Medicaid Status Research Articles

Medicaid Status is Independently Predictive of Increased Complications, Readmission, and Mortality Following Primary Total Shoulder Arthroplasty

In recent years, several studies have evaluated the effect of Medicaid insurance status on total shoulder arthroplasty (TSA) outcomes and have presented discordant findings. The purpose of this study is to determine if Medicaid status is an independent predictor of all-cause complications, readmission, revision, and mortality following elective primary TSA using a large, national administrative claims database. The Nationwide Readmissions Database (NRD) was queried to identify patients who underwent elective primary TSA from 2016 to 2020. Patients were propensity score matched in a 1:1 proportion based on age, sex, and discharge weight, yielding 15,374 Medicaid cases and 15,448 control cases. Patient demographic and discharge information, preoperative comorbidities, and postoperative outcomes were compared with bivariate analysis. Binary logistic regression was performed to account for the influence of variables other than Medicaid status on postoperative outcomes. Medicaid patients had higher rates of preoperative comorbidities, higher Charlson-Deyo Comorbidity Index scores, and lower household incomes than matched controls. Compared to controls, Medicaid patients undergoing TSA had higher odds of adverse clinical outcomes, including all-cause complications, readmission, and mortality within 180 days, along with other specific medical and implant-related complications including broken hardware, dislocation, prosthetic loosening, and surgical site infection. Medicaid status was independently predictive of increased rates of all-cause complications within 180 days, readmission within 180 days, dislocation, pneumonia, sepsis, and decreased rates of prosthetic loosening. Medicaid patients had an increased mean cost of $1,396 and increased mean length of stay of 0.4 days. Medicaid status was independently predictive of readmission, complications, and mortality within 180 days of primary TSA, as well as other specific medical and surgical complications. Medicaid patients experience higher admission costs and longer hospital stays compared to those with other insurance types. Medicaid status is a risk factor for adverse clinical outcomes, and orthopedic surgeons need to consider the multitude of disparities that Medicaid patients experience when determining surgical options, treatment plans, and hospital disposition.

Hospital and Patient Factors Associated with Length of Hospitalization in Patients Who Have Osteoarthritis Undergoing Primary Total Knee Arthroplasty: An Analysis of National Data

Background/Objective: By 2040, an estimated 3.5 million primary total knee arthroplasties (TKAs) will be performed each year in the United States (U.S.) osteoarthritis (OA) is the most common indication for primary TKA. We examined the association of hospital, regional, and patient-level factors with extended lengths of hospital stay (eLOS). MethodsPatients who have OA who underwent primary TKA from 2016 to 2019 were queried using a national inpatient database. We used the International Classification of Disease (ICD-10) codes to identify diagnoses and procedures. There were 2,592,469 patients who had OA who underwent primary TKA from 2016 to 2019. We used univariate and multivariable-adjusted logistic regression analyses to assess whether patient, payer, hospital, and geographic factors were associated with an eLOS. Predictive probabilities from multivariable analyses were used in the Area Under the Curve (AUC) analysis. ResultsPatient race and ethnicity, Medicaid or Medicare payer status, income, age/sex, and nearly all regional and hospital characteristics were independently associated with eLOS (> 3 days; ROC C-statistic = 0.74). Sensitivity analyses that used the most recent years of data from 2020 to 2021 (COVID-19 pandemic years), or adjusted for individual organ system complications reproduced the main results without much attenuation. ConclusionAge, sex, race, ethnicity, hospital location and teaching status, elective procedure designation, perioperative complications, and insurance payer status significantly influenced the length of hospital stay (LOS) for primary total knee arthroplasty (TKA) hospitalizations in the US. Recognized disparities were linked to longer hospital stays after primary TKA in patients who had osteoarthritis (OA). Implementing policies and interventions that target these factors could help shorten hospital stays for high-risk patients after primary TKA.

Relationship between Neighborhood-Level Social Risk Factor Measures and Presenting Glaucoma Severity Utilizing Multilevel Modeling

PurposeThe neighborhood and built environment social determinant of health domain has several social risk factors (SRF) that are modifiable through policy efforts. We investigated the impact of neighborhood-level SRFs on presenting glaucoma severity at a tertiary eye care center. DesignA cross-sectional study from August 2012 to May 2022 in the University of Michigan (UM) electronic health record (EHR). ParticipantsPatients with a diagnosis of any open angle glaucoma with ≥1 eye care visit at the UM Kellogg Eye Center and ≥1 reliable visual field. MethodsParticipants who met inclusion criteria were identified by ICD9/10 codes (365.x/H40.x). Data extracted from the EHR included patient demographics, address, presenting mean deviation (MD), and visual field reliability. Addresses were mapped to SRF measures at the census tract, block group, and county levels. Multilevel linear regression models were used to estimate the fixed effects of each SRF on MD, after adjusting for patient-level demographic factors and a random effect for neighborhood. Interactions between each SRF measure with patient-level race and Medicaid status were tested for an additive effect on MD. Main Outcome MeasuresThe main outcome measure was the effect of SRF on presenting MD. ResultsIn total, 4428 patients were included in the analysis who were, on average, 70.3 years old (standard deviation=11.9), 52.6% self-identified as female, 75.8% self-identified as White, and 8.9% had Medicaid. The median value of presenting MD was -4.94 dB (interquartile range=-11.45 to -2.07 dB). Neighborhood differences accounted for 4.4% of the variability in presenting MD. Neighborhood-level measures, including worse area deprivation (estimate, β=-0.31 per 1-unit increase; p<0.001), increased segregation (β=-0.92 per 0.1-unit increase in Theil’s H index; p<0.001), and increased neighborhood Medicaid (β=-0.68; p<0.001) were associated with worse presenting MD. Significant interaction effects with race and Medicaid status were found in several neighborhood-level SRF measures. ConclusionsAlthough patients’ neighborhood SRF measures accounted for a minority of the variability in presenting MD, most neighborhood-level SRFs are modifiable and were associated with clinically meaningful differences in presenting MD. Policies that aim to reduce neighborhood inequities to address allocation of resources could have lasting impacts on vision outcomes.

