行政院國家科學委員會專題研究計畫 成果報告
我國初次缺血性腦中風住院成本之預測
計畫類別： 個別型計畫 計畫編號： NSC92-2416-H-110-024- 執行期間： 92 年 08 月 01 日至 93 年 07 月 31 日 執行單位： 國立中山大學企業管理學系(所) 計畫主持人： 曾美君 共同主持人： 張谷州 報告類型： 精簡報告 處理方式： 本計畫可公開查詢中 華 民 國 93 年 10 月 15 日
我國初次缺血性腦中風住院成本之預測
中文摘要 過去二十年來，腦血管疾病一直高居台灣十大死因之第二位；而超過七成的腦血 管疾病是屬於缺血性腦中風。腦中風往往需要昂貴的重症加護醫療、長期照護與 復健，對病患、病患家屬以及整體社會產生很大的社會與經濟影響。由於臺灣已 經邁入世界衛生組織所謂的「高齡化社會」，腦血管疾病罹病比例預期會持續增 加，相關的醫療與長期照護需求亦將遽增。就醫療保健資源的有效運用而言，類 似重大疾病的治療成本之估算及預測更顯迫切需要。然而我國關於腦血管疾病的 相關社會、經濟與成本分析卻十分不足。 本研究延伸「我國初次缺血性腦中風之醫療成本分析」（NSC91-2416-H-110-027） 研究，以所得到的成本資料，併同病患住院初期可觀察到的病患基本特徵與診斷 資料，建構理論分析模型來預估我國缺血性腦中風的直接成本（意即相關醫療資 源的使用）。尤其，將特別著重疾病嚴重度與成本之關連。同時，透過迴歸分析之 模型驗證(model validation)方法來評估並改善所建構之模型的有效準確預測能力。 關鍵詞：成本、預測、缺血性腦中風 AbstractStroke has been the second leading cause of death in Taiwan since 1983, with ischemic stroke accounts for about 70%. Ischemic stroke, requiring costly acute hospitalization care and continuing inpatient/outpatient rehabilitation, consequently has a significant social and economic impact on patients, their families, and society as a whole. With an aging population, the costs for stroke care are likely to increase further in the future. Facing the limited healthcare resources, a better understanding of the costs of care, its determinants, and how they may be directly estimated on the basis of the patient’s early results is clearly warranted for a disease like stroke. Despite the obvious economic burden on individuals and society overall, there is still scant knowledge of the costs of stroke care in Taiwan.
The purpose of this study is, by extending a study of analyzing the direct medical costs caused by ischemic stroke in Taiwan (NSC91-2416-H-110-027), to predict the direct costs of acute hospitalization for patients with first-ever ischemic stroke in Taiwan, with particular attention given to the role of disease severity. Since the intent is to be predictive, we evaluate only those factors that can be assessed at admission. The model need to be properly validated to ascertain the predicted values from our model are likely to accurately predict responses on future patients or patients not used to develop the model.
Stroke is the second most common cause of mortality in Taiwan; in 2002 more than 12,000 residents died of cerebrovascular disease. Stroke patients, due to the associated high morbidity and mortality, are consistently high users of health care services. In facing
shortage of resources for healthcare, estimation of hospitalization cost for acute and long-term management of stroke is clearly warranted. However, to the best of our knowledge, no such study is available in Taiwan yet.
The purpose of the current analyses was to identify factors available on admission to predict costs of acute care hospitalization for patients with first-ever ischemic stroke in Taiwan. Greater attention, however, was given to the role of disease severity as it seemingly has not been well acknowledged in most cost analyses of stroke. Since the intent is for prediction, we evaluated only those factors that can be assessed at the time of admission, and bore in mind that an ideal model to predict costs should be simple and readily applicable.
