Evaluating a Mobile Integrated Health Transitional Care Program to Reduce Readmissions: Findings From a Quasi-Experimental Design

Abstract

Background:

The period following hospital discharge marks a precarious transition for older adults across medical, functional, and social domains. Paramedic-led mobile integrated health (MIH) programs offer a promising approach to bridging gaps in care, though their effectiveness in transitional care remains poorly characterized. The objective of this study was to evaluate the association of the Paramedic Assisted Community Evaluation (after) Discharge (PACED) intervention with 30-day rehospitalization rates among frail older adults.

Methods:

In this prospective observational study, the clinical team offered older adults discharged from an urban academic medical center a paramedic-delivered home visit within 72 hours. Patients who lived outside the program’s geographical catchment area, did not have a system-affiliated primary care doctor, or who declined visits, served as a comparator group to create a natural experiment. The primary outcome was 30-day rehospitalization. Secondary outcomes included 30-day emergency department (ED) utilization without admission. The characteristics of the sample and the content of the PACED visits for enrolled patients were presented descriptively. Modified Poisson regressions were used to estimate relative risks (RR) for 30-day rehospitalization and ED utilization, adjusting for demographics, comorbidities, frailty, and prior healthcare use with 95% confidence intervals (CIs).

Results:

Of 297 subjects, 190 received the PACED intervention (median age 82.7, 66.8% female) and 107 served as comparators (median age 81.3, 58.0% female). Thirty-day rehospitalization was lower among PACED participants (12.6% vs. 21.5%, adjusted RR 0.45, 95% CI 0.26–0.77, p=0.003), and they experienced fewer 30-day ED visits (11.5% vs. 18.7%, adjusted RR 0.58, 95% CI 0.33–0.98, p=0.05).

Conclusions:

An in-home Mobile Integrated Health (MIH) transitional care program for frail older adults after hospital discharge was associated with lower 30-day readmission rates. These findings highlight MIH as a promising model to support aging in place and suggest its potential value for adoption within age-friendly health systems.

Introduction

The time immediately after hospital discharge is marked by heightened medical risk for older adults. Within one month of going home, one in five will suffer an adverse event, and almost a quarter will be readmitted to the hospital.^{1, 2} These setbacks represent serious harms, including physical and emotional strain on patients and families, risk of delirium or iatrogenic harm, declining functional independence, increased likelihood of long-term disability or death, escalating healthcare costs, and penalties levied on health systems by the Center for Medicare and Medicaid Services.^2–4

For many older adults living with frailty, the challenges are even greater: their recovery is shaped not only by complex medical conditions and physiologic vulnerability but also by the influence of social circumstances and barriers to care that can make a transition particularly difficult.^{5, 6} Because of the clinical, emotional, and financial harm associated with unplanned rehospitalizations, a range of strategies have been studied to reduce their incidence.^{5, 7} Best practices for improving post-discharge outcomes emphasize multi-domain interventions with timely follow-up, patient and caregiver education, cross-team coordination, and intensive clinical monitoring.^{7, 8} Emerging transitional care models have demonstrated evidence of effectiveness in reducing readmissions and improving outcomes.^{2, 8, 9}

Despite this progress, healthcare systems continue to face clinical and operational challenges in implementing structured, scalable transitional care programs that can adequately support a medically and socially complex population.^9–11 Barriers to the sustainability of transitional care programs include operational complexity, funding constraints, and inadequate organizational infrastructure, particularly to provide community-based services.^{10, 12} Many programs require significant resources or depend on resources that are not universally available, which impedes the transferability of existing programs to additional settings.^10–13 These limitations have prevented the widespread adoption of any single model of transitional care.

Mobile integrated health (MIH) programs may help address these gaps. Defined as the use of mobile resources and assets to provide coordinated, community-based care, MIH programs deploy highly skilled community paramedics (CPs) to deliver in-home care in coordination with remote physicians.^14–16 Such models have emerged as a promising strategy to enhance care access.^{15, 17} MIH clinicians can perform clinical assessments, deliver treatments, coordinate with longitudinal care teams, and escalate care when needed.^{15, 16, 18} MIH programs are designed to serve as a collaborative interface linking health systems, community resources, and patients through EMS clinicians, providing an adaptive extension of care that enhances coordination within existing care teams, unlike traditional multidisciplinary models, which often operate as collections of siloed providers.¹⁵ MIH models have been associated with reduced utilization of emergency medical services (EMS), fewer emergency department (ED) visits, decreased hospitalizations, and improved patient satisfaction.^{14, 19, 20} MIH offers a uniquely adaptable and rapidly deployable care delivery model that builds on the widespread presence and infrastructure of EMS.

