Implementation Experience with a 30-Day Hospital Readmission Risk Score in a Large, Integrated Health System: A Retrospective Study.

BACKGROUND: Driven by quality outcomes and economic incentives, predicting 30-day hospital readmissions remains important for healthcare systems. The Cleveland Clinic Health System (CCHS) implemented an internally validated readmission risk score in the electronic medical record (EMR). OBJECTIVE: We evaluated the predictive accuracy of the readmission risk score across CCHS hospitals, across primary discharge diagnosis categories, between surgical/medical specialties, and by race and ethnicity. DESIGN: Retrospective cohort study. PARTICIPANTS: Adult patients discharged from a CCHS hospital April 2017-September 2020. MAIN MEASURES: Data was obtained from the CCHS EMR and billing databases. All patients discharged from a CCHS hospital were included except those from Oncology and Labor/Delivery, patients with hospice orders, or patients who died during admission. Discharges were categorized as surgical if from a surgical department or surgery was performed. Primary discharge diagnoses were classified per Agency for Healthcare Research and Quality Clinical Classifications Software Level 1 categories. Discrimination performance predicting 30-day readmission is reported using the c-statistic. RESULTS: The final cohort included 600,872 discharges from 11 Northeast Ohio and Florida CCHS hospitals. The readmission risk score for the cohort had a c-statistic of 0.6875 with consistent yearly performance. The c-statistic for hospital sites ranged from 0.6762, CI [0.6634, 0.6876], to 0.7023, CI [0.6903, 0.7132]. Medical and surgical discharges showed consistent performance with c-statistics of 0.6923, CI [0.6807, 0.7045], and 0.6802, CI [0.6681, 0.6925], respectively. Primary discharge diagnosis showed variation, with lower performance for congenital anomalies and neoplasms. COVID-19 had a c-statistic of 0.6387. Subgroup analyses showed c-statistics of > 0.65 across race and ethnicity categories. CONCLUSIONS: The CCHS readmission risk score showed good performance across diverse hospitals, across diagnosis categories, between surgical/medical specialties, and by patient race and ethnicity categories for 3 years after implementation, including during COVID-19. Evaluating clinical decision-making tools post-implementation is crucial to determine their continued relevance, identify opportunities to improve performance, and guide their appropriate use.

Predicting seizure freedom after epilepsy surgery, a challenge in clinical practice.

OBJECTIVE: The objective of this study was to compare the accuracy of clinical judgment in predicting seizure outcome after resective epilepsy surgery relative to two recently published statistical tools [the Epilepsy Surgery Nomogram (ESN) and the modified Seizure-Freedom score (m-SFS)]. METHODS: Details of presurgical evaluations of 20 patients who underwent epilepsy surgery were presented to 20 epilepsy experts. The final surgical treatment was also disclosed. The clinicians were asked to predict the likelihood of a good outcome (Engel 1) at 2 and 5 years in each case. The ESN and the m-SFS predictions were calculated with the data provided to the clinicians. The discriminative ability of clinical judgment, ESN, and m-SFS was assessed by calculating a concordance index (C-index). Expert opinion, the m-SFS and the ESN performances were compared using a Receiver Operating Characteristic (ROC) curve analysis. RESULTS: The mean age at surgery was 29 years (standard deviation [SD] = 14); 40% were male; 70% were right-handed, and thirteen (65%) had an Engel outcome 1 at 2 and 5 years. The mean C-index for the mean physician's prediction was 0.478 with a variance of 0.012. The ESN had an area under the curve (AUC) of 0.528 and 0.533 for the 2-year and 5-year predictions in comparison with the clinicians' predictions that was 0.476, and 0.466, respectively. For the m-SFS, the AUC at 2 years and 5 years was 0.539 and 0.539, respectively. No statistical difference was noted between the ESN and the clinicians or between m-SFS and the ESN, but there is a moderate statistical difference favoring the m-SFS to the clinicians (p 0.0960 and 0.0514, for 2 and 5 years). SIGNIFICANCE: Clinical judgment was not superior to the ESN nor to the m-SFS. Together with the interphysician's prediction variability, our findings reinforce the need for better tools to predict postoperative outcomes.

Predicting risk of pelvic floor disorders 12 and 20 years after delivery.