Concentration of High-Cost Head and Neck Cancer Surgical Patients.

Health care costs are disproportionately concentrated among a small number of patients. We sought to identify variables associated with high-cost patients and high hospital concentration of high-cost patients and to examine associations with short-term outcomes in head and neck cancer (HNCA) surgery. The Nationwide Inpatient Sample was used to identify 170,577 patients who underwent HNCA surgery in 2001-2011. High-cost patients were defined as patients whose costs of care were in the top decile, and high-concentration hospitals were defined as those whose percentage of high-cost patients was in the top decile. Multivariable regression was used to evaluate associations between cost and patient and hospital variables, postoperative complications, and in-hospital mortality. Costs associated with high-cost patients were 4.47-fold greater than the remaining 90% of patients. High-concentration hospitals treated 36% of all high-cost patients. High-cost patients were more likely to be non-white (OR = 2.08 [1.45-2.97]), have oral cavity cancer (OR = 1.21 [1.05-1.39]), advanced comorbidity (OR = 1.53 [1.31-1.77]), Medicaid (OR = 1.93 [1.62-2.31]) or self-pay payor status (OR = 1.72 [1.38-2.14]), income>50th percentile (OR = 1.25 [1.05-1.51]), undergo major procedures (OR = 3.52 [3.07-4.05]) and have non-routine discharge (OR = 7.50 [6.01-9.35]). High-concentration hospitals were more likely to be teaching hospitals (OR = 3.14 [1.64-6.05]) and less likely to be urban (OR = 0.20 [0.04-0.93]). After controlling for all other variables, high-cost patients were associated with an increased odds of mortality (OR = 8.00 [5.89-10.85]) and postoperative complications (OR = 5.88 [5.18-6.68]). High-concentration hospitals were associated with an increased odds of postoperative complications (OR = 1.31 [1.08-1.61]) but were not associated with increased mortality (OR = 0.98 [0.67-1.44]). High-cost HNCA surgical patients are associated with increased postoperative morbidity and mortality, and are disproportionately concentrated at teaching hospitals. 4 Laryngoscope, 2024.

Long-term ambient sulfur dioxide exposure during gestation and preterm birth in North Carolina, 2003–2015

BackgroundCoal-fired power plants are major contributors of ambient sulfur dioxide (SO2) air pollution. Epidemiological literature suggests an adverse association between SO2 exposure during gestation and preterm birth (PTB; <37 weeks completed gestation). PTB is strongly associated with infant mortality and increased risk for later life morbidities. ObjectiveWe investigated associations between SO2 and PTB in North Carolina and evaluated whether the associations were modified by race/ethnicity. MethodsWe assembled a retrospective, administrative cohort of singleton births in North Carolina from 2003 to 2015. We used US EPA EQUATES data to assign long-term SO2 gestational exposures to eligible births for the entire pregnancy and by trimester. We used multivariable generalized linear regression to estimate risk differences (RD (95%CI)) per 1-ppb increase in SO2, adjusted for gestational parent education, Medicaid status, marital status, and season of conception. Multi-pollutant models were additionally adjusted for other criteria air co-pollutants (O3, PM2.5, NO2). ResultsThe median SO2 (24-h average) across exposure windows was ∼1.5 (IQR: 1.8) ppb. The overall baseline risk for PTB was 8756 per 100,000 live births. When stratified by race/ethnicity, the baseline risk for PTB was 12215, 7824, and 7187 per 100,000 live births among non-Hispanic Black, non-Hispanic white, and Hispanic births, respectively. RDs per 1-ppb increase in SO2 averaged across the entire pregnancy were 317.0 (95%CI: 279.4, 354.5) and 568.2 (95%CI: 500.3, 636.1) per 100,000 live births for single- and multi-pollutant models, respectively. For the PTB multi-pollutant models, we observed similar RDs for non-Hispanic Black participants (669.6 [95%CI: 573.9, 765.2]) and non-Hispanic white participants (635.4 [95%CI: 557.2, 713.6]) with smaller RDs for Hispanic participants (336.8 [95%CI: 241.3, 432.2]). SignificanceThe results for our adjusted single- and multi-pollutant models showed adverse associations between SO2 and PTB, with some evidence of effect measure modification by race/ethnicity within subcategories of PTB.

