American Hospital Association Survey Research Articles

Rural-Urban Disparities in Hospital Services and Outcomes for Children With Medical Complexity

Limited availability of inpatient pediatric services in rural regions has raised concerns about access, safety, and quality of hospital-based care for children. This may be particularly important for children with medical complexity (CMC). To describe differences in the availability of pediatric services at acute care hospitals where rural- and urban-residing CMC presented for hospitalization; identify rural-urban disparities in health care quality and in-hospital mortality; and determine whether the availability of pediatric services at index hospitals or the experience of interfacility transfer modified rural-urban differences in outcomes. This retrospective cohort study examined all-payer claims data from Colorado, Massachusetts, and New Hampshire from 2012 to 2017. Analysis was conducted from May 2023 to July 2024. Participants included CMC younger than 18 years residing in these states and hospitalized during the study period. Rural or urban residence was determined using Rural-Urban Commuting Area codes. Hospitals were categorized as children's hospitals or general hospitals with comprehensive, limited, or no dedicated pediatric services using American Hospital Association survey data. Interfacility transfers between index and definitive care hospitals were identified using health care claims. In-hospital mortality, all-cause 30-day readmission, medical-surgical safety events, and surgical safety events were operationalized using Agency for Healthcare Research and Quality measure specifications. Among 36 943 CMC who experienced 79 906 hospitalizations, 16 525 (44.7%) were female, 26 034 (70.5%) were Medicaid-insured, and 34 008 (92.1%) were urban-residing. Rural-residing CMC were 6.55 times more likely to present to hospitals without dedicated pediatric services (rate ratio [RR], 6.55 [95% CI, 5.86-7.33]) and 2.03 times more likely to present to hospitals without pediatric beds (RR, 2.03 [95% CI, 1.88-2.21]) than urban-residing CMC, with no significant differences in interfacility transfer rates. In unadjusted analysis, rural-residing CMC had a 44% increased risk of in-hospital mortality (RR, 1.44 [95% CI, 1.03-2.02]) with no significant differences in other outcomes. Adjusting for clinical characteristics, the difference in in-hospital mortality was no longer significant. Index hospital type was not a significant modifier of observed rural-urban outcomes, but interfacility transfer was a significant modifier of rural-urban differences in surgical safety events. In this cohort study, rural-residing CMC were significantly more likely to present to hospitals without dedicated pediatric services. These findings suggest that efforts are justified to ensure that all hospital types are prepared to care for CMC.

Identifying opportunities for community EMS fall prevention

BackgroundFalls are a leading cause of morbidity and mortality among older adults in the United States. Current fall prevention interventions rely on provider referral or enrollment during inpatient admissions and require engagement and independence of the patient. Community emergency medical services (CEMS) are a unique opportunity to rapidly identify older adults at risk for falls and provide proactive fall prevention interventions in the home. We describe the demographics and treatment characteristics of the older adult population most likely to benefit from these interventions. Materials and methodsWe linked 2019 Healthcare Cost and Utilization Project Massachusetts State Emergency Department (ED) and State Inpatient Databases with American Hospital Association survey data to query ED encounters and inpatient admissions for adults age ≥55 with ED encounters for fall-related injury between July 1, 2019 and December 31, 2019. Univariable descriptive statistics assessed participant characteristics and bivariable tests of significance compared diagnoses, disposition, and hospital characteristics between older adults with and without an EMS encounter in the six months prior to the presenting fall. ResultsOf 66,027 older adults who presented with a fall to a Massachusetts ED in July-December 2019, 7,942 (11%) had a prior encounter with EMS in the preceding six months, most of which included an injury diagnosis (99%). Compared to older adults without previous EMS encounters, those with previous EMS encounters were more often in poorer health (17% vs. 10% with multiple or complex comorbidities, p < 0.001) and of lower socioeconomic status (12% vs. 8% in lowest neighborhood income quartile, p < 0.001; 10% vs. 6% enrolled in Medicaid, p < 0.001) compared to those without a prior EMS encounter. ConclusionsA significant proportion of older adults presenting to the ED with fall related injury have encounters with EMS in the preceding months. These participants are predisposed to poorer health and economic outcomes worsened by their fall and thus demonstrate a population that would benefit from CEMS fall prevention programs.

