Effects of Different Transitional Care Strategies on Outcomes after Hospital Discharge—Trust Matters, Too

Open AccessPublished:October 19, 2021DOI:https://doi.org/10.1016/j.jcjq.2021.09.012

      Abstract

      Background

      As health systems shift toward value-based care, strategies to reduce readmissions and improve patient outcomes become increasingly important. Despite extensive research, the combinations of transitional care (TC) strategies associated with best patient-centered outcomes remain uncertain.

      Methods

      Using an observational, prospective cohort study design, Project ACHIEVE sought to determine the association of different combinations of TC strategies with patient-reported and postdischarge health care utilization outcomes. Using purposive sampling, the research team recruited a diverse sample of short-term acute care and critical access hospitals in the United States (N = 42) and analyzed data on eligible Medicare beneficiaries (N = 7,939) discharged from their medical/surgical units. Using both hospital- and patient-reported TC strategy exposure data, the project compared patients “exposed” to each of five overlapping groups of TC strategies to their “control” counterparts. Primary outcomes included 30-day hospital readmissions, 7-day postdischarge emergency department (ED) visits and patient-reported physical and mental health, pain, and participation in daily activities.

      Results

      Participants averaged 72.3 years old (standard deviation =10.1), 53.4% were female, and most were White (78.9%). Patients exposed to one TC group (Hospital-Based Trust, Plain Language, and Coordination) were less likely to have 30-day readmissions (risk ratio [RR], 0.72; 95% confidence interval [CI] = 0.57–0.92, p < 0.001) or 7-day ED visits (RR, 0.72; 95% CI, 0.55–0.93, p < 0.001) and more likely to report excellent physical and mental health, greater participation in daily activities, and less pain (RR ranged from 1.11 to 1.15, p < 0.01).

      Conclusion

      In concert with care coordination activities that bridge the transition from hospital to home, hospitals’ clear communication and fostering of trust with patients were associated with better patient-reported outcomes and reduced health care utilization.

      Keywords

      Hospital discharge has been described as “patient safety's ‘perfect storm,’”
      • Clancy CM.
      Reengineering hospital discharge: a protocol to improve patient safety, reduce costs, and boost patient satisfaction.
      (p. 344) and a patient safety leader once noted that “nobody is responsible for coordinating care.”
      Kaiser Health News
      Health Care's ‘Dirty Little Secret’: No One May Be Coordinating Care.
      Care transitions, particularly from the hospital to home or post-acute care, represent a vulnerable period for acutely ill patients and their family caregivers,
      • Greenwald JL
      • Denham CR
      • Jack BW.
      The hospital discharge: a review of a high risk care transition with highlights of a reengineered discharge process.
      • Coleman EA.
      Falling through the cracks: challenges and opportunities for improving transitional care for persons with continuous complex care needs.
      • Mitchell SE
      • et al.
      Care Transitions from patient and caregiver perspectives.
      too often characterized by gaps in care and avoidable problems such as worsening symptoms,
      • Coleman EA
      • Williams MV.
      Executing high-quality care transitions: a call to do it right.
      • Kripalani S
      • et al.
      Promoting effective transitions of care at hospital discharge: a review of key issues for hospitalists.
      • Forster AJ
      • et al.
      Adverse events among medical patients after discharge from hospital.
      • Forster AJ
      • et al.
      The incidence and severity of adverse events affecting patients after discharge from the hospital.
      adverse effects from medications,
      • Greenwald JL
      • et al.
      Making inpatient medication reconciliation patient centered, clinically relevant and implementable: a consensus statement on key principles and necessary first steps.
      ,
      • Budnitz DS
      • et al.
      Emergency hospitalizations for adverse drug events in older Americans.
      unaddressed test results,
      • Roy CL
      • et al.
      Patient safety concerns arising from test results that return after hospital discharge.
      failed follow-up testing,
      • Moore C
      • McGinn T
      • Halm E.
      Tying up loose ends: discharging patients with unresolved medical issues.
      rehospitalizations,
      • Kripalani S
      • et al.
      Deficits in communication and information transfer between hospital-based and primary care physicians: implications for patient safety and continuity of care.
      • Jencks SF
      • Williams MV
      • Coleman EA.
      Rehospitalizations among patients in the Medicare fee-for-service program.
      • Mor V
      • et al.
      The revolving door of rehospitalization from skilled nursing facilities.
      and emergency department (ED) visits.
      • Li J
      • Young R
      • Williams MV
      Optimizing transitions of care to reduce rehospitalizations.
      ,
      • Coleman EA
      • et al.
      Preparing patients and caregivers to participate in care delivered across settings: the care transitions intervention.
      Despite increasing emphasis by the Centers for Medicare & Medicaid Services (CMS) on value-based payment initiatives (for example, Accountable Care Organizations [ACOs] and bundled payments), coordinating care remains a major challenge for many health systems.
      Recognition of these issues led to development of multicomponent transitional care (TC) models demonstrating improved care quality and reduced readmissions; for example, the Transitional Care Model,
      • Naylor M
      • et al.
      Comprehensive discharge planning for the hospitalized elderly: a randomized clinical trial.
      Care Transitions Intervention,
      • Coleman EA
      • et al.
      The care transitions intervention: results of a randomized controlled trial.
      Project RED (Re-Engineered Discharge),
      • Jack BW
      • et al.
      A reengineered hospital discharge program to decrease rehospitalization: a randomized trial.
      and Project BOOST (Better Outcomes by Optimizing Safe Transitions).
      • Williams MV
      • et al.
      Project BOOST implementation: lessons learned.
      Systematic reviews and studies of TC reveal that some multifaceted interventions enhance patient outcomes and reduce the risk of 30-day rehospitalizations by 20% to 40%.
      • Coleman EA
      • et al.
      Preparing patients and caregivers to participate in care delivered across settings: the care transitions intervention.
      ,
      • Naylor M
      • et al.
      Comprehensive discharge planning for the hospitalized elderly: a randomized clinical trial.
      ,
      • Jack BW
      • et al.
      A reengineered hospital discharge program to decrease rehospitalization: a randomized trial.
      ,
      • Hansen LO
      • et al.
      Interventions to reduce 30-day rehospitalization: a systematic review.
      • Naylor MD
      • et al.
      Comprehensive discharge planning and home follow-up of hospitalized elders: a randomized clinical trial.
      • Koehler BE
      • et al.
      Reduction of 30-day postdischarge hospital readmission or emergency department (ED) visit rates in high-risk elderly medical patients through delivery of a targeted care bundle.
      • Garåsen H
      • Windspoll R
      • Johnsen R.
      • Courtney M
      • et al.
      Fewer emergency readmissions and better quality of life for older adults at risk of hospital readmission: a randomized controlled trial to determine the effectiveness of a 24-week exercise and telephone follow-up program.
      • Rich MW
      • et al.
      A multidisciplinary intervention to prevent the readmission of elderly patients with congestive heart failure.
      • López Cabezas C
      • et al.
      Randomized clinical trial of a postdischarge pharmaceutical care program vs regular follow-up in patients with heart failure.
      • Morkisch N
      • et al.
      Components of the Transitional Care Model (TCM) to reduce readmission in geriatric patients: a systematic review.
      These models have been widely promoted through national programs supported by CMS and its Innovation Center (CMMI), including the Hospital Engagement Networks (HENs),

      Centers for Medicare & Medicaid Services. Partnership for Patients. Accessed Oct 7, 2021. https://innovation.cms.gov/innovation-models/partnership-for-patients.

      the Quality Improvement Organizations’ (QIO) strategic aim of Integrating Care for Populations and Communities (ICPC), and the Community-based Care Transitions Program (CCTP).

      Centers for Medicare & Medicaid Services. Community-based Care Transitions Program. Accessed Oct 7, 2021. http://innovation.cms.gov/initiatives/CCTP/.

      These efforts motivated thousands of hospitals, skilled nursing facilities (SNFs), and communities to implement various combinations of TC strategies to improve care transitions. Although these programs share some TC strategies, they vary,
      • Bradley EH
      • et al.
      Quality collaboratives and campaigns to reduce readmissions: what strategies are hospitals using?.
      and previous research has not delineated which program components are most effective or essential.
      The Patient-Centered Outcomes Research Institute (PCORI) funded Project ACHIEVE (Achieving Patient-Centered Care and Optimized Health In Care Transitions by Evaluating the Value of Evidence),
      • Scott AM
      • et al.
      Understanding facilitators and barriers to care transitions: insights from Project ACHIEVE site visits.
      to convene patients and family caregivers with nationally recognized health care researchers to identify which TC services and outcomes matter most to patients and family caregivers
      • Mitchell SE
      • et al.
      Care Transitions from patient and caregiver perspectives.
      and to rigorously evaluate ongoing natural experiments seeking to improve care transitions. The present study aimed to evaluate the associations of five different groups (bundles) of TC strategies with several patient-reported and utilization outcomes.
      • Li J
      • et al.
      Improving evidence-based grouping of transitional care strategies in hospital implementation using statistical tools and expert review.