The relative impact of labor induction versus improved labor management: Before and after the ARRIVE (a randomized trial of induction vs. expectant management) trial.

To evaluate the association of labor induction on cesarean delivery and other maternal and neonatal outcomes in low-risk, full-term patients in community hospitals during a period of concerted effort to safely prevent cesarean delivery. We performed a retrospective cohort study using the California Maternal Data Center comprised linked discharge diagnoses and birth certificate data for all low-risk, nulliparous, term, singleton, vertex (NTSV) individuals between 39 and 41 weeks from three Sacramento Valley community hospitals from 2016 to 2022 (N = 10,821) during a period of state-wide efforts to safely reduce cesarean rates. Logistic regression was used to calculate odds ratios (ORs) and adjusted odds ratios (aORs) after labor induction in two time periods before and after the ARRIVE trial. During the study period, labor induction increased from 14.7% to 23.1%. Controlling for maternal age, pre-pregnancy BMI, birthweight, maternal race and ethnicity, birthplace, English language, gestational age, Medicaid status, delivery year, and labor induction was associated with an increased aOR of 1.67 (95% CI 1.48-1.89) for cesarean delivery. We found a trend toward increased aOR of chorioamnionitis but no differences in blood transfusion, severe maternal morbidity, unexpected newborn complications, chorioamnionitis, operative vaginal delivery, maternal lacerations, and shoulder dystocia with labor induction. A decrease aOR of cesarean delivery was observed comparing all births in 2019-2021 to 2016-2018. Labor induction was associated with an increased aOR for cesarean delivery both before and after the ARRIVE trial. A decreased aOR for cesarean delivery was observed during the period of statewide efforts to safely reduce cesarean delivery both with and without labor induction.

Insurance Disparities in Quality of Care Among Patients With Head and Neck Cancer

Significant insurance status disparities have been demonstrated in head and neck cancer (HNC) outcomes. The effects of insurance status on HNC outcomes may be explained by differential access to high-quality care. To evaluate the association of insurance status with the quality of the treating hospital and receipt of guideline-compliant care among patients with HNC. This retrospective cohort study of data from the California Cancer Registry dataset linked with discharge records and hospital characteristics from the California Department of Health Care Access and Information included adult patients with HNC diagnosed between January 1, 2010, and December 31, 2019. Data were analyzed from May 10, 2023, to March 25, 2024. Insurance status: commercial, Medicare, Medicaid, uninsured, other, or unknown. Quality of the treating hospital (tertiles), receipt of National Comprehensive Cancer Network guideline-compliant care, and overall survival. A total of 23 933 patients (mean [SD] age, 64.8 [12.3] years; 75.3% male) met the inclusion criteria. Treatment in top-tertile hospitals (hazard ratio, 0.87; 95% CI, 0.79-0.95) was associated with improved overall survival compared with treatment in bottom-tertile hospitals. Medicare (odds ratio [OR], 0.78; 95% CI, 0.73-0.84), Medicaid (OR, 0.60; 95% CI, 0.54-0.66), and uninsured (OR, 0.38; 95% CI, 0.29-0.49) status were associated with lower likelihood of treatment in high-quality hospitals compared with commercial insurance. Among patients with advanced disease, Medicaid (OR, 0.72; 95% CI, 0.62-0.83) and uninsured (OR, 0.64; 95% CI, 0.44-0.93) patients were less likely to receive dual-modality therapy. Among patients with surgically resected advanced disease, Medicaid coverage (OR, 0.73; 95% CI, 0.58-0.93) was associated with lower likelihood of receiving adjuvant radiotherapy. This study found significant insurance disparities in quality of care among patients with HNC. These findings highlight the need for continued health insurance reform in the US to improve the quality of insurance coverage, in addition to expanding access to health insurance.

Prevalence and association of polycystic ovary syndrome and hidradenitis suppurativa in underrepresented groups

Hidradenitis suppurativa (HS) is an inflammatory disorder of follicular biology; androgens are believed to be involved in its pathogenesis. Polycystic ovary syndrome (PCOS) is similarly characterized by hyperandrogenism. Previous studies have found a lasting association of HS and PCOS. Socioeconomic status (SES) has been described as a comorbidity for both HS and PCOS that has not been accounted for in prior studies; we sought to investigate this association while adjusting for this. We also analyzed the prevalence of PCOS among HS patients. Using the All of Us database, female HS patients were stratified by PCOS diagnosis and compared by age, race, and ethnicity. Female HS patients were also nearest-neighbor propensity-score matched to controls at a 4:1 ratio, selecting for race, ethnicity, age, ever smoker, alcohol use disorder, obesity, type II diabetes, Medicaid status, and community deprivation index. Univariable and multivariable logistic regression was conducted to estimate the effect of HS on the presence of PCOS. The distribution of race among HS patients with PCOS was significantly different than HS patients without PCOS. A total of 1,022 female HS patients and 4,088 matched female controls were included. Significantly more patients carried a diagnosis of PCOS compared to controls (8.8% versus 4.3%, p < .001). In multivariable logistic regression, PCOS was significantly associated with HS [OR 1.71 (95% CI 1.34–2.17)]. This is the first study investigating the association of HS and PCOS within the All of Us database. We found that females with HS had a 1.34- to 2.17-fold increased odds of having PCOS, which is consistent with previous analyses. However, our analysis, in addition to controlling for common medical co-morbidities found in both HS and PCOS, also accounts for markers of SES at an individual and community level, further strengthening the association of HS with PCOS.

