Effect of Community Health Workers on 30-Day Hospital Readmissions in an Accountable Care Organization Population: A Randomized Clinical Trial | Physical Medicine and Rehabilitation | JAMA Network Open | JAMA Network
[Skip to Navigation]
Sign In
Visual Abstract. Effect of Community Health Workers on 30-Day Hospital Readmissions
Effect of Community Health Workers on 30-Day Hospital Readmissions
Figure 1.  Participant Recruitment Flowchart
Participant Recruitment Flowchart

CHW indicates community health worker.

Figure 2.  Postdischarge Outcomes at 30 Days for Participants Paired With Community Health Workers vs Usual Care
Postdischarge Outcomes at 30 Days for Participants Paired With Community Health Workers vs Usual Care

Adjusted for age, race/ethnicity, sex, number of hospitalizations, insurance, and living alone; discharge disposition was also applied for panel A only. ED indicates emergency department; error bars, 95% CI.

Table 1.  Patient Characteristics
Patient Characteristics
Table 2.  Patient-Reported Experience Outcomes at 30 Days
Patient-Reported Experience Outcomes at 30 Days
Table 3.  Types of CHW-Patient Contacts and Activities for Intervention Participants
Types of CHW-Patient Contacts and Activities for Intervention Participants
1.
Centers for Medicare & Medicaid Services. National Health Expenditure Data. Accessed December 18, 2017. https://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/NationalHealthExpendData
2.
Freeman  WJ, Weiss  AJ, Heslin  KC. Overview of U.S. hospital stays in 2016: variation by geographic region: statistical brief #246. In:  Healthcare Cost and Utilization Project (HCUP) Statistical Briefs. Agency for Healthcare Research and Quality; 2018.
3.
Auerbach  AD, Kripalani  S, Vasilevskis  EE,  et al.  Preventability and causes of readmissions in a national cohort of general medicine patients.   JAMA Intern Med. 2016;176(4):484-493. doi:10.1001/jamainternmed.2015.7863 PubMedGoogle ScholarCrossref
4.
McWilliams  JM, Hatfield  LA, Landon  BE, Hamed  P, Chernew  ME.  Medicare spending after 3 years of the Medicare shared savings program.   N Engl J Med. 2018;379(12):1139-1149. doi:10.1056/NEJMsa1803388 PubMedGoogle ScholarCrossref
5.
Hannan  EL, Racz  MJ, Walford  G,  et al.  Predictors of readmission for complications of coronary artery bypass graft surgery.   JAMA. 2003;290(6):773-780. doi:10.1001/jama.290.6.773 PubMedGoogle ScholarCrossref
6.
Aujesky  D, Mor  MK, Geng  M, Stone  RA, Fine  MJ, Ibrahim  SA.  Predictors of early hospital readmission after acute pulmonary embolism.   Arch Intern Med. 2009;169(3):287-293. doi:10.1001/archinternmed.2008.546 PubMedGoogle ScholarCrossref
7.
Harjai  KJ, Thompson  HW, Turgut  T, Shah  M.  Simple clinical variables are markers of the propensity for readmission in patients hospitalized with heart failure.   Am J Cardiol. 2001;87(2):234-237, A9. doi:10.1016/S0002-9149(00)01328-X PubMedGoogle ScholarCrossref
8.
Calvillo-King  L, Arnold  D, Eubank  KJ,  et al.  Impact of social factors on risk of readmission or mortality in pneumonia and heart failure: systematic review.   J Gen Intern Med. 2013;28(2):269-282. doi:10.1007/s11606-012-2235-x PubMedGoogle ScholarCrossref
9.
Hu  J, Gonsahn  MD, Nerenz  DR.  Socioeconomic status and readmissions: evidence from an urban teaching hospital.   Health Aff (Millwood). 2014;33(5):778-785. doi:10.1377/hlthaff.2013.0816 PubMedGoogle ScholarCrossref