      Methods

       Study Design

      A detailed description of Project ACHIEVE's methodology and the methodology for identifying the TC strategies and groups of them is described in detail in previous publications.
      • Scott AM
      • et al.
      Understanding facilitators and barriers to care transitions: insights from Project ACHIEVE site visits.
      ,
      • Li J
      • et al.
      Improving evidence-based grouping of transitional care strategies in hospital implementation using statistical tools and expert review.
      Briefly, this prospective observational cohort study collected primary survey data regarding Medicare patient exposure to TC strategies from 42 US short-term acute care and critical access hospitals and their hospitalized patients. Through multivariable analyses, these data were linked to Medicare claims data and Kaiser Permanente's clinical/administrative records to evaluate associations of groups of TC strategies with health care utilization and patient-reported outcomes (PROs). We followed the STROBE checklist (Appendix 1, available in online article) for reports of observational studies.
      • Vandenbroucke JP
      • et al.
      Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaboration.

       Data Sources/Measurement

      We measured patient exposure to 22 TC strategies—16 reported by hospitals and 6 reported by patients (for example, Plain Language Communication). Among hospital-reported strategies, all patients discharged from that hospital were considered “exposed” when the hospital reported application of all required activities to “all” or “most” patients. The process for TC strategy identification included literature review and expert opinion, focus groups with patients and family caregivers, a retrospective analysis of hospitals’ TC efforts, and input and feedback among the study's investigators and stakeholder/scientific advisors. This process is delineated in a prior publication.
      • Li J
      • et al.
      Improving evidence-based grouping of transitional care strategies in hospital implementation using statistical tools and expert review.
      Some of the strategies include risk selection stratification (for example, Identify High-Risk Patients and Intervene) in which the stratification process was conducted with all/most patients, and follow-up intervention or referral was applied as applicable. See Table 1 for the list of evaluated TC strategies and definitions.
      Table 1Transitional Care Strategies and Definitions
      TC Strategy NameDefinition
      Hospital Reported
      Transition TeamHospital uses specific transition team to coordinate transitional care services across hospital and postdischarge.
      Postdischarge Care ConsultationPosthospital discharge telephone call to follow up on postdischarge needs or provide education. Downstream providers provided with direct contact information for inpatient physician.
      Language AssessmentAssessment of spoken and written language preferences that influence each patient's capacity for informed decision making. Tailoring of the education, communication, and care plan to be responsive to language preferences.
      Patient Goal/Preference AssessmentIdentify what matters most to patients by assessing their preferences and health goals.
      Home VisitsPostdischarge home visits provided for patients receiving transitional care services.
      Medication ReconciliationMedication reconciliation review by outside pharmacies and/or primary care providers and availability of a designated person responsible for conducting medication reconciliation at discharge.
      Identify High-Risk Patients and InterveneIdentify high-risk patients and specific patient conditions that predispose patients for poor outcomes after discharge through evidence-based tools; includes follow-through with risk-based interventions to some extent.
      Patient/Family Caregiver Transitional Care Needs AssessmentAssess patient and family caregivers’ transitional care needs such as those related to their medical condition, psychosocial needs, and capabilities (for example, ability to afford medication, managing care at home).
      Timely Exchange of Critical Patient Information Among ProvidersTimely exchange of critical patient information (for example, patient hospital admission, discharge summary, goals/preferences) among providers across the continuum of care; includes outpatient and community service provider access to inpatient health care information.
      Referral to Community ServicesCoordinate community-based services in anticipation of hospital discharge and during transition.
      Teach Back for Information and SkillsIdentify patients’ learning styles and use teach back to confirm patients’ and caregivers' understanding. Provide opportunities for practicing new skills needed for self-care.
      Follow-Up AppointmentOutpatient follow-up appointments scheduled for patients prior to discharge.
      Transition Summary for Patients and Family CaregiversHospital consistently provides a patient-centered hospital summary to patients and caregivers listing important information regarding their care plan (for example, diagnoses, allergies, medications, contact information).
      Patient Reported
      Helpful Health Care ContactPatient possesses contact information for a health care professional who was helpful in addressing problems and concerns if contacted.
      Symptom ManagementProvision of information about what symptoms to monitor and receiving help if needed from health care professionals to manage any changes and unexpected problems with care since discharge.
      Plain Language Communication at HospitalCommunicate with patients—in the hospital—in simple, lay terms to improve patient understanding.
      Plain Language Communication at HomeCommunicate with patients—postdischarge—in simple, lay terms to improve patient understanding.
      Promote Trust at HospitalExpress care and concern for patients as individuals—in hospital setting.
      Promote Trust at HomeExpress care and concern for patients as individuals—postdischarge.
      Not in TC Groups*
      Identification of Family CaregiverIdentification of the family caregiver at the hospital.
      Interdisciplinary Approach

      Designated interdisciplinary team that facilitates the implementation of transitional care efforts and communication of patients’ discharge or transitional care needs. Electronic medical record also used to communicate patient TC needs among providers.
      Standard protocolsUse of standardized discharge template.
      *Although not included in analysis because of near universal application, these groups are assumed present in all transitional care (TC) groups.

       Hospital TC Strategy Implementation

      National hospital association collaborators on ACHIEVE (American Hospital Association, America's Essential Hospitals, and Joint Commission Resources) communicated with their members about Project ACHIEVE, soliciting interested sites. Purposive sampling in the hospital selection aimed to ensure sample diversity in urbanicity; safety-net status; critical access status; and participation in integrated delivery systems (13 Kaiser hospitals in one region), alternative payment models, and/or formal TC programs. From this effort we recruited 42 short-term acute care or critical access hospitals to participate in a TC strategy implementation survey and site visit and to recruit patients to complete a survey of TC experience (see Table 2). From October 2016 to December 2017, survey links were sent to hospital representatives through REDCap,

      University of Kentucky Center for Clinical and Translational Science. REDCap (Vanderbilt University). Accessed Oct 7, 2021. https://redcap.uky.edu/redcap/.

      a Web-based, HIPAA–compliant platform. To mitigate the potential for bias (for example, self-report, misclassification) from hospital representatives overreporting implementation of TC strategies or overestimating their consistent application across the hospital, we validated hospitals’ survey responses through subsequent site visits
      • Cook C
      • et al.
      and/or follow-up contact with hospital representatives.
      Table 2Characteristics of Study Hospitals Compared to National Samples
      Study Hospitals N = 42AHA Hospitals* N = 4,705
      Characteristicn(%)n(%)
      Region
      Midwest819.01,39629.7
      Northeast921.457112.1
      Puerto Rico00501.1
      South819.01,76237.4
      West1740.592119.6
      Unknown0050.1
      Total Licensed Beds
      <1 00614.31,75037.2
      100–2991228.61,19125.3
      ≥ 3002457.190919.3
      Unknown0085518.2
      Organizational Control
      Government, nonfederal819.01,01821.6
      Nongovernment, nonprofit3378.62,76458.7
      Investor-owned, for-profit12.488618.8
      Government, federal00370.8
      Study Hospitals

      N = 42
      CMS Hospitals
      Fiscal Year 2018 Center for Medicare & Medicaid Services (CMS) Impact File.


      N = 3,331
      n%n%
      Teaching Status
      Major teaching1638.138211.5
      Minor teaching2150.01,31639.5
      Nonteaching511.91,63349.0
      Urban/Rural Status
      Large urban2559.51,33940.2
      Other urban819.01,19735.9
      Rural921.479523.9
      * 2017 American Hospital Association (AHA) member survey, acute/critical care only.
      Fiscal Year 2018 Center for Medicare & Medicaid Services (CMS) Impact File.

       Patient Survey

      From June 2017 to April 2018, participating hospitals’ staff identified eligible patients and obtained HIPAA authorization to share their contact information for those interested in participating. The 13 Kaiser hospitals enrolled Medicare fee-for-service and Advantage beneficiaries, while all other hospitals enrolled only Medicare fee-for-service beneficiaries. A total of 17,639 patients were mailed survey packets (English or Spanish) 51 days postdischarge, in which they answered questions regarding their TC experience (for example, receipt of TC strategies) and patient-reported health outcomes. CMS mandated the 51-day delay following discharge, from an initial plan of 7 to 14 days, to avoid conflict with Hospital Consumer Assessment of Healthcare Providers and Systems (HCAHPS) surveys. Nonresponders received a postcard reminder after 7 days, a second survey after 24 days, and a phone call after 34 days; they were retired after up to five call attempts. On average, patients completed surveys 75 days postdischarge, with a 57.0% response rate. See the patient flowchart in Figure 1 and Sorra et al.’s companion article
      • Sorra J
      • et al.
      Development and psychometric properties of surveys to assess patient and caregiver experience with care transitions.
      for more information about the survey's administration methods and psychometric properties.
      Figure 1
      Figure 1This flowchart shows that, on average, patients completed surveys 75 days postdischarge with a 57.0% response rate.

