Predictive performance of comorbidity measures in administrative databases for diabetes cohorts

Abstract

BackgroundThe performance of comorbidity measures for predicting mortality in chronic disease populations and using ICD-9 diagnosis codes in administrative health data has been investigated in several studies, but less is known about predictive performance with ICD-10 data and for other health outcomes. This study investigated predictive performance of five comorbidity measures for population-based diabetes cohorts in administrative data. The objectives were to evaluate performance for: (a) disease-specific and general health outcomes, (b) data based on the ICD-9 and ICD-10 diagnoses, and (c) different age groups.MethodsPerformance was investigated for heart attack, stroke, amputation, renal disease, hospitalization, and death in all-age and age-specific cohorts. Hospital records, physician billing claims, and prescription drug records from one Canadian province were used to identify diabetes cohorts and measure comorbidity. The data were analysed using multiple logistic regression models and summarized using measures of discrimination, accuracy, and fit.ResultsIn Cohort 1 (n = 29,058), for which only ICD-9 diagnoses were recorded in administrative data, the Elixhauser index showed good or excellent prediction for amputation, renal disease, and death and performed better than the Charlson index. Number of diagnoses was a good predictor of hospitalization. Similar results were obtained for Cohort 2 (n = 41,925), in which both ICD-9 and ICD-10 diagnoses were recorded in administrative data, although predictive performance was sometimes higher. For age-specific models of mortality, the Elixhauser index resulted in the largest improvement in predictive performance in all but the youngest age group.ConclusionsCohort age and the health outcome under investigation, but not the diagnosis coding system, may influence the predictive performance of comorbidity measure for studies about diabetes populations using administrative health data.

Health and social predictors of applications to public housing: a population-based analysis

“…Descriptive and inference statistics were used to assess goodness of fit, including the Likelihood Ratio Test (LRT),32 33 which follows a χ 2 distribution with degrees of freedom equal to the difference in degrees of freedom between nested models.…”

Risk factors for infection, revision, death, blood transfusion and longer hospital stay 3 months and 1 year after primary total hip or knee arthroplasty

“…Most of the c-statistic values were close to or greater than 0.80 for each of the mortality outcomes, which suggests that each is a strong model in terms of discrimination. Lix and colleagues 19 suggested that a Brier score less than 0.25 is an acceptable prediction error where the smaller the value, the less the error. The Brier score for all models was well below this threshold.…”

Transcatheter aortic valve replacement in patients with systolic heart failure

“…11 The two coding systems also performed equally well in predicting 30-day and 1-year mortality using acute myocardial infarction mortality prediction rules. 12 Additional studies verified the ability of ICD-10-CM comorbidity measures such as the Elixhauser measures and the Charlson index to predict the occurrence of medical conditions or death in patients with diabetes 13 or to predict in-hospital case fatality in patients who had a stroke. 14 This report focuses on important issues in the ICD-10-CM/PCS transition for researchers.…”