- Email: [email protected]
Risk factor management and depression post-stroke: The value of an integrated model of care
Journal of Clinical Neuroscience 13 (2006) 84–90 www.elsevier.com/locate/jocn
Clinical study
Risk factor management and depression post-stroke: The value of an integrated model of care Jacques Joubert a,*, Chris Reid b, Lynette Joubert c, David Barton d, Denise Ruth e, David Jackson a, John O’. Sullivan f, Stephen M. Davis g b
a Department of Neurology, Royal Melbourne Hospital, Parkville, Australia Centre of Clinical Research Excellence in Therapeutics, Department of Epidemiology & Preventive Medicine, Monash University c School of Social Work, University of Melbourne, Parkville, Australia d Department of Psychiatry, Royal Melbourne Hospital, Parkville, Australia and Department of Psychological Medicine, Monash University, Prahran, Australia e Senior Fellow, Department of General Practice, University of Melbourne f Blackburn Clinic, Blackburn, Australia g Department of Neurology, Department of Surgery, University of Melbourne, Royal Melbourne Hospital, Parkville, Australia
Received 27 April 2005; accepted 6 July 2005
Abstract Background and purpose: Shared care initiatives, albeit commonly utilised in managing other chronic conditions have not been implemented in the area of stroke in Australia. The aims of this project were to adopt a shared care approach for stroke survivors comparing an experimental ‘‘shared–care’’ group with a ‘‘treatment-as-usual’’ control group in reference to the normalization or reversal of vascular risk factors and the detection and management of post-stroke depression. Methods: A randomised controlled experimental research design was implemented with participants randomized to an intervention or control group and followed over a 12-month period. The treatment group consisted of a randomly selected group of patients, discharged from an acute stroke unit and transferred into the shared care model. Risk factors for stroke and depression were compared between the two groups. Results: Of 97 patients originally included in the study, 17 dropped out. At 12 months, 80 patients remained for analysis (35 in the intervention group and 45 in the control group). The findings demonstrated positive trends for patients within the intervention group that were not found within the control group. The percentage of intervention patients reaching target systolic blood pressure (sBP) of 140 mmHg after 12 months tended to be greater than in the control group (p = 0.11, NS). In the intervention group, at 12 months, the total cholesterol greater than 5.18 mmol/L was 12.5% compared to 58.8% at discharge. In contrast this trend was not so distinct in the control group (57.7% to 42.9%). The percentage of patients reaching target (recommended) total cholesterol of 5.18 mmol/L was significantly greater in the shared care patients intervention group relative to the control group (p = 0.02). The average number of walks per week was also significantly greater in intervention group compared to the control group (p = 0.048). Moreover, 45% of the control group screened as depressed compared with 20% of the intervention group at 12 months (p = 0.06). Conclusions: This study demonstrates that major risk factors for recurrent stroke and vascular disease in general are better managed with the shared care model than with usual post-discharge care. The significantly reduced depression as found on the screening PHQ9 at 12 months indicated that the intervention was beneficial not only in the detection of depression but also treatment. Ó 2005 Published by Elsevier Ltd. Keywords: Stroke prevention; Risk factors; Depression
In stroke survivors, recurrence of a cerebrovascular event remains a potential risk that increases with time. A *
Corresponding author. Tel.: +61 3 94266073; fax: +61 3 94266094. E-mail address: [email protected] (J. Joubert).
0967-5868/$ - see front matter Ó 2005 Published by Elsevier Ltd. doi:10.1016/j.jocn.2005.07.003
population-based Australian study1 has shown that after a first-ever stroke, patients having survived 30 days have a 2-fold excess risk of death compared to the general population of the same age and sex, with a cumulative risk of
J. Joubert et al. / Journal of Clinical Neuroscience 13 (2006) 84–90
death at 10 years of 79%. One in four will die of a recurrent stroke and one in three will die of a cardiovascular event.1 In recent years, significant advances have been made in the prevention of recurrent stroke. Hachinski has postulated that up to 75% of cerebrovascular and cardiac disease could be avoided by the effective management of known vascular risk factors.2 Similarly, Yusuf calculated that in a smoker with vascular disease, smoking cessation and employing of four prevention strategies would result in approximately 80% risk reduction.3 Of 30-day stroke survivors, only 1 in 7 are accommodated permanently in a nursing home.1 The majority return home to the ongoing care of their general practitioners (GPs). GPs are the most important long-term providers of medical care for stroke survivors. Ideally, evidence-based recommendations for secondary stroke prevention transmitted to GPs should be translated into clinical practice at the primary care level. Depression, both in the acute and later stages after stroke is common, occurring in between 20% and 60% of survivors.4 Not only does depression adversely affect recovery and rehabilitation of stroke patients,4 but depressed stroke survivors have more than three times the risk of dying over the next decade than those without depression.5 GPs are the major service providers to people with mental health problems in the community6 and therefore central to the management of post-stroke depression. Despite the large volume of literature addressing secondary stroke prevention, the translation of evidencebased recommendations into practice by GPs is suboptimal both in the United States7 and the United Kingdom.8 To address this we believe that an integrated care model of service delivery could be an effective means of enhancing the translation of evidence-based guidelines for the detection and management of vascular risk factor and depression into practice at a primary care level. 1. Conceptualization of an integrated care model Over two years of extensive consultation with representatives from four Divisions of General Practice in Melbourne, Australia a model of integrated care for the stroke survivor was developed. Thus, GPs were involved at the earliest stages in the conceptualization of a dynamic, structured model that would link specialist stroke services bi-directionally with ongoing general practice care. Careful consideration was given to provision of clear directives for vascular risk factor management, as well as detection and treatment of depression according to evidence-based guidelines. A priority was not to increase the administrative burden on GPs but to facilitate management of risk factors and the detection and treatment of depression. Likewise, consideration was given to personal contact at key points between the specialist service (study neurologist) and the GP. A telephone tracking system provided pertinent ongoing information to the GP, gave patient and carer support, while the use of a flowchart facilitated risk factor management and provided feedback to the stroke service. This in
85
turn allowed for surveillance of the evolving risk factor profile and possible development of depression over time. Other important issues considered were the cost and sustainability of the model. 2. Melbourne stroke integrated care study The study developed from the concept of the shared care model. It was designed to evaluate whether the integrated care model is more effective than usual care in the longterm management of vascular risk factors and depression in non-institutionalized stroke survivors. A randomized control trial was funded by a Commonwealth of Australia General Practice Evaluation Program (GPEP) competitive grant. This paper describes the methodology and early findings of this study, the aims of which were to evaluate; 1) the effect of a shared care model on the management of the vascular risk factors for stroke according to accepted best-practice guidelines. 2) the effect of screening for post-stroke depression by a validated telephone assessment method with feedback to the GP in the context of a shared care model. 3) the effect of such a shared care model on stroke recurrence and long-term stroke-related mortality.