       Groups of TC Strategies

      TC strategy groups were defined a priori through a multistep process delineated in a prior publication.
      • Li J
      • et al.
      Improving evidence-based grouping of transitional care strategies in hospital implementation using statistical tools and expert review.
      The process included referencing the TC groups evaluated in ACHIEVE's prior retrospective study
      • Li J
      • et al.
      Information exchange among providers and patient-centeredness in transitional care: a five-year retrospective analysis.
      ; performing exploratory factor, latent class, and finite mixture model analyses on hospital and patient survey data to ascertain the natural clustering of hospitals’ implementation of and patients’ exposure to TC strategies; and confirming results through review by experts in TC, statistics, and comparative research methodologies. In addition to reflecting natural practice patterns of TC implementation across hospitals, the groups were designed to reflect conceptual or practical relevance and minimize overlap in TC strategies. Importantly, some TC strategies are reflected in multiple groups, which reflects natural practice patterns. Our approach resulted in five overlapping groups of TC strategies for analysis.
      The naming of TC strategy groups was a collaborative process among project primary investigators and key personnel. Over a series of three weekly meetings, the team reviewed and discussed qualitative (that is, related features of the composite strategies) and quantitative (for example, relationships of individual strategies with key outcomes) features of the TC strategies within each group to inform the naming process. For example, strategies with statistically significant associations with 30-day readmissions in bivariate analyses were emphasized in the name. Second, unifying characteristics of remaining strategies were assessd and highlighted. For example, Hospital-Based Trust, Plain Language, and Coordination, emerged from the observation that Promote Trust at Hospital and Plain Language Communication at Hospital were the only TC strategies in the group with significant bivariate relationships with reduced readmissions. Further, the term coordination reflected core features of the other TC strategies in that group: Postdischarge Care Consultation, Medication Reconciliation, Identify High-Risk Patients and Intervene, and Transition Summary for Patients and Family Caregivers. When draft names were formalized, team members were asked to voice agreement or disagreement and engage in discussion to achieve agreement on the final versions.
      Table 3 shows requisite TC strategies for each of the five groups. Exposure to each TC strategy group was entered as a dichotomous variable comparing those exposed to each group to their unexposed counterparts. Patients not receiving any of the specified five groups compose the “usual care” group (TC Control Group). These patients received multiple TC strategies (range 5 to 15), but not in the combinations in Table 3. See Table 4 for the TC strategies experienced by patients in this group.
      Table 3Transitional Care (TC) Strategy Groups and Their Component Strategies
      TC Strategy Groups (Patients Exposed)*Component TC Strategies
      Three TC strategies were almost universally applied and therefore are reasonably considered to be components of all groups: Identification of a Caregiver, Standard Protocols, and Interdisciplinary Approach. Definitions for all strategies are listed in Table 1.
      Patient Communication and Care Management

      (n = 2,158, 27.2%)
      • Patient Goal/Preference Assessment

      • Plain Language Communication at Hospital
      These TC strategies were measured from the patient survey.


      • Transition Summary for Patients and Family Caregivers

      • Helpful Health Care Contact
      These TC strategies were measured from the patient survey.


      OR Symptom Management
      These TC strategies were measured from the patient survey.


      • Plain Language Communication at Home
      These TC strategies were measured from the patient survey.


      • Postdischarge Care Consultation
      Hospital-Based Trust, Plain Language, and Coordination

      (n = 2,090, 26.3%)
      • Identify High-Risk Patients and Intervene

      • Plain Language Communication at Hospital
      These TC strategies were measured from the patient survey.


      • Promote Trust at Hospital
      These TC strategies were measured from the patient survey.


      • Medication Reconciliation

      • Transition Summary for Patients and Family Caregivers

      • Postdischarge Care Consultation
      Home-Based Trust, Plain Language, and Coordination

      (n = 1,979, 24.9%)
      • Transition Team

      • Follow-Up Appointment

      • Referral to Community Services

      • Home Visits

      • Promote Trust at Home
      These TC strategies were measured from the patient survey.


      • Plain Language Communication at Home
      These TC strategies were measured from the patient survey.
      Patient/Family Caregiver Assessment and Information Exchange Among Providers

      (n = 3,093, 39.0%)
      • Identify High-Risk Patients and Intervene

      • Patient/Family Caregiver Transitional Care Needs Assessment

      • Patient Goal/Preference Assessment

      • Timely Exchange of Critical Patient Information Among Providers
      Assessment and Teach Back

      (n = 508, 6.4%)
      • Language Assessment

      • Teach Back for Information and Skills

      • Postdischarge Care Consultation
      TC Control Group
      Patients were exposed to TC strategies but not in the groups defined above.


      (n = 2,042, 25.7%)
      • Not in any other group
      * Individual patients may be exposed to more than one group of TC strategies and included in analyses for each group.
      Three TC strategies were almost universally applied and therefore are reasonably considered to be components of all groups: Identification of a Caregiver, Standard Protocols, and Interdisciplinary Approach. Definitions for all strategies are listed in Table 1.
      These TC strategies were measured from the patient survey.
      § Patients were exposed to TC strategies but not in the groups defined above.
      Table 4Percentage of Patients in TC Control Group (N = 2,042) Exposed to Specific TC Strategies
      TC Strategy%
      Symptom Management19.2
      Promote Trust at Home30.5
      Teach Back for Information and Skills31.9
      Plain Language Communication at Home35.3
      Patient/Family Caregiver Transitional Care Needs Assessment41.4
      Postdischarge Care Consultation48.3
      Timely Exchange of Critical Patient Information Among Providers52.2
      Promote Trust at Hospital55.3
      Helpful Health Care Contact56.4
      Patient Goal/Preference Assessment58.0
      Plain Language Communication at Hospital59.7
      Follow-Up Appointment65.0
      Identify High-Risk Patients and Intervene65.8
      Referral to Community Services71.1
      Language Assessment80.1
      Transition Summary for Patients and Family Caregivers80.7
      Medication Reconciliation83.6
      Home Visits84.2
      Transition Team90.1
      Standard Protocols*93.9
      Interdisciplinary Approach*99.4
      Identification of Family Caregiver*100.0
      * Due to the near universal application of the final three strategies, they were not included in the transitional care (TC) strategy groups but were considered as being components of them.

       Clinical Claims Data

      Claims data of patients who completed surveys were obtained from CMS and Kaiser, including clinical encounters 2 years prior and 30 days after hospital admission (June 2015–July 2018).

       Participants

      Eligible patients were hospitalized on medical and surgical units at participating hospitals with Medicare insurance. Most Kaiser Permanente patients had Medicare Advantage. Otherwise, participants had Medicare Fee-for-Service. Exclusions included in-hospital deaths, transfers to another acute care hospital, discharges against medical advice, admissions for primary diagnosis of psychiatric condition, rehabilitation, or medical treatment of cancer; prisoners; or under suicide watch. Patients with active cancer treatment were excluded, as they often return for scheduled cancer treatment. Patients with mental illness as comorbidity (secondary diagnosis) were not excluded.
      Our study required power to detect meaningful differences in outcomes across the five groups of TC strategies. Therefore, we conducted a Monte Carlo simulation using ACHIEVE retrospective analysis data,
      • Li J
      • et al.
      Information exchange among providers and patient-centeredness in transitional care: a five-year retrospective analysis.
      which included five years of Medicare claims data from patients discharged from 370 hospitals. Results indicated a sample of 7,500 to 8,500 patients could detect absolute readmission rate differences of 2.1 to 2.9% (p = 0.05).

       IRB Approval

      Participants provided written or oral informed consent when completing the postdischarge survey. Institutional Review Boards (IRBs) at the University of Kentucky, Kaiser Permanente, and Westat approved the study protocol prior to data collection.

       Main Outcomes and Measures

      Primary outcomes included 30-day hospital readmissions and ED visits within 7 days to any hospital postdischarge, extracted from Medicare and Kaiser patient claims data, and PROs measured through the patient survey using validated instruments when available (for example, the National Institutes of Health Patient-Reported Outcomes Measurement Information System [PROMIS]
      • Cella D
      • et al.
      The Patient-Reported Outcomes Measurement Information System (PROMIS) developed and tested its first wave of adult self-reported health outcome item banks: 2005–2008.
      ). PROs included self-reported: physical health, mental health, pain, and participation in daily activities.
      • Cella D
      • et al.
      The Patient-Reported Outcomes Measurement Information System (PROMIS) developed and tested its first wave of adult self-reported health outcome item banks: 2005–2008.
      ,
      • Hays RD
      • et al.
      Development of physical and mental health summary scores from the Patient-Reported Outcomes Measurement Information System (PROMIS) global items.

