Display options
Share it on

Qual Life Res. 2015 Mar;24(3):521-8. doi: 10.1007/s11136-014-0746-0. Epub 2014 Jul 10.

Method variation in the impact of missing data on response shift detection.

Quality of life research : an international journal of quality of life aspects of treatment, care and rehabilitation

Carolyn E Schwartz, Tolulope T Sajobi, Mathilde G E Verdam, Veronique Sebille, Lisa M Lix, Alice Guilleux, Mirjam A G Sprangers

Affiliations

  1. DeltaQuest Foundation, Inc, Concord, MA, USA, [email protected].

PMID: 25008260 DOI: 10.1007/s11136-014-0746-0

Abstract

PURPOSE: Missing data due to attrition or item non-response can result in biased estimates and loss of power in longitudinal quality-of-life (QOL) research. The impact of missing data on response shift (RS) detection is relatively unknown. This overview article synthesizes the findings of three methods tested in this special section regarding the impact of missing data patterns on RS detection in incomplete longitudinal data.

METHODS: The RS detection methods investigated include: (1) Relative importance analysis to detect reprioritization RS in stroke caregivers; (2) Oort's structural equation modeling (SEM) to detect recalibration, reprioritization, and reconceptualization RS in cancer patients; and (3) Rasch-based item-response theory-based (IRT) models as compared to SEM models to detect recalibration and reprioritization RS in hospitalized chronic disease patients. Each method dealt with missing data differently, either with imputation (1), attrition-based multi-group analysis (2), or probabilistic analysis that is robust to missingness due to the specific objectivity property (3).

RESULTS: Relative importance analyses were sensitive to the type and amount of missing data and imputation method, with multiple imputation showing the largest RS effects. The attrition-based multi-group SEM revealed differential effects of both the changes in health-related QOL and the occurrence of response shift by attrition stratum, and enabled a more complete interpretation of findings. The IRT RS algorithm found evidence of small recalibration and reprioritization effects in General Health, whereas SEM mostly evidenced small recalibration effects. These differences may be due to differences between the two methods in handling of missing data.

CONCLUSIONS: Missing data imputation techniques result in different conclusions about the presence of reprioritization RS using the relative importance method, while the attrition-based SEM approach highlighted different recalibration and reprioritization RS effects by attrition group. The IRT analyses detected more recalibration and reprioritization RS effects than SEM, presumably due to IRT's robustness to missing data. Future research should apply simulation techniques in order to make conclusive statements about the impacts of missing data according to the type and amount of RS.

References

  1. J Palliat Med. 2004 Apr;7(2):233-45 - PubMed
  2. BMC Med Res Methodol. 2013 Feb 15;13:20 - PubMed
  3. Qual Life Res. 2011 Dec;20(10):1543-53 - PubMed
  4. Ann Oncol. 1999 Jul;10(7):775-82 - PubMed
  5. Qual Life Res. 2012 Feb;21(1):1-11 - PubMed
  6. Eur J Cancer. 2001 Sep;37(14):1729-35 - PubMed
  7. Psychooncology. 1999 Jul-Aug;8(4):344-54 - PubMed
  8. Health Qual Life Outcomes. 2005 Mar 30;3:21 - PubMed
  9. Qual Life Res. 2015 Mar;24(3):529-40 - PubMed
  10. Health Psychol. 1999 May;18(3):211-20 - PubMed
  11. Qual Life Res. 2011 Dec;20(10):1555-60 - PubMed
  12. Health Qual Life Outcomes. 2005 Sep 07;3:55 - PubMed
  13. Med Decis Making. 2000 Apr-Jun;20(2):186-93 - PubMed
  14. Nurs Res. 2000 May-Jun;49(3):176-80 - PubMed
  15. J Clin Epidemiol. 2002 Feb;55(2):176-83 - PubMed
  16. Diabet Med. 1993 Nov;10(9):851-4 - PubMed
  17. Qual Life Res. 2015 Mar;24(3):553-64 - PubMed
  18. Psychometrika. 2011 Apr 1;76(2):306-317 - PubMed
  19. Arch Intern Med. 2002 Jul 22;162(14):1611-8 - PubMed
  20. J Gerontol. 1993 May;48(3):P127-36 - PubMed
  21. Soc Sci Med. 1999 Jun;48(11):1549-61 - PubMed
  22. J Bone Joint Surg Am. 2006 Dec;88(12):2590-5 - PubMed
  23. Qual Life Res. 2005 Apr;14(3):611-27 - PubMed
  24. Qual Life Res. 2005 Apr;14(3):587-98 - PubMed
  25. Med Decis Making. 1998 Jul-Sep;18(3):278-86 - PubMed
  26. Qual Life Res. 2006 Nov;15(9):1533-50 - PubMed
  27. Qual Life Res. 2013 May;22(4):695-703 - PubMed
  28. Soc Sci Med. 1993 Aug;37(3):295-304 - PubMed
  29. Qual Life Res. 2000;9(6):603-15 - PubMed
  30. Acta Oncol. 1999;38(6):709-18 - PubMed
  31. Qual Life Res. 2011 Dec;20(10):1561-72 - PubMed
  32. Qual Life Res. 2005 Dec;14(10):2247-57 - PubMed
  33. Med Decis Making. 1990 Jan-Mar;10(1):58-67 - PubMed
  34. Qual Life Res. 2011 Dec;20(10):1527-40 - PubMed
  35. J Clin Epidemiol. 2009 Nov;62(11):1138-47 - PubMed
  36. Qual Life Res. 2005 Apr;14(3):599-609 - PubMed
  37. Qual Life Res. 2002 May;11(3):207-21 - PubMed
  38. J Palliat Med. 2005 Feb;8(1):121-35 - PubMed
  39. Br J Clin Psychol. 1995 May;34(2):289-300 - PubMed
  40. Soc Sci Med. 1991;32(9):981-7 - PubMed
  41. Qual Life Res. 2015 Mar;24(3):541-51 - PubMed

MeSH terms

Publication Types

Grant support