibrillation and general practice (random effect) 28 ‡Blood pressure data for 193 intensive target pa ents at six months and 182 at twelve months 29 *Blood pressure data for 198 standard target patients at six months and 197 at twelve months 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 https://mc.manuscriptcentral.com/bmj 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 23 of 35 BMJ
1 2 3 4 5 6 7 8 Confidential: Intensive target For Standard targetReview P value Only 9 arm arm 10 11 12 Pain 93/163 (57%) 89/173 (51%) 0.48 13 14 Breathlessness 53/148 (36%) 49/158 (31%) 0.53 15 16 Fatigue 75/149 (50%) 88/163 (54%) 0.36 17 Stiff joints 93/162 (57%) 99/176 (56%) 0.80 18 19 Sore eyes 35/148 (24%) 24/158 (15%) 0.08 20 Wheeziness 32/163 (20%) 28/175 (16%) 0.46 21 22 Headaches 27/151 (18%) 36/165 (22%) 0.22 23 24 Sleep difficulties 56/150 (37%) 66/163 (40%) 0.39 25 Dizziness 45/164 (27%) 39/173 (23%) 0.42 26 27 Loss of strength 44/148 (30%) 51/162 (31%) 0.52 28 Loss of libido 47/160 (29%) 50/171 (29%) 0.83 29 30 Impotence 29/129 (22%) 31/145 (21%) 0.54 31 32 Pins and needles 54/163 (33%) 44/176 (25%) 0.11 33 Cough 40/144 (28%) 49/160 (31%) 0.57 34 35 Swelling of 51/162 (31%) 49/177 (28%) 0.70 36 legs/ankles 37 38 Dry mouth 34/147 (23%) 36/161 (22%) 0.95 39 40 41 Table 4: Most frequent symptoms at 12 months 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 https://mc.manuscriptcentral.com/bmj BMJ Page 24 of 35
1 2 3 4 5 6 7 8 Confidential: For Review Only 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 Figure 2 Effect of intensive versus standard target on systolic BP at twelve months for different 37 patient sub-groups 38 39 Adjusted for baseline blood pressure, age group (<80, ≥80), gender, diabetes mellitus, atrial fibrillation and 40 general practice (random effect) 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 https://mc.manuscriptcentral.com/bmj Page 25 of 35 BMJ
1 2 3 Total patients on stroke register 4 14779 5 Excluded prior to study clinic – 5377 6 2495: onConfidential: 3+ antihypertensives For Review Only 7 2882: excluded by general practitioner‡ Invited to participate 8 9 9402 10 11 Declined/ did not 12 respond 8235 13 Total attending study clinic 14 15 1167 Excluded at study clinic - 638 16 586: exclusion criteria met† 17 42: patient choice 18 3: co-morbidity 19 1: error Randomised 20 6: no reason given 21 529 22 23 24 25 26 assigned to intensive target assigned to standard target 27 28 266 263 2 died 29 1 died 1 non-fatal major cardiovascular event 30 4 non-fatal major cardiovascular event 1 emigrated 31 32 withdrew consent 52 withdrew consent 32 29 not followed up for SBP at one year 33 28 not followed up for SBP at one year Number analysed Number analysed 34 Primary Outcome: 182 Primary Outcome: 197 35 Clinical Outcome: 266 Clinical Outcome: 263 36 37 38 39 40 https://mc.manuscriptcentral.com/bmj 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Page 26 of 35
1 2 3 4 5 Web Appendix:Confidential: For Review Only 6 7 Mean blood pressure (mm Hg) Mean difference from Effect size (mm Hg, 95% CI)† 8 baseline (mm Hg) 9 Baseline 12 months 10 Systolic blood pressure 11 12 Average of 5 th and 6 th Intensive target 137.3 (13.5) 122.7 (14.9) -14.7 (16.0) -3.24 (-5.8 to -0.64) 13 readings 14 Standard target 135.5 (13.3) 124.9 (13.2) -11.0 (16.6) .. 15 16 17 Average of 2 nd to 6 th Intensive target 140.0 (13.0) 125.0 (14.7) -15.1 (14.9) -3.25 (-5.8 to -0.67) 18 readings 19 Standard target 138.4 (12.7) 127.2 (13.6) -11.6 (16.2) .. 20 21 †Adjusted for baseline blood pressure, age group (<80, ≥80), gender, diabetes mellitus, atrial fibrillation and general practice (random effect) 22 23 Table i: Systolic blood pressure in intensive target and standard target groups 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 https://mc.manuscriptcentral.com/bmj 47 48 49 50 51 52 53 54 55 56 57 58 59 60 Page 27 of 35 BMJ
1 2 3 4 5 Confidential: Mean blood pressure (mmFor Hg) Mean Review difference from Effect size (mm Hg,Only 95% CI)† 6 baseline (mm Hg) 7 Baseline 12 months 8 Imputation method Systolic blood pressure 9 10 Multiple imputation Intensive target 142.9 (13.9) 126.8 (14.9) -16.1 (15.4) -3.19 (-5.72 to -0.65) 11 12 Standard target 142.2 (13.3) 129.6 (14.7) -12.6 (17.4) .. 13 14 15 Group mean Intensive target 142.9 (14.0) 127.4 (12.2) -15.5 (15.0) -2.33 (-4.33 to -0.32) 16 17 Standard target 142.2 (13.4) 129.4 (12.8) -12.8 (16.6) .. 18 19 20 Last value carried Intensive target 142.9 (14.0) 130.0 (15.4) -12.9 (15.5) -1.79 (-4.15 to 0.57) 21 forward 22 Standard target 142.2 (13.4) 131.3 (15.6) -10.8 (16.8) 23 24 25 †Adjusted for baseline blood pressure, age group (<80, ≥80), gender, diabetes mellitus, atrial fibrillation and general practice (random effect) 26 Table ii: Systolic blood pressure in intensive target and standard target groups with imputation of missing data 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 https://mc.manuscriptcentral.com/bmj 47 48 49 50 51 52 53 54 55 56 57 58 59 60 BMJ Page 28 of 35