       Statistical Analysis

      We used multivariable statistical models to control for potentially confounding variables at the patient, hospital, and community level, based on the CMS risk standardized hospitalwide all-cause readmission methodology, ACHIEVE site visits,
      • Scott AM
      • et al.
      Understanding facilitators and barriers to care transitions: insights from Project ACHIEVE site visits.
      ,
      • Li J
      • et al.
      Improving evidence-based grouping of transitional care strategies in hospital implementation using statistical tools and expert review.
      patient/family caregiver interviews,
      • Mitchell SE
      • et al.
      Care Transitions from patient and caregiver perspectives.
      and guidance from the study's stakeholder and scientific advisors (Appendix 2, available in online article). Appendix 3 delineates model covariates for all outcomes. Briefly, health care utilization covariates included (a) patient demographics, comorbidities/prior health care utilization, and psychosocial factors (for example, health literacy); (b) hospital structural factors (for example, SNF ownership, Kaiser affiliation); and (c) community factors (for example, SNF beds per 100,000 residents). PRO covariates included patient demographics, comorbidities, and psychosocial factors.
      • Cella D
      • et al.
      The Patient-Reported Outcomes Measurement Information System (PROMIS) developed and tested its first wave of adult self-reported health outcome item banks: 2005–2008.
      ,
      • Hays RD
      • et al.
      Development of physical and mental health summary scores from the Patient-Reported Outcomes Measurement Information System (PROMIS) global items.
      We tested independent associations of model covariates with each outcome through analysis of variance (ANOVA) and t-tests before adjusting for all covariates. To adjust for multiple comparisons across the five TC groups and six outcomes assessed, assuming 30 possible hypotheses, we used a false discovery rate (FDR) adjustment—specifically, the Benjamini-Hochberg adjustment. A prespecified upper bound of FDR of 0.05 was used for these adjustment calculations. PRO data were collapsed into three categories (for example, Poor/Fair, Good, and Very Good/Excellent) for a multinomial regression analysis with Poor/Fair as the reference.
      For health care utilization outcomes, we applied a generalized linear mixed model (GLMM) with random effects for hospitals to each dichotomous outcome. SAS 9.4 (SAS Institute Inc., Cary, North Carolina) PROC GLIMMIX was used, following this equation: logit(P(Readmissionij=1))=β0+β1*PatientDemographic+β2*Disease+β3*Community+β4*Hospital+β5*TCGroup+bi0+εij— eq (1), where i is the index hospital, j is each index admission claim, bi0is the random effects across hospitals where εij is the error term, andβi are the vectors of coefficients of related covariates. This generated odds ratios that were then converted to risk ratios (RRs), given the large sample size. This conversion aimed to enhance understanding of the magnitude of the results.

       Missing Data

      To determine record “completeness,” we applied the HCAHPS survey

      Centers for Medicare & Medicaid Services. CAPHPS Hospital Survey (HCAHPS) Quality Assurance Guidelines, ver. 12.0. Mar 2017. Accessed Oct 7, 2021. http://www.hcahpsonline.org/globalassets/hcahps/quality-assurance/qag_v12.0_2017.pdf.

      criterion requiring at least 50% completion of applicable-to-all (ATA) questions. “Applicable to all” indicates if a question was applicable to all respondents, excluding questions skipped based on previous responses. Sensitivity analyses using 90% ATA completion rate showed no significant difference in responses. Missingness in final models varied between 6% and 11%.

      Results

      Compared to all US hospitals (Table 2), the 42 participating hospitals were more often located in the West (40.5% vs. 19.6%), had at least 300 beds (57.1% vs 19.3%), were teaching (88.1% vs. 51.0%), and were nonprofit (78.6% vs. 58.7%). Thirteen (31.0%) were from Kaiser Permanente Southern California. On average, patients completed surveys 75 days postdischarge, with a 57.0% response rate. See the patient flowchart in Figure 1.
      Table 5 summarizes patient sample demographics (N = 7,939). The majority were female (53.4%), White (78.9%), and high school educated (59.0%), with an average age of 72.3 years (standard deviation [SD] = 10.1). About 10% of respondents were Black; 14.7% Hispanic. Patient exposure to TC groups was lowest for Assessment and Teach Back (6.4%, n = 508) and highest for Patient/Family Caregiver Assessment and Information Exchange Among Providers (39.0%, n = 3,093). Demographics across TC groups were similar, although Patient Communication and Care Management had a higher proportion of White patients (82.7%), while the TC Control Group had higher proportions of disabled patients (16.5% vs. 11.9% overall) and dual-eligible patients (19.1% vs. 15.5% overall).*
      Table 5Patient Characteristics Overall and by TC Group
      OverallNo Group

      (not in any group)

      N = 2,042

      (25.7%)
      Patient

      Communication and Care Management

      N = 2,158

      (27.2%)
      Hospital-Based Trust, Plain Language, and Coordination

      N = 2,090

      (26.3%)
      Home-Based Trust,

      Plain Language, and Coordination

      N = 1,979

      (24.9%)
      Patient/Family

      Caregiver Assessment and Information Exchange Among

      Providers

      N = 3,093

      (39.0%)
      Assessment

      and Teach Back

      N = 508 (6.4%)
      n%n%n%n%n%n%n%
      Gender
      Female4,23753.371,09353.531,09350.651,08251.771,10255.681,63652.8926852.76
      Male3,70246.6394946.471,06549.351,00848.2387744.321,45747.1124047.24
      Race from Survey
      Unknown3444.33612.99773.571416.75864.351765.69152.95
      Other5096.41864.211396.441939.231236.222628.47438.46
      Black82210.3520810.191587.321677.9926013.1434611.196713.19
      White6,26478.901,68782.621,78482.671,58976.031,51076.302,30974.6538375.39
      Medicare Eligibility Type
      Age6,26278.881,59277.961,67377.531,67280.001,59280.442,46279.6041281.10
      Disability94111.8533616.4531514.601708.1320510.362377.665110.04
      ESRD2202.77793.87793.66452.15552.78481.5581.57
      QMB5717.1922010.771848.531044.981155.811374.43458.86
      SLMB4976.261045.091064.911346.411296.522347.57407.87
      Dual Medicaid Eligible
      Yes1,23215.5239019.1034315.8927613.2128414.3540513.099318.31
      TC, transitional care; ESRD, end state renal disease; QMB, Qualified Medicare Beneficiary; SLMB, Specified Low-Income Medicare Beneficiary.
      Patient-reported outcome data were reported for 95% to 97% of patients. Mental health was the only PRO for which a majority of patients responded “excellent/very good” (60.8%). About 10% of patients had an unplanned 30-day readmission; 8.6% had an ED visit within 7 days postdischarge (Appendix 4).

       Unadjusted Associations with Primary Outcomes

      Prior to risk adjustment, the TC group labeled Hospital-Based Trust, Plain Language, and Coordination was associated with a range of positive outcomes, while the TC Control Group was associated with a range of poor outcomes (Appendices 5a and 5b). Patient Communication and Care Management and Home-Based Trust, Plain Language, and Coordination were associated with positive PROs, while Patient/Family Caregiver Assessment and Information Exchange Among Providers was associated with excellent/very good mental health.

       Risk-Adjusted Associations

       Health Care Utilization Outcomes

      One TC group, Hospital-Based Trust, Plain Language, and Coordination, was significantly associated with lower risk of 30-day readmissions (RR, 0.72; 95% confidence interval [CI] = 0.57–0.92, p < 0.01) and 7-day ED visits (RR, 0.72; 95% CI = 0.55–0.93, p < 0.01) in the risk-adjusted analysis (Table 6).
      Table 6Risk-Adjusted Associations Between TC Groups and Patient Outcomes*
      Health Care UtilizationSelf-Reported Health StatusParticipation in Daily ActivitiesPain in Past Week
      30-Day Readmission7-Day ED VisitsPhysicalMental
      TC GroupRR (95% CI)RR (95% CI)Outcome Level
      Reference = Poor/Fair.
      RR (95% CI)RR (95% CI)Outcome Level
      Reference = A little/Not at all.
      RR (95% CI)Outcome Level
      Reference = Constantly/Almost daily.
      RR (95% CI)
      Patient Communication and Care Management0.96

      (0.77–1.19)
      1.01

      (0.80–1.28)
      Good1.05

      (0.97–1.13)
      1.11

      (1.01–1.20)
      Moderate1.05

      (0.95–1.14)
      Sometimes0.95

      (0.85–1.06)
      Very Good/ Excellent1.05

      (0.99–1.12)
      1.06
      p value is less than or equal to 0.01.


      (1.02–1.10)
      Mostly/ Completely1.08
      p value is less than or equal to 0.01.


      (1.02–1.14)
      Once in past week/Not at all1.00

      (0.93–1.08)
      Hospital-Based Trust, Plain Language, and Coordination0.72
      p value is less than or equal to 0.01.


      (0.57–0.92)
      0.72
      p value is less than or equal to 0.01.


      (0.55–0.93)
      Good1.06

      (0.99–1.14)
      1.22
      p value is less than 0.001.


      (1.13–1.30)
      Moderate1.22
      p value is less than 0.001.


      (1.11–1.31)
      Sometimes1.09

      (0.98–1.21)
      Very Good/ Excellent1.15
      p value is less than 0.001.


      (1.09–1.21)
      1.13
      p value is less than 0.001.


      (1.09–1.16)
      Mostly/ Completely1.14
      p value is less than 0.001.


      (1.08–1.20)
      Once in past week/Not at all1.11
      p value is less than or equal to 0.01.


      (1.03–1.19)
      Home-Based Trust, Plain Language, and Coordination0.95

      (0.75–1.21)
      0.89

      (0.70–1.12)
      Good1.08

      (1.00–1.16)
      1.17
      p value is less than 0.001.


