Attitudes on Artificial Intelligence use in Pediatric Care From Parents of Hospitalized Children.

INTRODUCTION: Artificial intelligence (AI) may benefit pediatric healthcare, but it also raises ethical and pragmatic questions. Parental support is important for the advancement of AI in pediatric medicine. However, there is little literature describing parental attitudes toward AI in pediatric healthcare, and existing studies do not represent parents of hospitalized children well. METHODS: We administered the Attitudes toward Artificial Intelligence in Pediatric Healthcare, a validated survey, to parents of hospitalized children in a single tertiary children's hospital. Surveys were administered by trained study personnel (11/2/2021-5/1/2022). Demographic data were collected. An Attitudes toward Artificial Intelligence in Pediatric Healthcare score, assessing openness toward AI-assisted medicine, was calculated for seven areas of concern. Subgroup analyses were conducted using Mann-Whitney U tests to assess the effect of race, gender, education, insurance, length of stay, and intensive care unit (ICU) admission on AI use. RESULTS: We approached 90 parents and conducted 76 surveys for a response rate of 84%. Overall, parents were open to the use of AI in pediatric medicine. Social justice, convenience, privacy, and shared decision-making were important concerns. Parents of children admitted to an ICU expressed the most significantly different attitudes compared to parents of children not admitted to an ICU. CONCLUSIONS: Parents were overall supportive of AI-assisted healthcare decision-making. In particular, parents of children admitted to ICU have significantly different attitudes, and further study is needed to characterize these differences. Parents value transparency and disclosure pathways should be developed to support this expectation.

Technical/Algorithm, Stakeholder, and Society (TASS) barriers to the application of artificial intelligence in medicine: A systematic review.

INTRODUCTION: The use of artificial intelligence (AI), particularly machine learning and predictive analytics, has shown great promise in health care. Despite its strong potential, there has been limited use in health care settings. In this systematic review, we aim to determine the main barriers to successful implementation of AI in healthcare and discuss potential ways to overcome these challenges. METHODS: We conducted a literature search in PubMed (1/1/2001-1/1/2023). The search was restricted to publications in the English language, and human study subjects. We excluded articles that did not discuss AI, machine learning, predictive analytics, and barriers to the use of these techniques in health care. Using grounded theory methodology, we abstracted concepts to identify major barriers to AI use in medicine. RESULTS: We identified a total of 2,382 articles. After reviewing the 306 included papers, we developed 19 major themes, which we categorized into three levels: the Technical/Algorithm, Stakeholder, and Social levels (TASS). These themes included: Lack of Explainability, Need for Validation Protocols, Need for Standards for Interoperability, Need for Reporting Guidelines, Need for Standardization of Performance Metrics, Lack of Plan for Updating Algorithm, Job Loss, Skills Loss, Workflow Challenges, Loss of Patient Autonomy and Consent, Disturbing the Patient-Clinician Relationship, Lack of Trust in AI, Logistical Challenges, Lack of strategic plan, Lack of Cost-effectiveness Analysis and Proof of Efficacy, Privacy, Liability, Bias and Social Justice, and Education. CONCLUSION: We identified 19 major barriers to the use of AI in healthcare and categorized them into three levels: the Technical/Algorithm, Stakeholder, and Social levels (TASS). Future studies should expand on barriers in pediatric care and focus on developing clearly defined protocols to overcome these barriers.

Confidence-based laboratory test reduction recommendation algorithm.

