Attitudes and Current Practice of Primary Care Physicians in Acute Stroke Management
Background and Purpose— Stroke patients often report that primary care physicians (PCPs) are their first medical contact after onset of symptoms. We studied PCP attitudes and current practice in early management of suspected stroke patients.
Methods— A cross-sectional survey was conducted among 714 general practitioners, internists, and neurologists providing acute primary care for stroke patients in 4 different regions in Germany. PCP attitudes and practices were assessed with standardized questionnaires and case vignettes presenting suspected stroke patients contacting PCPs either by phone or in practice. Factors influencing the decision of the PCPs to admit patients with clear stroke symptoms as medical emergency to hospital were assessed using multivariate analysis.
Results— In total, 395 PCPs participated in the study (55.3%). Most PCPs agreed that stroke (94.7%) and transient ischemic attack (84.8%) were medical emergencies. In case vignettes, admission to hospital as medical emergency was preferred management after first contact to patients with clear stroke symptoms by phone or in practice (68.9% and 65.6%, respectively). Outpatient clarification was the preferred option of PCPs in patients with unclear stroke symptoms contacting PCPs by phone or in practice (54.7% and 75.5%, respectively) and in transient ischemic attack patients (50.9%). Working as general practitioner (odds ratio, 0.3; 95% confidence interval, 0.2 to 0.6) and practice location outside metropolitan area (P=0.002) independently decreased probability of admitting suspected stroke patients as medical emergency when first contact to PCPs was by phone. PCP agreement that all stroke and transient ischemic attack patients must be admitted to hospital increased probability for early hospitalization of stroke and transient ischemic attack patients when first contact was in practice (odds ratio, 2.0; 95% confidence interval, 1.2 to 3.3, and odds ratio, 1.8; 95% confidence interval, 1.1 to 2.8, respectively).
Conclusions— Stroke and transient ischemic attack were well recognized as medical emergencies by PCPs in our study. However, only two-thirds of PCP would immediately admit stroke suspected patients with clear symptoms to hospital as medical emergency.
There is convincing evidence that specific therapeutic and management strategies reduce stroke-related mortality and morbidity.1–3 The majority of these therapies are more effective if administered as soon as possible after stroke onset. Some therapies even require the start of the treatment within a given time period.4,5 Therefore, early hospitalization of suspected stroke patients for immediate diagnostic intervention as well as adequate treatment and management is crucial to reduce burden of stroke for society.2,4 A major demand for early admission of stroke patients to hospital is an effective prehospital management.3,6 In observational studies among hospitalized stroke patients, primary care physicians (PCPs) are the first medical contact person after onset of stroke symptoms in 25% to 45%.7–11 Therefore, it is of great importance that PCPs correctly recognize stroke signs and symptoms as, for example, potential time delays minimize the option to initiate tissue plasminogen activator treatment.5 According to current recommendations, PCPs should immediately initiate referral of a suspected stroke patient to an adequate hospital.3 However, there are limited data available on attitudes and current referral practice of PCPs in the very early management of stroke suspected patients.
Therefore, we conducted a cross-sectional survey among PCPs on their perception of stroke and transient ischemic attack (TIA) as medical emergencies as well as their current practice in referral of patients suspected of stroke or TIA.
Sample and Setting
The study was conducted among general practitioners, internists, and neurologists providing acute primary care for stroke patients in 4 geographically different regions of North Rhine-Westphalia, Germany. As study areas, 2 mainly urban (Muenster, Dortmund) and 2 mainly rural regions (Arnsberg, Luedenscheid) were chosen. Within the urban areas, the study was performed in existing networks of PCPs covering ≈60% of all PCPs in these areas. In the rural areas, a random sample of 50% of all PCPs was drawn from the official list of the Westphalian Board of Physicians. In total, 714 PCPs in the 4 regions were invited to participate. The study was performed from September 2005 to March 2006.