      (1.07–1.25)
      Moderate1.11

      (1.02–1.21)
      Sometimes1.08

      (0.96–1.19)
      Very Good/ Excellent1.08

      (1.02–1.15)
      1.08
      p value is less than 0.001.


      (1.04–1.12)
      Mostly/ Completely1.09
      p value is less than 0.001.


      (1.03–1.15)
      Once in past week/Not at all0.98

      (0.91–1.06)
      Patient/Family Caregiver Assessment and Information Exchange Among Providers0.98

      (0.72–1.31)
      0.85

      (0.64–1.11)
      Good0.91

      (0.83–0.99)
      1.05

      (0.95–1.14)
      Moderate1.00

      (0.90–1.10)
      Sometimes0.98

      (0.88–1.09)
      Very Good/ Excellent0.87
      p value is less than 0.001.


      (0.80–0.94)
      0.98

      (0.93–1.03)
      Mostly/ Completely0.98

      (0.91–1.04)
      Once in past week/Not at all0.97

      (0.89–1.04)
      Assessment and Teach Back1.47

      (0.95–2.22)
      1.06

      (0.69–1.60)
      Good0.89

      (0.77–1.01)
      0.90

      (0.75–1.04)
      Moderate0.78
      p value is less than or equal to 0.01.


      (0.64–0.94)
      Sometimes0.93

      (0.76–1.10)
      Very Good/ Excellent0.89

      (0.78–0.99)
      0.92

      (0.83–1.00)
      Mostly/ Completely1.00

      (0.91–1.10)
      Once in past week/Not at all0.91

      (0.79–1.03)
      TC Control Group

      (not in any TC group)
      0.97

      (0.74–1.25)
      0.79

      (0.60–1.04)
      Good0.92

      (0.83–1.01)
      0.96

      (0.85–1.05)
      Moderate1.00

      (0.89–1.10)
      Sometimes0.95

      (0.83–1.09)
      Very Good/ Excellent0.83
      p value is less than 0.001.


      (0.75–0.91)
      0.92
      p value is less than 0.001.


      (0.86–0.97)
      Mostly/ Completely0.94

      (0.86–1.01)
      Once in past week/Not at all0.91

      (0.82–1.01)
      * Adjustment for multiple comparisons performed using a prespecified upper bound false discovery rate of 0.05 with the Benjamini-Hochberg adjustment (Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc Series B Stat Methodol. 1995;57:289–300). Statistical significance with adjustment for multiple comparisons noted with a bold font.
      TC, transitional care; ED, emergency department; RR. risk ratio; CI, confidence interval.
      Reference = Poor/Fair.
      Reference = A little/Not at all.
      § Reference = Constantly/Almost daily.
      || p value is less than or equal to 0.01.
      # p value is less than 0.001.

       Clinical PROs

      Hospital-Based Trust, Plain Language, and Coordination was positively associated with all PROs, with RRs ranging from 1.11 (95% CI = 1.03–1.19, p = 0.009) for low/no pain to 1.15 (95% CI = 1.09–1.21, p < 0.001) for excellent/very good physical health (Table 6).
      Home-Based Trust, Plain Language, and Coordination was associated with excellent/very good mental health (RR, 1.08; 95% CI = 1.04–1.12, p < 0.001) and greater participation in daily activities (RR, 1.09; 95% CI = 1.03–1.15, p = 0.004.), as was Patient Communication and Care Management: excellent/very good mental health (RR, 1.06; 95% CI = 1.02–1.10, p = 0.009) and greater participation in daily activities (RR, 1.08; 95% CI = 1.02–1.14, p = 0.01).
      Conversely, Patient/Family Caregiver Assessment and Information Exchange Among Providers was associated with lower likelihood of excellent/very good physical health (RR, 0.87; 95% CI = 0.80–0.94, p < 0.001). The TC Control Group was associated with lower likelihood of both excellent/very good physical health (RR, 0.83; 95% CI = 0.75–0.91, p < 0.001) and excellent/very good mental health (RR, 0.92; 95% CI = 0.86–0.97. p < 0.001).