BACKGROUND: We propose a new deep learning model to identify unnecessary hemoglobin (Hgb) tests for patients admitted to the hospital, which can help reduce health risks and healthcare costs. METHODS: We collected internal patient data from a teaching hospital in Houston and external patient data from the MIMIC III database. The study used a conservative definition of unnecessary laboratory tests, which was defined as stable (i.e., stability) and below the lower normal bound (i.e., normality). Considering that machine learning models may yield less reliable results when trained on noisy inputs containing low-quality information, we estimated prediction confidence to assess the reliability of predicted outcomes. We adopted a "select and predict" design philosophy to maximize prediction performance by selectively considering samples with high prediction confidence for recommendations. Our model accommodated irregularly sampled observational data to make full use of variable correlations (i.e., with other laboratory test values) and temporal dependencies (i.e., previous laboratory tests performed within the same encounter) in selecting candidates for training and prediction. RESULTS: The proposed model demonstrated remarkable Hgb prediction performance, achieving a normality AUC of 95.89% and a Hgb stability AUC of 95.94%, while recommending a reduction of 9.91% of Hgb tests that were deemed unnecessary. Additionally, the model could generalize well to external patients admitted to another hospital. CONCLUSIONS: This study introduces a novel deep learning model with the potential to significantly reduce healthcare costs and improve patient outcomes by identifying unnecessary laboratory tests for hospitalized patients.

Complication rates in very low and extremely low birth weight infants following laparotomy: a prospective study.

INTRODUCTION: Surgical site occurrences (SSO), including surgical site infection, dehiscence, and incisional hernia, are complications following laparotomy. SSO rates in premature neonates are poorly understood. We hypothesize that SSO rates are higher among extremely low birth weight (ELBW) infants compared to very low birth weight (VLBW) infants and strive to determine the optimal abdominal closure method for these infants. METHODS: We conducted a prospective observational study of infants < 1.5 kg (kg) undergoing laparotomy at two institutions from 1/1/2020 to 5/1/2022. Patients were grouped by weight and closure; SSO rates were computed and the association tested using Fisher's exact test. RESULTS: We identified 59 patients and 104 total operations. At initial surgery, 37 patients weighed < 1 kg (ELBW); 22 patients weighed 1-1.5 kg (VLBW). Complication rate for ELBW was 6(16%) vs. 2(9%) in VLBW, but not significant (p = 0.45). More complications followed a single-layer compared to a two-layer closure (18 vs. 2), but not significant (p = 0.30). CONCLUSIONS: SSO rates are higher for ELBW infants undergoing laparotomy, and fewer complications follow two-layer closure. However, these findings did not reach statistical significance. Further studies are needed to identify modifiable factors to reduce postoperative complications in these infants.

Patterns of Care for Colorectal Liver Metastasis Within an Integrated Health System: Secular Trends and Outcomes.

BACKGROUND: Utilization of evidence-based treatments for patients with colorectal liver metastasis (CRC-LM) outside high-volume centers is not well-characterized. We sought to describe trends in treatment and outcomes, and identify predictors of therapy within a nationwide integrated health system. METHODS: Observational cohort study of patients with CRC-LM treated within the Veterans Affairs (VA) health system (1998-2012). Secular trends and outcomes were compared on the basis of treatment type. Multivariate regression was used to identify predictors of no treatment (chemotherapy or surgery). RESULTS: Among 3270 patients, 57.3 % received treatment (chemotherapy and/or surgery) during the study period. The proportion receiving treatment doubled (38 % in 1998 vs. 68 % in 2012; trend test, p < 0.001), primarily driven by increased use of chemotherapy (26 vs. 57 %; trend test, p < 0.001). Among patients having surgery (16 %), the proportion having ablation (10 vs. 61.9 %; trend test, p < 0.001) and multimodality therapy (15 vs. 67 %; trend test, p < 0.001) increased significantly over time. Older patients [65-75 years: odds ratio (OR) 1.65, 95 % confidence interval (CI) 1.39-1.97; >75 years: OR 3.84, 95 % CI 3.13-4.69] and those with high comorbidity index (Charlson ≥3: OR 1.47, 95 % CI 1.16-1.85) were more likely to be untreated. Overall survival was significantly different based on treatment strategy (log-rank p < 0.001). CONCLUSIONS: The proportion of CRC-LM patients receiving treatment within the largest integrated health system in the US (VA health system) has increased substantially over time; however, one in three patients still does not receive any treatment. Future initiatives should focus on increasing treatment among older patients as well as on evaluating reasons leading to the no-treatment approach and increased use of ablation procedures.

