Skilled Nursing Facility–to–Home Transitions After Heart Failure Hospitalization: A Mixed-Methods Study of Communication, Self-care, Medication, and Follow-up

Abstract

Background:

Skilled nursing facilities (SNFs) play a critical role in post-acute recovery for older adults with heart failure (HF), yet the transition from SNF to home remains a vulnerable and understudied phase of care. Although discharge guidelines emphasize clear communication, HF-specific self-care education, medication management, and follow-up coordination, little is known about how these practices are implemented during SNF-to-home transitions.

Methods:

We conducted a convergent mixed-methods study across four non-profit SNFs, integrating data from post-discharge patient and caregiver surveys, structured medical record abstraction of discharge instructions, and semi-structured staff interviews. Eligible patients were Medicare beneficiaries aged ≥65 years discharged from SNF to home following HF hospitalization, with SNF stays ≤60 days. Quantitative data were analyzed using descriptive statistics, while qualitative data underwent thematic and directed content analysis. Findings were triangulated across data sources to identify key challenges and actionable strategies.

Results:

Among 150 respondents, 59% reported receiving written discharge instructions; however, HF-specific self-care elements (e.g., daily weight monitoring, low salt diet) were documented in only 15–41% of instructions. Although 87% reported receiving a medication list, only 53% had it reflected in the discharge instructions, and adherence support was infrequently addressed (24%). Follow-up coordination was similarly discordant: 37% of respondents reported a scheduled primary care appointment, compared with 13% documented in discharge instructions. Staff interviews revealed non-standardized discharge workflows, workforce constraints, and reliance on verbal education, contributing to variability in patient preparation and communication across care settings.

Conclusions:

SNF-to-home transitions after HF hospitalization are marked by discordance between patient-reported education and written documentation, as well as inconsistent medication and follow-up coordination. These gaps represent modifiable vulnerabilities during a high-risk recovery period. Standardized HF-focused discharge workflows and strengthened cross-setting communication may improve transitional care, while long-term solutions must address structural and workforce constraints.

Introduction

Heart failure (HF) is a leading cause of hospitalization among older adults, accounting for over 1 million hospitalizations annually in the United States.¹ Many patients with HF are older adults with multimorbidity^2–5 and functional decline,^6,7 which often worsens during hospitalization.⁸ Consequently, more than one in five patients hospitalized for HF are discharged to skilled nursing facilities (SNFs) for short-term rehabilitation^9,10 before returning home.

The transition from SNF to home represents a vulnerable and understudied phase of care. Nearly one in four patients discharged from SNF to home after HF hospitalization are rehospitalized within 30 days.¹¹ During this period, patients may experience evolving care plans, medication changes, and fragmented communication with outpatient clinicians, all of which contribute to the complexity of this transition.

HF guidelines from the American College of Cardiology and American Heart Association¹² recommend patient-centered discharge planning, including clear written instructions, HF-specific self-care education, and timely follow-up. However, limited evidence exists on how these practices are implemented during SNF-to-home transitions.

To address these gaps, we conducted a convergent mixed-methods study to examine (1) how patients and caregivers experience the SNF-to-home transition after HF hospitalization, (2) how discharge instructions are documented in SNFs, and (3) the structural factors shaping the quality of these transitions.

Methods

Data availability

The de-identified quantitative data underlying this article will be made available upon reasonable request to the corresponding author, subject to institutional review and data use agreements. Qualitative interview transcripts are not publicly available because of confidentiality protections. Study materials, including the survey instruments, medical record abstraction tool, and interview guide, are provided in the Supplemental Material.

Study design

We conducted a convergent mixed-methods study¹³ across four non-profit SNFs. A convergent mixed-methods approach integrates quantitative and qualitative data collected concurrently to provide a comprehensive understanding of the phenomenon under study.¹³ The study integrated three data sources: a post-discharge patient and caregiver survey, structured medical record abstraction of discharge instructions, and semi-structured interviews with SNF staff. By triangulating data from these sources, we aimed to identify key challenges and actionable strategies to strengthen HF transitions from SNF to home. This study followed the GRAMMS framework for reporting mixed methods studies.¹⁴ Protocols were approved by the Institutional Review Boards at NYU Grossman School of Medicine and the University of California, San Francisco (UCSF).

Setting

The four SNFs were non-profit facilities located in large US cities and represented a broad range of CMS Care Compare overall quality ratings, including one 1-star facility, one 2-star facility, one 4-star facility, and one 5-star facility. Additional facility characteristics are provided in the Supplemental Material (Table S1).

Participants and recruitment

Patients and caregivers

Inclusion criteria:

Eligible patients were Medicare beneficiaries aged ≥65 years discharged from SNF to home after SNF admission following a HF-related hospitalization, with a SNF length of stay ≤60 days. Eligibility was limited to English-speaking patients or those with an English-speaking legally authorized representative (caregiver proxies). Patients with cognitive or communication impairment were enrolled via their legally authorized representative following a standardized cognitive capacity screening protocol. Eligibility was confirmed through manual chart review. Detailed information on patient identification, screening, and the cognitive assessment protocol is in the Supplemental Material (Supplemental Method 1, Supplemental Method 2).