BACKGROUND: Little progress has been made in the prevention of pelvic floor disorders, despite their significant health and economic impact. The identification of women who are at risk remains a key element in targeting prevention and planning health resource allocation strategies. Although events around the time of childbirth are recognized clinically as important predictors, it is difficult to counsel women and to intervene around the time of childbirth because of an inability to convey a patient's risk accurately in the presence of multiple risk factors and the long time lapse, which is often decades, between obstetric events and the onset of pelvic floor disorders later in life. Prediction models and scoring systems have been used in other areas of medicine to identify patients who are at risk for chronic diseases. Models have been developed for use before delivery that predict short-term risk of pelvic floor disorders after childbirth, but no models that predict long-term risk exist. OBJECTIVE: The purpose of this study was to use variables that are known before and during childbirth to develop and validate prognostic models that will estimate the risks of these disorders 12 and 20 years after delivery. STUDY DESIGN: Obstetric variables were collected from 2 cohorts: (1) women who gave birth in the United Kingdom and New Zealand (n=3763) and (2) women from the Swedish Medical Birth Register (n=4991). Pelvic floor disorders were self-reported 12 years after childbirth in the United Kingdom/New Zealand cohort and 20 years after childbirth in the Swedish Register. The cohorts were split so that data during the first half of the cohort's time period were used to fit prediction models, and validation was performed from the second half (temporal validation). Because there is currently no consensus on how to best define pelvic floor disorders from a patient's perspective, we chose to fit the data for each model using multiple outcome definitions for prolapse, urinary incontinence, fecal incontinence, ≥1 pelvic floor disorder, and ≥2 pelvic floor disorders. Model accuracy was measured in the following manner: (1) by ranking an individual's risk among all subjects in the cohort (discrimination) with the use of a concordance index and (2) by observing whether the predicted probability was too high or low (calibration) at a range of predicted probabilities with the use of visual plots. RESULTS: Models were able to discriminate between women who experienced bothersome symptoms or received treatment at 12 and 20 years, respectively, for pelvic organ prolapse (concordance indices, 0.570, 0.627), urinary incontinence (concordance indices, 0.653, 0.689), fecal incontinence (concordance indices, 0.618, 0.676), ≥1 pelvic floor disorders (concordance indices, 0.639, 0.675), and ≥2 pelvic floor disorders (concordance indices, 0.635, 0.619). Route of delivery and family history of each pelvic floor disorder were strong predictors in most models. Urinary incontinence before and during the index pregnancy was a strong predictor for the development of all pelvic floor disorders in most models 12 years after delivery. The 12- and 20-year bothersome symptoms or treatment for prolapse models were accurate when predictions were provided for risk from 0% to approximately 15%. The 12- and 20-year primiparous model began to over predict when risk rates reached 20%. When we predicted bothersome symptoms or treatment for urinary incontinence, the 12-year models were accurate when predictions ranged from approximately 5-60%; the 20-year primiparous models were accurate from 5% and 80%. For bothersome symptoms or treatment for fecal incontinence, the 12- and 20-year models were accurate from 1-15% risk and began to over predict at rates at >15% and 20%, respectively. CONCLUSION: Models may provide an opportunity before birth to identify women who are at low risk of the development of pelvic floor disorders and may provide institute prevention strategies such as pelvic floor muscle training, weight control, or elective cesarean section for women who are at higher risk. Models are provided at http://riskcalc.org/UR_CHOICE/.

A model to predict risk of blood transfusion after gynecologic surgery.

BACKGROUND: A model that predicts a patient's risk of receiving a blood transfusion may facilitate selective preoperative testing and more efficient perioperative blood management utilization. OBJECTIVE: We sought to construct and validate a model that predicts a patient's risk of receiving a blood transfusion after gynecologic surgery. STUDY DESIGN: In all, 18,319 women who underwent gynecologic surgery at 10 institutions in a single health system by 116 surgeons from January 2010 through June 2014 were analyzed. The data set was split into a model training cohort of 12,219 surgeries performed from January 2010 through December 2012 and a separate validation cohort of 6100 surgeries performed from January 2013 through June 2014. In all, 47 candidate risk factors for transfusion were collected. Multiple logistic models were fit onto the training cohort to predict transfusion within 30 days of surgery. Variables were removed using stepwise backward reduction to find the best parsimonious model. Model discrimination was measured using the concordance index. The model was internally validated using 1000 bootstrapped samples and temporally validated by testing the model's performance in the validation cohort. Calibration and decision curves were plotted to inform clinicians about the accuracy of predicted probabilities and whether the model adds clinical benefit when making decisions. RESULTS: The transfusion rate in the training cohort was 2% (95% confidence interval, 1.72-2.22). The model had excellent discrimination and calibration during internal validation (bias-corrected concordance index, 0.906; 95% confidence interval, 0.890-0.928) and maintained accuracy during temporal validation using the separate validation cohort (concordance index, 0.915; 95% confidence interval, 0.872-0.954). Calibration curves demonstrated the model was accurate up to 40% then it began to overpredict risk. The model provides superior net benefit when clinical decision thresholds are between 0-50% predicted risk. CONCLUSION: This model accurately predicts a patient's risk of transfusion after gynecologic surgery facilitating selective preoperative testing and more efficient perioperative blood management utilization.

