Log Link Research Articles

Development of Conflict-Based Safety Performance Function for Safety Evaluation of Urban Roundabouts in India

This study evaluates the safety performance of urban roundabouts in India using conflict-based Safety Performance Functions (SPFs). High-resolution aerial videos were utilized to extract vehicle trajectories, and the Surrogate Safety Assessment Model (SSAM) was employed for conflict identification. Data were collected from 27 roundabouts across various Indian cities, focusing on non-lane-based heterogeneous traffic conditions. Four SPFs were developed utilizing a Generalized Linear Model with Negative Binomial (NB) distribution and log link function for different proximity levels based on Time-to-Collision (TTC) and Post-Encroachment Time (PET) values. Key factors influencing conflict frequency included conflicting traffic volume, occupancy time, ratio of approaching and conflicting traffic volume, percentage of cars and motorcycles, inscribed circle diameter, roundabout symmetry, and number of legs. The developed conflict frequency prediction models fitted well and demonstrated robust predictive performance. This proactive framework for roundabout safety assessment can be followed to enhance road safety in developing nations.

Abstract 4137461: Elevated Lipoprotein(a) Increases Cardiovascular-Related Healthcare Resources Utilization and Costs among US Medicare, Medicaid, and Commercial Enrollees with ASCVD

Background: Elevated lipoprotein(a) [Lp(a)] is a recognized genetic, independent, and causal risk driver for atherosclerotic cardiovascular disease (ASCVD). Less is known about how elevated Lp(a) affects downstream cardiovascular (CV)-related healthcare resource utilization (HCRU) and costs in patients with ASCVD. Objective: To assess the association of Lp(a) levels with CV-related HCRU (hospitalizations, outpatient and emergency department [ED) visits), and CV-related costs for patients with ASCVD in a national-scale multi-insurance cohort. Methods: Patients with ASCVD were pooled from US Medicare, commercial, and Medicaid health plans from 2017 to 2022. Lp(a) levels (mg/dL and nmol/L) were linked at the patient level with HealthVerity Lab Data. Elevated Lp(a) was defined as Lp(a)≥70mg/dL (≥ 175 nmol/L) and normal Lp(a) was defined as Lp(a)<30mg/dL (<75 nmol/L). Generalized linear models with log link negative binomial distribution (HCRU) and log link gamma distribution (costs) adjusted for age, sex, race, insurance type, comorbidities, statin use, and prior hospitalizations, measured the association of Lp(a) levels with the annualized rate of CV-related hospitalizations, ED visits, and outpatient visits (incidence rate ratio [IRR]) and CV-related costs (annualized cost ratio). Results: A total of 180,240 ASCVD patients were identified with Lp(a) values. Median (IQR) Lp(a) was 14 mg/dL (4 mg/dL – 48 mg/dL) or 35 nmol/L (12 nmol/L – 114 nmol/L). Mean (SD) age was 68.6 (12.0) years with 49.9% being female; 72.8% of patients were from Medicare, followed by commercial (21.8%) and Medicaid (5.4%). Fully adjusted CV-related HCRU was significantly higher in those with elevated Lp(a) compared to normal Lp(a): inpatient hospitalizations (IRR=1.11), outpatient visits (IRR=1.11) and ED visits (IRR=1.06), all p <0.001 (IRR and 95% CI reported in Table 1). Patients with elevated Lp(a) had 14% higher CV-related annualized healthcare costs than patients with normal Lp(a) (p <0.001) with significantly higher CV-related HCRU beginning with Lp(a) levels >30 mg/dL (>75 nmol/L). Conclusion: Elevated Lp(a) was significantly associated with greater CV-related HCRU and costs in patients with ASCVD.

Abstract 4136509: Children in Foster Care with Heart Disease Have Greater Resource Utilization and Mortality

