Primary Care Physician–Reported Secondary and Tertiary Stroke Prevention Practices
A Comparison Between the United States and the United Kingdom
Background and Purpose Stroke is a major healthcare problem in both the United States (US) and the United Kingdom (UK). Little comparative data are available concerning how generalist physicians in the two countries approach the management of patients at high risk of stroke.
Methods Contemporaneous surveys of random samples of primary care physician–reported stroke prevention practices were performed in the US and UK from 1993 to 1994. The US National Survey of Physician Practices for the Secondary and Tertiary Prevention of Ischemic Stroke included 254 noninternist primary care physicians. The UK Survey of the Care of Patients With Stroke in General Practice obtained responses from 661 general practitioners. The two surveys used many of the same questions, allowing for direct comparisons of reported stroke prevention practices between American and British physicians.
Results More than 80% of American physicians reported a variety of services (24-hour electrocardiography, echocardiography, brain CT scan, brain MR scan, carotid ultrasonography, cerebral angiography) as readily available. These same services were readily available to less than 10% of physicians in the UK (P<.001 for each comparison). Although physicians in the UK reported prescribing lower doses of aspirin for stroke prevention than physicians in the US, the proportions of physicians using aspirin were not different. In contrast, almost 70% of physicians in the US responded that they always or often anticoagulate patients with nonvalvular atrial fibrillation compared with 7% of British physicians (P<.001). Whereas 70% of American versus 14% of British physicians reported obtaining carotid ultrasound studies in patients with asymptomatic bruits (P<.001), physicians in the UK more commonly reported referring this type of patient to neurologists (46% versus 21%, P<.001). For patients with recent carotid-distribution transient ischemic attack or minor stroke, physicians in the US more commonly reported referral to neurologists (55% versus 45%, P=.022), referral to surgeons (39% versus 19%, P<.001), the performance of carotid ultrasonography (80% versus 11%, P<.001), echocardiography (45% versus 5%, P<.001), 24-hour electrocardiography (49% versus 4%, P<.001), brain CT scan (72% versus 3%, P<.001), and the prescription of anticoagulants (53% versus 4%, P<.001).
Conclusions These data show significant differences in stroke prevention practices as reported by primary care physicians practicing in the US and UK. Some of these differences may be related to the relative availability of specific services in the two countries, potentially leading to overutilization in the US and underutilization in the UK in certain circumstances.
- atrial fibrillation
- carotid endarterectomy
- diagnostic imaging
- Great Britain
- stroke prevention
- United States
Stroke remains a major healthcare problem in both the US and the UK.1 2 In each country, stroke is the third leading cause of death and a major cause of disability among adults, yet stroke is frequently preventable. In the US, both prevention and primary care are receiving increasing emphasis, with managed care systems placing generalist physicians as “gatekeepers,” controlling the patient’s access to both specialists and laboratory testing.3 The organization of medical care in the UK and western Europe has long emphasized the central role of the primary care physician, with patients generally having access to specialists only when the generalist physician has requested consultation.4 5
A previous survey-based study found significant differences in the strategies used for the diagnosis and treatment of neurological disease, including TIA and completed stroke, among British and American neurologists.6 This study was performed in the late 1980s, before the results of randomized trials of carotid endarterectomy were available, and included American neurologists trained in a single residency program. Little comparative data are available concerning how generalist physicians in the US and UK approach the management of patients at high risk of stroke. The US National Survey of Physician Practices for the Secondary and Tertiary Prevention of Ischemic Stroke was stratified by physician specialty and included a random sample of noninternist primary care physicians.7 At the same time the US survey was conducted, a similar study of the care of patients with cerebrovascular disease as reported by generalist physicians was performed in the UK. The two surveys used many of the same questions, allowing for direct comparisons of reported stroke prevention practices between American and British physicians. We sought to determine how generalist physicians in the two countries viewed the availability of stroke prevention services and how they approached the evaluation and treatment of a series of typical patients at elevated stroke risk. Insights gained through these comparisons may have important implications for how physicians in each healthcare system approach the management of patients with cerebrovascular disease and provide a framework for future comparative outcome studies.
