Improving Telehealth Transition of Care Programs Focused on Readmission Reduction

Abstract

Introduction:

Suboptimal transitional care from the hospital to home can result in poor health outcomes, increased costs, and readmissions. Telehealth-based transitional care programs have shown some improvements in readmission rates; however, it is unclear why some patients benefit while others do not. This study evaluated a connected transitional care (CTC) program that provided high-risk patients with timely post-discharge telehealth appointments conducted by a nurse practitioner. Our focus was on understanding why some patients participating in the program benefit and are not readmitted while others are readmitted.

Methods:

We analyzed readmission rates for patients referred to the program and compared those who engaged, by completing a telehealth visit, to those who did not. For those patients who did engage, we conducted chart reviews of a subset of patients who were not readmitted compared with those who were readmitted to extract themes and understand differences that could serve to improve the CTC program.

Results:

Of 1374 patients referred to the CTC program, 443 (32.2%) engaged by completing a telehealth visit. Those who engaged in the program had a readmission rate of 18.7% compared with 21.3% for those who did not, resulting in a relative risk reduction of 12%. Chart reviews comparing patients who engaged and were not readmitted (32 charts reviewed) with those who were readmitted (18 charts reviewed) revealed several differences. Patients who were not readmitted were seen sooner after discharge, had greater family/caregiver involvement, had social needs addressed, required less language interpretation, and had fewer instances of altered mental status.

Conclusions:

This study suggests that a telehealth transition program may reduce readmissions, although a more rigorous statistical analysis is needed. Importantly, the qualitative chart review suggests several areas for improvement, including engaging family/caregivers, providing better social need support, and developing ways to support behavioral health.

Patients transitioning care from a hospital to the home are at a time of heightened risk of adverse outcomes and readmissions.¹ Adverse outcomes and readmissions are also costly and are estimated to account for more than $17 billion in avoidable Medicare expenditures.² Transitional care programs have been developed and implemented to lessen the vulnerability that patients may face and to address exorbitant costs.³

To incentivize new transitional care models and the adoption of transitional care programs, Medicare has offered health care facilities a way to bill for transitional care services (Medicare Transitional Care Management⁴) to promote providing high-risk patients with the care they need immediately after discharge. To further reduce readmissions and restrict payments to hospitals that had excess readmission rates, the Hospital Readmissions Reduction Program (HRRP) was launched in 2012. Since the implementation of the HRRP, numerous interventions to reduce readmissions have been developed by hospitals and health care systems across the United States.

However, the evidence supporting the effectiveness of these care transition models and the effectiveness of the HRRP and its policies on reducing readmissions is mixed.^1,5,6 Some studies cite a reduction in readmissions due to either proactive discharge advocates who provided personalized discharge instruction materials before patient departure or programs featuring telehealth interventions shortly after discharge.^7,8 One such program experienced a significant and sustained reduction in total readmission rates through a nurse-led telehealth-based solution that emphasized increased clinical outreach to high-risk patients, resulting in improved access to follow-up care.⁸ Although telehealth seems like a promising pathway, past studies, including a meta-analysis, have shown that telehealth readmission reduction programs may only benefit certain patients while others see no benefit.^1,9 Further, many of these studies were conducted outside of the United States, and the countries where these studies were conducted have different policies that may impact the services that can be provided to patients.

The current study aims to address the research gap around telehealth transitional care programs by understanding why only certain patients benefit from these programs. Specifically, we used a qualitative chart review approach to understand why some high-risk patients who are referred to and participate in a telehealth-based transitional care program are still readmitted, while others seem to benefit and are not readmitted.

CTC PROGRAM BACKGROUND

Program Description

The CTC program facilitates a safe discharge plan by providing timely postdischarge telehealth appointments with a nurse practitioner for patients unable to meet with their primary care provider (PCP) in a timely manner. It aims to provide continuity of care from the inpatient to outpatient settings.

The CTC program consists of a connected transitional care nurse practitioner (CTCNP) and a connected transitional care medical assistant (CTCMA). The CTCNP reviews the patient’s discharge plan, completes medication reconciliation, amends prescriptions, provides orders and/or referrals to necessary services, completes assessments, and provides guidance for medical concerns. Overall, the CTCNP serves as a bridge to the patient’s primary care physician (PCP). Complementing the role of the CTCNP, the CTCMA manages the referral worklist, contacts patients, schedules appointments with the CTCNP, assists with prior authorization needs, and guides patients to establishing care with a PCP or acquiring an appointment with a specialist.

