1932

Abstract

Uncontrolled confounding due to unmeasured confounders biases causal inference in health science studies using observational and imperfect experimental designs. The adoption of methods for analysis of bias due to uncontrolled confounding has been slow, despite the increasing availability of such methods. Bias analysis for such uncontrolled confounding is most useful in big data studies and systematic reviews to gauge the extent to which extraneous preexposure variables that affect the exposure and the outcome can explain some or all of the reported exposure-outcome associations. We review methods that can be applied during or after data analysis to adjust for uncontrolled confounding for different outcomes, confounders, and study settings. We discuss relevant bias formulas and how to obtain the required information for applying them. Finally, we develop a new intuitive generalized bias analysis framework for simulating and adjusting for the amount of uncontrolled confounding due to not measuring and adjusting for one or more confounders.

Loading

Article metrics loading...

/content/journals/10.1146/annurev-publhealth-032315-021644
2017-03-20
2024-06-17
Loading full text...

Full text loading...

/deliver/fulltext/publhealth/38/1/annurev-publhealth-032315-021644.html?itemId=/content/journals/10.1146/annurev-publhealth-032315-021644&mimeType=html&fmt=ahah

Literature Cited

  1. Angus DC. 1.  2015. Fusing randomized trials with big data: the key to self-learning health care systems?. JAMA 314:8767–68 [Google Scholar]
  2. Arah OA. 2.  2008. The role of causal reasoning in understanding Simpson's paradox, Lord's paradox, and the suppression effect: covariate selection in the analysis of observational studies. Emerg. Themes Epidemiol. 5:5 [Google Scholar]
  3. Arah OA, Chiba Y, Greenland S. 3.  2008. Bias formulas for external adjustment and sensitivity analysis of unmeasured confounders. Ann. Epidemiol. 18:8637–46 [Google Scholar]
  4. Arah OA, Sudan M, Olsen J, Kheifets L. 4.  2013. Marginal structural models, doubly robust estimation, and bias analysis in perinatal and paediatric epidemiology. Paediatr. Perinat. Epidemiol. 27:3263–65 [Google Scholar]
  5. Axelson O, Steenland K. 5.  1988. Indirect methods of assessing the effects of tobacco use in occupational studies. Am. J. Ind. Med. 13:1105–18 [Google Scholar]
  6. Breslow NE, Day NE. 6.  1980. Statistical Methods in Cancer Research 1 The Analysis of Case-Control Studies Int. Agency Res. Cancer Sci. Publ. 32. Lyon, Fr.: Int. Agency Res. Cancer [Google Scholar]
  7. Bross IDJ. 7.  1966. Spurious effects from an extraneous variable. J. Chronic Dis. 19:6637–47 [Google Scholar]
  8. Bross IDJ. 8.  1967. Pertinency of an extraneous variable. J. Chronic Dis. 20:7487–95 [Google Scholar]
  9. Brumback B, Greenland S, Redman M, Kiviat N, Diehr P. 9.  2003. The intensity-score approach to adjusting for confounding. Biometrics 59:2274–85 [Google Scholar]
  10. Cai Z, Brumback BA. 10.  2015. Model-based standardization to adjust for unmeasured cluster-level confounders with complex survey data. Stat. Med. 34:152368–80 [Google Scholar]
  11. Cornfield J, Haenszel W, Hammond EC, Lilienfeld AM, Shimkin MB, Wynder EL. 11.  1959. Smoking and lung cancer: recent evidence and a discussion of some questions. J. Natl. Cancer Inst. 22:173–203 [Google Scholar]
  12. Faries D, Peng X, Pawaskar M, Price K, Stamey JD, Seaman JW. 12.  Evaluating the impact of unmeasured confounding with internal validation data: an example cost evaluation in type 2 diabetes. Value Health 16:2259–66 [Google Scholar]
  13. Flanders WD, Khoury MJ. 13.  1990. Indirect assessment of confounding: graphic description and limits on effect of adjusting for covariates. Epidemiology 1:3239–46 [Google Scholar]
  14. Gail MH, Wacholder S, Lubin JH. 14.  1988. Indirect corrections for confounding under multiplicative and additive risk models. Am. J. Ind. Med. 13:1119–30 [Google Scholar]
  15. Goto A, Arah OA, Goto M, Terauchi Y, Noda M. 15.  2013. Severe hypoglycaemia and cardiovascular disease: systematic review and meta-analysis with bias analysis. BMJ 347:f4533 [Google Scholar]
  16. Greenland S. 16.  1996. Basic methods for sensitivity analysis of biases. Int. J. Epidemiol. 25:61107–16 [Google Scholar]
  17. Greenland S. 17.  2003. The impact of prior distributions for uncontrolled confounding and response bias. J. Am. Stat. Assoc. 98:46147–54 [Google Scholar]