Background: While children and youth in foster care (CYFC) have greater healthcare needs generally, data are lacking in heart disease. We characterize CYFC with heart disease, assess the impacts of being in foster care on outcomes, and examine validity of ICD-10 code Z62.21 for identification of CYFC in administrative data. Methods: In this single-center, retrospective cohort study of all cardiac surgical patients (<18yo) admitted 2017-2022, CYFC were identified using institutional Electronic Medical Records via natural language processing or ICD codes, with confirmation by chart review. Demographics, medical characteristics and outcomes of CYFC were compared to children not in foster care. Generalized linear models assessed associations with primary outcomes of mortality at current follow-up (logit link) and total hospital days (log link), controlling for clinical factors and length of follow-up. We separately added Child Opportunity Index (COI) and payer to models. This cohort was then studied in the administrative pediatric health information system (PHIS) dataset to calculate ICD code sensitivity and specificity. Results: Of 4,431 cardiac surgical patients, 67 (1.5%) were in foster care. CYFC did not have significantly higher mortality (9% vs 5.8%), 30-day readmission or cardiac reoperation, number of complex chronic conditions (CCC), RACHS-2, or distance from center compared to non-CYFC. CYFC had lower COI (median 61 vs 72, p=0.016) and more Medicaid coverage (87% vs 40%, p<0.001). Length of stay after index surgery was similar; however, cumulative hospital days during the study were significantly higher for CYFC (median 22 [IQR 10,45] vs 12 [6,27], p<0.001). Adjusting for RACHS-2, CCC, age, readmission, reoperation, and follow-up duration, CYFC had higher odds of mortality (2.7 [95%CI 1.0,6.9] p=0.046) and more total hospital days (p=0.008). With COI or payer in the model, CYFC was not associated with mortality but remained an independent risk factor for more total hospital days. Only 16/67 (23.9%) of CYFC were coded with the Z62.21 in PHIS, yielding sensitivity of 21.9% (95%CI 13.1,33.1), and specificity of 100% (99.9,100). Conclusion: CYFC with heart disease have a relative social disadvantage, greater hospital resource utilization, and long-term mortality despite similar illness severity. Improved capture of foster care status in administrative data is needed to guide future studies and reduce health disparities for the vulnerable population of CYFC.

The role of the construction industry in the recovery of processing enterprises during the crisis

The article discusses the consequences of the crisis period for the work of processing enterprises in Ukraine. It is established that hostilities on the territory of our state and general instability lead to a decrease in production and force processing enterprises to cease their activities. Many enterprises are experiencing serious financial difficulties due to work restrictions, problems with the supply of raw materials, and the loss of industrial facilities. In the summer of 2022, the percentage of companies that closed or were in the process of ceasing operations was 46.8 %. A significant consequence of the war for the Ukrainian economy at the moment is the rapid changes in the territorial location and the change in the symmetry of the processing industry of Ukraine. The war also affected the sectoral structure of the processing industry, producing significant shifts and changes in the volume of products sold. The current political situation in Ukraine and the severance of business chains with the temporarily occupied territories require processing enterprises to search for new markets for products, sources of supply of raw materials, change logistics links, increase the production of raw materials and their own products. During the study, it was found that certain key aspects of the impact on the activities of processing enterprises were previously insufficiently studied. It is determined that the construction industry is of critical importance for the processing industry of the State during the crisis, which is designed to contribute to the recovery of this industry: to develop infrastructure, to build and modernize production facilities, which allows to introduce new technologies and optimize production processes, to improve the conditions of production itself, to build energy facilities to provide processing enterprises with stable and affordable energy supply, to apply modern construction technologies that will create environmentally friendly and energy-efficient production. For the restoration and development of the processing industry of Ukraine, taking into account the impact of the construction industry, we propose to apply the following measures: widespread implementation of the enterprise relocation program with the expansion of state support; the use of state programs for the reconstruction of property and lending, the development of international communications and the European vector for the recovery of the post-war industry based on the ESG concept and the principles of building back better, compliance with the principles of energy efficiency and energy saving, and the application of decarbonization measures. These tools are designed to provide stimulating processes in the processing industry to achieve and exceed the prewar level of the volume of competitive and high-quality products. The development of the construction industry plays an important role in the restoration of processing enterprises in Ukraine, so it is important to create favorable conditions for their cooperation, to develop comprehensive development strategies aimed at ensuring sustainable economic growth and support.

Decompositions of Information Criteria in Joint Model Under Probit Link for Response

ABSTRACTThe item response and response time are naturally encountered in variety of computerized testing, which are jointly analysed to achieve efficient model fitting. Several novel Bayesian assessment criteria, which quantify the enhancement in the fit of one dataset given the other, are proposed by utilizing decompositions of the deviance information criterion (DIC) and the logarithm of the pseudo‐marginal likelihood (LPML) under a logistic link. However, the calculation of these criteria under a probit link, a commonly employed alternative, remains unexplored. Moreover, decomposing DIC and LPML under probit link presents greater challenges than in the logistic model due to the intricate additional dimension integral. This paper addresses this gap by demonstrating the application of Bayesian model assessment criteria with a normal ogive model in the context of item response data within a joint modelling framework. Additionally, we also illustrate how to implement Bayesian sampling via a powerful R package ‘nimble’. Extensive simulation studies are conducted to evaluate the empirical performance of those criteria. Further, a detailed analysis of the Programme for International Student Assessment (PISA) data is carried out to illustrate the methodology.

Fetal growth ultrasound in obese patients for the detection of growth abnormalities.