      Discussion

      To our knowledge, Project ACHIEVE is the largest known prospective TC study, collecting data from 42 diverse hospitals across the nation and from nearly 8,000 Medicare beneficiaries regarding their TC experiences, self-reported outcomes, and health care utilization. Of the five groups of TC strategies assessed, the three that focused on patient- and family caregiver–centered care were consistently associated with more participation in daily activities and better mental health. Further, the Hospital-Based Trust, Plain Language, and Coordination TC group that included hospital providers’ conveyance of trust, plain language communication, “tailored care planning” (that is, medication reconciliation and high-risk patient interventions), and “bridging efforts” (transition summary and postdischarge care consultation) was additionally associated with higher ratings on all PROs and lower likelihood of 7-day ED visits and 30-day readmissions. Our results suggest that implementation of the TC strategies in this group collectively address the spectrum of patients’ needs to ensure a successful transition from the hospital. These results confirm prior research that bridging processes and tailored care planning are essential,
      • Li J
      • Young R
      • Williams MV
      Optimizing transitions of care to reduce rehospitalizations.
      • Coleman EA
      • et al.
      Preparing patients and caregivers to participate in care delivered across settings: the care transitions intervention.
      • Naylor M
      • et al.
      Comprehensive discharge planning for the hospitalized elderly: a randomized clinical trial.
      • Coleman EA
      • et al.
      The care transitions intervention: results of a randomized controlled trial.
      • Jack BW
      • et al.
      A reengineered hospital discharge program to decrease rehospitalization: a randomized trial.
      • Williams MV
      • et al.
      Project BOOST implementation: lessons learned.
      • Hansen LO
      • et al.
      Interventions to reduce 30-day rehospitalization: a systematic review.
      • Naylor MD
      • et al.
      Comprehensive discharge planning and home follow-up of hospitalized elders: a randomized clinical trial.
      ,
      • Leppin AL
      • et al.
      Preventing 30-day hospital readmissions: a systematic review and meta-analysis of randomized trials.
      and suggest that establishing trust and communicating clearly with patients should also be foundational parts of future transitional care.
      As health care becomes increasingly complex, our findings contribute to a growing body of literature suggesting that clarity and trust between patients and providers are a prerequisite for high-quality care.
      • Saha S
      • Beach MC.
      The impact of patient-centered communication on patients’ decision making and evaluations of physicians: a randomized study using video vignettes.
      • Thom DH
      • Hall MA
      • Pawlson LG.
      Measuring patients’ trust in physicians when assessing quality of care.
      • Thom DH.
      Physician behaviors that predict patient trust.
      • Safran DG
      • et al.
      Linking primary care performance to outcomes of care.
      • Birkhaüer J
      • et al.
      Trust in the health care professional and health outcome: a meta-analysis.
      • Fiscella K
      • et al.
      Patient trust: is it related to patient-centered behavior of primary care physicians?.
      • Murray B
      • McCrone S.
      An integrative review of promoting trust in the patient-primary care provider relationship.
      • Shan L
      • et al.
      Patient satisfaction with hospital inpatient care: effects of trust, medical insurance and perceived quality of care.
      • Benkert R
      • et al.
      Trust, mistrust, racial identity and patient satisfaction in urban African American primary care patients of nurse practitioners.
      • Bohnert ASB
      • et al.
      Ratings of patient-provider communication among veterans: serious mental illnesses, substance use disorders, and the moderating role of trust.
      • Blendon RJ
      • Benson JM
      • Hero JO.
      Public trust in physicians—U.S. medicine in international perspective.
      • First Smith CP.
      do no harm: institutional betrayal and trust in health care organizations.
      • Chandra S
      • Mohammadnezhad M
      • Ward P.
      Trust and communication in a doctor-patient relationship: a literature review.
      • Clever SL
      • et al.
      Does doctor-patient communication affect patient satisfaction with hospital care? Results of an analysis with a novel instrumental variable.
      • Musa D
      • et al.
      Trust in the health care system and the use of preventive health services by older Black and White adults.
      In fact, although many of the strategies within the Hospital-Based Trust, Plain Language, and Coordination TC group were also in other TC groups, they were not significantly associated with the same range of positive outcomes as when combined with Promote Trust at Hospital. The strategies focusing on trust and communication possibly complement those focusing on care planning and transition management so that the full benefit of the latter is achieved. As such, these findings and this TC group map closely to what patients and family caregivers said they want from care transitions in ACHIEVE's prior focus groups and interviews, and what made care transitions feel safe and trustworthy.
      • Mitchell SE
      • et al.
      Care Transitions from patient and caregiver perspectives.
      Specifically, patients and family caregivers wanted to feel (1) cared for and about (for example, Promote Trust), (2) prepared to implement the care plan (for example, Plain Language Communication, Transition Summary for Patients and Family Caregivers), and (3) clear accountability from the health care system regarding whom to contact for help (for example, Postdischarge Care Consultation).
      • Mitchell SE
      • et al.
      Care Transitions from patient and caregiver perspectives.
      Our research is consistent with previous studies of trust in patient care, finding that trust between patients and providers is consistently associated with better care plan adherence,
      • Safran DG
      • et al.
      Linking primary care performance to outcomes of care.
      positive health outcomes
      • Safran DG
      • et al.
      Linking primary care performance to outcomes of care.
      • Birkhaüer J
      • et al.
      Trust in the health care professional and health outcome: a meta-analysis.
      • Fiscella K
      • et al.
      Patient trust: is it related to patient-centered behavior of primary care physicians?.
      • Murray B
      • McCrone S.
      An integrative review of promoting trust in the patient-primary care provider relationship.
      and patient satisfaction.
      • Safran DG
      • et al.
      Linking primary care performance to outcomes of care.
      ,
      • Fiscella K
      • et al.
      Patient trust: is it related to patient-centered behavior of primary care physicians?.
      ,
      • Shan L
      • et al.
      Patient satisfaction with hospital inpatient care: effects of trust, medical insurance and perceived quality of care.
      • Benkert R
      • et al.
      Trust, mistrust, racial identity and patient satisfaction in urban African American primary care patients of nurse practitioners.
      • Bohnert ASB
      • et al.
      Ratings of patient-provider communication among veterans: serious mental illnesses, substance use disorders, and the moderating role of trust.
      Unfortunately, public trust in the health system has declined,
      • Blendon RJ
      • Benson JM
      • Hero JO.
      Public trust in physicians—U.S. medicine in international perspective.
      in part due to high health care costs, perceived institutional betrayal,
      • First Smith CP.
      do no harm: institutional betrayal and trust in health care organizations.
      poor communication,
      • Chandra S
      • Mohammadnezhad M
      • Ward P.
      Trust and communication in a doctor-patient relationship: a literature review.
      ,
      • Clever SL
      • et al.
      Does doctor-patient communication affect patient satisfaction with hospital care? Results of an analysis with a novel instrumental variable.
      and lack of care continuity.
      • Thom DH
      • Hall MA
      • Pawlson LG.
      Measuring patients’ trust in physicians when assessing quality of care.
      ,
      • Musa D
      • et al.
      Trust in the health care system and the use of preventive health services by older Black and White adults.
      Although this study did not seek to define the specific provider behaviors that engender trust in care transitions, a body of prior literature documents that feeling cared for by health care providers has repeatedly been shown to increase trust.
      • Thom DH.
      Physician behaviors that predict patient trust.
      ,
      • Greene J
      • Ramos C.
      A mixed methods examination of health care provider behaviors that build patients’ trust.
      ,
      • Hillen MA
      • et al.
      How can communication by oncologists enhance patients’ trust? An experimental study.
      Other provider behaviors that have been shown to correspond with increased trust include effective, patient-centered, and/or empathetic communication
      • Saha S
      • Beach MC.
      The impact of patient-centered communication on patients’ decision making and evaluations of physicians: a randomized study using video vignettes.
      ,
      • Greene J
      • Ramos C.
      A mixed methods examination of health care provider behaviors that build patients’ trust.
      ,
      • Zwingmann J
      • et al.
      Effects of patient-centered communication on anxiety, negative affect, and trust in the physician in delivering a cancer diagnosis: a randomized, experimental study.
      ; competence
      • Greene J
      • Ramos C.
      A mixed methods examination of health care provider behaviors that build patients’ trust.
      ,
      • Hillen MA
      • et al.
      How can communication by oncologists enhance patients’ trust? An experimental study.
      ; and honesty.
      • Thom DH
      • Hall MA
      • Pawlson LG.
      Measuring patients’ trust in physicians when assessing quality of care.
      ,
      • Hillen MA
      • et al.
      How can communication by oncologists enhance patients’ trust? An experimental study.
      Importantly, the TC strategy group associated with the most positive outcomes was also distinguished by tailored care planning and bridging activities. Hospital discharge is a vulnerable time for patients who, while recovering, must assume new self-care responsibilities, adjust medications and diet, and monitor new and evolving symptoms. Unfortunately, patients and family caregivers are frequently omitted from care planning, leaving them inadequately prepared,
      • Coleman EA
      • Berenson RA.
      Lost in transition: challenges and opportunities for improving the quality of transitional care.
      ,
      • Snow V
      • et al.
      Transitions of care consensus policy statement: American College of Physicians, Society of General Internal Medicine, Society of Hospital Medicine, American Geriatrics Society, American College Of Emergency Physicians, and Society for Academic Emergency Medicine.
      lacking clarity about who is responsible for their care,
      • Mitchell SE
      • et al.
      Care Transitions from patient and caregiver perspectives.
      ,
      • O'Leary KJ
      • et al.
      Hospitalized patients’ understanding of their plan of care.
      or unsure about whom to contact with questions.
      • Coleman EA
      • Berenson RA.
      Lost in transition: challenges and opportunities for improving the quality of transitional care.
      ,
      • Snow V
      • et al.
      Transitions of care consensus policy statement: American College of Physicians, Society of General Internal Medicine, Society of Hospital Medicine, American Geriatrics Society, American College Of Emergency Physicians, and Society for Academic Emergency Medicine.
      Bridging activities that prepare patients for their “new normal” have been associated with decreased health care utilization.
      • Hansen LO
      • et al.
      Interventions to reduce 30-day rehospitalization: a systematic review.
      ,
      • Rennke S
      • et al.
      Hospital-initiated transitional care interventions as a patient safety strategy: a systematic review.
      Our results suggest that when combined with promotion of trust and plain language communication, their benefits may include improved PROs.
      Unexpectedly, two TC strategy groups characterized by patient assessments (for example, risk, language, patient goals) and information exchange among care settings were associated with poorer outcomes. It is possible that assessment results were insufficiently acted upon or that patient/family caregivers were not engaged in information exchanged. Another possible explanation could be unmeasured/nonreported variance of fidelity or adaptation in strategy implementation, as these groups also included only hospital-reported TC strategies.

       Limitations

      Despite the strengths of this large, prospective cohort study, it has limitations. First, the study's observational design precludes the ability to attribute strong causation to the associations found between TC groups and outcomes. Nonetheless, rigorous observational studies are necessary to evaluate ongoing natural experiments.
      • Li J
      • et al.
      Project ACHIEVE—using implementation research to guide the evaluation of transitional care effectiveness.
      ,
      • Black N.
      Why we need observational studies to evaluate the effectiveness of health care.
      Our hospital and patient samples may differ from nonparticipants in unmeasured ways. Due to the need to ensure adequate implementation of varying groups of TC strategies for analysis, we purposefully recruited hospitals using evidence-based TC programs, or members of an integrated delivery system (for example, Kaiser Permanente). As a result, the sampled hospitals may be better resourced to provide TC than other US hospitals. Notably, the TC Control Group included a higher proportions of patients who were disabled or dual-eligible than the other TC groups. Patients in the TC Control Group also reported poorer physical and mental health. This might reflect inequitable health care received by such patients; that is, they were not provided coordinated care that included important TC strategies grouped together. Increasing research documents how such inequity is associated with lack of trust of the health system and poorer outcomes.

      Bipartisan Policy Center. Delivery System Reform: Improving Care for Individuals Dually Eligible for Medicare and Medicaid. Sep 2016. Accessed Oct 7, 2021. https://bipartisanpolicy.org/report/dually-eligible-medicare-medicaid/.

      • Alberti PM
      • Baker MC.
      Dual eligible patients are not the same: how social risk may impact quality measurement's ability to reduce inequities.
      • Gorges RJ
      • Sanghavi P
      • Konetzka RT.
      A national examination of long-term care setting, outcomes, and disparities among elderly dual eligibles.
      Project ACHIEVE provides additional support for these harmful associations.
      Patient and hospital survey data were self-reported and subject to differences in interpretation and recall bias. To mitigate these potential biases among hospitals, we validated hospital survey responses through site visits and follow-up contact (for example, phone call, e-mails). However, lacking consent from nonparticipating patients to collect demographic data, we were unable to compare them to participants. The required delay in patient survey administration (> 51 days postdischarge) may have exacerbated patient recall bias and selection bias, in that sicker patients may have died or become incapable of participation. In addition, patients with cognitive impairment who lacked consenting capacity were excluded from participation. Thus, participants—10.4% of whom had a 30-day unplanned readmission—were likely healthier than hospitalized Medicare beneficiaries overall, 15.3% of whom had a 30-day readmission.

      Centers for Medicare and Medicaid Services. FY2017 national average rate. Accessed October 27, 2021. https://data.cms.gov/provider-data/#2017-annual-files.

      Although patient-reported exposure (to 6 of the 22 TC strategies) and outcomes are strengths of the study, some exposure and outcome data were collected simultaneously. Therefore, it is possible that patients perceiving “good” outcomes were more likely to credit hospitals with providing certain TC strategies. Nonetheless, all TC groups with patient-reported strategies also required hospital-reported strategies, and results were consistent across patient-reported and health care utilization outcomes, lending validity to our findings.
      Finally, because Project ACHIEVE ultimately aims to provide implementation recommendations to hospitals, and because many TC strategies are either not directly observable or are not applicable to all patients (for example, home visit, referral to community services), our approach to measuring TC strategy implementation is largely hospital-centric. Consequently, apart from the six TC strategies we believed essential to assess through patients’ responses, we lack a complete view of patients’ individual exposure to TC strategies, which may vary even within hospitals. To minimize this potential, we targeted patient recruitment to the floors/units where hospitals focused their TC efforts and complemented hospital self-report of implementation with site visits and follow-up calls. Although future research should identify the groups of strategies that work specifically within certain at-risk populations or communities, we believe our findings provide useful guidance to all hospitals. Nonetheless, the results may not be fully applicable to younger (that is, non-Medicare) patients or to smaller, nonacademic, or for-profit hospitals. Future research should also include prospective experimental or quasi-experimental studies of transitional care models that include transitional care strategies from the most successful group.