Standardized questionnaires were developed. For investigating attitudes of PCPs concerning acute management of stroke and TIA patients, participants were asked to indicate their level of agreement with nine general statements on a 4-item Likert scale.12 PCP current practice of early management of stroke suspected patients were investigated by presenting 5 case scenarios. Case scenarios described in detail age, sex, comorbidities, and clinical signs and neurological symptoms of a specific patient. Because the first contact of patients with suspected stroke to the PCP will occur either on the phone or in person in the office, case vignettes included both settings. Two scenarios asked about PCP management of suspected stroke patients when first contact to patient was on the phone, one with clear and one with unclear stroke symptoms. The 3 remaining scenarios investigated PCP management of suspected stroke patients after first contact to patients in PCP practice, one with clear, one with unclear stroke symptoms, and one with clear signs of a TIA 1 night before. For each case scenario, PCPs were asked to indicate their initial reaction in the specific situation from predefined options. These response options were combined into outpatient clarification (including, eg, further laboratory or diagnostic tests by the PCP or referral of the patient to another physician in an outpatient setting), admission to hospital as a medical emergency, admission to hospital not as a medical emergency, and other actions. The case scenarios were constructed by the study group and critically reviewed by an experienced general practitioner and a clinical neurologist, both not involved in the study. A copy of the case vignettes is available on request from the study group. Finally, PCP preferences for selecting a suitable hospital for admitting identified stroke patients were assessed. Demographic variables of the PCP and the PCP practice characteristics were documented. The questionnaire was adapted in accordance with a previous Australian study.13 A pretest was performed with 25 PCPs for investigating acceptance of the questionnaire and correct understanding of the case vignettes.
Questionnaires were self-administered. Each PCP received a standardized mailed questionnaire with a personal cover letter and a stamped self-addressed envelope. If the PCP did not answer, 2 postal reminders were sent out within 3-week intervals. If still no questionnaire returned, the PCP was contacted by phone.
The t test was used to test differences in continuous variables, and the χ2 test was used for those in proportions. Multivariate logistic regression analysis was performed to identify factors independently influencing the decision of the PCP to admit a stroke or a TIA patient with clear symptoms to hospital as medical emergency in relation to PCP demographic and practice characteristics. Odds ratio and corresponding 95% confidence intervals were calculated by running different regression models for suspected stroke patients with first contact by phone, first contact in practice, and for TIA patient with first contact in practice. The impact of the following variables on probability of admission to hospital as a medical emergency were investigated: sex, work experience, discipline of PCP (general practitioner versus internist and neurologist), working in group practice, location of practice, mean number of acute stroke patients treated per year, mean distance of emergency medical services (EMS) to next hospital, agreement that every stroke or TIA patient is a medical emergency and agreement that all stroke or TIA patient must be admitted to hospital (agree or rather agree versus disagree or rather disagree). Analyses were controlled for study area. Statistical significance of the resulting coefficients was tested using the likelihood ratio test. Variables in multivariate analyses were eliminated using backward-elimination procedure. Analyses were restricted to data without missing values. All tests were 2-tailed and statistical significance was determined at an α level of 0.05. Statistical analyses were performed with SPSS 12.0 software packet.
The Ethics Committee of the Westphalian Board of Physicians and the Medical Faculty of the University of Muenster approved this study.
Of 714 eligible PCPs, 395 completed the questionnaire (55.3%). From the remaining 319 PCPs, 31.4% indicated lack of time, 21.9% general refusal of scientific studies, 16.6% lack of interest in the topic, 4.7% other reasons, and 0.6% absentee of financial gratification as reasons for not participating; 16.6% were not reached by phone after several attempts and 8.2% gave agreement for participation by telephone, but no questionnaire returned.
The mean (standard deviation) age of PCPs was 49.5 (8.2) years, and 23.3% were female. Half of PCPs (50.3%) indicated working in a group practice; the majority of respondents were general practitioners (72.7%). Further characteristics are shown in Table 1. Nearly half of the PCPs treated ≤10 patients with acute stroke per year; the mean distance of an EMS to the next hospital was ≤20 minutes for 97.5%. Despite a greater proportion of practices located in metropolitan areas and a shorter mean distance of EMS to the next hospital in the urban samples, no other statistical significant differences were observed between the urban and the rural PCP samples.