US National Survey of Physician Practices for the Secondary and Tertiary Prevention of Ischemic Stroke
Details of the survey methodology and analytic procedures have previously been presented.7 The study population consisted of all physicians, excluding residents and fellows, who were practicing in the US, actively involved in patient care, and listed in the 1993 American Medical Association’s Physician Masterfile under one of several specialties including family practice specialists and general practice physicians (noninternist primary care physicians), internists, neurologists, and surgeons (vascular surgeons and neurosurgeons). A stratified random sample was then chosen by selecting physicians from a list ordered by region of the country for each specialty (each specialty formed a stratum). Survey instruments were mailed to approximately equal numbers of the randomly selected physicians from each specialty (n=2000 including 478 noninternist primary care physicians). Survey data were collected between August 1993 and February 1994. Physicians who did not respond to the initial mailing within 7 weeks were sent a second questionnaire with a reminder note from the Principal Investigator. After an additional 7-week period, nonrespondents were contacted by telephone and either prompted to return their questionnaire or asked to complete the questionnaire by telephone.
The self-administered questionnaire included 23 items and was divided into two distinct parts. The first part of the survey consisted of items that described physician and practice characteristics and the availability of selected procedures and tests. In addition, there were questions concerning basic stroke prevention strategies. These included items that queried (1) the dose of aspirin used for stroke prevention (does not prescribe aspirin, prescribes <325 mg/d, 325 mg/d, 650 mg/d, or ≥975 mg/d) and (2) the frequency of monitoring coagulation profiles after the initial adjustment period in patients taking warfarin (does not prescribe warfarin, once a month, once every other month, once every 3 months, once every 6 months, or once a year or less). The second part of the questionnaire consisted of a series of vignettes that probed the reported use of diagnostic studies and management strategies for patients with defined conditions placing them at elevated stroke risk. These included patients with (1) asymptomatic nonvalvular atrial fibrillation, (2) an asymptomatic carotid bruit, and (3) a recent (within 1 month) carotid-distribution TIA or minor stroke.
UK Survey of the Care of Patients With Stroke in General Practice
The study population consisted of a random sample (1 in 20) of general practitioners in England and Wales who were identified from a list of principals in general practice (n=1399) held by the UK Department of Health. Between April and June 1994, each of the selected practitioners was mailed a questionnaire with a covering letter explaining that the study was designed to describe the way general practitioners manage patients at increased risk of stroke. Two additional mailings were sent to nonresponders.
The self-administered questionnaire included 20 items. As with the US study, the UK survey consisted of items that described physician and practice characteristics and the availability of selected procedures and tests. In addition, there were items that queried (1) the dose of aspirin used for stroke prevention (does not prescribe aspirin, prescribes 75 mg/d, 150 mg/d, 300 mg/d, 600 mg/d, or ≥900 mg/d) and (2) the frequency of monitoring coagulation profiles after the initial adjustment period in patients taking warfarin (does not prescribe warfarin, once a month, once every 2 months, once every 3 months, once every 6 months, or once a year or less). The questionnaire also included a series of vignettes that probed the reported use of the same diagnostic studies and management strategies asked of American physicians for patients over age 60 with (1) asymptomatic atrial fibrillation without heart murmur, (2) an asymptomatic carotid bruit, and (3) a recent (within 1 month) carotid-distribution TIA or minor stroke.
Response Categories and Data Analysis
In both surveys, the availability of the selected services was rated as either being provided by the physician him/herself, readily available, or not readily available. The responses of physicians who indicated that they either performed the procedure themselves or had the service readily available were combined (as “readily available”) and compared with those for whom the service was not readily available. The physicians reported using the indicated tests, procedures, or modalities always, often, sometimes, seldom, or never in a given clinical setting. Because the categories do not represent meaningful or exactly specified differences in clinical practice, and to further simplify the presentation of the analyses, the responses “always or often” and “seldom or never” were combined (“always/often” and “seldom/never”). Because of differences in formulations between the two countries, doses of aspirin were compared as <300 mg/d, 300 to 325 mg/d, and >325 mg/d. The significance of differences between the responses of American and British physicians was determined with χ2 statistics with continuity correction.
