Analytic strategies to adjust confounding using exposure propensity scores and disease risk scores:: Nonsteroidal antiinflammatory drugs and short-term mortality in the elderly

被引:141
作者
Stürmer, T
Schneeweiss, S
Brookhart, MA
Rothman, KJ
Avorn, J
Glynn, RJ
机构
[1] Harvard Univ, Sch Med, Brigham & Womens Hosp, Div Pharmacoepidemiol & Pharmacoecon, Boston, MA 02120 USA
[2] Harvard Univ, Sch Med, Brigham & Womens Hosp, Div Prevent Med, Boston, MA USA
关键词
anti-inflammatory agents; non-steroidal; bias (epidemiology); cohort studies; confounding factors (epidemiology); epidemiologic methods; research design;
D O I
10.1093/aje/kwi106
中图分类号
R1 [预防医学、卫生学];
学科分类号
1004 ; 120402 ;
摘要
Little is known about optimal application and behavior of exposure propensity scores (EPS) in small studies. In a cohort of 103,133 elderly Medicaid beneficiaries in New Jersey, the effect of nonsteroidal antiinflammatory drug use on 1-year all-cause mortality was assessed (1995-1997) based on the assumption that there is no protective effect and that the preponderance of any observed effect would be confounded. To study the comparative behavior of EPS, disease risk scores, and "conventional" disease models, the authors randomly resampled 1,000 subcohorts of 10,000, 1,000, and 500 persons. The number of variables was limited in disease models, but not EPS and disease risk scores. Estimated EPS were used to adjust for confounding by matching, inverse probability of treatment weighting, stratification, and modeling. The crude rate ratio of death was 0.68 for users of nonsteroidal antiinflammatory drugs. "Conventional" adjustment resulted in a rate ratio of 0.80 (95% confidence interval: 0.77, 0.84). The rate ratio closest to 1 (0.85) was achieved by inverse probability of treatment weighting (95% confidence interval: 0.82, 0.88). With decreasing study size, estimates remained further from the null value, which was most pronounced for inverse probability of treatment weighting (n = 500: rate ratio = 0.72, 95% confidence interval: 0.26, 1.68). In this setting, analytic strategies using EPS or disease risk scores were not generally superior to "conventional" models. Various ways to use EPS and disease risk scores behaved differently with smaller study size.
引用
收藏
页码:891 / 898
页数:8
相关论文
共 40 条
[1]  
[Anonymous], PHARMACOEPIDEMIOLOGY
[2]   Rare outcomes, common treatments: Analytic strategies using propensity scores [J].
Braitman, LE ;
Rosenbaum, PR .
ANNALS OF INTERNAL MEDICINE, 2002, 137 (08) :693-695
[3]  
Cepeda MS, 2000, PHARMACOEPIDEM DR S, V9, P103, DOI 10.1002/(SICI)1099-1557(200003/04)9:2<103::AID-PDS477>3.0.CO
[4]  
2-6
[5]   Comparison of logistic regression versus propensity score when the number of events is low and there are multiple confounders [J].
Cepeda, MS ;
Boston, R ;
Farrar, JT ;
Strom, BL .
AMERICAN JOURNAL OF EPIDEMIOLOGY, 2003, 158 (03) :280-287
[6]   Nonsteroidal anti-inflammatory drugs, apoptosis, and colon-cancer chemoprevention [J].
Chan, TA .
LANCET ONCOLOGY, 2002, 3 (03) :166-174
[7]  
COOK EF, 1989, J CLIN EPIDEMIOL, V42, P317
[8]   Nonnarcotic analgesic use and the risk of hypertension in US women [J].
Dedier, J ;
Stampfer, MJ ;
Hankinson, SE ;
Willett, WC ;
Speizer, FE ;
Curhan, GC .
HYPERTENSION, 2002, 40 (05) :604-608
[9]   PROGNOSTIC MODELS AND THE PROPENSITY SCORE [J].
DRAKE, C ;
FISHER, L .
INTERNATIONAL JOURNAL OF EPIDEMIOLOGY, 1995, 24 (01) :183-187
[10]   EFFECTS OF MISSPECIFICATION OF THE PROPENSITY SCORE ON ESTIMATORS OF TREATMENT EFFECT [J].
DRAKE, C .
BIOMETRICS, 1993, 49 (04) :1231-1236