Impact of hospital volume and facility characteristics on postoperative outcomes after hepatectomy: A mediation analysis

BackgroundThe impact of hospital procedural volume on outcomes after hepatectomy relative to other facility-related factors remains unclear. We sought to define the comparative impact of hospital volume compared with other facility-related factors on postoperative outcomes among Medicare beneficiaries undergoing hepatectomy. MethodsData on patients who underwent hepatectomy between 2013 and 2021 were collected from the Medicare Standard Analytic Files and linked with facility-level data from the American Hospital Association Survey databases. Hospital volume was stratified into high- (top 10%) and low-volume centers. Propensity score matching was used to account for variable imbalances in patient characteristics among high-compared with low-volume centers. Mediation analysis was employed to delineate facility-related factors responsible for the impact of hospital volume on outcomes with a specific focus on incidence of complications, in-hospital mortality, and failure to rescue. ResultsThe analytic cohort included 22,969 patients from 340 institutions. After propensity score matching, receipt of surgery at a high-volume center was associated with a lower likelihood of postoperative complications (39.9% vs 41.7%, P = .01), in-hospital mortality (2.2% vs 2.8%, P = .02), and failure to rescue (5.4% vs 6.5%, P = .04) versus low-volume centers. Mediation analysis revealed that hospital capacity (bed capacity and nurse-to-bed ratio) contributed the most to the variations in risk of complications and in-hospital mortality, whereas liver transplant program status had the largest impact on failure to rescue. ConclusionsHospital volume is a significant determinant of postoperative outcomes after hepatectomy, with hospital capacity and liver transplant program status being important mediators of this effect. Centralization and optimal resource distribution are important to achieve favorable outcomes following liver resection.

Hospital-based skilled nursing facility survival: Organizational and market-level predictors.

Rising health care costs and consequent increases in Medicare reimbursements have led to many payment reforms over the years. Implementation of the prospective payment system (PPS) for hospitals in 1983 incentivized hospitals to either purchase skilled nursing facilities (SNFs) or utilize their excess capacity to establish one within the hospital. With PPS reimbursement being applied to SNFs in 1998, prior monetary incentives for hospitals to own an SNF disappeared. However, despite the reduction in numbers, many hospitals continued to operate their hospital-based skilled nursing facilities (HBSNFs). This study examines the organizational and market-level factors associated with the survival of HBSNFs using the population ecology of organizations framework. Using American Hospital Association survey data, event histories of all U.S. acute care hospitals with an open HBSNF in 1998 were plotted to examine if a hospital closed its HBSNF during a 22-year period (1998-2020). The primary independent variables included hospital size, ownership, total margin, market competition, and Medicare Advantage penetration. The independent and control variables were lagged by 1 year. Cox regressions were conducted to estimate the hazard ratios capturing the risk of HBSNF closure. The results showed that HBSNFs located in large, not-for-profit hospitals and those operating in less competitive markets had greater odds of surviving. The HBSNF administrators of small, for-profit hospitals and those operating in highly competitive markets could utilize the findings of this study to judiciously allocate slack resources to their HBSNFs to keep those open given the current emphasis on continuity of care by regulatory bodies.

Extended-Spectrum Beta-Lactamase Producing Enterobacterales Infections in the United States, 2012-2021

Background: The 2022 Special Report: COVID-19 U.S. Impact on Antimicrobial Resistance identified continued increases in the rate of extended- spectrum beta-lactamase producing (ESBL) infections in the United States from 2017 through 2020. Using similar data sources and methodology, we examined the trends of species-specific ESBL infections from 2012-2021. Methods: We identified a cohort of patients from the PINC AI and BD Research Insights databases with a clinical culture yielding a Klebsiella pneumoniae or Escherichia coli isolate with accompanying susceptibility testing. E. coli or K. pneumoniae isolates non-susceptible to ceftriaxone, cefotaxime, ceftazidime, or cefepime were considered suggestive of ESBL production. Isolates from patients with no culture yielding the same resistance phenotype of interest in the previous 14 days were counted as an incident case. Community-onset (CO) cultures were obtained ≤ day 3 of hospitalization; hospital-onset (HO) cultures were obtained ≥ day 4. We used a raking procedure to determine weights for extrapolating the number of discharges included in our sample to match the distribution of discharges, stratified by bed size, U.S. census division, urban/rural designation, and teaching status, for U.S. hospitals included in the American Hospital Association survey. We evaluated rates over time due to the changes in number of hospitalizations during the COVID-19 pandemic. Results were stratified by HO and CO, and sterile and non-sterile specimen sources. Results: In 2021, there were 48,936 ESBL K. pneumoniae and 153,112 ESBL E. coli infections among approximately 32 million discharges. Overall, most infections were CO and from non-sterile specimens. From 2012-2021, the rate of ESBL K. pneumoniae increased from 9.54 to 15.28 per 10,000 discharges. ESBL E. coli infections increased from 2012-2020 (30.18 to 51.32 per 10,000 discharges), then declined in 2021 (47.81 per 10,000 discharges) (Table 1, Figure 1). The proportion of non-sterile ESBL E. coli declined from 88% in 2012 to 83% in 2021, and the proportion of non-sterile ESBL K. pneumoniae was 85-87% over the study period (Figure 2). Conclusion: ESBL E. coli and K. pneumoniae infections increased from 2012-2021, although the CO ESBL E. coli rate decreased between 2020 and 2021. Understanding changes in culturing practices over time may provide insights into the increased proportion of ESBL E. coli from sterile sites. Additionally, further investigation into differences in organism trends, particularly in 2021, may inform prevention strategies.