Association of patient-reported experience of care with cancer clinical trial participation.

11117 Background: Few individuals diagnosed with cancer participate in clinical trials. Patient-reported experience of care (PEC), assessing patients’ perspectives on interactions with healthcare providers and systems, is important in evaluating quality of care. To help identify patient-centered strategies that may enhance trial enrollment, we examined whether PEC may predict enrollment in clinical trials. Methods: This study utilized Surveillance, Epidemiology, and End Results-Consumer Assessment of Healthcare Providers and Systems (SEER-CAHPS) linked data, including SEER registry data, CAHPS PEC surveys, and Medicare claims. Individuals in SEER-CAHPS who were diagnosed with lung cancer 2000-2018 and completed a CAHPS survey ≥6 months following diagnosis were included. The study outcome, clinical trial enrollment after cancer diagnosis, was determined using Medicare claims; among individuals who enrolled in clinical trials, only those completing a CAHPS survey before trial enrollment were included. PEC was assessed using 5 global ratings (Overall Care, Personal Doctor, Specialist, Health Plan, and Prescription Drug Plan) and 6 composite measures (Doctor Communications, Getting Needed Care, Getting Care Quickly, Getting Needed Drugs, Care Coordination, and Customer Service); all were scored 0 (worse PEC) to 10 (best PEC). Associations of trial participation and PEC were assessed using multivariable logistic regression, controlling for age, race/ethnicity, dual Medicaid status, fee-for-service vs Medicare Advantage plan at diagnosis, and physical problems limiting activities, and were weighted to correspond to the overall SEER population. Results: The study included 5,361 individuals diagnosed with lung cancer; 4.8% of the weighted population enrolled in a clinical trial. Patients with higher PEC ratings for their Specialist (odds ratio [OR] for a 1-point higher PEC score 0.86, 95% CI 0.76, 0.98); for their Health Plan (OR 0.87, 95% CI 0.80, 0.96); and for Getting Needed Care (OR 0.89, 95% CI 0.81, 0.98) had significantly (p&lt;0.05) lower oddsof having enrolled in a clinical trial. Among individuals without physical limitations, association of PEC and trial enrollment were similar. Among individuals with physical limitations, higher PEC ratings for Getting Care Quickly were associated with higher odds (OR 1.20, 95% CI 1.03, 1.39) of having enrolled in a trial. Conclusions: In this population, reporting worse PEC in certain domains was significantly associated with increased clinical trial enrollment. This suggests patients less happy with their clinical care may be more likely to enroll in trials, perhaps to receive different types of care and/or from different providers. PEC associations with trial enrollment differ between patients with vs without physical limitations, suggesting different strategies may be effective for increasing trial enrollment among these two sub-groups.

Exploring Sociodemographic Factors in Allergic Fungal Rhinosinusitis in a Northern California Patient Population.

Allergic fungal rhinosinusitis (AFRS) is a subtype of chronic rhinosinusitis (CRS) that has previously been associated with younger age and Black patients. However, the role of demographic and socioeconomic factors in AFRS severity remains to be fully elucidated. The objective of this study was to determine whether demographic and socioeconomic factors are associated with incidence of AFRS, as well as with disease severity in Northern California. A retrospective cohort study was conducted of adult patients with AFRS and CRS from 2010 to 2019. AFRS was determined by the Bent and Kuhn criteria, and severity was assessed by radiographic evidence of cranioorbital invasion and other clinical parameters. Chi-square and t-test were used to assess demographic and socioeconomic differences between AFRS and CRS cohorts, and multivariable logistic regression was used to assess risk factors for severe AFRS. Black patients represented 26.2% (55/210 patients) of the AFRS group and 4.9% (842/17,300 patients) of the CRS group, with pairwise comparison of race/ethnicity categories showing that the AFRS group had significantly higher proportions of Black race/ethnicity compared with other race/ethnicities (p < 0.01). AFRS and CRS groups differed significantly by age, with mean ages of 48.7 and 51.0 years, respectively (p = 0.04). There were no significant differences in gender, Medicaid status, comorbidities, and socioeconomic status measures. Multivariate logistic regression showed that Black patients had higher odds of having severe AFRS (adjusted odds ratio = 2.29; 95% confidence interval: 1.18-4.45). AFRS has a unique predilection for Black patients, and severe disease is also more likely in this population.

