Elsevier

Annals of Epidemiology

Volume 14, Issue 7, August 2004, Pages 507-516
Annals of Epidemiology

Review of the performance of methods to identify diabetes cases among vital statistics, administrative, and survey data

https://doi.org/10.1016/j.annepidem.2003.09.016Get rights and content

Abstract

Purpose

The ability to identify prevalent cases of diagnosed diabetes is crucial to monitoring preventative care practices and health outcomes among persons with diagnosed diabetes.

Methods

We conducted a comprehensive literature review to assess and summarize the validity of various strategies for identifying individuals with diagnosed diabetes and to examine the factors influencing the validity of these strategies.

Results

We found that studies using either administrative data or survey data were both adequately sensitive (i.e., identified the majority of cases of diagnosed diabetes) and highly specific (i.e., did not identify the individuals as having diabetes if they did not). In contrast, studies based on cause-of-death data from death certificates were not sensitive, failing to identify about 60% of decedents with diabetes and in most of these studies, researchers did not report specificity or positive predictive value.

Conclusions

Surveillance is critical for tracking trends in diabetes and targeting diabetes prevention efforts. Several approaches can provide valuable data, although each has limitations. By understanding the limitations of the data, investigators will be able to estimate diabetes prevalence and improve surveillance of diabetes in the population.

Introduction

Diabetes is a serious and growing public health problem in the United States 1., 2., 3.. Public health surveillance of diabetes and its complications is critical for defining the burden of the disease, identifying high-risk groups, developing strategies to reduce the economic burden and human cost of diabetes, formulating health-care policy, and evaluating progress in disease prevention and control 4., 5., 6.. Health-care organizations, such as managed care, have also developed systems to track and monitor patients with diabetes (4). Such systems were designed to decrease costs, clarify health-care utilization patterns, monitor health-care outcomes, implement case-management programs, evaluate quality of care improvement efforts, and address accountability demands from accreditation agencies, employers, and other purchasers of health care.

Public health and managed-care organizations typically use multiple data sources to track and monitor chronic diseases such as diabetes 4., 7., 8., 9.. These include vital statistics, survey data, and administrative databases (e.g., billing data such as Medicare data, hospitalization data, physician encounter data, and pharmacy data). Because several effective, but underused, preventive-care practices are known to reduce complications related to diabetes, a major focus of diabetes surveillance efforts has been on preventive-care practices and health outcomes among persons with diagnosed diabetes. Understanding the performance of different surveillance methods to accurately identify prevalent cases of diagnosed diabetes is important in these surveillance efforts. However, strategies to identify individuals with diagnosed diabetes from various data sources have not been systematically evaluated. Therefore, we conducted a comprehensive literature review to summarize the validity of various methods for identifying individuals with diagnosed diabetes and to examine the factors influencing the validity within each study.

Section snippets

Methods

We searched Medline for articles published from 1966 through mid-2002 that reported a measure of validity of identifying persons with diabetes, such as sensitivity, specificity, positive predictive value, and kappa, or that contained sufficient data to allow us to calculate those measures. Since this is a review of the available literature, we did not assess the reported validity in each article.

We limited our search to English language articles from developed countries and excluded articles

Death certificates

We found 17 articles on studies that examined the sensitivity of cause-of-death data from death certificates in identifying decedents with diabetes 21., 22., 23., 24., 25., 26., 27., 28., 29., 30., 31., 32., 33., 34., 35., 36., 37. (Table 1). None of the 17 examined the specificity or positive predictive value (PPV) of the cause-of-death data. The sensitivity of cause-of-death data in these studies ranged from 7.1% to 68.6% with a median of 36%. As would be expected, studies using any listed

Discussion

In this review we found that administrative databases and health surveys did a relatively good job of identifying persons with diagnosed diabetes. We found these data sources adequately sensitive, (i.e., identified the majority of persons with diagnosed diabetes), and highly specific (i.e., did not identify individuals as having diabetes when they did not). In contrast, cause-of-death data from death certificates were not sensitive, failing to identify roughly 60% of decedents with diabetes.

References (68)

  • A.D. Morris et al.

    Adherence to insulin treatment, glycemic control, and ketoacidosis in insulin-dependent diabetes mellitus. The DARTS/MEMO Collaboration. Diabetes Audit and Research in Tayside Scotland. Medicines Monitoring Unit

    Lancet

    (1997)
  • W.S. Weintraub et al.