To examine the impact of maternal obesity on fetal growth abnormalities including fetal growth restriction (FGR) and large for gestational age (LGA) fetuses. Secondary analysis from the Nulliparous Pregnancy Outcomes Study: Monitoring Mothers-to-be (nuMoM2b). The study excluded individuals with pregestational or gestational diabetes, chronic hypertension, and other major maternal medical conditions. Ultrasound assessments were performed at 16 - 21 and 22 - 29 weeks of gestation. Our exposure was the presence of pre-pregnancy obesity. Our primary outcome was rates of fetal growth abnormalities identified by ultrasound, defined as FGR (estimated fetal weight [EFW] or abdominal circumference < 10th percentile) or LGA (EFW > 90 th percentile) among obese compared to non-obese women. A secondary analysis was performed after limiting ultrasound performed from 28-29 weeks. To estimate adjusted relative risks (aRR) with 95% confidence intervals (95%CIs), we used generalized linear models with Poisson distribution and log link using robust error variance, adjusting for the predefined covariates. Of 7,354 participants, 1,443 (19.6%) had pre-pregnancy obesity while 5,911 (80.4%) did not. Pre-pregnancy obesity compared with normal weight was associated with an increased risk of fetal growth abnormalities both at 16-21 weeks (16.0% vs. 13.2%; aRR 1.23; 95%CI 1.06-1.42) and 22-29 weeks (16.0% vs. 12.1%; aRR 1.33; 95%CI 1.14-1.54). Furthermore, pre-pregnancy obesity compared with normal weight was associated with an increased risk of LGA both at 16-21 weeks (12.5% vs. 10.3%; aRR 1.24; 95%CI 1.05-1.47) and 22-29 weeks (10.6% vs. 6.9%; aRR 1.66; 95%CI 1.38-2.01). In a secondary analysis limited to the ultrasound 28-29 weeks, both fetal growth abnormalities and LGA were associated with the presence of obesity. In any of the analyses, pre-pregnancy obesity was not associated with FGR compared to normal weight. Maternal obesity is associated with an increased risk of fetal growth abnormalities and LGA fetuses.

Familial coaggregation and shared genetic influence between major depressive disorder and gynecological diseases.

The mechanism underlying the co-occurrence of major depressive disorder (MDD) and gynecological diseases remains unclear. This study aimed to investigate the familial co-aggregation and shared genetic loading between MDD and gynecological diseases, namely dysmenorrhea, endometriosis, uterine leiomyomas (UL), and polycystic ovary syndrome (PCOS). Overall, 2,121,632 females born 1970-1999 with parental information were enrolled from the Taiwan National Health Insurance Research Database (NHIRD); 25,142 same-sex twins and 951,779 persons with full-sibling(s) were selected. Genome-wide genotyping data were available for 67,882 unrelated female participants from the Taiwan Biobank linked to the NHIRD. A generalized linear model with a logistic link function was used to examine the associations of individual history, family history in parents/full-siblings/same-sex twins, and polygenic risk scores (PRS) for MDD with the risk of gynecological diseases; generalized estimating equations were used to consider the non-independence of data. Both parents affected with MDD was associated with four gynecological diseases, and its magnitude of association was higher than either affected parent; maternal MDD showed a higher magnitude of association than paternal MDD. Full-siblings of patients with MDD had a higher risk of four gynecological diseases; same-sex twins of patients with MDD had a greater association with dysmenorrhea and PCOS. PRS for MDD was associated with dysmenorrhea and endometriosis. Familial co-aggregation was observed in the co-occurrence of MDD and four gynecological diseases. There exists a shared polygenic liability between MDD and dysmenorrhea and endometriosis. Individuals with MDD-affected relatives or a higher PRS for MDD should be monitored for gynecological diseases.

Real-World Assessment of Economic and Clinical Outcomes in Thai Patients With Respiratory Syncytial Virus Infection Across Age Groups: A Retrospective Cohort Analysis.

Respiratory syncytial virus (RSV) is an important cause of acute lower respiratory infections worldwide, including Thailand. This study aimed to assess clinical and economic burdens of RSV infections across different age groups in Thailand. A retrospective cohort study was conducted using data from a tertiary care hospital from 2014 to 2021. Patients who tested at least one positive RSV were included and stratified into five age groups (< 2, 2-5, 5-18, 18-65, and > 65 years). Healthcare resource utilization, direct medical costs, and clinical outcomes were analyzed with descriptive statistics. Generalized linear models with gamma distributions and log link were used to model cost outcomes. Costs were reported in 2021 US dollars (USD), with 1 USD = 31.98 Thai Baht. A total of 2122 RSV-positive patients were identified, half of which (1097) were hospitalized. The median (interquartile range [IQR]) total hospitalization costs ranged from USD780 (IQR: USD488-USD1185) in those < 2 years to USD2231 (IQR: USD1250-USD4989) in those aged 65+ years. Case fatality rates among hospitalized patients also varied from 2.5% to 28.4% depending on age. Increased age, presence of comorbidities, and need for critical care were associated with higher hospitalization costs. Among RSV-positive patients, younger children experienced the greatest burden, but poorer outcomes were observed in older adults. Higher costs were associated with older age, comorbidities and critical care needs. Understanding RSV economic burdens is crucial for assessing the cost-effectiveness and public health value of vaccination programs that prioritize at-risk groups to mitigate the public health impact.

