Risk-scoring model for prediction of non-home discharge in epithelial ovarian cancer patients

Factors Associated with Discharge Destination in Patients with Bone Metastases

Background and Objectives: The discharge destination of patients with advanced cancer correlates with their quality of life. Patients with bone metastases often undergo lifestyle changes owing to pain and activity limitations. However, there are few reports on factors related to the discharge destination of patients with bone metastases. This study aimed to elucidate the factors associated with the discharge destination of patients with bone metastases. Methods: This study included 278 patients diagnosed with bone metastases who were admitted to the University of Tsukuba Hospital between April 2015 and March 2020. This study examined discharge destination, occurrence of skeletal-related events (SREs), primary lesions, locations of bone metastases, functional ambulation categories (FAC), age, and length of hospital stay. A binomial logistic regression analysis was conducted to compare the home and non-home discharge groups. Results: Of the 278 patients, 142 were discharged to home, 89 were discharged to somewhere other than home (non-home), and 47 died. The discharge destination was associated with spinal cord compression (SCC) (odds ratio [OR] 3.37, 95% confidence interval [CI] 1.35-8.43), hypercalcemia (OR 6.84, 95% CI 1.09-42.76), and FAC at admission (OR 0.45, 95% CI 0.35-0.58). The admission FAC cut-off value for discharge to home was determined to be 1.5 (area under the curve [AUC] 0.79, sensitivity 77.5%, specificity 68.5%). Conclusions: Factors associated with discharge destination were identified. The walking ability required for discharge to home was FAC 1.5, meaning that the patient needed one person to assist in preventing falls when walking on level ground. A cut-off value for FAC on admission for predicting outcomes was identified, suggesting the importance of gait ability assessment on admission.

Nonhome Discharge in Patients Undergoing Pelvic Reconstructive Surgery: A National Analysis

IMPORTANCE

Discharge to home after surgery has been recognized as a determinant of long-term survival and is a common concern in the elderly population.

OBJECTIVE

The aim of the study was to determine the incidence and risk factors for nonhome discharge in patients undergoing major surgery for pelvic organ prolapse.

STUDY DESIGN

We performed a retrospective cohort study using the American College of Surgeons National Surgical Quality Improvement Program Database from 2010 to 2018. We included patients who underwent sacrocolpopexy, vaginal colpopexy, and colpocleisis. We compared perioperative characteristics in patients who were discharged home versus those who were discharged to a nonhome location. Stepwise backward multivariate logistic regression was then used to control for confounding variables and identify independent predictors of nonhome discharge.

RESULTS

A total of 38,012 patients were included in this study, 209 of whom experienced nonhome discharge (0.5%). Independent predictors of nonhome discharge included preoperative weight loss (adjusted odds ratio [aOR], 5.9; 95% confidence interval [CI], 1.3-27.5), dependent health care status (aOR, 5.0; 95% CI, 2.6-9.5), abdominal hysterectomy (aOR, 2.3; 95% CI, 1.4-3.7), American Society of Anesthesiologists class 3 or greater (aOR, 2.0; 95% CI, 1.5-2.7), age (aOR, 1.1; 95% CI, 1.05-1.09), operative time (aOR, 1.005; 95% CI, 1.003-1.006), laparoscopic hysterectomy (aOR, 0.6; 95% CI, 0.4-1.0), and laparoscopic sacrocolpopexy (aOR, 0.5; 95% CI, 0.3-0.8).

CONCLUSIONS

In patients undergoing surgery for pelvic organ prolapse, nonhome discharge is associated with various indicators of frailty, including age, health care dependence, and certain comorbidities. An open surgical approach increases the risk of nonhome discharge, while a laparoscopic approach is associated with lower risk.

Systematic review of prediction models for postacute care destination decision-making

OBJECTIVE

This article reports a systematic review of studies containing development and validation of models predicting postacute care destination after adult inpatient hospitalization, summarizes clinical populations and variables, evaluates model performance, assesses risk of bias and applicability, and makes recommendations to reduce bias in future models.