  18. Greenland S. 18.  2004. Bounding analysis as an inadequately specified methodology. Risk Anal 24:51085–92 [Google Scholar]
  19. Greenland S. 19.  2005. Multiple-bias modelling for analysis of observational data (with discussion). J. R. Stat. Soc. Ser. A 168:2267–306 [Google Scholar]
  20. Greenland S. 20.  2009. Bayesian perspectives for epidemiologic research. III: Bias analysis via missing-data methods. Int. J. Epidemiol. 38:61662–73 [Google Scholar]
  21. Greenland S. 21.  2014. Sensitivity analysis and bias analysis. Handbook of Epidemiology W Ahrens, I Pigeot 685–706 New York: Springer, 2nd ed.. [Google Scholar]
  22. Greenland S, Pearl J, Robins JM. 22.  1999. Causal diagrams for epidemiologic research. Epidemiology 10:137–48 [Google Scholar]
  23. Helmich E, Boerebach BCM, Arah OA, Lingard L. 23.  2015. Beyond limitations: improving how we handle uncertainty in health professions education research. Med. Teach. 37:111–8 [Google Scholar]
  24. Jain SH, Rosenblatt M, Duke J. 24.  2014. Is big data the new frontier for academic-industry collaboration?. JAMA 311:212171 [Google Scholar]
  25. Kaufmann SHE, Fletcher HA, Guzman CA, Ottenhoff THM. 25.  2015. Big data in vaccinology: introduction and section summaries. Vaccine 33:405237–40 [Google Scholar]
  26. Klungsøyr O, Sexton J, Sandanger I, Nygård JF. 26.  2009. Sensitivity analysis for unmeasured confounding in a marginal structural Cox proportional hazards model. Lifetime Data Anal 15:2278–94 [Google Scholar]
  27. Larson EB. 27.  2013. Building trust in the power of “big data” research to serve the public good. JAMA 309:232443–44 [Google Scholar]
  28. Lash TL, Fink AK. 28.  2003. Semi-automated sensitivity analysis to assess systematic errors in observational data. Epidemiology 14:4451–58 [Google Scholar]
  29. Lash TL, Fox MP, Fink AK. 29.  2011. Applying Quantitative Bias Analysis to Epidemiologic Data New York: Springer [Google Scholar]
  30. Lash TL, Fox MP, MacLehose RF, Maldonado G, McCandless LC, Greenland S. 30.  2014. Good practices for quantitative bias analysis. Int. J. Epidemiol. 43:61969–85 [Google Scholar]
  31. Lee W-C. 31.  2011. Bounding the bias of unmeasured factors with confounding and effect-modifying potentials. Stat. Med. 30:91007–17 [Google Scholar]
  32. Li L, Brumback BA, Weppelmann TA, Morris JG, Ali A. 32.  2016. Adjusting for unmeasured confounding due to either of two crossed factors with a logistic regression model. Stat. Med. 35:183179–88 [Google Scholar]
  33. Lin DY, Psaty BM, Kronmal RA. 33.  1998. Assessing the sensitivity of regression results to unmeasured confounders in observational studies. Biometrics 54:3948 [Google Scholar]
  34. Lipsitch M, Tchetgen ET, Cohen T. 34.  2010. Negative controls: a tool for detecting confounding and bias in observational studies. Epidemiology 21:3383–88 [Google Scholar]
  35. Luna X, De, Waernbaum I, Richardson TS. 35.  2011. Covariate selection for the nonparametric estimation of an average treatment effect. Biometrika 98:4861–75 [Google Scholar]
  36. MacLehose RF, Kaufman S, Kaufman JS, Poole C. 36.  2005. Bounding causal effects under uncontrolled confounding using counterfactuals. Epidemiology 16:4548–55 [Google Scholar]
  37. McCandless LC, Gustafson P, Levy A. 37.  2007. Bayesian sensitivity analysis for unmeasured confounding in observational studies. Stat. Med. 26:112331–47 [Google Scholar]
  38. McCandless LC, Richardson S, Best N. 38.  2012. Adjustment for missing confounders using external validation data and propensity scores. J. Am. Stat. Assoc. 107:49740–51 [Google Scholar]
  39. McCulloch CE, Searle SR, Neuhaus JM. 39.  2009. Generalized, Linear, and Mixed Models Hoboken, NJ: John Wiley & Sons [Google Scholar]
  40. Myers JA, Rassen JA, Gagne JJ, Huybrechts KF, Schneeweiss S. 40.  et al. 2011. Effects of adjusting for instrumental variables on bias and precision of effect estimates. Am. J. Epidemiol. 174:111213–22 [Google Scholar]
  41. Pearl J. 41.  2009. Causal inference in statistics: an overview. Stat. Surv. 3:96–146 [Google Scholar]
  42. Pearl J. 42.  2009. Causality: Models, Reasoning and Inference New York: Cambridge Univ. Press, 2nd ed.. [Google Scholar]
  43. Pearl J. 43.  2011. Invited commentary: understanding bias amplification. Am. J. Epidemiol. 174:111223–27 [Google Scholar]
  44. Phillips CV. 44.  2003. Quantifying and reporting uncertainty from systematic errors. Epidemiology 14:4459–66 [Google Scholar]
  45. Porta M. 45.  2014. A Dictionary of Epidemiology New York: Oxford Univ. Press, 6th ed.. [Google Scholar]