Few studies have evaluated the effectiveness of MIH-based transitional care programs, particularly among frail older adults. To address this gap, we sought to evaluate the impact of a MIH-facilitated transitional care intervention called the Paramedic Assisted Community Evaluation (after) Discharge (PACED) program. A feasibility pilot of PACED in a limited cohort has been previously described.²¹ That study informed iterative refinement of the intervention, including protocol modifications and workflow integration.²¹ The present manuscript builds on that work by evaluating the effectiveness of the iterated PACED program after incorporation into routine clinical practice, in a larger, more generalized population. The objective of this study is to evaluate the effect of the PACED intervention on 30-day rehospitalization risk in community-dwelling older adults with frailty. We hypothesized that undergoing a PACED visit would be associated with a lower risk of early readmission and ED utilization.

Methods

Study Settings and Participants

This prospective observational study was conducted at a tertiary academic medical center located in the northeastern United States, serving a diverse population across urban, suburban, and rural regions with approximately 200,000 annual patient encounters. Following completion of a feasibility pilot of PACED²¹, the intervention was revised based on stakeholder feedback and operational findings and subsequently integrated into routine clinical care. Refinements included adding an on-call clinical pharmacist for consultation, formal coordination with the inpatient nurse navigators to prioritize the timing of the PACED visit before other services, and the use of a shorter, structured note template. The current analysis includes a larger eligible population and approximately twice the number of enrolled participants compared with the pilot, permitting a more rigorous evaluation of clinical effectiveness and reflecting real-world implementation at scale. Data from patients included in the pilot study were included in this effectiveness evaluation.

Between July 2023 and June 2025, patients discharged from medical inpatient or observation units to independent or assisted living settings were screened for participation. Eligibility criteria included age ≥ 65 years, residing within a 30-mile radius of the hospital (aligned with the institution’s MIH service area), having an established relationship with a system-affiliated primary care provider (PCP) (to ensure effective inter-team communication), and having an eFrailty index ≥0.24, a threshold consistent with moderate-severe frailty, which is in turn associated with an elevated risk of 30- and 90-day ED visits and hospital readmissions.²²

Exclusion criteria included patients discharged to post-acute institutional care facilities, including skilled nursing facilities and subacute rehabilitation. We also excluded patients admitted for elective day procedures (e.g., endoscopy) and those residing outside the designated MIH catchment area. The study protocol was approved by the affiliated medical school’s Institutional Review Board. The study team aligned the best practices and reporting approach to the STROBE checklist for reporting cohort studies.²³

Intervention

The PACED intervention is comprised of a single post-discharge home visit conducted by a CP, designed to provide transitional clinical support between hospital discharge and outpatient follow-up. Visits are performed within 72 hours of discharge. Before referral to the PACED program, a dedicated nurse navigator on the PACED staff coordinated care for all enrolled patients. The navigator coordinated scheduling with the inpatient care management team to prioritize the PACED visit ahead of other home-based services and clinic-based follow-ups when applicable. PACED was implemented within routine clinical care and did not exclude patients receiving concurrent home health services, including post-discharge nursing visits.

The components of the intervention are summarized in Figure 1. In addition to a clinical evaluation, medication reconciliation, and assessment for unmet social drivers of health needs, the program screens for delirium (3D-CAM instrument) and falls (Timed-Up-and-Go Test).^{24, 25} Each visit is conducted under real-time physician supervision via a HIPAA-compliant video telehealth platform. This includes a face-to-face conversation during the visit between the patient, family members, the CP, and the physician. A structured template shared by the CP and supervising physician is used to document relevant clinical and social findings. The intervention workflow and documentation were refined iteratively over the pilot period in response to feedback from frontline clinicians.²¹ The structured PACED note is routed to the patient’s PCP at the conclusion of the visit.

Figure 1: Elements of the PACED Intervention.

PACED Elements. Patients who received a PACED visit underwent a structured home assessment in which paramedics evaluated multiple domains of transitional care associated with rehospitalization risk.