MATERIALS AND METHODS

A systematic literature review was conducted following PRISMA guidelines and the Cochrane Prognosis Methods Group criteria. Online databases were searched in June 2020 to identify all published studies in this area. Data were extracted based on the CHARMS checklist, and studies were evaluated based on predictor variables, validation, performance in validation, risk of bias, and applicability using the Prediction Model Risk of Bias Assessment Tool (PROBAST) tool.

RESULTS

The final sample contained 28 articles with 35 models for evaluation. Models focused on surgical (22), medical (5), or both (8) populations. Eighteen models were internally validated, 10 were externally validated, and 7 models underwent both types. Model performance varied within and across populations. Most models used retrospective data, the median number of predictors was 8.5, and most models demonstrated risk of bias.

DISCUSSION AND CONCLUSION

Prediction modeling studies for postacute care destinations are becoming more prolific in the literature, but model development and validation strategies are inconsistent, and performance is variable. Most models are developed using regression, but machine learning methods are increasing in frequency. Future studies should ensure the rigorous variable selection and follow TRIPOD guidelines. Only 14% of the models have been tested or implemented beyond original studies, so translation into practice requires further investigation.

Does the Emergency Surgery Score predict failure to discharge the patient home? A nationwide analysis

BACKGROUND

The Emergency Surgery Score (ESS) is a point-based scoring system validated to predict mortality and morbidity in emergency general surgery (EGS). In addition to demographics and comorbidities, ESS accounts for the acuity of disease at presentation. We sought to examine whether ESS can predict the destination of discharge of EGS patients, as a proxy for quality of life at discharge.

METHODS

Using the 2007 to 2017 American College of Surgeons National Surgical Quality Improvement Program database, we identified all EGS patients. EGS cases were defined as per American College of Surgeons National Surgical Quality Improvement Program as those performed by a general surgeon within a short interval from diagnosis or the onset of related symptomatology, when the patient's well-being and outcome may be threatened by unnecessary delay and patient's status could deteriorate unpredictably or rapidly. Emergency Surgery Score patients were then categorized by their discharge disposition to home versus rehabilitation or nursing facilities. All patients with missing ESS or discharge disposition and those discharged to hospice, senior communities, or separate acute care facilities were excluded. Emergency Surgery Score was calculated for each patient. C statistics were used to study the correlation between ESS and the destination of discharge.

RESULTS

Of 6,485,915 patients, 84,694 were included. The mean age was 57 years, 51% were female, and 79.6% were discharged home. The mean ESS was 5. Emergency Surgery Score accurately and reliably predicted the discharge destination with a C statistic of 0.83. For example, ESS of 1, 10, and 20 were associated with 0.9%, 56.5%, and 100% rates of discharge to a rehabilitation or nursing facility instead of home.

CONCLUSION

Emergency Surgery Score accurately predicts which EGS patients require discharge to rehabilitation or nursing facilities and can thus be used for preoperatively counseling patients and families and for improving early discharge preparations, when appropriate.

LEVEL OF EVIDENCE

Prognostic and epidemiological, level III.

Predictive Model and Online Calculator for Discharge Disposition in Brain Tumor Patients

BACKGROUND

In the era of value-based payment models, it is imperative for neurosurgeons to eliminate inefficiencies and provide high-quality care. Discharge disposition is a relevant consideration with clinical and economic ramifications in brain tumor patients. We developed a predictive model and online calculator for postoperative non-home discharge disposition in brain tumor patients that can be incorporated into preoperative workflows.

METHODS

We reviewed all brain tumor patients at our institution from 2017 to 2019. A predictive model of discharge disposition containing preoperatively available variables was developed using stepwise multivariable logistic regression. Model performance was assessed using receiver operating characteristic curves and calibration curves. Internal validation was performed using bootstrapping with 2000 samples.

