Background and Purpose: Stroke patients appear to have improved outcomes when cared for by neurologists, but the mechanism by which improved outcome is achieved is unclear. This study compares 30-day cause-specif ic rehospitalization, 30-day mortality, and specific processes of care for patients treated by a neurologist only, a generalist only, a neurologist and a generalist (i.e., collaborative care), or by another specialist during the index hospitalization.Methods: This study uses Cox regression to analyze claims and enrollment data from 44,099 Medicare beneficiaries 65 years of age and older and discharged with acute ischemic stroke from 1998 to 2000 in 11 US metropolitan regions. Results: Patients seen by neurologists had more severe strokes than patients seen by generalists, though patients seen by generalists had more comorbidities. Patients seen by neurologists (alone or collaboratively) had a 10 and 16% lower risk of 30-day mortality, respectively. Patients seen by a neurologist only had a 12% lower risk of rehospitalization for infections and aspiration pneumonitis. In contrast, patients seen by neurologists had a higher risk of rehospitalization for atherosclerotic (cardiovascular and non-acute cerebrovascular) disease. Patients seen by neurologists were more likely to be discharged to inpatient rehabilitation, had longer lengths of stay, and were more likely to receive warfarin after discharge. Conclusions: Results support the hypothesis that neurologists improve outcomes specifically by reducing the potential for aspiration (through increased swallowing evaluations) or by improving functioning (through use of rehabilitation therapy). Future studies should continue to examine the mechanisms by which neurologists may achieve better outcomes in stroke care.

Stroke patients appear to have improved mortality and reduced disability when cared for by neurologists during the inpatient hospitalization [1,2,3,4], but neurologists also perform more tests [1, 3] and are significantly more expensive [1]. The added value of neurology care for stroke patients continues to be hotly debated [5, 6] and is fueled by the relative lack of studies identifying specific mechanisms through which care by a neurologist can be shown to improve outcomes such as mortali ty [7].

As pressure to contain the cost of health care increases, it is critical to begin understanding the organizational arrangements and mechanisms by which neurologists (alone or in collaboration with generalists) may achieve improved outcomes in stroke care [8]. Neurologists do have a different practice style when compared to generalist physicians [1, 3], including increased use of antiplatelets [3], occupational and speech therapy [3], anticoagulants [1], and discharge to inpatient rehabilitation facilities [1] as well as increased diagnostic testing [1, 3], and longer length of stay [1]. The goal of this study is to compare 30-day cause-specific rehospitalization, 30-day mortality, and specific processes of care for older stroke patients treated by a neurologist only, a generalist only, both a neurologist and a generalist, or by another specialist during the index hospitalization.

We identified 44,099 Medicare beneficiaries 65 years of age and older and discharged with acute ischemic stroke from 1998 to 2000 in 11 metropolitan regions based on an ICD-9 diagnosis code of 434 or 436 in the first position on the discharge diagnosis list. We obtained administrative data from the Centers for Medicare and Medicaid Services and from a large national health maintenance organization (HMO). Neurology care or generalist care (family practitioners, general practitioners, or internists) was identified by the presence of one or more claims during the index hospital admission. The dependent variables were the time (in days) from index hospital admission to death or rehospitalization (excluding admission to a rehabilitation facility or inpatient rehabilitation unit). Additional dependent variables representing the process of care included discharge destination, length of stay for the index admission, and warfarin use. Use of warfarin after discharge was proxied by claims for prothrombin time tests within 30 days after the stroke admission date (or prior to rehospitalization if rehospitalized within 30 days).

Primary diagnoses for the first rehospitalization within 30 days of the index admission date were categorized using Clinical Classification Software [9]. Control variables included individual sociodemographics, neighborhood socioeconomics (through Census 2000), HMO membership, comorbidities [10], mechanical ventilation [11], and placement/revision of a gastrostomy tube [12].