Primary Tumor Resection and Multimodality Treatment for Patients with Metastatic Colon Cancer.

BACKGROUND: Although controversial, recent data suggest a benefit associated with primary tumor resection (PTR) in metastatic colon cancer (mCC) patients. However, utilization of the various management strategies over time relative to surgery, in particular multimodality treatment (MMT), as well as the impact of age on treatment remains unclear. STUDY DESIGN: Historical cohort study of mCC patients in the National Cancer Data Base (1998-2009). Temporal trends in treatment utilization (chemotherapy, PTR alone, MMT) were evaluated. Using a landmark approach, the association between treatment, age, and risk of death was evaluated with multivariable Cox regression, including interaction. RESULTS: Among 103,100 mCC patients, PTR decreased 50.1 % during the study period, whereas MMT and chemotherapy increased 27.4 and 104.8 %, respectively (trend test, p < 0.001). Patients aged ≥75 years were the only group for whom PTR alone was the most common intervention over time and performed more commonly (33.8 %) than MMT (23.8 %) in the most recent study year. Relative to MMT, risk of death was higher for all other management strategies. The sequence of PTR and chemotherapy (reference-surgery first) did not affect risk of death (chemotherapy first-1.05 [0.95-1.15]), as long resection was a part of MMT (PTR alone-1.16 [1.08-1.23]). Patient age did not impact the relative benefit associated with competing management strategies. CONCLUSIONS: Although the benefit associated with PTR in mCC patients is a function of MMT, PTR alone remains a common management strategy among older patients. Given the aging U.S. population, exploring provider biases and patient preferences may be necessary to optimize management of mCC patients.

Ninety-day readmission after colorectal cancer surgery in a Veterans Affairs cohort.

BACKGROUND: Readmissions following colorectal surgery are common. However, there are limited data examining unplanned readmissions (URs) after colorectal cancer (CRC) surgery. The goal of this study was to identify reasons and predictors of UR, and to examine their clinical impact on CRC patients. METHODS: A retrospective cohort study using a prospective CRC surgery database of patients treated at a VA tertiary referral center was performed (2005-2011). Ninety-day URs were recorded and classified based on reason for readmission. Clinical impact of UR was measured using a validated classification for postoperative complications. Multivariate logistic regression analyses were performed to identify predictors of UR. RESULTS: 487 patients were included; 104 (21%) required UR. Although the majority of UR were due to surgical reasons (n = 72, 69%), medical complications contributed to 25% of all readmission events. Nearly half of UR (n = 44, 40%) had significant clinical implications requiring invasive interventions, intensive care unit stays, or led to death. After multivariate logistic regression, the following independent predictors of UR were identified: African-American race (odds ratio [OR] 0.47 [0.27-0.88]), ostomy creation (OR 2.50 [1.33-4.70]), and any postoperative complication (OR 4.36 [2.48-7.68]). CONCLUSIONS: Ninety-day URs following colorectal cancer surgery are common, and represent serious events associated with worse outcomes. In addition to postoperative complications, surgical details that can be anticipated (i.e., ileostomy creation) and medical events unrelated to surgery, both contribute as important and potentially preventable reasons for UR. Future studies should focus on developing and examining interventions focused at improving the process of perioperative care for this high-risk population.

Analysis of model development strategies: predicting ventral hernia recurrence.