RESULTS

Our cohort included 2335 patients who underwent 2586 surgeries with a 16% non-home discharge rate. Significant predictors of non-home discharge were age >60 years (odds ratio [OR], 2.02), African American (OR, 1.73) or Asian (OR, 2.05) race, unmarried status (OR, 1.48), Medicaid insurance (OR, 1.90), admission from another health care facility (OR, 2.30), higher 5-factor modified frailty index (OR, 1.61 for 5-factor modified frailty index ≥2), and lower Karnofsky Performance Status (increasing OR with each 10-point decrease in Karnofsky Performance Status). The model was well calibrated and had excellent discrimination (optimism-corrected C-statistic, 0.82). An open-access calculator was deployed (https://neurooncsurgery.shinyapps.io/discharge_calc/).

CONCLUSIONS

A strongly performing predictive model and online calculator for non-home discharge disposition in brain tumor patients was developed. With further validation, this tool may facilitate more efficient discharge planning, with consequent improvements in quality and value of care for brain tumor patients.

Predictors of Non-home Discharge after Cytoreductive Surgery and Hyperthermic Intraperitoneal Chemotherapy

BACKGROUND

Using a national database of cytoreductive surgery and hyperthermic intraperitoneal chemotherapy (CRS/HIPEC) recipients, we sought to determine risk factors for nonhome discharge (NHD) in a cohort of patients.

METHODS

Patients undergoing CRS/HIPEC at any one of 12 participating sites between 2000 and 2017 were identified. Univariate analysis was used to compare the characteristics, operative variables, and postoperative complications of patients discharged home and patients with NHD. Multivariate logistic regression was used to identify independent risk factors of NHD.

RESULTS

The cohort included 1593 patients, of which 70 (4.4%) had an NHD. The median [range] peritoneal cancer index in our cohort was 14 [0-39]. Significant predictors of NHD identified in our regression analysis were advanced age (odds ratio [OR], 1.09; 95% confidence interval [CI], 1.05-1.12; P < 0.001), an American Society of Anesthesiologists (ASA) score of 4 (OR, 2.87; 95% CI, 1.21-6.83; P = 0.017), appendiceal histology (OR, 3.14; 95% CI 1.57-6.28; P = 0.001), smoking history (OR, 3.22; 95% CI, 1.70-6.12; P < 0.001), postoperative total parenteral nutrition (OR, 3.14; 95% CI, 1.70-5.81; P < 0.001), respiratory complications (OR, 7.40; 95% CI, 3.36-16.31; P < 0.001), wound site infections (OR, 3.12; 95% CI, 1.58-6.17; P = 0.001), preoperative hemoglobin (OR, 0.81; 95% CI, 0.70-0.94; P = 0.006), and total number of complications (OR, 1.41; 95% CI, 1.16-1.73; P < 0.001).

CONCLUSIONS

Early identification of patients at high risk for NHD after CRS/HIPEC is key for preoperative and postoperative counseling and resource allocation, as well as minimizing hospital-acquired conditions and associated health care costs.

Evaluation of non-completion of intraperitoneal chemotherapy in patients with advanced epithelial ovarian cancer

Objective

To identify factors associated with non-completion of intraperitoneal with intravenous chemotherapy [IP/IV] in women with epithelial ovarian cancer (EOC).

Methods

This was an Institutional Review Board approved, retrospective cohort study in women with stage III EOC following optimal cytoreductive surgery (CRS) (<1 cm) followed by IP/IV chemotherapy from 2000–2016. Demographic, surgical, and oncologic variables were collected. Pearson χ² test and 2 sample t-test evaluated for variables associated with IP/IV chemotherapy completion. Kaplan-Meier survival analysis was performed for progression-free survival (PFS) and overall survival (OS).

Results

Of 96 women, 71.9% (n=69) completed 6 cycles of IP/IV chemotherapy. The majority had high grade serous histology (n=82; 85.4%) and stage IIIC disease (n=83; 86.5%). Common reasons for IP/IV chemotherapy discontinuation were grade 3–4 gastrointestinal (n=10; 37.0%), neurologic (n=6; 22.2%), hematologic (n=3; 11.1%), renal toxicities (n=3; 11.1%) and port infections (n=3; 11.1%). Incidence of IP port complications was 20.8% (n=20). Port complications (48.0% vs. 11.6%; p<0.001) and hospitalization during chemotherapy (29.6% vs. 2.9%; p<0.001) were more frequent in patients who discontinued IP/IV chemotherapy. Patients who completed IP/IV chemotherapy had higher rates of home discharge following CRS (92.2% vs. 72.0%; p<0.01) and lower Eastern Cooperative Oncology Group (ECOG) score (0 vs. 1.0; p=0.04). There was no significant difference in PFS (p=0.51) nor OS (p=0.38) between the cohorts.