Exclusion criteria:

Exclusion criteria included non-English-speaking patients without an English-speaking proxy, SNF stays >60 days, discharge against medical advice, or hospital readmission from SNF for >7 days post-enrollment. We limited the cohort to short-term SNF stays (≤60 days) to focus on post-acute rehabilitation following HF hospitalization, as longer stays often reflect chronic custodial or extended rehabilitation needs with distinct discharge processes. We also excluded patients with prolonged hospital readmissions (>7 days) during the SNF stay, as such interruptions represent substantial changes in clinical trajectory and may initiate a new episode of post-acute care, complicating interpretation of the SNF-to-home transition following the index hospitalization.

Patients (or legally authorized representatives for patients with cognitive impairment) were approached in person at the SNF or by telephone during or shortly after the SNF stay. Those who agreed to participate in this study provided written or verbal informed consent prior to engaging in any research activities. Patients with cognitive impairment who were unable to complete the survey independently were represented by a caregiver proxy following a standardized cognitive capacity screening protocol. Caregivers also completed surveys when patients preferred proxy assistance. Separate informed consent was obtained for all participants, including caregiver respondents.

SNF staff

Key informants were purposively selected¹⁵ with assistance from site medical directors to capture diverse roles in the discharge process (nurses, social workers, case managers, physicians, nurse practitioners, administrators). Eligibility required ≥6 months at the facility, ≥1 year of post-acute care experience, involvement in SNF discharge, and experience with HF patients. Snowball sampling¹⁶ identified additional staff informants. For semi-structured staff interviews, written or verbal informed consent was obtained prior to participation.

Data sources and measures

Post-discharge survey

The post-discharge survey is provided in the Supplemental Material (eAppendices 1–2) and was developed through an iterative process involving literature review, expert consultation, and pilot testing with individuals who had family members experience SNF-to-home transitions. Survey questions were mapped to CMS SNF Conditions of Participation^17,18 and Joint Commission standards^19,20 to ensure alignment with federal and accreditation requirements for safe discharge practices. Survey questions assessed receipt and self-reported understanding of written discharge instructions, HF-specific self-care education, medication education, and follow-up scheduling, with open-ended questions exploring discharge challenges and successes. Caregivers completed an adapted version of the survey (eAppendix 2) for patients with cognitive impairment or those who preferred their caregiver to complete it. Patient-level sociodemographic data were also collected. Additional details on survey development and validation are provided in the Supplemental Material (Supplemental Method 3).

Medical record abstraction

We developed a structured medical record abstraction tool (Supplemental Material, eAppendix 3) to assess the content and quality of patient-facing discharge instructions, informed by CMS discharge planning requirements¹⁷ and the retired CMS HF-1 core measure.²¹ The tool included 26 items focused on discharge instructions, including reason for index hospitalization, HF self-care elements, presence of a medication list, contact information, and scheduled follow-up. Additionally, we also collected the frequency with which discharge summaries documented transmission to the patient’s primary care provider (PCP), given that communication with outpatient clinicians is a core expectation for continuity of care and a requirement under CMS discharge planning standards. Detailed patient-level variables were abstracted only for enrolled participants who provided informed consent and completed the post-discharge survey.

Patient-level information on age, sex, Medicaid status, and comorbidities was abstracted from the SNF electronic health record (EHR). Frailty was assessed using the FRAIL-NH scale, a validated 7-item instrument designed for nursing home residents (range 0–13; higher scores indicate greater frailty), and also abstracted from the EHR.²² Cognitive impairment was measured using the Cognitive Function Scale (CFS),²³ derived from the Minimum Data Set 3.0 and categorized patients as cognitively intact, mildly impaired, moderately impaired, or severely impaired. Abstractions were performed by trained personnel using standardized protocols and underwent secondary review by a second research team member to ensure accuracy. Detailed abstraction procedures are in the Supplemental Material, Supplemental Method 4.

Staff interviews

Staff interviews were conducted at three SNFs. A semi-structured interview guide was developed to explore current discharge practices, barriers, and facilitators to safe SNF-to-home transition for patients with HF. Interviews were conducted in person, by telephone, or video conference, audio-recorded, and professionally transcribed. The interview guide is in Supplemental Material (eAppendix 4). Additional details on interview guide development and data collection are in the Supplemental Material (Supplemental Method 5)

Data management

Survey and abstraction data were collected and managed in REDCap.²⁴ Interview transcripts were managed in Dedoose (SocioCultural Research Consultants, LLC, Manhattan Beach, CA).

Analysis

Quantitative analysis

We summarized survey and abstraction data using descriptive statistics. Primary process measures included receipt and presence of written discharge instructions, HF self-care elements, provision of medication lists and adherence supports, and outpatient follow-up. We compared patient/caregiver recall with discharge instructions documentation for key HF self-care, medication management, and follow-up elements. Analyses were performed in SAS 9.4 (SAS Institute, Cary, NC).