Association of glucagon-like peptide-1 receptor agonist use and rates of acute myocardial infarction, stroke and overall mortality in patients with type 2 diabetes mellitus in a large integrated health system.

AIMS: To assess the potential impact of glucagon-like peptide-1 receptor agonist (GLP-1RA) exposure on cardiovascular disease (CVD) and mortality outcomes in patients with type 2 diabetes (T2D), using a large retrospective cohort. RESEARCH DESIGN AND METHODS: Patients who had T2D between 2005 and 2014 (N = 105 862) were identified from the electronic health record system at Cleveland Clinic using a validated electronic phenotype. A time-dependent, Cox, multiple regression analysis was used to assess the association between GLP-1RA exposure and risk of acute myocardial infarction (AMI), stroke/cerebrovascular accident (CVA), and overall mortality, as well as the composite of all three outcomes. The findings were further evaluated by assessing the effect of GLP-1RAs on the same variables in patients with and without prior CVD. The model adjusted for differences in demographic information, hypertension, laboratory/vital signs, history of outcomes, and T2D medications. RESULTS: There were significantly lower rates of AMI (hazard ratio [HR] 0.80, 95% confidence interval [CI] 0.65 to 0.99; P = .045), CVA (HR 0.82, 95% CI 0.74 to 0.91, P < .001), overall mortality (HR 0.48, 95% CI 0.41 to 0.57; P < .001), and the composite outcome (HR 0.82, 95% CI 0.74 to 0.91; P < .002) during the consolidated time that patients were exposed to GLP-1RAs compared to corresponding rates during intervals without GLP-1RA exposure. GLP-1RA treatment was associated with a significant decrease in CVA, mortality, and the composite outcome in patients with and without established CVD, not significantly affecting AMI in these subgroups. CONCLUSIONS: GLP-1RA exposure was found to be associated with a reduction in the risk of cardiovascular events observed and overall mortality among patients with T2D with and without established CVD, after adjusting for potential confounders.

Clinical Use of PCA3 and TMPRSS2:ERG Urinary Biomarkers in African-American Men Undergoing Prostate Biopsy.

PURPOSE: Prostate specific antigen has decreased performance characteristics for the detection of prostate cancer in African-American men. We evaluated urinary PCA3 and TMPRSS2:ERG in a racially diverse group of men. MATERIALS AND METHODS: After institutional review board approval, post-examination urine was prospectively collected before prostate biopsy. PCA3 and TMPRSS2:ERG RNA copies were quantified using transcription mediated amplification assays (Hologic, San Diego, California). Prediction models were created using standard of care variables (age, race, family history, prior biopsy, abnormal digital rectal examination) plus prostate specific antigen. Decision curve analysis was performed to compare the net benefit of PCA3 and TMPRSS2:ERG. RESULTS: Of 304 patients 182 (60%) were African-American and 139 (46%) were diagnosed with prostate cancer (69% African-American). PCA3 and TMPRSS2:ERG scores were greater in men with prostate cancer, 3 or more cores, 33.3% or more cores, greater than 50% involvement of greatest biopsy core and Epstein significant prostate cancer (p <0.01). PCA3 added to the standard of care plus prostate specific antigen model for the detection of any prostate cancer in the overall cohort (0.747 vs 0.677, p <0.0001) in African-American men only (0.711 vs 0.638, p=0.0002) and nonAfrican-American men (0.781 vs 0.732, p=0.0016). PCA3 added to the model for the prediction of high grade prostate cancer for the overall cohort (0.804 vs 0.78, p=0.0002) and African-American men only (0.759 vs 0.717, p=0.0003) but not nonAfrican-American men. Decision curve analysis demonstrated improvement with the addition of PCA3. For African-American men TMPRSS2:ERG did not improve concordance statistics for the detection of prostate cancer. CONCLUSIONS: For African-American men urinary PCA3 improves the ability to predict the presence of any and high grade prostate cancer. However, the TMPRSS2:ERG urinary assay does not add significantly to standard tools.