10.
Goldfield  NI, McCullough  EC, Hughes  JS,  et al.  Identifying potentially preventable readmissions.   Health Care Financ Rev. 2008;30(1):75-91.PubMedGoogle Scholar
11.
Jack  HE, Arabadjis  SD, Sun  L, Sullivan  EE, Phillips  RS.  Impact of community health workers on use of healthcare services in the United States: a systematic review.   J Gen Intern Med. 2017;32(3):325-344. doi:10.1007/s11606-016-3922-9 PubMedGoogle ScholarCrossref
12.
We analyzed 25 population health interventions—and these 2 give the best ‘bang for your buck.’ Advisory Board. March 22, 2018. Accessed July 19, 2018. https://www.advisory.com/blog/2018/03/population-health-interventions
13.
Kangovi  S, Mitra  N, Grande  D, Long  JA, Asch  DA.  Evidence-based community health worker program addresses unmet social needs and generates positive return on investment.   Health Aff (Millwood). 2020;39(2):207-213. doi:10.1377/hlthaff.2019.00981 PubMedGoogle ScholarCrossref
14.
American Public Health Association. Support for community health worker leadership in determining workforce standards for training and credentialing. Accessed July 19, 2018. https://www.apha.org/policies-and-advocacy/public-health-policy-statements/policy-database/2015/01/28/14/15/support-for-community-health-worker-leadership
15.
Lohr  AM, Ingram  M, Nuñez  AV, Reinschmidt  KM, Carvajal  SC.  Community-clinical linkages with community health workers in the United States: a scoping review.   Health Promot Pract. 2018;19(3):349-360. doi:10.1177/1524839918754868 PubMedGoogle ScholarCrossref
16.
Malcarney  M-B, Pittman  P, Quigley  L, Horton  K, Seiler  N.  The changing roles of community health workers.   Health Serv Res. 2017;52(suppl 1):360-382. doi:10.1111/1475-6773.12657 PubMedGoogle ScholarCrossref
17.
The Rural Health Information Hub, Health Resources and Services Administration. Community Health Workers Toolkit. Accessed July 20, 2018. https://www.ruralhealthinfo.org/toolkits/community-health-workers
18.
Balcazar  H, Rosenthal  EL, Brownstein  JN, Rush  CH, Matos  S, Hernandez  L.  Community health workers can be a public health force for change in the United States: three actions for a new paradigm.   Am J Public Health. 2011;101(12):2199-2203. doi:10.2105/AJPH.2011.300386 PubMedGoogle ScholarCrossref
19.
Commonwealth of Massachusetts. Core Competencies for Community Health Workers. Revised May 2014. Accessed July 19, 2018. https://www.mass.gov/service-details/core-competencies-for-community-health-workers
20.
Reinschmidt  KM, Ingram  M, Morales  S,  et al.  Documenting community health worker roles in primary care: contributions to evidence-based integration into health care teams, 2015.   J Ambul Care Manage. 2017;40(4):305-315. doi:10.1097/JAC.0000000000000178 PubMedGoogle ScholarCrossref
21.
Brownstein  JN, Chowdhury  FM, Norris  SL,  et al.  Effectiveness of community health workers in the care of people with hypertension.   Am J Prev Med. 2007;32(5):435-447. doi:10.1016/j.amepre.2007.01.011 PubMedGoogle ScholarCrossref
22.
Rothschild  SK, Martin  MA, Swider  SM,  et al.  Mexican American trial of community health workers: a randomized controlled trial of a community health worker intervention for Mexican Americans with type 2 diabetes mellitus.   Am J Public Health. 2014;104(8):1540-1548. doi:10.2105/AJPH.2013.301439 PubMedGoogle ScholarCrossref