3. Participants and methods In 2000 and 2001, a prospective randomized control trial was undertaken in the stroke units of the Royal Melbourne Hospital and the Western General Hospital. Patients included in the study were randomized to an intervention or control group and were followed over a 12-month period. The study was conducted according to the Declaration of Helsinki and was approved by the Research and Ethics Committee of the Royal Melbourne Hospital and Western Hospital. 4. Inclusion and exclusion criteria All consecutive patients (aged 20 years and older) of both sexes, admitted with transient ischemic attack or completed stroke (cerebral infarction or hemorrhage) confirmed by computerized tomography scan were considered for inclusion in the study. Eligibility was determined only after imaging and clinical status had been reviewed. The inclusion criteria admitted patients with both first or recurrent stroke who would return to their GPs for management after discharge, and who were willing to be contacted for repeat assessments over a 12-month period. Patients were excluded if they were discharged to nursing homes, had serious or life-threatening co-morbidities (such as cancer), were non-English speaking, refused to participate, died while in hospital, were cognitively impaired to the extent that they could not cooperate with follow-up visits with their GP (mini-mental test score less than 23), were
86
J. Joubert et al. / Journal of Clinical Neuroscience 13 (2006) 84–90
Fig. 1. Diagrammatic Representation of Integrated Care Model.
significantly aphasic and those living more than two hours drive by car from either hospital. Following confirmation of eligibility and the provision of informed consent, patients were randomized into an intervention or control group according to a computergenerated process. If more than one stroke patient was treated by the same GP, all subsequent patients were allocated to the same condition as the first patient to prevent contamination of the sample. 5. Clinical data and in-hospital activities For intervention participants, clinical evaluation inhospital included neurological and clinical status, Rankin Stroke Severity assessment, the Barthel Index, Frenchay Aphasia Screening Test (FAST), Modified Mini Mental State Examination, the Australian Quality of Life assessment (AQoL) and the Patient Health Questionnaire Depression Model (PHQ-9). Identical assessments were administered to the control participants, with the exception of the PHQ-9. When available, the following laboratory results were recorded: lipid profile, blood glucose level (BGL), hemoglobin A1C, International Normalised Ratio (INR), urea and electrolytes, computerized tomography (CT), magnetic resonance imaging (MRI), carotid ultrasound, radiology of the chest, transoesophageal echocardiography (TOE) and transthoracic echocardiography (TTE). Stroke subtypes were categorized as 1) lacunar infarction, 2) large vessel atherothrombotic infarction, 3) cardio-embolic infarction and 4) cerebral hemorrhage. Prior to discharge, each patient’s risk factor profile was recorded. These included: hypertension, diabetes mellitus, dyslipidaemia, atrial fibrillation, cardiac source for embolism, significant carotid artery stenosis and a history of transient ischaemic attack, previous stroke, excessive alcohol consumption, smoking history and obesity (BMI > 25). In the hospital, the role of the study coordinator was explained to patients randomized to the intervention group.
The process of integrated care (see Fig. 1) including the importance of the effective management of modifiable risk factors, the institution of beneficial life-style changes such as regular exercise, limiting alcohol intake and cessation of smoking where appropriate were discussed. Patients randomized into the control group were informed that the coordinator would contact them in 12 months to review diagnostic test results and clinical data from the previous year. 5.1. Carer identification A suitable family member or friend was nominated as carer. The integrated care model was explained to them and their integral role in the process of care was emphasized. 5.2. Post-discharge shared care GPs of patients in the intervention group were sent explanatory letters regarding the integrated care program and a comprehensive but succinct discharge summary detailing initial presentation, social circumstances, known risk factors for stroke, relevant investigations, discharge clinical status and medication, final stroke diagnosis as well as discharge planning. A shared care package was prepared for the GP. This contained goals and the recommendations for risk factor management according to clinical practice and evidence-based guidelines,9 a flow chart designed for the serial recording of modifiable vascular risk factors and depression, references from the stroke literature and recommendations for treatment of depression in stroke patients. For the intervention group, visits were arranged with the GP’s office for 2 weeks, 3 months, 6 months, 9 months and 12 months post-discharge. For control patients, GPs were contacted at 12 months and collection of study-specific data was obtained.
J. Joubert et al. / Journal of Clinical Neuroscience 13 (2006) 84–90
5.3. Role of study neurologist The study neurologist met with coordinators on a weekly basis to discuss clinical and auditing issues related to the integrated process. Prior to discharge, the risk factor profile and discharge summary of each patient was verified by the study neurologist. The study neurologist telephoned the patients’ GPs, discussing the clinical details of the patients and seeking the GPs’ participation in the study. In this model, patients are reviewed shortly after discharge at the stroke clinic, where the study neurologist emphasizes the importance of correct risk factor management and discusses the integrated model with both the patient and carer. The neurologist remains available for telephone advice to the GPs throughout the study. 5.4. Telephone tracking A week before each GP visit, the study coordinator telephoned the patient to assess their progress. Discussion took place with both patient and nominated carer. Data collected at this pre-GP-visit telephone assessment included the AQoL, current problems experienced by the patient or family related to the stroke, measures of ambulation (distance and frequency), current cigarette smoking, alcohol consumption, number of GP visits in the last three months, and changes to medication. In addition, the Patient Health Questionnaire Depression Module (PHQ-9)10 was administered by telephone. This is a screening tool for depression based on the DSM-1V criteria for diagnosing major depression developed in and for primary care and validated for telephone use. It provides a measure of severity of depression which can be repeated to inform treatment decisions. The information collected from the patient at the pre-GP-visit telephone assessments was transmitted by facsimile to the GP prior to each of the scheduled GP consultations during the study period. After each GP visit, the patient was telephoned to ascertain whether the visit took place and if there were any changes to medication and management. Nonattendance prompted re-scheduling of the appointment. This process of pre-and post visit telephone assessment was conducted at every three-monthly scheduled GP visit.
