For Publishers
DiscoveryMetadataPeer reviewHostingPublishing
For Researchers
JoinPublishReviewCollect
Register
Blog
About
Search
Advanced search
My ScienceOpen
Search
For Publishers
For Researchers
Blog
About
21
views
7
references
 Top references
cited by
23
    
0 reviews
 Review
0
comments
 Comment
0
recommends
+1 Recommend
0 collections
    0
    shares
      318
      similar
       All similar
      • Record: found
      • Abstract: found
      • Article: found
      Is Open Access

      Addressing Missing Data in Patient‐Reported Outcome Measures (PROMS): Implications for the Use of PROMS for Comparing Provider Performance

      research-article

      Read this article at

      ScienceOpenPublisherPMC
      Bookmark
          There is no author summary for this article yet. Authors can add summaries to their articles on ScienceOpen to make them more accessible to a non-specialist audience.

          Abstract

          Patient‐reported outcome measures (PROMs) are now routinely collected in the English National Health Service and used to compare and reward hospital performance within a high‐powered pay‐for‐performance scheme. However, PROMs are prone to missing data. For example, hospitals often fail to administer the pre‐operative questionnaire at hospital admission, or patients may refuse to participate or fail to return their post‐operative questionnaire. A key concern with missing PROMs is that the individuals with complete information tend to be an unrepresentative sample of patients within each provider and inferences based on the complete cases will be misleading. This study proposes a strategy for addressing missing data in the English PROM survey using multiple imputation techniques and investigates its impact on assessing provider performance. We find that inferences about relative provider performance are sensitive to the assumptions made about the reasons for the missing data. © 2015 The Authors. Health Economics Published by John Wiley & Sons Ltd.

          Related collections

          Most cited references7

          • Record: found
          • Abstract: found
          • Article: found
          Is Open Access

          Factors associated with non-response in routine use of patient reported outcome measures after elective surgery in England

          Andrew Hutchings, Jenny Neuburger, Kirstin Grosse Frie … (2012)
          Background Patient-reported outcome measures are increasingly being used to compare providers. We studied whether non-response rates to post-operative questionnaires are associated with patients' characteristics and organisational features of providers. Methods 131 447 patients who underwent a hip or knee replacement, hernia repair or varicose vein surgery in 2009-10 in England. Multivariable logistic regression to calculate adjusted odds ratios of non-response for characteristics of the patients and organisational characteristics of providers. Multiple imputation was used for missing patient characteristics. Providers were included as random effects. Results Response rates to the post-operative questionnaire were 85.1% for hip replacement (n = 37 961), 85.3% for knee replacements (n = 44 422), 72.9% for hernia repair (n = 34 964), and 64.8% for varicose vein surgery (n = 14 100). Across the four procedures, there were higher levels of non-response in men (odds ratios 1.03 [95% CI 0.95-1.11] - 1.35 [1.25-1.46]), younger patients (those under 55 years 3.01 [2.72-3.32] - 6.05 [5.49-6.67]), non-white patients (1.24 [1.11-1.38] - 2.08 [1.89-2.31]), patients in the most deprived quintile of socio-economic status (1.47 [1.34-1,62] - 1.86 [1.71-2.03]), those who lived alone (1.11 [0.99-1.23] - 1.27 [1.18-1.36]) and those who had been assisted when completing their pre-operative questionnaire (1.26 [1.10-1.46] -1.67 [1.56-1.79]). Non-response rates were also higher in patients who had poorer pre-operative health (three or more comorbidities: 1.14 [0.96-1.35] - 1.45 [1.30-1.63]). Providers' patient recruitment rates before surgery and the timing of pre-operative questionnaire administration did not affect the rates of response to post-operative questionnaires. Conclusion If non-response can be shown to be associated with outcome, then rates of non-response to post-operative questionnaires would need to be taken into account when these measures are being used to compare the performance of providers or to evaluate surgical procedures.
          Article 0 comments Cited 48 times – based on 0 reviews     Review now
            Bookmark
            • Record: found
            • Abstract: found
            • Article: not found

            Nonresponse bias in a survey of patient perceptions of hospital care.

            Eric Chamot, T Perneger, Patrick Bovier (2005)
            Incomplete participation is of particular concern for surveys of patient perceptions of care because patients who have negative opinions may be least likely to participate. We sought to examine indirect evidence of nonresponse bias. We re-analyzed data from a cross-sectional patient survey. Our subjects were patients discharged from a Swiss hospital (n = 2156). We measured the following: (1) an observed problem score, based on 15 key items of the Picker Patient Experience questionnaire, (2) a predicted problem score, and (3) a participation propensity score. The latter scores were computed for all eligible patients, including those who did not return the survey, from routinely available baseline data. The participation rate was 70% (n = 1518), and the mean problem score was 29.9 (SD 23.8). Early respondents reported significantly fewer problems than late respondents (28.6 versus 32.9, P = 0.001). Participation propensity scores were progressively lower in early respondents (mean 74.2), late respondents (70.7), and nonrespondents (63.9, P < 0.001); the pattern was similar for predicted problem scores (early respondents: 29.5; late respondents: 30.5; nonrespondents: 33.4, P < 0.001). The propensity to participate was negatively associated with the problem score (Pearson r = -0.19). Finally, predictors of participation were similar to predictors of problem scores. The tendency to participate in the survey was negatively associated with the report of problems during hospitalization. Nevertheless, increasing participation from 30% to 70% had only a modest influence on the final conclusions of the survey.
            Article 0 comments Cited 38 times – based on 0 reviews     Review now
              Bookmark
              • Record: found
              • Abstract: found
              • Article: not found