23.
Wells  KJ, Luque  JS, Miladinovic  B,  et al.  Do community health worker interventions improve rates of screening mammography in the United States: a systematic review.   Cancer Epidemiol Biomarkers Prev. 2011;20(8):1580-1598. doi:10.1158/1055-9965.EPI-11-0276 PubMedGoogle ScholarCrossref
24.
Percac-Lima  S, Grant  RW, Green  AR,  et al.  A culturally tailored navigator program for colorectal cancer screening in a community health center: a randomized, controlled trial.   J Gen Intern Med. 2009;24(2):211-217. doi:10.1007/s11606-008-0864-x PubMedGoogle ScholarCrossref
25.
Krieger  JW, Takaro  TK, Song  L, Weaver  M.  The Seattle-King County Healthy Homes Project: a randomized, controlled trial of a community health worker intervention to decrease exposure to indoor asthma triggers.   Am J Public Health. 2005;95(4):652-659. doi:10.2105/AJPH.2004.042994 PubMedGoogle ScholarCrossref
26.
Venditti  EM, Zgibor  JC, Vander Bilt  J,  et al.  Mobility and Vitality Lifestyle Program (MOVE UP): a community health worker intervention for older adults with obesity to improve weight, health, and physical function.   Innov Aging. 2018;2(2):igy012. doi:10.1093/geroni/igy012 PubMedGoogle Scholar
27.
Kangovi  S, Mitra  N, Grande  D,  et al.  Patient-centered community health worker intervention to improve posthospital outcomes: a randomized clinical trial.   JAMA Intern Med. 2014;174(4):535-543. doi:10.1001/jamainternmed.2013.14327 PubMedGoogle ScholarCrossref
28.
Balaban  RB, Galbraith  AA, Burns  ME, Vialle-Valentin  CE, Larochelle  MR, Ross-Degnan  D.  A patient navigator intervention to reduce hospital readmissions among high-risk safety-net patients: a randomized controlled trial.   J Gen Intern Med. 2015;30(7):907-915. doi:10.1007/s11606-015-3185-x PubMedGoogle ScholarCrossref
29.
Kangovi  S, Mitra  N, Norton  L,  et al.  Effect of community health worker support on clinical outcomes of low-income patients across primary care facilities: a randomized clinical trial.   JAMA Intern Med. 2018;178(12):1635-1643. doi:10.1001/jamainternmed.2018.4630 PubMedGoogle ScholarCrossref
30.
Kangovi  S, Mitra  N, Grande  D, Huo  H, Smith  RA, Long  JA.  Community health worker support for disadvantaged patients with multiple chronic diseases: a randomized clinical trial.   Am J Public Health. 2017;107(10):1660-1667. doi:10.2105/AJPH.2017.303985 PubMedGoogle ScholarCrossref
31.
Finkelstein  A, Zhou  A, Taubman  S, Doyle  J.  Health care hotspotting—a randomized controlled trial.   N Engl J Med. 2020;382(2):152-162. doi:10.1056/NEJMsa1906848 PubMedGoogle ScholarCrossref
32.
Carter  J, Walton  A, Donelan  K, Thorndike  A.  Implementing community health worker-patient pairings at the time of hospital discharge: a randomized control trial.   Contemp Clin Trials. 2018;74:32-37. doi:10.1016/j.cct.2018.09.013 PubMedGoogle ScholarCrossref
33.
Carter  J, Ward  C, Wexler  D, Donelan  K.  The association between patient experience factors and likelihood of 30-day readmission: a prospective cohort study.   BMJ Qual Saf. 2018;27(9):683-690. doi:10.1136/bmjqs-2017-007184 PubMedGoogle ScholarCrossref
34.
Carter  J, Ward  C, Thorndike  A, Donelan  K, Wexler  DJ.  Social Factors and Patient Perceptions Associated With Preventable Hospital Readmissions.   J Patient Exp. 2020;7(1):19-26. doi:10.1177/2374373518825143PubMedGoogle ScholarCrossref