Impact of the COVID-19 pandemic on mortality and health services utilization costs of patients diagnosed with lung cancer

IntroductionHealthcare service disruptions due to the COVID-19 pandemic may have caused worse health outcomes and resulted in more expensive treatments for patients diagnosed with lung cancer in Alberta, Canada. MethodsA population-based retrospective cohort design was used to compare 1-year survival, mortality, and health services utilization costs of patients diagnosed with lung cancer pre- (March 17th, 2018 to March 16th, 2019), intra- (March 17th, 2020 to March 16th, 2021), and post-pandemic (March 17th, 2021 to March 16th, 2022). Kaplan-Meier curves and Cox regressions were used for estimating survival and hazard ratios. General linear models with gamma family and log link were used for estimating health services utilization costs. All costs were converted to 2024 Canadian dollars (CA$1~US$0.74). ResultsIn total, 2332, 2271, and 2408 individual patients were diagnosed with lung cancer in the pre-, intra-, and post-COVID, respectively. The survival at 365 days after diagnosis was 50%, 49%, and 51% for patients diagnosed with lung cancer pre-, intra-, and post-pandemic, respectively. Multivariate Cox regressions showed that patients diagnosed intra- and post-pandemic had a significantly (15% and 10%) higher probability of death at 1 year after diagnosis as compared to those diagnosed pre-pandemic (intra- vs. pre- HR=1.15, p=0.001 and post- vs. pre- HR=1.10, p=0.023). Patients diagnosed with lung cancer intra-pandemic had the highest health services utilization cost ($59.000) per patient per year, followed by post-pandemic ($55,510) and pre-pandemic ($51,640). Compared to pre-pandemic, the health services utilization cost intra-pandemic was 15.3% ($7,859) higher and post-pandemic was 7.5% ($3,887) higher. Although significantly higher than pre-, post-pandemic patients had a lower hazard ratio and health services utilization cost compared to intra-pandemic. ConclusionsLung cancer patients diagnosed during COVID-19 pandemic had significantly increased mortality and health services utilization costs compared to pre-pandemic, however, these impacts improved right post-pandemic.

Clinical characteristics and influencing factors of serious fall injuries among older inpatients: A secondary analysis of multicenter cross-sectional administrative data

BackgroundInpatient falls are common adverse events in hospitals, particularly among older adults. However, current research on falls in the older population is limited by small sample sizes and reliance on single-center designs without distinguishing the different levels of fall injuries. ObjectiveTo examine the circumstances of falls in older inpatients and investigate the factors influencing serious fall injuries. DesignA secondary analysis of multicenter cross-sectional administrative data from the China National Database of Nursing Quality (CNDNQ). Setting(s)67 hospitals in Hunan, China. Participants5143 fall incidents from 5048 older inpatients were included from January 2019 to December 2022. MethodsDescriptive statistics (means, standard deviations, frequencies, and percentages) were used to describe fall characteristics. Given the multilevel data structure, a Generalized Linear Mixed Model (GLMM) with a Logistic link function was employed, treating hospitals and wards as random effects. Univariate and multivariate analyses were conducted. ResultsThe prevalences of fall-related injuries and serious fall injuries in this study were 63.3 % and 25.0 %, respectively. Nurse-to-patient ratio at the time of fall ≤0.04 (aOR = 1.522, 95 % CI: 1.297–1.785), female gender (aOR = 1.304, 95 % CI: 1.137–1.497), a lengthy hospital stay (aOR = 1.009, 95 % CI: 1.001–1.018) were associated with an increased risk of serious fall injuries. Surgical (aOR = 0.655, 95 % CI: 0.482–0.889) and orthopedic wards (aOR = 0.649, 95 % CI: 0.449–0.939) exhibited lower risks of serious fall injuries than internal medicine wards. The time period from 8:00 to 15:59 was associated with a higher risk of serious fall injuries compared to the period from 0:00 to 7:59 (aOR = 1.416, 95 % CI: 1.171–1.713). Compared to lying, standing (aOR = 2.594, 95 % CI: 1.733–3.883), getting in/out of bed (aOR = 1.403, 95 % CI: 1.009–1.951), walking or rehabilitating (aOR = 3.039, 95 % CI: 2.194–4.208), toileting (aOR = 1.411, 95 % CI: 1.038–1.917), and showering (aOR = 2.170, 95 % CI: 1.391–3.385) were associated with an increased risk of serious fall injuries. ConclusionsFall-related injuries and serious fall injuries are highly prevalent among older inpatients. The study provides insights for targeted fall prevention strategies, improving nursing management and patient safety. RegistrationNot registered.