Exploring COVID-19 census burdens by US hospital characteristics: Implications of quality reporting at rural and critical access hospitals.

By assessing longitudinal associations between COVID-19 census burdens and hospital characteristics, such as bed size and critical access status, we can explore whether pandemic-era hospital quality benchmarking requires risk-adjustment or stratification for hospital-level characteristics. We used hospital-level data from the US Department of Health and Human Services including weekly total hospital and COVID-19 censuses from August 2020 to August 2023 and the 2021 American Hospital Association survey. We calculated weekly percentages of total adult hospital beds containing COVID-19 patients. We then calculated the number of weeks each hospital spent at Extreme (≥20% of beds occupied by COVID-19 patients), High (10%-19%), Moderate (5%-9%), and Low (<5%) COVID-19 stress. We assessed longitudinal hospital-level COVID-19 stress, stratified by 15 hospital characteristics including joint commission accreditation, bed size, teaching status, critical access hospital status, and core-based statistical area (CBSA) rurality. Among n = 2582 US hospitals, the median(IQR) weekly percentage of hospital capacity occupied by COVID-19 patients was 6.7%(3.6%-13.0%). 80,268/213,383 (38%) hospital-weeks experienced Low COVID-19 census stress, 28% Moderate stress, 22% High stress, and 12% Extreme stress. COVID-19 census burdens were similar across most hospital characteristics, but were significantly greater for critical access hospitals. US hospitals experienced similar COVID-19 census burdens across multiple institutional characteristics. Evidence-based inclusion of pandemic-era outcomes in hospital quality reporting may not require significant hospital-level risk-adjustment or stratification, with the exception of rural or critical access hospitals, which experienced differentially greater COVID-19 census burdens and may merit hospital-level risk-adjustment considerations.

Association of Hospitals' Experience with Bundled Payment for Care Improvement Model with the Diffusion of Acute Hospital Care at Home.

To examine whether hospitals' experience in a prior payment model incentivizing care coordination is associated with their decision to adopt a new payment program for a care delivery innovation. Data were sourced from Medicare fee-for-service claims in 2017, the list of participants in Bundled Payment for Care Improvement initiatives (BPCI and BPCI-Advanced), the list of hospitals approved for Acute Hospital Care at Home (AHCaH) between November 2020 and August 2022, and the American Hospital Association Survey. Retrospective cohort study. Hospitals' adoption of AHCaH was measured as a function of hospitals' BPCI experiences. Hospitals' BPCI experiences were categorized into five mutually exclusive groups: (1) direct BPCI participation, (2) indirect participation through physician group practices (PGPs) after dropout, (3) indirect participation through PGPs only, (4) dropout only, and (5) no BPCI exposure. All data are derived from pre-existing sources. General acute hospitals eligible for both BPCI initiatives and AHCaH are included. Of 3248 hospitals included in the sample, 7% adopted AHCaH as of August 2022. Hospitals with direct BPCI experience had the highest adoption rate (17.7%), followed by those with indirect participation through BPCI physicians after dropout (11.8%), while those with no exposure to BPCI were least likely to participate (3.2%). Hospitals that adopted AHCaH were more likely to be located in communities where more peer hospitals participated in the program (median 10.8% vs. 0%). After controlling for covariates, the association of the adoption of AHCaH with indirect participation through physicians after dropout was as strong as with early BPCI adopter hospitals (average marginal effect: 5.9 vs. 6.2 pp, p < 0.05), but the other categories were not. Hospitals that participated in the bundled payment model either directly or indirectly PGPs were more likely to adopt a care delivery innovation requiring similar competence in the next period.

What happens to rural hospitals during a ransomware attack? Evidence from Medicare data.