Response Rates and Practice Characteristics
Of the 478 US primary care physicians initially mailed surveys, completed responses were obtained from 254 (114 general practitioners and 140 family practice specialists; sample response rate, 53%). In comparison, 661 of 1399 British primary care physicians (47%) returned a completed questionnaire.
The American noninternist primary care physicians who responded to the survey were a mean of 49 years of age; 53% were in solo practice, 26% practiced in small groups, 12% in large groups, and 2% in academic medical centers. Thirty-seven percent had their practices in large metropolitan areas, 36% in small metropolitan areas, and 27% in rural locations. Twenty-seven percent practiced in the central portion of the country, 17% in the northeast, 36% in the south, and 20% in the west. Twenty percent of US generalist physicians had practices in which more than 50% of their patients had risk factors for stroke.
The British physicians responding to the survey had been qualified as doctors for a median of 18 years. These physicians cared for a median of 4 patients with acute stroke during the preceding 6 months. Forty-one percent of the physicians were in “training practices” in which they or one of their colleagues acts as the mentor for new physicians during the last portion of their formal post–medical school training.
Perceived Service Availability and Common Practices
Fig 1⇓ presents the perceived availabilities of the indicated services as reported by American and British generalist physicians. All of the differences are significant (P<.001). Although the difference in the reported availability of ECG is small, the differences in the perceived availabilities of the other services are dramatic. More than 80% of US noninternist primary care physicians reported ready access to 24-hour ECG, echocardiography, brain CT scan, brain MR scan, carotid ultrasonography, and cerebral angiography. These same procedures were viewed as being readily available to less than 10% of generalist physicians in the UK.
The majority of physicians in both countries reported that they prescribed one adult aspirin tablet (300 to 325 mg) or less daily. Proportionally more British physicians (93%) than American physicians (37%) responded that they prescribed less than one adult tablet daily (P<.0001).
Physicians practicing in the US reported that they performed coagulation profiles in stable anticoagulated patients more frequently than British physicians. Whereas 82% of American generalist physicians who use warfarin reported that they monitor their patients at least monthly, only 58% of British physicians who use warfarin reported that they monitor their patients this frequently (P=.0001). Eighteen percent of American physicians perform coagulation studies less frequently compared with 42% of British physicians.
Nonvalvular Atrial Fibrillation
American physicians were asked how frequently they would refer to a cardiologist, obtain a 24-hour ECG, prescribe aspirin, or prescribe anticoagulation for an asymptomatic patient with nonvalvular atrial fibrillation. British physicians were asked how frequently they would perform the same actions for a patient with atrial fibrillation and no heart murmur. The American physicians were not given the age of the patient; British physicians were asked about their practice for patients over 60 years of age. Other comorbid conditions such as hypertension and congestive heart failure were not specified. A higher proportion of physicians practicing in the US responded that they always or often refer their patients to a cardiologist than physicians practicing in the UK, but the difference was not large (Fig 2⇓, top left panel). Relatively more American physicians obtain a 24-hour ECG (Fig 2⇓, top right panel) and prescribe anticoagulants (Fig 2⇓, bottom right panel). There was no difference in the proportions of physicians prescribing aspirin.
Asymptomatic Carotid Bruit
Fig 3⇓ compares the responses of American and British physicians for an asymptomatic patient with a carotid bruit. Relatively fewer physicians practicing in the US reported that they always or often refer such patients to neurologists (Fig 3⇓, top left panel), but substantially more obtain carotid ultrasonography (Fig 3⇓, bottom left panel). There was no difference in the proportions of physicians responding that they always or often refer these patients to surgeons (Fig 3⇓, top right panel), and most physicians in either country reported that they would prescribe aspirin (Fig 3⇓, bottom right panel).