Long COVID Is Associated with Decreased Quality of Life and Increased Mental Disability

COVID-19’s impact on quality of life (QoL) has been studied; however, data on the effects of long COVID on QoL and mental disability remain limited. This study aimed to investigate whether long COVID is associated with decreased QoL and increased mental disability. We analyzed data from adults (≥18 years) in the 2022 Medical Expenditure Panel Survey (MEPS). Physical and mental QoL were assessed using the Veterans RAND 12-Item Health Survey, while mental disability was measured with the Kessler Index. Long COVID was defined as experiencing COVID-related symptoms for over 3 months. To account for skewness in the outcome variables, the study employed survey-weighted generalized linear regression with a log link and gamma distribution. The analysis included 153,330,408 subjects (unweighted n = 10,975). The average age of those with and without long COVID was similar (51 vs. 52 years, p = 0.46). Subjects with long COVID had a 3% lower physical QoL (β = −0.03, p = 0.02), a 4% lower mental QoL (β = −0.04, p &lt; 0.01), and a 26% higher mental disability score (β = 0.26, p &lt; 0.01) compared to those without long COVID. These findings highlight the urgent need for targeted strategies to address long COVID and support patients in recovering their pre-COVID QoL and mental health.

Matrix Factorization and Prediction for High-Dimensional Co-Occurrence Count Data via Shared Parameter Alternating Zero Inflated Gamma Model

High-dimensional sparse matrix data frequently arise in various applications. A notable example is the weighted word–word co-occurrence count data, which summarizes the weighted frequency of word pairs appearing within the same context window. This type of data typically contains highly skewed non-negative values with an abundance of zeros. Another example is the co-occurrence of item–item or user–item pairs in e-commerce, which also generates high-dimensional data. The objective is to utilize these data to predict the relevance between items or users. In this paper, we assume that items or users can be represented by unknown dense vectors. The model treats the co-occurrence counts as arising from zero-inflated Gamma random variables and employs cosine similarity between the unknown vectors to summarize item–item relevance. The unknown values are estimated using the shared parameter alternating zero-inflated Gamma regression models (SA-ZIG). Both canonical link and log link models are considered. Two parameter updating schemes are proposed, along with an algorithm to estimate the unknown parameters. Convergence analysis is presented analytically. Numerical studies demonstrate that the SA-ZIG using Fisher scoring without learning rate adjustment may fail to find the maximum likelihood estimate. However, the SA-ZIG with learning rate adjustment performs satisfactorily in our simulation studies.

Racial and ethnic disparities in COVID-19 hospital cost of care.

Although studies have evaluated the hospital cost of care associated with treating patients with COVID-19, there are no studies that compare the hospital cost of care among racial and ethnic groups based on detailed cost accounting data. The aims of this study were to provide a detailed description of the hospital costs of COVID-19 based on individual resources during the hospital stay and standardized costs that do not rely on inflation adjustment and evaluate the extent to which hospital total cost of care for patients with COVID-19 differs by race and ethnicity. This study used electronic medical record data from an urban academic medical center in Chicago, Illinois USA. Hospital cost of care was calculated using accounting data representing the cost of the resources used to the hospital (i.e., cost to the hospital, not payments). A multivariable generalized linear model with a log link function and inverse gaussian distribution family was used to calculate the average marginal effect (AME) for Black, White, and Hispanic patients. A second regression model further compared Hispanic patients by preferred language (English versus Spanish). In our sample of 1,853 patients, the average adjusted cost of care was significantly lower for Black compared to White patients (AME = -$5,606; 95% confidence interval (CI), -$10,711 to -$501), and Hispanic patients had higher cost of care compared to White patients (AME = $8,539, 95% CI, $3,963 to $13,115). In addition, Hispanic patients who preferred Spanish had significantly higher cost than Hispanic patients who preferred English (AME = $11,866; 95% CI $5,302 to $18,431). Total cost of care takes into account both the intensity of the treatment as well as the duration of the hospital stay. Thus, policy makers and health systems can use cost of care as a proxy for severity, especially when looking at the disparities among different race and ethnicity groups.