              Multiple imputation methods for handling missing data in cost-effectiveness analyses that use data from hierarchical studies: an application to cluster randomized trials.

              Manuel Gomes, Richard Grieve, Karla Diaz-Ordaz … (2013)
              Multiple imputation (MI) has been proposed for handling missing data in cost-effectiveness analyses (CEAs). In CEAs that use cluster randomized trials (CRTs), the imputation model, like the analysis model, should recognize the hierarchical structure of the data. This paper contrasts a multilevel MI approach that recognizes clustering, with single-level MI and complete case analysis (CCA) in CEAs that use CRTs. We consider a multilevel MI approach compatible with multilevel analytical models for CEAs that use CRTs. We took fully observed data from a CEA that evaluated an intervention to improve diagnosis of active labor in primiparous women using a CRT (2078 patients, 14 clusters). We generated scenarios with missing costs and outcomes that differed, for example, according to the proportion with missing data (10%-50%), the covariates that predicted missing data (individual, cluster-level), and the missingness mechanism: missing completely at random (MCAR), missing at random (MAR), or missing not at random (MNAR). We estimated incremental net benefits (INBs) for each approach and compared them with the estimates from the fully observed data, the "true" INBs. When costs and outcomes were assumed to be MCAR, the INBs for each approach were similar to the true estimates. When data were MAR, the point estimates from the CCA differed from the true estimates. Multilevel MI provided point estimates and standard errors closer to the true values than did single-level MI across all settings, including those in which a high proportion of observations had cost and outcome data MAR and when data were MNAR. Multilevel MI accommodates the multilevel structure of the data in CEAs that use cluster trials and provides accurate cost-effectiveness estimates across the range of circumstances considered.
              Article 0 comments Cited 24 times – based on 0 reviews     Review now
                Bookmark
                All references

                Author and article information

                Journal
                Journal ID (nlm-ta): Health Econ
                Journal ID (iso-abbrev): Health Econ
                Journal ID (doi): 10.1002/(ISSN)1099-1050
                Journal ID (publisher-id): HEC
                Title: Health Economics
                Publisher: John Wiley and Sons Inc. (Hoboken )
                ISSN (Print): 1057-9230
                ISSN (Electronic): 1099-1050
                Publication date (Electronic): 05 March 2015
                Publication date (Print): May 2016
                Volume: 25
                Issue: 5 ( doiID: 10.1002/hec.v25.5 )
                Pages: 515-528
                Affiliations
                [ 1 ] Department of Health Services Research and PolicyLondon School of Hygiene and Tropical Medicine LondonUK
                [ 2 ] Centre for Health EconomicsUniversity of York YorkUK
                Author notes
                [*] [* ] Correspondence to: Department of Health Services Research and Policy, London School of Hygiene and Tropical Medicine, 15‐17 Tavistock Place, London WC1H 9SH, UK. E‐mail: manuel.gomes@ 123456lshtm.ac.uk



                Article
                Publisher ID: HEC3173 Other ID: HEC-14-0313.R1
                DOI: 10.1002/hec.3173
                PMC ID: 4973682
                PubMed ID: 25740592
                SO-VID: 035cabb6-cf6d-4736-bbe5-b708c339bec7
                Copyright © © 2015 The Authors. Health Economics Published by John Wiley & Sons Ltd.
                License:

                This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.

                History
                Date received : 27 June 2014
                Date revision received : 27 November 2014
                Date accepted : 10 February 2015
                Page count
                Pages: 14
                Categories
                Subject: Research Article
                Subject: Research Articles
                Custom metadata
                source-schema-version-number 2.0
                component-id hec3173
                header-id hec3173-hdr-0001
                cover-date May 2016
                details-of-publishers-convertor Converter:WILEY_ML3GV2_TO_NLMPMC version:4.9.4 mode:remove_FC converted:04.08.2016

                ScienceOpen disciplines: Economics of health & social care
                Keywords: missing data,multiple imputation,patient‐reported outcome measures,provider performance,missing not at random
                Data availability:
                ScienceOpen disciplines: Economics of health & social care
                Keywords: missing data, multiple imputation, patient‐reported outcome measures, provider performance, missing not at random

                Comments

                Comment on this article

                scite_

                Similar content318

                See all similar

                Cited by22

                See all cited by

                Most referenced authors107

                See all reference authors