Perception of Stroke and TIA as Medical Emergencies
PCP views of stroke and TIA as medical emergencies are presented in Table 2. Most of PCP perceived every stroke (95%) and every TIA (85%) as medical emergency. However, 39% of PCP agreed or rather agreed that TIA patients could receive outpatient diagnosis and treatment. We further asked about PCP reasons for not admitting stroke patients to hospital (multiple answers possible). If PCPs reported to treat stroke patients in an outpatient setting, main reasons for not hospitalizing stroke patients were severe comorbidities of stroke patient (36.7%), lack of therapeutic consequences of hospital admission for the patient (32.9%), or patients’ refusal of hospital admission (23.8%).
Current Practice of Acute Stroke Management
Table 3 summarizes PCP responses to the 5 case vignettes indicating current practice of acute management of patients with suspected stroke and TIA 1 night before. Admission to hospital as medical emergency was the preferred management option of PCP after first contact to a patient with clear stroke symptoms by phone (69%) and in practice (66%). Most PCP voted for outpatient clarification in patients with unclear stroke symptoms presenting by phone (55%) or in practice (76%) and in TIA patients (51%). In multivariate analysis, a patient with clear stroke symptoms who contacted the PCP by phone was less likely to be admitted to hospital as a medical emergency when location of the doctor’s office was outside a metropolitan area (P=0.002) and the respondent was working as a general practitioner (odds ratio, 0.3; 95% confidence interval, 0.2 to 0.6; Table 4). Agreement of PCPs to the statement that all stroke patients and all patients with TIA must be admitted to hospital showed a significant increase in the probability of admitting a suspected stroke patient (odds ratio, 2.0; 95% confidence interval, 1.2 to 3.3) and a TIA patient (odds ratio, 1.8, 95% confidence interval, 1.1 to 2.8) to hospital as an emergency when first contact to this patient was in office.
Criteria for Selecting a Hospital for Acute Stroke Patients
PCP criteria for selecting a suitable hospital for admitting identified stroke patients who had first medical contact to PCPs are shown in the Figure. Approximately 45% of PCP stated that they would admit all stroke patients to a specialized stroke center; only 5% did not want to influence the selection process.
Our cross-sectional study provides the first data to our knowledge on current referral patterns of PCP in the very early management of patients with suspected stroke or TIA after first contact to PCPs. The majority of PCPs in our study agreed that acute stroke and TIA were medical emergencies. However, only two-thirds of PCPs would immediately admit stroke suspected patients with clear symptoms to hospital as a medical emergency. Case vignettes also indicated that patients with unclear stroke symptoms and TIA patients were more likely to be initially diagnosed and treated as outpatients. Factors independently influencing early management of stroke suspected patients differ depending on whether first contact to the PCP was by phone or in practice. Discipline of the PCP and localization of practice influenced likelihood of patients to be admitted as a medical emergency when first contact to the PCP was on the phone. In patients contacting PCPs in practice, agreement of PCPs that every stroke or TIA must be admitted to hospital increased the probability of admitting these patients as medical emergency immediately.