Cox regression was used to examine the relationship of physician specialty to 30-day rehospitalization and mortality. Because death is a competing risk for rehospitalization (i.e., patients cannot be rehospitalized after dying), patients who died were censored in the rehospitalization model at the date of death. Patients who were rehospitalized and subsequently died contributed to both the rehospitalization and mortality models. HMO patients who disenrolled were censored at their disenrollment date in rehospitalization models but not in mortality models (mortality data was available for HMO patients who disenrolled). Adjusted mean predicted probabilities and 95% confidence intervals were bootstrapped by replicating analyses 1,000 times. All statistical tests used robust variance estimates that allowed for clustering of patients within hospitals.

Descriptive Characteristics

Patients seen by neurologists (either alone or collaboratively) had more severe stroke and those with collaborative care also had more comorbidities in addition to more severe stroke (table 1). Patients seen by a generalist only had more comorbidities and less severe stroke.

Table 1

Key characteristics of hospitalized acute stroke patients by neurology care category (n = 44,099)

Key characteristics of hospitalized acute stroke patients by neurology care category (n = 44,099)
Key characteristics of hospitalized acute stroke patients by neurology care category (n = 44,099)

30-Day Mortality and Rehospitalization

Patients who were seen by a neurologist only or who had collaborative care had significantly lower 30-day mortality when compared to patients seen by a generalist only (table 2). Patients seen by a neurologist only also had a borderline lower risk of 30-day rehospitalization.

Table 2

Hazard ratios (HR) and 95% confidence intervals (CI) for the relationship between neurology care and 30-day mortality and rehospitalization (n = 44,099)

Hazard ratios (HR) and 95% confidence intervals (CI) for the relationship between neurology care and 30-day mortality and rehospitalization (n = 44,099)
Hazard ratios (HR) and 95% confidence intervals (CI) for the relationship between neurology care and 30-day mortality and rehospitalization (n = 44,099)

Cause-Specific 30-Day Rehospitalization

To examine specific mechanisms that might be associated with improved outcome, we examined cause-specific rehospitalization (table 3). Patients seen by a neurologist only had a significant 12% decreased risk of rehospitalization for infections and aspiration pneumonitis, but showed a significant 17% increased risk of rehospitalization for heart disease. In addition, patients with collaborative care had a significant 19% increased risk of rehospitalization for non-acute cerebrovascular disease. With the exception of an increased risk of rehospitalization for ‘signs, symptoms, and ill-defined conditions’ among patients with no physician claims, there were no other significant differences in the risk of rehospitalization.

Table 3

Hazard ratios (HR) and 95% confidence intervals (CI) for the relationship between neurology care and 30-day cause-specific rehospitalization (n = 44,099)

Hazard ratios (HR) and 95% confidence intervals (CI) for the relationship between neurology care and 30-day cause-specific rehospitalization (n = 44,099)
Hazard ratios (HR) and 95% confidence intervals (CI) for the relationship between neurology care and 30-day cause-specific rehospitalization (n = 44,099)

Discharge Destination, Length of Stay, and Warfarin Use

Neurologists and generalists differed significantly in the process of care for stroke patients (table 4). After adjustment, patients who saw neurologists (either alone or collaboratively) were significantly more likely to be discharged to an inpatient rehabilitation facility when compared to patients who saw a generalist only, and patients who saw neurologists only were much less likely to be discharged to a skilled nursing facility. Patients who saw neurologists (either alone or collaboratively) also had somewhat longer lengths of stay. Finally, patients who saw neurologists showed greater warfarin use after discharge.

