BACKGROUND: Adjusting for large numbers of covariates ascertained from patients' health care claims data may improve control of confounding, as these variables may collectively be proxies for unobserved factors. Here, we develop and test an algorithm that empirically identifies candidate covariates, prioritizes covariates, and integrates them into a propensity-score-based confounder adjustment model. METHODS: We developed a multistep algorithm to implement high-dimensional proxy adjustment in claims data. Steps include (1) identifying data dimensions, eg, diagnoses, procedures, and medications; (2) empirically identifying candidate covariates; (3) assessing recurrence of codes; (4) prioritizing covariates; (5) selecting covariates for adjustment; (6) estimating the exposure propensity score; and (7) estimating an outcome model. This algorithm was tested in Medicare claims data, including a study on the effect of Cox-2 inhibitors on reduced gastric toxicity compared with nonselective nonsteroidal anti-inflammatory drugs (NSAIDs). RESULTS: In a population of 49,653 new users of Cox-2 inhibitors or nonselective NSAIDs, a crude relative risk (RR) for upper GI toxicity (RR = 1.09 [95% confidence interval = 0.91-1.30]) was initially observed. Adjusting for 15 predefined covariates resulted in a possible gastroprotective effect (0.94 [0.78-1.12]). A gastroprotective effect became stronger when adjusting for an additional 500 algorithm-derived covariates (0.88 [0.73-1.06]). Results of a study on the effect of statin on reduced mortality were similar. Using the algorithm adjustment confirmed a null finding between influenza vaccination and hip fracture (1.02 [0.85-1.21]). CONCLUSIONS: In typical pharmacoepidemiologic studies, the proposed high-dimensional propensity score resulted in improved effect estimates compared with adjustment limited to predefined covariates, when benchmarked against results expected from randomized trials.
BACKGROUND: Adjusting for large numbers of covariates ascertained from patients' health care claims data may improve control of confounding, as these variables may collectively be proxies for unobserved factors. Here, we develop and test an algorithm that empirically identifies candidate covariates, prioritizes covariates, and integrates them into a propensity-score-based confounder adjustment model. METHODS: We developed a multistep algorithm to implement high-dimensional proxy adjustment in claims data. Steps include (1) identifying data dimensions, eg, diagnoses, procedures, and medications; (2) empirically identifying candidate covariates; (3) assessing recurrence of codes; (4) prioritizing covariates; (5) selecting covariates for adjustment; (6) estimating the exposure propensity score; and (7) estimating an outcome model. This algorithm was tested in Medicare claims data, including a study on the effect of Cox-2 inhibitors on reduced gastric toxicity compared with nonselective nonsteroidal anti-inflammatory drugs (NSAIDs). RESULTS: In a population of 49,653 new users of Cox-2 inhibitors or nonselective NSAIDs, a crude relative risk (RR) for upper GI toxicity (RR = 1.09 [95% confidence interval = 0.91-1.30]) was initially observed. Adjusting for 15 predefined covariates resulted in a possible gastroprotective effect (0.94 [0.78-1.12]). A gastroprotective effect became stronger when adjusting for an additional 500 algorithm-derived covariates (0.88 [0.73-1.06]). Results of a study on the effect of statin on reduced mortality were similar. Using the algorithm adjustment confirmed a null finding between influenza vaccination and hip fracture (1.02 [0.85-1.21]). CONCLUSIONS: In typical pharmacoepidemiologic studies, the proposed high-dimensional propensity score resulted in improved effect estimates compared with adjustment limited to predefined covariates, when benchmarked against results expected from randomized trials.
Authors: Daniel H Solomon; Sebastian Schneeweiss; Robert J Glynn; Yuka Kiyota; Raisa Levin; Helen Mogun; Jerry Avorn Journal: Circulation Date: 2004-04-19 Impact factor: 29.690
Authors: Jeremy A Rassen; Robert J Glynn; Kenneth J Rothman; Soko Setoguchi; Sebastian Schneeweiss Journal: Pharmacoepidemiol Drug Saf Date: 2011-12-08 Impact factor: 2.890
Authors: Krista F Huybrechts; M Alan Brookhart; Kenneth J Rothman; Rebecca A Silliman; Tobias Gerhard; Stephen Crystal; Sebastian Schneeweiss Journal: Am J Epidemiol Date: 2011-09-20 Impact factor: 4.897
Authors: Jessica A Myers; Jeremy A Rassen; Joshua J Gagne; Krista F Huybrechts; Sebastian Schneeweiss; Kenneth J Rothman; Marshall M Joffe; Robert J Glynn Journal: Am J Epidemiol Date: 2011-10-24 Impact factor: 4.897
Authors: Issa J Dahabreh; Radley C Sheldrick; Jessica K Paulus; Mei Chung; Vasileia Varvarigou; Haseeb Jafri; Jeremy A Rassen; Thomas A Trikalinos; Georgios D Kitsios Journal: Eur Heart J Date: 2012-06-17 Impact factor: 29.983
Authors: Derek W Brown; Stacia M DeSantis; Thomas J Greene; Vahed Maroufy; Ashraf Yaseen; Hulin Wu; George Williams; Michael D Swartz Journal: Stat Med Date: 2020-04-16 Impact factor: 2.373
Authors: Jessica M Franklin; Sebastian Schneeweiss; Jennifer M Polinski; Jeremy A Rassen Journal: Comput Stat Data Anal Date: 2014-04 Impact factor: 1.681
Authors: Jeff Y Yang; Michael Webster-Clark; Jennifer L Lund; Robert S Sandler; Evan S Dellon; Til Stürmer Journal: Gastrointest Endosc Date: 2019-04-30 Impact factor: 9.427