Costs and Healthcare Utilization of Heart Disease by COVID-19 Diagnosis and Race and Ethnicity

Introduction: Heart disease (HD) poses a significant health and economic burden in the United States, with considerable variations in outcomes across different racial and ethnic groups. The COVID-19 pandemic has further highlighted disparities in healthcare utilization and costs associated with HD. ac Methods: We used the 2021 Merative MarketScan Medicaid claims database to estimate medical costs and healthcare utilization associated with HD by racial and ethnic groups and COVID-19 diagnosis status. Our study focused on individuals aged ≥18 years old continuously enrolled in a non-capitated insurance plan in 2021. The outcome measures included total medical expenditures and healthcare utilization, including numbers of emergency department (ED) visits and inpatient admissions, and length of inpatient stay. We employed a generalized linear model with a family of gamma and log links was used for medical costs, and a negative binomial regression for healthcare utilization. Three-way interactions of HD, COVID-19 diagnosis, and race and ethnicity categories were implemented after adjusting for age, sex, and comorbidities. We reported average marginal effects with 95% CIs. Results: Among 1,008,166 Medicaid beneficiaries, 8% had HD in 2021. The cost associated with HD was $10,819 per beneficiary in 2021 (95% CI, 10,292–11,347; P<0.001). The cost was $15,840 (95% CI, 14,389–17,291; P<0.001) for non-Hispanic Black individuals, $9,945 (95% CI, 9,172–10,718; P<0.001) for non-Hispanic White, and $8,511 (95% CI, 7,490–9,531; P<0.001) for Hispanic individuals. Individuals with a COVID-19 diagnosis ($19,638) had $9,541 (95% CI, 7,049–12,032; P<0.001) higher costs associated with HD than those without COVID-19 ($10,098) (P<0.001). Individuals with HD had higher numbers of ED visits (0.937 per beneficiary, 95% CI, 0.913–0.960), inpatient admissions (0.463 per beneficiary, 95% CI, 0.455–0.471), and average length of stay (2.541 days per admission, 95% CI, 2.405–2.677) than those without HD. Conclusions: Our findings showed that costs and healthcare utilization associated with HD are substantial in all racial and ethnic groups and the highest among non-Hispanic Black individuals. Furthermore, individuals with a COVID-19 diagnosis had approximately two times higher costs associated with heart disease compared with individuals without a COVID-19 diagnosis.

Comparing direct medical care costs of patients with bladder cancer who received an ileal conduit vs. neobladder in the year following cystectomy

PurposeBladder cancer is 1 of the most costly cancers, however there is limited research on medical care costs by type of urinary diversion. The objective of our study was to compare medical care costs of the 2 most common urinary diversions in the year following radical cystectomy. MethodsThe Bladder Cancer Quality of Life Study included patients diagnosed with bladder cancer who underwent radical cystectomy and received an ileal conduit (IC, n = 821) or neobladder (NB, n = 181) in 3 integrated health systems. Medical care costs per patient per quarter were estimated for the year following cystectomy. Multivariable generalized linear models with a gamma distribution and log link were used to estimate mean monthly medical care costs (2022 USD$), adjusted for patient demographic and clinical characteristics. ResultsIn multivariable analysis, mean monthly costs per quarter were not significantly different between IC and NB patients in the 12 months following cystectomy. Overall, mean monthly costs in IC and NB patients were highest during the first quarter and decreased thereafter. Factors associated with higher mean costs across all quarters included presence of any complications and advanced tumor stage at cystectomy (all P < 0.001). ConclusionOur study addresses an important knowledge gap by quantifying the medical costs of bladder cancer patients by urinary diversion type and comparing costs of different treatment approaches. Studies that assess patient-reported outcomes and out-of-pocket costs, by urinary diversion type, are warranted to inform treatment decision-making and cost conversations.

Health care contact days in older adults enrolled in SWOG cancer clinical trials.