Abstract 6470: Real-world analysis of insurance status and survival in patients with selected GI cancers

Abstract Background: Health insurance is a critical determinant of healthcare access and outcomes in gastrointestinal (GI) cancers. Epidemiological data have shown that insurance-associated healthcare outcomes vary by diagnosis, geographic location, and demographics. Thus, it is critical for cancer providers to understand local impacts of insurance status on patient outcomes. Here, we present real-world outcomes for pancreatic, hepatobiliary, and colorectal cancer (CRC) patients in 11 Midwest hospitals based on insurance status. Methods: We identified patients diagnosed with colorectal, pancreatic, and hepatobiliary cancers within 11 Midwest hospitals from 2011-2021 with North American Association of Central Cancer Registries data. Health insurance was defined as private, Medicaid, Medicare, and uninsured. Effect of insurance status at diagnosis on overall survival (OS) in selected cancers combined and by cancer type were calculated using a Cox proportional hazard (CoxPH) model controlling for age, sex, race, stage, time to treatment initiation (TTI), and Charlson Comorbidity Index (CCI). We evaluated differences in TTI by insurance status using t-tests. We analyzed stage and treatment modality by insurance status using chi-square tests. Post-hoc analysis of TTI by OS used CoxPH. Results: Relative to private insurance, Medicaid and Medicare insurance status significantly reduced OS for all selected GI cancers combined (HR 1.38, 95% CI 1.17 - 1.62 and HR 1.25, 95% CI 1.11 - 1.40, respectively; p&amp;lt;0.001) and for CRC (HR 1.49, 95% CI 1.12 - 1.98 and HR 1.28, 95% CI 1.06 - 1.54, respectively; p&amp;lt;0.01). Uninsured patients had significantly reduced OS only for pancreas cancer (HR 6.36, 95% CI 2.56 - 15.8; p&amp;lt;0.001). No significant associations between insurance status and OS were noted in Hepatobiliary cancer. TTI was significantly associated with insurance status (p&amp;lt;0.001) and mean TTI for Medicaid (35.8 days) and Medicare (27.9 days) patients was longer than mean TTI for privately insured patients (23.3 days; p&amp;lt;0.01). TTI for uninsured patients was not significantly different than for privately insured. Analysis of TTI and OS showed no increased hazard for patients with TTI of 36 days versus 23 days (HR 0.87, 95% CI 0.75 - 1.02; p=0.085). Treatment modality and stage did not differ as a function of insurance status in this dataset. Discussion: Our results expand previous work by demonstrating a survival impact for GI cancer patients with Medicaid or Medicare insurance. This disparity persists despite controlling for known prognostic factors such as age, stage, CCI, and TTI. Notably, TTI varied by insurance suggesting it may account for a portion of this disparity but did not independently drive survival. Further work is required to identify and develop targeted interventions to address other drivers of reduced survival in those without private insurance. Citation Format: Rebecca AbuAyed, Joshua Schwanke, Brooke Patterson, McKenzie White, Schelomo Marmor, Eric Jensen, Christopher Tignanelli, Emil Lou, Ajay Prakash. Real-world analysis of insurance status and survival in patients with selected GI cancers [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 6470.

Reducing Disparities: Regional Anesthesia Blocks for Mastectomy with Reconstruction Within Standardized Regional Anesthesia Pathways.

Recent data suggest disparities in receipt of regional anesthesia prior to breast reconstruction. We aimed to understand factors associated with block receipt for mastectomy with immediate tissue expander (TE) reconstruction in a high-volume ambulatory surgery practice with standardized regional anesthesia pathways. Patients who underwent mastectomy with immediate TE reconstruction from 2017 to 2022 were included. All patients were considered eligible for and were offered preoperative nerve blocks as part of routine anesthesia care. Interpreters were used for non-English speaking patients. Patients who declined a block were compared with those who opted for the procedure. Of 4213 patients who underwent mastectomy with immediate TE reconstruction, 91% accepted and 9% declined a nerve block. On univariate analyses, patients with the lowest rate of block refusal were white, non-Hispanic, English speakers, patients with commercial insurance, and patients undergoing bilateral reconstruction. The rate of block refusal went down from 12 in 2017 to 6% in 2022. Multivariable logistic regression demonstrated that older age (p = 0.011), Hispanic ethnicity (versus non-Hispanic; p = 0.049), Medicaid status (versus commercial insurance; p < 0.001), unilateral surgery (versus bilateral; p = 0.045), and reconstruction in earlier study years (versus 2022; 2017, p < 0.001; 2018, p < 0.001; 2019, p = 0.001; 2020, p = 0.006) were associated with block refusal. An established preoperative regional anesthesia program with blocks offered to all patients undergoing mastectomy with TE reconstruction can result in decreased racial disparities. However, continued differences in age, ethnicity, and insurance status justify future efforts to enhance preoperative educational efforts that address patient hesitancies in these subpopulations.

Sociodemographic Disparities in Preoperative Visual Acuity and Cataract Surgery Utilization in the San Francisco Bay Area.