87
the coordinator and the data was entered into a database. Specifically, the data sought from the GP included the patients’ blood pressure and medication, LDL, HDL, total cholesterol, tryglicerides, INR and Warfarin dosage if indicated, blood sugar level (fasting), carotid stenosis and antiplatelet medication. In the case of best-practice standards not being met, the coordinator or the study neurologist contacted the GP and options for review, diagnostic testing and treatment were discussed. 5.7. Twelve-month follow-up Both the intervention and the control participants were followed up at 12 months and data regarding current problems, depression, frequency and amount of ambulation, smoking and alcohol consumption, number of GP visits in the last 3 months, changes to medication and the AQoL were obtained from a telephone interview. At a final interview conducted at patients’ homes, questionnaires including The London Handicap Scale, The Carer Strain Index, The PIE assessment, Clinical Status, Modified Rankin Scale, Modified Mini Mental State Examination, The Barthel Index, The MOS Social Support Survey, and the FAST were administered. Patients’ access to services and care in the community was also addressed. 5.8. Data analysis and storage All data was recorded on pre-formatted forms in hard copy and entered into the study database. Analysis was conducted using SPSS Version 11 statistical software (SPSS, 2003). Differences in the intervention group and control group were analyzed using independent t-test analysis (continuous variables) and Chi-square analysis (categorical variables). Analyses were conducted on participants completing 12 months follow-up. 6. Results A total of 421 confirmed stroke patients were admitted collectively to the Royal Melbourne Hospital and the Western Hospital between December 2000 and November 2001. 224 patients were considered eligible for inclusion
5.5. GP consultation two weeks after discharge At the first GP consultation after discharge, the coordinator met with the patient, carer and GP at the GP’s office. The coordinator explained to the GP the use of the shared care folder and flow chart and that prior to future three two monthly visits, telephone assessments would be faxed to the GP. Following the first consultation, the GP completed the flowchart and faxed this information to the coordinator. 5.6. Three-monthly progress tracking After each three-monthly GP visit, the flow charts were transmitted by facsimile to the stroke unit scrutinized by
Table 1 Baseline characteristics of participants
Age % Male Systolic BP Diastolic BP Total cholesterol HDL cholesterol LDL cholesterol Glucose HbA1c Body mass index
Intervention (n = 46)
Control (n = 51)
P
64.7 ± 14.9 51% 134 ± 16 75 ± 11 5.1 ± 0.9 1.19 ± 0.33 3.1 ± 1.0 6.4 ± 2.4 8.2 ± 3.8 27.5 ± 5.6
68.21 ± 12.5 49% 131 ± 18 77 ± 11 5.5 ± 1.1 1.21 ± 0.48 3.5 ± 1.1 6.9 ± 2.6 7.5 ± 3.8 29.0 ± 5.8
ns ns ns ns ns ns ns ns ns ns
88
J. Joubert et al. / Journal of Clinical Neuroscience 13 (2006) 84–90
in the study. Of these 77 patients refused, 53 the family declined to be involved, four patients were participating in other research programs and two primary care clinicians refused to participate. A further 17 patients were not seen prior to discharge and two patients were transferred to other acute hospitals. 126 patients remained eligible for participation and a total of 97 patients were randomised. 17 patients withdrew after randomisation with the reasons for dropout including; reconsideration of participation, re-location overseas, later admission to a nursing home, inadequate GP support, patient reluctance to visit GP and non-vascular death. At baseline, there was no statistical difference between the control and the intervention groups regarding age, sex, stroke severity, stroke pathology, mean cholesterol level, blood sugar and blood pressure (see Table 1). 7. Changes in risk factors 7.1. Blood pressure Comparison at 12 months showed that the systolic blood pressure (sBP) in the control group increased by an average of 4 mm Hg to 136 mm Hg, while that of the intervention group decreased by an average of 2 mm Hg to 132 mm Hg (p = ns). The difference between groups was not
significant (p = 0.071). The percentage of patients reaching target sBP of 140 mm Hg or less after 12 months in shared care tended to be greater than in the control group (p = 0.11) (see Figs. 2 and 3). 7.2. Total cholesterol Whilst the number of patients with elevated cholesterol was similar at discharge from hospital for both groups, at 12 months the mean total cholesterol for the control group was 5.1 mm/L while that of the intervention group was 4.7 mmol/L (p = 0.16) (see Fig. 4). Moreover, the percentage of patients reaching the (recommended) total cholesterol of 5.18 mmol/L was greater in the intervention group (p = 0.02) (see Fig. 5). 7.3. Physical activity Exercise participation, assessed as the number of walks per week, increased over the twelve-month period in the intervention group compared to the control group (p = 0.048) (see Fig. 6). 7.4. Depression Almost half (45%) of the control group screened as depressed at 12 months following discharge. Compared with
136.58
137 136
5.4
134.49
134 133
Tr eatment (n=35) Control (n=45)
132.34
132.29
Mean mmol/L
mm Hg
135
5.53
5.6
5.2
5.14
5.1 Treatment (n=31)
5
Control (n=33) 4.78
4.8
132 4.6
131
4.4
130 Systolic Blood Pressure at Discharge
Systolic Blood Pressure at 12 Months
Fig. 2. Mean systolic blood pressure at discharge and 12 months (treatment n = 35, control n = 45) (ns).
Total Cholesterol at Total Cholesterol at Discharge 12 Months
Fig. 4. Mean cholesterol at discharge and 12 months (treatment n = 32, control n = 33) (ns). 70
52.3 Treatment: %of patients sBP< 140mmHG (n=35)
Percentage (%)
52 51
Control: %of patients sBP < 140mmHG (n=45)
50 49 48
47.7
50 Percentage (%)
53
62.9
60
35.6
40 30 20
47 10
46 45
0
12 Months
Fig. 3. Percentage of patients with systolic blood pressure less than 140 mm HG at 12 months (treatment n = 35, control n = 45) (ns).
Treatment (n=35)
Control (n=45)
Fig. 5. Percentage of patients with cholesterol less than 5.18 mmol/L at 12 months (treatment n = 35, control n = 45) (p = 0.24).
J. Joubert et al. / Journal of Clinical Neuroscience 13 (2006) 84–90 5
4.5
Mean no. of walks
4.5 4
3.5
3.5
3.4
3.1
3 2.5
Treatment (n=35)
2
Control (n=45)
1.5 1 0.5 0
Walks per week at Walks per week at 12 discharge months
Percentage (%)
Fig. 6. Mean number of walks per week at discharge and 12 months (treatment n = 35, control n = 45) (ns).
50 45 40 35 30 25 20 15 10 5 0
45
Treatment (n=35)
20
Control (n=45)
Depression as per PHQ#9 score at 12 months
Fig. 7. Depression as measured by PHQ9 at discharge and 12 months (treatment n = 35, control, n = 45) (ns).
this, 20% of the intervention group screened as depressed at 12 months suggesting that the intervention program was associated with reduced rates of depression (p = 0.06). Feedback from GPs was uniformly positive regarding the value and ease of administration of the screening tool (Fig. 7). 8. Discussion This study describes the early results of an integrated care model addressing vascular risk factors and chronic post-stroke depression in stroke survivors. The model combines a structured approach to best-practice management, a tracking system to improve compliance with evidencebased guidelines and ongoing support from a specialist service. Despite the small sample size the early results indicate trends toward significance with blood pressure control, cholesterol levels and physical activity in the intervention group compared to control patients exposed to usual care. There was a similar positive trend between the intervention and control group with regard to screening for depression. A definition of shared care is ‘primary and secondary care clinicians having joint contemporaneous responsibility’ for patient care, including routine management and monitoring.11 A review in 1999 of shared care in Australia showed that there were examples for conditions such as diabetes, asthma and mental health, but none for stroke.11
89
