Review Article
The validity of administrative data to identify hip fractures is high—a systematic review

https://doi.org/10.1016/j.jclinepi.2012.10.004Get rights and content

Abstract

Objective

To determine the validity of the diagnostic algorithms for osteoporosis and fractures in administrative data.

Study Design and Setting

A systematic search was conducted to identify studies that reported the validity of a diagnostic algorithm for osteoporosis and/or fractures using administrative data.

Results

Twelve studies were reviewed. The validity of the diagnosis of osteoporosis in administrative data was fair when at least 3 years of data from hospital and physician visit claims were used (area under the receiver operating characteristic [ROC] curve [AUC] = 0.70) or when pharmacy data were used (with or without the use of hospital and physician visit claims data, AUC > 0.70). Nonetheless, the positive predictive values (PPVs) were low (<0.60). There was good evidence to support the use of hospital data to identify hip fractures (sensitivity: 69–97%; PPV: 63–96%) and the addition of physician claims diagnostic and procedural codes to hospitalization diagnostic codes improved these characteristics (sensitivity: 83–97%; PPV: 86–98%). Vertebral fractures were difficult to identify using administrative data. There was some evidence to support the use of administrative data to define other fractures that do not require hospitalization.

Conclusions

Administrative data can be used to identify hip fractures. Existing diagnostic algorithms to identify osteoporosis and vertebral fractures in administrative data are suboptimal.

Introduction

What is new?

  • This is the first systematic review to be conducted to determine the validity of diagnostic algorithms for osteoporosis and fractures in administrative data.

  • Findings from this systematic review suggest that administrative data can be used to identify hip fractures.

  • However, existing diagnostic algorithms to identify osteoporosis and vertebral fractures in administrative data are suboptimal and require further development.

Osteoporosis is a disease characterized by decreased bone mass and increased fracture risk. It is associated with significant morbidity, including fractures, impaired health-related quality of life, and increased mortality [1], [2]. In addition, the burden of disease related to osteoporosis and fractures is projected to increase markedly as the population ages. As such, osteoporosis and fracture risk have become important public health problems warranting surveillance [2].

Health administrative data are defined as information passively collected, often by government and health care providers, for the purpose of managing health care delivery. Often, the databases exist primarily for reimbursement purposes [3]. In Canada (as in other countries), health administrative databases record population-level data (because almost all citizens are covered by the physician billing and hospital databases in each province). Different administrative databases contain information on important variables, including demographic characteristics, hospitalizations, in- and outpatient physician visits and services, filled prescriptions, and vital status. Health administrative data have thus been used to pursue population-level health outcomes research and disease surveillance. In the field of osteoporosis, health administrative data have been used to study the epidemiology of the disease and as a tool for disease surveillance [4], [5], [6], [7], [8], [9].

Because health administrative databases are generally not designed for the purposes of conducting health research per se, health researchers using these sources of data had to develop definitions and/or algorithms to identify the diagnoses of interest in the databases ad hoc. Not surprisingly, though, the results and interpretation of those studies may be affected by the validity of the given definitions and/or algorithms. Because administrative data have the potential to be a rich source of data to pursue population-based research in osteoporosis and fracture risk assessment, we wished to review the validity of existing algorithms to diagnose osteoporosis and fractures in administrative data. We formulated the following tentative question: How valid are diagnoses of osteoporosis and hip, vertebral, and other fractures in administrative databases?

Section snippets

Methods

We undertook a systematic review of the literature concerning the validity of diagnoses of osteoporosis and fractures in administrative data. The study is reported according to the preferred reporting items for systematic reviews and meta-analyses (PRISMA) guidelines (http://www.prisma-statement.org). The authors planned and designed the study, and defined the variables of interest and the search strategy by consensus, with the assistance of a professional librarian. Details of the study

Results

The electronic search yielded 208 unique citations (Fig. 1). From those, 190 were excluded at the stage of title and abstract review. The kappa coefficient between reviewers at this phase was 0.79, demonstrating substantial agreement [11]. Differences were resolved by consensus. Among the 18 studies identified for full-text review, 11 were excluded (insufficient data to calculate the test characteristics of the diagnostic algorithms [n = 5], no validation performed [n = 2], report not in English or

Discussion

In this systematic review of the literature, we found that algorithms to identify hip fractures in administrative data appear valid, and test characteristics improve when physician claims and procedural codes are used in addition to hospitalization data. Algorithms to identify osteoporosis and vertebral fractures in administrative data are available but have limited sensitivity and low PPVs. This should be acknowledged when the current algorithms are used and further research should be

References (29)

  • R. Spasoff

    Epidemiologic methods for health policy

    (1999)
  • W.D. Leslie et al.

    Trends in hip fracture rates in Canada

    JAMA

    (2009)
  • P. Vestergaard et al.

    Osteoporosis is markedly underdiagnosed: a nationwide study from Denmark

    Osteoporos Int

    (2005)
  • K.M. Sanders et al.

    Age- and gender-specific rate of fractures in Australia: a population-based study

    Osteoporos Int

    (1999)
  • Cited by (70)

    View all citing articles on Scopus

    Funding: This study was funded in part by the Canadian Arthritis Network. Dr Hudson is supported by New Investigator awards from the Canadian Institutes of Health Research and the Fonds de la Recherche en Santé du Québec. Dr Lacaille holds the Mary Pack Chair in Arthritis Research from UBC and The Arthritis Society of Canada. The funding sources had no role in the design of the study, analysis of the data, preparation of the manuscript, and decision to submit for publication.

    View full text