Qualitative analysis

We used a combination of qualitative and quantitative content analysis to analyze open-ended survey responses.^25,26 We organized data analysis around discharge challenges and successes. We first used qualitative content analysis to systematically examine the open-ended responses to obtain a condensed understanding and description of content.²⁵ We used a data-driven (inductive) approach to analysis whereby open coding was performed by two reviewers to identify salient and elevated topics of importance within the data set. Codes were then grouped into higher order categories. Quantitative content analysis was then performed to count coding categories.²⁶ This was conducted for the purpose of providing a more detailed assessment of how frequently certain codes were mentioned. Additional details for survey coding are in the Supplemental Material, Supplemental Method 5.

For staff interview data, we used thematic analysis to systematically identify recurring patterns and themes in the data.^27,28 Team members independently performed a data-driven (inductive) approach without imposing pre-existing frameworks. For each interview transcript, two trained analysts conducted line-by-line coding. Analysts met regularly to compare codes to ensure consistent interpretation, resolve discrepancies through negotiated consensus, and to iteratively refine a shared codebook.²⁹ Codes were then grouped into broader categories and synthesized into themes.

To ensure theoretical and methodological rigor and credibility in qualitative analysis of data, several strategies were employed.^30–32 These included sustained immersion and familiarization with the data, ensuring multivocality, reflexive memoing, independent coding followed by group reviewing to reach coding consensus, maintenance of an electronic paper trial of analytic decisions, and frequent research team meetings to review code and theme development. Data analysis was managed using Dedoose (SocioCultural Research Consultants, LLC, Manhattan Beach, CA).

Mixed-methods integration

The research team used informational matrices to triangulate³³ findings across data sources by domain, juxtaposing quantitative frequencies with qualitative data from patients, caregivers, and staff. Integration focused on areas of convergence and discrepancy (e.g., recall vs documentation) and on structural factors shaping discharge quality. Results are organized into four domains: (1) communication and documentation, (2) HF self-care education, (3) medications, and (4) follow-up coordination. These domains were identified during the analysis and based on patterns identified from both qualitative and quantitative data. Within each domain, we integrate findings from surveys, medical record abstraction, and staff interviews to highlight patient/caregiver experiences, structural barriers, and areas for improvement. Additional details are in the Supplemental Material, Supplemental Method 5.

Results

Enrollment and sample

A total of 472 patients were screened for the study, of whom 176 were enrolled (Figure 1). After post-enrollment exclusions and withdrawals, 150 participants completed the post-discharge survey and had discharge instructions abstracted. Of these surveys, 69 were completed by patients and 81 by caregiver proxies. All participants were admitted to the SNF following a hospitalization in which heart failure was documented as a primary or significant contributing diagnosis.

Figure 1 -. Flow Diagram of Enrolled Patients:

A total of 472 patients met initial screening criteria: age ≥65 years, Medicare beneficiary status, and planned discharge to home after a skilled nursing facility (SNF) stay. Of these, 296 patients were not enrolled due to inability to reach after discharge (101), active or passive declination (187), or lack of an English-speaking legally authorized representative (8).

Among the 176 enrolled patients, 15 were excluded post-enrollment for meeting exclusion criteria (SNF stay >60 days, discharge against medical advice, or hospital readmission from SNF for >7 days), and two rescinded consent. Nine participants did not complete the post-discharge survey.

Twenty staff from three SNFs participated in semi-structured interviews (average 43 minutes, range 24–60 minutes). Interviewees represented diverse professional roles, including nurses, social workers, case managers, physicians, nurse practitioners, and an administrator; the distribution of roles and facilities is detailed in Table S2.

Patient baseline characteristics are shown in Table 1. Patients were older adults (mean age 83.7±8.3 years), and 53% were female. Common comorbidities included kidney disease (49%) and diabetes (42%). The mean FRAIL-NH score was 4.0±2.5, consistent with mild to moderate frailty in post-acute nursing home populations. Cognitive impairment, assessed using the Cognitive Function Scale, was present in 26% of participants (16% mild, 8% moderate). Approximately one-third (35%) lived alone.

Table 1.

Characteristics of Enrolled Patients (N=150)

Characteristic	Value
Demographics
Age, years, mean ± SD	83.7 ± 8.3
Female sex, n (%)	79 (52.7)
Insurance, n (%):
Medicaid*	30 (20.0)
Race/ethnicity, n (%) †
White	90 (60.0)
Black	32 (22.0)
Hispanic	11 (7.3)
Asian	11 (7.3)
Other	3 (2.0)
Refused to answer	6 (4.0)
Education, n (%)
Less than high school	13 (8.7)
High school graduate/GED	29 (19.3)
College/vocational school	57 (38.0)
Graduate school	44 (29.3)
Refused to answer	7 (4.7)
Household income, n (%) ‡
<$20,000	23 (15.3)
$20,000-$50,000	30 (20.0)
$50,001-$80,000	21 (14.0)
$80,001-$100,000	9 (6.0)
>$100,000	24 (16.0)
Refused to answer/Don’t know	42 (28.0)
Household composition, n (%)
Lives alone	52 (34.7%)
Lives with 1 other person	56 (37.3%)
Lives with 2 other persons	26 (17.3%)
Lives with 3+ other persons	10 (6.7%)
Lives in Assisted Living Facility	5 (3.3%)
Refused to answer	1 (0.7%)
Marital status, n (%)
Married	48 (32.0)
Widowed	39 (26.0)
Single	36 (24.0)
Divorced	18 (12.0)
Separated	5 (3.3)
Living with partner	2 (1.3)
Refused to answer	2 (1.3)
Clinical characteristics & Comorbidities, n (%)
COVID-19 positive, n (%)||	9 (6.0)
Diabetes	63 (42.0)
Kidney disease	73 (48.7)
Cancer	47 (31.3)
Cirrhosis	2 (1.3)
Level of Frailty
FRAIL-NH score, mean ± SD#	4.0 ± 2.5
Cognitive impairment, n (%) **
Intact	110 (73.3)
Mild	24 (16.0)
Moderate	12 (8.0)
Food insecurity, n (%) ††
Food insecurity present	9 (6.0)
No food insecurity	137 (91.3)
Refused to answer	4 (2.7)
Housing stability, n (%)
Stable housing expected	142 (94.7)
Housing concerns	3 (2.0)
Uncertain	3 (2.0)
Refused to answer	2 (1.3)