Newly implemented enhanced recovery pathway positively impacts hospital length of stay.

BACKGROUND: Enhanced recovery pathways (ERPs) are thought to improve surgical outcomes by standardizing perioperative patient care established in evidence-based literature. The objective of this study was to determine the impact of a colorectal surgery ERP on hospital length of stay (LOS) and other patient outcomes. METHODS: This is a comparative effectiveness study of patients undergoing elective colorectal surgery 2 years prior (pre-ERP group) and 2 years after (ERP group) implementation of an ERP program. The primary outcome was hospital LOS. Secondary outcomes included postoperative complications, 30-day readmissions, and 30-day reoperations. Multivariable regression analyses were utilized to control for patient factors, general health factors, diagnosis, surgeon, colon versus rectal operations, and open versus minimally invasive operations-laparoscopic and robotic. An ERP checklist was developed to track adherence to components of the pathway. RESULTS: The study population included 1036 patients: 523 in the pre-ERP group and 513 in the ERP group. Unadjusted LOS was significantly shorter in the ERP group than the control pre-ERP group [3 (IQR 3.5) vs 5 days (IQR 4.6); p < 0.0001]. Multivariable regression analysis confirmed the reduction in LOS, controlling for age, colon/rectum procedure, open/laparoscopic/robotic approach, primary diagnosis, and alvimopan use. Postoperative outcomes were not significantly different between groups except for 30-day readmissions, which were unexpectedly higher in the ERP group (14.6 vs 8.7 %, p = 0.04). CONCLUSIONS: A newly implemented ERP on a dedicated colorectal surgery service in an academic non-university hospital setting resulted in shorter hospital LOS, but increased readmissions, for patients undergoing elective open and minimally invasive colon and rectal surgery. Future multi-institutional studies are needed to understand the impact of ERP on postoperative complications and readmissions.

Glycopeptides versus ß-lactams for the prevention of surgical site infections in cardiovascular and orthopedic surgery: a meta-analysis.

OBJECTIVE: To compare the efficacy of glycopeptides and ß-lactams in preventing surgical site infections (SSIs) in cardiac, vascular, and orthopedic surgery. BACKGROUND: The cost-effectiveness of switching from ß-lactams to glycopeptides for preoperative antibiotic prophylaxis has been controversial. ß-Lactams are generally recommended in clean surgical procedures, but they are ineffective against resistant gram-positive bacteria. METHODS: PubMed, International Pharmaceuticals Abstracts, Scopus, and Cochrane were searched for randomized clinical trials comparing glycopeptides and ß-lactams for prophylaxis in adults undergoing cardiac, vascular, or orthopedic surgery. Abstracts and conference proceedings were included. Two independent reviewers performed study selection, data extraction, and assessment of risk of bias. RESULTS: Fourteen studies with a total of 8952 patients were analyzed. No difference was detected in overall SSIs between antibiotic types. However, compared with ß-lactams, glycopeptides reduced the risk of resistant staphylococcal SSIs by 48% (relative risk, 0.52; 95% confidence interval, 0.29-0.93; P = 0.03) and enterococcal SSIs by 64% (relative risk, 0.36; 95% confidence interval, 0.16-0.80; P = 0.01), but increased respiratory tract infections by 54% (relative risk, 1.54; 95% confidence interval, 1.19-2.01; P ≤ 0.01). Subgroup analysis of cardiac procedures showed superiority of ß-lactams in preventing superficial and deep chest SSIs, susceptible staphylococcal SSIs, and respiratory tract infections. CONCLUSIONS: Glycopeptides reduce the risk of resistant staphylococcal SSIs and enterococcal SSIs, but increase the risk of respiratory tract infections. Additional high-quality randomized clinical trials are needed as these results are limited by high risk of bias.

Stability of Patient Social Determinants of Health With a Focus on Food Insecurity.