35.
Fisher  SR, Graham  JE, Krishnan  S, Ottenbacher  KJ.  Predictors of 30-day readmission following inpatient rehabilitation for patients at high risk for hospital readmission.   Phys Ther. 2016;96(1):62-70. doi:10.2522/ptj.20150034 PubMedGoogle ScholarCrossref
36.
Ottenbacher  KJ, Karmarkar  A, Graham  JE,  et al.  Thirty-day hospital readmission following discharge from postacute rehabilitation in fee-for-service Medicare patients.   JAMA. 2014;311(6):604-614. doi:10.1001/jama.2014.8 PubMedGoogle ScholarCrossref
37.
Vasan  A, Morgan  JW, Mitra  N,  et al.  Effects of a standardized community health worker intervention on hospitalization among disadvantaged patients with multiple chronic conditions: a pooled analysis of three clinical trials.   Health Serv Res. 2020;55(suppl 2):894-901. doi:10.1111/1475-6773.13321 PubMedGoogle ScholarCrossref
38.
Nuti  LA, Lawley  M, Turkcan  A,  et al.  No-shows to primary care appointments: subsequent acute care utilization among diabetic patients.   BMC Health Serv Res. 2012;12(1):304. doi:10.1186/1472-6963-12-304 PubMedGoogle ScholarCrossref
39.
Ellis  DA, McQueenie  R, McConnachie  A, Wilson  P, Williamson  AE.  Demographic and practice factors predicting repeated non-attendance in primary care: a national retrospective cohort analysis.   Lancet Public Health. 2017;2(12):e551-e559. doi:10.1016/S2468-2667(17)30217-7 PubMedGoogle ScholarCrossref
40.
Gimpel  N, Marcee  A, Kennedy  K, Walton  J, Lee  S, DeHaven  MJ.  Patient perceptions of a community-based care coordination system.   Health Promot Pract. 2010;11(2):173-181. doi:10.1177/1524839908320360 PubMedGoogle ScholarCrossref
41.
Brooks  BA, Davis  S, Frank-Lightfoot  L, Kulbok  PA, Poree  S, Sgarlata  L.  Building a Community Health Worker Program: The Key to Better Care, Better Outcomes, & Lower Costs. CommunityHealth Works; 2018.
42.
Covert  H, Sherman  M, Miner  K, Lichtveld  M.  Core competencies and a workforce framework for community health workers: a model for advancing the profession.   Am J Public Health. 2019;109(2):320-327. doi:10.2105/AJPH.2018.304737 PubMedGoogle ScholarCrossref
43.
Rosenthal  EL, Rush  CH, Allen  C. Understanding scope and competencies: a contemporary look at the United States community health worker field: progress report of the Community Health Worker (CHW) Core Consensus Project. Accessed October 5, 2020. https://sph.uth.edu/dotAsset/28044e61-fb10-41a2-bf3b-07efa4fe56ae.pdf
44.
Ingram  M, Reinschmidt  KM, Schachter  KA,  et al.  Establishing a professional profile of community health workers: results from a national study of roles, activities and training.   J Community Health. 2012;37(2):529-537. doi:10.1007/s10900-011-9475-2 PubMedGoogle ScholarCrossref
45.
Arvey  SR, Fernandez  ME.  Identifying the core elements of effective community health worker programs: a research agenda.   Am J Public Health. 2012;102(9):1633-1637. doi:10.2105/AJPH.2012.300649 PubMedGoogle ScholarCrossref
46.
Lapidos  A, Lapedis  J, Heisler  M.  Realizing the value of community health workers—new opportunities for sustainable financing.   N Engl J Med. 2019;380(21):1990-1992. doi:10.1056/NEJMp1815382 PubMedGoogle ScholarCrossref
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    Views 5,172
    Citations 0
    Original Investigation
    Public Health
    May 20, 2021