64 Background: Health care contact days—days with health care contact outside the home—are a measure of how much of a patient’s life is consumed by health care. This number of contact days are especially relevant for older adults with advanced cancer facing limited survival. One concern is that patients with different sociodemographic and clinical backgrounds may be at particular risk of increased contact days. Thus, there is a critical need to understand factors associated with contact days. Methods: We linked data from 6 SWOG trials in advanced cancer to Medicare claims data. These trials encompassed a variety of primary cancer types [prostate (S9916, S0421), lung (S0003, S0124, S0819), and pancreas (S0205)]. Health care contact days (days with ambulatory, emergency department, inpatient, or facility-based care) in the first 12 months were calculated using Medicare database files. A day with &gt;1 source of contact counted as 1. We examined sociodemographic (age, race, ethnicity, sex, insurance status, rurality, area deprivation index) and clinical (prognostic risk, cancer type, performance status) predictors of contact days, using negative binomial regression, with a log link and clustering by study ID, including an offset variable for duration of observation. We examined predictors in univariate and multivariable analysis. We report relative risk (RRs) estimates and 95% CI. Results: We included 1,429 patients (median age, 71 years, 7.6% Black, 2.7% Hispanic, 21.4% female, and 23.5% rural residence). Of these, 741 (51.9%) patients died in the first year. The median number of contact days was 48, out of a median of 350 days of observation. At a trial level, median percentage of contact days ranged from 14% (prostate) to 27% (lung). The most common sources of contact days were ambulatory clinician visits (median=17), tests (median=12), and treatments (median=11). In multivariable regression, factors associated with increased contact days included increasing age (per year, RR=1.02 [1.01-1.02], p&lt;.001), insurance status (Medicare alone or with Private or Medicaid insurance, vs Military/VA, none, or other insurance, RR=2.47 [2.16-2.83], p&lt;.001), higher prognostic risk score (RR=1.14 [1.04-1.25], p=.004), and type of cancer (pancreas, RR=1.69 [1.51-1.89] and lung, RR=1.69 [1.54-1.85] vs prostate, p&lt;.001 for both). Conclusions: This novel linkage of trial and claims data to evaluate contact days and associated factors found that older trial participants spent up to one in four days with health care contact. Ambulatory appointments were the most common source of contact days. Encouragingly, rural residence and neighborhood deprivation were not associated with contact days, increasing confidence that participants from diverse geographic areas may receive equitable care in trials. These data highlight the frequency of contact days for trial participants, while identifying factors to consider when benchmarking contact days.

Association of serum glucose/potassium index levels with poor long-term prognosis in patients with Aneurysmal Subarachnoid Hemorrhage

IntroductionThe glucose/potassium index (GPI) has been reported as a predictor biomarker of in-hospital complications in patients with aneurysmal subarachnoid hemorrhage (aSAH). ObjectivesTo determine the association between GPI and functional outcome at 3–6 months after discharge in patients diagnosed with aSAH in a Peruvian referral hospital during 2018–2021. Materials and MethodsWe conducted a retrospective cohort observational study with a secondary database in patients with aSAH during 2018–2021 in a Peruvian referral hospital. We evaluated the relationship between GPI values and motor functionality from 3 m to 6 months post-discharge, using a Poisson family generalized linear model with Log link function and robust variance according to categorization of good and poor outcome. We considered a value of p<0.05 as statistically significant. We used restricted cubic splines with five nodes to evaluate the linear correlation between the 2 main variables. Results212 patients were included in the analysis. 21.1 % and 19.3 % had poor outcome at 3 and 6 months after discharge, respectively. Multivariate analysis showed that GPI was not associated with poor outcome at 3 (RR=0.999; 95 %CI=0.979–1.018) or 6 months after discharge (RR=0.979; 95 %CI=0.979–1.020). On the other hand, Splines plots showed no correlation between GPI and modified Rankin. ConclusionsDespite the usefulness of GPI to prognosticate in-hospital complications, its association with functional outcome is inconclusive.

Association of social isolation and depressive symptoms with workplace productivity loss in a multi-ethnic Asian study

The association of health-related productivity loss (HRPL) with social isolation and depressive symptoms is not well studied. We aimed to examine the association of social isolation and depressive symptoms with productivity loss. Data on employed adults aged 21 years and above were derived from the Population Health Index (PHI) study conducted by the National Healthcare Group (NHG) on community-dwelling adults, residing in the Central and Northern residential areas of Singapore. The severity of depressive symptoms and social isolation were assessed using the 9-item Patient Health Questionnaire (PHQ-9) and Lubben Social Network Scale-6 (LSNS-6) respectively. Productivity loss was assessed using the Work Productivity and Activity Impairment Questionnaire (WPAI). We used Generalised Linear Models, with family gamma, log link for the analysis. Models were adjusted for socio-demographic variables (including age, gender, ethnicity, employment status, housing type) and self-reported chronic conditions (including the presence of diabetes, hypertension, and dyslipidemia). There were 2,605 working (2,143 full-time) adults in this study. The median reported percentage of unadjusted productivity loss was 0.0%, 10.0% and 20.0% for participants with social isolation, depressive symptoms, and both, respectively. In the regression analysis, mean productivity loss scores were 2.81 times (95% Confidence Interval: 2.12, 3.72) higher in participants with depressive symptoms than those without. On the other hand, social isolation was not found to be associated with productivity loss scores (1.17, 95% Confidence Interval: 0.96, 1.42). The interaction term of depressive symptoms with social isolation was statistically significant, with an effect size of 1.89 (95% Confidence Interval: 1.04, 3.44). It appeared that productivity loss was amplified when social isolation and depressive symptoms were concomitant. Our results suggested significant associations of social isolation and depressive symptoms with productivity loss. These findings highlighted the potential impact of social isolation and depressive symptoms on work performance and drew attention to the importance of having a holistic work support system that promotes social connectedness, mental wellbeing and work productivity.

