A systematic review identifies valid comorbidity indices derived from administrative health data

Yurkovich, Marko; Aviña‐Zubieta, J Antonio; Thomas, Jamie M.; Gorenchtein, Mike; Lacaille, Diane

doi:10.1016/j.jclinepi.2014.09.010

Cited by 272 publications

(262 citation statements)

References 88 publications

Supporting

Mentioning

250

Contrasting

Unclassified

Order By: Relevance

“…Maternal comorbidities were compared in women who did and did not report influenza immunization via the Elixhauser index, a comorbidity index comprising a comprehensive set of 30 comorbidities defined with ICD-9-CM codes from administrative data. 18 Relative risk (RR) and 95% CI were used to compare immunized and unimmunized mothers across the primary infant outcomes of interest: presence of ILI, laboratory-confirmed influenza, and hospitalization with laboratory-confirmed influenza. A multivariable Poisson regression model with robust error variance was used to compare the RR of infant influenza-related outcomes based on maternal influenza vaccine status, adjusted for maternal characteristics including age, birth month, insurance type, residence location, and hospital region.…”

Section: Discussionmentioning

confidence: 99%

Influenza in Infants Born to Women Vaccinated During Pregnancy

et al. 2016

View full text Add to dashboard Cite

show abstract

Section: Discussionmentioning

confidence: 99%

Influenza in Infants Born to Women Vaccinated During Pregnancy

et al. 2016

View full text Add to dashboard Cite

show abstract

“…We strongly suggest that measurements of comorbidities such as the Elixhauser Index or the Romano adaptation of the Charlson Index be added to the PPMI study and to similar database studies in the future. 29 Finally, about 25% of our patients had a family history of PD and it is possible that some of them were carriers of a genetic mutation. Many genetic forms have an influence on PD phenotype and therefore the presence of patients with genetic forms of PD could be a possible confounding factor in our study.…”

Section: Allmentioning

confidence: 92%

Age at onset and Parkinson disease phenotype

et al. 2016

View full text Add to dashboard Cite

Objective: To explore clinical phenotype and characteristics of Parkinson disease (PD) at different ages at onset in recently diagnosed patients with untreated PD. Methods:We have analyzed baseline data from the Parkinson's Progression Markers Initiative database. Four hundred twenty-two patients with a diagnosis of PD confirmed by DaTSCAN imaging were divided into 4 groups according to age at onset (onset younger than 50 years, 50-59 years, 60-69 years, and 70 years or older) and investigated for differences in side, type and localization of symptoms, occurrence/severity of motor and nonmotor features, nigrostriatal function, and CSF biomarkers.Results: Older age at onset was associated with a more severe motor and nonmotor phenotype, a greater dopaminergic dysfunction on DaTSCAN, and reduction of CSF a-synuclein and total tau. The most common presentation was the combination of 2 or 3 motor symptoms (bradykinesia, resting tremor, and rigidity) with rigidity being more common in the young-onset group. In about 80% of the patients with localized onset, the arm was the most affected part of the body, with no difference across subgroups. Conclusions:Although the presentation of PD symptoms is similar across age subgroups, the severity of motor and nonmotor features, the impairment of striatal binding, and the levels of CSF biomarkers increase with age at onset. The variability of imaging and nonimaging biomarkers in patients with PD at different ages could hamper the results of future clinical trials.

show abstract

“…Prior studies have reported that when the objective is to predict mortality, disease prevalence captured from multiple health care settings generally performs better (higher c-statistic values) than disease prevalence assessments limited to a single health care setting. 4,15,28,29 For this reason, diagnostic ascertainment in this study included information recorded in inpatient, outpatient, and emergency care settings. …”

Section: Look Back Periodsmentioning

confidence: 99%

“…One reason cited for expanding the look back period beyond a single health encounter is to limit discharge coding bias, whereby existing and less serious chronic health conditions are less likely to be recorded within EHRs during acute health encounters and vice versa. 1,14 In a recent systematic review, Yurkovich et al 15 concluded that combining health data from outpatient and inpatient encounters yields comorbidity indices with better predictive performance for mortality outcomes than inpatient data alone. Evidence also suggests that longer look back is associated with comorbidity indices with better predictive performance for a wide array of health outcomes.…”