We examined whether cataract surgery utilization and preoperative visual acuity were associated with patient-specific factors, including ocular findings and comorbidities, general biomedical factors, and/or sociodemographic factors. Retrospective, cross-sectional study. We reviewed the electronic health records of patients from 2012 to 2022 who were examined and followed for at least 2years by an eye care provider at University of California San Francisco Health (UCSF Health)and who had cataract in at least one eye associated with best-corrected visual acuity of 20/25 or worse. Data include ocular factors (e.g., best-corrected visual acuity, lens opacity grade, diagnoses of glaucoma, and/or age-related macular degeneration), biomedical comorbidities, and sociodemographic factors including race/ethnicity, health insurance coverage, primary language spoken, and social vulnerability index. Logistic and multivariate regression analyses. We examined cataract surgery utilization and preoperative best-corrected visual acuity. Compared to White and Asian patients, Black patients had the lowest rates of cataract surgery utilization and the poorest mean preoperative visual acuities, with Hispanic patients following in second place in both categories. However, when the analysis controlled for sociodemographic and biomedical factors, Medicaid insurance and speaking Chinese as a primary language emerged as significant associations. In addition, higher cataract surgery utilization rates were associated with worse preoperative best-corrected visual acuity, a concurrent diagnosis of glaucoma, and a concurrent diagnosis of macular degeneration. Worse preoperative visual acuity was associated with Spanish or Chinese language preference, Medicaid status, and glaucoma diagnosis; poorer preoperative visual acuity was only weakly correlated withincreased social vulnerability. After adjusting for other biomedicalandsociodemographic variables, havingMedicaid insurance and being a non-English speaker were the factorsmost notably associated with reduced cataract surgery utilization and poorer preoperative visual acuity. Health insurance and language barriers, as well as other biomedical and sociodemographic factors, may explain a large proportion of the racial disparities inboth cataract surgery utilization and preoperative visual acuity observed among Black and Hispanic patients. Chinese-speaking patients with limited English proficiency are a vulnerable subgroup that exhibits lower rates of cataract surgery utilization and higher degrees of visual loss prior to undergoingcataract surgery compared to other Asian patients.

Early Correlates of School Readiness Before and During the COVID-19 Pandemic Linking Health and School Data

Many known correlates of kindergarten readiness are captured in developmental and social screenings in primary care; little is known about how primary care data predicts school readiness. To identify early Kindergarten Readiness Assessment (KRA) correlates by linking electronic health record (EHR) data with school district KRA data and to examine potential outcomes of the COVID-19 pandemic using KRA scores between 2018 and 2021. This was a retrospective cohort study linking a large primary care practice (PCP) with school assessment data. Linkage used patient name, date of birth, and address. The setting was an urban school district and PCP affiliated with an academic medical center. Students had a KRA score from fall of 2018, 2019, or 2021 (no 2020 KRA due to the COVID-19 pandemic) and at least 1 prior well-child visit at the PCP. Exposures included year KRA administered, reported child race and ethnicity, child sex, interpreter for medical visits, child ever failed Ages & Stages Questionnaire (ASQ) 18 to 54 months, ever rarely read to, Medicaid status, food insecurity, housing insecurity, problems with benefits, and caregiver depressive symptoms. KRA score (continuous), with a possible range of 0 to 300 (passing score = 270). A total of 3204 PCP patients (mean [SD] age, 67 [4] months; 1612 male [50.3%]; 2642 Black [82.5%]; 94 Hispanic [2.9%]; 244 White [7.6%]) were matched to their KRA score. Mean (SD) KRA scores were significantly lower in 2021 (mean [SD], 260.0 [13.0]; 214 of 998 [21.4%]) compared with 2019 (mean [SD], 262.7 [13.5]; 317 of 1114 [28.5%]) and 2018 (mean [SD], 263.5 [13.6]; 351 of 1092 [32.1%]), a pattern mirrored in the larger school district. In the linear regression final model (n = 2883), the following binary variables significantly lowered the child's KRA score (points lowered [95% CI]) below a mean passing score of 270.8: child ever failed ASQ after 18 months (-6.7; 95% CI, -7.7 to -5.6), Medicaid insured (-5.7; 95% CI, -9.0 to -2.3), Hispanic ethnicity (-3.8; 95% CI, -6.9 to -0.6), requires interpreter (-3.6; 95% CI, -7.1 to -0.1), 2021 year (-3.5; 95% CI, -4.7 to -2.3), male sex (-2.7; 95% CI, -3.7 to -1.8), ever rarely read to (-1.5; 95% CI, -2.6 to -0.4), and food insecurity (-1.2; 95% CI, -2.4 to -0.1). Race, caregiver depression, housing insecurity, and problems receiving benefits were not associated with KRA scores in final model. Findings of this cohort study suggest a deleterious association of the COVID-19 pandemic with early learning and development. There may be potential for PCPs and school districts to collaborate to identify and mitigate risks much earlier.