Conclusion

In this series, the rate of IP/IV chemotherapy completion is high. Non-home discharge and higher ECOG status following CRS are associated with IP/IV chemotherapy non-completion and should be considered in treatment planning.

Risk for non-home discharge following surgery for ischemic mitral valve disease

OBJECTIVES

To determine the frequency and risk factors for non-home discharge (NHD) and its association with clinical outcomes and quality of life (QOL) at 1 year following cardiac surgery in patients with ischemic mitral regurgitation (IMR).

METHODS

Discharge disposition was evaluated in 552 patients enrolled in trials of severe or moderate IMR. Patient and in-hospital factors associated with NHD were identified using logistic regression. Subsequently, association of NHD with 1-year mortality, serious adverse events (SAEs), and QOL was assessed.

RESULTS

NHD was observed in 30% (154/522) with 25% (n = 71/289) in moderate and 36% (n = 83/233) in patients with severe IMR (unadjusted P = .006), a difference not significant after including age (5-year change: adjusted odds ratio [adjOR], 1.52; 95% confidence interval [CI], 1.35-1.72; P < .001), diabetes (adjOR, 1.94; 95% CI, 1.27-2.94; P = .002), and previous heart failure (adjOR, 1.64; 95% CI, 1.06-2.52; P = .03). Odds of NHD were increased for patients with postoperative SAEs (adjOR, 1.85; 95% CI, 1.19-2.86; P = .01) but not based on type of cardiac surgery. Greater rates of death and SAEs were observed in NHD patients at 1 year: adjusted hazard ratio, 4.29 (95% CI, 2.14-8.59; P < .001) and adjusted rate ratio, 1.45 (95% CI, 1.03-2.02; P = .03), respectively. QOL did not differ significantly between groups.

CONCLUSIONS

NHD is common following surgery for IMR, influenced by older age, diabetes, previous heart failure, and postoperative SAEs. These patients may be at greater risk of death and subsequent SAEs after discharge. Discussion of NHD with patients may have important implications for decision-making and guiding expectations following cardiac surgery.

Assessment of poor functional status and post-acute care needs following primary ovarian cancer debulking surgery

INTRODUCTION

Poor baseline functional status is associated with adverse surgical outcomes. Additionally, decline in the postoperative setting may result in the delay of additional treatments, impacting overall survival. This study assesses the incidence and risk factors for functional decline following primary ovarian cancer debulking surgery in previously independent women using discharge location as a surrogate.

METHODS

All patients with a postoperative diagnosis of ovarian cancer who underwent surgical debulking and had documentation of discharge location were identified using the 2011-2012 American College of Surgeons National Surgical Quality Improvement Program database. Patients were excluded if their baseline functional status was dependent or partially dependent, or if they died before discharge. Discharge destination was dichotomized as home versus non-home. Descriptive data included demographics, comorbidities, and perioperative outcomes. Multivariable logistic regression was used to evaluate the association of clinical and surgical factors on discharge destination.

RESULTS

1786 patients met the criteria for analysis; 120 (6.7%) patients were discharged to non-home. Differences between home and non-home discharges included age (53.2% vs 83.3% ≥60), body mass index (26.5 vs 27.8 median), comorbidities (45.2% vs 64.2% with ≥1), and complications (8.6% vs 30.0% with ≥1, all p<0.05). In multivariable logistic regression analyses, only increasing age and complications were independently associated with discharge to non-home. Those age ≥70 had 9.0 times the risk (95% CI 3.5 to 23.4; p<0.001) as age <50. The presence of one or more postoperative complications carried 4.5 times (95% CI 2.9 to 7.0; p<0.001) the risk of those without complications. 30 day mortality was also increased in patients discharged to non-home.