^*All participants had Medicare (inclusion criterion); percentages reflect those with dual Medicare-Medicaid coverage.

^†Participants could select more than one race, so percentages add up to >100%.

^‡Missing data: n=1

^||COVID-19 data collection initiated May 2020; missing data for participants enrolled prior to this date: n=18 (12.0%).

^#Frailty scores range from 0–13, with higher scores indicating greater frailty. Missing data: n=6 (4.0%).

^**Cognitive function categories based on Cognitive Function Scale derived from Brief Interview for Mental Status (BIMS) score, or Cognitive Performance Scale (CPS) score when BIMS unavailable. Missing data: n=4 (2.7%).

^††Food insecurity defined as responding “often true” or “sometimes true” to either food security question.

Communication and documentation

Of 150 survey respondents, 88 (59%) reported receiving written discharge instructions; among these, 76 (92%) found them easy to read and understand, although only 59 (71%) reported having the opportunity to ask questions about them. Discharge instructions varied in content (see next section, HF self-care education); when the reason for hospitalization was documented, only 26 (17%) used plain language (see Supplemental Material, Supplemental Method 4, for details on how this was assessed).

Patients and caregivers reported variable staff involvement in discharge preparation. Social workers were most frequently cited as assisting with discharge preparation (67%), followed by nurses (50%) and doctors (36%) (Table S3a). Physical and occupational therapists were also noted to provide guidance during the discharge process (Table S3b).

Most respondents (105, 70%) reported receiving several days' notice before SNF discharge, slightly fewer (99, 66%) felt their preferences were considered in discharge planning. Open-ended survey responses frequently cited communication problems (n=67), including inadequate discharge education, medication miscommunication, and limited communication with the primary caregiver (Table S4). One respondent [ID 79, caregiver] shared that the social worker “wasn’t aware of the discharge and didn’t have any information,” adding that the family felt “as if they did all the work.” Staff interviews echoed these experiences, emphasizing role ambiguity, non-standardized communication, and limited time: “I think it’s too busy, and I don’t think someone physically takes the time to tell the person, ‘Hey make sure you have a blood pressure monitor at home…’” [ID 2, social worker]

HF self-care education

Staff reported that variability in discharge communication extended to HF self-care education; written instructions were often generic and templated, with few HF-specific action plans. A physician [ID 65] noted, “The discharge instructions don’t really address anything heart failure–specific… they list problems and medications and follow-ups, but not lifestyle-management instructions, diet, or anything.” Instead, some staff described using their own discretion to provide individualized teaching at discharge, including written materials they felt were more effective, but noted that this was inconsistently delivered table).

Consistent with these reports, patient and caregiver perspectives indicated gaps in HF-specific guidance. In open-ended survey responses, 49 respondents reported receiving no heart care instructions (Table S5). One respondent [ID 188, a caregiver who is also a cardiac nurse] stated, “They didn’t tell me about daily weights… [or] sodium-restricted diet… signs and symptoms of fluid overload, when to call the doctor.” Across core elements, patient and caregiver recall exceeded what was documented in written discharge instructions (Table 2). For example, daily weight monitoring was recalled by 57 (38%) but documented for 23 (15%); and written guidance to call for symptoms consistent with volume overload appeared in only 8% of discharge instructions. Similarly, exercise/activity recommendations were recalled by 91 (61%) versus 18 (12%) documented. Dietary guidance showed closer agreement, with 69 (46%) recalling low-salt advice versus 62 (41%) documented.