Abstract WMP104: Influence of Living Arrangements and Within Household Communications on Blood Pressure Control Among Stroke Patients

Introduction: Social support plays a vital role in shaping health behaviors and may impact disease outcomes such as blood pressure (BP) control. We aim to study the association between household living arrangements and BP control for stroke patients, over time and by household communication type. Methods: Discharge Educational Strategies for Reduction of Vascular Events (DESERVE), which aimed to improve post-stroke BP, randomized stroke survivors to skills-based intervention or usual care. Baseline, 6- and 12-month BP was measured according to AHA guidelines with BP control defined as no hypertension present or if follow-up BP was within normal range. We grouped baseline household arrangements as nuclear, multigeneration, spouse, alone, or other and household communication as engaging in conversations about stroke or not. We estimated adjusted odds of BP control by living arrangements and within household communication controlled for age, gender, race/ethnicity, education, Medicaid, and intervention status. Results: Among 540 participants with baseline BP data (72% with hypertension), 361 and 478 had 6- and 12-month data, respectively. Most patients lived in nuclear households (31%), followed by living alone (25%). About 36% participants achieved well-controlled BP after 6 months, increasing to 37% after 12 months. Living in a nuclear arrangement was associated with 2.59 (95% CI: 1.13-5.94) times higher odds of BP control at 6 months. Among participants with multigenerational families, those engaged in conversations about stroke within household showed 5.52 (1.25-24.3) times higher odds of BP control at 12 months. Conclusions: Our findings suggest that living arrangements are positively related to BP control outcomes in stroke patients, and communication with household members on health topics modifies this association. Within household social support to stroke patients may offer an opportunity for targeted secondary stroke prevention strategies.

Outcomes and Predictors of 30-Day Readmission in Patients with Hepatocellular Carcinoma Undergoing Transarterial Chemoembolization between 2016 and 2018

Background: Hepatocellular carcinoma (HCC) is the third leading cause of cancer death worldwide. The 5-year survival rate for liver cancer in the US has improved from 3% four decades ago to 20% now. Transarterial chemoembolization (TACE) is the treatment of choice for stage B/intermediate-stage HCC. Complications of TACE include hepatic encephalopathy, liver failure, post-embolization syndrome, duodenal ulcers, liver abscesses, acute cholecystitis, and injury to the biliary tract. This study evaluates the 30-day readmission rate and predictors of readmission among patients with HCC undergoing TACE. Methods: The 2016–2018 Healthcare Cost and Utilization Project (HCUP) database, which includes the National Readmission Database (NRD), was used. All adult patients with HCC who underwent TACE were identified using the International Classification of Diseases (ICD-10). The rate of 30-day readmissions after TACE and the associated diagnoses were identified. Logistic regression was used to obtain adjusted odds ratios for variables associated with 30-day readmission. Results: A total of 566 patients underwent TACE between 2016–2018. Sixty-five patients were excluded due to death and unavailability of 30-day readmission data. The procedure was performed in large (80.4%), metro-teaching hospitals (94.5%). Mean patient age was 65.1 ± 9.9 years, and 74% of patients were male. Among the 501 patients, 81 (16.2%) were readmitted within 30 days. The mean age for readmitted patients was 63.2 ± 11.0 and 69.1% were male. The mean length of stay at readmission was 5.5 ± 7.3 days. A total of 7.4% of patients had neurological disorders, 17.3% had weight loss, 30.9% had fluid and electrolyte imbalance, and 21.0% had hepatic encephalopathy. The most common primary diagnoses at 30-day readmission were liver cell carcinoma, sepsis, and liver failure. Univariate analysis for variables associated with 30-day readmission included hepatic encephalopathy (OR 3.45; 95% CI 1.8–6.62; p = 0.0002), underlying neurological disorders (OR 3.28; 95% CI 1.16–9.3; p = 0.03), weight loss (OR 2.82; 95% CI 1.42–5.61; p = 0.003), and Medicaid status (OR 1.74; 95% CI 1.05–2.88; p = 0.03). Multivariable analysis showed hepatic encephalopathy (OR 2.91; 95% CI 1.4, 6.04; p = 0.04) and weight loss (OR 2.37; 95% CI 1.13–4.96; p = 0.02) were associated with hospital readmission. Conclusions: Weight loss and hepatic encephalopathy were predictors for 30-day readmission after a TACE procedure for HCC.

Long-term exposure to PM2.5 species and all-cause mortality among Medicare patients using mixtures analyses

BackgroundThe relationship between long-term exposure to PM2.5 and mortality is well-established; however, the role of individual species is less understood. ObjectivesIn this study, we assess the overall effect of long-term exposure to PM2.5 as a mixture of species and identify the most harmful of those species while controlling for the others. MethodsWe looked at changes in mortality among Medicare participants 65 years of age or older from 2000 to 2018 in response to changes in annual levels of 15 PM2.5 components, namely: organic carbon, elemental carbon, nickel, lead, zinc, sulfate, potassium, vanadium, nitrate, silicon, copper, iron, ammonium, calcium, and bromine. Data on exposure were derived from high-resolution, spatio-temporal models which were then aggregated to ZIP code. We used the rate of deaths in each ZIP code per year as the outcome of interest. Covariates included demographic, temperature, socioeconomic, and access-to-care variables. We used a mixtures approach, a weighted quantile sum, to analyze the joint effects of PM2.5 species on mortality. We further looked at the effects of the components when PM2.5 mass levels were at concentrations below 8 μg/m3, and effect modification by sex, race, Medicaid status, and Census division. ResultsWe found that for each decile increase in the levels of the PM2.5 mixture, the rate of all-cause mortality increased by 1.4% (95% CI: 1.3%–1.4%), the rate of cardiovascular mortality increased by 2.1% (95% CI: 2.0%–2.2%), and the rate of respiratory mortality increased by 1.7% (95% CI: 1.5%–1.9%). These effects estimates remained significant and slightly higher when we restricted to lower concentrations. The highest weights for harmful effects were due to organic carbon, nickel, zinc, sulfate, and vanadium. ConclusionsLong-term exposure to PM2.5 species, as a mixture, increased the risk of all-cause, cardiovascular, and respiratory mortality.

