![• A synthesis of the six practice guidelines in 2006
finds that, “[ACEIs] or ARBs are recommended for a
patient with HTN and comorbidities such as [heart
failure] HF, myocardial infarction (MI), diabetes
mellitus [(DM)], chronic kidney disease [(CKD)], and
recurrent stroke.1”
• Both ACEIs and ARBs are known to cause anemia.
3](https://image.slidesharecdn.com/an-application-of-doubly-robust-estimation-johnson2606-120502115121-phpapp01/85/An-Application-of-Doubly-Robust-Estimation-JOHNSON-3-320.jpg)
Ad
More Related Content
What's hot (19)
Similar to An Application of Doubly Robust Estimation JOHNSON (20)
Ad
More from HMO Research Network (20)
Ad
An Application of Doubly Robust Estimation JOHNSON
- 1. An Application of Doubly Robust Estimation Brian P. Johnson, MPH, Charles E. Gessert, MD, MPH, Colleen M. Renier, BS, Jeanette A. Palcher, BA, Adnan Ajmal, MBBS HMORN conference, May 2, 2012
- 2. Background • Angiotensin converting enzyme inhibitors (ACEIs) and angiotensin receptor blockers (ARBs) are FDA-approved for the treatment of hypertension (HTN)1. – Captopril was first FDA-approved ACEI in 1981. (http://en.wikipedia.org/wiki/ACE_inhibitor, accessed April 23, 2012) – Losartan was first FDA-approved ARB in 1995. (http://en.wikipedia.org/wiki/Discovery_and_development_of_angiotensin_r eceptor_blockers, accessed April 23, 2012) 2
- 3. • A synthesis of the six practice guidelines in 2006 finds that, “[ACEIs] or ARBs are recommended for a patient with HTN and comorbidities such as [heart failure] HF, myocardial infarction (MI), diabetes mellitus [(DM)], chronic kidney disease [(CKD)], and recurrent stroke.1” • Both ACEIs and ARBs are known to cause anemia. 3
- 4. Study Overview • In 2009, Dr. Ajmal initiated a retrospective study of Essentia Health patient records to assess change in Hgb within a population who had been prescribed either ACEI or ARB between 2005 and 2009. • Particularly interested in patients with CKD, which is defined as a glomerular filtration rate (GFR) < 60 4
- 5. Inclusion/Exclusion Criteria (abbreviated) • Inclusion criteria – Prior primary care (PC) provided by Essentia Health (EH) – Aged 40 to 70 years and initially prescribed ACEI or ARB, but not both, by an EH PC physician – Baseline and followup (F/U) Hgb values before and after initiation of ACEI or ARB – History of DM, CHF, and/or HTN – Baseline GFR before and after initiation of ACEI or ARB • Exclusion criteria – Underlying conditions associated with anemia, or – Other conditions or treatments that might affect Hgb level during the F/U period 5
- 6. Subject Characteristics Class of drug ACEI ARB Baseline Covariate (N=551) (N=190) p-value Demographics Age 57.36 +/- 7.73 57.85 +/- 7.65 0.45 Sex (female) 230 (41.7) 104 (54.7) <0.01 Comorbidities DM 191 (34.7) 72 (37.9) 0.43 HTN 481 (87.3) 176 (92.6) 0.05 CHF 29 (5.3) 18 (9.5) 0.06 CKD 85 (15.4) 39 (20.5) 0.11 Laboratory Hgb 14.65 +/- 1.42 14.29 +/- 1.52 <0.01 6
- 7. Estimated Effects of Covariates Outcome Model Logistic Model Effect on F/U Hgb in gm/dL (95% CI) OR of ARB* Baseline Covariate Overall (N=741) (95% CI) Treatment initiation date (years) -0.06 (-0.12, -0.00) 0.85 (0.72, 0.99) Demographics Age (years) 0.00 (-0.01, 0.01) 1.00 (0.98, 1.03) Sex (female) -0.33 (-0.48, -0.19) 1.46 (1.01, 2.11) Comorbidities DM -0.03 (-0.17, 0.11) 1.09 (0.77, 1.55) HTN 0.11 (-0.10, 0.32) 1.94 (1.05, 3.59) CHF 0.39 ( 0.12, 0.67) 1.79 (0.94, 3.41) CKD -0.16 (-0.34, 0.03) 1.15 (0.73, 1.80) Laboratory Hgb 0.60 ( 0.55, 0.65) 0.88 (0.77, 1.00) * Odds of receiving ARB relative to odds of receiving ACEI
- 8. Evident Confounding • CHF status infers an increase in F/U Hgb and more CHF subjects were on ARBs – Clinical explanation is that CHF patients are hemodiluted at baseline and treatment for CHF increases Hgb concentration • More females were on ARBs than on ACEIs and F/U Hgb differs per sex, even while accounting for baseline Hgb • Similar issues with HTN, baseline Hgb, and when treatment was initiated 8
- 9. Causal Inference • Counterfactuals – Suppose each individual in the population has a potential outcome (e.g., F/U Hgb,) for each exposure (e.g., ACEI and ARB.) – Potential outcomes are estimated so as to be unbiased • Average causal effect (ACE) – The difference of the mean potential outcomes and mean of the difference between potential outcomes – If all confounders are measured, potential outcomes and exposures are independent which permits unbiased estimation of ACE 9