Hospitals are increasingly the target of cybersecurity threats, including ransomware attacks. Little is known about how ransomware attacks affect care at rural hospitals. We used data on hospital ransomware attacks from the Tracking Healthcare Ransomware Events and Traits database, linked to American Hospital Association survey data and Medicare fee-for-service (FFS) claims data from 2016 to 2021. We measured Medicare FFS volume and revenue in the inpatient, outpatient, and emergency room setting-at the hospital-week level. We then conducted a stacked event study analysis, comparing hospital volume and revenue at ransomware-attacked and nonattacked hospitals before and after attacks. Ransomware attacks severely disrupted hospital operations-with comparable effects observed at rural versus urban hospitals. During the first week of the attack, inpatient admissions volume fell by 14.7% at rural hospitals (P=.04) and 16.9% at urban hospitals (P=.01)-recovering to preattack levels within 2-3 weeks. Outpatient visits fell by 35.3% at rural hospitals (P<.01) and 22.0% at urban hospitals (P=.03) during the first week. Emergency room visits fell by 10.0% at rural hospitals (P=.04) and 19.3% at urban hospitals (P=.01). Travel time and distance to the closest nonattacked hospital was 4-7 times greater for rural ransomware-attacked hospitals than for urban ransomware-attacked hospitals. Ransomware attacks disrupted hospital operations in rural and urban areas. Disruptions of similar magnitudes may be more detrimental in rural areas, given the greater distances patients must travel to receive care and the outsized impact that lost revenue may have on rural hospital finances.

REPORTED PRICES FOR HIGH VOLUME HAND SURGERY IN THE ERA OF PRICE TRANSPARENCY: IMPLICATIONS FOR FUTURE POLICY ITERATIONS.

In 2021, the United States enacted a law requiring hospitals to report prices for healthcare services. Across several healthcare services, poor compliance and wide variation in pricing was found. This study aims to investigate variation in reporting and listed prices by hospital features for high-volume hand surgeries including Carpal Tunnel release, Trigger Finger Release, De Quervain Tenosynovitis Release, and Carpometacarpal Arthroplasty. The Turquoise Health price transparency database was used to obtain listed prices and linked to hospital characteristics from the 2021 Annual American Hospital Association Survey. This study used descriptive statistics and generalized linear regression. The analytic cohort included 2,652 hospitals from across the US. The highest rate of price reporting was in the Midwest (52%, n=836) and lowest in the South (39%, n=925). Compared to commercial insurers, ($3,609, 95% CI: $3,414 to $3,805) public insurance rates were significantly lower (Medicare: $1,588, 95% CI: $1,484 to $1,693, adjusted difference = -$2,021, p<0.001, Medicaid: $1,403, (95% CI: $1,194 to $1,612, adjusted difference = -$2,206, p<0.001). Listed rates for self-pay patients were not statistically different from commercial rates. Although pricing for high volume elective hand surgeries is frequently reported, a high proportion of hospitals do not report prices. These data highlight the need for future transparency policy to include pricing for high-volume hand surgery to give patients the ability to make financially informed choices. These results are a valuable aid for surgeons and patients to promote financially conscious decisions.

Chronic Disease Prevalence in the US: Sociodemographic and Geographic Variations by Zip Code Tabulation Area.

We examined the geographic distribution and sociodemographic and economic characteristics of chronic disease prevalence in the US. Understanding disease prevalence and its impact on communities is crucial for effective public health interventions. Data came from the American Community Survey, the American Hospital Association Survey, and the Centers for Disease Control and Prevention's PLACES. We used quartile thresholds for 10 chronic diseases to assess chronic disease prevalence by Zip Code Tabulation Areas (ZCTAs). ZCTAs were scored from 0 to 20 based on their chronic disease prevalence quartile. Three prevalence categories were established: least prevalent (score ≤6), moderately prevalent (score 7-13), and highest prevalence (score ≥14). Community characteristics were compared across categories and spatial analyses to identify clusters of ZCTAs with high disease prevalence. Our study showed a high prevalence of chronic disease in the southeastern region of the US. Populations in ZCTAs with the highest prevalence showed significantly greater socioeconomic disadvantages (ie, lower household income, lower home value, lower educational attainment, and higher uninsured rates) and barriers to health care access (lower percentage of car ownership and longer travel distances to hospital-based intensive care units, emergency departments, federally qualified health centers, and pharmacies) compared with ZCTAs with the lowest prevalence. Socioeconomic disparities and health care access should be addressed in communities with high chronic disease prevalence. Carefully directed resource allocation and interventions are necessary to reduce the effects of chronic disease on these communities. Policy makers and clinicians should prioritize efforts to reduce chronic disease prevalence and improve the overall health and well-being of affected communities throughout the US.