Fletcher et al. BMC Cardiovascular Disorders 2010, 10:37 1 http://www.biomedcentral.com/1471-2261/10/37 2 3 4 5 STUDY PROTOCOL Open Access 6 7 8 Confidential: For Review Only 9 Protocol for Past BP: a randomised controlled trial 10 11 of different blood pressure targets for people 12 13 with a history of stroke of transient ischaemic 14 15 attack (TIA) in primary care 16 17 Kate Fletcher1*, Jonathan Mant2, Richard McManus1, Sarah Campbell1, Jonathan Betts1, Clare Taylor1, 18 Satnam Virdee1, Sue Jowett1, Una Martin1, Sheila Greenfield1, Gary Ford3, Nick Freemantle1, FD Richard Hobbs1 19 20 21 22 Abstract 23 Background: Blood pressure (BP) lowering in people who have had a stroke or transient ischaemic attack (TIA) 24 leads to reduced risk of further stroke. However, it is not clear what the target BP should be, since intensification of 25 therapy may lead to additional adverse effects. PAST BP will determine whether more intensive BP targets can be 26 achieved in a primary care setting, and whether more intensive therapy is associated with adverse effects on 27 quality of life. 28 Methods/Design: This is a randomised controlled trial (RCT) in patients with a past history of stroke or TIA. 29 Patients will be randomised to two groups and will either have their blood pressure (BP) lowered intensively to a 30 target of 130 mmHg systolic, (or by 10 mmHg if the baseline systolic pressure is between 125 and 140 mmHg) 31 32 compared to a standard group where the BP will be reduced to a target of 140 mmHg systolic. Patients will be 33 managed by their practice at 1-3 month intervals depending on level of BP and followed-up by the research team 34 at six monthly intervals for 12 months. 35 610 patients will be recruited from approximately 50 general practices. The following exclusion criteria will be 36 applied: systolic BP <125 mmHg at baseline, 3 or more anti-hypertensive agents, orthostatic hypotension, diabetes 37 mellitus with microalbuminuria or other condition requiring a lower treatment target or terminal illness. 38 The primary outcome will be change in systolic BP over twelve months. Secondary outcomes include quality of 39 life, adverse events and cardiovascular events. 40 In-depth interviews with 30 patients and 20 health care practitioners will be undertaken to investigate patient and 41 healthcare professionals understanding and views of BP management. 42 Discussion: The results of this trial will inform whether intensive blood pressure targets can be achieved in people 43 who have had a stroke or TIA in primary care, and help determine whether or not further research is required 44 before recommending such targets for this population. 45 Trial Registration: ISRCTN29062286 46 47 48 49 50 51 52 53 54 * Correspondence: [email protected] 55 1Primary Care Clinical Sciences, Clinical Sciences Building University of Birmingham, Edgbaston Birmingham UK, B15 2TT 56 Full list of author information is available at the end of the article 57 58 © 2010 Fletcher et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in 59 any medium, provided the original work is properly cited. 60 https://mc.manuscriptcentral.com/bmj Page 29 of 35 BMJ
Fletcher et al. BMC Cardiovascular Disorders 2010, 10:37 Page 2 of 8 1 http://www.biomedcentral.com/1471-2261/10/37 2 3 4 5 Background Long term management of blood pressure following 6 Stroke is the third largest cause of death in England, and stroke and TIA is predominantly carried out in primary 7 the single largest cause of adult disability[1]. A National care. Recent studies of blood pressure control in this 8 AuditConfidential: Office report (2005) highlighted the high costFor of setting Review paint a mixed picture Only of implementation of 9 stroke to the NHS: about £2.8 billion per year in direct guidelines. In a study of seven general practices in 10 care costs, and an additional £1.8 billion per year cost to South Birmingham in 2002, 63% of patients with a pre- 11 the wider economy due to lost productivity and disabil- vious stroke or TIA had BP above the 140 mmHg tar- 12 ity [2]. Recent estimates suggest that between 30-45% of get, and 80% above the 130 mmHg target [11]. 68% of 13 strokes are recurrent events, [3] so more effective sec- these patients were prescribed BP lowering therapy. An 14 ondary prevention could result in significant savings: the analysis of general practice data on the QRESEARCH 15 National Audit Office estimates that preventing just 2% database for 2002-2004 found that of all patients with 16 of strokes in England in a year could save care costs of incident stroke, blood pressure was not recorded in 25% 17 over £37 million [2]. NICE has identified better control of patients, and where it was recorded, it was above the 18 of hypertension as one of the interventions that are cost 140 mmHg target in 47% [12]. An analysis of the impact 19 saving for the NHS [4]. of the Quality Outcomes Framework (QOF) carried out 20 The PROGRESS trial demonstrated that blood pres- for the National Audit Office found that the proportion 21 sure lowering is beneficial in reducing risk of stroke of people with a history of stroke or TIA who had their 22 amongst both hypertensive and non-hypertensive indivi- BP measured in the preceding 15 months rose from 23 duals with a history of stroke or TIA recruited in sec- 89% to 95% between 2004 and 2005, and the proportion 24 ondary care immediately after their cerebrovascular with a BP below 150 mmHg (the target level for the 25 event. In this trial, patients were randomised to either a QOF) rose from 69% to 80% suggesting some improve- 26 combination of an ACE inhibitor and thiazide diuretic ment [13]. An analysis of the care of over three thou- 27 against double placebo, or an ACE inhibitor alone sand patients who had a TIA during 2004-5 found that 28 against single placebo. The decision to randomise to 60% had a BP equal to or below the 140 mmHg target, 29 one or two agents was made by the supervising physi- though only 50% were on any blood pressure lowering 30 cian on the basis of whether or not they thought it was therapy [14]. 