    Can cardiovascular clinical characteristics be identified and outcome models be developed from an in-patient claims database?

    Am J Cardiol

    (1999)
  • D.M. Kriegsman et al.

    Self-report and general practitioner information on the presence of chronic disease in community-dwelling elderly: A study of the accuracy of patients' self-reports and on determinants of inaccuracy

    J Clin Epidemiol

    (1996)
  • L.M. Martin et al.

    Validation of self-reported chronic conditions and health services in a managed care population

    Am J Prev Med

    (2000)
  • S.A. Kaye et al.

    Increased incidence of diabetes mellitus in relation to abdominal adiposity in older women

    J Clin Epidemiol

    (1991)
  • M.I. Harris et al.

    Prevalence of diabetes, impaired fasting glucose and impaired glucose tolerance in US adults. The Third National Health and Nutrition Examination Survey, 1988–1994

    Diabetes Care

    (1998)
  • H. King et al.

    Global burden of diabetes, 1995–2025: Prevalence, numerical estimates, and projections

    Diabetes Care

    (1998)
  • American Diabetes Association

    Economic consequences of diabetes mellitus in the US in 1997

    Diabetes Care

    (1998)
  • M.M. Engelgau et al.

    Use of services by diabetes patients in managed care organizations. Development of a diabetes surveillance system. CDC Diabetes in Managed Care Work Group

    Diabetes Care

    (1998)
  • M.A. Safran et al.

    The war against diabetes: How will we know if we are winning?

    Diabetes Care

    (1999)
  • Centers for Disease Control and Prevention (CDC)

    Preventive-care practices among persons with diabetes—United States, 1995 and 2001

    MMWR Morb Mortal Wkly Rep

    (2002)
  • S.B. Thacker et al.

    Public health surveillance for chronic conditions: A scientific basis for decisions

    Stat Med

    (1995)
  • S.M. Teutsch

    Considerations in planning a surveillance system

  • R.A. Goodman et al.

    Evaluating public health surveillance

  • M. Beckett et al.

    Do health interview surveys yield reliable data on chronic illness among older respondents?

    Am J Epidemiol

    (2000)
  • T. Bush et al.

    Self-report and medical record report agreement of selected medical conditions in the elderly

    Am J Public Health

    (1989)
  • A. Gittelson et al.

    Studies on the reliability of vital and health records: I. Comparison of cause of death and hospital discharge diagnoses

    Am J Public Health

    (1979)
  • A.M. Lagaay et al.

    Validation of medical history taking as part of a population based survey in subjects aged 85 and over

    BMJ

    (1992)
  • S. Shea et al.

    Reliability of the behavioral risk factor survey in a triethnic population

    Am J Epidemiol

    (1991)
  • A.D. Stein et al.

    Reproducibility of responses to telephone interviews: Demographic predictors of discordance in risk factor status

    Am J Epidemiol

    (1995)
  • A.D. Stein et al.

    The Behavioral Risk Factor Surveillance System questionnaire: Its reliability in a statewide sample

    Am J Public Health

    (1993)
  • S. Tretli et al.

    Reliability of questionnaire information on cardiovascular disease and diabetes: Cardiovascular disease study in Finnmark County

    J Epidemiol Community Health

    (1982)
  • K. Rockwood et al.

    Incidence and outcomes of diabetes mellitus in elderly people: Report from the Canadian Study of Health and Aging

    CMAJ

    (2000)
  • E.M. Andresen et al.

    Underreporting of diabetes on death certificates, King County, Washington

    Am J Public Health

    (1993)
  • Cited by (128)

    • The association between oil spill cleanup-related total hydrocarbon exposure and diabetes

      2022, Environmental Research
      Citation Excerpt :

      Strengths of this analysis of oil spill workers include its use of THC exposure estimates, based on measurements, as opposed to using job title as a proxy for exposure, the ability to identify undiagnosed DM cases via measured HbA1c, a relatively large sample size, and the ability to control for important demographic, lifestyle, and anthropometric factors, including measured BMI. In addition, it examined incident/newly identified diabetes cases rather than diabetes mortality, allowing us to account for well-managed diabetes among the still living and avoiding the documented under-ascertainment, possibly differential, resulting from use of cause of death data from vital records (Centers for Disease Control, 2017; Cheng et al., 2008; McEwen et al., 2006; Saydah et al., 2004). It is also the first study of the association between oil spill response and cleanup-related exposures and risk of DM.

    View all citing articles on Scopus
    View full text