2455. National Trends in the Rates of Drug-resistant Bacteria Commonly Associated with Healthcare in U.S. Acute Care Hospitals, 2019-2021

Abstract Background The 2022 Special Report: COVID-19 U.S. Impact on Antimicrobial Resistance identified significant increases in the rate of hospital-onset (HO) infections during the year 2020 for bacteria commonly associated with healthcare. Using the same data sources and methodology, we provide updated estimates for the year 2021. Methods We identified a cohort of patients from the PINC AI PHD and BD Research Insights databases with a clinical culture yielding an isolate of an organism of interest with accompanying susceptibility testing. Isolates from patients with no culture yielding the same resistance phenotype of interest in the previous 14 days were counted as an incident case. Community-onset (CO) cultures were obtained ≤ day 3 of hospitalization; HO cultures were obtained ≥ day 4. We used a raking-procedure to determine weights for extrapolating the number of discharges included in our sample to match the distribution of discharges, stratified by bed size, U.S. census division, urban/rural designation, and teaching status, for U.S. hospitals included in the American Hospital Association survey. We examined temporal trends using a multivariable logistic model incorporating a survey design with the corresponding weights and hospital designation as the specific cluster controlling for hospital characteristics, month of discharge, proportion of patients in specific age group, proportion of male patients, and database. Results were stratified by HO and CO. Results From 2019-2021, the overall number of hospitals contributing data was 640. HO rates were significantly higher in 2020 and 2021 compared with 2019. The HO rates of VRE, MDRPseu, and CRE were significantly higher in 2021 when compared to 2020 (Table 1). For CO infections, we found that MRSA rates decreased significantly during the study period, while VRE and CRAB were significantly higher in 2021 compared to 2019 (Table 1). Conclusion Hospital onset antibiotic resistant infection rates in 2021 remain higher than pre-pandemic rates with increasing rates for several pathogens in 2021 compared to 2020, while CO MRSA declined. As infections caused by resistant pathogens represent a serious threat to patient safety, further study of factors contributing to the continued increases and decreases may inform prevention strategies. Disclosures All Authors: No reported disclosures

Abstract 14181: Interhospital Transfer for Heart Failure in the United States: A Patient and Hospital-Level Analysis From GWTG-HF

Background: Cardiovascular diseases, particularly heart failure (HF), are common reasons for interhospital transfer (IHT) in the United States. IHT targets patients requiring a higher level of care, and has been associated with increased cost, length of stay (LOS), and mortality. We aim to describe the characteristics and outcomes of patients undergoing IHT, as well as variation in the use of IHT across hospitals. Methods: Patients hospitalized for HF who were admitted from another hospital were compared to those who were not within the Get With The Guidelines-Heart Failure registry between January 2015 and June 2022. Hospital characteristics were obtained from the 2018 American Hospital Association Survey. Clinical data from the receiving hospital were analyzed. Results: Overall, 662,575 patients across 622 hospitals (156,263 patients at 127 transplant centers; 19,852 patients at 55 rural sites) were included, of which 36,375 (5.5%) arrived as IHT. IHT patients were younger (70.0 vs 73.0 years), more male (58 vs 52%), white (72 vs 66%), and more often had private insurance (31% vs 27%). IHT patients had a lower median left ventricular ejection fraction (37 vs 45%), median blood pressure (130/74 vs 139/77 mmHg), and a longer mean LOS (6.3 vs 5.0 days). In a multivariable model, IHT patients had higher in-hospital mortality (OR 1.85, 95% CI 1.76-1.95), were less likely to discharge home (OR 0.79, 95% CI 0.77-0.81), and were more likely to be prescribed most guideline-directed therapies (Figure). Conclusions: More than 1 in 20 patients hospitalized for HF was transferred from another hospital. Sex, racial, economic, and geographic differences exist in which HF patients get transferred, and rates of IHT varied across hospitals. IHT was associated with higher mortality and longer LOS compared with non-transferred patients.

Improving Value in Colorectal Cancer Care: An Economic Analysis of Enhanced Recovery Protocols at a Community Hospital.