Table 4

Adjusted predicted means or probabilities and 95% confidence intervals (CI) for discharge destination, length of stay, and warfarin use within 30 days, by neurology care category (n = 44,099)

Adjusted predicted means or probabilities and 95% confidence intervals (CI) for discharge destination, length of stay, and warfarin use within 30 days, by neurology care category (n = 44,099)
Adjusted predicted means or probabilities and 95% confidence intervals (CI) for discharge destination, length of stay, and warfarin use within 30 days, by neurology care category (n = 44,099)

Cause-Specific 30-Day Rehospitalization, Discharge Destination, and Warfarin Use

To evaluate whether discharge destination and warfarin use explained cause-specific 30-day rehospitalization, we included these variables in the rehospitalization models for infections and aspiration pneumonitis, heart disease, and non-acute cerebrovascular disease. Including discharge destination and warfarin use in the cause-specific rehospitalization models (above) did not explain the rehospitalization results for heart disease and non-acute cerebrovascular disease. However, in the model for infections and aspiration pneumonitis, the hazard ratio was no longer statistically significant after including discharge destination and warfarin use (hazard ratio = 0.90, 95% confidence interval 0.80, 1.02). This change was primarily related to the inclusion of discharge destination.

This study corroborates previous studies [1, 2] that found care by a neurologist is associated with better outcomes but also found no evidence that collaborative care by a generalist and neurologist is associated with better outcomes than care by a neurologist alone. This contrasts with studies showing improved care for patients with myocardial infarction [13] and congestive heart failure [14] when treated collaboratively compared with care by either cardiologists or generalists alone. Because patients who saw a neurologist alone had more severe stroke but fewer comorbidities, it is possible that our finding is related to unidentified factors (e.g., unmeasured comorbidities) that we could not control for in this study. However, patients who received collaborative care had both more severe stroke and similar comorbidities when compared to patients who received generalist care only, yet still showed decreased mortality. It is also possible that other unmeasured differences between patients might also explain our results through residual confounding.

Our results represent the first population-based evidence indicating that care by a neurologist is associated with lower risk of rehospitalizations for infections and aspiration. This is consistent with a large meta-analysis suggesting that stroke units may lower the risk of ‘complications of immobility’ that could lead to death (examples included sepsis, venous thromboembolism, and decubitus ulceration), although the authors viewed their analyses as only suggestive [7]. However, it is important to note that stroke units are based on a collaborative care model characterized by collaborative multidisciplinary team care. Only a minority of these trials incorporated formal neurological department care. Results are also consistent with the hypotheses that neurologists improve outcomes specifically by reducing the potential for aspiration (through increased swallowing evaluations) or by improving functioning (through use of rehabilitation therapy) [3]. In contrast to previous work [7], we found no evidence that neurology care was associated with lower risk of cardiovascular complications and, in fact, our data suggested the opposite association.

There are a variety of complex factors that may explain these differences in outcomes. Different practice styles by neurologists when compared to generalists may be particularly relevant [1, 3]. Important aspects of neurologist practice style include increased use of interventions by other specialties (occupational and speech therapy) [3] and use of other facilities (inpatient rehabilitation facilities [1]). Through increased referrals to these other specialties and facilities, neurologists represent an initial contact that may have complex effects on subsequent outcomes. This may be supported by our results suggesting that discharge destination partially explained the association between infections and aspiration pneumonitis. However, we did not have data on occupational and speech therapy, or on swallowing evaluations, which may also contribute to explaining these results.

Administrative data provide large sample sizes to study cause-specific rehospitalizations, but such data also have substantial limitations. By using the primary discharge diagnosis code, we improve the likelihood of identifying true stroke but may bias the sample toward more benign outcomes as non-primary position patients have a larger comorbidity burden and higher 30-day case fatality [15]. This approach may have led us to underestimate outcomes compared to the entire population of stroke patients.

Given the substantial burden associated with caring for older stroke patients in the United States, future studies should continue to examine the mechanisms by which neurologists working alone or collaboratively may achieve better outcomes for stroke care. If additional studies support our findings, the specific processes of care that lead to these improved outcomes need to be identified. Improved understanding of how neurologists might translate care processes into better outcomes can support both increased use of neurologists in caring for stroke and education for generalists about specific processes of stroke care that may improve outcome.