To our knowledge, this is still the case. Cerebrovascular disease incidence is particularly amenable to reduction by risk factor management.2,3,9,13 Approximately 44,000 Australians suffer a stroke each year.12 Stroke is expensive23 and secondary prevention by effectively translating evidence-based guidelines into practice by a model of shared care has significant potential economic benefits. In excess of 50% of stroke survivors have residual disability with impairment of function.9 After a first stroke, the risk of recurrence is about 7% per year.14 Stroke recurrence frequently compounds the already present disability, increasing rehabilitation and hospitalization costs and contributing to high long-term mortality.9 Stroke survivors are at particular risk of ischemic cardiovascular events; one third of deaths at 10 years result from cardiovascular disease.1 Measures for the secondary prevention of stroke may have an additional benefit in that the profile of risk factors for stroke and coronary heart disease are very similar.9 After TIA or mild ischemic stroke, the use of anticoagulation in patients with atrial fibrillation or revascularisation of carotid artery stenosis of between 70% to 99% is recommended. If these two conditions are not present, the use of low-dose aspirin is of demonstrated benefit.9 Randomised controlled trials of lowering both systolic and diastolic blood pressure have shown relative risk reduction of stroke by about 28%13 and the Heart Outcomes Prevention Evaluation (HOPE) Trial showed that the use of ramipril lowered stroke risk independent of blood pressure reduction.15 Simvastatin is effective in secondary stroke prevention and observational studies have suggested that tight diabetic control and smoking cessation are effective secondary stroke prevention measures.13 Accordingly, the American Heart Association has published a consensus statement with detailed guidelines and recommendations for secondary prevention of stroke.9 Despite this information, translating research into practice has lagged.7,8,16,17 A study comparing secondary prevention practices of GPs in the United States and United Kingdom revealed that often, they did not follow published recommendations.7 Gibbs et al. found a ‘‘marked under use of antiplatelet and anticoagulant agents’’ in patients with a new diagnosis of stroke or TIA.8 Mead et al.17 found that in patients with atrial fibrillation, only 20% of GPs interviewed would consider instituting anticoagulation. Low treatment rates (ranging from 39% to 62%) have been reported elsewhere.16,18 Interestingly, in comparison, compliance with secondary prevention of myocardial infarction is relatively good.19 The barriers to translating recommendations into practice are not obvious.20 Whether this can be attributed to a lack of awareness of evidence and/or a resistance to changing clinical practice remains unknown. Even at a dedicated stroke follow-up clinic, where 92% of patients had a primary care physician ‘‘with whom the clinic was in touch’’, patient follow-up after two years indicated that compliance with
90
J. Joubert et al. / Journal of Clinical Neuroscience 13 (2006) 84–90
recommendations for secondary prevention was poor.21 A review of possible barriers indicated that physicians will not implement guidelines if unfamiliar with them, if there are no available tools, such as a reminder system, and if they are unconvinced that a guideline will impact on patient outcomes.20 Moreover, lifestyle changes particularly in the elderly, can be difficult to effect. In this study, the lack of sustained motivation due to a lack of awareness by patients and carers of the potential long-term dangers of risk factor neglect was evident by the reasons given by the intervention group drop-outs. This highlights the need to provide ongoing, detailed information about stroke and risk factor management that is physician- and patient-focused.20,21 Nevertheless, an examination of potential interventions, such as point-of-care reminders and academic detailing, were found to be effective.20 Supplying physicians with evidence-based guidelines for hypertension control has resulted in their implementation.21 In another study,22 weekly telephone feedback to physicians of hypertensive control resulted in a mean 6.0 mmHg decrease in patients exposed to this intervention compared to a 2.8 mmHg increase in diastolic blood pressure in the control group. Because opinions vary among physicians as to the significance of risk factors and how aggressively they should be managed, it is essential to widely disseminate evidence-based guidelines. Not only should physicians be involved in the educational process, but also carers and patients. Use of modern information technology should be used to track adherence to guidelines by physicians. In this study, the trend to reduction of depression in the intervention group, according to the screening tool, is noteworthy, but the reasons for this are unclear. The telephone tracking system, with early recognition of depression and feedback to physicians, may account for this but other factors may play a role. In this study, because of the different nature of acute and chronic post-stroke depression4, comparison of the rate of depression in the intervention and control groups was done at 12 months and not baseline. The difference between the two groups is the basis for a more detailed future study. The study demonstrates that the screening tool (PHQ-9) is readily administered by telephone. The positive trends in this study demonstrate that through using a shared care model in the stroke survivor, improved management of vascular risk factors is possible and that screening for depression with feed-back to the GP may result in decreased post-stroke depression at 12 months. As yet it is uncertain what elements of the model are effective in reducing the barriers to risk factor management and the recognition and treatment of post-stroke depression at primary care level. Future studies should address this issue. References 1. Hardie K, Hankey GJ, Jamrozik K, Broadhurst RJ, Anderson C. Ten-Year Survival After First-Ever Stroke in the Perth Community Stroke Study. Stroke 2003;34:1842–6.
2. Hachinski V. Stroke: 30 years of progress. Stroke 2002;33:1–4. 3. Yusuf S. Two decades of progress in preventing vascular disease. The Lancet 2002;360:2–3. 4. Carota A, Bogousslavsky J. Poststroke Depression Ischemic Stroke: Advances in Neurology, 92. Philadelphia: Lippincott Williams & Wilkins; 2003. 5. Everson SA, Ruberts RE, Goldberg DE, et al. Depressive symptoms and increased risk of stroke mortality over a 29-year period. Arch Intern Med 1998;158:1133–8. 6. CLIPP (Consultation and Liaison in Primary-Care Psychiatry) Manual. Victorian Government Department of Human Services, Melbourne; 2000. Available at: http://www.health.vic.gov.au/mentalhealth/publications/clipp. 7. Goldstein LB, Farmer A, Matchar DB. Primary Care Physician – Reported Secondary and Tertiary Stroke Prevention Practices. A Comparison Between the United States and United Kingdom. Stroke 1997;28:746–51. 8. Gibbs RG, Newson R, Lawrenson R, Greenhalgh RM, Davies AH. Diagnosis and Initial Management of Stroke and Transient Ischemic Attack Across UK Health Regions from 1992 to 1996. Experience of a National Primary Care Database. Stroke 2001;32: 1085–90. 9. Wolf PA, Clegett GP, Easton JD, et al. Preventing Ischemic Stroke in Patients With Prior Stroke and Transient Ischemic Attack. A Statement for Healthcare Professionals; From the Stroke Council of the American Heart Association. Stroke 1999;30:1991–4. 10. Spitzer R, Kroenke K, Williams J. Validity and utility of a self-report version of the PRIME-MD: the PHQ Primary Care Study. JAMA 1999;282:1737–44. 11. Harris MF, Davies GP. Shared care through divisions of general practice: Moving forward. Med J Aus 85–8. 12. Stroke Australia Task Force. National Stroke Strategy. Melbourne: National Stroke Foundation; 1997. 13. Hankey GJ. Risk Factor Management to Prevent Stroke. Ischemic Stroke: Advances in Neurology, 92. Philadelphia: Lippincott Williams & Wilkins; 2003. 14. Hankey GJ, Jamrozik J, Broadhurst RH, et al. Long-term risk of first recurrent stroke in the Perth Community Stroke Study. Stroke 1998;29:2491–500. 15. Heart Outcomes Prevention Evaluation (HOPE) Study investigators. Effects of angiotensin converting-enzyme inhibitor, ramipril, on cardiovascular events in high risk-patients. N Engl J Med 2000;342:145–53. 16. Cohen N, Almoznino-Sarafan D, et al. Warfarin for Stroke Prevention Still Underused in Atrial Fibrillation. Stroke 2000;31: 1217–22. 17. Mead GE, Murray H, McCullum CN, O’Neill PA. How do general practitioners manage patients at risk from stroke?. Br J Clin Pract 1996;50:426–30. 18. Sappok T, Faulstich A, Stuckert E, Kruck H, Marx P, Koennecke HC. Compliance with secondary prevention of ischemic stroke a prospective evaluation. Stroke 2001;32:1884–9. 19. Bradley F, Morgan S, Smith H, Mant D. Preventive care for patients following myocardial infarction. Fam Pract 1997;14:220–6. 20. Ornstein SM. Translating Research into Practice Using Electronic Medical Records. The PPRNet-TRIP Project: Primary and Secondary Prevention of Coronary Heart Disease and Stroke. Top Health Inf Manage 2002;22:52–8. 21. Joseph LN, Babikian VL, Allen NC, Winter MR. Risk factor modification in stroke prevention –the experience of a stroke clinic. Stroke 1999;30:16–20. 22. Friedman RH, Kazis LE, Jette A, Smith MB, Stollerman J, Togerson K, Carey K. A telecommunications system for monitoring and counseling patients with hypertension: impact on medication adherence and blood pressure control. Am J Hypertens 1996;9:285– 92. 23. Matchar DB. The value of stroke prevention treatment. Neurology 1998;51:S31–5.