A standardized escalation protocol was established to guide the response to clinical concerns identified during the visit. All clinical problems are communicated to the staffing physician and the patient’s PCP. Unmet social or functional needs identified during PACED visits are escalated to the nurse navigator. Urgent findings requiring same-day intervention (prescription errors, missing durable medical devices, or clinical concerns needing immediate attention) are managed by the physician staffing the visit, in consultation with the patient’s PCP or the geriatrician on call as needed. Patients requiring emergency evaluation (including patients with high-acuity medical complaints or unstable vital signs) are transported to the ED by a separate transporting EMS unit.

Positive 3D-CAM delirium assessments are escalated in real time to the staffing physician, and a determination is made as to whether further diagnostic workup and an alternative disposition plan are necessary. For TUG results, CPs attempted to immediately mitigate environmental hazards. If additional mobility equipment is needed or there are other fall concerns, the CP consulted with the care navigator and staffing physician. If an error or discrepancy was identified between the discharge medication list and what the patient has in their home, the CP consults with the staffing physician and on-call clinical pharmacist to remediate the concern.

Comparison Group

A natural comparator group was identified from patients who met all primary eligibility criteria but were excluded due to residence outside the designated geographic catchment area of the MIH program, lack of affiliation with a health system primary care provider, or because they declined or were not offered the intervention. Clinical outcomes for these participants were abstracted from the electronic health record (EHR) by nonblinded study personnel for the period spanning July 2023 to July 2025.

Measures and Analysis

This study used a quasi-experimental, observational design with a convenience sample; therefore, we did not perform an a priori power calculation. The primary outcome of this study was 30-day rehospitalization. The secondary outcome was 30-day ED utilization without admission. These outcomes were measured via direct review of encounters documented within our health system’s EHR, supplemented by use of the “Care Everywhere” health information exchange, which provides access to data from surrounding hospital systems.

We also conducted a preplanned exploratory sub-analysis to assess whether the effect of the intervention varied across key demographic and clinical subgroups. Subgroups included age (<85 vs ≥85 years), sex (male vs female), relative eFrailty index (<0.5 vs ≥0.5), presence of chronic cardiorespiratory disease (chronic obstructive pulmonary disease [COPD] or heart failure), dementia diagnosis, number of hospitalizations the prior year, and primary diagnosis type at index hospitalization. The intervention and control subjects’ demographic and clinical characteristics were summarized descriptively; baseline characteristics were compared using chi-square tests for categorical variables and t-tests or Wilcoxon rank-sum tests for continuous variables, as appropriate.

To estimate the association of PACED visits with the risk of 30-day rehospitalization, we used a modified Poisson regression with robust standard errors to directly estimate adjusted relative risks (RRs). The model was adjusted for potential confounders identified a priori, including age, sex, frailty index, comorbid conditions, index hospitalization length of stay, and number of hospitalizations in the previous twelve months. We also conducted a stratified analysis, fitting separate modified Poisson regression models for each subgroup. In each stratified model, we estimated the adjusted relative risk of 30-day rehospitalization for the PACED and control group, using the same covariate set as in the primary model. We evaluated the Poisson model for overdispersion by comparing the residual deviance to the residual degrees of freedom. All models were assessed for goodness-of-fit and multicollinearity. Statistical significance was defined as a two-sided p-value < 0.05. Analyses were performed using Stata version 17.0 (StataCorp, College Station, TX).

Results

Study Participants

In total, 297 subjects were included in the analysis: 190 in the PACED group and 107 in the control group. A STROBE diagram depicting screening and inclusion is illustrated in Figure 2. Table 1 summarizes the demographic characteristics of the intervention group (median age 82.7, 66.8% female) and the control group (median age 81.3, 58.0% female). There was no significant difference between the groups concerning age, sex, or number of previous hospitalizations in the preceding 12 months. The intervention group had a higher median eFrailty index (0.42 versus 0.39, p=0.001) and a higher prevalence of COPD (28.0% versus 6.7%, p=0.03). There was no other detected difference between comorbidities in the PACED group versus the controls, including a history of dementia.

Figure 2: STROBE Flow Diagram.

