Use of Machine Learning and Statistical Algorithms to Predict Hospital Length of Stay Following Colorectal Cancer Resection: A South African Pilot Study

Okechinyere J Achilonu; June Fabian; Brendan Bebington; Elvira Singh; Gideon Nimako; Rene M J C Eijkemans; Eustasius Musenge

doi:10.3389/fonc.2021.644045

Use of Machine Learning and Statistical Algorithms to Predict Hospital Length of Stay Following Colorectal Cancer Resection: A South African Pilot Study

Front Oncol. 2021 Oct 1:11:644045. doi: 10.3389/fonc.2021.644045. eCollection 2021.

Authors

Okechinyere J Achilonu¹, June Fabian^{2

3}, Brendan Bebington^{3

4}, Elvira Singh^{1

5}, Gideon Nimako^{1

6}, Rene M J C Eijkemans⁷, Eustasius Musenge¹

Affiliations

¹ Division of Epidemiology and Biostatistics, School of Public Health, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa.
² Medical Research Council/Wits University Rural Public Health and Health Transitions Research Unit (Agincourt), School of Public Health, Faculty of Health Sciences, University of Witwatersrand, Johannesburg, South Africa.
³ Wits Donald Gordon Medical Centre, School of Clinical Medicine, Faculty of Health Sciences, University of Witwatersrand, Johannesburg, South Africa.
⁴ Department of Surgery, Faculty of Health Science, University of the Witwatersrand Faculty of Science, Parktown, Johannesburg, South Africa.
⁵ National Cancer Registry, National Health Laboratory Service, Johannesburg, South Africa.
⁶ Industrialization, Science, Technology and Innovation Hub, African Union Development Agency (AUDA-NEPAD), Johannesburg, South Africa.
⁷ Julius Center for Health Sciences and Primary Care, University Medical Center, Utrecht University, Utrecht, Netherlands.

Abstract

The aim of this pilot study was to develop logistic regression (LR) and support vector machine (SVM) models that differentiate low from high risk for prolonged hospital length of stay (LOS) in a South African cohort of 383 colorectal cancer patients who underwent surgical resection with curative intent. Additionally, the impact of 10-fold cross-validation (CV), Monte Carlo CV, and bootstrap internal validation methods on the performance of the two models was evaluated. The median LOS was 9 days, and prolonged LOS was defined as greater than 9 days post-operation. Preoperative factors associated with prolonged LOS were a prior history of hypertension and an Eastern Cooperative Oncology Group score between 2 and 4. Postoperative factors related to prolonged LOS were the need for a stoma as part of the surgical procedure and the development of post-surgical complications. The risk of prolonged LOS was higher in male patients and in any patient with lower preoperative hemoglobin. The highest area under the receiving operating characteristics (AU-ROC) was achieved using LR of 0.823 (CI = 0.798-0.849) and SVM of 0.821 (CI = 0.776-0.825), with each model using the Monte Carlo CV method for internal validation. However, bootstrapping resulted in models with slightly lower variability. We found no significant difference between the models across the three internal validation methods. The LR and SVM algorithms used in this study required incorporating important features for optimal hospital LOS predictions. The factors identified in this study, especially postoperative complications, can be employed as a simple and quick test clinicians may flag a patient at risk of prolonged LOS.

Keywords: South Africa; colorectal cancer; hospital length of stay; internal validation techniques; logistic regression; prediction models; risk factors; support vector machine.