Clinical study
Risk factor management and depression post-stroke: The value of an integrated model of care Jacques Joubert a,*, Chris Reid b, Lynette Joubert c, David Barton d, Denise Ruth e, David Jackson a, John O’. Sullivan f, Stephen M. Davis g b
a Department of Neurology, Royal Melbourne Hospital, Parkville, Australia Centre of Clinical Research Excellence in Therapeutics, Department of Epidemiology & Preventive Medicine, Monash University c School of Social Work, University of Melbourne, Parkville, Australia d Department of Psychiatry, Royal Melbourne Hospital, Parkville, Australia and Department of Psychological Medicine, Monash University, Prahran, Australia e Senior Fellow, Department of General Practice, University of Melbourne f Blackburn Clinic, Blackburn, Australia g Department of Neurology, Department of Surgery, University of Melbourne, Royal Melbourne Hospital, Parkville, Australia
Received 27 April 2005; accepted 6 July 2005
Abstract Background and purpose: Shared care initiatives, albeit commonly utilised in managing other chronic conditions have not been implemented in the area of stroke in Australia. The aims of this project were to adopt a shared care approach for stroke survivors comparing an experimental ‘‘shared–care’’ group with a ‘‘treatment-as-usual’’ control group in reference to the normalization or reversal of vascular risk factors and the detection and management of post-stroke depression. Methods: A randomised controlled experimental research design was implemented with participants randomized to an intervention or control group and followed over a 12-month period. The treatment group consisted of a randomly selected group of patients, discharged from an acute stroke unit and transferred into the shared care model. Risk factors for stroke and depression were compared between the two groups. Results: Of 97 patients originally included in the study, 17 dropped out. At 12 months, 80 patients remained for analysis (35 in the intervention group and 45 in the control group). The findings demonstrated positive trends for patients within the intervention group that were not found within the control group. The percentage of intervention patients reaching target systolic blood pressure (sBP) of 140 mmHg after 12 months tended to be greater than in the control group (p = 0.11, NS). In the intervention group, at 12 months, the total cholesterol greater than 5.18 mmol/L was 12.5% compared to 58.8% at discharge. In contrast this trend was not so distinct in the control group (57.7% to 42.9%). The percentage of patients reaching target (recommended) total cholesterol of 5.18 mmol/L was significantly greater in the shared care patients intervention group relative to the control group (p = 0.02). The average number of walks per week was also significantly greater in intervention group compared to the control group (p = 0.048). Moreover, 45% of the control group screened as depressed compared with 20% of the intervention group at 12 months (p = 0.06). Conclusions: This study demonstrates that major risk factors for recurrent stroke and vascular disease in general are better managed with the shared care model than with usual post-discharge care. The significantly reduced depression as found on the screening PHQ9 at 12 months indicated that the intervention was beneficial not only in the detection of depression but also treatment. Ó 2005 Published by Elsevier Ltd. Keywords: Stroke prevention; Risk factors; Depression
In stroke survivors, recurrence of a cerebrovascular event remains a potential risk that increases with time. A *
Corresponding author. Tel.: +61 3 94266073; fax: +61 3 94266094. E-mail address: [email protected] (J. Joubert).
0967-5868/$ - see front matter Ó 2005 Published by Elsevier Ltd. doi:10.1016/j.jocn.2005.07.003
population-based Australian study1 has shown that after a first-ever stroke, patients having survived 30 days have a 2-fold excess risk of death compared to the general population of the same age and sex, with a cumulative risk of
J. Joubert et al. / Journal of Clinical Neuroscience 13 (2006) 84–90
death at 10 years of 79%. One in four will die of a recurrent stroke and one in three will die of a cardiovascular event.1 In recent years, significant advances have been made in the prevention of recurrent stroke. Hachinski has postulated that up to 75% of cerebrovascular and cardiac disease could be avoided by the effective management of known vascular risk factors.2 Similarly, Yusuf calculated that in a smoker with vascular disease, smoking cessation and employing of four prevention strategies would result in approximately 80% risk reduction.3 Of 30-day stroke survivors, only 1 in 7 are accommodated permanently in a nursing home.1 The majority return home to the ongoing care of their general practitioners (GPs). GPs are the most important long-term providers of medical care for stroke survivors. Ideally, evidence-based recommendations for secondary stroke prevention transmitted to GPs should be translated into clinical practice at the primary care level. Depression, both in the acute and later stages after stroke is common, occurring in between 20% and 60% of survivors.4 Not only does depression adversely affect recovery and rehabilitation of stroke patients,4 but depressed stroke survivors have more than three times the risk of dying over the next decade than those without depression.5 GPs are the major service providers to people with mental health problems in the community6 and therefore central to the management of post-stroke depression. Despite the large volume of literature addressing secondary stroke prevention, the translation of evidencebased recommendations into practice by GPs is suboptimal both in the United States7 and the United Kingdom.8 To address this we believe that an integrated care model of service delivery could be an effective means of enhancing the translation of evidence-based guidelines for the detection and management of vascular risk factor and depression into practice at a primary care level. 1. Conceptualization of an integrated care model Over two years of extensive consultation with representatives from four Divisions of General Practice in Melbourne, Australia a model of integrated care for the stroke survivor was developed. Thus, GPs were involved at the earliest stages in the conceptualization of a dynamic, structured model that would link specialist stroke services bi-directionally with ongoing general practice care. Careful consideration was given to provision of clear directives for vascular risk factor management, as well as detection and treatment of depression according to evidence-based guidelines. A priority was not to increase the administrative burden on GPs but to facilitate management of risk factors and the detection and treatment of depression. Likewise, consideration was given to personal contact at key points between the specialist service (study neurologist) and the GP. A telephone tracking system provided pertinent ongoing information to the GP, gave patient and carer support, while the use of a flowchart facilitated risk factor management and provided feedback to the stroke service. This in
85
turn allowed for surveillance of the evolving risk factor profile and possible development of depression over time. Other important issues considered were the cost and sustainability of the model. 2. Melbourne stroke integrated care study The study developed from the concept of the shared care model. It was designed to evaluate whether the integrated care model is more effective than usual care in the longterm management of vascular risk factors and depression in non-institutionalized stroke survivors. A randomized control trial was funded by a Commonwealth of Australia General Practice Evaluation Program (GPEP) competitive grant. This paper describes the methodology and early findings of this study, the aims of which were to evaluate; 1) the effect of a shared care model on the management of the vascular risk factors for stroke according to accepted best-practice guidelines. 2) the effect of screening for post-stroke depression by a validated telephone assessment method with feedback to the GP in the context of a shared care model. 3) the effect of such a shared care model on stroke recurrence and long-term stroke-related mortality.