Table 2:

Heart Failure Self-Care Instructions – Triangulation of Survey, Discharge Documentation, & Qualitative Data

Self-Care Instruction	Patient/Caregiver Reported, n (%)	Documented in Discharge Instructions, n (%)	Qualitative Data
Monitoring Instructions
Monitor symptoms/health problems	70 (46.7%)	–†	“The discharge instructions don’t really address anything heart failure–specific… they list problems and medications and follow-ups, but not lifestyle-management instructions, diet, or anything.” [ID 65, physician] “For me, that’s different than other social workers, that sets me apart, is that I review this booklet with them. It’s our cardiopulmonary rehabilitation education booklet. I would consider this as a part of my discharge process because one of the main things I want to make sure they understand is exactly what congestive heart failure is…This booklet is very helpful in that. It has weight tables that you can log your weight. It encourages you to do that daily, for example, so that you can track your weight. It also has a really nice list of healthy foods” [ID 1, social worker] “They didn’t tell me about daily weights… [or] sodium-restricted diet… signs and symptoms of fluid overload, when to call the doctor.” [ID 188, caregiver] “I’m not sure if there’s anything handwritten or printed materials…there may be advice given at the end.” [ID 43, physician] “I say to monitor your symptoms. If you feel like you are using [or] requiring more pills, or you are seeing that you have more leg swelling, then you should follow-up with your doctor. Maybe you need to change your medications.” [ID 53, physician] “[On the discharge instructions] it’s mainly the weight and also the symptoms, like if they're feeling shortness of breath or chest pain…I think it’s included. I don’t remember. But these are the symptoms we do go over, like, read them. Sometimes it's hard for patients to remember all these… If a patient [has] cognitive impairment, we do tell the families, too…These are the symptoms to expect and whatnot. [ID 22, physician] “They have this thing called the swag bag…I think it comes with a scale. It comes with a blood pressure machine. It comes with a booklet, healthier food choices.” [ID 14, nurse]
Monitor for ankle swelling	61 (40.7%)	–†
Weight monitoring	57 (38.3%)	23 (15.3%)
Dietary Recommendations
Low salt diet	69 (46.3%)	62 (41.3%)	“I got several pieces of paper…on the paper [were] some dietary restrictions like no salt. And that was kind of it, and I mean, I read that portion…nobody went over it with me.” [ID 152, patient] “I think the dietitians do a very good job of talking about diet.” [ID 8, nurse practitioner] “To be honest, it’s hard to tell them, ‘cause it’s printed. I know it’s low sodium, but…I know they’re not gonna follow it at home. They’re gonna go back to the regular food they eat.” [ID 16, nurse]
Ask for low salt when eating out	38 (25.7%)	–†
Limit fluid intake	–*	16 (10.7%)
Activity and Lifestyle
Exercise recommendations	91 (61.1%)	18 (12.0%)	“[I received] some general suggestions about continuing physical therapy and occupational therapy” (ID 115, patient] “Do you have any desire to stop smoking? We can always help you. If they don’t want the patch, and they’re very—’cause I know a lotta times, some doctors use Wellbutrin to help them through for smoking cessation…‘Do you want us to talk to the doctor about how to stop the smoking?’ Trust me, 90 percent of the time, they’ll tell you no.” [ID 3, nurse case manager] “I try to get to patients while they’re here to quit smoking if they do smoke.” [ID 6, physician]
Smoking cessation guidance	12 (8.0%)‡	0 (0%)
Avoid getting sick (flu shot, avoid ill people)	41 (27.3%)	0 (0%)
Emergency Instructions
Who to call if symptoms worsen	69 (46.0%)	–†	“If anything worsens, if they’re feeling short of breath or anything, then they should call 911 or call the doctor.” [ID 53, physician] “I was told to take daily weights and notify a doctor if weight gain was more than thresholds stated; I was also told to take daily blood pressure and limit fluid intake.” [ID 169, caregiver] “If you are gaining weight, call your doctor, this many pounds, more than three to four pounds and you are having symptoms to tell—and avoid salty diet. We do tell all the patients throughout their stay and also on their discharge.” [ID 22, physician]
Call doctor for weight gain	–*	13 (8.7%)
Call doctor for volume overload symptoms	–*	12 (8.0%)
Call doctor for chest pain/palpitations	–*	0 (0%)

n=150 for all analyses

^*Specific survey question not asked; data available only from medical record abstraction.

^†Specific data not abstracted; data available only from survey.

^‡Smoking cessation guidance was assessed in all survey respondents (n=150). Not all patients were active smokers at the time of discharge; percentages reflect the full sample.

Medications

Gaps in communication and documentation were also noted with the management of medications during the SNF-to-home transition. Though most respondents reported receiving a medication list (131, 87%), only 79 (53%) had this list reflected in the discharge instructions documented in the EHR (Table 3). Medication changes were common: 65 (43%) respondents reported new prescriptions, 42 (28%) reported dose or frequency changes, and 29 (19%) reported discontinuations. Recommendations for adherence support strategies, such as pillboxes or reminders, were infrequently addressed—only 35 respondents (24%) recalled receiving such recommendations, and none were documented in written instructions.