31 safe to randomise an individual patient to two agents. In summary, although there is some evidence that 32 Mean blood pressure in the intervention arm was blood pressure lowering in people who have had a 33 reduced from 147 mmHg systolic by 9 mmHg (SE 0.3), stroke or TIA is beneficial, there is no clear guidance on 34 and this was associated with a 28% reduction in stroke what the target BP should be. Furthermore, data col- 35 risk [5]. The positive result of the PROGRESS trial lected from primary care suggest that guidelines from 36 raises a supplementary question: by how much should the British Hypertension Society and Intercollegiate 37 blood pressure be lowered? No randomised trials have Stroke Working Party are not being fully implemented. 38 specifically compared different target blood pressures This research is designed to support implementation of 39 (BP) in the post-stroke/TIA population. Observational the guidelines by both addressing the gaps in the under- 40 lying evidence base, and testing a specific mechanism 41 data (although not collected specifically in people with a for implementation of blood pressure lowering. 42 history of stroke or TIA) suggest that the lower the 43 blood pressure, the lower the risk of vascular mortality, Methods/Design 44 at least down to 115 mmHg systolic [6,7]. There is 45 some evidence from PROGRESS to support this, in that Study aims 46 the sub-group of patients whose baseline BP was The primary aim of Past BP is to determine whether a 47 between 120 and 140 mmHg who were randomised to more intensive target BP for people with stroke or TIA in 48 combination therapy had a significantly reduced risk of a pragmatic primary care setting will lead to a lower BP. 49 stroke compared with control, though this benefit was Secondary aims of the research are to: 50 not observed in patients who were randomised to a sin- • 51 gle agent [8]. Guidelines have tended to interpret this determine the impact of a more intensive BP target 52 evidence by recommending a target of 130 mmHg for on patient quality of life; 53 systolic blood pressure in people with cerebrovascular • identify the barriers to implementation of more 54 disease [9,10]. However, the question remains whether intensive blood pressure lowering; 55 such a target is prudent in general (42% were rando- • to explore whether the potential benefits associated 56 mised to a single agent and gained no benefit) and with intensive blood pressure lowering might be out- 57 whether it is achievable in primary care (PROGRESS weighed by potential adverse effects on quality of life 58 was secondary care based) [5]. and costs. 59 60 https://mc.manuscriptcentral.com/bmj BMJ Page 30 of 35
Fletcher et al. BMC Cardiovascular Disorders 2010, 10:37 Page 3 of 8 1 http://www.biomedcentral.com/1471-2261/10/37 2 3 4 5 The latter point will be explored by economic model- clinical computer system to identify all patients on the 6 ling. If there is uncertainty as to whether there is net stroke/TIA register. Where possible, the computer 7 benefit from intensive blood pressure lowering, then search will exclude patients with clear exclusion criteria 8 therewouldbeacaseforatrialofdifferenttargetsforConfidential: For(see Review table 1). The GP will alsoOnly remove patients for 9 BP lowering in primary care that is powered to detect whom a study invitation would be inappropriate (for 10 differences in clinical end-points. example, those with a terminal illness). Patients with no 11 The study will also investigate patients’ understanding clear exclusion criteria at this stage will be sent a letter 12 and beliefs about the relationship between blood pres- inviting them to attend a study baseline clinic 13 sure and stroke, and patients and healthcare profes- appointment. 14 sionals experience of participating in the study, which Baseline clinic appointment 15 may contribute to the success or otherwise of the This clinic appointment is carried out by a Research 16 intervention. Nurse. At this appointment the nurse will: confirm the 17 stroke/TIA diagnosis through review of medical records 18 Study design and setting and patient interview; determine whether there are any 19 Past BP is a primary care based pragmatic randomised exclusion criteria present; and collect baseline data (see 20 controlled trial (RCT) in which people with stroke or table 2). If the patient is eligible and willing to take part, 21 TIA are randomised to an intensive blood pressure (BP) the nurse will also gain written informed consent prior 22 target group (target 130 mmHg systolic, or 10 mmHg to randomisation, and will telephone the randomisation 23 reduction in systolic BP if baseline BP 125 - 140 service to obtain treatment group allocation. 