Analysis of Uveal Melanoma 5-Year Survival Rates by Medicaid Status: A Nationwide Analysis

ABSTRACT Purpose The survival outcomes of patients with primary uveal melanomas based on Medicaid status have not been previously discussed in the literature. Methods The Surveillance, Epidemiology, and End Results Medicaid database were utilized to identify patients with primary uveal melanomas diagnosed between 2006 and 2013. The Kaplan–Meier method was utilized to construct 5-year survival curves in adult, non-elderly patients. Log-rank testing was used to determine differences in survival rates, and multivariate Cox proportional hazards modeling was utilized to perform adjusted survival analysis. Results A total of 1,765 patients were included (Medicaid: 81, non-Medicaid: 1684). A total of 1683 (95.4%) were White. The average age was 51.75 years (SD = 9.5 years). Medicaid patients were more likely to be unmarried, live in a high poverty neighborhood, and live in a rural area (all p < .001). We observed no significant difference in 5-year survival rates between those enrolled in Medicaid (86.6%, 95% CI: 79.1%1–94.7%) and those not enrolled in Medicaid (85.5, 95% CI: 83.8%–87.2%) (p = .80). After controlling for socioeconomic and clinical factors, Medicaid enrollment was not associated with an increased risk of mortality compared to non-Medicaid enrollment. Age (aHR: 1.04, 95% CI: 1.02–1.06, p < .001) and tumor size >10 mm (aHR: 3.04, 95% CI: 1.49–6.21, p = .002) were associated with an increased risk of mortality. Conclusion Medicaid enrollment was not associated with worse cancer-specific 5-year survival. Further research needs to be elicited to better understand the role of Medicaid enrollment in patients with primary uveal melanoma.

Abstract 13006: Telemedicine Disparities in Ambulatory Cardiology Patients in a Large Academic Health System

Introduction: Although the COVID-19 pandemic prompted rapid expansion of telemedicine, the effect of system-wide digital health strategies on vulnerable populations remains to be understood. Given video visits offer several advantages over audio-only visits, telemedicine encounters are typically preferentially scheduled as video visits; however, access complexities may lead to audio-only encounters and exacerbate the digital divide. Aim: We aimed to identify sociodemographic characteristics associated with telemedicine visit modality in the outpatient cardiology population in a large academic health center. Methods: We assessed all outpatient cardiology telemedicine visits from 1/1/2020-12/31/2021 via the Johns Hopkins Telemedicine Equity Dashboard, which sources its data from the electronic medical record. Our primary outcome was visit modality, with an adverse outcome defined as video-to-audio visit conversion. The first telemedicine encounter for each unique patient was included (accounting for 85% of all telemedicine encounters during study period). We assessed predictors of video-to-audio visit conversion using simple and multivariate logistic regression with relevant pre-determined sociodemographic variables (see Figure ). Results: There were 47,961 total telemedicine encounters, spread between 39,381 unique patients. Odds of conversion from video to audio-only visit increased with age, with the highest odds of conversion to audio-only visit in patients age 75+. Non-White race, Medicaid, Medicare and self-pay insurance status, and higher social deprivation index (SDI) quintile were also associated with increased odds of audio-only modality. Conclusion: Current telemedicine strategies would benefit from further analysis and targeted intervention to develop a more equitable digital care delivery system, particularly for patients that are older, non-White, do not have private insurance, and live in areas with higher SDI.

Linking WIC program and HMO administrative data to study the impact of WIC participation

BackgroundThe Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) serves low-income women and children up to age five in the U.S. While numerous studies document the health impacts of WIC participation, these studies often lack adequate control groups of families eligible for WIC but not participating. This study sought to link a large WIC administrative dataset with electronic health records from a Health Maintenance Organization (HMO) with the goal of supporting rigorous evaluations of the impact of WIC on child health outcomes beyond the age of five. MethodsChildren’s WIC-eligibility status was determined by Medicaid status. Since no common unique identifier existed, we linked WIC and HMO child records using child’s first and last names, date of birth, sex, and address history using exact and fuzzy matching based on sound and the edit distance between two alphanumeric strings. A fuzzy-match score quantified uncertainty. All analysis was conducted by WIC research staff; no identifying information about WIC participants was exchanged. ParticipantsLos Angeles County children seen at an HMO (n = 974,451) between 2008 and 2016 and children enrolled in WIC between 2003 and 2018. Results17.3% of HMO children definitely matched a WIC-participating child, 12% possibly matched, 14.3% were eligible but did not match a WIC-participating child and 56.4% were not WIC eligible. ConclusionsWith appropriate data confidentiality safeguards in place, linking administrative datasets from federal programs with electronic health records is a cost-efficient way to identify a viable comparison group necessary to document the impact of government investments.