Recent (Within 1 Month) TIA or Minor Carotid-Distribution Stroke
In contrast to patients with asymptomatic bruit, a relatively higher proportion of American generalist physicians reported that they always or often refer patients with recent TIA or minor stroke to neurologists (Fig 4⇓, top left panel) and surgeons (Fig 4⇓, top right panel). Substantially more American than British physicians obtain carotid ultrasonography (Fig 4⇓, middle left panel), echocardiography (Fig 4⇓, middle right panel), 24-hour ECG (Fig 4⇓, bottom left panel), and brain CT scan (Fig 4⇓, bottom right panel). Most (>80%) of both American and British physicians prescribe aspirin for these patients; however, substantially more physicians practicing in the US responded that they always or often prescribe an anticoagulant (53% of physicians in the US versus 4% of physicians in the UK; P<.001).
Our study is unique in that it compares the results of contemporaneous surveys of the reported stroke prevention practices of random samples of generalist physicians practicing in the US and the UK. The two surveys shared many virtually identically worded questions. To further enhance comparability, only the responses of American noninternist primary care physicians were included in the analyses. These surveys reveal significant differences in both the perceived availability of relevant diagnostic modalities and in the management of patients at elevated risk of stroke as reported by generalist physicians in the two countries. There are differences in reported practice in both situations in which clinical trial data are available to help guide management and in which clinical trial data are not available or are not convincing.
Many of the differences in practice revealed by these surveys may be directly related to differences in the organization of health care between the two countries. British generalist physicians perceived many tests as being not readily available. This relative lack of availability undoubtedly has an impact on how British physicians approach the management of patients at risk for stroke compared with physicians practicing in the US. For example, British physicians report limited access to carotid ultrasonography, and therefore they responded that they use this test less frequently than American physicians for patients with asymptomatic bruit or recent carotid-distribution TIA/minor stroke. If a British physician believes that carotid ultrasonography is warranted, one strategy would be to refer the patient to a hospital physician such as a neurologist who could obtain the test. In fact, more generalist physicians in the UK reported that they would refer a patient with an asymptomatic bruit to a neurologist than generalist physicians in the US. However, relatively fewer British physicians reported referring patients with a recent carotid-distribution TIA or minor stroke to either hospital physicians (eg, neurologists) or surgeons. The reason for this latter difference is uncertain. These surveys were conducted approximately 2 years after the results were available from North American and European randomized trials indicating that endarterectomy was efficacious in selected symptomatic patients with high-grade stenosis of the carotid artery.8 9 10 Since British physicians infrequently reported obtaining carotid ultrasonography, these data suggest that despite the results of these clinical trials, many physicians in the UK were still not considering patients for carotid endarterectomy. This could be due to a lack of awareness of the trial results, concern about surgical complication rates, the physicians’ attitudes concerning carotid endarterectomy, or barriers to obtaining services.
The difference in the use of anticoagulants by physicians in the UK and the US is also striking. At the time the surveys were performed, the results of several randomized trials concerning the use of warfarin in patients with nonvalvular atrial fibrillation had been published11 12 13 or had recently become available.14 15 Although clinical details that could influence the use of anticoagulants in individual patients with atrial fibrillation were not reported, approximately 75% of American physicians reported always or often prescribing anticoagulants, whereas approximately 75% of British physicians responded that they seldom or never prescribe this medication. Since physicians in both countries have access to the same medical literature, unless there are dramatic differences in the respective populations of patients with atrial fibrillation, this difference in the reported frequency of use of anticoagulants is unexplained. However, concern about complication rates and difficulties in providing continuing monitoring of warfarin treatment may be relevant factors.16 17
Approximately half of American primary care physicians reported that they always or often obtain a 24-hour ECG in asymptomatic patients with nonvalvular atrial fibrillation compared with less than 10% of generalist physicians in the UK. Again, this relative lack of use of 24-hour ECG by British physicians may reflect the reported lack of ready availability of the test, or the ready availability of the test to American physicians may be resulting in its overuse. The role of 24-hour ECG in the management of patients with asymptomatic atrial fibrillation, particularly in regard to decisions related to the use of antithrombotic therapy, is uncertain.18 Therefore, it may be that US physicians are obtaining the study too frequently.