PACED STROBE Diagram. Patients were screened for age, eFrailty index, and disposition from the hospital using the electronic health record. They were subsequently assessed for geographical location and presence of an affiliated PCP. An additional 39 patients were not approached or declined a PACED visit; those excluded based on those criteria were included in the study as the comparator group. The remaining 190 eligible participants underwent a PACED visit.

Table 1:

Subject Demographics

	PACED (n=190)	Control (n=107)	p
Age
Median (IQR)	82.7 (12)	81.3 (11)	0.14
Sex n (%)
Female	127 (66.8)	63 (58.0)	0.14
Male	63 (36.0)	44 (41.2)
Frailty Index
Median (IQR)	0.42 (0.17)	0.39 (0.16)	0.001
Race n (%)
White	173 (91.1)	95 (88.7)	0.004
Asian	14 (7.4)	2 (1.8)
Black/African American	1 (0.5)	9 (8.4)
American Indian	2 (1.1)	1 (0.93)
Ethnicity
Not Hispanic or Latino	165 (86.8)	92 (85.9)	0.61
Hispanic or Latino	25 (13.2)	15 (14.0)
Comorbid Conditions n (%)
Chronic Obstructive Pulmonary Disease	46 (24.2)	12 (11.2)	0.008
Congestive Heart Failure	31 (16.3)	11 (10.3)	0.17
Coronary Artery Disease	47 (24.7)	31 (29.0)	0.59
Dementia	22 (11.6)	19 (17.7)	0.42
Diabetes	65 (34.2)	31 (29.0)	0.47
Hyperlipemia	126 (66.3)	61 (57.0)	0.59
Hypertension	162 (85.3)	99 (92.5)	0.09
Osteoporosis	50 (26.3)	19 (17.8)	0.32
Stroke/CVA	20 (10.5)	8 (7.5)	0.11
Previous hospitalizations in 12 months before index admission
Median (IQR)	2 (3)	2 (2)	0.49

PACED Visit Characteristics

The clinical findings from PACED visits are summarized in Supplemental Table 1. A medication reconciliation was completed in 179 patients (94.2%). Medication errors were identified in 73 patients (38.4%) and corrected during the visit for 68 patients (93.2%). A total of 36 patients (18.9%) had less than a 30-day supply of medications.

CPs documented acute clinical concerns for 57 patients (30.0%) and additional therapeutic needs in 71 patients (37.4%). Emergent safety concerns were identified in 3 patients (1.6%). Positive delirium screens occurred in 6 patients (3.2%), orthostatic vital signs in 4 patients (2.1%), and environmental fall risks in 7 patients (3.7%). A total of 175 patients (92.1%) had a Timed Up and Go (TUG) greater than 14 seconds, an indicator associated with frailty and increased fall risk.^{25, 26}

Unmet nonclinical health needs were identified in 67 patients (35.3%), most commonly requiring help with transportation (20.0%), housework (12.1%), shopping/meal preparation (10.0%), transferring/walking (4.2%), and challenges with medication access/adherence (6.3%). Disposition following the visit was treatment at home for 187 patients (98.4%) and transportation to the ED for 3 patients (1.6%).

Clinical Outcomes

The crude and adjusted relative risk of 30-day acute care utilizations are summarized in Table 2. The PACED group had 12 rehospitalizations (12.6%) versus 23 in the comparator group (21.5%), yielding an adjusted RR of 0.45 (CI 0.26–0.77). ED visits without hospitalization were observed in 22 (11.5%) participants in the PACED group and 20 (18.7%) participants in the comparator group (adjusted RR 0.58, CI 0.33–0.98).

Table 2:

Participant Health Utilization Outcomes, All Subjects

	PACED (n=190) (n,%)	Control (n=107) (n, %)	Crude RR (CI)	p	Adjusted RR (CI)	p
30-Day Rehospitalization	24 (12.6)	23 (21.5)	0.54 (0.32–0.91)	0.02	0.45 (0.26–0.77)	0.003
30-Day ED encounter (no hospitalization)	22 (11.5)	20 (18.7)	0.62 (0.32–0.90)	0.02	0.58 (0.33–0.98)	0.05

Clinical outcomes by patient subgroup are depicted in Figure 3 and summarized in Supplemental Table 2. When stratified by demographic and clinical subgroups, the PACED intervention was associated with a more significant risk reduction of 30-day hospitalization in men, patients >85, patients with fewer than 2 hospitalizations before index hospitalization, and patients without CHF/COPD. Additionally, a more significant effect was noted in patients whose primary diagnosis at index hospitalization was for a pulmonary, gastrointestinal, or hematological/oncological diagnosis.