This study was supported by a grant (R01-AG19747) from the National Institute of Aging.

1.
Mitchell JB, Ballard DJ, Whisnant JP, Ammering CJ, Samsa GP, Matchar DB: What role do neurologists play in determining the costs and outcomes of stroke patients? Stroke 1996;27:1937–1943.
2.
Smith MA, Shahar E, McGovern PG, Kane RL, Doliszny KM, Arnett DK, Luepker RV: HMO membership and patient age and the use of specialty care for hospitalized patients with acute stroke: the Minnesota Stroke Survey. Med Care 1999;37:1186–1198.
3.
Goldstein LB, Matchar DB, Hoff-Lindquist J, Samsa GP, Horner RD: VA Stroke Study: neurologist care is associated with increased testing but improved outcomes. Neurology 2003;61:792–796.
4.
Kaste M, Palomaki H, Sarna S: Where and how should elderly stroke patients be treated? A randomized trial. Stroke 1995;26:249–253.
5.
Lees KR: Stroke is best managed by a neurologist: battle of the titans. Stroke 2003;34:2764–2765.
6.
Caplan L: Stroke is best managed by neurologists. Stroke 2003;34:2763.
7.
Stroke Unit Trialists’ Collaboration: How do stroke units improve patient outcomes? A collaborative systematic review of the randomized trials. Stroke 1997;28:2139–2144.
8.
Diez-Tejedor E, Fuentes B: Acute care in stroke: do stroke units make the difference? Cerebrovasc Dis 2001;11(suppl 1):31–39.
9.
Clinical Classifications Software (ICD-9-CM): Summary and Downloading Information. Rockville, Agency for Healthcare Research and Quality, 2003.
10.
Elixhauser A, Steiner C, Harris DR, Coffey RM: Comorbidity measures for use with administrative data. Med Care 1998;36:8–27.
11.
Horner RD, Sloane RJ, Kahn KL: Is use of mechanical ventilation a reasonable proxy indicator for coma among Medicare patients hospitalized for acute stroke? Health Serv Res 1998;32:841–859.
12.
Quan H, Parsons GA, Ghali WA: Validity of procedure codes in International Classification of Diseases, 9th revision, clinical modification administrative data. Med Care 2004;42:801–809.
13.
Ayanian JZ, Guadagnoli E, McNeil BJ, Cleary PD: Treatment and outcomes of acute myocardial infarction among patients of cardiologists and generalist physicians. Arch Intern Med 1997;157:2570–2576.
14.
Chin MH, Wang JC, Zhang JX, Sachs GA, Lang RM: Differences among geriatricians, general internists, and cardiologists in the care of patients with heart failure: a cautionary tale of quality assessment. J Am Geriatr Soc 1998;46:1349–1354.
15.
Tirschwell DL, Longstreth WT Jr: Validating administrative data in stroke research. Stroke 2002;33:2465–2470.
Copyright / Drug Dosage / Disclaimer
Copyright: All rights reserved. No part of this publication may be translated into other languages, reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying, recording, microcopying, or by any information storage and retrieval system, without permission in writing from the publisher.
Drug Dosage: The authors and the publisher have exerted every effort to ensure that drug selection and dosage set forth in this text are in accord with current recommendations and practice at the time of publication. However, in view of ongoing research, changes in government regulations, and the constant flow of information relating to drug therapy and drug reactions, the reader is urged to check the package insert for each drug for any changes in indications and dosage and for added warnings and precautions. This is particularly important when the recommended agent is a new and/or infrequently employed drug.
Disclaimer: The statements, opinions and data contained in this publication are solely those of the individual authors and contributors and not of the publishers and the editor(s). The appearance of advertisements or/and product references in the publication is not a warranty, endorsement, or approval of the products or services advertised or of their effectiveness, quality or safety. The publisher and the editor(s) disclaim responsibility for any injury to persons or property resulting from any ideas, methods, instructions or products referred to in the content or advertisements.