    Effect of Community Health Workers on 30-Day Hospital Readmissions in an Accountable Care Organization Population: A Randomized Clinical Trial

    Author Affiliations
    • 1Department of Medicine, Massachusetts General Hospital, Boston
    • 2Division of General Internal Medicine, Massachusetts General Hospital, Boston
    • 3Mongan Institute, Massachusetts General Hospital, Boston
    • 4Brandeis University, Waltham, Massachusetts
    • 5Division of General Internal Medicine, Massachusetts General Hospital, Harvard Medical School, Boston
    JAMA Netw Open. 2021;4(5):e2110936. doi:10.1001/jamanetworkopen.2021.10936
    Key Points

    Question  Does community health worker care reduce 30-day hospital readmissions in inpatient adults participating in accountable care organization plans?

    Findings  In this randomized clinical trial including 550 adults, intervention patients were significantly less likely to experience 30-day hospital readmissions than control participants. In post hoc subgroup analysis, the effect remained significant for participants discharged to short-term rehabilitation but not for those discharged home.

    Meaning  In this study, community health workercare improved postdischarge outcomes in clinically complex patients insured by accountable care organizations, particularly for those discharged to short-term rehabilitation.

    Abstract

    Importance  Value-based care within accountable care organizations (ACOs) has magnified the importance of reducing preventable hospital readmissions. Community health worker (CHW) interventions may address patients’ unmet psychosocial and clinical care needs but have been underused in inpatient and postdischarge care.

    Objective  To determine if pairing hospitalized patients with ACO insurance with CHWs would reduce 30-day readmission rates.

    Design, Setting, and Participants  This randomized clinical trial was conducted in 6 general medicine hospital units within 1 academic medical center in Boston, Massachusetts. Participants included adults hospitalized from April 1, 2017, through March 31, 2019, who had ACO insurance and were at risk for 30-day readmission based on a hospital readmission algorithm. The main inclusion criterion was frequency of prior nonelective hospitalizations (≥2 in the past 3 months or ≥3 in the 12 months prior to enrollment). Data were analyzed from February 1, 2018, through March 3, 2021.

    Intervention  CHWs met with intervention participants prior to discharge and maintained contact for 30 days postdischarge to assist participants with clinical access and social resources via telephone calls, text messages, and field visits. CHWs additionally provided psychosocial support and health coaching, using motivational interviewing, goal-setting, and other behavioral strategies. The control group received usual care, which included routine care from primary care clinics and any outpatient referrals made by hospital case management or social work at the time of discharge.

    Main Outcomes and Measures  The primary outcome was 30-day hospital readmissions. Secondary outcomes included 30-day missed primary care physician or specialty appointments.

    Results  A total of 573 participants were enrolled, and 550 participants (mean [SD] age, 70.1 [15.7] years; 266 [48.4%] women) were included in analysis, with 277 participants randomized to the intervention group and 273 participants randomized to the control group. At baseline, participants had a mean (SD) of 3 (0.8) hospitalizations in the prior 12 months. There were 432 participants (78.5%) discharged home and 127 participants (23.1%) discharged to a short rehabilitation stay prior to returning home. Compared with participants in the control group, participants in the intervention group were less likely to be readmitted within 30 days (odds ratio [OR], 0.44; 95% CI, 0.28-0.90) and to miss clinic appointments within 30 days (OR, 0.56; 95% CI, 0.38-0.81). A post hoc subgroup analysis showed that compared with control participants, intervention participants discharged to rehabilitation had a reduction in readmissions (OR, 0.09; 95% CI, 0.03-0.31), but there was no significant reduction for those discharged home (OR, 0.68; 95% CI, 0.41-1.12).

    Conclusions and Relevance  This randomized clinical trial found that pairing ACO-insured inpatient adults with CHWs reduced readmissions and missed outpatient visits 30 days postdischarge. The effect was significant for those discharged to short-term rehabilitation but not for those discharged home.

    Trial Registration  ClinicalTrials.gov Identifier: NCT03085264

    Introduction

    Hospitalizations account for one-third of US health care costs.1,2 Readmissions in the first 30 days after hospital discharge are common, and approximately 27% of adult 30-day readmissions are estimated to be preventable.3 Increasing movement toward value-based care within accountable care organizations (ACOs) has magnified the importance of reducing preventable readmissions.4 Factors associated with hospital readmissions include medical complexity,5 clinical comorbidities,6,7 and social determinants of health.8-10 Many health care organizations have increasing interest in developing interventions within the ACO framework that address clinical care gaps and unmet social needs. Integration of community health workers (CHWs) is one of few interventions that have generated promising outcomes in terms of reducing hospitalizations and health care costs.11-13