3. Participants and methods In 2000 and 2001, a prospective randomized control trial was undertaken in the stroke units of the Royal Melbourne Hospital and the Western General Hospital. Patients included in the study were randomized to an intervention or control group and were followed over a 12-month period. The study was conducted according to the Declaration of Helsinki and was approved by the Research and Ethics Committee of the Royal Melbourne Hospital and Western Hospital. 4. Inclusion and exclusion criteria All consecutive patients (aged 20 years and older) of both sexes, admitted with transient ischemic attack or completed stroke (cerebral infarction or hemorrhage) confirmed by computerized tomography scan were considered for inclusion in the study. Eligibility was determined only after imaging and clinical status had been reviewed. The inclusion criteria admitted patients with both first or recurrent stroke who would return to their GPs for management after discharge, and who were willing to be contacted for repeat assessments over a 12-month period. Patients were excluded if they were discharged to nursing homes, had serious or life-threatening co-morbidities (such as cancer), were non-English speaking, refused to participate, died while in hospital, were cognitively impaired to the extent that they could not cooperate with follow-up visits with their GP (mini-mental test score less than 23), were
86
J. Joubert et al. / Journal of Clinical Neuroscience 13 (2006) 84–90
Fig. 1. Diagrammatic Representation of Integrated Care Model.
significantly aphasic and those living more than two hours drive by car from either hospital. Following confirmation of eligibility and the provision of informed consent, patients were randomized into an intervention or control group according to a computergenerated process. If more than one stroke patient was treated by the same GP, all subsequent patients were allocated to the same condition as the first patient to prevent contamination of the sample. 5. Clinical data and in-hospital activities For intervention participants, clinical evaluation inhospital included neurological and clinical status, Rankin Stroke Severity assessment, the Barthel Index, Frenchay Aphasia Screening Test (FAST), Modified Mini Mental State Examination, the Australian Quality of Life assessment (AQoL) and the Patient Health Questionnaire Depression Model (PHQ-9). Identical assessments were administered to the control participants, with the exception of the PHQ-9. When available, the following laboratory results were recorded: lipid profile, blood glucose level (BGL), hemoglobin A1C, International Normalised Ratio (INR), urea and electrolytes, computerized tomography (CT), magnetic resonance imaging (MRI), carotid ultrasound, radiology of the chest, transoesophageal echocardiography (TOE) and transthoracic echocardiography (TTE). Stroke subtypes were categorized as 1) lacunar infarction, 2) large vessel atherothrombotic infarction, 3) cardio-embolic infarction and 4) cerebral hemorrhage. Prior to discharge, each patient’s risk factor profile was recorded. These included: hypertension, diabetes mellitus, dyslipidaemia, atrial fibrillation, cardiac source for embolism, significant carotid artery stenosis and a history of transient ischaemic attack, previous stroke, excessive alcohol consumption, smoking history and obesity (BMI > 25). In the hospital, the role of the study coordinator was explained to patients randomized to the intervention group.
The process of integrated care (see Fig. 1) including the importance of the effective management of modifiable risk factors, the institution of beneficial life-style changes such as regular exercise, limiting alcohol intake and cessation of smoking where appropriate were discussed. Patients randomized into the control group were informed that the coordinator would contact them in 12 months to review diagnostic test results and clinical data from the previous year. 5.1. Carer identification A suitable family member or friend was nominated as carer. The integrated care model was explained to them and their integral role in the process of care was emphasized. 5.2. Post-discharge shared care GPs of patients in the intervention group were sent explanatory letters regarding the integrated care program and a comprehensive but succinct discharge summary detailing initial presentation, social circumstances, known risk factors for stroke, relevant investigations, discharge clinical status and medication, final stroke diagnosis as well as discharge planning. A shared care package was prepared for the GP. This contained goals and the recommendations for risk factor management according to clinical practice and evidence-based guidelines,9 a flow chart designed for the serial recording of modifiable vascular risk factors and depression, references from the stroke literature and recommendations for treatment of depression in stroke patients. For the intervention group, visits were arranged with the GP’s office for 2 weeks, 3 months, 6 months, 9 months and 12 months post-discharge. For control patients, GPs were contacted at 12 months and collection of study-specific data was obtained.
J. Joubert et al. / Journal of Clinical Neuroscience 13 (2006) 84–90
5.3. Role of study neurologist The study neurologist met with coordinators on a weekly basis to discuss clinical and auditing issues related to the integrated process. Prior to discharge, the risk factor profile and discharge summary of each patient was verified by the study neurologist. The study neurologist telephoned the patients’ GPs, discussing the clinical details of the patients and seeking the GPs’ participation in the study. In this model, patients are reviewed shortly after discharge at the stroke clinic, where the study neurologist emphasizes the importance of correct risk factor management and discusses the integrated model with both the patient and carer. The neurologist remains available for telephone advice to the GPs throughout the study. 5.4. Telephone tracking A week before each GP visit, the study coordinator telephoned the patient to assess their progress. Discussion took place with both patient and nominated carer. Data collected at this pre-GP-visit telephone assessment included the AQoL, current problems experienced by the patient or family related to the stroke, measures of ambulation (distance and frequency), current cigarette smoking, alcohol consumption, number of GP visits in the last three months, and changes to medication. In addition, the Patient Health Questionnaire Depression Module (PHQ-9)10 was administered by telephone. This is a screening tool for depression based on the DSM-1V criteria for diagnosing major depression developed in and for primary care and validated for telephone use. It provides a measure of severity of depression which can be repeated to inform treatment decisions. The information collected from the patient at the pre-GP-visit telephone assessments was transmitted by facsimile to the GP prior to each of the scheduled GP consultations during the study period. After each GP visit, the patient was telephoned to ascertain whether the visit took place and if there were any changes to medication and management. Nonattendance prompted re-scheduling of the appointment. This process of pre-and post visit telephone assessment was conducted at every three-monthly scheduled GP visit.