Enhanced recovery protocols (ERPs) have been shown to improve the outcomes of gastrointestinal cancer care, leading to reduced morbidity of gastrointestinal treatment and reduced delays in systemic therapy. ERP implementation has also previously shown a reduction in length of stay (LOS) without changing the readmission rate; however, the economic cost associated with these measures has not yet been quantified. The aim of this study was to evaluate the economic costs of ERP implementation for colorectal cancer at a community hospital. The Diagnostic Related Group (DRG) codes were used to assess costs associated with the hospitalizations of cases in the ERP versus non-ERP groups. The American Hospital Association (AHA) Annual Survey from 1999 to 2015 was used to provide the expenses per day for inpatient hospitalization in the United States. Postoperative LOS, average healthcare costs, and postoperative complications between ERP-protocol and non-ERP protocol groups were analyzed using analysis of variance (ANOVA) and independent t-tests. The AHA survey estimated that $2,265 was incurred per day for non-profit hospitals in Florida and $2,346 was incurred per day for the United States. For all DRG codes, the ERP-participating group was associated with a shorter LOS and reduced health care costs. LOS-associated cost was compared between ERP and non-ERP groups: for DRG 329, the total savings was $162,118.8 (n = 12 non-ERP versus n = 8 ERP, P = 4.39 × 10-18); for DRG 330, $314,552.64 (n = 36 non-ERP versus n = 24 ERP, P = 2.72 × 10-22); and for DRG 331, $89,302.73 (n = 11 non-ERP versus n = 23 for ERP, P = 4.19 × 10-20). The implementation of an ERP protocol for colorectal cancer was associated with significantly reduced costs in a community hospital.

Substance Use Disorder Program Availability in Safety-Net and Non–Safety-Net Hospitals in the US

Safety-net hospitals (SNHs) are ideal sites to deliver addiction treatment to patients with substance use disorders (SUDs), but the availability of these services within SNHs nationwide remains unknown. To examine differences in the delivery of different SUD programs in SNHs vs non-SNHs across the US and to determine whether these differences are increased in certain types of SNHs depending on ownership. This cross-sectional analysis used data from the 2021 American Hospital Association Annual Survey of Hospitals to examine the associations of safety-net status and ownership with the availability of SUD services at acute care hospitals in the US. Data analysis was performed from January to March 2022. This study used 2 survey questions from the American Hospital Association survey to determine the delivery of 5 hospital-based SUD services: screening, consultation, inpatient treatment services, outpatient treatment services, and medications for opioid use disorder (MOUD). A total of 2846 hospitals were included: 409 were SNHs and 2437 were non-SNHs. The lowest proportion of hospitals reported offering inpatient treatment services (791 hospitals [27%]), followed by MOUD (1055 hospitals [37%]), and outpatient treatment services (1087 hospitals [38%]). The majority of hospitals reported offering consultation (1704 hospitals [60%]) and screening (2240 hospitals [79%]). In multivariable models, SNHs were significantly less likely to offer SUD services across all 5 categories of services (screening odds ratio [OR], 0.62 [95% CI, 0.48-0.76]; consultation OR, 0.62 [95% CI, 0.47-0.83]; inpatient services OR, 0.73 [95% CI, 0.55-0.97]; outpatient services OR, 0.76 [95% CI, 0.59-0.99]; MOUD OR, 0.6 [95% CI, 0.46-0.78]). With the exception of MOUD, public or for-profit SNHs did not differ significantly from their non-SNH counterparts. However, nonprofit SNHs were significantly less likely to offer all 5 SUD services compared with their non-SNH counterparts (screening OR, 0.52 [95% CI, 0.41-0.66]; consultation OR, 0.56 [95% CI, 0.44-0.73]; inpatient services OR, 0.45 [95% CI, 0.33-0.61]; outpatient services OR, 0.58 [95% CI, 0.44-0.76]; MOUD OR, 0.61 [95% CI, 0.46-0.79]). In this cross-sectional study of SNHs and non-SNHs, SNHs had significantly lower odds of offering the full range of SUD services. These findings add to a growing body of research suggesting that SNHs may face additional barriers to offering SUD programs. Further research is needed to understand these barriers and to identify strategies that support the adoption of evidence-based SUD programs in SNH settings.

Hospital Characteristics Associated With Clinically Integrated Network Participation.