24 mmHg) or a standard BP target group (target 140 Once the treatment allocation has been obtained, 25 mmHg systolic). patients in the intensive target group and any patients 26 We will also use qualitative methodologies to investi- who have been randomised to the standard treatment 27 gate patient and healthcare professionals understanding group whose BP is above the target of 140 mmHg will 28 and views of BP management. Grounded theory meth- see a GP in order to have their treatment intensified 29 ods will guide sampling, data collection and data analy- using the study specific treatment protocol (see figure 1). 30 sis [15,16]. Sampling will be done purposively to allow Patients in the standard group whose BP is below target 31 for the maximum variety of characteristics. Semi-struc- will receive an appointment to see the practice nurse 32 tured interviews will be carried out [17] and will con- three months post randomisation. 33 tinue until new concepts are no longer being generated Randomisation 34 and theoretical saturation is reached. The randomisation will use minimisation to balance the 35 Ethical Considerations randomised groups on the basis of age (<80, ≥80), sex, 36 Full ethical approval for this study has been obtained diabetes mellitus, atrial fibrillation (because of the diffi- 37 from Warwickshire Research Ethics Committee, refer- culties of obtaining accurate BP measurements in this 38 ence 08/H1211/121. A Data Monitoring Committee and group), baseline systolic BP and practice. 39 a Trial Steering Committee will monitor the progress of Patient follow-up procedures 40 the RCT. Patients will be followed up from trial entry for 12 41 months. Follow up will be carried out in several ways. 42 Randomised controlled trial Firstly, practice nurses (PNs) will see patients at 1-3 43 Study Interventions monthly intervals, depending on BP and treatment allo- 44 ’ 45 Management of both treatment groups will follow study cation. (see figure 2) The PN will take a patient sBPand 46 specific treatment protocols that reflect the current refer them to the GP if the BP is above target, or if the 47 NICE guidelines [18]. However, the thresholds for inter- patient is having problems with adverse effects to their 48 vention between the two treatment groups are different: BP medication. The GP will then adjust the medication, 49 all patients in the intensive target arm will have their BP following the study treatment protocol (see figures 1). At 50 lowering therapy intensified at trial entry since the tar- 6 and 12 months patients will be followed up by a 51 get will be automatically below their baseline BP research nurse (RN) where details of primary and sec- 52 whereas only those patients in the standard arm whose ondary outcomes will be collected. (see table 2). Finally, 53 BP is above 140/90 mmHg will have their therapy inten- the records of patients will be flagged at the NHS Central 54 sified at the outset. Register. 55 Identification of eligible patients Patient outcome measures 56 Eligible patients will be identified from general practices Theprimaryoutcomemeasureisachangeinsystolic 57 from the Central England Primary Care Research Net- blood pressure between baseline and 12 months. Blood 58 work and from the Midlands Research Practice Consor- pressure measurements are performed in a standardised 59 tium (MidReC). Each practice will run a search of their way, using BHS validated automated electronic monitors 60 https://mc.manuscriptcentral.com/bmj Page 31 of 35 BMJ
Fletcher et al. BMC Cardiovascular Disorders 2010, 10:37 Page 4 of 8 1 http://www.biomedcentral.com/1471-2261/10/37 2 3 4 5 Table 1 Study inclusion and exclusion criteria 6 Inclusion criteria 7 On practice TIA/stroke register 8 ExclusionConfidential: criteria For Review Only 9 Systolic BP < 125 mmHg at baseline; 10 Already taking 3 or more anti-hypertensive agents; orthostatic hypotension (>20 mmHg postural change in systolic BP after 1 minute standing) 11 Patient already has a treatment target of 130 mmHg systolic BP specified 12 Unable to provide informed consent. 13 Insufficient corroborative evidence of stroke/TIA from medical record and patient interview 14 15 Table 2 Timing and content of study assessments 16 17 Baseline data: research nurse administered 18 Socio-demographic characteristics: 19 Age; Ethnicity; Gender; Postcode 20 Validation of stroke/transient ischaemic attack: 21 Review of medical records with patient history 22 Clinical measures: 23 Six blood pressure (BP) measurements, calculating mean of 2nd and 3rd measurements and recording details of: arm used; arm circumference; 24 BP cuff size; and time BP measurement started 25 24 hour ambulatory BP recording 26 Medical history 27 Previous history of angina, myocardial infarction, heart failure, atrial fibrillation, CABG/angioplasty (balloon)/or stent, peripheral vascular disease, 28 diabetes, chronic kidney disease. Current prescription medications. Smoking status and alcohol intake 29 Patient questionnaires - self-completion 30 Health related quality of life assessed by the SF-36 [20] and EQ-5 D [19] 31 Disability assessed by the Modified Rankin Scale [26] 32 Medication Adherence Report Schedule (MARS) for BP treatment [27] 33 Symptoms/side effects questionnaire 34 Patient questionnaire - research nurse completion 35 Cognitive function assessed by the Mini Mental State [21] 36 Eligibility and consent 37 Review inclusion and exclusion criteria and record outcome of consent process 38 Patient follow