BACKGROUND: There have been many attempts to identify variables associated with ventral hernia recurrence; however, it is unclear which statistical modeling approach results in models with greatest internal and external validity. We aim to assess the predictive accuracy of models developed using five common variable selection strategies to determine variables associated with hernia recurrence. METHODS: Two multicenter ventral hernia databases were used. Database 1 was randomly split into "development" and "internal validation" cohorts. Database 2 was designated "external validation". The dependent variable for model development was hernia recurrence. Five variable selection strategies were used: (1) "clinical"-variables considered clinically relevant, (2) "selective stepwise"-all variables with a P value <0.20 were assessed in a step-backward model, (3) "liberal stepwise"-all variables were included and step-backward regression was performed, (4) "restrictive internal resampling," and (5) "liberal internal resampling." Variables were included with P < 0.05 for the Restrictive model and P < 0.10 for the Liberal model. A time-to-event analysis using Cox regression was performed using these strategies. The predictive accuracy of the developed models was tested on the internal and external validation cohorts using Harrell's C-statistic where C > 0.70 was considered "reasonable". RESULTS: The recurrence rate was 32.9% (n = 173/526; median/range follow-up, 20/1-58 mo) for the development cohort, 36.0% (n = 95/264, median/range follow-up 20/1-61 mo) for the internal validation cohort, and 12.7% (n = 155/1224, median/range follow-up 9/1-50 mo) for the external validation cohort. Internal validation demonstrated reasonable predictive accuracy (C-statistics = 0.772, 0.760, 0.767, 0.757, 0.763), while on external validation, predictive accuracy dipped precipitously (C-statistic = 0.561, 0.557, 0.562, 0.553, 0.560). CONCLUSIONS: Predictive accuracy was equally adequate on internal validation among models; however, on external validation, all five models failed to demonstrate utility. Future studies should report multiple variable selection techniques and demonstrate predictive accuracy on external data sets for model validation.

Postoperative transitional care needs in the elderly: an outcome of recovery associated with worse long-term survival.

OBJECTIVE: To characterize transitional care needs (TCNs) after colorectal cancer (CRC) surgery and examine their association with age and impact on overall survival (OS). BACKGROUND: TCNs after cancer surgery represent additional burden for patients and are associated with higher short-term mortality. They are not well-characterized in CRC patients, particularly in the context of a growing elderly population, and their effect on long-term survival is unknown. METHODS: A retrospective cohort study of CRC patients (N = 486) having curative surgery at a tertiary referral center (2002-2011) was conducted. Outcomes included TCNs (home health or nonhome destination at discharge) and OS. Patients were compared on the basis of age: young (<65 years), old (65-74 years), and oldest (≥75 years). Multivariate logistic regression models were used to examine the association of age with TCNs, and OS was compared on the basis of TCNs and stage, using the Kaplan-Meier method. RESULTS: TCNs were required by 130 patients (27%). The oldest patients had highest TCNs (49%) compared with the other age groups (P < 0.01), with rehabilitation services as their primary TCNs (80%). After multivariate analysis, patients 75 years or older had significantly increased TCN risk (odds ratio, 4.7; 95% confidence interval, 2.6-8.5). TCN was associated with worse OS for patients with early- and advanced stage CRC (P < 0.001). CONCLUSIONS: TCNs after CRC surgery are common and significantly increased in patients 75 years or older, represent an outcome of postoperative recovery, and are associated with worse long-term survival. Preoperative identification of higher risk populations should be used for patient counseling, advanced preoperative planning, and to implement strategies targeted at minimizing TCNs.

Systematic review of educational interventions for ostomates.

BACKGROUND: Stoma-related complications lead to increased hospital length of stay and readmissions. Although education of new ostomates is widely recommended, there is a lack of data regarding effective evidence-based educational interventions to prevent or decrease these complications. OBJECTIVE: The aim of this study was to systematically review the literature for educational interventions for new ostomates designed to decrease stoma-related complications. DATA SOURCES: PubMed was searched for studies on educational interventions for new ostomates. STUDY SELECTION: Studies were included if they were in English, targeted adult stoma patients, and evaluated an educational intervention at the time of stoma creation. INTERVENTION: Educational interventions were performed. MAIN OUTCOME MEASURES: The outcomes of interest were length of stay, complications, and readmissions. RESULTS: We found 1706 articles of which 7 met the inclusion criteria. Two were randomized controlled trials, and the rest were cohort studies. The overall quality of the studies was low. Each study used a unique intervention. However, all incorporated a specialized colorectal or ostomy nurse. Of the 5 studies that evaluated length of stay, 2 found a reduction in length of stay associated with the intervention, but 3 found no difference. Two studies found a reduction in complications, but 2 found no difference. Of the 3 studies that evaluated readmissions, none found a difference in the intervention group compared with the control group. LIMITATIONS: This study is limited by the search of a single database and the inclusion of only English language studies. CONCLUSION: Education is a key component of patient care; however, evidence to support an improvement in clinical outcomes is lacking. Further study is needed by the use of rigorous designs to craft a feasible educational intervention that will lead to improved patient care and outcomes.