The effectiveness of COVID-19 vaccination in preventing hospitalisation and mortality: A nationwide cross-sectional study in Iran.

The pandemic of the coronavirus disease 2019 (COVID-19) led to a global health crisis, prompting widespread vaccination efforts to reduce severe outcomes. In this study, we assessed the impact of mass COVID-19 vaccination on hospitalisation and mortality rates in Iran, where over 83% of the vaccinated population received inactivated virus vaccines. Using retrospective, cross-sectional analysis, we examined data from the Iran Health Insurance Organisation, covering 41 million individuals from 20 February 2020 to 20 March 2022. We analysed hospital records from 956 Iranian hospitals, focusing on inpatient stays, short-term hospitalisations, and emergency department visits. Study outcomes included COVID-19 hospital admissions and associated mortality. We used negative binomial regression to compare hospital admission rates between periods, while we used a poison regression model with a log link to assess mortality risks before and after vaccination. Among 806 076 hospital admissions, 57 599 deaths were recorded. COVID-19 hospitalisations increased with age, and women had slightly higher admission rates than men. Advanced age and male sex correlated with higher mortality rates. Hospital admissions rose to 1178.66 per million population per month post-vaccination compared to 459.78 pre-vaccination. The incidence rate ratio was 2.09 (95% confidence interval (CI) = 1.90-2.32, P < 0.001), mainly due to the Delta variant. In contrast, post-vaccination mortality rates decreased from 111.33 to 51.66 per 1000 admissions per month. Post-vaccination, COVID-19 mortality significantly decreased, with a relative risk being 0.61 (95% CI = 0.60-0.62, P < 0.001) across all age groups and sexes. The Delta variant increased hospital admissions among vaccinated individuals, but widespread vaccination significantly reduced COVID-19-related mortality.

Real world analysis of treatment change and response in adults with attention-deficit/hyperactivity disorder (ADHD) alone and with concomitant psychiatric comorbidities: results from an electronic health record database study is the United States

BackgroundThe objectives of this study were to examine the association of psychiatric comorbidities and patient characteristics with treatment change and response as well as to assess the association between treatment change and healthcare resource utilization (HCRU) among adult patients with attention-deficit/hyperactivity disorder (ADHD) and psychiatric comorbidities.MethodsDe-identified electronic health records from the NeuroBlu Database (2002–2021) were used to select patients ≥ 18 years with ADHD who were prescribed ADHD-specific medication. The index date was set as the first prescription of ADHD medication. The outcomes were treatment change (discontinuation, switch, add-on, or drop) and HCRU (inpatient, outpatient, composite) within 12 months of follow-up. Cox proportional-hazard model was used to assess the association between clinical and demographic patient characteristics and treatment change, while generalized linear model with negative binomial distribution and log link function was used to assess the association between key risk factors linked to treatment change and HCRU rates.ResultsA total of 3,387 patients with ADHD were included (ADHD only: 1,261; ADHD + major depressive disorder (MDD): 755; ADHD + anxiety disorder: 467; ADHD + mood disorder: 164). Nearly half (44.8%) of the study cohort experienced a treatment change within the 12-month follow-up period. Treatment switch and add-on were more common in patients with ADHD and comorbid MDD and anxiety disorder (switch: 18.9%; add-on: 20.5%) compared to other cohorts (range for switch: 8.5–13.6%; range for add-on: 8.9–12.1%) Survival analysis demonstrated that the probability of treatment change within 12 months from treatment initiation in the study cohort was estimated to be 42.4%. Outpatient visit rates statistically significantly increased from baseline (mean [SD] 1.03 [1.84] visits/month) to 3 months post-index (mean [SD] 1.62 [1.91] visits/month; p < 0.001), followed by a gradual decline up to 12 months post-index. Being prescribed both a stimulant and a non-stimulant at index date was statistically significantly associated with increased risk of treatment change (adjusted hazard ratio: 1.64; 95% CI: 1.13, 2.38; p = 0.01).ConclusionsThis real-world study found that treatment change was common among patients with ADHD and psychiatric comorbidities. These findings support the need for future studies to examine the unmet medical and treatment needs of this complex patient population.