Identifying prognostic factors for clinical outcomes and costs in four high-volume surgical treatments using routinely collected hospital data

Between-Hospital and Between-Physician Variation in Outcomes and Costs in High- and Low-Complex Surgery: A Nationwide Multilevel Analysis

OBJECTIVES

Clinicians and policy makers are increasingly exploring strategies to reduce unwarranted variation in outcomes and costs. Adequately accounting for case mix and better insight into the levels at which variation exists is crucial for such strategies. This nationwide study investigates variation in surgical outcomes and costs at the level of hospitals and individual physicians and evaluates whether these can be reliably compared on performance.

METHODS

Variation was analyzed using 92 330 patient records collected from 62 Dutch hospitals who underwent surgery for colorectal cancer (n = 6640), urinary bladder cancer (n = 14 030), myocardial infarction (n = 31 870), or knee osteoarthritis (n = 39 790) in the period 2018 to 2019. Multilevel regression modeling with and without case-mix adjustment was used to partition variation in between-hospital and between-physician components for in-hospital mortality, intensive care unit admission, length of stay, 30-day readmission, 30-day reintervention, and in-hospital costs. Reliability was calculated for each treatment-outcome combination at both levels.

RESULTS

Across outcomes, hospital-level variation relative to total variation ranged between ≤ 1% and 15%, and given the high caseloads, this typically yielded high reliability (> 0.9). In contrast, physician-level variation components were typically ≤ 1%, with limited opportunities to make reliable comparisons. The impact of case-mix adjustment was limited, but nonnegligible.

CONCLUSIONS

It is not typically possible to make reliable comparisons among physicians due to limited partitioned variation and low caseloads. Nevertheless, for hospitals, the opposite often holds. Although variation-reduction efforts directed at hospitals are thus more likely to be successful, this should be approached cautiously, partly because level-specific variation and the impact of case mix vary considerably across treatments and outcomes.

Explainable time-series deep learning models for the prediction of mortality, prolonged length of stay and 30-day readmission in intensive care patients

Background

In-hospital mortality, prolonged length of stay (LOS), and 30-day readmission are common outcomes in the intensive care unit (ICU). Traditional scoring systems and machine learning models for predicting these outcomes usually ignore the characteristics of ICU data, which are time-series forms. We aimed to use time-series deep learning models with the selective combination of three widely used scoring systems to predict these outcomes.

Materials and methods

A retrospective cohort study was conducted on 40,083 patients in ICU from the Medical Information Mart for Intensive Care-IV (MIMIC-IV) database. Three deep learning models, namely, recurrent neural network (RNN), gated recurrent unit (GRU), and long short-term memory (LSTM) with attention mechanisms, were trained for the prediction of in-hospital mortality, prolonged LOS, and 30-day readmission with variables collected during the initial 24 h after ICU admission or the last 24 h before discharge. The inclusion of variables was based on three widely used scoring systems, namely, APACHE II, SOFA, and SAPS II, and the predictors consisted of time-series vital signs, laboratory tests, medication, and procedures. The patients were randomly divided into a training set (80%) and a test set (20%), which were used for model development and model evaluation, respectively. The area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and Brier scores were used to evaluate model performance. Variable significance was identified through attention mechanisms.

Results

A total of 33 variables for 40,083 patients were enrolled for mortality and prolonged LOS prediction and 36,180 for readmission prediction. The rates of occurrence of the three outcomes were 9.74%, 27.54%, and 11.79%, respectively. In each of the three outcomes, the performance of RNN, GRU, and LSTM did not differ greatly. Mortality prediction models, prolonged LOS prediction models, and readmission prediction models achieved AUCs of 0.870 ± 0.001, 0.765 ± 0.003, and 0.635 ± 0.018, respectively. The top significant variables co-selected by the three deep learning models were Glasgow Coma Scale (GCS), age, blood urea nitrogen, and norepinephrine for mortality; GCS, invasive ventilation, and blood urea nitrogen for prolonged LOS; and blood urea nitrogen, GCS, and ethnicity for readmission.

Conclusion

The prognostic prediction models established in our study achieved good performance in predicting common outcomes of patients in ICU, especially in mortality prediction. In addition, GCS and blood urea nitrogen were identified as the most important factors strongly associated with adverse ICU events.