up for BP control 39 40 GP appointment at any time patient BP medication review required 41 Action taken to treat/monitor side effects 42 Action taken to treat BP above target using study algorithm (see figure 1) 43 Make appointment with practice nurse at appropriate interval (see figure 2) 44 Practice nurse follow up - 1-3 month intervals 45 Six BP readings, as per baseline data collection 46 Side effects of BP medication 47 Refer patient to GP or make further appointment with practice nurse (see figure 2) 48 Patient follow up by research team 49 Research nurse follow up - 6 and 12 months post randomisation 50 Six BP readings, as per baseline data collection 51 24 hour ambulatory BP recording (12 month f/u only) 52 Check details of patient visits to GP and practice nurse 53 Diagnosis of key medical conditions (as per baseline data collection) since baseline or the previous research nurse follow up 54 All hospital admissions or outpatient visits since baseline or the previous research nurse follow up 55 Record of medications introduced since baseline or the previous research nurse follow up 56 Monitor compliance with repeat medication since baseline or previous research nurse follow up 57 Completion of patient questionnaires, as per baseline data collection. 58 Obtaining information on patients who died 59 Records flagged at NHS central register 60 https://mc.manuscriptcentral.com/bmj BMJ Page 32 of 35
Fletcher et al. BMC Cardiovascular Disorders 2010, 10:37 Page 5 of 8 1 http://www.biomedcentral.com/1471-2261/10/37 2 3 4 5 6 7 8 Confidential: For Review Only 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 Figure 1 Summary of Algorithm for BP control. 34 35 36 supplied and validated for the study. The patient will be measures of BP (change in diastolic and mean daytime 37 seated for 5 minutes and then 6 measurements will be ambulatory systolic BP between baseline and twelve 38 taken at minute intervals. The second and third mea- months); measures of adherence (including GP adher- 39 surements are averaged to give the reading. As the ence to protocol and patient adherence to prescribed 40 intensive target group have their BP monitored more medication); quality of life (EQ5 D [19]; SF36 [20]); side 41 frequently than those in the standard group, there may effects, tolerability and adverse events; clinical outcomes 42 ‘ ’ 43 be some diminution of the white coat effect in this (including major cardiovascular events [composite of 44 group; the mean of readings 2 to 6 and the mean of 5 fatal and non-fatal stroke, myocardial infarction or fatal 45 and 6 will be used to monitor for this. Any differences coronary heart disease and other cardiovascular death], between the groups in the primary outcome should be all cause mortality, cognitive function [21] and hospital 46 th th 47 sustained in the mean of the 5 and 6 readings (by admissions). Key secondary events (stroke; myocardial 48 which time any accommodation effect is likely to have infarction; fatal coronary heart disease and other cardio- 49 worn off), enabling us to determine whether accommo- vascular death) will be reviewed by independent clini- 50 dation has a significant effect in this study. 24 hour cians blinded to treatment to ensure unbiased coding of 51 blood pressure recordings using an ambulatory sphyg- these events. 52 momanometer will be recorded at baseline and at twelve Sample size 53 months. 24 hour ambulatory measurement will be unaf- Randomisation of 610 patients (305 per arm), with 12 54 fected by accommodation, so will provide further evi- months of follow-up, will detect a 5 mmHg difference in 55 dence as to whether or not this was significant in this systolic BP between groups with 90% power and at a 56 study. significance level of 5% assuming a standard deviation of 57 A variety of secondary outcome measures are also 17.5 mmHg (a conservative estimate of standard devia- 58 assessed during the RN follow up appointments. (see tion falling between 16, a figure derived from the same 59 table 2) Key secondary outcomes are: additional sort of patients as included in this study [11] and 19, 60 https://mc.manuscriptcentral.com/bmj Page 33 of 35 BMJ
Fletcher et al. BMC Cardiovascular Disorders 2010, 10:37 Page 6 of 8 1 http://www.biomedcentral.com/1471-2261/10/37 2 3 4 5 patients. This will generate 5,625 patients on practice 6 TIA/stroke registers (from the QOF data, the overall 7 prevalence of TIA/stroke in primary care is 1.5%). From 8 Confidential: Forour Review analysis of South Birmingham Only data [11], we antici- 9 pate that 13% of these patients will be ineligible because 10 they are already on three or more anti-hypertensives, 11 and 28% because they will not fulfil the diagnostic cri- 12 teria for a history of stroke or TIA for the study [24]. 13 We assume that 30% of patients will respond to the 14 invitation to attend a study clinic, that 24% of them will 15 be ineligible due to a systolic BP below 125, and a 16 further 15% will decline to take part after discussion 17 with the research nurse. This equates to the recruitment 18 of 12 patients per practice with an average list size of 19 7,500. 