    CHWs are trained to have basic knowledge of clinical conditions and to provide health coaching using motivational interviewing and psychosocial support.14-16 CHWs also have expertise in social determinants of health and can assist patients with economic, health care access–associated, social, educational, and environmental resources to help close gaps in patient health care.17,18 Specifically, CHWs can augment patient engagement by strengthening patient connections to primary care and identifying relevant community-, state-, or federal-based patient resources to meet unmet patient needs (eg, food, housing, transportation).19,20

    While most CHW studies have focused on specific disease-based cohorts,21-26 some randomized clinical trials have examined the effectiveness of using CHWs to improve postdischarge outcomes in adult inpatient and outpatient general medicine populations. In a study of 222 Medicaid inpatients randomized to a CHW intervention, Kangovi et al27 found that having at least one 30-day readmission did not differ between the intervention and control groups; however, there was a significant reduction in the number of intervention participants with 2 or more 30-day readmissions. In a randomized clinical trial of 1009 inpatients older than 60 years, Balaban et al28 found that CHW care reduced 30-day readmission rates in intervention participants compared with control participants. Another randomized clinical trial by Kangovi et al29 of 592 primary care outpatients tested a 6-month CHW intervention and demonstrated a reduced likelihood of readmission. Two other randomized clinical trials examining 6-month interventions with CHWs demonstrated no significant difference in hospitalizations.30,31

    To determine the effect of CHW care delivery on 30-day readmissions within an ACO population, we conducted a randomized clinical trial to test a 30-day CHW intervention for patients admitted to the internal medicine service in an academic medical center in Boston, Massachusetts. The hypothesis was that CHW care delivery initiated in the hospital and extending for 30 days after discharge would reduce 30-day readmissions compared with usual care.

    Methods

    This randomized clinical trial was approved by the of Partners Human Research Committee (Trial Protocol in Supplement 1; eAppendix in Supplement 2). All enrolled patients provided written informed consent for study enrollment. This study is reported following the Consolidated Standards of Reporting Trials (CONSORT) reporting guideline for randomized clinical trials.

    Trial Design

    The Community Care Transitions study was a randomized clinical trial was designed to improve health care outcomes for patients at high risk for readmission by pairing CHWs with inpatients for 30 days after discharge.32 CHW care delivery was implemented in partnership with clinical teams that were unaware of outcomes until trial completion. The CHWs addressed and integrated patient-identified needs (eg, food, housing, transportation) into the clinical care plans and used motivational interviewing and psychosocial support strategies to improve adherence to clinical care.

    Setting and Participants

    The study was conducted at Massachusetts General Hospital, a 999-bed teaching hospital in Boston, Massachusetts. Six internal medicine hospital units (or clinical wards) were used for trial recruitment. Each unit had similar percentages of 30-day readmissions with no differences in the diagnoses or ages of hospitalized patients.

    Eligibility criteria were developed based on findings from 2 prior studies by Carter et al.33,34 Briefly, patients aged 18 years or older who were admitted to 1 of the study units were potentially eligible if they met the high-risk hospital criteria for 30-day readmission, with threshold of 16% or greater readmission risk. This was determined by a Massachusetts General Hospital–based 12-factor risk algorithm that included prior hospitalizations, fall risk, wound care needs, and frailty. After being identified by this algorithm, patients were evaluated by research staff for additional study inclusion criteria: prior history of 2 or more nonelective hospitalizations in the 3 months prior to enrollment or 3 or more nonelective hospitalizations in the 12 months prior to enrollment; participation in a hospital-based ACO benefit (Medicare, Medicaid, or private insurance); living within a 20-mile radius of the main hospital; having access to a working telephone; being fluent in English; having 1 or more unmet care-related needs identified during inpatient multidisciplinary rounds (eg, difficulties with medication management, appointment scheduling, access to transportation, or social support); and having a primary care physician (PCP). Patients were ineligible if they were experiencing homelessness, unable to provide consent owing to cognitive impairment, or had an invoked health care proxy or prisoner status (Figure 1). Patients with preexisting outpatient program support (ie, integrated care management services with telephonic nursing, social work, and clinical resource care coordination), home nursing, or other supportive programming (eg, physical therapy) were eligible for participation in the trial. Patients living in nursing homes or discharged to long-term care were excluded prior to randomization.