Data from two different study periods were examined by means of multiple linear regression. One data set included 368 patients with first-ever ischemic stroke consecutively admitted, within 48 hours of onset, to the Department of Neurology, Chang Gung Memorial Hospital, Kaohsiung, Taiwan between September 1998 and October 1999. The other data included 217 patients with the same inclusion criteria but admitted between September 2002 and April 2003 in the same study hospital. (Another set of data from another hospital is under processed.) The hospital is a 1900-beded non-profit proprietary hospital, providing medical-center-level health care in an area with a population of approximately three million in southern Taiwan.
In this study, costs were the hospital charges because real costs were not available. The charges could be a fairly good proxy for costs as they are the reimbursement claims made by the hospital to the Bureau of National Health Insurance in Taiwan. Because our intention was to study the cost of acute care hospitalization, the discharge date was recorded as the date of the patient died or was discharged to home, another hospital, a rehabilitation facility, or any places other than the Department of Neurology in KCHMH. The data sets contain demographic, clinical, and administrative elements. Whereas the information on inpatient charges was obtained from the discharge database of the hospital, other data were
prospectively collected at admission.
Demographic characteristics included age, sex, comorbidity (history of hypertension, diabetes mellitus or hypercholesterolemia), smoking, congestive heart failure, valvular heart disease, atrial fibrillation, history of cardiac disease (history of arrhythmia, angina pectoris, ischemic heart disease, and/or abnormalities identified from the initial electrocardiogram).
Clinical characteristics included stroke severity measured by the NIH Stroke Scale (NIHSS), functional independence status at admission measured using the modified Barthel Index (MBI) on a scale of 0 to 20 (20=normal), hours after stroke onset (within 24 hours or not) and stroke subtype.
We noted that physicians often disagree about the diagnosis of subtype of ischemic stroke. We nevertheless chose to determine the ischemic stroke subtype according to the criteria of the Trial of ORG10172 in Acute Stroke Treatment. As in our study the number of patients with cardioembolism subtype of stroke was relatively small, stroke subtype was dichotomized as small-vessel occlusion versus others in our analysis.
Descriptive analyses of the independent variables and hospitalization costs were performed to obtain a global impression. The hospitalization costs, which were very positively skewed, were logarithmically transformed to normalize the distribution. We also noted that NIHSS and MBI have the kind of nonlinear relationships with the costs, possibly due to the impact of in-hospital mortality and/or discharge decisions on hospitalization costs. To account for the noticeable nonlinear effects, we assessed the necessarily to create multiple dichotomous variables. In this regard, for the purpose of this analysis, we grouped NIHSS variable into five categories (0-6, 7-10, 11-15, 16-25 and 25-38) and made the 0-6 group the reference group. Similarly, we grouped MBI variable into four categories (0-5, 6-11, 12-18, 19-20), with 19-20 being our reference group.
Self-Evaluation of the study
Cost of illness studies are necessary for determining the economic impact of disease and for assigning the material and human resources required for prevention, diagnosis, treatment, and follow-up of patients with different diseases. Our study therefore provides valuable information to healthcare decision-makers regarding resource utilization in acute care of ischemic stroke. To increase the usefulness of the model, our model need to be externally validated and consequently different set of data is warranted. However, it takes time and tremendous efforts to get another set of data which is comparable to our study cohort. Fortunately, we have made some progress and, hopefully, the results of this study could be sent to journals by the end of the year. Meanwhile, a research note on measurement and classification of stroke severity (written in Chinese) has been accepted by a local journal.
References
1. Adams HP Jr, Leclerc JR, Bluhmki E, et al. Measuring outcomes as a function of baseline severity of ischemic stroke. Cerebrovasc Dis. 2004;18:124-129.
2. Altman DG, Royston P. What do we mean by validating a prognostic model? Stat
Med. 2000;19:453-473.
3. Andersen CK, Andersen K, Kragh-Sorensen P. Cost function estimation: the choice of a model to apply to dementia. Health Econ. 2000;9:397-409.
4. Baird AE, Dambrosia J, Janket S, et al. A three-item scale for the early prediction of stroke recovery. Lancet. 2001;357:2095-2099.
