Mp39-02 Development and Validation of a Risk Prediction Model for Prostate Cancer Diagnosis Within a Statewide Quality Improvement Collaborative
Year: 2016
Type: article
Study design: Observational Study
Abstract: metastatic bladder cancer patients treated with RC between 2000 and 2009.All ICD-9 diagnostic codes underlying the 17 groupings of comorbid conditions contributing to the Deyo et al. implementation of CCI were identified for the 12-month interval preceding RC.Logistic regression models quantified the ability of the original Deyo implementation of CCI to predict 90-day mortality, after adjusting for T stage, N stage, age, gender and race.Subsequently, we tested different models with fewer than 17 groupings of comorbid conditions and relied on Akaike's information criterion (AIC) to identify the most parsimonious and informative combination of comorbidities for 90-day mortality.2,000 bootstrap resamples was used for internal validation.RESULTS: Overall, 761 (10.8%) patients died after 90-day from surgery.The original unaltered Deyo that relied on all 17 groupings of comorbid conditions resulted in 69.7% ability to predict 90-day mortality after RC.Analyses aimed at identifying the most parsimonious and informative model resulted in the inclusion of 6 of the 17 original groupings of comorbid conditions: congestive heart failure, cerebrovascular disease, chronic pulmonary disease, rheumatologic disease, diabetes without complications and renal disease.The accuracy of this model was 69.9%.The univariate AUC of 6 groupings of comorbid conditions was 55.5% for congestive heart failure, 54.8% for chronic pulmonary disease, 53.4% for renal disease, 53.3% for diabetes without complications, 52.5% for cerebrovascular disease and 51.2% rheumatologic disease.CONCLUSIONS: The Deyo implementation of CCI that relies on 6 groupings of comorbid conditions yield accuracy that is virtually equal to that of the original index with all 17 groupings of comorbid conditions.The use of the shorter comorbidity scoring system proposed here is more efficient and may be recommended for purpose of further validation.
Source: The Journal of Urology
Language: English
Subfield: Pulmonary and Respiratory Medicine
Field: Medicine
Domain: Health Sciences
Open Access status: bronze
OpenAlex ID: W2325093144