Emulating a target trial in case-control designs: an application to statins and colorectal cancer

被引:39
作者
Dickerman, Barbra A. [1 ]
Garcia-Albeniz, Xabier [1 ,2 ]
Logan, Roger W. [1 ]
Denaxas, Spiros [3 ,4 ,5 ]
Hernan, Miguel A. [1 ,6 ,7 ]
机构
[1] Harvard TH Chan Sch Publ Hlth, Dept Epidemiol, 677 Huntington Ave, Boston, MA 02115 USA
[2] RTI Hlth Solut, Barcelona, Spain
[3] UCL, Inst Hlth Informat Res, London, England
[4] UCL, Hlth Data Res UK HDR UK London, London, England
[5] Alan Turing Inst, London, England
[6] Harvard TH Chan Sch Publ Hlth, Dept Biostat, Boston, MA USA
[7] Harvard MIT Div Hlth Sci & Technol, Boston, MA USA
基金
英国医学研究理事会; 美国国家卫生研究院;
关键词
Case-control; causal inference; comparative effectiveness; electronic health records; target trial; CARDIOVASCULAR-DISEASE; RISK; VALIDATION; PLEA;
D O I
10.1093/ije/dyaa144
中图分类号
R1 [预防医学、卫生学];
学科分类号
1004 ; 120402 ;
摘要
Background: Previous case-control studies have reported a strong association between statin use and lower cancer risk. It is unclear whether this association reflects a benefit of statins or is the result of design decisions that cannot be mapped to a (hypothetical) target trial (that would answer the question of interest). Methods: We outlined the protocol of a target trial to estimate the effect of statins on colorectal cancer incidence among adults with low-density lipoprotein (LDL) cholesterol below 5 mmol/L. We then emulated the target trial using linked electronic health records of 752 469 eligible UK adults (CALIBER 1999-2016) under both a cohort design and a case-control sampling of the cohort. We used pooled logistic regression to estimate intention-to-treat and per-protocol effects of statins on colorectal cancer, with adjustment for baseline and time-varying risk factors via inverse-probability weighting. Finally, we compared our case-control effect estimates with those obtained using previous case-control procedures. Results: Over the 6-year follow-up, 3596 individuals developed colorectal cancer. Estimated intention-to-treat and per-protocol hazard ratios were 1.00 (95% confidence interval [CI]: 0.87, 1.16) and 0.90 (95% CI: 0.71, 1.12), respectively. As expected, adequate case-control sampling yielded the same estimates. By contrast, previous case-control analytical approaches yielded estimates that appeared strongly protective (odds ratio 0.57, 95% CI: 0.36, 0.91, for >= 5 vs. <5 years of statin use). Conclusions: Our study demonstrates how to explicitly emulate a target trial using case-control data to reduce discrepancies between observational and randomized trial evidence. This approach may inform future case-control analyses for comparative effectiveness research.
引用
收藏
页码:1637 / 1646
页数:10
相关论文
共 21 条
[1]  
[Anonymous], 2008, Mordern Epidemiology
[2]   The association between 3-hydroxy-3-methylglutaryl conenzyme A inhibitor use and breast carcinoma risk among postmenopausal women - A case-control study [J].
Boudreau, DM ;
Gardner, JS ;
Malone, KE ;
Heckbert, SR ;
Blough, DK ;
Daling, JR .
CANCER, 2004, 100 (11) :2308-2316
[3]   Statins for Prevention of Cardiovascular Disease in Adults Evidence Report and Systematic Review for the US Preventive Services Task Force [J].
Chou, Roger ;
Dana, Tracy ;
Blazina, Ian ;
Daeges, Monica ;
Jeanne, Thomas L. .
JAMA-JOURNAL OF THE AMERICAN MEDICAL ASSOCIATION, 2016, 316 (19) :2008-2024
[4]   Statins and cancer risk - A meta-analysis [J].
Dale, KM ;
Coleman, CI ;
Henyan, NN ;
Kluger, J ;
White, CM .
JAMA-JOURNAL OF THE AMERICAN MEDICAL ASSOCIATION, 2006, 295 (01) :74-80
[5]   UK phenomics platform for developing and validating electronic health record phenotypes: CALIBER [J].
Denaxas, Spiros ;
Gonzalez-Izquierdo, Arturo ;
Direk, Kenan ;
Fitzpatrick, Natalie K. ;
Fatemifar, Ghazaleh ;
Banerjee, Amitava ;
Dobson, Richard J. B. ;
Howe, Laurence J. ;
Kuan, Valerie ;
Lumbers, R. Tom ;
Pasea, Laura ;
Patel, Riyaz S. ;
Shah, Anoop D. ;
Hingorani, Aroon D. ;
Sudlow, Cathie ;
Hemingway, Harry .
JOURNAL OF THE AMERICAN MEDICAL INFORMATICS ASSOCIATION, 2019, 26 (12) :1545-1559
[6]   Data Resource Profile: Cardiovascular disease research using linked bespoke studies and electronic health records (CALIBER) [J].
Denaxas, Spiros C. ;
George, Julie ;
Herrett, Emily ;
Shah, Anoop D. ;
Kalra, Dipak ;
Hingorani, Aroon D. ;
Kivimaki, Mika ;
Timmis, Adam D. ;
Smeeth, Liam ;
Hemingway, Harry .
INTERNATIONAL JOURNAL OF EPIDEMIOLOGY, 2012, 41 (06) :1625-1638
[7]   Avoidable flaws in observational analyses: an application to statins and cancer [J].
Dickerman, Barbra A. ;
Garcia-Albeniz, Xabier ;
Logan, Roger W. ;
Denaxas, Spiros ;
Hernan, Miguel A. .
NATURE MEDICINE, 2019, 25 (10) :1601-+
[8]  
Emberson JR, 2012, PLOS ONE, V7, DOI [10.1371/journal.pone.0029849, 10.1016/S0140-6736(12)60367-5]
[9]   The risk of cancer in users of statins [J].
Graaf, MR ;
Beiderbeck, AB ;
Egberts, ACG ;
Richel, DJ ;
Guchelaar, HJ .
JOURNAL OF CLINICAL ONCOLOGY, 2004, 22 (12) :2388-2394
[10]   Comparison of dynamic treatment regimes via inverse probability weighting [J].
Hernán, MA ;
Lanoy, E ;
Costagliola, D ;
Robins, JM .
BASIC & CLINICAL PHARMACOLOGY & TOXICOLOGY, 2006, 98 (03) :237-242