INTRODUCTION/OBJECTIVES: Many health systems screen patients for social determinants of health and refer patients with social needs to community service organizations for assistance. However, few studies have examined how social determinants of health change over time in the same individuals. METHODS: We examined patients screened by The MetroHealth System in Cleveland, Ohio for 11 social determinants of health, including food insecurity, financial strain, transportation, housing stability, utilities affordability, other housing problems, intimate partner violence, social connection, physical activity, daily stress, and digital connectivity. We determined changes in these social determinants among patients screened at baseline and again after 6 to 18 months of follow-up. We further examined correlates of changes in food insecurity, because it is a common need among our patients and leads to numerous referrals to community organizations for assistance. RESULTS: A substantial majority of patients had no change in each social determinant. For example, among 18 038 patients screened twice for food insecurity, 13 913 (77.1%) did not screen positive for food insecurity at baseline and follow-up and 1726 (9.6%) screened positive for food insecurity at both times. A total of 1080 (6.0%) did not screen positive for food insecurity at baseline but screened positive at follow-up while 1319 (7.3%) screened positive for food insecurity at baseline but not at follow-up. Among patients screening positive for food insecurity at baseline, screening negative at follow-up was independently associated with being age ≥60 years (odds ratio [OR] = 1.46, 95% confidence interval [CI] = 1.02-2.10), not screening positive for financial strain (OR = 1.64, CI = 1.27-2.13), not screening positive for housing problems (OR = 1.65, CI = 1.28-2.13), and not screening positive for intimate partner violence (OR = 1.45, OR = 1.02-2.08). A longer duration between baseline and follow-up screening was also independently associated with not screening positive for food insecurity at follow-up. Being referred for food assistance was not associated with absence of food insecurity at follow-up (OR = 0.71, CI = 0.47-1.08). CONCLUSIONS: Most patients report no change in specific social determinants of health over 6 to 18 months. Examining changes may identify subgroups at greatest risk for persistence of adverse determinants and help to evaluate the impact of assistance efforts.

A Randomized Trial to Address Food Insecurity and Promote Smoking Cessation Among Low-Income Adults.

Many low-income adults who smoke also have unmet social needs, such as food insecurity, which can serve as a barrier to smoking cessation. We developed a novel intervention to jointly address smoking cessation and food insecurity and assessed its feasibility, acceptability, and preliminary outcomes. We enrolled participants who screened for food insecurity, reported smoking daily, and were ready to quit. All participants received 3 months of resources navigation from a community health worker through monthly telephone calls for referrals and check-ins for smoking cessation and food access resources. Participants randomized to the intervention group received an economic intervention equivalent to the cost of 1 week of groceries/month for 3 months. We randomized 55 participants who were smoking on average 13 cigarettes/day. The trial was feasible and acceptable based on 3-month retention rates (80%) and end-of-study qualitative feedback (91% would recommend the study to others). At 3 months, participants in the intervention versus control group reported a longer length of abstinence from smoking and had a higher proportion of serious quit attempts. Results from this pilot study suggest the importance of attending to social needs, particularly food insecurity, as a strategy to promote smoking cessation among low-income adults who smoke.

Clustering of Social Determinants of Health Among Patients.

INTRODUCTION/OBJECTIVES: Many health systems screen patients for social determinants of health and refer patients with social needs to community organizations for assistance. Understanding how social determinants cluster together may help guide assistance programs. METHODS: This study examined patients screened by The MetroHealth System in Cleveland, Ohio for 9 social determinants, including food insecurity, financial strain, transportation limitations, inability to pay for housing or utilities, intimate partner violence, social isolation, infrequent physical activity, daily stress, and lack of internet access. Clustering analyses were performed to determine which combination of social determinants occurred together more often than would be expected if each determinant were independent of each other. RESULTS: Among 23 161 screened patients, there were 19 dyads, 13 triads, and one tetrad of social determinants that clustered together. The most prevalent triad of food insecurity, social isolation, and inability to pay for housing or utilities occurred among 1095 patients but would be expected to occur among 284 patients, for an observed/expected ratio of 3.85 (95% confidence interval 3.64-4.07). In multivariate analyses, younger, Black, and lower income patients were 2 to 3 times more likely to have this triad compared to older, White, and wealthier patients. CONCLUSIONS: Social determinants of health frequently cluster together, and such clustering is associated with patient demographic characteristics. Further work is needed to determine how social determinant clusters impact health and cost outcomes and to develop programs that can address multiple co-existing social needs.

Food insecurity, social needs, and smoking status among patients in a county hospital system.