  46. Robins JM, Rotnitzky A, Scharfstein DO. 46.  2000. Sensitivity analysis for selection bias and unmeasured confounding in missing data and causal inference models. Statistical Models in Epidemiology, the Environment, and Clinical Trials ME Halloran, D Berry 1–94 New York: Springer [Google Scholar]
  47. Rosenbaum PR. 47.  2002. Observational Studies New York: Springer, 2nd ed.. [Google Scholar]
  48. Rosenbaum PR. 48.  2010. Design of Observational Studies New York: Springer [Google Scholar]
  49. Rothman KJ, Greenland S, Lash TL. 49.  2008. Modern Epidemiology Philadelphia: Lippincott Williams & Wilkins, 3rd ed.. [Google Scholar]
  50. Schlesselman JJ. 50.  1978. Assessing effects of confounding variables. Am. J. Epidemiol. 108:13–8 [Google Scholar]
  51. Schneeweiss S. 51.  2006. Sensitivity analysis and external adjustment for unmeasured confounders in epidemiologic database studies of therapeutics. Pharmacoepidemiol. Drug Saf. 15:5291–303 [Google Scholar]
  52. Stamey JD, Beavers DP, Faries D, Price KL, Seaman JW. 52.  2014. Bayesian modeling of cost-effectiveness studies with unmeasured confounding: a simulation study. Pharm. Stat. 13:194–100 [Google Scholar]
  53. Steenland K. 53.  2004. Monte Carlo sensitivity analysis and Bayesian analysis of smoking as an unmeasured confounder in a study of silica and lung cancer. Am. J. Epidemiol. 160:4384–92 [Google Scholar]
  54. Stürmer T, Glynn RJ, Rothman KJ, Avorn J, Schneeweiss S. 54.  2007. Adjustments for unmeasured confounders in pharmacoepidemiologic database studies using external information. Med. Care 45:10 Suppl. 2S158–65 [Google Scholar]
  55. Stürmer T, Rothman KJ, Avorn J, Glynn RJ. 55.  2010. Treatment effects in the presence of unmeasured confounding: dealing with observations in the tails of the propensity score distribution—a simulation study. Am. J. Epidemiol. 172:7843–54 [Google Scholar]
  56. Stürmer T, Schneeweiss S, Avorn J, Glynn RJ. 56.  2005. Adjusting effect estimates for unmeasured confounding with validation data using propensity score calibration. Am. J. Epidemiol. 162:3279–89 [Google Scholar]
  57. Stürmer T, Schneeweiss S, Rothman KJ, Avorn J, Glynn RJ. 57.  2007. Performance of propensity score calibration—a simulation study. Am. J. Epidemiol. 165:101110–18 [Google Scholar]
  58. Sudan M, Kheifets L, Arah OA, Olsen J. 58.  2013. Cell phone exposures and hearing loss in children in the Danish national birth cohort. Paediatr. Perinat. Epidemiol. 27:3247–57 [Google Scholar]
  59. Uddin MJ, Groenwold RHH, Ali MS, de Boer A, Roes KCB. 59.  et al. 2016. Methods to control for unmeasured confounding in pharmacoepidemiology: an overview. Int. J. Clin. Pharm. 38:3714–23 [Google Scholar]
  60. VanderWeele TJ. 60.  2010. Bias formulas for sensitivity analysis for direct and indirect effects. Epidemiology 21:4540–51 [Google Scholar]
  61. VanderWeele TJ. 61.  2013. Unmeasured confounding and hazard scales: sensitivity analysis for total, direct, and indirect effects. Eur. J. Epidemiol. 28:2113–17 [Google Scholar]
  62. VanderWeele TJ. 62.  2016. Mediation analysis: a practitioner's guide. Annu. Rev. Public Health 37:17–32 [Google Scholar]
  63. Vanderweele TJ, Arah OA. 63.  2011. Bias formulas for sensitivity analysis of unmeasured confounding for general outcomes, treatments, and confounders. Epidemiology 22:142–52 [Google Scholar]
  64. Vanderweele TJ, Mukherjee B, Chen J. 64.  2012. Sensitivity analysis for interactions under unmeasured confounding. Stat. Med. 31:222552–64 [Google Scholar]
  65. Vanderweele TJ, Shpitser I. 65.  2011. A new criterion for confounder selection. Biometrics 67:41406–13 [Google Scholar]
  66. VanderWeele TJ, Shpitser I. 66.  2013. On the definition of a confounder. Ann. Stat. 41:1196–220 [Google Scholar]
  67. Westreich D, Greenland S. 67.  2013. The table 2 fallacy: presenting and interpreting confounder and modifier coefficients. Am. J. Epidemiol. 177:4292–98 [Google Scholar]
  68. Yanagawa T. 68.  1984. Case-control studies: assessing the effect of a confounding factor. Biometrika 71:1191–94 [Google Scholar]
/content/journals/10.1146/annurev-publhealth-032315-021644
Loading
/content/journals/10.1146/annurev-publhealth-032315-021644
Loading

Data & Media loading...

  • Article Type: Review Article
This is a required field
Please enter a valid email address
Approval was a Success
Invalid data
An Error Occurred
Approval was partially successful, following selected items could not be processed due to error