      Conclusion

      To our knowledge, Project ACHIEVE is the largest patient-centered study of Medicare patients undergoing hospital discharge transitions. Our findings contribute to a body of research suggesting that a patient- and family caregiver–centered approach to the hospital discharge TC process can improve care transitions
      • Mitchell SE
      • et al.
      Care Transitions from patient and caregiver perspectives.
      ,
      • Safran DG
      • et al.
      Linking primary care performance to outcomes of care.
      • Birkhaüer J
      • et al.
      Trust in the health care professional and health outcome: a meta-analysis.
      • Fiscella K
      • et al.
      Patient trust: is it related to patient-centered behavior of primary care physicians?.
      • Murray B
      • McCrone S.
      An integrative review of promoting trust in the patient-primary care provider relationship.
      • Shan L
      • et al.
      Patient satisfaction with hospital inpatient care: effects of trust, medical insurance and perceived quality of care.
      by communicating plainly and expressing care and concern for patients.
      • Zulman DM
      • et al.
      Practices to foster physician presence and connection with patients in the clinical encounter.
      • Teutsch C.
      Patient-doctor communication.
      • Riess H
      • Kraft-Todd G.E.M.P.A.T.H.Y
      a tool to enhance nonverbal communication between clinicians and their patients.
      Our findings build upon prior research by demonstrating that these communication strategies, combined with tailored care planning and bridging processes, may reduce postdischarge acute care utilization and improve the patient-centered clinical outcomes of transitional care.
      Centers for Medicare & Medicaid Services
      Trust matters.

       Funding and Disclaimer

      Research reported in this manuscript was funded through a Patient-Centered Outcomes Research Institute (PCORI) Award (TC-1403-14049). The statements in this manuscript are solely the responsibility of the authors and do not necessarily represent the views of the Patient-Centered Outcomes Research Institute (PCORI), its Board of Governors, or Methodology Committee.

       Conflicts of Interest

      All authors report no conflicts of interest.

       Acknowledgments

      The present study required a tremendous team effort. The authors would like to acknowledge the significant contributions of the following individuals in the conduct of this study. University of Kentucky: Chris Cook, MS; Megan McIntosh; and Dan Cleland, MS, for their assistance with project coordination and administration. Kaiser Permanente: Heather Watson, MBA; Sandra Koyama, MD; Dan Huynh, MD; Maria Taitano, MD; Ernest Shen, PhD; Angel Alem, MPH; Thearis Osuji, MPH; and Janet Shim, MS, for their assistance in project conceptualization and coordination. Boston Medical Center: Suzanne Mitchell, MD; and Jessica Howard Martin, MA, MPH, for their assistance in project conceptualization. Telligen: Brianna Gass, MPH; Christine LaRocca, MD; and Lacy McFall for their assistance in project conceptualization. Westat: Deborah Carpenter, RN, MSN; John Rauch, PhD; and Betsy Kaeberle, for project conceptualization, coordination, and the design and conduct of patient surveys. University of Pennsylvania: Mary Naylor, PhD, RN; Karen Hirschman, PhD, MSW; Mark Pauly, PhD; and Kathleen McCauley, PhD, RN, for their assistance in project conceptualization and coordination. The authors also wish to thank Chad Boult, MD, MPH, MBA, for his careful review and helpful editing suggestions.

      Appendix. SUPPLEMENTARY MATERIALS

      References

        • Clancy CM.
        Reengineering hospital discharge: a protocol to improve patient safety, reduce costs, and boost patient satisfaction.
        Am J Med Qual. 2009; 24: 344-346
        • Kaiser Health News
        Health Care's ‘Dirty Little Secret’: No One May Be Coordinating Care.
        Rabin RC. 2012; (Apr 30,. Accessed Oct 7, 2021)
        • Greenwald JL
        • Denham CR
        • Jack BW.
        The hospital discharge: a review of a high risk care transition with highlights of a reengineered discharge process.
        J Patient Saf. 2007; 3: 97-106
        • Coleman EA.
        Falling through the cracks: challenges and opportunities for improving transitional care for persons with continuous complex care needs.
        J Am Geriatr Soc. 2003; 51: 549-555
        • Mitchell SE
        • et al.
        Care Transitions from patient and caregiver perspectives.
        Ann Fam Med. 2018; 16: 225-231
        • Coleman EA
        • Williams MV.
        Executing high-quality care transitions: a call to do it right.
        J Hosp Med. 2007; 2: 287-290
        • Kripalani S
        • et al.
        Promoting effective transitions of care at hospital discharge: a review of key issues for hospitalists.
        J Hosp Med. 2007; 2: 314-323
        • Forster AJ
        • et al.
        Adverse events among medical patients after discharge from hospital.
        CMAJ. 2004 Feb 3; 170: 345-349
        • Forster AJ
        • et al.
        The incidence and severity of adverse events affecting patients after discharge from the hospital.
        Ann Intern Med. 2003 Feb 4; 138: 161-167
        • Greenwald JL
        • et al.
        Making inpatient medication reconciliation patient centered, clinically relevant and implementable: a consensus statement on key principles and necessary first steps.
        J Hosp Med. 2010; 5: 477-485
        • Budnitz DS
        • et al.
        Emergency hospitalizations for adverse drug events in older Americans.
        N Engl J Med. 2011 Nov 24; 365: 2002-2012
        • Roy CL
        • et al.
        Patient safety concerns arising from test results that return after hospital discharge.
        Ann Intern Med. 2005 Jul 19; 143: 121-128
        • Moore C
        • McGinn T
        • Halm E.
        Tying up loose ends: discharging patients with unresolved medical issues.
        Arch Intern Med. 2007 Jun 25; 167: 1305-1311
        • Kripalani S
        • et al.
        Deficits in communication and information transfer between hospital-based and primary care physicians: implications for patient safety and continuity of care.
        JAMA. 2007 Feb 28; 297: 831-841
        • Jencks SF
        • Williams MV
        • Coleman EA.
        Rehospitalizations among patients in the Medicare fee-for-service program.
        N Engl J Med. 2009 Apr 2; 360: 1418-1428
        • Mor V
        • et al.
        The revolving door of rehospitalization from skilled nursing facilities.
        Health Aff (Millwood). 2010; 29: 57-64
        • Li J
        • Young R
        • Williams MV
        Optimizing transitions of care to reduce rehospitalizations.
        Cleve Clin J Med. 2014; 81: 312-320
        • Coleman EA
        • et al.
        Preparing patients and caregivers to participate in care delivered across settings: the care transitions intervention.
        J Am Geriatr Soc. 2004; 52: 1817-1825
        • Naylor M
        • et al.
        Comprehensive discharge planning for the hospitalized elderly: a randomized clinical trial.
        Ann Intern Med. 1994 Jun 15; 120: 999-1006
        • Coleman EA
        • et al.
        The care transitions intervention: results of a randomized controlled trial.
        Arch Intern Med. 2006 Sep 25; 166: 1822-1828
        • Jack BW
        • et al.
        A reengineered hospital discharge program to decrease rehospitalization: a randomized trial.
        Ann Intern Med. 2009 Feb 3; 150: 178-187
        • Williams MV
        • et al.
        Project BOOST implementation: lessons learned.
        South Med J. 2014; 107: 455-465
        • Hansen LO
        • et al.
        Interventions to reduce 30-day rehospitalization: a systematic review.
        Ann Intern Med. 2011 Oct 18; 155: 520-528
        • Naylor MD
        • et al.
        Comprehensive discharge planning and home follow-up of hospitalized elders: a randomized clinical trial.
        JAMA. 1999 Feb 17; 281: 613-620
        • Koehler BE
        • et al.
        Reduction of 30-day postdischarge hospital readmission or emergency department (ED) visit rates in high-risk elderly medical patients through delivery of a targeted care bundle.
        J Hosp Med. 2009; 4: 211-218
        • Garåsen H
        • Windspoll R
        • Johnsen R.
        Intermediate care at a community hospital as an alternative to prolonged general hospital care for elderly patients: a randomised controlled trial. 7. BMC Public Health, 2007 May 2: 68
        • Courtney M
        • et al.
        Fewer emergency readmissions and better quality of life for older adults at risk of hospital readmission: a randomized controlled trial to determine the effectiveness of a 24-week exercise and telephone follow-up program.
        J Am Geriatr Soc. 2009; 57: 395-402
        • Rich MW
        • et al.
        A multidisciplinary intervention to prevent the readmission of elderly patients with congestive heart failure.
        N Engl J Med. 1995 Nov 2; 333: 1190-1195
        • López Cabezas C
        • et al.
        Randomized clinical trial of a postdischarge pharmaceutical care program vs regular follow-up in patients with heart failure.
        Farm Hosp. 2006; 30: 328-342
        • Morkisch N
        • et al.
        Components of the Transitional Care Model (TCM) to reduce readmission in geriatric patients: a systematic review.
        BMC Geriatr. 2020 Sep 11; 20: 345
      1. Centers for Medicare & Medicaid Services. Partnership for Patients. Accessed Oct 7, 2021. https://innovation.cms.gov/innovation-models/partnership-for-patients.

