Systematic variable reduction for simplification of incisional hernia risk prediction instruments

Phoebe B McAuliffe; Jesse Y Hsu; Robyn B Broach; Yuliya Borovskiy; Adrienne N Christopher; Martin P Morris; John P Fischer

doi:10.1016/j.amjsurg.2022.03.003

Systematic variable reduction for simplification of incisional hernia risk prediction instruments

Am J Surg. 2022 Jul;224(1 Pt B):576-583. doi: 10.1016/j.amjsurg.2022.03.003. Epub 2022 Mar 8.

Authors

Phoebe B McAuliffe¹, Jesse Y Hsu², Robyn B Broach³, Yuliya Borovskiy⁴, Adrienne N Christopher⁵, Martin P Morris⁶, John P Fischer⁷

Affiliations

¹ University of Pennsylvania, Division of Plastic Surgery, 3400 Civic Center Boulevard, Philadelphia, PA, 19104, USA. Electronic address: Phoebe.mcauliffe@pennmedicine.upenn.edu.
² University of Pennsylvania, Division of Biostatistics, Department of Biostatistics, Epidemiology and Informatics, 423 Guardian Drive, Philadelphia, PA, 19104, USA. Electronic address: Jesse.hsu@pennmedicine.upenn.edu.
³ University of Pennsylvania, Division of Plastic Surgery, 3400 Civic Center Boulevard, Philadelphia, PA, 19104, USA. Electronic address: Robyn.broach@pennmedicine.upenn.edu.
⁴ University of Pennsylvania Health System, Data Analytics Team, 3400 Civic Center Blvd, Philadelphia, PA, 19104, USA. Electronic address: Yuliya.borovskiy@pennmedicine.upenn.edu.
⁵ Thomas Jefferson University, Department of Surgery, 211 S 9th St, Philadelphia, PA, 19107, USA. Electronic address: anc008@jefferson.edu.
⁶ Thomas Jefferson University, Department of Surgery, 211 S 9th St, Philadelphia, PA, 19107, USA. Electronic address: marty.p.morris@gmail.com.
⁷ University of Pennsylvania, Division of Plastic Surgery, 3400 Civic Center Boulevard, Philadelphia, PA, 19104, USA. Electronic address: John.fischer2@pennmedicine.upenn.edu.

PMID: 35282872
DOI: 10.1016/j.amjsurg.2022.03.003

Abstract

Background: Incisional hernia (IH) is a complex, costly and difficult to manage surgical complication. We aim to create an accurate and parsimonious model to assess IH risk, pared down for practicality and translation in the clinical environment.

Methods: Institutional abdominal surgical patients from 2002 to 2019 were identified (N = 102,281); primary outcome of IH, demographic factors, and comorbidities were extracted. A 32-variable Cox proportional hazards model was generated. Reduced-variable models were created by systematic removal of variables 1-4 and 23-25 at a time.

Results: The c-statistic of the full 32-variable model was 0.7232. Four comorbidities decreased accuracy of the model: COPD, paralysis, cancer and combined autoimmune/hereditary collagenopathy or AAA diagnosis. The model with those 4 comorbidities removed had the highest c-statistic (0.7291). The most reduced model included 7 variables and had a c-statistic of 0.7127.

Conclusion: Accuracy of an IH predictive model is only marginally affected by a vast reduction in end-user inputs.

Keywords: Abdominal surgery; Incisional hernia; Laparotomy; Point of care tool; Risk prediction; Variable reduction.

MeSH terms

Abdomen / surgery
Herniorrhaphy / adverse effects
Humans
Incisional Hernia* / epidemiology
Incisional Hernia* / etiology
Incisional Hernia* / surgery
Proportional Hazards Models
Risk Factors