Simplified risk prediction indices do not accurately predict 30-day death or readmission after discharge following colorectal surgery

Abstract

Background: Risk-prediction indices are one of many tools implemented to guide efforts to decreasereadmissions. Butusing simplified models to predict a complex process can prove challenging. Additionally, no risk-prediction index has been developed for patients undergoing colorectal surgery. Therefore, we evaluated the performance of a widely-utilized simplified index (LACE) and a novel index in modeling readmissions in this patient population. Methods: Using a retrospective, split-sample cohort, patients discharged after colorectal surgery were identified within the inpatient databases of the Healthcare Cost and Utilization Project for New York, California, and Florida, 2006-2014. The primary outcome was death or readmission within 30 days after discharge. Multivariable logistic regression models incorporated patient comorbidities, postoperative complications, and hospitalization details, and were evaluated using the C statistic. Results: 440,742 patients met eligibility criteria. The rate of death or readmission within 30 days after discharge was 14.0% (n = 61,757). When applied to surgical patients, the LACE index demonstrated a poor model fit (C = 0.631). Model fit improved significantly yet remained poor (C = 0.654; p < 0.001) with the addition of variables known to be associated with readmission after colorectal surgery: age, indication for surgery, and creation of a new ostomy. A novel, simplified model also yielded a poor model fit (C = 0.660).

Readmissions Within 48 Hours of Discharge: Reasons, Risk Factors, and Potential Improvements

“…[1][2][3] Readmissions have been commonly recognized as important cost drivers, 4 and several large-scale studies using national or statewide data focused on identifying potentially targetable predictors of hospital readmission. [4][5][6][7][8] However, institution-specific discrepancies due to variable care protocols, discharge criteria, or surgical settings may impede uncritical extrapolation and application of the results. Moreover, large patient cohorts are required to identify less common and potentially preventable reasons for readmission that can help focus researchers on specific outcomes of interest, such as very early readmissions.…”

A Single-Institution Analysis of Targeted Colorectal Surgery Enhanced Recovery Pathway Strategies That Decrease Readmissions

“…Stapler et al: Interventions That Decrease Readmission e738 prediction has not been successfully modeled based on variables thought to be risk factors after colorectal surgery. 23 In contrast, another study developed a readmission risk calculator based on American College of Surgeons National Surgical Quality Improvement Program data that identified patients at high risk for readmission after colorectal surgery. Risk factors for readmission included patients with an ileostomy and those who developed postoperative SSIs.…”

Care Fragmentation and Mortality in Readmission after Surgery for Hepatopancreatobiliary and Gastric Cancer: A Patient-Level and Hospital-Level Analysis of the Healthcare Cost and Utilization Project Administrative Database

“…We have done this in a previous population-level analysis of readmissions after GI surgery with the intent of improving, not restricting, generalizability of population-level research. 22 In our clinical practice, patients with LOS beyond a subjective thresholdearound 21 daysehave become a unique cohort of patients that diverge from a generalizable risk profile for post-discharge events and risk-skew data like length of stay or cost. The outlier outcomes of these patients is certainly multifactorial, but may be due to rare complications, the interaction of pre-existing comorbidities, and postoperative complications, or extensive operations (eg, multivisceral, vascular resection) with accompanying risk that is not well-captured using administrative coding.…”

“…Our group has previously shown that, even with surgeryspecific variables and existing validated models, achieving adequate model fit when modeling readmissions using administrative data is challenging when applying models to specific patient populations. 22 Limitations of this dataset include the inability to track patients across state lines, the aforementioned limitation of the reliability of coding from administrative data, and the inexact estimation of cost from administrative charges using a conversion that varies between hospitals and years. The states selected may not be representative of states with lower population density, and we do intend to test our models using data from less populated states and data from our own institution.…”

“…Less than 2 days was chosen to reduce or eliminate the potential inclusion of patients with incorrectly or overcoded procedures, assuming that the anticipated LOS for all procedures of interest be at least 2 days. Eliminating patients with LOS greater than 21 days has previously been performed in surgical population-level analyses, 22 with rationale provided in the Discussion section. Data were sourced from subsequent inpatient stays and emergency room visits (HCUP State Emergency Department [ED] databases).…”