Tobacco use in the U.S. is increasingly concentrated among populations with socioeconomic disadvantages such as food insecurity. Building on prior studies showing that food insecurity increases odds of cigarette smoking, the current study sought to examine how food insecurity and other social needs, particularly financial strain, transportation barriers, and housing/utility insecurity, were associated with smoking status among adult patients seen in a county hospital system. We analyzed data from the electronic health record of patients from The MetroHealth System (Cleveland, Ohio, USA), covering a two-year period since implementation of social determinants of health assessments (2019-2021; N = 45,151 patients). Logistic regression analyses were used to examine associations with smoking status. Compared to the overall smoking prevalence (21 %), smoking was higher among patients screening for transportation barriers (41 %), financial strain (39 %), food insecurity (34 %), and housing/utility insecurity (27 %). Each of these social needs was independently associated with increased odds of current smoking (all p < 0.05). Smoking prevalence increased sequentially as the number of social needs increased; with each addition of a social need, there was a dose-response association with higher odds of current smoking (adjusted ORs ranged from 1.56 to 3.76, all p < 0.001), and current smoking specifically among ever smoking patients (adjusted ORs ranged from 1.39 to 3.01, all p < 0.001). There was substantial overlap among several social needs and smoking status. Alongside improving access to evidence-based cessation treatments and services, the findings raise the possibility that addressing social needs might reduce barriers to quitting and thereby reduce tobacco use disparities.

Food as Medicine Clinic: Early Results and Lessons Learned.

INTRODUCTION: Hospital-based food pantries are commonly used to address food insecurity. However, few studies have examined the impact of these food pantries on patients with chronic health conditions. In this study, we sought to assess the effect of a hospital-based food pantry clinic on self-reported dietary changes, health outcomes, and resource utilization. METHODS: This study included food insecure participants with suboptimally controlled congestive heart failure, hypertension, or diabetes who visited a Food as Medicine (FAM) clinic at an academic healthcare system between October 2018 and November 2019. The clinic provided a three-day supply of food for participants and their families up to two times per month for up to 12 months. Baseline, three-month, and six-month surveys were used to assess dietary behaviors, and electronic health record (EHR) data were used to assess health outcomes and utilization. Multivariable Poisson regression was used to explore variables associated with FAM clinic use. RESULTS: At three months, participants self-reported improved dietary behaviors, including increased consumption of fruits and vegetables as snacks and an increased variety of fruits and vegetables consumed. There were no statistically significant changes in clinical or healthcare utilization measures, despite small absolute improvements in systolic blood pressure (SBP), hospitalizations, and emergency department (ED) visits. There was a weak association between FAM clinic visit frequency and changes in dietary behaviors. CONCLUSION: Among patients with chronic diseases, the use of the FAM clinic was associated with improved self-reported dietary behaviors and a nonsignificant improvement in health outcomes and resource utilization.

A Framework for Evaluating Social Determinants of Health Screening and Referrals for Assistance.

INTRODUCTION/OBJECTIVES: Many health systems screen patients for social determinants of health and refer patients with social needs to community service organizations for assistance. We developed a framework based on sequential steps to evaluate this process. METHODS: We reviewed efforts by The MetroHealth System in Cleveland, Ohio and identified 6 sequential steps: patient screened, has social needs, consents to referral to a service organization, referral placed, referral accepted, and referral outcome. Referral outcomes were categorized as resolved (organization provided requested service or patient self-resolved problem), or unresolved (patient unable to be contacted or declined assistance). We then determined the numbers of patients with food insecurity who completed each step, how completion differed by patient characteristics and service organization, and reasons for failure to complete specific steps. RESULTS: We used the framework to evaluate screening and assistance steps among 5741 patients who attended a COVID-19 vaccine clinic from February 15-March 31, 2021 and were followed through April 30, 2021. The percentage of patients who completed each step ranged from 17-98%. Step completion differed by patient age, patient race, and clinic. Of 360 referrals accepted by community organizations, 98 (27%) were resolved. The most common reasons for unresolved referrals were inability of service organization to contact patients (151), no reason stated (71), and patients declined service (30). CONCLUSIONS: A framework based on sequential steps may be used to evaluate social determinants of health screening and assistance programs. Further work is needed to address reasons for failure to complete steps, to include patient perspectives, and to determine long-term outcomes.

A model to predict risk of postpartum infection after Caesarean delivery.

PURPOSE: The purpose of this study is to build and validate a statistical model to predict infection after caesarean delivery (CD). METHODS: Patient and surgical variables within 30 d of CD were collected on 2419 women. Postpartum infection included surgical site infection, urinary tract infection, endomyometritis and pneumonia. The data were split into model development and internal validation (1 January-31 August; N = 1641) and temporal validation subsets (1 September-31 December; N = 778). Logistic regression models were fit to the data with concordance index and calibration curves used to assess accuracy. Internal validation was performed with bootstrapping correcting for bias. RESULTS: Postoperative infection occurred in 8% (95% CI 7.3-9.9), with 5% meeting CDC criteria for surgical site infections (SSI) (95% CI 4.1-5.8). Eight variables were predictive for infection: increasing BMI, higher number of prior Caesarean deliveries, emergent Caesarean delivery, Caesarean for failure to progress, skin closure using stainless steel staples, chorioamnionitis, maternal asthma and lower gestational age. The model discriminated between women with and without infection on internal validation (concordance index = 0.71 95% CI 0.67-0.76) and temporal validation (concordance index = 0.70, 95% CI 0.62, 0.78). CONCLUSIONS: Our model accurately predicts risk of infection after CD. Identification of patients at risk for postoperative infection allows for individualized patient care and counseling.

