Goodness-of-fit tests for logistic regression models when data are collected using a complex sampling design

Department of Biostatistics, Virginia Commonwealth University, 1101 East Marshall St., B1-066, Richmond, VA 23298-0032, USA; Division of Epidemiology and Biostatistics and Department of Statistics, School of Public Health, The Ohio State University, 320 West Tenth Ave., M200 Starling-Loving Hall, Columbus, OH 43210, USA; University of Massachusetts, 128 Worcester Road, Stowe, VT 05672-4320, USA
Computational Statistics & Data Analysis 01/2007; DOI: 10.1016/j.csda.2006.07.006
Source: RePEc

ABSTRACT Logistic regression models are frequently used in epidemiological studies for estimating associations that demographic, behavioral, and risk factor variables have on a dichotomous outcome, such as disease being present versus absent. After the coefficients in a logistic regression model have been estimated, goodness-of-fit of the resulting model should be examined, particularly if the purpose of the model is to estimate probabilities of event occurrences. While various goodness-of-fit tests have been proposed, the properties of these tests have been studied under the assumption that observations selected were independent and identically distributed. Increasingly, epidemiologists are using large-scale sample survey data when fitting logistic regression models, such as the National Health Interview Survey or the National Health and Nutrition Examination Survey. Unfortunately, for such situations no goodness-of-fit testing procedures have been developed or implemented in available software. To address this problem, goodness-of-fit tests for logistic regression models when data are collected using complex sampling designs are proposed. Properties of the proposed tests were examined using extensive simulation studies and results were compared to traditional goodness-of-fit tests. A Stata ado function svylogitgof for estimating the F-adjusted mean residual test after svylogit fit is available at the author's website

  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Objective The objectives of this study are to examine racial and ethnic differences in suicidal behaviour, its main risk factors, and the effect of the risk factors on suicidal behaviour in young adults in the United States. Design Using nationally representative data (n = 10,585) from Add Health, we calculate the prevalence of suicidal behavior and associated risk factors for non-Hispanic White, non-Hispanic Black, and Hispanic youth (aged 18-26) using logistic regression models of suicidal ideation stratified by race. Results Non-Hispanic White and Hispanic young adults have higher rates of suicidal ideation than their non-Hispanic Black counterparts, but racial/ethnic differences in attempts are not statistically significant. Non-Hispanic Whites and Hispanic young adults are more likely to possess key risk factors for suicide. With the exception of substance use variables (i.e. alcohol and marijuana use) which appear to be more conducive to suicidal ideation in non-Hispanic Black than in non-Hispanic White young adults, the effects of risk factors appear to be similar across race/ethnicity. Conclusion The higher prevalence of suicidal ideation in non-Hispanic White and Hispanic young adults may be driven by their greater exposure to risk factors, as opposed to differences in the effects of these risk factors. More research is needed to uncover why non-Hispanic White and Hispanic young adults have higher rates of suicidal ideation than their non-Hispanic Black counterparts; yet, rates of suicide attempts are comparable and non-Hispanic White young adults have the highest rate of completed suicides.
    Ethnicity and Health 10/2013; · 1.20 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Objectives  The aim of this study was to identify common risk factors for patient-reported medical errors across countries. In country-level analyses, differences in risks associated with error between health care systems were investigated. The joint effects of risks on error-reporting probability were modelled for hypothetical patients with different health care utilization patterns. Design  Data from the Commonwealth Fund's 2010 lnternational Survey of the General Public's Views of their Health Care System's Performance in 11 Countries. Setting  Representative population samples of 11 countries were surveyed (total sample = 19 738 adults). Utilization of health care, coordination of care problems and reported errors were assessed. Regression analyses were conducted to identify risk factors for patients' reports of medical, medication and laboratory errors across countries and in country-specific models. Results  Error was reported by 11.2% of patients but with marked differences between countries (range: 5.4-17.0%). Poor coordination of care was reported by 27.3%. The risk of patient-reported error was determined mainly by health care utilization: Emergency care (OR = 1.7, P < 0.001), hospitalization (OR = 1.6, P < 0.001) and the number of providers involved (OR three doctors = 2.0, P < 0.001) are important predictors. Poor care coordination is the single most important risk factor for reporting error (OR = 3.9, P < 0.001). Country-specific models yielded common and country-specific predictors for self-reported error. For high utilizers of care, the probability that errors are reported rises up to P = 0.68. Conclusions  Safety remains a global challenge affecting many patients throughout the world. Large variability exists in the frequency of patient-reported error across countries. To learn from others' errors is not only essential within countries but may also prove a promising strategy internationally.
    Health expectations: an international journal of public participation in health care and health policy 02/2012; · 1.80 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: The aim of this study was to explore the role of self-beliefs in predicting postschool outcomes for deaf young adults in transition from secondary settings. Three self-level constructs were explored: self-concept, self-determination, and expectations about the future. This study utilized data from the National Longitudinal Transition Study-2 (NLTS2) that collected longitudinal data from youths with disabilities across the nation, 550 of which were deaf or hard of hearing and met the selection criteria in this study. This study examined the relationships between these deaf adolescents’ self-beliefs and actual future achievements that were reached as they transitioned to adult life, in three domains: life, employment, and education. Despite the generally positive self-beliefs of deaf individuals, which were, in some cases, related to postschool outcomes, the self-beliefs assessed in this study did not emerge as comprehensive predictors of postschool attainments. Findings suggest that for deaf individuals, successfully navigating transitions to adult life involves dimensions beyond individual agency. Positive self-beliefs are clearly a part of successfully attaining postschool outcomes, but deaf individuals may not have full access to equitable opportunities to capitalize on these beliefs.
    Journal of Developmental and Physical Disabilities 06/2014; · 0.89 Impact Factor


Available from