The percentage of PCPs in our study who agreed or rather agreed that every stroke is a medical emergency was comparable to a previous Australian study among general practitioners (95% versus 97%, respectively).13 In addition, in both studies similar numbers of PCPs thought that all stroke patients should be admitted to hospital (75% versus 74%, respectively).13 However, in our study only two-thirds of PCPs would admit a patient with clear stroke symptoms to hospital as medical emergency. This gap between awareness and concrete action might be attributable to the fact that currently no widely accepted guidelines for PCPs are available for the correct identification and management of suspected stroke in patients in the very early stage. The low rate of PCPs who would admit a patient with a TIA to hospital as a medical emergency in our study might reflect a tendency to neglect an acute TIA as serious disease when symptoms are no longer present at first contact. However, there is strong evidence for a high risk of stroke after TIA of 10% at 7 days, and up to 20% at 1 month.14–17 It is worth remarking that ≈50% of strokes occur within the first 48 hours after TIA onset.16 A considerable variation in the management of patients with TIA was also demonstrated in a previous study that presented case vignettes of TIA patients to a convenience sample of practicing neurologists.18 One further study analyzing medical records of patients with stroke and TIA in PCP practices showed that 10% of patients with stroke but only 2% of patients with TIA were admitted to hospital on the day of the index visit while most patients were completely managed as outpatients.19
In our study, general practitioners were less likely to admit a stroke patient to hospital as a medical emergency when first contact to the patient was by phone compared with internists or neurologists. Previous studies reported delayed hospital admission of suspected stroke patients when first contact was to general practitioners,7–9,11,20 but no differentiation was made between phone and personal contact. General practitioners take the main responsibility for the referral of their patients to specialized outpatient care or to hospital.21 Therefore, general practitioners might prefer involvement in diagnosis and treatment of their patients to a greater extent22 and might also prefer examining the patients personally before involving a specialist or admitting the patient to hospital. There was also a decreased likelihood that a suspected stroke patient who contacted the PCP by phone was admitted to hospital as a medical emergency when the location of doctor’s practice was outside a metropolitan area. Because of broader catchment areas of EMS outside metropolitan areas, arrival times of EMS are longer in rural areas compared with urban areas.23 Thus, PCPs working outside a metropolitan area might perceive that there is no substantial time gain for immediately calling an EMS compared with an outpatient clarification of symptoms. However, mean distance of EMS to next hospital showed no significant influence on the attitude of a PCP to admit a suspected stroke patient to hospital as a medical emergency. We found that a patient with suspected stroke or TIA with first contact in office was managed more often as a medical emergency when PCP agreed with the statement that every stroke and every TIA patient must be admitted to hospital. This is in accordance with recommendations of current guidelines emphasizing that health care professionals must consider stroke as a medical emergency to ensure immediate hospital admission of suspected stroke patients.2,3,24 The likelihood of being admitted to hospital as a medical emergency was lower when the patient was seen in practice compared with first contact by phone. This result might reflect that if PCPs are unable to verify the described symptoms clearly on the phone, they tend to act more cautious and admit patients to hospital.
Our study has several limitations. First, results were based on self-reported stroke awareness and referral patterns of PCP. We were not able to determine whether PCPs really expressed their actual practice when being approached by a stroke or a TIA patient. Second, we cannot exclude that responders had different awareness of acute stroke and TIA than nonresponders. Approximately 40% of nonattending PCP indicated refusal of scientific studies or lack of interest in the topic as reason for not participating. Thus, this subgroup might be less aware of stroke and TIA being medical emergencies. Third, because of differences between health care systems, our results might not be completely transferable to other countries. Thus, further research is needed to confirm our findings.
Different factors influenced the decision of PCPs to hospitalize patients with clear symptoms as medical emergencies depending on whether first contact to the PCP was by phone or in practice. Therefore, future studies have to consider carefully potential variations in ways suspected stroke patients first contact their PCPs. A positive association was found between the agreement that all stroke and TIA patients must be admitted to hospital and the management of patients with suspected stroke and TIA as medical emergencies. Thus, programs increasing awareness of PCPs to perceive stroke as medical emergency might be effective.
The authors thank the Westphalian Board of Physicians for supporting our study. In addition, the authors express their honest gratitude to all primary care physicians who participated in our study.
Sources of Funding
This study was funded by an unrestricted grant of the German Stroke Foundation, Guetersloh.
The study was presented in part as an oral presentation at the 15th European Stroke Conference in Brussels, May 16–19, 2006.
- Received July 13, 2006.
- Revision received October 20, 2006.
- Accepted November 21, 2006.
A systems approach to immediate evaluation and management of hyperacute stroke. Experience at eight centers and implications for community practice and patient care. The National Institute of Neurological Disorders and Stroke (NINDS) rt-PA Stroke Study Group. Stroke. 1997; 28: 1530–1540.
