Clinical Research Directory

Experimental Trial of an Employment-Based Housing Program for Adults at Risk of Homelessness

Employment-Based Housing (EBH) is a novel model that helps unstably housed individuals obtain housing and on-site employment where they are housed. The long-term goal of this project is to examine how housing and vocational services can be integrated into hospital emergency departments (EDs) as a multi-sectoral intervention. The investigators central hypothesis is that EBH provides housing, employment, and health insurance which will reduce acute care use and ameliorate non-medical determinants of health more than standard care models. The rationale is that EBH addresses multiple upstream factors that can prevent recurring use of acute care, and the combination of housing with employment services will have positive synergistic effects. This project has three specific aims: 1) Conduct a three-arm randomized clinical trial comparing EBH to passive (resource list) and active control interventions (basic vocational services) on acute care use as the primary outcome and secondary outcomes related to housing, employment, and health-related quality of life; 2) Understand the experiences of subgroups of participants receiving EBH; and 3) Identify barriers and facilitators to adoption of EBH in public healthcare systems in other cities nationally. To achieve these aims, an EBH program will be integrated into the ED of the largest safety net hospital system in Houston, Texas- the 4th largest city in the United States. The project will include a three-arm randomized trial enrolling 750 participants (250 participants per arm). The three-arm trial allows for the investigation of the unique contribution of housing over other non-medical determinants of health (employment and insurance). Participants will be randomized to one of the three arms and followed for 12 months.

EWSs and 28-Day Mortality in Geriatric ED Patients

This prospective observational cohort study evaluated the prognostic performance of commonly used early warning scores for predicting 28-day all-cause mortality among geriatric patients presenting to the emergency department with non-traumatic conditions. Patients aged 65 years and older were consecutively screened during the study period. Demographic characteristics, comorbidities, vital signs, level of consciousness, blood gas parameters, complete blood count parameters, frailty status, and early warning scores were recorded at emergency department presentation or within the first hour of admission. The evaluated scoring systems included National Early Warning Score (NEWS/NEWS2), Modified Early Warning Score (MEWS), quick Sequential Organ Failure Assessment (qSOFA), Rapid Emergency Medicine Score (REMS), Cardiac Arrest Risk Triage (CART), and Hamilton Early Warning Score (HEWS) score. The primary outcome was 28-day all-cause mortality. The study also examined whether age, comorbidity burden, frailty, laboratory markers, and hemodynamic parameters were independently associated with 28-day mortality in this population.

CMSL Ambulatory Sensitive Condition Nudge Study 2

The project aims to evaluate a nurse-led intervention to reduce inappropriate emergency department (ED) use among adult patients seen at Geisinger's Community Medicine Service Line (CMSL) clinics. The intervention occurs immediately following an appointment where a patient receives a diagnosis of an ambulatory sensitive condition (ASC; i.e., a condition considered to be a risk factor for near-term ED use). The evaluation will compare eligible patients with an ASC who were randomly assigned to receive follow-up outreach (patient portal message and/or call) from a nurse (who was automatically prompted via the Epic electronic health record system to initiate outreach) with those who were randomly assigned to receive standard care. Analyses will be intent-to-treat. The primary outcome is ED use in the week (i.e., 7 days) following the appointment. We ran an earlier version of this intervention (NCT06798389). The current study is modified based on results and clinical guidance. Specifically, more conditions will be included as qualifying ASCs for enrollment. Patients under 30 will be excluded. And rather than calling all patients as in the original study, patient portal users may be contacted via the portal instead of or in addition to a phone call. Finally, in the first study, the intervention was differentially effective by age group (\<45, 45-64, 65+). Our primary analysis will be conducted separately by age group, though we will also conduct an analysis combining across age groups. We will run the study until we reach at least 4,330 patients in each of the following age groups: patients aged 30-45, patients aged 45-64, patients aged 65+. Therefore, our estimated sample size is at least 4,330x3 = 12,990. We may be required to do an interim data pull and/or stop the study early at the direction of clinical or operational leaders.

Predictors of Emergency Department Use in Frail Patients

When admitted to the emergency department (ED), elderly non-autonomous patients show high risk of adverse health outcomes. The prompt identification of ED use risk factors in such population is hence needed. While cognitive impairment is a known clinical risk factor, biomarkers of most prevalent dementias have been scarcely investigated as possible ED use predictors. Within this context, this prospective study aims at exploring whether plasma phospho-tau181 and cerebrovascular burden can predict ED use at 6 months in elderly non-autonomous patients, irrespective of frailty.

Integrated Telemedicine Program Evaluation

This project will evaluate a virtual, on-demand telemedicine program for High emergency department (ED) utilizers with comorbid physical health conditions and behavioral health correlates. Researchers will randomize patients to have either full access to the full intervention (treatment group) or to receive only standard care (control group). Analyses will be intent-to-treat. The primary outcome is the number of ED visits 120 days after the first Best Practice Alert (BPA) firing. The research team plans to enroll 3200 patients in this study. However, randomization may end due to system constraints on December 31, 2025 before reaching that target. This sample provides 80% power to detect a 35% relative reduction in ED utilization using a two-tailed test with an alpha of .15, assuming 40% compliance with the program.