Suture versus preperitoneal polypropylene mesh for elective umbilical hernia repairs.

BACKGROUND: Repair of primary ventral hernias (PVH) such as umbilical hernias is a common surgical procedure. There is a paucity of risk-adjusted data comparing suture versus mesh repair of these hernias. We compared preperitoneal polypropylene (PP) repair versus suture repair for elective umbilical hernia repair. METHODS: A retrospective review of all elective open PVH repairs at a single institution from 2000-2010 was performed. Only patients with suture or PP repair of umbilical hernias were included. Univariate analysis was conducted and propensity for treatment-adjusted multivariate logistic regression. RESULTS: There were 442 elective open PVH repairs performed; 392 met our inclusion criteria. Of these patients, 126 (32.1%) had a PP repair and 266 (67.9%) underwent suture repair. Median (range) follow-up was 60 mo (1-143). Patients who underwent PP repair had more surgical site infections (SSIs; 19.8% versus 7.9%, P < 0.01) and seromas (14.3% versus 4.1%, P < 0.01). There was no difference in recurrence (5.6% versus 7.5%, P = 0.53). On propensity score-adjusted multivariate analysis, we found that body mass index (odds ratio [OR], 1.10) and smoking status (OR, 2.3) were associated with recurrence. Mesh (OR, 2.34) and American Society of Anesthesiologists (OR, 1.95) were associated with SSI. Only mesh (OR, 3.41) was associated with seroma formation. CONCLUSIONS: Although there was a trend toward more recurrence with suture repair in our study, this was not statistically significant. Mesh repair was associated with more SSI and seromas. Further prospective randomized controlled trial is needed to clarify the role of suture and mesh repair in PVH.

Review of stoma site and midline incisional hernias after stoma reversal.

BACKGROUND: The incidence of incisional hernias after stoma reversal is not well reported. The aim of this study was to systematically review the literature reporting data on incisional hernias after stoma reversal. We evaluated both the incidence of stoma site and midline incisional hernias. METHODS: A systematic review identified studies published between January 1, 1980, and December 31, 2012, reporting the incidence of incisional hernia after stoma reversal at either the stoma site or at the midline incision (in cases requiring laparotomy). Pediatric studies were excluded. Assessment of risk of bias, detection method, and essential study-specific characteristics (follow-up duration, stoma type, age, body mass index, and so forth) was done. RESULTS: Sixteen studies were included in the analysis; 1613 patients had 1613 stomas formed. Fifteen studies assessed stoma site hernias and five studies assessed midline incisional hernias. The median (range) incidence of stoma site incisional hernias was 8.3% (range 0%-33.9%) and for midline incisional hernias was 44.1% (range 8.7%-58.1%). When evaluating only studies with a low risk of bias, the incidence for stoma site incisional hernias is closer to one in three and for midline incisional hernias is closer to one in two. CONCLUSION: Stoma site and midline incisional hernias are significant clinical complications of stoma reversals. The quality of studies available is poor and heterogeneous. Future prospective randomized controlled trials or observational studies with standardized follow-up and outcome definitions/measurements are needed.

Prevalence of surgical site infection at the stoma site following four skin closure techniques: a retrospective cohort study.