Patients are connected to the program either through a referral before discharge from an inpatient hospital stay or within a week of discharge from the hospital.

Identification of Patients and Referral to the CTC Program

To determine which patients may benefit from the CTC program most, a readmission prevention solution model evaluates certain criteria of interest. This proprietary, electronic health record–based solution evaluates patient data such as clinical information and previous hospitalization history to establish a risk of readmission. Each patient receives a risk score that quantifies their likelihood of being readmitted and stratifies the patients into 3 groups: low-risk, moderate-risk, and high-risk of readmission. For our analysis, we only evaluated high-risk patients. In addition, to be considered for the program, patients must also: (1) have an impending discharge to home/home health or have recently been discharged in the last week, (2) be unable to visit their PCP in the clinically recommended time window postdischarge, (3) have access to a computer/mobile device and reliable internet, and (4) are willing and able to participate in telehealth (following education). Several members of the care team may be involved in recommending patients to the program, including case management, hospitalists, and/or care transitions nurses.

METHODS

Participants and Inclusion/Exclusion Criteria

To be eligible for inclusion in the analysis, patients must have had (1) a classification of inpatient during their hospital visit, (2) a readmission risk classification of high, (3) had a referral to the CTC program, and (4) have been discharged within the study period of the second quarter of 2022 through the third quarter of 2024. Patients who were referred to hospice and/or dying within 30 days of discharge were excluded from this analysis. The sample population is that of 2 suburban hospitals within a large mid-Atlantic health care system. Analyses are based on a readmission period of 30 days.

Study Design

The overall analysis included a 2-pronged approach. First, we examined readmission rates for those who were referred and engaged in the CTC program compared with those who were referred and did not engage in the program. We utilized the total number of high-risk patients referred to the program (1374) within our study period as the denominator for analyses and calculated rates based on 30-day readmissions.

Second, we sought to understand factors that may contribute to patients participating in the program and still being readmitted compared with those who participate and are not readmitted. To do this, we randomly selected and reviewed the NP notes of 50 high-risk patients who were referred to the CTC program and participated with readmission and with no readmission. The inclusion criteria for this analysis remained the same as above. All charts reviewed were those of high-risk patients who were referred to the CTC program.

The patient charts were evaluated by a registered nurse who is a patient safety and human factors subject matter expert. Each chart review was based on a pre-established template that provided a clear process for review and promoted consistency in each chart review. The review template consisted of 5 main sections: intervention, safety risks, follow-up order, referrals, and readmission/no-readmission details. The Intervention section involved evaluating any interventions that took place, such as but not limited to the CTC visit itself, medication reconciliation, and the administration of interpretive services. Any Safety Risks, such as fall risks, oxygen needs, mobility challenges, and use of medical equipment, were noted. The Follow-Up section reviewed whether specialists needed to be consulted and if there was a home health plan in place. Referrals and their associated details were also evaluated, and lastly, whether or not the patient was readmitted and the details supporting any Readmission were reviewed. Each chart evaluation was synthesized into themes and presented in the results section. The reviewer was not blinded to patient participation in the CTC program or patient readmission status. This study was approved by the Institutional Review Board.

RESULTS

Preliminary Analysis of Program Effectiveness and Patient Demographics

Although a comprehensive statistical analysis was not conducted, the rates of readmission across different groups were compared. Figure 1 shows the number of patients who completed a CTCNP visit and whether they were readmitted. Table 1 shows the patient demographics for patients referred to the program and for those patients who completed and did not complete a CTCN visit segmented by readmission status.

FIGURE 1.

Patient flow diagram of study period sample—high-risk patients.

TABLE 1.