Predicting 30-Day All-Cause Readmission Risk for Subjects Admitted With Pneumonia at the Point of Care.

BACKGROUND: The pneumonia 30-d readmission rate has been endorsed by the National Quality Forum as a quality metric. Hospital readmissions can potentially be lowered by improving in-hospital care, transitions of care, and post-discharge disease management programs. The purpose of this study was to create an accurate prediction model for determining the risk of 30-d readmission at the point of discharge. METHODS: The model was created using a data set of 1,295 hospitalizations at the Cleveland Clinic Main Campus with pneumonia over 3 y. Candidate variables were limited to structured variables available in the electronic health record. The final model was compared with the Centers for Medicare and Medicaid Services (CMS) model among subjects 65 y of age and older (n = 628) and was externally validated. RESULTS: Three hundred thirty subjects (25%) were readmitted within 30 d. The final model contained 13 variables and had a bias-corrected C statistic of 0.74 (95% CI 0.71-0.77). Number of admissions in the prior 6 months, opioid prescription, serum albumin during the first 24 h, international normalized ratio and blood urea nitrogen during the last 24 h were the predictor variables with the greatest weight in the model. In terms of discriminative performance, the Cleveland Clinic model outperformed the CMS model on the validation cohort (C statistic 0.69 vs 0.60, P = .042). CONCLUSIONS: The proposed risk prediction model performed better than the CMS model. Accurate readmission risk prediction at the point of discharge is feasible and can potentially be used to focus post-acute care interventions in a high-risk group of patients.

Patient Characteristics Associated With Severe Hypoglycemia in a Type 2 Diabetes Cohort in a Large, Integrated Health Care System From 2006 to 2015.

OBJECTIVE: To identify severe hypoglycemia events, defined as emergency department visits or hospitalizations for hypoglycemia, in patients with type 2 diabetes receiving care in a large health system and to identify patient characteristics associated with severe hypoglycemia events. RESEARCH DESIGN AND METHODS: This was a retrospective cohort study from January 2006 to December 2015 using the electronic medical record in the Cleveland Clinic Health System (CCHS). Participants included 50,439 patients with type 2 diabetes receiving care in the CCHS. Number of severe hypoglycemia events and associated patient characteristics were identified. RESULTS: The incidence proportion of severe hypoglycemia increased from 0.12% in 2006 to 0.31% in 2015 (P = 0.01). Compared with patients who did not experience severe hypoglycemia, those with severe hypoglycemia had similar median glycosylated hemoglobin (HbA1c) levels. More patients with severe hypoglycemia versus those without had a prior diagnosis of nonsevere hypoglycemia (9% vs. 2%, P < 0.001). Logistic regression confirmed an increased odds for severe hypoglycemia with insulin, sulfonylureas, increased number of diabetes medications, history of nonsevere hypoglycemia (odds ratio [OR] 3.01, P < 0.001), HbA1c <6% (42 mmol/mol) (OR 1.95, P < 0.001), black race, and increased Charlson comorbidity index. Lower odds of severe hypoglycemia were noted with higher BMI and use of metformin, dipeptidyl peptidase 4 inhibitors, and glucagon-like peptide 1 agonists. CONCLUSIONS: In this retrospective study of patients with type 2 diabetes with severe hypoglycemia, patient characteristics were identified. Patients with severe hypoglycemia had previous nonsevere hypoglycemia diagnoses more frequently than those without. Identifying patients at high risk at the point of care can allow for change in modifiable risk factors and prevention of severe hypoglycemia events.

Next-generation prostate cancer risk calculator for primary care physicians.