BACKGROUND/AIMS: Surgical site infection (SSI) is a common complication of stoma reversal. Studies have suggested that different skin closures affect SSI rates. Our aim was to determine which skin closure technique following stoma reversal leads to the lowest rate of SSI. METHODS: We conducted a retrospective review of all adult patients undergoing stoma reversal at a single institution (2005-2011) and compared the rate of SSI following four skin closure techniques: primary closure (PC), secondary closure (SC), loose PC (LPC), and circular closure (CC). Univariate analysis included χ(2) or Fisher's exact test and ANOVA or Kruskal-Wallis H test for categorical and continuous data, respectively. A multivariate logistic regression model was created to identify predictors of SSI. RESULTS: One hundred and forty-six patients were identified: 40 (27%) PC, 68 (47%) SC, 20 (14%) LPC, and 18 (12%) CC. CC was less likely to have SSI (6%) compared to PC (43%), SC (16%), and LPC (15%; p < 0.01). Increasing body mass index was a predictor of SSI (odds ratio 1.11, 95% confidence interval 1.04-1.12, p < 0.01). CC was associated with the lowest odds of developing SSI [0.07 (0.01-0.63), p = 0.02]. CONCLUSIONS: SSI rate was the lowest for stomas that were closed with CC.

Environmental Impact of Traveling to the Annual American Pediatric Surgical Association Meeting.

PURPOSE: Research has demonstrated negative environmental impacts from in-person conferences. Nonetheless, there are benefits to in-person meetings. The 2023 American Pediatric Surgical Association (APSA) meeting was mostly attended in-person. To understand the environmental impact, this study quantifies the travel emissions generated from that meeting. METHODS: The 2023 APSA meeting was held in Orlando, FL. Using a de-identified list of attendees, the distance between the attendee's home city and Orlando was determined. If ≤ 200 miles, it was assumed the attendee drove. If > 200 miles, the distance between the closest airport and Orlando International Airport was determined. Travel emissions factors represent emissions per person-mile traveled. The Environmental Protection Agency (EPA) Greenhouse Gas Inventory emissions factors for carbon dioxide (CO2), methane (CH4), and nitrous oxide (N2O) were multiplied by travel distances to determine the emissions generated from each attendee. These were aggregated to determine the total meeting travel emissions. The EPA Greenhouse Gas Equivalencies Calculator was used to convert the emissions to a relatable outcome. RESULTS: There were 757 in-person and 135 virtual attendees. Fifty attendees drove and 707 attendees flew. This generated 267,279 kg CO2, 1222 gm CH4, and 8486 gm N2O; equivalent to the emissions generated from the average annual use of 60 gasoline-powered passenger vehicles in the United States. CONCLUSION: Based on attendance to the 2023 APSA meeting, there is a preference for meeting in-person, though the associated environmental cost should be recognized. Based on these results, APSA should consider strategies to mitigate the environmental impact of its annual meeting. LEVELS OF EVIDENCE: N/A.

Patient Satisfaction, chronic pain, and functional status following laparoscopic ventral hernia repair.

BACKGROUND: Ventral hernia repairs are one of the most common surgeries performed. Symptoms are the most common motivation for repair. Unfortunately, outcomes of repair are typically measured in recurrence and infection rather than patient focused results. We correlated factors associated with decreased patient satisfaction, chronic pain, and diminished functional status following laparoscopic ventral hernia repair (LVHR) METHODS: A retrospective study of 201 patients from two affiliated institutions was performed. Patient satisfaction, chronic abdominal pain, pain scores, and Activities Assessment Scale results were obtained in 122 patients. Results were compared with univariate and multivariate analysis. RESULTS: Thirty-two (25.4%) patients were dissatisfied with their LVHR while 21 (17.2%) patients had chronic abdominal pain and 32 (26.2%) patients had poor functional status following LVHR. Decreased patient satisfaction was associated with perception of poor cosmetic outcome (OR 17.3), eventration (OR 10.2), and chronic pain (OR 1.4). Chronic abdominal pain following LVHR was associated with incisional hernia (OR 9.0), recurrence (OR 4.3), eventration (OR 6.0), mesh type (OR 1.9), or ethnicity (OR 0.10). Decreased functional status with LVHR was associated with mesh type used (OR 3.7), alcohol abuse (OR 3.4), chronic abdominal pain (OR 1.3), and age (OR 1.1). CONCLUSIONS: One-fourth of patients have poor quality outcome following LVHR. These outcomes are affected by perception of cosmesis, eventration, chronic pain, hernia type, recurrence, mesh type, and patient characteristics/co-morbidities. Closing central defects and judicious mesh selection may improve patient satisfaction and function. Focus on patient-centered outcomes is warranted.