DISCUSSION

6.7% of previously independent ovarian cancer patients were discharged to non-home following surgery. Major risk factors for non-home include older age, comorbidities, and postoperative complications. Efforts to optimize baseline functional status and minimize surgical complications may improve discharge rates to non-home and postoperative functional status.

Comparison of Surgeon Assessment to Frailty Measurement in Abdominal Aortic Aneurysm Repair

BACKGROUND

Endovascular abdominal aortic aneurysm repair (EVAR) allows us to intervene on patients otherwise considered poor candidates for open repair. Despite its importance in determining operative approach, no comparison has been made between the subjective "eyeball test" and an objective measurement of preoperative frailty for EVAR patients.

MATERIALS AND METHODS

Patients undergoing elective EVAR were identified in the Vascular Quality Initiative (VQI) database (2003-2017). Patients were classified "unfit" based on a surgeon-reported variable. Frailty was defined using the VQI-derived Risk Analysis Index, which includes sex, age, BMI, renal failure, congestive heart failure, dyspnea, preoperative ambulation, and functional status. The association between fitness and/or frailty and adverse outcomes was determined by logistic regression.

RESULTS

A total of 11,694 patients undergoing elective EVAR were included of which only 18.1% were "unfit," whereas 34.6% were "frail" and overall 43.6% "unfit or frail." Patients deemed "unfit" or "frail" had significantly increased odds of mortality, complications, and nonhome discharge (P < 0.001), and both frailty and unfitness generated negative predictive values for these outcomes greater than 93%. In adjusted logistic regression, the addition of objective frailty significantly improved model performance in predicting nonhome discharge (C-statistic 0.65 versus 0.71, P < 0.001) and complications (0.59 versus 0.61, P = 0.01), but similarly predicted mortality (0.74 versus 0.73, P = 0.99).

CONCLUSIONS

Preoperative frailty assessment provides a useful objective measure of risk stratification as an adjunct to a physician's clinical intuition. The addition of frailty expands the pool of high-risk patients who are more likely to experience adverse postoperative events after elective EVAR and may benefit from uniquely tailored perioperative interventions.

Improving the post-acute care discharge score (PACD) by adding patients' self-care abilities: A prospective cohort study

Background

Reducing delays in hospital discharge is important to improve transition processes and reduce health care costs. The recently proposed post-acute care discharge score focusing on the self-care abilities before hospital admission allows early identification of patients with a need for post-acute care. New limitations in self-care abilities identified during hospitalization may also indicate a risk. Our aim was to investigate whether the addition of the post-acute care discharge score and a validated self-care instrument would improve the prognostic accuracy to predict post-acute discharge needs in unselected medical inpatients.

Methods

We included consecutive adult medical and neurological inpatients. Logistic regression models with area under the receiver operating characteristic curve were calculated to study associations of post-acute discharge score and self-care index with post-acute discharge risk. We calculated joint regression models and reclassification statistics including the net reclassification index and integrated discrimination improvement to investigate whether merging the self-care index and the post-acute discharge score leads to better diagnostic accuracy.

Results

Out of 1342 medical and 402 neurological patients, 150 (11.18%) and 94 (23.38%) have reached the primary endpoint of being discharged to a post-acute care facility. Multivariate analysis showed that the self-care index is an outcome predictor (OR 0.897, 95%CI 0.864–0.930). By combining the self-care index and the post-acute care discharge score discrimination for medical (from area under the curve 0.77 to 0.83) and neurological patients (from area under the curve 0.68 to 0.78) could be significantly improved. Reclassification statistics also showed significant improvements with regard to net reclassification index (14.2%, p<0.05) and integrated discrimination improvement (4.83%, p<0.05).

Conclusions

Incorporating an early assessment of patients’ actual intrahospital self-care ability to the post-acute care discharge score led to an improved prognostic accuracy for identifying adult, medical and neurological patients at risk for discharge to a post-acute care facility.