Section: Fortin Et Almentioning

confidence: 99%

“…In this study, mortality outcomes were selected because they are the most commonly used end points in validation studies of comorbidity measures and allow for cross-study comparisons. 15 However, alternative outcomes, such as hospital readmissions, have been shown to be better predicted by longer look back periods and would require additional validation. 17 While implementing the VW point system, we observed an inverse relationship between several comorbidity groups and the mortality outcomes.…”

Section: Health Condition(s) Index Only or (Ci) Points 1-year Look Bmentioning

confidence: 99%

See 1 more Smart Citation

Optimal look back period and summary method for Elixhauser comorbidity measures in a US population-based electronic health record database

Fortin¹,

Crispo²,

Cohen³

et al. 2017

OAMS

View full text Add to dashboard Cite

Background: Comorbidity risk-adjustment tools are widely used in health database research to control for clinical differences between individuals, but they need to be validated a priori. This study aimed to identify the optimal parameters for predicting all-cause inhospital mortality using Quan's enhanced Elixhauser comorbidity measures (ECMs) in the US-based Cerner Health Facts ® (HF) electronic health record database. Methods: Health care recipients aged 18-89 years between 2002 and 2011 were included. Prevalent comorbidities recorded, 1) during the index encounter; 2) in the prior year; and 3) in the prior 2 years were identified using the ECMs. Multiple logistic regression models, with inhospital mortality at index and at 1 year as the predicted outcomes, were fitted with comorbidities summarized as binary indicators, total counts, or weighted scores for the three look back periods. Baseline variables included sex and age. The receiver operating characteristic (ROC) curves of the competing models were compared with a non-parametric Mann-Whitney U test to identify the optimal parameters. Results: A sample of 3,273,298 unique health care recipients were included, of whom 31,298 (1.0%) and 50,215 (1.5%) died during the index encounter and within the 1-year follow-up, respectively. Models of comorbidity based on binary and weighted indicators had near-identical performance and were statistically better than the models based on total counts (p < 0.0001). Discrimination of inhospital mortality was highest with a look back period limited to the index encounter, while inhospital mortality at 1 year was best predicted with 1 year of look back (p < 0.0001). Conclusion: In Cerner HF, the binary and weighted methods for summarizing the Quan ECM were the best predictors of all-cause inhospital mortality at index and at 1 year. Observed differences in predictive performance between models with diagnostic ascertainment periods of up to 2 years of look back were statistically significant but not practically important. Keywords: comorbidity, ICD-9, electronic health records, risk adjustment, mortality, statistical modeling IntroductionRisk-adjustment measures of patient comorbidity are commonly used in health database research and are associated with short-and long-term mortality, hospital costs, inpatient length of stay (LOS), physician visits, and hospital readmissions. [1][2][3][4][5][6][7][8][9] The strong discriminatory performance of two Elixhauser comorbidity measure (ECM) variants, by Quan et al 10 and the Agency for Healthcare Research and Quality (AHRQ), 11 for predicting inhospital mortality was recently confirmed in the Cerner Health Facts ® (HF) database. 12 HF is a longitudinal electronic health record (EHR) data source populated by health care centers located across the continental US in compliance with the US Health Insurance Portability and Accountability Act (HIPAA). HIPAA-compliant data repositories must adhere to strict patient de-identification procedures to ensure the privacy and protection of personal ...

show abstract

A systematic review identifies valid comorbidity indices derived from administrative health data

Cited by 272 publications

References 88 publications

Influenza in Infants Born to Women Vaccinated During Pregnancy

Influenza in Infants Born to Women Vaccinated During Pregnancy

Age at onset and Parkinson disease phenotype

Optimal look back period and summary method for Elixhauser comorbidity measures in a US population-based electronic health record database

Contact Info

Product

Resources

About