Takayasu arteritis complicating pregnancy in adolescence.

Pregnant adolescent patients afflicted with Takayasu arteritis represent a clinical entity not seen by many. The care of such patients is often managed by multidisciplinary teams, where vascular surgeons are asked to provide input on cardiovascular implications during and after a pregnant state. Knowledge and understanding of the interaction between the two conditions allows for well-informed decision making and favorable outcomes with pregnancy, as well as proper long-term follow-up and care with appropriate clinicians.

Postoperative mortality and need for transitional care following liver resection for metastatic disease in elderly patients: a population-level analysis of 4026 patients.

OBJECTIVES: The goal of this study was to characterize the association of age with postoperative mortality and need for transitional care following hepatectomy for liver metastases. METHODS: A retrospective cohort study using the Nationwide Inpatient Sample (2005-2008) was performed. Patients undergoing hepatectomy for liver metastases were categorized by age as: Young (aged <65 years); Old (aged 65-74 years), and Oldest (aged ≥75 years). Multivariate logistic regression analyses were performed to identify predictors of in-hospital mortality and need for transitional care (non-home discharge). RESULTS: A total of 4026 patients were identified; 36.6% (n = 1475) were elderly (aged ≥65 years). Rates of in-hospital mortality and non-home discharge increased with advancing age group [1.3% vs. 2.2% vs. 3.3% (P = 0.005) and 2.1% vs. 6.1% vs. 18.3% (P < 0.001), respectively]. Independent predictors of in-hospital mortality were age within the Oldest category [odds ratio (OR) 2.21, 95% confidence interval (CI) 1.19-4.12] and a Deyo Comorbidity Index score of ≥3 (OR 6.95, 95% CI 3.55-13.60). Independent predictors for need for transitional care were age within the Old group (OR 2.44, 95% CI 1.66-3.58), age within the Oldest group (OR 8.48, 95% CI 5.87-12.24), a Deyo score of 1 (OR 2.00, 95% CI 1.40-2.85), a Deyo score of 2 (OR 4.70, 95% CI 2.93-7.56), a Deyo score of ≥3 (OR 6.41, 95% CI 3.67-11.20), and female gender (OR 1.56, 95% CI 1.15-2.11). CONCLUSIONS: Although increasing age was associated with higher risk for in-hospital mortality, the absolute risk was low and within accepted ranges, and comorbidity was the primary driver of mortality. Conversely, need for transitional care was significantly more common in elderly patients. Therefore, liver resection for metastases is safe in well-selected elderly patients, although consideration should be made for potential transitional care needs.

External Validation of a Laboratory Prediction Algorithm for the Reduction of Unnecessary Labs in the Critical Care Setting.

BACKGROUND: Unnecessary laboratory tests contribute to iatrogenic harm and are a major source of waste in the health care system. We previously developed a machine learning algorithm to help clinicians identify unnecessary laboratory tests, but it has not been externally validated. In this study, we externally validate our machine learning algorithm. METHODS: To externally validate the machine learning algorithm that was originally trained on the Medical Information Mart for Intensive Care (MIMIC) III database, we tested the algorithm in a separate institution. We identified and abstracted data for all patients older than 18 years admitted to the intensive care unit at Memorial Hermann Hospital in Houston, Texas (MHH) from January 1, 2020 to November 13, 2020. Using the transfer learning style, we performed external validation of the machine learning algorithm. RESULTS: A total of 651 MHH patients were included. The model performed well in predicting abnormality (area under the curve [AUC] 0.98 for MIMIC III and 0.89 for MHH). The model performed similarly in predicting transitions from normal laboratory range to abnormal (AUC 0.71 for MIMIC III and 0.70 for MHH). The performance of the model in predicting the actual laboratory value was also similar in the MIMIC III (accuracy 0.41) and MHH data (0.45). CONCLUSIONS: We externally validated the machine learning model and showed that the model performed similarly, supporting the generalizability to other settings. While this model demonstrated good performance for predicting abnormal labs and transitions, it does not perform well enough for prediction of laboratory values in most clinical applications.