87
the coordinator and the data was entered into a database. Specifically, the data sought from the GP included the patients’ blood pressure and medication, LDL, HDL, total cholesterol, tryglicerides, INR and Warfarin dosage if indicated, blood sugar level (fasting), carotid stenosis and antiplatelet medication. In the case of best-practice standards not being met, the coordinator or the study neurologist contacted the GP and options for review, diagnostic testing and treatment were discussed. 5.7. Twelve-month follow-up Both the intervention and the control participants were followed up at 12 months and data regarding current problems, depression, frequency and amount of ambulation, smoking and alcohol consumption, number of GP visits in the last 3 months, changes to medication and the AQoL were obtained from a telephone interview. At a final interview conducted at patients’ homes, questionnaires including The London Handicap Scale, The Carer Strain Index, The PIE assessment, Clinical Status, Modified Rankin Scale, Modified Mini Mental State Examination, The Barthel Index, The MOS Social Support Survey, and the FAST were administered. Patients’ access to services and care in the community was also addressed. 5.8. Data analysis and storage All data was recorded on pre-formatted forms in hard copy and entered into the study database. Analysis was conducted using SPSS Version 11 statistical software (SPSS, 2003). Differences in the intervention group and control group were analyzed using independent t-test analysis (continuous variables) and Chi-square analysis (categorical variables). Analyses were conducted on participants completing 12 months follow-up. 6. Results A total of 421 confirmed stroke patients were admitted collectively to the Royal Melbourne Hospital and the Western Hospital between December 2000 and November 2001. 224 patients were considered eligible for inclusion
5.5. GP consultation two weeks after discharge At the first GP consultation after discharge, the coordinator met with the patient, carer and GP at the GP’s office. The coordinator explained to the GP the use of the shared care folder and flow chart and that prior to future three two monthly visits, telephone assessments would be faxed to the GP. Following the first consultation, the GP completed the flowchart and faxed this information to the coordinator. 5.6. Three-monthly progress tracking After each three-monthly GP visit, the flow charts were transmitted by facsimile to the stroke unit scrutinized by
Table 1 Baseline characteristics of participants
Age % Male Systolic BP Diastolic BP Total cholesterol HDL cholesterol LDL cholesterol Glucose HbA1c Body mass index
Intervention (n = 46)
Control (n = 51)
P
64.7 ± 14.9 51% 134 ± 16 75 ± 11 5.1 ± 0.9 1.19 ± 0.33 3.1 ± 1.0 6.4 ± 2.4 8.2 ± 3.8 27.5 ± 5.6
68.21 ± 12.5 49% 131 ± 18 77 ± 11 5.5 ± 1.1 1.21 ± 0.48 3.5 ± 1.1 6.9 ± 2.6 7.5 ± 3.8 29.0 ± 5.8
ns ns ns ns ns ns ns ns ns ns
88
J. Joubert et al. / Journal of Clinical Neuroscience 13 (2006) 84–90
in the study. Of these 77 patients refused, 53 the family declined to be involved, four patients were participating in other research programs and two primary care clinicians refused to participate. A further 17 patients were not seen prior to discharge and two patients were transferred to other acute hospitals. 126 patients remained eligible for participation and a total of 97 patients were randomised. 17 patients withdrew after randomisation with the reasons for dropout including; reconsideration of participation, re-location overseas, later admission to a nursing home, inadequate GP support, patient reluctance to visit GP and non-vascular death. At baseline, there was no statistical difference between the control and the intervention groups regarding age, sex, stroke severity, stroke pathology, mean cholesterol level, blood sugar and blood pressure (see Table 1). 7. Changes in risk factors 7.1. Blood pressure Comparison at 12 months showed that the systolic blood pressure (sBP) in the control group increased by an average of 4 mm Hg to 136 mm Hg, while that of the intervention group decreased by an average of 2 mm Hg to 132 mm Hg (p = ns). The difference between groups was not
significant (p = 0.071). The percentage of patients reaching target sBP of 140 mm Hg or less after 12 months in shared care tended to be greater than in the control group (p = 0.11) (see Figs. 2 and 3). 7.2. Total cholesterol Whilst the number of patients with elevated cholesterol was similar at discharge from hospital for both groups, at 12 months the mean total cholesterol for the control group was 5.1 mm/L while that of the intervention group was 4.7 mmol/L (p = 0.16) (see Fig. 4). Moreover, the percentage of patients reaching the (recommended) total cholesterol of 5.18 mmol/L was greater in the intervention group (p = 0.02) (see Fig. 5). 7.3. Physical activity Exercise participation, assessed as the number of walks per week, increased over the twelve-month period in the intervention group compared to the control group (p = 0.048) (see Fig. 6). 7.4. Depression Almost half (45%) of the control group screened as depressed at 12 months following discharge. Compared with
136.58
137 136
5.4
134.49
134 133
Tr eatment (n=35) Control (n=45)
132.34
132.29
Mean mmol/L
mm Hg
135
5.53
5.6
5.2
5.14
5.1 Treatment (n=31)
5
Control (n=33) 4.78
4.8
132 4.6
131
4.4
130 Systolic Blood Pressure at Discharge
Systolic Blood Pressure at 12 Months
Fig. 2. Mean systolic blood pressure at discharge and 12 months (treatment n = 35, control n = 45) (ns).
Total Cholesterol at Total Cholesterol at Discharge 12 Months
Fig. 4. Mean cholesterol at discharge and 12 months (treatment n = 32, control n = 33) (ns). 70
52.3 Treatment: %of patients sBP< 140mmHG (n=35)
Percentage (%)
52 51
Control: %of patients sBP < 140mmHG (n=45)
50 49 48
47.7
50 Percentage (%)
53
62.9
60
35.6
40 30 20
47 10
46 45
0
12 Months
Fig. 3. Percentage of patients with systolic blood pressure less than 140 mm HG at 12 months (treatment n = 35, control n = 45) (ns).
Treatment (n=35)
Control (n=45)
Fig. 5. Percentage of patients with cholesterol less than 5.18 mmol/L at 12 months (treatment n = 35, control n = 45) (p = 0.24).
J. Joubert et al. / Journal of Clinical Neuroscience 13 (2006) 84–90 5
4.5
Mean no. of walks
4.5 4
3.5
3.5
3.4
3.1
3 2.5
Treatment (n=35)
2
Control (n=45)
1.5 1 0.5 0
Walks per week at Walks per week at 12 discharge months
Percentage (%)
Fig. 6. Mean number of walks per week at discharge and 12 months (treatment n = 35, control n = 45) (ns).
50 45 40 35 30 25 20 15 10 5 0
45
Treatment (n=35)
20
Control (n=45)
Depression as per PHQ#9 score at 12 months
Fig. 7. Depression as measured by PHQ9 at discharge and 12 months (treatment n = 35, control, n = 45) (ns).
this, 20% of the intervention group screened as depressed at 12 months suggesting that the intervention program was associated with reduced rates of depression (p = 0.06). Feedback from GPs was uniformly positive regarding the value and ease of administration of the screening tool (Fig. 7). 8. Discussion This study describes the early results of an integrated care model addressing vascular risk factors and chronic post-stroke depression in stroke survivors. The model combines a structured approach to best-practice management, a tracking system to improve compliance with evidencebased guidelines and ongoing support from a specialist service. Despite the small sample size the early results indicate trends toward significance with blood pressure control, cholesterol levels and physical activity in the intervention group compared to control patients exposed to usual care. There was a similar positive trend between the intervention and control group with regard to screening for depression. A definition of shared care is ‘primary and secondary care clinicians having joint contemporaneous responsibility’ for patient care, including routine management and monitoring.11 A review in 1999 of shared care in Australia showed that there were examples for conditions such as diabetes, asthma and mental health, but none for stroke.11
89