Figure 3: Stratified Risk of 30-day Rehospitalization, PACED Group Vs. Comparators.

Stratified Risk of 30-day Rehospitalization, PACED Group Vs. Comparators. Subgroup analyses are displayed by frailty index, sex, age, number of prior hospitalizations, comorbid conditions and index hospitalization diagnosis category. Diamonds represent point estimates, and horizontal lines represent 95% confidence intervals.

Discussion

In this study, an MIH-facilitated transitional care model for frail older adults discharged in the community was associated with a 55% relative reduction in 30-day rehospitalization risk compared with patients receiving usual care, suggesting that this model may contribute to improved short-term post-discharge outcomes in a high-risk population. These findings suggest that MIH should be considered as a promising strategy to support patients, coordinate care, and identify risk factors for adverse events to help older adults navigate the transitional periods of their health trajectories.

Our findings should be interpreted in the context of a previously published pilot feasibility study of this program.²¹ That initial project established operational feasibility, demonstrated high protocol fidelity, and identified workflow refinements necessary for real-world implementation. This study builds on the pilot by evaluating outcomes associated with the iterated intervention after complete integration into routine practice, using a larger eligible cohort. This graduated reporting, from feasibility and implementation to effectiveness evaluation, was designed to reflect best practices for systematically translating complex interventions into sustainable clinical practice.^{27, 28}

The identification and communication of actionable high-risk findings during PACED visits may be a mechanism of the observed reduction in rehospitalization in the treated population. Nearly 40% of patients were found to have at least one medication error, most of which were corrected in real time, reducing the risk of adverse drug events, a leading cause of readmission.²⁹ In addition, numerous patients lacked an adequate supply of their medications. CPs frequently identified acute clinical concerns and additional therapeutic needs, enabling timely escalation before these issues progressed to hospital-level events. Targeted screening for known risk factors for rehospitalization, including delirium, fall risk, and unmet social needs, further allowed the care team to address multifactorial contributors.^{6, 30–32} The use of MIH during these visits highlights the benefit of active embedded medical direction, an enabling mechanism that translates CP access, trust, and clinical skills into integrated and impactful care.

Subgroup analyses suggest that the benefits of the PACED intervention were not uniform across all patient types. Patients aged ≥ 85 years appeared to derive greater benefit from the intervention, although there was minimal difference in patients stratified by frailty.

Patients without a COPD or CHF diagnosis who received PACED services also appeared to have lower risks of 30-day rehospitalization, highlighting that the intervention may be most effective in subsets of older adults who have less tenuous comorbid conditions. Tailoring transitional care interventions to the unique risks of older adults aging in place and directing interventions to patients most likely to benefit may maximize impact and optimize resource utilization.

Other intensive transitional care initiatives have shown effectiveness in reducing rehospitalization by enhancing coordination and providing intensive supervision during transitions.^{2, 8, 33, 34} Programs have reported a 20–44% reduction in early rehospitalization coupled with substantial healthcare savings.^{2, 35, 36} However, despite their evidence of effectiveness, replicating existing options for transitional care delivery remains problematic due to scalability barriers. Communication and information sharing are often fragmented, with poor coordination among healthcare teams, which can lead to delays and errors.^{9, 10, 37} Systemic and policy barriers, such as fragmented funding and weak incentives for follow-up, particularly in rural and underserved regions, create further gaps.^{9, 10, 37, 38} Together, these limitations highlight persistent challenges in delivering comprehensive, scalable, and sustainable transitional models.