      2. Centers for Medicare & Medicaid Services. Community-based Care Transitions Program. Accessed Oct 7, 2021. http://innovation.cms.gov/initiatives/CCTP/.

        • Bradley EH
        • et al.
        Quality collaboratives and campaigns to reduce readmissions: what strategies are hospitals using?.
        J Hosp Med. 2013; 8: 601-608
        • Scott AM
        • et al.
        Understanding facilitators and barriers to care transitions: insights from Project ACHIEVE site visits.
        Jt Comm J Qual Patient Saf. 2017; 43: 433-447
        • Li J
        • et al.
        Improving evidence-based grouping of transitional care strategies in hospital implementation using statistical tools and expert review.
        BMC Health Serv Res. 2021 Jan 7; 21: 35
        • Vandenbroucke JP
        • et al.
        Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaboration.
        Ann Intern Med. 2007 Oct 16; 147: W163-W194
      3. University of Kentucky Center for Clinical and Translational Science. REDCap (Vanderbilt University). Accessed Oct 7, 2021. https://redcap.uky.edu/redcap/.

        • Cook C
        • et al.
        It Takes (at Least) Two: Partnerships and Collaborations in Transitional Care. Poster presented at the 12th Annual Conference on the Science of Dissemination and Implementation in Health, Arlington, VA2019 (Dec 4–6)
        • Sorra J
        • et al.
        Development and psychometric properties of surveys to assess patient and caregiver experience with care transitions.
        BMC Health Serv Res. 2021; 21 (Aug 9): 785
        • Li J
        • et al.
        Information exchange among providers and patient-centeredness in transitional care: a five-year retrospective analysis.
        Center for Health Services Research, Lexington, KY2020
        • Cella D
        • et al.
        The Patient-Reported Outcomes Measurement Information System (PROMIS) developed and tested its first wave of adult self-reported health outcome item banks: 2005–2008.
        J Clin Epidemiol. 2010; 63: 1179-1794
        • Hays RD
        • et al.
        Development of physical and mental health summary scores from the Patient-Reported Outcomes Measurement Information System (PROMIS) global items.
        Qual Life Res. 2009; 18: 873-880
      4. Centers for Medicare & Medicaid Services. CAPHPS Hospital Survey (HCAHPS) Quality Assurance Guidelines, ver. 12.0. Mar 2017. Accessed Oct 7, 2021. http://www.hcahpsonline.org/globalassets/hcahps/quality-assurance/qag_v12.0_2017.pdf.

        • Leppin AL
        • et al.
        Preventing 30-day hospital readmissions: a systematic review and meta-analysis of randomized trials.
        JAMA Intern Med. 2014; 174: 1095-1107
        • Saha S
        • Beach MC.
        The impact of patient-centered communication on patients’ decision making and evaluations of physicians: a randomized study using video vignettes.
        Patient Educ Couns. 2011; 84: 386-392
        • Thom DH
        • Hall MA
        • Pawlson LG.
        Measuring patients’ trust in physicians when assessing quality of care.
        Health Aff (Milwwod). 2004; 23: 124-132
        • Thom DH.
        Physician behaviors that predict patient trust.
        J Fam Pract. 2001; 50: 323-328
        • Safran DG
        • et al.
        Linking primary care performance to outcomes of care.
        J Fam Pract. 1998; 47: 213-220
        • Birkhaüer J
        • et al.
        Trust in the health care professional and health outcome: a meta-analysis.
        PLoS One. 2017 Feb 7; 12e0170988
        • Fiscella K
        • et al.
        Patient trust: is it related to patient-centered behavior of primary care physicians?.
        Med Care. 2004; 42: 1049-1055
        • Murray B
        • McCrone S.
        An integrative review of promoting trust in the patient-primary care provider relationship.
        J Adv Nurs. 2015; 71: 3-23
        • Shan L
        • et al.
        Patient satisfaction with hospital inpatient care: effects of trust, medical insurance and perceived quality of care.
        PLoS One. 2016 Oct 18; 11e0164366
        • Benkert R
        • et al.
        Trust, mistrust, racial identity and patient satisfaction in urban African American primary care patients of nurse practitioners.
        J Nurs Scholarsh. 2009; 41: 211-219
        • Bohnert ASB
        • et al.
        Ratings of patient-provider communication among veterans: serious mental illnesses, substance use disorders, and the moderating role of trust.
        Health Commun. 2011; 26: 267-274
        • Blendon RJ
        • Benson JM
        • Hero JO.
        Public trust in physicians—U.S. medicine in international perspective.
        N Engl J Med. 2014 Oct 23; 371: 1570-1572
        • First Smith CP.
        do no harm: institutional betrayal and trust in health care organizations.
        J Multidiscip Healthc. 2017 Apr 3; 10: 133-144
        • Chandra S
        • Mohammadnezhad M
        • Ward P.
        Trust and communication in a doctor-patient relationship: a literature review.
        Arch Med. 2018; 3: 36
        • Clever SL
        • et al.
        Does doctor-patient communication affect patient satisfaction with hospital care? Results of an analysis with a novel instrumental variable.
        Health Serv Res. 2008; 43: 1505-1519
        • Musa D
        • et al.
        Trust in the health care system and the use of preventive health services by older Black and White adults.
        Am J Public Health. 2009; 99: 1293-1299
        • Greene J
        • Ramos C.
        A mixed methods examination of health care provider behaviors that build patients’ trust.
        Patient Educ Couns. 2021; 104: 1222-1228
        • Hillen MA
        • et al.
        How can communication by oncologists enhance patients’ trust? An experimental study.
        Ann Oncol. 2014; 25: 896-901
        • Zwingmann J
        • et al.
        Effects of patient-centered communication on anxiety, negative affect, and trust in the physician in delivering a cancer diagnosis: a randomized, experimental study.
        Cancer. 2017 Aug 15; 123: 3167-3175
        • Coleman EA
        • Berenson RA.
        Lost in transition: challenges and opportunities for improving the quality of transitional care.
        Ann Intern Med. 2004 Oct 5; 141: 533-536
        • Snow V
        • et al.
        Transitions of care consensus policy statement: American College of Physicians, Society of General Internal Medicine, Society of Hospital Medicine, American Geriatrics Society, American College Of Emergency Physicians, and Society for Academic Emergency Medicine.
        J Hosp Med. 2009; 4: 364-370
        • O'Leary KJ
        • et al.
        Hospitalized patients’ understanding of their plan of care.
        Mayo Clin Proc. 2010; 85: 47-52
        • Rennke S
        • et al.
        Hospital-initiated transitional care interventions as a patient safety strategy: a systematic review.
        Ann Intern Med. 2013 Mar 5; 158: 433-440
        • Li J
        • et al.
        Project ACHIEVE—using implementation research to guide the evaluation of transitional care effectiveness.
        BMC Health Serv Res. 2016 Feb 19; 16: 70
        • Black N.
        Why we need observational studies to evaluate the effectiveness of health care.
        BMJ. 1996 May 11; 312: 1215-1218
      5. Bipartisan Policy Center. Delivery System Reform: Improving Care for Individuals Dually Eligible for Medicare and Medicaid. Sep 2016. Accessed Oct 7, 2021. https://bipartisanpolicy.org/report/dually-eligible-medicare-medicaid/.

        • Alberti PM
        • Baker MC.
        Dual eligible patients are not the same: how social risk may impact quality measurement's ability to reduce inequities.
        Medicine (Baltimore). 2020 Sep 18; 99: e22245
        • Gorges RJ
        • Sanghavi P
        • Konetzka RT.
        A national examination of long-term care setting, outcomes, and disparities among elderly dual eligibles.
        Health Aff (Millwood). 2019; 38: 1110-1118
      6. Centers for Medicare and Medicaid Services. FY2017 national average rate. Accessed October 27, 2021. https://data.cms.gov/provider-data/#2017-annual-files.

        • Zulman DM
        • et al.
        Practices to foster physician presence and connection with patients in the clinical encounter.
        JAMA. 2020 Jan 7; 323: 70-81
        • Teutsch C.
        Patient-doctor communication.
        Med Clin North Am. 2003; 87: 1115-1145
        • Riess H
        • Kraft-Todd G.E.M.P.A.T.H.Y
        a tool to enhance nonverbal communication between clinicians and their patients.
        Acad Med. 2014; 89: 1108-1112
        • Centers for Medicare & Medicaid Services
        Hospital Readmissions Reduction Program (HRRP). 2021; (Updated: Aug 6.) Accessed Oct 7, 2021