Table 3:

Medication Management - Triangulation of Survey, Discharge Documentation, & Qualitative Data

Medication Component	Patient/Caregiver Reported, n (%)	Documented in Discharge Instructions, n (%)	Qualitative Data
Medication Lists and Education
Received medication list	131 (87.3%)	79 (52.7%)	“She was sent home with medications with no instructions.” [ID 104, caregiver] “It’s a part of our discharge process that a nurse reviews those meds with them and goes over the meds list.” [ID 1, social worker] “When I prepare the discharge summary or they’re getting ready for discharge, I do go through their medications. Often, they’re discharged on the medications that they were getting during their stay.” [ID 65, physician] “At discharge, the providers went over everything. They told her to follow the medications. They made sure to pass on the medication instructions so she could understand them, including her diabetic medication as well.” (ID 81, caregiver)
Understood medication instructions	124 (83.2%)	–*
Told to use medication reminders/pillbox	35 (23.8%)	0 (0%)
Medication Changes
Any medication changes	78 (52.0%)	–†	“We were quite confused on which medications to take. There was so much importance put on the doctors finding the correct dosage of each medication and when we went home, it was not very clear for us at all.” [ID132, caregiver] “We really don’t say, ‘You stopped taking this medication while you were here, so do not continue that medication when you go home.’ I do tell them to check their medications on the list versus what they actually have at home, so they don’t take anything that’s not on the list.” [ID 8, nurse practitioner]
New medications prescribed	65 (43.3%)	–†
Dose/frequency changes	42 (28.0%)	–†
Medications discontinued	29 (19.3%)	–†

^*Specific data not abstracted; data available only from survey.

^†Medications were abstracted from the medical record; however, detailed abstraction results are beyond the scope of this analysis and are not presented here.

Most respondents (124, 83%) reported understanding the medication instructions; however, 21 (14%) reported low or no confidence in understanding which new medications to start and which prior medications to discontinue, and 24 (16%) reported low or no confidence in understanding the purpose of each medication (Table S6). Open-ended survey responses described gaps in medication communication (Table 3, Table S4). One caregiver [ID 104] noted, “She was sent home with medications with no instructions,” while another [ID 103, caregiver] described, “His medications were not available on the day of discharge because he was discharged on a Sunday and the pharmacy was closed.”

In staff interviews, they emphasized that medications were a discharge priority, with prescriptions sent to the patient’s pharmacy and nurses reviewing lists at discharge. Staff also noted that explicit stop/start guidance was inconsistently delivered: “We really don’t say, ‘You stopped taking this medication while you were here, so do not continue that medication when you go home.’ I do tell them to check their medications on the list versus what they actually have at home, so they don’t take anything that’s not on the list.” [ID 8, nurse practitioner]. Staff interviews also provided context for medication-related logistical challenges experienced by patients/families on discharge. “Home care was able to help us to make sure that the medication got to the patient, but she was discharged close to the weekend, so they didn’t get there until the following week…that was a failure in terms of how we communicated with the patient” [Subject ID 52, nurse practitioner].

Follow-up coordination and post-discharge services

Organization of post-SNF follow-up appointments was variable. Of the patients and caregivers surveyed, 55 (37%) reported that a follow-up appointment with a PCP had been scheduled prior to discharge, yet such appointments were documented in only 20 (13%) of discharge instructions (Table 4). Among the documented PCP appointments, 8 were scheduled within 7 days of discharge. Cardiology follow-up was documented in 28 (19%) discharge instructions. Of the 95 patients who did not have follow-up arranged, 44 (46%) respondents reported scheduling it independently. Less than half of respondents (69, 46%) reported they were told whom to call if the patient’s symptoms worsened; similarly, a phone number to call with problems was only included in 71 (47%) of discharge instructions.

Table 4:

Outpatient Follow-up - Triangulation of Survey, Discharge Documentation, & Qualitative Data

Follow-up Component	Patient/Caregiver Reported, n (%)	Documented in Discharge Instructions, n (%)	Qualitative Data
Primary Care Follow-up
Has a primary care physician	140 (93.3%)	–†	“Outpatient provider, honestly, we don’t do anything. Ideally, we should have—at least fax the discharge summary or something. All we say is, ‘Okay, take this discharge summary with you to see—when you go see your doctor.’ That’s all we say. There’s no communication.” [ID 15, nurse practitioner] “Prior to their discharge, we reach out to their primary care doctors….just make sure they have appointments for follow up, for their cardiologist, primary care.” [ID 52, nurse practitioner] “We just tell the resident, “You need to follow-up with your primary care doctor.” We don't give the primary care doctor a call telling them that they left already.” [ID 50, social worker] “With the heart-failure patients, we do tend to be more aggressive about making sure that they have a set follow-up appointment in terms of a day and a time because with some of the other patients who might not be already be set up, we will tell them, “Make sure you get an appointment with your PCP.” For heart failure and…other [fragile] patients …we ensure that they have an appointment before they leave, usually with PCP and with cardiology.” [ID 6, physician] “We do PCP follow-ups within a 30-day window, because we e-script everybody home with a 30-day supply of medications. What we tell them with the anticipation is that within those 30 days, you have to start following up with your doctors, because once your 30 days are done, you cannot come back to us for another set of prescriptions. [ID 5, social worker]
PCP aware of SNF stay	95 (67.8%)*	–†
PCP appointment scheduled before discharge	55 (36.7%)	20 (13.3%)
PCP appointment within 7 days	–‡	8 (40.0%)||
Specialty Follow-up
Other appointments scheduled	61 (40.7%)	–†	“Some of them do come with the cardiologist appointment already. If they have a cardiologist appointment coming pretty soon, we leave it. We do make a cardiology appointment for the heart-failure patients, especially. Most of them are between 7 to 10 days, especially the brittle heart-failure patients. Some of them are stable heart-failure patients that [follow-up] within two weeks. It's not their main symptom that they were admitted, but they do have the follow-up appointment.” [ID 22, physician] “I was informed how to keep in touch with doctors, and was given a call to set up future appointments.” [ID 19, caregiver]
Cardiology appointment scheduled	–‡	28 (18.7%)
Home Services
Visiting nurse arranged	–‡	97 (64.7%)	“I was told to exercise every day, got a part-time nurse to come to my house, a home attendant to help, and to continue seeing my primary care physician. I was also told to make an appointment with my cardiologist.” (ID 171, patient) “…we always ask the resident… ‘do you want in-home or do you wanna do outpatient?’ I would say most opt for in-home.” [ID 5, social worker] “Our biggest resource…is the home care…makin’ an appointment and tellin’ somebody to go to an appointment, you have a 40 percent chance, 30 to 40 percent chance of them doin’ it if they wanna do it. If you refer them to home care, that’s why most of the time we refer them to the home care, at least you got a nurse that’s gonna be going to the house.” [ID 3, nurse case manager]
Home health aide arranged	–‡	81 (54.0%)
Physical therapy arranged	–‡	102 (68.0%)
Other home services arranged	–‡	93 (62.4%)#
Additional Services
Outpatient tests scheduled	24 (16.0%)	–†	“They get a follow-up appointment with instructions to bring their discharge summary to their follow-up appointment.” [ID 41, administrator] “We tell them, ‘You have to see within the week,’ especially for those who are on Coumadin.” [ID 14, nurse]
Told to keep appointments	104 (69.8%)	–†
Contact Information	69 (46.0%)**	71 (47.3%)††