Using a Guided Machine Learning Ensemble Model to Predict Discharge Disposition following Meningioma Resection

Objective  Machine learning (ML) algorithms are powerful tools for predicting patient outcomes. This study pilots a novel approach to algorithm selection and model creation using prediction of discharge disposition following meningioma resection as a proof of concept. Materials and Methods  A diversity of ML algorithms were trained on a single-institution database of meningioma patients to predict discharge disposition. Algorithms were ranked by predictive power and top performers were combined to create an ensemble model. The final ensemble was internally validated on never-before-seen data to demonstrate generalizability. The predictive power of the ensemble was compared with a logistic regression. Further analyses were performed to identify how important variables impact the ensemble. Results  Our ensemble model predicted disposition significantly better than a logistic regression (area under the curve of 0.78 and 0.71, respectively, p  = 0.01). Tumor size, presentation at the emergency department, body mass index, convexity location, and preoperative motor deficit most strongly influence the model, though the independent impact of individual variables is nuanced. Conclusion  Using a novel ML technique, we built a guided ML ensemble model that predicts discharge destination following meningioma resection with greater predictive power than a logistic regression, and that provides greater clinical insight than a univariate analysis. These techniques can be extended to predict many other patient outcomes of interest.

Identifying modifiable and non-modifiable risk factors associated with prolonged length of stay after hysterectomy for uterine cancer

OBJECTIVE

We examined the influence of modifiable (intraoperative factors and complications) and non-modifiable (clinical and demographic characteristics) factors on length of stay (LOS) for women who underwent hysterectomy for uterine cancer.

METHODS

The National Surgical Quality Improvement Program database was used to identify women who underwent hysterectomy for uterine cancer from 2006 to 2015. The association between demographic, preoperative, intraoperative, and postoperative factors and LOS was examined. The primary outcome was prolonged LOS (>75th an3 >90th percentiles). Model fit statistics were used to assess the importance of each group of characteristics.

RESULTS

Of 19,084 women identified, 6082 (31.9%) underwent abdominal and 13,002 (68.1%) underwent minimally invasive hysterectomy. In the abdominal hysterectomy group, the 75th and 90th percentiles for LOS were 5 and 8days, respectively. All risk factors combined accounted for 23.6% of the variation in LOS >75th percentile. Demographic characteristics explained 4.0%, preoperative factors 7.0%, intraoperative factors 7.9%, and postoperative characteristics 9.7% of variation in prolonged LOS. In the minimally invasive group, the 75th and 90th percentiles for LOS were 1 and 2days, respectively. The combined risk factors explained 16.2% of the variation in prolonged LOS. Demographic characteristics accounted for 6.2%, preoperative factors 4.1%, intraoperative factors 6.9%, and postoperative characteristics 1.3% of variation in prolonged LOS. Similar patterns were seen when prolonged LOS was defined as >90th percentile.

CONCLUSION

Perioperative risk factors account for approximately one quarter of the variation in prolonged LOS. Overall, a substantial proportion of the variation in LOS remains unexplained by measurable patient and hospital factors which may limit the utility of LOS as a quality metric for endometrial cancer.

Demographic, presentation, and treatment factors and racial disparities in ovarian cancer hospitalization outcomes

BACKGROUND

This study examines whether racial disparities in hospitalization outcomes persist between African-American and White women with ovarian cancer after matching on demographic, presentation, and treatment factors.

METHODS

Using data from the Nationwide Inpatient Sample database, 5,164 African-American ovarian cancer patients were sequentially matched with White patients on demographic (e.g., age, income), presentation (e.g., stage, comorbidities), and treatment (e.g., surgery, radiation) factors. Racial differences in-hospital length of stay, post-operative complications, and in-hospital mortality were evaluated using conditional logistic regression models.

RESULTS

White ovarian cancer patients had relatively higher odds of post-operative complications when matched on demographics (OR 1.35, 95% CI 1.05, 1.74), and presentation (OR 1.28, 95% CI 1.00, 1.65) but not when additionally matched on treatment (OR 1.03, 95% CI 0.78, 1.35). African-American patients had longer in-hospital length of stay (6.96 ± 7.21 days) compared with White patients when matched on demographics (6.37 ± 7.07 days), presentation (6.48 ± 7.16 days), and treatment (6.53 ± 7.59 days). Compared with African-American patients, White patients experienced lower odds of in-hospital mortality when matched on demographics (OR 0.78, 95% CI 0.66, 0.92), but this disparity was no longer significant when additionally matched on presentation (OR 0.88, 95% CI 0.75, 1.04) and treatment (OR 0.95, 95% CI 0.81, 1.12).