Increased integration of physician organizations and hospitals into health systems has not necessarily improved clinical integration or patient outcomes. However, federal regulators have issued favorable opinions for clinically integrated networks (CINs) as a way to pursue coordination between hospitals and physicians. Hospital organizational affiliations, including independent practice associations (IPA), physician-hospital organizations (PHOs), and accountable care organizations (ACOs), may support CIN participation. No empirical evidence, however, exists about factors associated with CIN participation. Data from the 2019 American Hospital Association survey (n = 4405) were analyzed to quantify hospital CIN participation. Multivariable logistic regression models were estimated to examine whether IPA, PHO, and ACO affiliations were associated with CIN participation, controlling for market factors and hospital characteristics. In 2019, 34.6% of hospitals participated in a CIN. Larger, not-for-profit, and metropolitan hospitals were more likely to participate in CINs. In adjusted analyses, hospitals participating in CINs were more likely to have an IPA (9.5% points, P < 0.001), a PHO (6.1% points, P < 0.001), and ACO (19.3% points, P < 0.001) compared with hospitals not participating in a CIN. Over one-third of hospitals participate in a CIN, despite limited evidence about their effectiveness in delivering value. Results suggest that CIN participation may be a response to integrative norms. Future work should attempt to better define CIN participation and strive to disentangle overlapping organizational participation.

Understanding Variation in In-hospital Mortality After Major Surgery in the United States.

We aimed to quantify the contributions of patient characteristics (PC), hospital structural characteristics (HC), and hospital operative volumes (HOV) to in-hospital mortality (IHM) after major surgery in the United States (US). The volume-outcome relationship correlates higher HOV with decreased IHM. However, IHM after major surgery is multifactorial, and the relative contribution of PC, HC, and HOV to IHM after major surgery is unknown. Patients undergoing major pancreatic, esophageal, lung, bladder, and rectal operations between 2006 and 2011 were identified from the Nationwide Inpatient Sample linked to the American Hospital Association survey. Multilevel logistic regression models were constructed using PC, HC, and HOV to calculate attributable variability in IHM for each. Eighty thousand nine hundred sixty-nine patients across 1025 hospitals were included. Postoperative IHM ranged from 0.9% for rectal to 3.9% for esophageal surgery. Patient characteristics contributed most of the variability in IHM for esophageal (63%), pancreatic (62.9%), rectal (41.2%), and lung (44.4%) operations. HOV explained < 25% of variability for pancreatic, esophageal, lung, and rectal surgery. HC accounted for 16.9% and 17.4% of the variability in IHM for esophageal and rectal surgery. Unexplained variability in IHM was high in the lung (44.3%), bladder (39.3%), and rectal (33.7%) surgery subgroups. Despite recent policy focus on the volume-outcome relationship, HOV was not the most important contributor to IHM for the major organ surgeries studied. PC remains the largest identifiable contributor to hospital mortality. Quality improvement initiatives should emphasize patient optimization and structural improvements, in addition to investigating the yet unexplained sources contributing to IHM.

Statewide Perinatal Quality Improvement, Teamwork, and Communication Activities in Oklahoma and Texas.

The purpose of this study was to describe statewide perinatal quality improvement (QI) activities, specifically implementation of Alliance for Innovation on Maternal Health (AIM) patient safety bundles and use of teamwork and communication tools in obstetric units in Oklahoma and Texas. In January-February 2020, we conducted a survey of AIM-enrolled hospitals in Oklahoma (n = 35) and Texas (n = 120) to gather data on obstetric unit organization and QI processes. Data were linked to hospital characteristics information from the 2019 American Hospital Association survey and hospitals' maternity levels of care from state agencies. We generated descriptive statistics for each state and created an index to summarize adoption of QI processes. We fitted linear regression models to examine how this index varied by hospital characteristics and self-reported ratings for patient safety and AIM bundle implementation. Most obstetric units had standardized clinical processes for obstetric hemorrhage (94% Oklahoma; 97% Texas), massive transfusion (94% Oklahoma; 97% Texas), and severe hypertension in pregnancy (97% Oklahoma; 80% Texas); regularly conducted simulation drills for obstetric emergencies (89% Oklahoma; 92% Texas); had multidisciplinary QI committees (61% Oklahoma; 83% Texas); and conducted debriefs after major obstetric complications (45% Oklahoma; 86% Texas). Few obstetric units offered recent staff training on teamwork and communication to their staff (6% Oklahoma; 22% Texas); those who did were more likely to employ specific strategies to facilitate communication, escalate concerns, and manage staff conflicts. Overall, adoption of QI processes was significantly higher in hospitals in urban than rural areas, teaching than nonteaching, offering higher levels of maternity care, with more staff per shift, and greater delivery volume (all P < .05). The QI adoption index scores were strongly associated with respondents' ratings for patient safety and implementation of maternal safety bundles (both P < .001). Adoption of QI processes varies across obstetric units in Oklahoma and Texas, with implications for implementing future perinatal QI initiatives. Notably, findings highlight the need to reinforce support for rural obstetric units, which often face greater barriers to implementing patient safety and QI processes than urban units.