n=150 for all calculations unless otherwise noted

^*Among those with a PCP (n=140).

^†Specific data not abstracted; data available only from survey.

^‡Specific survey question not asked; data available only from medical record abstraction.

^||Among those with documented PCP appointments (n=20).

^#Occupational therapy was most often listed as the other home services scheduled.

^**Corresponds to survey question “Were you told who to call if your symptoms got worse”?

^††Corresponds to medical abstraction item “Is there a specific phone number to call with problems listed anywhere in the discharge?”

Home services arrangements were often documented in discharge instructions (visiting nurse: 65%, home health aide: 54%, home physical therapy: 68%, other services—primarily occupational therapy—62%), yet respondents still described gaps in durable medical equipment setup and transportation planning (Table S4). A respondent [ID 76, caregiver] described, “…wasn’t given a working wheelchair at discharge…they were supposed to help set a ride home, which they didn’t.”

Communication across care settings was limited. Among the 140 patients with a primary care clinician, 95 (68%) of respondents reported that the PCP was aware of the patient’s SNF stay. Among those who reported that the PCP was aware of the SNF stay, 53 respondents indicated that the patient or family member informed the PCP. Medical record abstraction identified only one discharge summary that was documented as sent to an outpatient clinician. Staff perspectives reflected these experiences, as one nurse practitioner summarized, “We don’t do anything… There’s no communication.” [ID 15]

Discussion

In this convergent mixed-methods study, we identify meaningful safety gaps in SNF-to-home transitions for older adults recovering after HF hospitalization. Although many patients and caregivers recalled receiving verbal discharge education, written discharge instructions frequently lacked HF-specific self-care components, and documentation of follow-up arrangements and communication with outpatient clinicians was limited. By triangulating patient and caregiver reports, discharge instructions, and staff perspectives, we demonstrate discrepancies between perceived and documented discharge practices—gaps that may not be apparent when relying on a single data source. Given the central role of volume status monitoring, medication continuity, and coordinated follow-up in preventing HF decompensation, these gaps represent clinically meaningful vulnerabilities during a high-risk period of recovery.^11,34–36

The multidisciplinary nature of discharge planning in SNFs presents unique structural challenges. Our data demonstrate that discharge responsibilities were distributed across multiple team members without clear leadership, which may contribute to the communication gaps we observed in key components of HF transitional care. These structural factors—combined with staffing constraints, including fewer licensed nursing staff, higher patient-to-staff ratios, and more limited onsite physician and nurse practitioner coverage compared to hospitals³⁷—as well as workforce instability^38,39 —may impair consistent implementation of HF discharge processes that depend on clear ownership, patient continuity, and patient/family rapport. These findings are particularly relevant in light of the recent federal rescission of minimum staffing requirements for SNFs.⁴⁰

While prior work has focused primarily on hospital-to-home transitions in HF, comparatively little empirical research has examined the SNF-to-home transition as a distinct phase of care. Patients discharged from SNFs after HF hospitalization are often medically complex, with multimorbidity, frailty, and cognitive impairment that may impair their ability to self-manage HF after returning home. In this context, reliance on verbal education, incomplete documentation of HF self-care instructions, and inconsistent communication about medication changes may increase the risk of fluid overload or other complications that commonly precipitate early readmission.