Another unexplained difference is the relatively higher proportion of US generalist physicians who reported prescribing anticoagulants for patients with recent carotid-distribution TIA or minor stroke. In a previous report, we noted that noninternist primary care physicians in the US more frequently reported using anticoagulants for patients without surgical contraindications who had a recent carotid-distribution TIA or minor stroke and more than 70% stenosis of the ipsilateral extracranial carotid artery compared with neurologists.19 Nearly one in four noninternist primary care physicians responded that they would seldom or never use carotid endarterectomy in this setting.20 Thirty-nine percent of the US noninternist primary care physicians who responded that they would seldom or never use carotid endarterectomy also reported that they always or often use anticoagulants for these patients.19 This implied that an unproved therapy (anticoagulation) may be used by some US physicians in place of a therapy of proven value (carotid endarterectomy) in this particular clinical scenario. Thus, although generalist physicians in the UK may be using anticoagulants less frequently than might be expected for patients with atrial fibrillation, US primary care physicians may be using anticoagulants more frequently than expected for patients with recent TIA or minor stroke.
The optimal dose of aspirin for stroke prevention remains a subject of considerable controversy.21 22 23 24 Despite this controversy between proponents of “high-dose” and “low-dose” aspirin, we previously reported that most American physicians, regardless of specialty, prescribe 325 mg/d, with only 4% prescribing doses of ≥650 mg/d.7 A higher proportion of physicians in the UK apparently use even lower doses of aspirin for stroke prevention than physicians in the US.
There are several limitations to these analyses that should be stressed. Although the response rates for these surveys (53% in the US and 47% in the UK) compare favorably to other nationally based physician surveys in these two countries,25 26 the precise reasons that individual physicians chose not to respond are unknown, and the possibility of some bias cannot be completely excluded. Second, we surveyed physician practices as reported by the physicians themselves. Physician practices as reported in a survey may not reflect their actual practices.27 Third, although the use of simple clinical vignettes to probe physician practices is useful for improving the comparability of the results, these simple scenarios lack clinical details that are often important in actual clinical decision making (eg, the previous discussion of anticoagulation for atrial fibrillation). Fourth, we could not control for possible population differences between the UK and the US that could potentially affect clinical care. Although age-adjusted incidence rates for stroke are similar in the two countries,28 age-adjusted stroke mortality rates are somewhat higher in the UK.29 Fifth, we did not include the responses of American internists in these comparative analyses. Although American internists frequently provide primary care, their training and the scope of their practices are dissimilar from generalist physicians in the UK. Finally, we have no data to address whether the differences in practice suggested by our surveys actually affect health-related outcomes (such as stroke-related mortality) or quality of life.
Despite limitations, these data begin to provide insights into how the treatment of patients at elevated risk of stroke differs in these distinctive healthcare systems. How these differences between the US and UK affect outcome and how stroke prevention practices in the US will change with the continuing development of managed care are important questions that will require further study.
Selected Abbreviations and Acronyms
|TIA||=||transient ischemic attack|
The US National Survey of Physician Practices for the Secondary and Tertiary Prevention of Ischemic Stroke was performed as part of the Stroke Prevention Patient Outcomes Research Team (PORT) and was funded through contract 282-91-0028 from the US Agency for Health Care Policy Research. The UK Survey of the Care of Patients With Stroke in General Practice was funded by the Anglia and Oxford Regional Health Authority Health Promotion Fund. The authors wish to thank Drs Arthur J. Bonito, Gregory Samsa, Pamela W. Duncan, Gordon H. DeFriese, Eugene Z. Oddone, and John E. Paul for their contributions to the design and/or conduct of the US National Survey of Physician Practices for the Secondary and Tertiary Prevention of Ischemic Stroke and Drs Simon Winner, Jonathan Michaels, Simon Street, and Diana Harwood for their help in conducting the UK Survey of the Care of Patients With Stroke in General Practice.
- Received November 19, 1996.
- Revision received December 26, 1996.