5. Barber M, Fail M, Shields M, et al. Validity and reliability of estimating the Scandinavian stroke scale score from medical records. Cerebrovasc Dis. 2004;17:224-227.
6. Caro JJ, Huybrechts KF, Kelley HE. Predicting treatment costs after acute ischemic stroke on the basis of patient characteristics at presentation and early dysfunction.
Stroke. 2001;32:100-106.
7. Chang KC, Tseng MC, Weng HH, et al. Prediction of length of stay of first-ever ischemic stroke. Stroke. 2002;33:2670-2674.
8. Chang KC, Tseng MC: Costs of acute care of first-ever ischemic stroke in Taiwan.
Stroke. 2003;34:e219-e221.
9. Chiu L, Hong C-T, Shyu W-C, et al. Estimation of costs due to hospitalization for first-ever stroke patients in northern Taiwan. Chung Hua I Hsueh Tsa Chih (Taipei). 1999;62:261-267. [In Chinese]
10. Counsell C, Dennis M. Systematic review of prognostic models in patients with acute stroke. Cerebrovasc Dis. 2001;12:159–170.
11. Diringer MN, Edwards DF, Mattson DT, et al. Predictors of acute hospital costs for treatment of ischemic stroke in an academic center. Stroke. 1999;30:724-728. 12. Evers S, Voss G, Nieman F, et al. Predicting the cost of hospital stay for stroke
patients: the use of diagnosis related groups. Health Policy. 2002;61:21-42.
13. Evers SM, Struijs JN, Ament AJ, et al. International comparison of stroke cost studies.
Stroke. 2004;35:1209-1215.
14. Georgiadis D, Schwarz S, Aschoff A, et al. Hemicraniectomy and moderate hypothermia in patients with severe ischemic stroke. Stroke. 2002;33:1584-1588.
15. Harrell FE, Jr., Lee KL, Mark DB. Multivariable prognostic models: issues in developing models, evaluating assumptions and adequacy, and measuring and reducing errors. Stat Med. 1996;15:361-387.
16. Holloway RG, Benesch CG, Rahilly CR, et al. A systematic review of cost-effectiveness research of stroke evaluation and treatment. Stroke.
1999;30:1340-1349.
17. Lipscomb J, Ancukiewicz M, Parmigiani G, Hasselblad V, Samsa G, Matchar DB. Predicting the cost of illness: a comparison of alternative models applied to stroke. Med
Decis Making. 1998;18(Suppl):S39-S56.
18. Mamoli A, Censori B, Casto L, et al. An analysis of the costs of ischemic stroke in an Italian stroke unit. Neurology. 1999;53:112-126.
19. Martinez-Vila E, Irimia P. The cost of stroke. Cerebrovasc Dis. 2004;17(Suppl 1):124-129.
20. Payne KA, Huybrechts KF, Caro JJ, et al. Long term cost-of-illness in stroke: an international review. Pharmacoeconomics. 2002;20:813-825.
21. Reed SD, Blough DK, Meyer K, et al. Inpatient costs, length of stay, and mortality for cerebrovascular events in community hospitals. Neurology. 2001;57:305-314. 22. Rundek T, Mast H, Hartmann A, et al. Predictors of resource use after acute
hospitalization: the Northern Manhattan Stroke Study. Neurology. 2000;55:1180-1187. 23. Schlegel D, Kolb SJ, Luciano JM, et al. Utility of the NIH Stroke Scale as a predictor of
hospital disposition. Stroke. 2003;34:134-137.
24. Sundberg G, Bagust A, Terent A. A model for costs of stroke services. Health Policy. 2003;63:81-94.
25. Yoneda Y, Uehara T, Yamasaki H, Kita Y, Tabuchi M, Mori E. Hospital-based study of the care and cost of acute ischemic stroke in Japan. Stroke. 2003;34:718-724.