INTRODUCTION: Current prostate cancer risk calculators are limited in impact because only a probability of having prostate cancer is provided. We developed the next generation of prostate cancer risk calculator that incorporates life expectancy in order to better evaluate prostate cancer risk in context to a patient's age and comorbidity. METHODS: We combined two cohorts to develop the new risk calculator. The first was 5638 subjects who all underwent a prostate biopsy for prostate cancer detection. The second was 979 men diagnosed with prostate cancer with long-term survival data. Two regression models were used to create multivariable nomograms and an online prostate cancer risk calculator was developed. RESULTS: Of the 5638 patients who underwent a prostate biopsy, 629 (11%) were diagnosed with aggressive prostate cancer (Gleason Score 7[4+3] or more). Of the 979 patients who underwent treatment for prostate cancer, the 10-year overall survival (OS) was 49.6% (95% confidence interval [CI] 46.6-52.9). The first multivariable nomogram for cancer risk had a concordance index of 0.74 (95% CI 0.72, 0.76), and the second nomogram to predict survival had a concordance index of 0.71 (95% CI 0.69-0.72). The next-generation prostate cancer risk calculator was developed online and is available at: http://riskcalc.org/ProstateCA_Screen_Tool. CONCLUSIONS: We have developed the next-generation prostate cancer risk calculator that incorporates a patient's life expectancy based on age and comorbidity. This approach will better evaluate prostate cancer risk. Future studies examining other populations will be needed for validation.

Effect of glycemic control on the Diabetes Complications Severity Index score and development of complications in people with newly diagnosed type 2 diabetes.

BACKGROUND: The aim of the present study was to assess the longitudinal accumulation of diabetes-related complications and the effect of glycemic control on the Diabetes Complications Severity Index (DCSI) score in people with newly diagnosed type 2 diabetes (T2D). METHODS: A retrospective cohort study was conducted using electronic health records from a large integrated healthcare system. People with newly diagnosed T2D were identified between 2005 and 2016 and stratified by initial HbA1c category (<7%, <8%, ≥8%). The DCSI scores were determined for each study year, and the cumulative incidence of diabetes-related complications was assessed. A Cox proportional hazard model was used to evaluate the effect of baseline HbA1c and worsening glycemic (HbA1c) control on longitudinal changes in DCSI scores. RESULTS: Of 32 174 people identified as having newly diagnosed T2D, 14 016 (44%), 21 657 (67%), and 9983 (31%) had an initial or baseline HbA1c <7%, <8%, and ≥8%, respectively. Ten years after diabetes diagnosis, retinopathy, chronic kidney disease, coronary heart disease, and neuropathy were diagnosed in 22%, 29%, 24%, and 36% of people. Baseline HbA1c did not affect the observed trend in longitudinal changes in DCSI scores throughout the 11-year period. For people in each of the initial HbA1c groups (<7%, <8%, ≥8%), worsening or persistently poor glycemic control was significantly associated with a 10%, 19%, or 16% increase in the risk of experiencing an increased DCSI score, respectively (all P < 0.01). CONCLUSIONS: Baseline glycemic control had no apparent effect on longitudinal changes in DCSI score. Worsening or persistently poor glycemic control was associated with an increased risk of an increase in the DCSI score.

External validation of the TIMI risk score for secondary cardiovascular events among patients with recent myocardial infarction.

BACKGROUND AND AIMS: Risk stratification of patients with recent myocardial infarction (MI) for subsequent cardiovascular (CV) events helps identify patients most likely to benefit from secondary prevention therapies. This study externally validated a new risk score (TRS2˚P) for secondary events derived from the TRA2°P-TIMI 50 trial among post-MI patients from two large health care systems. METHODS: This retrospective cohort study included 9618 patients treated for acute MI at either the Cleveland Clinic (CC) or Geisinger Health System (GHS) between 2008 and 2013. Patients with a clinic visit within 2-52 weeks of MI were included and followed for CV death, repeat MI, and ischemic stroke through electronic medical records (EMR). The TRS2˚P is based on nine factors determined through EMR documentation. Discrimination and calibration of the TRS2˚P were quantified in both patient populations. RESULTS: MI patients at CC and GHS were older, had more comorbidities, received fewer medications, and had higher 3-year event rates compared to subjects in the TRA2°P trial: 31% (CC), 33% (GHS), and 10% (TRA2°P-TIMI 50). The proposed risk score had similar discrimination across the three cohorts with c-statistics of 0.66 (CC), 0.66 (GHS), and 0.67 (TRA2°P-TIMI 50). A strong graded relationship between the risk score and event rates was observed in all cohorts, though 3-year event rates were consistently higher within TRS2°P strata in the CC and GHS cohorts relative to TRA2˚P-TIMI 50. CONCLUSIONS: The TRS2˚P demonstrated consistent risk discrimination across trial and non-trial patients with recent MI, but event rates were consistently higher in the non-trial cohorts.