MIH programs may overcome common limitations of traditional transitional care programs by leveraging their integration within health systems, adaptability, and efficient use of existing infrastructure. Unlike siloed interventions, optimized MIH teams function as an extension of the health system, facilitating real-time communication and coordinated information sharing among acute, post-acute, and community-based teams.^{15, 39} Their operational flexibility allows services to be tailored to the complex needs of older adults, providing episodic, rapid-response care, while the widespread availability of EMS agencies reduces the need for new infrastructure. EMS clinicians are skilled at delivering clinical care in the home, including high-level diagnostic and therapeutic interventions for acute issues affecting frail, community-dwelling older adults.¹⁵ Their integration with public safety and healthcare systems, along with the trust they hold in the community, normalizes their presence in patients’ homes. By leveraging existing EMS networks, MIH programs can deploy resources efficiently, reduce gaps in care, even in rural or resource-constrained areas, and provide a resilient alternative to fragmented transitional programs. Effective interprofessional practice, which is intrinsic to MIH, further enhances care by combining the trust, access, and clinical abilities of EMS clinicians with the complementary skills of other healthcare professionals, creating a more efficient and responsive model that addresses both the medical and social needs of older adults.

This study has limitations. It was observational, quasi-experimental, and non-randomized, so despite adjusting for covariates, residual unmeasured confounding may exist, and outcomes should be interpreted as associations rather than causal effects of the PACED intervention. Classifying patients who declined the intervention and patients who were outside of the MIH program’s geographical catchment area as a control group may have introduced selection bias, particularly since geographical distance from hospital resources impacts health outcomes in older adults.⁴⁰ Additionally, we did not differentiate between inpatient versus observation index admissions, between patients who had other home health services and those who did not, or between patients who were homebound and those who were more mobile. Imbalances in those characteristics may have introduced unmeasured bias into the study results. We did not compare the study cohorts’ clinical outcomes to those of less frail older adults or patients discharged to institutional settings; understanding how the participants’ outcomes compared to these other populations may have further contextualized the study’s findings.

The study was single-center, limiting generalizability, and the cohort may have been underpowered for subgroup analyses. Although the intervention was standardized, visit content and quality could vary, potentially affecting outcomes. Data were obtained from medical records, which may be subject to misclassification or abstractor error, and some rehospitalizations outside the system may have been missed, although EHR health information exchange review indicated most patients exclusively used the study hospital system. Finally, we did not measure patient-centered measures such as satisfaction or quality of life.

Beyond scientific limitations, pragmatic factors may also constrain the sustainability and scalability of MIH-led transitional care programs. Despite growing adoption and evidence supporting MIH’s effectiveness, policy and reimbursement structures have not kept pace.¹⁵ Most programs are not covered by insurance and depend on grants, payor contracts, or health system support.^{15, 41, 42} Ongoing advocacy by clinicians, health system leadership, and policymakers will shape the future role and integration of MIH in clinical practice.

MIH models for the peri-acute care of older adults advance the practice of caring for this population at home, a key priority for patients and their medical teams.⁴³ By providing in-home assessment, chronic disease management, medication reconciliation, and timely follow-up, MIH reduces the need for burdensome travel to clinical settings and supports independence. For clinicians, MIH enhances care coordination, improves communication across care teams, and allows earlier identification of health changes that may prevent hospitalizations. Despite these benefits, further research is needed to optimize MIH for geriatric populations, including identifying which interventions are most effective in preventing functional decline and hospitalization, understanding patient and caregiver experiences, and tailoring program workflows to accommodate cognitive impairment, multimorbidity, and social drivers of health. Further study is also warranted to determine the optimal way to identify patients most likely to derive benefit from more intense clinical supervision at home.

An MIH-facilitated transitional care intervention was associated with reduced risk of 30-day rehospitalization among frail older adults following hospital discharge. These findings suggest that leveraging CPs to deliver structured, in-home transitional care may represent a scalable strategy to address a period of heightened vulnerability for older adults. While additional research is needed to confirm effectiveness and sustainability across diverse settings, our results provide early evidence that MIH-based approaches can strengthen transitional care and reduce avoidable acute care utilization.

Key Points:

• A paramedic-led Mobile Integrated Health (MIH) transitional care intervention was associated with lower 30-day hospital readmissions among frail older adults after hospital discharge

• Paramedics identified and addressed frequent medication discrepancies, acute clinical issues, and unmet social needs

• MIH represents a flexible model that leverages existing infrastructure to deliver needed care across the continuum, extending healthcare capacity beyond hospitals and clinics to better support older adults in their homes and communities.

Why this paper matters

Older adults face high risks of rehospitalization and care fragmentation after discharge. This study demonstrates that paramedic-delivered transitional visits can reduce early post-discharge healthcare utilization, offering a community-based model to promote safer transitions and support aging in place by leveraging existing mobile infrastructure.