20 21 Statistical analysis 22 The principal analyses will use generalised linear mod- 23 els, accounting for baseline BP as a patient level covari- 24 ate, and practices as random effects and compare 25 differences in systolic BP (primary outcome), and differ- 26 ences in diastolic BP, quality of life, adherence and fre- 27 quency of adverse effects (secondary outcomes). We will 28 look at effect on systolic BP lowering in pre-specified 29 sub-groups: diabetes; atrial fibrillation; and age group. 30 Clinical event rates will be monitored by treatment allo- 31 cation by the Data Monitoring & Ethics Committee, but 32 only aggregated rates will be made available to the 33 investigators. 34 35 Economic evaluation 36 Decision analytical modelling will be undertaken to 37 synthesise data from the trial and the literature in order 38 Figure 2 follow up procedure. to determine whether potential benefits of intensive 39 blood pressure lowering (by lowering the risk of stroke) 40 are outweighed by potential adverse effects on quality of 41 the standard deviation in the PROGRESS Study [5]). life. Ultimately the model analyses will inform whether a 42 further trial, powered to detect differences in clinical 43 The calculation assumes that: a 5 mmHg difference in end-points, is required. 44 systolic BP is of clinical significance, leading to a 20% 45 reduction in major vascular events;[22] that there will A Markov model will be constructed to consider 46 be 5% mortality at six months, and a further 10% of intensive target and standard target strategies for blood 47 patients will not have their BP measured at six months. pressure lowering in patients with a history of stroke or 48 With regard to ambulatory blood pressure measure- TIA. The clinical events of importance in the model are 49 ment, one of the secondary outcomes of the study, ran- further stroke events, myocardial infarction (MI) and 50 domisation of 450 patients (225 per arm) will detect a other cardiovascular related mortality. Data from the 51 4 mmHg difference in systolic BP between groups with trial and literature will inform the probability of these 52 90% power and at a significance level of 5% assuming a events occurring and the risk reduction afforded by the 53 standard deviation of 11.7 mmHg [23]. This calculation alternative strategies. Attached to each health state will 54 assumes that 80% of patients will have ambulatory be associated health state utility values (quality of life) 55 blood pressure measured at 12 months. in order that quality-adjusted life years (QALYs) can be 56 To recruit patients from primary care, an estimate of calculated. Quality of life on each treatment strategy will 57 the number of practices is required. Approximately 50 be obtained from the trial data on EQ-5 D, and previous 58 practices with an average list size of 7,500 will be studies will inform post-stroke and post-MI values. In 59 needed in order to recruit the required number of addition, in order that cost-effectiveness analyses can be 60 https://mc.manuscriptcentral.com/bmj BMJ Page 34 of 35
Fletcher et al. BMC Cardiovascular Disorders 2010, 10:37 Page 7 of 8 1 http://www.biomedcentral.com/1471-2261/10/37 2 3 4 5 undertaken, the model will be populated with costs of minutes, and will be audio taped and transcribed 6 the therapies prescribed in each strategy and acute and verbatim. 7 long term costs of further cardiovascular events. Data Analysis 8 InConfidential: order to explore uncertainties in the analyses, ForData Review collection and analysis wiOnlyll be iterative, occurring 9 deterministic sensitivity analysis is proposed to test the as data collection in the interviews proceeds. Data will 10 robustness of the model when varying key model para- be analysed using a thematic approach, based on the 11 meters and structural assumptions. Probabilistic sensitiv- principles of ‘Framework’ analysis [25] and using Frame- 12 ity analysis will be undertaken to incorporate the work software. The research team will actively contri- 13 uncertainty around parameter values and quantify the bute to the development of the analysis and conceptual 14 overall decision uncertainty, and inform whether further framework and their different disciplinary and profes- 15 research is required. sional backgrounds will maximise theoretical sensitivity 16 [16]. 17 Qualitative Study Time plan 18 Sampling Patient recruitment began in July 2009 and is planned to 19 A purposively selected sample of 30 patients (10 each continue until February 2011. By October 2009, 23 20 from intervention and control and 10 patients who patients (4% of target) have been recruited into the trial. 21 declined the invitation to participate), and 20 healthcare Interviews will commence in January 2010 and are 22 professionals (Health Care Assistants (HCAs), nurses expected to be completed by February 2011. 23 and GPs) will be selected for interview. 24 Patient Sampling Strategy Sampling will be carried out Discussion 25 on the basis of study arm (intervention or control), with The results of this trial and the health economic analysis 26 a further group of people who did not consent to parti- will provide insight into the role of intensive blood pres- 27 cipate in the trial also being invited to attend for an sure targets for people who have had a stroke or TIA. If 28 interview. Within each group, participants will be the trial is negative and a significant difference in systo- 29 selected on the basis of: age (tertiles); socio-economic lic blood pressure is not observed between the two 30 status (using IMD scores); number of different classes of study arms, then the embedded qualitative work will be 31 medications; and whether they have had a stroke or a of importance to determine why low blood pressure tar- 32 TIA. A researcher will randomly select patients from gets did not lead to lower blood pressure. If the trial is 33 these categories, ensuring that similar numbers of positive, then the critical question remains as to whether 34 patients in all categories are included. striving for lower blood pressure targets is appropriate. 