Demographics for Patients Referred to the CTC Program

Characteristic	Overall (%)*	Order, visit, no readmission†	Order, visit, readmission†	Order, no visit, no readmission‡	Order, no visit, readmission‡
Total	1374 (100)	360	83	732	199
Sex
Female	682 (49.6% of 1374)	178 (49.4% of 360)	50 (60.2% of 83)	353 (51.8% of 732)	101 (50.8% of 199)
Male	692 (50.3%)	182 (50.6%)	33 (39.8%)	379 (51.8%)	98 (49.2%)
Age
29 and under	12 (<1)	1 (0.3%)	1 (1.2%)	8 (1.1%)	2 (1.0%)
30-39	53 (3.8)	9 (2.5%)	2 (2.4%)	28 (3.8%)	14 (7.0%)
40-49	77 (7.2%)	23 (6.4%)	3 (2.6%)	41 (5.6%)	10 (5.0%)
50-59	187 (13.6)	57 (15.8%)	11 (13.3%)	90 (12.3%)	29 (14.6%)
60-69	409 (29.7)	115 (31.9%)	26 (31.3%)	212 (29.0%)	56 (28.1%)
70-79	393 (28.6)	96 (26.7%)	24 (28.9%)	214 (29.2%)	59 (29.6%)
80-89	182 (13.2)	41 (11.4%)	15 (18.1%)	104 (14.2%)	22 (11.1%)
90-99	60 (4.3)	18 (5.0%)	1 (1.2%)	34 (4.6%)	7 (3.5%)
99 and over	1 (<1)	0 (0.0%)	0 (0.0%)	1 (0.1%)	0 (0.0%)
Race
White	350 (25.4)	80 (22.2%)	23 (27.7%)	183 (25.0%)	64 (32.2%)
Black or African American	976 (71)	261 (72.5%)	55 (66.3%)	526 (71.9%)	134 (67.3%)
Other	43 (3.1)	14 (3.9%)	5 (6.0%)	20 (2.7%)	1 (0.5%)
Unknown	5 (<1)	3 (0.8%)	0 (0.0%)	2 (0.3%)	0 (0.0%)
Ethnicity
Hispanic/Latino	22 (1.6)	9 (2.5%)	5 (6.0%)	10 (1.4%)	2 (1.0%)
Not Hispanic/Latino	1347 (98)	339 (94.2%)	75 (90.4%)	704 (96.2%)	195 (98.0%)
Unknown	5 (<1)	12 (3.3%)	3 (3.6%)	18 (2.5%)	2 (1.0%)

^*Denotes the total number of patients referred to the program within the study period.

^†Denotes the total number of patients who were referred to the program, completed a visit with a CTCNP, and their outcome.

^‡Denotes the total number of patients who were referred to the program, did not complete a visit with a CTCNP, and their outcome.

To establish a baseline of whether there is an association between participation in the program and readmissions, patient totals from both groups (participated versus did not participate) were recorded, and their respective percentages were calculated. Overall, 1374 (42%) out of a total of 3273 high-risk patients were referred to the program within our study period. Of these referred patients, 32.2% (443 of 1374) completed a telehealth appointment with the CTCNP. Of the 32% who completed an appointment, 18.7% (83 of 443) were readmitted. Comparatively, 67.8% (931 of 1374) of patients referred to the program did not complete an appointment with the CTCNP, and those patients experienced a slightly higher readmission rate of 21.3% (199 of 931).

Chart Review Themes Associated With Patients Completing the CTCNP and Readmission

Charts from 32 patients who completed the CTCNP visit and were not readmitted were compared with charts reviewed from 18 patients who completed the visit and were readmitted. Table 2 provides a summary of the themes identified from the 2 populations. Patients who were not readmitted had a fewer average number of days between referral and first contact compared with those who were readmitted (5.0 versus 5.9 d). Several social/economic themes emerged including patients who were not readmitted having greater family/caregiver involvement (34.4% versus 22.2%), having social needs addressed (25% versus 16.7%), did not require language translation (0% versus 16.7%), and did not have altered mental status (0% versus 33.3%). One medication theme was noted, which was that patients who were not readmitted had a slightly higher rate of taking their medications that were prescribed at discharge compared to those who were readmitted (93.8% versus 88.9%).

TABLE 2.

Frequency of Themes Found in Chart Reviews

	Order—visit—no readmission (% of total)	Order—visit—readmission (% of total)
Total	32	18
Avg. # of days between CTC order and first contact	5.0	5.9
Social/economic themes
Family/caregiver involvement	11 (34.4)	4 (22.2)
Social needs addressed	8 (25)	3 (16.7)
Language interpreter needed	0 (0)	3 (16.7)
Altered mental status noted	0 (0)	6 (33.3)
Medication themes
NP supported prescription/refills	5 (15.6)	3 (16.7)
NP discusses med compliance	32 (100)	18 (100)
Patient taking meds prescribed at discharge	30 (93.8)	16 (88.9)

