Hospital costs of Balloon Pulmonary Angioplasty (BPA) procedure and management for CTEPH patients: An observational study based on the French national hospital discharge database (PMSI)

Introduction Since 2014, Balloon Pulmonary Angioplasty (BPA) has become an emerging and complementary strategy for chronic thromboembolic hypertension (CTEPH) patients who are not suitable for pulmonary endarterectomy (PEA) or who have recurrent symptoms after the PEA procedure. Objective To assess the hospital cost of BPA sessions and management in CTEPH patients. Methods An observational retrospective cohort study of CTEPH-adults hospitalized for a BPA between January 1st, 2014 and June 30th, 2016 was conducted in the 2 centres performing BPA in France (Paris Sud and Grenoble) using the French national hospital discharge database (PMSI-MCO). Patients were followed until 6 months or death, whichever occurred first. Follow-up stays were classified as stays with BPA sessions, for BPA management or for CTEPH management based on a pre-defined algorithm and a medical review using type of diagnosis (ICD-10), delay from last BPA procedure stay and length of stay. Hospital costs (including medical transports) were estimated from National Health Insurance perspective using published official French tariffs from 2014 to 2016 and expressed in 2017 Euros. Results A total of 191 patients were analysed; mainly male (53%), with a mean age of 64,3 years. The first BPA session was performed 1.1 years in median (IQR 0.3–2.92) after the first PH hospitalisation. A mean of 3 stays with BPA sessions per patient were reported with a mean length of stay of 8 days for the first stay and 6 days for successive stays. The total hospital cost attributable to BPA was € 4,057,825 corresponding to €8,764±3,435 per stay and €21,245±12,843 per patient. Results were sensitive to age classes, density of commune of residence and some comorbidities. Conclusions The study generated robust real-world data to assess the hospital cost of BPA sessions and management in CTEPH patients within its first years of implementation in France.

Sex-Related Disparities in the Incidence and Outcomes of Community-Acquired Pneumonia among Type 2 Diabetes Patients: A Propensity Score-Matching Analysis Using the Spanish National Hospital Discharge Database for the Period 2016–2019

(1) Background: To analyze incidence, clinical characteristics, procedures, and in-hospital outcomes among patients hospitalized with community-acquired pneumonia (CAP) according to the presence of T2DM in Spain (2016–2019) and to assess the role of gender among those with T2DM. (2) Methods: Using the Spanish National Hospital Discharge Database, we estimated hospitalized CAP incidence. Propensity score matching was used to compare population subgroups. (3) Results: CAP was coded in 520,723 patients, of whom 140,410 (26.96%) had T2DM. The hospitalized CAP incidence was higher in patients with T2DM (both sexes) (IRR 4.25; 95% CI 4.23–4.28). The hospitalized CAP incidence was higher in men with T2DM than in women with T2DM (IRR 1.46; 95% CI 1.45–1.47). The hospitalized CAP incidence among T2DM patients increased over time; however, the in-hospital mortality (IHM) decreased between 2016 and 2019. IHM was higher among non-T2DM men and women than matched T2DM men and women (14.23% and 14.22% vs. 12.13% and 12.91%; all p < 0.001, respectively), After adjusting for confounders, men with T2DM had a 6% higher mortality risk than women (OR 1.06; 95% CI 1.02–1.1). (4) Conclusions: T2DM is associated with a higher hospitalized CAP incidence and is increasing overtime. Patients hospitalized with CAP and T2DM have lower IHM. Male sex is a significant risk factor for mortality after CAP among T2DM patients.

Characterization of neonatal opioid withdrawal syndrome in Arizona from 2010-2017

In this paper, we describe a population of mothers who are opioid dependent at the time of giving birth and neonates exposed to opioids in utero who experience withdrawal following birth. While there have been studies of national trends in this population, there remains a gap in studies of regional trends. Using data from the Arizona Department of Health Services Hospital Discharge Database, this study aimed to characterize the population of neonates with neonatal opioid withdrawal syndrome (NOWS) and mothers who were opioid dependent at the time of giving birth, in Arizona. We analyzed approximately 1.2 million electronic medical records from the Arizona Department of Health Services Hospital Discharge Database to identify patterns and disparities across socioeconomic, ethnic, racial, and/or geographic groupings. In addition, we identified comorbid conditions that are differentially associated with NOWS in neonates or opioid dependence in mothers. Our analysis was designed to assess whether indicators such as race/ethnicity, insurance payer, marital status, and comorbidities are related to the use of opioids while pregnant. Our findings suggest that women and neonates who are non-Hispanic White and economically disadvantaged, tend be part of our populations of interest more frequently than expected. Additionally, women who are opioid dependent at the time of giving birth are unmarried more often than expected, and we suggest that marital status could be a proxy for support. Finally, we identified comorbidities associated with neonates who have NOWS and mothers who are opioid dependent not previously reported.

Prediction of In-Hospital Mortality from Administrative Data: A Sequential Pattern Mining Approach

Study of trajectory of care is attractive for predicting medical outcome. Models based on machine learning (ML) techniques have proven their efficiency for sequence prediction modeling compared to other models. Introducing pattern mining techniques contributed to reduce model complexity. In this respect, we explored methods for medical events’ prediction based on the extraction of sets of relevant event sequences of a national hospital discharge database. It is illustrated to predict the risk of in-hospital mortality in acute coronary syndrome (ACS). We mined sequential patterns from the French Hospital Discharge Database. We compared several predictive models using a text string distance to measure the similarity between patients’ patterns of care. We computed combinations of similarity measurements and ML models commonly used. A Support Vector Machine model coupled with edit-based distance appeared as the most effective model. Indeed discrimination ranged from 0.71 to 0.99, together with a good overall accuracy. Thus, sequential patterns mining appear motivating for event prediction in medical settings as described here for ACS.