- 10. • Estimate ACE by regression modeling – Unbiased if regression model is correctly specified • Estimate ACE by inverse probability weighting – propensity to be exposed to one of the treatments is captured by an estimated probability – Unbiased if propensity model is correctly specified • Doubly-robust (DR) – Combine regression and propensity models – Unbiased estimate of ACE if either model is correct • Using SAS %dr macro of Funk et al. (2011) (See http://www.unc.edu/~mfunk/dr) 10
- 11. Augmented Inverse Probability Weighted Estimator • 11
- 12. • 12
- 13. Causal Effect Estimates (subset of subjects) DR DR DR treatment Class estimate estimate effect Subject of drug F/U Hgb (ACEI) (ARB) (ARB - ACEI) 10003* ACEI 14.4 14.19 14.39 0.20 10005 ARB 17.3 15.06 24.19 9.13 10008 ACEI 14.9 14.87 14.98 0.11 10009 ACEI 16.7 16.89 16.06 -0.83 10016* ARB 15.2 15.21 14.34 -0.87 10022 ARB 13.0 13.27 11.83 -1.44 10038 ACEI 14.0 13.56 15.46 1.90 10039 ACEI 14.4 14.58 13.35 -1.23 10047* ACEI 15.0 15.10 14.06 -1.05 10048 ARB 14.1 13.40 15.12 1.72 13
- 14. Average Causal Effect Estimates Average F/U Hgb Standard Confidence Class of drug (gm/dL) Deviation Interval p-value ACEI 14.31 1.43 (14.21, 14.42) ARB 14.48 2.02 (14.33, 14.62) Difference 0.16 2.06 ( 0.01, 0.31) 0.03 14
- 15. Conclusion • Causal estimates address the question, “what if everyone were treated with ARB relative to if everyone were treated with ACEI.” • ACE can be estimated even when confounding exists • Estimated ACE suggests F/U Hgb is higher when ARBs rather than ACEIs are prescribed, but the mean difference may not be clinically meaningful. 15
- 16. Further Research • Variable and model selection, including interactions, and sensitivity analysis – Age, for example, doesn’t seem to be important, but it’s in the models. – Effect of baseline Hgb may be different for the sexes • Use all cases, not just complete 16
- 17. References 1. Miller AE, Cziraky M, Spinler SA. ACE inhibitors versus ARBs: comparison of practice guidelines and treatment selection considerations. Formulary. 2006;41:274–284. 2. Funk MJ, Westreich D, Wiesen C, Sturmer T, Brookhart MA, Davidian M. Doubly robust estimation of causal effects. Am J Epidemiol. Apr 1 2011;173(7):761-767. 17
- 18. Bibliography Lunceford JK, Davidian M. Stratification and weighting via the propensity score in estimation of causal treatment effects: A comparative study. Stat Med. Oct 15 2004;23(19):2937-2960. Robins JM, Rotnitzky A, Zhao LP. Estimation of Regression Coefficients When Some Regressors Are Not Always Observed. J Amer Statistical Assoc. 1994;89(427):846-863. 18
Editor's Notes
- #6: Want stable PCP study population, aged 40 to 70, with a history of DM, CHF, and/or HTN, complete labs, and no confounding conditions.
- #7: Bivariate summaries. Difference in sex, HTN, and Hgb between the two groups. The difference is Hgb may be driven by the difference in sex, but it’s a difference nonetheless.
- #8: Multivariate models. Rows is red indicate the covariate has a significant effect and imbalance between the groups. Rows in orange indicate the covariate either has a significant effect or that there’s an imbalance between the groups.
- #9: Hard to ignore confounding with a clear conscience and the results may not be trustworthy.
- #12: The individual DR estimates are transparent, but the ACE in this form is not particularly intuitive.The z indicates treatment groupThe y is the actual FUP HgbThe p is the estimated probability, or propensity, of getting ARBThe m is the predicted FUP Hgb from the respective group
- #13: The ACE in this form is more intuitive, but the individual DR estimates are not transparent.Indicator functions are like switches.The mean of the numerators in the first terms, which are the residuals, are zero by definition, but the mean of the ratios are not. The mean of the difference of the ratios, however, is expected to be small if large or small values of p are not associated with large or small values of residuals. When this is true, the last two terms make the greater contribution to the ACE.Note that the first two terms honor the actual treatment received while the second make use of the concept of potential outcomes.
- #14: The starred subjects must be particularly like one or the other treatment groups.