CONCLUSION

Racial disparities in ovarian cancer hospitalization outcomes persisted after adjusting for demographic and presentation factors; however these differences were eliminated after additionally accounting for treatment factors. More studies are needed to determine the factors driving racial differences in ovarian cancer treatment in otherwise similar patient populations.

Non-routine discharge disposition is associated with post-discharge complications and 30-day readmissions following craniotomy for brain tumor resection

Several studies have reported an association between high-volume brain tumor centers and greater rates of routine discharge disposition in the context of better outcomes. However, the relationship between in-hospital complications, discharge destination, and postoperative adverse events (AEs) remains unexplored. The purpose of this study was thus to use a large, prospectively collected database to examine the association between discharge destination, post-discharge complications, readmissions, and reoperations among patients undergoing craniotomy for brain tumor. The 2011-2014 National Surgical Quality Improvement (NSQIP) database was employed to identify all adult patients who underwent a craniotomy for brain tumor resection. Demographics, comorbidities, and perioperative variables were collected for each patient. Univariate statistics with subsequent binary logistic regression analyses were used to explore the relationship between these perioperative factors and postoperative events, including major post-discharge complications, minor post-discharge AEs, readmissions, and return to the operating room (ROR). Significant variables such as demographics, comorbidities, operative time, body mass index, ASA classification and pre-discharge complications were controlled for in each model. Of the 14,854 patients identified, 11,409 (77.9%) were discharged home. After controlling for comorbidities and in-hospital AEs, discharge to skilled rehabilitation was an independent predictor of major post-discharge complications (OR 1.74, 95% CI 1.31-2.30, p < 0.001), minor post-discharge events (OR 1.60, 95% CI 1.07-2.41, p = 0.024), and ROR (OR 1.68, 95% CI 1.27-2.22, p < 0.001). Discharge to a care facility was predictive of major complications (OR 1.51, 95% CI 1.04-2.19, p = 0.030) and ROR (OR 2.02, 95% CI 1.46-2.80, p < 0.001). These factors may be considered in discharge planning and further outcomes studies for patients undergoing resection.

Nomogram to predict non-home discharge following pancreaticoduodenectomy in a national cohort of patients

BACKGROUND

Despite the development of pathways to enhance recovery and discharge to home, a significant proportion of patients are discharged to inpatient facilities after pancreaticoduodenectomy (PD). The aim of this study was to determine the rate of non-home discharge (NHD) following PD in a national cohort of patients and to develop predictive nomograms for NHD.

METHODS

The National Surgical Quality Improvement Program was used to construct and validate pre- and postoperative nomograms for NHD following PD.

RESULTS

A total of 6856 patients who underwent PD were identified, of which 927 (13.5%) had an NHD. The independent preoperative predictors of NHD were being female, older age, higher BMI, low serum albumin, >10% weight loss, ASA class III/IV, and being diagnosed with a bile duct/ampullary neoplasm or neuroendocrine tumor. A preoperative nomogram was constructed with a concordance index of 0.77. When postoperative variables were added to the model, the concordance index increased to 0.82. The postoperative predictors of NHD were return to the operating room, length of stay of ≥14 days, and any inpatient complications.

CONCLUSIONS

These nomograms could be useful risk assessment tools to predict NHD after PD and therefore facilitate patient counseling and improve resource utilization and discharge planning.

Preoperative Frailty Is Associated With Discharge to Skilled or Assisted Living Facilities After Urologic Procedures of Varying Complexity

OBJECTIVE

To evaluate the association between frailty and postoperative discharge destination after different types of commonly performed urologic procedures in older patients.