A STUDY ON ELECTRONIC HEALTH RECORDS

Electronic health records (EHRs) were introduced to enhance patient outcomes and care quality. The adoption of EHRs and patient outcomes were compared in this study. Using State Inpatient Databases connected to the 2011 American Hospital Association survey, we conducted an observational study. Six sizable, diversified states' worth of surgical and medical patients were included. We conducted univariate analyses and created hierarchical regression models that examined the relationships between mortality, readmission rates, and complications with the extent of EHR utilisation. Using a difference-in-differences methodology, we assessed the impact of EHR adoption on outcomes from 2008 to 2011. Patients seeking medical and surgical care went to hospitals with no EHR (3.5%), partial EHR (55.2%), and full EHR (41.3%) systems. Patients at hospitals with full EHR had the lowest rates of inpatient mortality, readmissions, and hospitalizations in univariate studies. Keywords: Electronic Health Records, Bar Coding, Speech Recognition, Patient Confidentiality.

Abstract 13171: Cardiac Rehabilitation and COVID-19: Nationwide Declines in Access and Participation Among Medicare Beneficiaries

Introduction: The effect of the COVID-19 pandemic on availability of and participation in cardiac rehabilitation (CR) participation is unknown. Methods: We used Medicare Fee-for-Service claims, American Hospital Association surveys, and Rural Urban Commuting Area codes to evaluate CR center availability and CR participation (01/2019-12/21). Results: Medicare beneficiaries participated in a mean ± SD of 56,898 ± 2,046 CR sessions per month from 01/2019 - 02/2020. Immediately after the announcement of the public health emergency in 03/2020, CR sessions declined by 93% (to 3,989 sessions in 04/2020) (Figure). The monthly CR sessions recovered gradually through 12/21, but CR participation remained 17% lower than pre-pandemic levels (54,730 ± 2,340 sessions/month in Q1 2019 vs. 45,209 ± 326 sessions/month in Q4 2021, p&lt;0.01). Prior to the pandemic, Medicare beneficiaries received CR at 2,631 ± 8 CR centers. Only 688 centers were in operation in 04/2020 (a 74% decline from pre-pandemic levels), with slow and incomplete recovery in the following months. The number of CR centers in 12/2021 was 5% lower than pre-pandemic levels (2,620 ± 6 centers available in Q1 2019 vs. 2,485 ± 8 centers available in Q4 2021, p &lt;0.01). Ownership status (private/not for profit, private/for profit and public/municipal), teaching status affiliation (major, minor and none), and location (metro, micro and rural) were not associated with post-pandemic CR center availability. Conclusions: The COVID-19 pandemic has been associated with persistent declines in the availability of CR centers and participation in CR sessions among Medicare beneficiaries. Future studies should examine the impact of these closures on health outcomes and equity. Meeting the Million Hearts Initiative’s goal of equitably increasing CR participation will require scalable innovation in CR financing and delivery.

Emergency Department Pediatric Readiness Among US Trauma Centers: A Machine Learning Analysis of Components Associated With Survival.

We used machine learning to identify the highest impact components of emergency department (ED) pediatric readiness for predicting in-hospital survival among children cared for in US trauma centers. ED pediatric readiness is associated with improved short-term and long-term survival among injured children and part of the national verification criteria for US trauma centers. However, the components of ED pediatric readiness most predictive of survival are unknown. This was a retrospective cohort study of injured children below 18 years treated in 458 trauma centers from January 1, 2012, through December 31, 2017, matched to the 2013 National ED Pediatric Readiness Assessment and the American Hospital Association survey. We used machine learning to analyze 265 potential predictors of survival, including 152 ED readiness variables, 29 patient variables, and 84 ED-level and hospital-level variables. The primary outcome was in-hospital survival. There were 274,756 injured children, including 4585 (1.7%) who died. Nine ED pediatric readiness components were associated with the greatest increase in survival: policy for mental health care (+8.8% change in survival), policy for patient assessment (+7.5%), specific respiratory equipment (+7.2%), policy for reduced-dose radiation imaging (+7.0%), physician competency evaluations (+4.9%), recording weight in kilograms (+3.2%), life support courses for nursing (+1.0%-2.5%), and policy on pediatric triage (+2.5%). There was a 268% improvement in survival when the 5 highest impact components were present. ED pediatric readiness components related to specific policies, personnel, and equipment were the strongest predictors of pediatric survival and worked synergistically when combined.