35 Health care professionals sampling strategy Practices If we observe no difference in adverse event rates or 36 participating in the study were selected to ensure a quality of life between the two arms of the trial, then it 37 range of practice characteristics are represented, includ- is likely that aiming for lower blood pressure targets will 38 ing practice size and socio-economic status. A be worthwhile, given the benefits of reduced stroke risk 39 researcher will randomly select 20 practitioners from that were observed in the PROGRESS trial [5]. This will 40 these practices and send an invitation letter and infor- be tested by our economic analysis. If, on the other 41 mation sheet, inviting them to take part in an interview. hand, the lower blood pressures are at the cost of higher 42 Patients and practitioners who fail to respond to invi- adverseeventrates,thenitmaybethatafurthertrial 43 tation or who do not wish to participate will be replaced powered to detect differences in clinical end-points will 44 45 by another patient/practitioner with similar characteris- be required to guide clinical practice. 46 tics. This process will continue until theoretical satura- 47 tion is achieved and interviews cease. Additional material 48 Interviews 49 Semi-structured, face-to-face, in-depth interviews will be Additional file 1: Interview guides. This file contains a copy of the ’ interview guides for Patients and Health Care Professionals. 50 carried out in patients own homes or in other suitable 51 locations, or with healthcare professionals in the sur- 52 gery, and will be conducted by a researcher trained in 53 qualitative interviewing techniques. Fully informed con- Acknowledgements 54 sent will be obtained from interviewees at the start of We would also like to acknowledge the help and support provided by the Stroke Research Network, the Primary Care Research Network and the West 55 the interview, and a consent form signed. An interview Midlands Research Consortium (MidReC). 56 topic guide will be used (see additional file 1) which will This programme receives financial support from the National Institute for 57 then be modified and refined during the first interviews. Health Research (NIHR) Programme Grants for Applied Research funding scheme. The views and opinions expressed in this editorial are those of the 58 Each interview is expected to last between 60 and 90 59 60 https://mc.manuscriptcentral.com/bmj Page 35 of 35 BMJ
Fletcher et al. BMC Cardiovascular Disorders 2010, 10:37 Page 8 of 8 1 http://www.biomedcentral.com/1471-2261/10/37 2 3 4 author(s) and do not necessarily reflect those of the NHS, NIHR or the 12. Hippisley Cox J, Pringle M, Ryan R: Stroke: prevalence, incidence and care 5 Department of Health. in general practices 2002 to 2004. Final Report to the National Stroke 6 Audit Team, Royal College of Physician. QRESEARCH, Nottingham 2004. 7 Author details 13. Hippisley Cox J, Fenty J, Langford G, Pringle M, Coupland C: Report to the 1PrimaryConfidential: Care Clinical Sciences, Clinical Sciences Building University ofFor NationalReview Audit Office: Quality of care Only for stroke and TIA in general 8 Birmingham, Edgbaston Birmingham UK, B15 2TT. 2General Practice & practice using the new GMS contract indicators. QRESEARCH, Nottingham 9 Primary Care Research Unit, Department of Public Health & Primary Care, 2005. 10 University of Cambridge, UK, CB2 0SR. 3Clinical Research facility Royal Victoria 14. Mant J, Barton P, Ryan R, et al: What is the optimum model of service 11 Infirmary Newcastle Upon Tyne UK, NE1 4LP. delivery for transient ischaemic attack? Report for the National Co- ordinating Centre for NHS Service Delivery and Organisation R&D. 2007. 12 Authors’ contributions 15. Glaser B, Strauss A: The discovery of grounded theory: strategies for 13 JM conceived the study. JM, KF, SC, GF, RH, RM and UM designed the RCT; qualitative research. Chicago: Aldine 1967. 14 KF, SV and SG were responsible for the design of the qualitative element; SJ, 16. Strauss A, Corbin J: Basics of qualitative research, grounded theory JM, CT and RM participated in the design of the health economics; JB, CT, procedures. London: Sage 1990. 15 and JM designed the treatment algorithm and NF participated in the design 17. Britten N: Qualitative interviews in medical research. BMJ 1995, 16 of the statistics. KF drafted the manuscript and all authors contributed to its 311:251-253. 17 revision. All authors have given final approval of the version to be published. 18. National Collaborating Centre for Chronic Conditions: Management of JM is the study guarantor. hypertension in adults in primary care: partial update. London: Royal 18 College of Physicians 2006. 