To compensate for perceived gaps, some staff provided supplemental materials, such as educational booklets, tailored to patient needs; however, this practice was inconsistent and varied depending on the staff member handling the discharge. These variations may partially explain the discrepancies between patient and caregiver recall and documented discharge instructions. While some patients may have received additional information, this was not reflected in the EHR, contributing to variability in discharge education.

While sustained workforce investments are critical for long-term improvements, immediate interventions could focus on standardizing SNF discharge planning using SNF transitional care frameworks, such as Connect-Home,⁴¹ and explicitly tailoring them for HF-specific risks. Specifically, we propose a low-burden, standardized HF-transition bundle for patients transitioning from SNF to home that includes an individualized self-care plan with explicit diuretic instructions, monitoring guidance with clear action thresholds, caregiver engagement, and scheduled follow-up. Existing EHR infrastructure may help support implementation of these practices through HF-specific discharge templates, checklists, and clinical decision-support tools that prompt completion of key transition tasks and facilitate more consistent patient education and documentation. Communication between SNFs and outpatient clinicians could be strengthened through direct, secure electronic transmission of discharge summaries and reconciled medication lists at the time of SNF discharge. Greater interoperability and health information exchange across care settings may facilitate continuity during transitions.⁴² In settings with limited on-site staffing, external supports—such as pharmacist-led medication reconciliation^43,44 and telehealth HF^45,46 education consults—may help ensure that critical transition tasks are completed reliably. Special attention to weekend discharge workflows, including advance prescription fulfillment and early scheduling of home services, may further mitigate predictable operational barriers.

This study has limitations. It focused on Medicare beneficiaries aged ≥65 years discharged home after short post-acute stays (≤60 days) and was conducted in two major cities, limiting generalizability to for-profit facilities, long-stay residents, younger or non-Medicare populations, and non-urban settings. Enrollment was restricted to English-speaking patients or those with an English-speaking caregiver, and many surveys were completed by caregivers, introducing potential selection, recall, and social desirability bias. Although caregivers completed surveys for many participants, we did not conduct caregiver-specific qualitative interviews or analyses. In addition, detailed patient characteristics were collected only after enrollment, informed consent, and completion of the post-discharge survey. Therefore, we were unable to compare enrolled and non-enrolled individuals with respect to frailty, cognitive status, or other patient-level characteristics. Our assessment of discharge instruction quality and cross-setting communication relied on EHR documentation, which may under-capture verbal education or communication not documented in the EHR. Importantly, the participating SNFs were non-profit facilities with established interdisciplinary discharge practices; therefore, findings should be interpreted as reflecting opportunities to enhance documentation and standardization rather than deficiencies in clinical care. In addition, staff interviews did not include physical or occupational therapists, whose perspectives may have provided additional insights into the discharge planning process. Future research should examine how SNF-to-home transitions vary across patient, caregiver, and facility contexts. Larger studies are needed to evaluate whether transition quality differs by patient sociodemographic characteristics, level of frailty, cognitive impairment, or caregiver involvement, and to identify which populations may be most vulnerable to gaps during the transition from SNF to home. Examining how discharge practices vary across facility contexts is another important area for future research.

Conclusion

SNF-to-home transitions for older adults with HF are characterized by predictable, remediable gaps: incomplete discharge communication and documentation, inconsistent HF self-care guidance, and gaps in medication transition planning and follow-up coordination. The discordance between patient and caregiver recall and written documentation underscores reliance on verbal education that is not consistently reinforced across care settings. Structural and workforce constraints, as described by staff, impede consistent, person-centered discharge preparation. In the current policy environment, near-term improvements should prioritize implementation of standardized, low-burden HF transition workflows, while long-term solutions must address workforce capacity to sustainably improve post-acute care transitions.

Supplementary Material

Supplemental Methods 1–5

Tables S1–S6

eAppendices 1–4

References 51–52

What Is New?

• Most research on heart failure transitions has focused on hospital-to-home discharge, despite the large number of older adults who transition from skilled nursing facilities (SNFs) to home after heart failure hospitalization.

• Using a mixed-methods approach that integrated patient and caregiver surveys, discharge documentation, and staff interviews, we identified substantial gaps in heart failure–specific discharge communication, medication transition planning, and follow-up coordination during SNF-to-home transitions.

• We also found discordance between what patients and caregivers recalled receiving and what was documented in the medical record, highlighting limitations of relying on a single source to assess discharge quality.

What Are the Clinical Implications?

• Older adults discharged from SNFs after heart failure hospitalization may return home without consistent written guidance regarding self-care, medication changes, or follow-up plans.

• Standardized heart failure–focused discharge processes, including structured education, medication reconciliation, caregiver engagement, and improved communication with outpatient clinicians, may help address these gaps.

• Existing transitional care frameworks and electronic health record–based tools could be adapted to improve the consistency and reliability of SNF-to-home transitions for patients with heart failure.

Non-standard Abbreviations and Acronyms

CFS

Cognitive Function Scale

CMS

Centers for Medicare & Medicaid Services

EHR

Electronic Health Record

FRAIL-NH

Frailty scale for nursing home residents

GRAMMS

Good Reporting of A Mixed Methods Study

HF

Heart Failure

PCP

Primary Care Provider

SNF

Skilled Nursing Facility