MATERIALS AND METHODS

Using data from the American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) from 2011 to 2013, we identified commonly performed inpatient urologic procedures among patients aged 65 and older. We then assessed the effect of frailty, measured by the NSQIP Frailty Index (NSQIP-FI), on discharge to a skilled or assisted living facility using logistic regression and assessed the heterogeneity of this effect across procedures using 2-level random effects modeling.

RESULTS

Overall, 1144 out of 20,794 (5.5%) urologic cases, representing 19 different procedures, resulted in discharge to a skilled or assisted living facility. Cystectomy and large transurethral resection of bladder tumor had the highest percentage (16.3%). Twenty-five percent of patients undergoing urology procedures were frail (NSQIP-FI 0.18+), including 9.8% of patients discharged to a facility. Even after adjustment for year, age, race, type of anesthesia, smoking status, recent weight loss, and whether or not the procedure was elective, frailty was strongly associated with discharge to a facility (adjusted odds ratio 3.1 [96% confidence interval 2.5, 3.8] for NSQIP-FI 0.18+ compared to NSQIP FI 0). This finding was consistent across most procedures of varying complexity with an overall effect of odds ratio 1.6 (95% confidence interval 1.5, 2.0).

CONCLUSION

Increasing frailty is associated with discharge to a skilled or assisted living facility across most inpatient urologic procedures evaluated, regardless of complexity. This information is important for preoperative counseling with patients undergoing urologic surgery.

Can the risk of non-home discharge after resection of gastric adenocarcinoma be predicted: a seven-institution study of the US Gastric Cancer Collaborative

BACKGROUND

There are no validated methods to preoperatively identify patients with increased risk of discharge to skilled nursing facilities following resection of gastric cancer. We sought to identify preoperative predictors of non-home discharge to optimize transition of care to skilled nursing facility.

METHODS

Patients who underwent resection of gastric cancer from 2000 to 2012 from seven participating institutions of the US Gastric Cancer Collaborative were analyzed. Fisher's exact tests, Student t tests, and logistic regression analyses identified preoperative variables associated with non-home discharge. A prediction tool was created and validated through c-indices. Survival analysis was conducted according to the methods of Kaplan and Meier.

RESULTS

Out of the 918 patients identified, 93 (10 %) were discharged to nonhome location. Univariate analysis identified advancing age, American Society of Anesthesiology (ASA) score, hypertension, decreasing preoperative albumin, and lack of neoadjuvant chemotherapy as risk factors for non-home discharge (NHD). Multivariable analysis identified advanced age (odds ratio (OR) = 1.07, 95 % confidence interval (CI) = 1.04-1.10, p < 0.0001), depressed preoperative albumin (OR = 2.17, 95 % CI = 1.47-3.19, p = 0.0001), and total gastrectomy (OR = 2.56, 95 % CI = 1.53-4.3, p = 0.0003) as risk factors for NHD. The c-index of the model and the validation population were 0.76 and 0.8, respectively. Additionally, there was an association of decreased overall survival in patients discharged to nonhome location (35.5 months, home discharge, vs 12 months, NHD, p < 0.0001).

CONCLUSIONS

Older patients with compromised nutritional status have greater risk of NHD following resection of gastric cancer. The prediction tool can augment preoperative planning to optimize transition of care to skilled nursing facility.

Alternative performance measures for prediction models

As a performance measure for a prediction model, the area under the receiver operating characteristic curve (AUC) is insensitive to the addition of strong markers. A number of measures sensitive to performance change have recently been proposed; however, these relative-performance measures may lead to self-contradictory conclusions. This paper examines alternative performance measures for prediction models: the Lorenz curve-based Gini and Pietra indices, and a standardized version of the Brier score, the scaled Brier. Computer simulations are performed in order to study the sensitivity of these measures to performance change when a new marker is added to a baseline model. When the discrimination power of the added marker is concentrated in the gray zone of the baseline model, the AUC and the Gini show minimal performance improvements. The Pietra and the scaled Brier show more significant improvements in the same situation, comparatively. The Pietra and the scaled Brier indices are therefore recommended for prediction model performance measurement, in light of their ease of interpretation, clinical relevance and sensitivity to gray-zone resolving markers.