Costs associated with postoperative intra-abdominal abscess in pediatric perforated appendicitis: A retrospective cohort study.

BACKGROUND: Intra-abdominal abscess, the most common complication after perforated appendicitis, is associated with considerable economic burden. However, costs of intra-abdominal abscesses in children are unknown. We aimed to evaluate resource utilization and costs attributable to intra-abdominal abscess in pediatric perforated appendicitis. METHODS: A single-center retrospective analysis was performed of children (<18 years) who underwent appendectomy for perforated appendicitis (2013-2019). Hospital costs incurred during the index admission and within 30 postoperative days were obtained from the hospital accounting system and inflated to 2019 USD. Generalized linear models were used to determine excess resource utilization and costs attributable to intra-abdominal abscess after adjusting for confounders. RESULTS: Of 763 patients, 153 (20%) developed intra-abdominal abscesses. Eighty-one patients with intra-abdominal abscesses (53%) underwent percutaneous abscess drainage. Intra-abdominal abscess was independently associated with a nearly 8-fold increased risk of 30-day readmission (adjusted risk ratio, 7.8 [95% confidence interval, 4.7-13.0]). Patients who developed an intra-abdominal abscess required 6.1 excess hospital bed days compared to patients without intra-abdominal abscess (95% confidence interval, 5.3-7.0). Adjusted mean hospital costs for patients with intra-abdominal abscess totaled $27,394 (95% confidence interval, $25,688-$29,101) versus $15,586 (95% confidence interval, $15,102-$16,069) for patients without intra-abdominal abscess. Intra-abdominal abscess was associated with an incremental cost of $11,809 (95% confidence interval, $10,029-$13,588). Hospital room costs accounted for 66% of excess costs. CONCLUSION: Postoperative intra-abdominal abscess nearly doubled pediatric perforated appendicitis costs, primarily due to more hospital bed days and associated room costs. Intra-abdominal abscesses resulted in estimated excess costs of $1.8 million during the study period. Even small reductions in intra-abdominal abscess rates or hospital bed days could yield substantial health care savings.

Variability in opioid prescribing practices, knowledge, and beliefs: A survey of providers caring for pediatric surgical patients.

BACKGROUND/PURPOSE: Comprehensive opioid stewardship programs require collective stakeholder alignment and proficiency. We aimed to determine opioid-related prescribing practices, knowledge, and beliefs among providers who care for pediatric surgical patients. METHODS: A single-center, cross-sectional survey was conducted of attending physicians, residents, and advanced practice providers (APPs), who managed pediatric surgical patients. RESULTS: Of 110 providers surveyed, 75% completed the survey. Over half of respondents (n = 43, 52%) reported always/very often prescribing opioids at discharge, with residents reporting the highest rate (66%). Provider types had varying prescribing patterns, including what types of opioids and non-opioids they prescribed. There was a lack of formal training, particularly among residents, of which only 42% reported receiving formal opioid prescribing education. Finally, although only 28% of providers felt that the opioid epidemic affects children, 48% believed pediatric providers' prescribing patterns contributed to the opioid epidemic as a whole, and 80% reported changing their prescribing practices in response. CONCLUSIONS: Significant variability exists in opioid prescribing practices, knowledge, and beliefs among providers who care for pediatric surgical patients. Effective opioid stewardship requires comprehensive policies, pediatric specific guidelines, and education for all providers caring for children to align provider proficiency and optimize prescribing patterns.