Adams H, Adams R, Del Zoppo G, Goldstein LB. Guidelines for the early management of patients with ischemic stroke: 2005 guidelines update a scientific statement from the Stroke Council of the American Heart Association/American Stroke Association. Stroke. 2005; 36: 916–923.
Hack W, Kaste M, Bogousslavsky J, Brainin M, Chamorro A, Lees K, Leys D, Kwiecinski H, Toni P, Langhorne P, Diener C, Hennerici M, Ferro J, Sivenius J, Gunnar N, Bath P, Olsen TS, Gugging M. European stroke initiative recommendations for stroke management-update 2003. Cerebrovasc Dis. 2003; 16: 311–337.
Adams HP, Adams RJ, Brott T, del Zoppo GJ, Furlan A, Goldstein LB, Grubb RL, Higashida R, Kidwell C, Kwiatkowski TG, Marler JR, Hademenos GJ. Guidelines for the early management of patients with ischemic stroke: A scientific statement from the Stroke Council of the American Stroke Association. Stroke. 2003; 34: 1056–1083.
Hacke W, Donnan G, Fieschi C, Kaste M, von Kummer R, Broderick JP, Brott T, Frankel M, Grotta JC, Haley EC Jr, Kwiatkowski T, Levine SR, Lewandowski C, Lu M, Lyden P, Marler JR, Patel S, Tilley BC, Albers G, Bluhmki E, Wilhelm M, Hamilton S. Association of outcome with early stroke treatment: pooled analysis of ATLANTIS, ECASS, and NINDS rt-PA sroke trials. Lancet. 2004; 363: 768–774.
Derex L, Adeleine P, Nighoghossian N, Honnorat J, Trouillas P. Factors influencing early admission in a French stroke unit. Stroke. 2002; 33: 153–159.
Fogelholm R, Murros K, Rissanen A, Ilmavirta M. Factors delaying hospital admission after acute stroke. Stroke. 1996; 27: 398–400.
Harraf F, Sharma AK, Brown MM, Lees KR, Vass RI, Kalra L. A multicentre observational study of presentation and early assessment of acute stroke. BMJ. 2002; 325: 17–20.
Wester P, Radberg J, Lundgren B, Peltonen M. Factors associated with delayed admission to hospital and in-hospital delays in acute stroke and TIA: a prospective, multicenter study. Stroke. 1999; 30: 40–48.
Fink A. How to ask survey questions. The survey kit, 2nd ed. Thousand Oaks, London, India: Sage Publications; 2003.
Middleton S, Sharpe D, Harris J, Corbett A, Lusby R, Ward J. Case scenarios to assess Australian general practitioners’ understanding of stroke diagnosis, management, and prevention. Stroke. 2003; 34: 2681–2686.
Coull AJ, Lovett JK, Rothwell PM. Population based study of early risk of stroke after transient ischaemic attack or minor stroke: Implications for public education and organisation of services. BMJ. 2004; 328: 326–328.
Daffertshofer M, Mielke O, Pullwitt A, Felsenstein M, Hennerici M. Transient ischemic attacks are moretThan “Ministrokes.” Stroke. 2004; 35: 2453–2458.
Kleindorfer D, Panagos P, Pancioli A, Khoury J, Kissela B, Woo D, Schneider A, Alwell K, Jauch E, Miller R, Moomaw C, Shukla R, Broderick JP. Incidence and short-term prognosis of transient ischemic attack in a population-based study. Stroke. 2005; 36: 720–723.
Harbison J, Hossain O, Jenkinson D, Davis J, Louw SJ, Ford GA. Diagnostic accuracy of stroke referrals from primary care, emergency room physicians, and ambulance staff using the face arm speech test. Stroke. 2003; 34: 71–76.
RettG NRW. Rettungsgesetz NRW §17 (online). Available at http://www.agn-online.de/rettg_nrw.html. 1992.