DISCUSSION

Transitional care programs are thought to ease the transition for patients from one care setting to another. These interventions are designed to prevent poor outcomes due to substandard care coordination and, in turn, avoid readmission. While the design of these programs may vary in their inclusion of telehealth or other care modalities, the intended outcomes remain the same. In our descriptive review of the CTC program, we found participation in the CTC to be associated with reduced readmissions, although a more rigorous statistical analysis is needed. Importantly, our chart review of a subset of patients who participated in the CTC program and were not readmitted compared with those who were readmitted revealed several interesting themes. Most of these themes were generally categorized as social/economic themes. Surprisingly, medication themes were not that different between the 2 groups. Based on these themes, we have several recommendations for improving telehealth-based transitional care programs.

Prioritize Timely Follow-up After Discharge

Our data show that patients who were not readmitted tended to have an average of 5 days between discharge and their CTCNP appointment compared with 5.9 days for those who were readmitted. The facilitation of early and effective communication and the fostering of trust among the entire care team have been recognized as important features of transitional care processes and addressing the unique needs of their patients.¹⁰ Once communication is established, addressing problems related to medications, patient education, and discharge instructions while also engaging family caregivers becomes more easily accessible.

Increase Family/Caregiver Involvement

Our chart review suggests family/caregiver involvement may be important for reduced readmissions. CTC programs should focus on engaging family/caregivers. In the absence of family/caregivers, programs can attempt to engage the patient to a greater degree by using the patient activation measure (PAM).¹¹ The creators cite factors such as self-management of symptoms and collaboration with providers as items that are likely to improve health outcomes and are reflected in the scale. Use of scales such as PAM can act as a method to engage with patients on a deeper level when the use of family caregivers is not possible.¹¹

Include Processes to Address Social, Language, and Behavioral Needs¹²

Our data show that patients who were readmitted may not have had their social needs addressed, required language interpretation, and had higher rates of altered mental status compared with those who were not readmitted. These factors should be addressed through additional services that can either be provided during the NP visit or planned for after the initial visit. One challenge is finding and aligning patients with experts and community-based programs that may be able to provide these services. Establishing these connections as part of a CTC program and facilitating connections between these services and the patient and/or caregiver is critical.¹¹

Interestingly, only 30% of high-risk patients referred for the intervention successfully received it; low adherence rates are a common finding among postdischarge interventions. Identifying the root causes for patient declines or no-shows in this program would ensure more patients are able to receive the intervention and thus benefit from safer discharges and fewer hospital returns. Interviewing patients who participated and did not participate in the CTC program is an important next step.

Lastly, continuity of care is essential for better patient outcomes.^13,14 Primary care physicians are uniquely positioned to fully understand the needs of the patient and serve as a point of contact for specialists. Patients who were discharged from a hospital that is closely associated with their primary care provider were more likely to receive timely follow-up care than those discharged from other hospitals within and outside their health system.¹⁵ Thus, it is imperative that CTC programs facilitate connections to primary care and not serve as a replacement for primary care.

Limitations

There are limitations to our study. We examined percentages of readmission versus no readmission for those who participated and did not participate. A more rigorous statistical analysis is needed to show program effectiveness. Our analysis compared patients receiving the intervention to those referred but who did not engage. The patients who were referred but did not engage may be systematically different in risk, leading to a biased result that shows greater program effectiveness. This limitation could be mitigated by a randomized trial design. We reviewed a limited sample of charts from high-risk readmission patients, and additional chart reviews of either high-risk patients or other groups, such as low- and moderate-risk, may have revealed other themes of interest. In addition, at the time of evaluation, the CTCNP conducted only 1 telehealth follow-up visit and was not able to ensure that patients followed through with verified appointments. In turn, the chart review did not include details on the outcome of these subsequent appointments, which may have included useful information. The chart reviewer examined only narrative notes from the telehealth visit, meaning the reviews were limited to what the CTCNP deemed important enough to record. Finally, the chart reviewer was not blinded to whether the patient was referred to the CTC program, engaged in the program, and readmission status, which could lead to biased assessments.

CONCLUSIONS

Although preliminary, our results suggest that a telehealth care transition program may reduce readmissions. Several factors may be contributing to patients participating in the program and still being readmitted, including a lack of family/caregiver involvement, social needs not being addressed, requiring language interpretation, and having an altered mental status. These insights provide opportunities to redesign care transition programs to address these needs so that more patients can benefit from such programs.