Jump to comment:

• Adjusting Hospital Scores to Account for Differences in Patient Characteristics
  Layla Parast, Marc Elliott
  Published on: 6 November 2018

• Published on: 6 November 2018

  Adjusting Hospital Scores to Account for Differences in Patient Characteristics

  • Layla Parast, Statistician RAND Corporation
  • Other Contributors:
    • Marc Elliott, Statistician

  We read with interest the recent article by Rubens et al. “Impact of Patient Characteristics on the Canadian Patient Experiences Survey–Inpatient Care: Survey Analysis from an Academic Tertiary Care Centre.” This work adds to the literature by examining patient characteristics associated with response patterns on the Canadian Patient Experience Inpatient Care Survey. The authors conclude that caution is needed when comparing performance between different entities assessed by this survey, as observed differences could be explained by variation in patient mix rather than variation in performance. While we agree that patient characteristics are likely to explain some of the variation in observed patient experience scores, as has been found in similar settings in the US and UK (Elliott et al. 2009, Paddison et al. 2012), we do not believe that this is problematic for the validity of patient experience scores. Following the common practice of adjusting entity scores for patient characteristics such as self-reported health, education, and age allows for valid comparisons as if all hospitals had treated the same patient population. In practice, one typically adjusts for the average within-hospital difference associated with patient characteristics not under the control of the hospital via a linear regression mode with hospital intercepts.

  In the US inpatient setting, for example, Elliott et al. 2009 showed that self-reported health status had the greatest explanatory pow...

  Show More

  We read with interest the recent article by Rubens et al. “Impact of Patient Characteristics on the Canadian Patient Experiences Survey–Inpatient Care: Survey Analysis from an Academic Tertiary Care Centre.” This work adds to the literature by examining patient characteristics associated with response patterns on the Canadian Patient Experience Inpatient Care Survey. The authors conclude that caution is needed when comparing performance between different entities assessed by this survey, as observed differences could be explained by variation in patient mix rather than variation in performance. While we agree that patient characteristics are likely to explain some of the variation in observed patient experience scores, as has been found in similar settings in the US and UK (Elliott et al. 2009, Paddison et al. 2012), we do not believe that this is problematic for the validity of patient experience scores. Following the common practice of adjusting entity scores for patient characteristics such as self-reported health, education, and age allows for valid comparisons as if all hospitals had treated the same patient population. In practice, one typically adjusts for the average within-hospital difference associated with patient characteristics not under the control of the hospital via a linear regression mode with hospital intercepts.

  In the US inpatient setting, for example, Elliott et al. 2009 showed that self-reported health status had the greatest explanatory power, followed by education, service line and age, when examining experience measures on the Hospital Consumer Assessment of Health Care Providers and Systems (HCAHPS) Survey. Current publicly-available HCAHPS documentation describes how hospital scores are adjusted to account for differences in the self-reported health status and other characteristics of survey respondents across hospitals(HCAHPS 2018). In this instance, the authors’ work could be used to similarly adjust the scores of Canadian hospitals to produce patient experience scores that would enable valid comparisons across hospitals serving different patient populations.

   
  References

  Elliott, Marc N., Alan M. Zaslavsky, Elizabeth Goldstein, William Lehrman, Katrin Hambarsoomians, Megan K. Beckett, and Laura Giordano. 2009. "Effects of survey mode, patient mix, and nonresponse on CAHPS® hospital survey scores." Health services research 44 (2p1):501-518

  HCAHPS. Hospital Consumer Assessment of Healthcare providers and Systems (HCAHPS) Quality Assurance Guidelines, Version 12.0, March 2017. http://www.hcahpsonline.org/globalassets/hcahps/quality-assurance/qag_v1... Accessed June 1, 2018.

  Paddison, Charlotte, Marc Elliott, Richard Parker, Laura Staetsky, Georgios Lyratzopoulos, John L. Campbell, and Martin Roland. 2012. "Should measures of patient experience in primary care be adjusted for case mix? Evidence from the English General Practice Patient Survey." BMJ Qual Saf 21:634-640.

  Show Less

  Conflict of Interest:
  None declared.

  • Back to top