19 Competing interests 19. The EuroQol Group: EQ-5 D user guide, Rotterdam, Netherlands, The 20 GF or his institution has received payment for educational, advisory and EuroQol Group. 1996. 21 research activities related to BP lowering drugs in the last 12 months from 20. Ware EJ: SF-36 Health Survey: Manual and Interpretation Guide. Boston, the following pharmaceutical companies: Boehringer Ingelheim, Servier. Massachusetts, The Health Institute, New England Medical Center 1993, 4:3. 22 FDRH has received limited research support in terms of BP devices from 21. Folstein MF, Folstein SE: “Mini-Mental State” A practical method for 23 Microlife and BpTRU and occasional sponsorship or speaker fees from a grading the cognitive state of patients for the clinician. J Psychiatric Res 24 number of pharmaceutical companies that market anti-hypertensives. 1975, 12:189-198. RJM is funded by an NIHR Career Development Fellowship and has no 22. Staessen JA, Wang J-G, Thijs L: Cardiovascular prevention and blood 25 conflicts with respect to this paper. pressure reduction: a quantitative overview updated until 1 March 2003. 26 NF has received funding for research and consulting from several companies Journal of Hypertension 2003, 21:1055-76. 27 who manufacture treatments for cardiovascular disease. 23. Campbell P, Ghuman N, Wakefield D, Wolfson L, White WB: Long-term None of the remaining authors have any conflicts of interest. reproducibility of ambulatory blood pressure is superior to office blood 28 pressure in the very elderly. J Hum Hypertens 2010. 29 Received: 23 June 2010 Accepted: 9 August 2010 24. Mant J, McManus RJ, Hare R, Mayer P: Identification of stroke in the 30 Published: 9 August 2010 community: a comparison of three methods. British Journal of General 31 Practice 2003, 53:520-524. 25. Ritchie J, Lewis J, (Eds): Qualitative Research Practice: A Guide for Social 32 References 1. National stroke strategy. [http://www.dh.gov.uk/en/ Science Students and Researchers. London: Sage 2003. 33 Publicationsandstatistics/Publications/PublicationsPolicyAndGuidance/ 26. Bamford J, Sandercock P, Dennis M, Burn J, Warlow C: A prospective study 34 DH_081062]. of acute cerebrovascular disease in the community: the Oxfordshire Community Stroke Study: 1981-86. Incidence, case fatality rates and 35 2. National Audit Office: Reducing brain damage: faster access to better stroke care. London: the Stationery Office 2005. overall outcome at one year of cerebral infarction, primary intracerebral 36 3. Mant J, Wade DT, Winner S: Health care needs assessment: stroke. Health and sub-arachnoid haemorrhage. J Neurol Neurosurg Psychiatry 1990, 37 care needs assessment: the epidemiologically based needs assessment reviews, 53:16-22. 38 First series Oxford: Radcliffe Medical PressStevens A, Raftery J, Mant J, 27. Horne R: The medication adherence report scale. University of Brighton: Simpson S , 2 2004, 141-244. Brighton, UK 2004. 39 4. National Institute for Health & Clinical Excellence: Cost Saving Guidance. 40 NICE 2010 [http://www.nice.org.uk/usingguidance/ Pre-publication history benefitsofimplementation/costsavingguidance.jsp]. The pre-publication history for this paper can be accessed here: 41 http://www.biomedcentral.com/1471-2261/10/37/prepub 42 5. PROGRESS Collaborative Group: Randomised trial of a perindopril-based BP lowering regimen among 6105 individuals with previous stroke or doi:10.1186/1471-2261-10-37 43 transient ischaemic attack. Lancet 2001, 358:1033-41. Cite this article as: Fletcher et al.: Protocol for Past BP: a randomised 44 6. Prospective Studies Collaboration: Age-specific relevance of usual blood controlled trial of different blood pressure targets for people with a 45 pressure to vascular mortality: a meta-analysis of individual data for one history of stroke of transient ischaemic attack (TIA) in primary care. BMC million adults in 61 prospective studies. Lancet 2002, 360:1903-13. Cardiovascular Disorders 2010 10:37. 46 7. Law MR, Morris JK, Wald NJ: Use of blood pressure lowering drugs in the 47 prevention of cardiovascular disease: meta-analysis of 147 randomised 48 trials in the context of expectations from prospective epidemiological studies. BMJ 2009, 338:b1665. 49 8. Arima H, Chalmers J, Woodward M, for the PROGRESS Group, et al: Lower Submit your next manuscript to BioMed Central 50 target BPs are safe and effective for the prevention of recurrent stroke. and take full advantage of: 51 Journal of Hypertension 2006, 24:1201-08. 52 9. Intercollegiate Stroke Working Party: National clinical guidelines for stroke. • Convenient online submission London: Royal College of Physicians, 3 2008. 53 10. Williams B, Poulter NR, Brown MJ, et al: Guidelines for management of • Thorough peer review 54 hypertension: report of the 4th working party of the BHS, 2004 - BHS IV. • No space constraints or color figure charges Journal of Human Hypertension 2004, 18:139-185. 55 • Immediate publication on acceptance 11. Mant J, McManus RJ, Hare R: Applicability to primary care of national 56 clinical guidelines on blood pressure lowering for people with stroke: • Inclusion in PubMed, CAS, Scopus and Google Scholar 57 cross sectional study. BMJ 2006, 332:635-637. • Research which is freely available for redistribution 58 59 Submit your manuscript at 60 www.biomedcentral.com/submit https://mc.manuscriptcentral.com/bmj