    Enrollment and Randomization

    Patients were identified with a hospital-based database that generated a daily report of inpatients at high risk for readmission. Eligible patients were discussed at multidisciplinary rounds daily on each inpatient unit with CHW staff, case managers, unit nurses, bedside nurses, and physicians who reviewed the potential barriers to discharge. After confirming eligibility, research staff approached patients for enrollment. After providing consent and agreeing to enrollment, patients were randomized by study staff via block randomization using the team statistician’s preloaded Excel spreadsheet version 16.0 (Microsoft) block with 6 participants per block to either the intervention or the control group. To allow enough time for enrollment procedures, most participants had to be enrolled prior to finalization of their discharge plans; therefore, it was not feasible to preidentify which participants would be discharged to home or to a short-term rehabilitation facility.

    Intervention Care

    CHWs met with intervention participants and inpatient multidisciplinary teams prior to discharge to establish goals identified by patients and clinical teams. After discharge, CHWs delivered care using multiple communication strategies, including telephone calls, text messages, home visits, rehabilitation facility visits, and field visits (eg, accompanying patient for medical or social service encounters). CHWs provided health coaching, as well as assistance with obtaining any needed clinical access or social resources. CHWs were trained in CHW core competencies,19 consisting of motivational interviewing, goal-setting, behavior change, and psychosocial support.

    Three CHWs delivered the study intervention, and all had experience working with or living in communities similar to those of participants. All CHWs received 1 month of protocol training led by the study CHW supervisor. CHW core competencies were emphasized during training and applied to case-based scenarios. CHWs documented all encounters in the electronic medical record (EMR) (eg, enrollment notes, progress notes for in-person visits and telephone calls); all patient and care team interactions were documented by CHWs in a REDCap database (Vanderbilt University). Each patient’s clinical team members were copied on all CHW EMR notes and contacted directly by CHWs during the intervention when needed. CHW staff also communicated directly with any practitioners of each patient’s outpatient primary care based–support, such as care management, nursing, or physical therapy, to streamline care delivery.

    Usual Care

    Usual care was defined as routine postdischarge care from the hospital discharge team. Any outpatient referrals made by hospital case management (eg, visiting nursing association, physical therapy, occupational therapy) or social work (eg, meal preparation, transportation, elder care services) at the time of discharge were also considered part of usual care.

    Measures

    Study participants completed an enrollment questionnaire adapted from a previous survey instrument.26 This survey instrument was derived from some standard established measures of patient experience for benchmarking as well as validated questions generated by prestudy qualitative interviews with patients and physicians. Questionnaire domains included health-related social needs (eg, food, housing, transportation), perceptions of their physical and mental health, confidence in their ability to care for themselves after discharge, satisfaction with inpatient care, perceived likelihood of readmission, understanding of the care plan, and ability to independently perform activities of daily living. Basic demographic characteristics, insurance status, primary diagnosis associated with admission, and major medical and psychiatric comorbidities were collected by EMR review. All participants were asked to complete a 30-day postdischarge questionnaire that included questions to assess perceived likelihood of 30-day readmission and confidence in caring for oneself outside the hospital.

    Outcomes

    The primary outcome was 30-day hospital readmission during the study period. Prespecified secondary outcomes were 30-day postdischarge missed outpatient physician appointments and emergency department (ED) visits. Outcomes were determined by the number of participants with 1 or more hospital admissions, ED visits (including ED observation stays but excluding ED visits resulting in hospitalization), or missed appointments during the 30 days after discharge from the index hospitalization. Automated email notifications of Partners Healthcare admissions and ED visits were generated by a hospital readmissions database and sent to study staff within 1 hour of occurrence. Hospital admissions were captured at Partners Healthcare–affiliated hospitals in Massachusetts (3 academic hospitals; 6 community hospitals) and New Hampshire (1 community hospital). Missed appointments were captured similarly by a hospital database and tabulated quarterly. Since the EMR was unable to capture encounters outside Partners Healthcare, participants were asked about clinical encounters during a 30-day postdischarge questionnaire. Owing to the nature of the intervention, patients and CHW staff were unable to be blinded to the study treatment arms. Study staff collecting EMR and patient-reported outcomes and performing data analysis were blinded. Clinical outcomes were adjudicated independently of the trial.

