# Development of the multivariate administrative data cystectomy model and its impact on misclassification bias

## TL;DR

This study shows that using a statistical model improves accuracy in identifying cystectomy cases from hospital data, reducing errors compared to traditional coding methods.

## Contribution

A multivariate logistic regression model using administrative data reduces misclassification bias in identifying cystectomy cases.

## Key findings

- The model accurately predicted cystectomy types with high c-statistics and zero calibration error.
- Using the model significantly reduced misclassification bias compared to billing codes for both incontinent and continent diversions.
- Administrative data accuracy can be improved by probabilistic imputation instead of relying on individual billing codes.

## Abstract

Misclassification bias (MB) is the deviation of measured from true values due to incorrect case assignment. This study compared MB when cystectomy status was determined using administrative database codes vs. predicted cystectomy probability.

We identified every primary cystectomy-diversion type at a single hospital 2009–2019. We linked to claims data to measure true association of cystectomy with 30 patient and hospitalization factors. Associations were also measured when cystectomy status was assigned using billing codes and by cystectomy probability from multivariate logistic regression model with covariates from administrative data. MB was the difference between measured and true associations.

500 people underwent cystectomy (0.12% of 428 677 hospitalizations). Sensitivity and positive predictive values for cystectomy codes were 97.1% and 58.6% for incontinent diversions and 100.0% and 48.4% for continent diversions, respectively. The model accurately predicted cystectomy-incontinent diversion (c-statistic [C] 0.999, Integrated Calibration Index [ICI] 0.000) and cystectomy-continent diversion (C:1.000, ICI 0.000) probabilities. MB was significantly lower when model-based predictions was used to impute cystectomy-diversion type status using for both incontinent cystectomy (F = 12.75; p < .0001) and continent cystectomy (F = 11.25; p < .0001).

A model using administrative data accurately returned the probability that cystectomy by diversion type occurred during a hospitalization. Using this model to impute cystectomy status minimized MB. Accuracy of administrative database research can be increased by using probabilistic imputation to determine case status instead of individual codes.

The online version contains supplementary material available at 10.1186/s12874-024-02199-1.