- Accepted December 30, 1996.
- Copyright © 1997 by American Heart Association
American Heart Association. 1993 Stroke Facts. Dallas, Tex: American Heart Association; 1992.
Wolfe CDA, Burney PGJ. Is stroke mortality on the decline in England? Am J Epidemiol. 1992;136:558-565.
Burkholder JS, Dudjak LA. The midlevel practitioner role: one medical center’s experience. AACN Clin Issues. 1994;5:369-403.
Joint Working Party of the Welsh Council of the Royal College of General Practitioners and the Welsh General Medical Services Committee. Patient care and the general practitioner. BMJ. 1994;309:1144-1147.
Gervas J, Perez Fernandez M, Starfield BH. Primary care, financing and gatekeeping in western Europe. Fam Pract. 1994;11:307-317.
Goldstein LB, Bonito AJ, Matchar DB, Duncan PW, DeFriese GH, Oddone EZ, Paul JE, Akin DR, Samsa GP. US National Survey of Physician Practices for the Secondary and Tertiary Prevention of Stroke: design, service availability, and common practices. Stroke. 1995;26:1607-1615.
Stroke Prevention in Atrial Fibrillation Investigators. Stroke Prevention in Atrial Fibrillation Study. Circulation. 1991;84:527-539.
Boston Area Anticoagulation for Atrial Fibrillation Investigators. The effect of low-dose warfarin on the risk of stroke in patients with nonrheumatic atrial fibrillation. N Engl J Med. 1990;22:1505-1558.
Ezekowitz MD, Bridgers SL, James KE, Carliner NH, Colling CL, Gornick CC, Krause-Steinrauf H, Kurtzke JF, Nazarian SM, Radford MJ, Rickles FR, Shabetai R, Deykin D. Warfarin in the prevention of stroke associated with nonrheumatic atrial fibrillation. N Engl J Med. 1992;327:1406-1412.
Sweeney KG, Pereira Gray DJ, Steele RJ, Evans PH. Commentary: caution needed in introducing warfarin treatment. BMJ. 1995;311:560-561.
Shakespeare J. Anticoagulation in patients with atrial fibrillation: GPs struggle to meet demand. BMJ. 1994;308:415.
Page RL, Wilkinson WE, Clair WK, McCarthy EA, Pritchett EL. Asymptomatic arrhythmias in patients with symptomatic paroxysmal atrial fibrillation and paroxysmal supraventricular tachycardia. Circulation. 1994;89:224-227.
Goldstein LB, Bonito AJ, Matchar DB, Duncan PW, Samsa GP. US National Survey of Physician Practices for the Secondary and Tertiary Prevention of Ischemic Stroke: medical therapy in patients with carotid artery stenosis. Stroke. 1996;27:1473-1478.
Goldstein LB, Bonito AJ, Matchar DB, Duncan PW, Samsa GP. US National Survey of Physician Practices for the Secondary and Tertiary Prevention of Ischemic Stroke: carotid endarterectomy. Stroke. 1996;27:801-806.
Barnett HJM. Aspirin in stroke prevention: an overview. Stroke. 1990;21(suppl IV):IV-40-IV-43.
Dyken ML, Barnett HJM, Easton JD, Fields WS, Fuster V, Hachinski V, Norris JW, Sherman DG. Low-dose aspirin and stroke: ‘it ain’t necessarily so.’ Stroke. 1992;23:1395-1399.
Thran SL, Berk ML. Surveys of physicians: an overview. In: Proceedings of the International Conference on Establishment Surveys. Alexandria, Va: American Statistical Association; 1994:83-92.
McAvoy BR, Kaner EFS. General practice postal surveys: a questionnaire too far? BMJ. 1996;313:732-733.
Dennis M, Bamford J, Sandercock P, Warlow C. Prognosis of transient ischemic attacks in the Oxfordshire Community Stroke Project. Stroke. 1990;21:848-853.
Bonita R, Stewart A, Beaglehole R. International trends in stroke mortality: 1970-1985. Stroke. 1990;21:989-992.