To our knowledge, this is still the case. Cerebrovascular disease incidence is particularly amenable to reduction by risk factor management.2,3,9,13 Approximately 44,000 Australians suffer a stroke each year.12 Stroke is expensive23 and secondary prevention by effectively translating evidence-based guidelines into practice by a model of shared care has significant potential economic benefits. In excess of 50% of stroke survivors have residual disability with impairment of function.9 After a first stroke, the risk of recurrence is about 7% per year.14 Stroke recurrence frequently compounds the already present disability, increasing rehabilitation and hospitalization costs and contributing to high long-term mortality.9 Stroke survivors are at particular risk of ischemic cardiovascular events; one third of deaths at 10 years result from cardiovascular disease.1 Measures for the secondary prevention of stroke may have an additional benefit in that the profile of risk factors for stroke and coronary heart disease are very similar.9 After TIA or mild ischemic stroke, the use of anticoagulation in patients with atrial fibrillation or revascularisation of carotid artery stenosis of between 70% to 99% is recommended. If these two conditions are not present, the use of low-dose aspirin is of demonstrated benefit.9 Randomised controlled trials of lowering both systolic and diastolic blood pressure have shown relative risk reduction of stroke by about 28%13 and the Heart Outcomes Prevention Evaluation (HOPE) Trial showed that the use of ramipril lowered stroke risk independent of blood pressure reduction.15 Simvastatin is effective in secondary stroke prevention and observational studies have suggested that tight diabetic control and smoking cessation are effective secondary stroke prevention measures.13 Accordingly, the American Heart Association has published a consensus statement with detailed guidelines and recommendations for secondary prevention of stroke.9 Despite this information, translating research into practice has lagged.7,8,16,17 A study comparing secondary prevention practices of GPs in the United States and United Kingdom revealed that often, they did not follow published recommendations.7 Gibbs et al. found a ‘‘marked under use of antiplatelet and anticoagulant agents’’ in patients with a new diagnosis of stroke or TIA.8 Mead et al.17 found that in patients with atrial fibrillation, only 20% of GPs interviewed would consider instituting anticoagulation. Low treatment rates (ranging from 39% to 62%) have been reported elsewhere.16,18 Interestingly, in comparison, compliance with secondary prevention of myocardial infarction is relatively good.19 The barriers to translating recommendations into practice are not obvious.20 Whether this can be attributed to a lack of awareness of evidence and/or a resistance to changing clinical practice remains unknown. Even at a dedicated stroke follow-up clinic, where 92% of patients had a primary care physician ‘‘with whom the clinic was in touch’’, patient follow-up after two years indicated that compliance with
90
J. Joubert et al. / Journal of Clinical Neuroscience 13 (2006) 84–90
recommendations for secondary prevention was poor.21 A review of possible barriers indicated that physicians will not implement guidelines if unfamiliar with them, if there are no available tools, such as a reminder system, and if they are unconvinced that a guideline will impact on patient outcomes.20 Moreover, lifestyle changes particularly in the elderly, can be difficult to effect. In this study, the lack of sustained motivation due to a lack of awareness by patients and carers of the potential long-term dangers of risk factor neglect was evident by the reasons given by the intervention group drop-outs. This highlights the need to provide ongoing, detailed information about stroke and risk factor management that is physician- and patient-focused.20,21 Nevertheless, an examination of potential interventions, such as point-of-care reminders and academic detailing, were found to be effective.20 Supplying physicians with evidence-based guidelines for hypertension control has resulted in their implementation.21 In another study,22 weekly telephone feedback to physicians of hypertensive control resulted in a mean 6.0 mmHg decrease in patients exposed to this intervention compared to a 2.8 mmHg increase in diastolic blood pressure in the control group. Because opinions vary among physicians as to the significance of risk factors and how aggressively they should be managed, it is essential to widely disseminate evidence-based guidelines. Not only should physicians be involved in the educational process, but also carers and patients. Use of modern information technology should be used to track adherence to guidelines by physicians. In this study, the trend to reduction of depression in the intervention group, according to the screening tool, is noteworthy, but the reasons for this are unclear. The telephone tracking system, with early recognition of depression and feedback to physicians, may account for this but other factors may play a role. In this study, because of the different nature of acute and chronic post-stroke depression4, comparison of the rate of depression in the intervention and control groups was done at 12 months and not baseline. The difference between the two groups is the basis for a more detailed future study. The study demonstrates that the screening tool (PHQ-9) is readily administered by telephone. The positive trends in this study demonstrate that through using a shared care model in the stroke survivor, improved management of vascular risk factors is possible and that screening for depression with feed-back to the GP may result in decreased post-stroke depression at 12 months. As yet it is uncertain what elements of the model are effective in reducing the barriers to risk factor management and the recognition and treatment of post-stroke depression at primary care level. Future studies should address this issue. References 1. Hardie K, Hankey GJ, Jamrozik K, Broadhurst RJ, Anderson C. Ten-Year Survival After First-Ever Stroke in the Perth Community Stroke Study. Stroke 2003;34:1842–6.
2. Hachinski V. Stroke: 30 years of progress. Stroke 2002;33:1–4. 3. Yusuf S. Two decades of progress in preventing vascular disease. The Lancet 2002;360:2–3. 4. Carota A, Bogousslavsky J. Poststroke Depression Ischemic Stroke: Advances in Neurology, 92. Philadelphia: Lippincott Williams & Wilkins; 2003. 5. Everson SA, Ruberts RE, Goldberg DE, et al. Depressive symptoms and increased risk of stroke mortality over a 29-year period. Arch Intern Med 1998;158:1133–8. 6. CLIPP (Consultation and Liaison in Primary-Care Psychiatry) Manual. Victorian Government Department of Human Services, Melbourne; 2000. Available at: http://www.health.vic.gov.au/mentalhealth/publications/clipp. 7. Goldstein LB, Farmer A, Matchar DB. Primary Care Physician – Reported Secondary and Tertiary Stroke Prevention Practices. A Comparison Between the United States and United Kingdom. Stroke 1997;28:746–51. 8. Gibbs RG, Newson R, Lawrenson R, Greenhalgh RM, Davies AH. Diagnosis and Initial Management of Stroke and Transient Ischemic Attack Across UK Health Regions from 1992 to 1996. Experience of a National Primary Care Database. Stroke 2001;32: 1085–90. 9. Wolf PA, Clegett GP, Easton JD, et al. Preventing Ischemic Stroke in Patients With Prior Stroke and Transient Ischemic Attack. A Statement for Healthcare Professionals; From the Stroke Council of the American Heart Association. Stroke 1999;30:1991–4. 10. Spitzer R, Kroenke K, Williams J. Validity and utility of a self-report version of the PRIME-MD: the PHQ Primary Care Study. JAMA 1999;282:1737–44. 11. Harris MF, Davies GP. Shared care through divisions of general practice: Moving forward. Med J Aus 85–8. 12. Stroke Australia Task Force. National Stroke Strategy. Melbourne: National Stroke Foundation; 1997. 13. Hankey GJ. Risk Factor Management to Prevent Stroke. Ischemic Stroke: Advances in Neurology, 92. Philadelphia: Lippincott Williams & Wilkins; 2003. 14. Hankey GJ, Jamrozik J, Broadhurst RH, et al. Long-term risk of first recurrent stroke in the Perth Community Stroke Study. Stroke 1998;29:2491–500. 15. Heart Outcomes Prevention Evaluation (HOPE) Study investigators. Effects of angiotensin converting-enzyme inhibitor, ramipril, on cardiovascular events in high risk-patients. N Engl J Med 2000;342:145–53. 16. Cohen N, Almoznino-Sarafan D, et al. Warfarin for Stroke Prevention Still Underused in Atrial Fibrillation. Stroke 2000;31: 1217–22. 17. Mead GE, Murray H, McCullum CN, O’Neill PA. How do general practitioners manage patients at risk from stroke?. Br J Clin Pract 1996;50:426–30. 18. Sappok T, Faulstich A, Stuckert E, Kruck H, Marx P, Koennecke HC. Compliance with secondary prevention of ischemic stroke a prospective evaluation. Stroke 2001;32:1884–9. 19. Bradley F, Morgan S, Smith H, Mant D. Preventive care for patients following myocardial infarction. Fam Pract 1997;14:220–6. 20. Ornstein SM. Translating Research into Practice Using Electronic Medical Records. The PPRNet-TRIP Project: Primary and Secondary Prevention of Coronary Heart Disease and Stroke. Top Health Inf Manage 2002;22:52–8. 21. Joseph LN, Babikian VL, Allen NC, Winter MR. Risk factor modification in stroke prevention –the experience of a stroke clinic. Stroke 1999;30:16–20. 22. Friedman RH, Kazis LE, Jette A, Smith MB, Stollerman J, Togerson K, Carey K. A telecommunications system for monitoring and counseling patients with hypertension: impact on medication adherence and blood pressure control. Am J Hypertens 1996;9:285– 92. 23. Matchar DB. The value of stroke prevention treatment. Neurology 1998;51:S31–5.