    Power Calculation

    For the power calculation, we assumed a readmission rate of 18% for usual care and a readmission rate of 13% with use of CHW care delivery, requiring a sample size of 1200 adults (600 intervention and 600 control) with more than 90% power to reject the null hypothesis that the readmission rate was greater than 13% using 1-sided binomial testing with 5% type I error. However, we were unable to enroll 1200 patients within the study funding timeline; therefore, recruitment was stopped prior to achieving the planned sample size.

    Statistical Analysis

    Demographic characteristics and baseline survey item responses were summarized between intervention and control groups. Bivariate analyses, using Pearson χ2 tests for categorical variables and t test for continuous variables, were performed to assess differences between binary clinical outcomes (30-day hospital readmission, missed appointments, ED visits) and study arms, as well as demographic characteristics and survey item responses. For every clinical outcome, a logistic regression model was applied to obtain unadjusted and adjusted odds ratios (ORs) to assess the intervention effect. Covariates adjusted for in the model included age, race/ethnicity, sex, number of hospitalizations, insurance, living alone status, and discharge disposition. Race/ethnicity was self-reported and collected as a potential confounder of readmissions, missed appointments, and ED visits. For 30-day patient experience outcomes, patient responses at admission and 30 days after discharge were compared, and a difference-in-differences analysis was performed. P values were 2-sided, and P < .05 was considered statistically significant. A separate univariate analysis was performed to identify types of CHW-patient contact along with categories of resources or care delivery administered. All analyses were performed using SAS statistical software version 9.4 (SAS Institute). Data were analyzed from February 1, 2018, through March 3, 2021.

    Results

    Patients were enrolled April 1, 2017, through March 31, 2019. A total of 1843 patients at high risk for readmission were screened, 1230 patients were deemed eligible, and 573 patients were randomized to the intervention (288 patients) or control (285 patients) groups (Figure 1). Six participants (1.0%), including 4 from the intervention group and 2 from the control group, withdrew from the study, and 17 participants (3.1%), including 7 from the intervention group and 10 from the control group, died prior to completing the study and were excluded. The remaining 550 participants were included in the final analyses, with 277 participants in the intervention group and 273 participants in the control group.

    The mean age (SD) of participants was 70.1 (15.7) years, and 266 (48.4%) were women (Table 1). The most common insurance was Medicare (388 participants [70.5%]). All trial participants had a mean (SD) of 3 (0.8) hospitalizations in the 12 months prior to index hospitalization (Table 1). Overall, 67 participants (24.5%) in the control group and 60 participants (21.7%) in the intervention group were discharged to rehabilitation. The mean (SD) length of participant rehabilitation stay was 3.9 (1.1) days. Identified covariates were balanced in intervention and control groups.

    Overall, compared with the control group, fewer participants in the intervention group were readmitted in the 30 days after hospital discharge (67 participants [24.5%] vs 35 participants [12.6%]; OR, 0.44; 95% CI, 0.28-0.90; P < .001) (Figure 2A). In a post hoc subgroup analysis of this cohort (eTable 2 in Supplement 2), we found that intervention participants discharged to rehabilitation demonstrated a reduction of 32.3 percentage points in readmissions compared with control participants (3 participants [5.0%] vs 25 participants [37.3%]; OR, 0.09; 95% CI, 0.03-0.31; P < .001) (Figure 2B), but the difference observed in intervention vs control participants discharged home was not statistically significant (32 participants [14.7%] vs 42 participants [20.4%]; OR, 0.68; 95% CI, 0.41-1.12; P = .13) (Figure 2C). A total of 4 readmissions occurred outside the hospital system and were identified by the patient questionnaire.

    Fewer intervention than control participants had missed appointments (61 participants [22.0%] vs 92 participants [33.7%]; OR, 0.56; 95% CI, 0.38-0.81; P = .002), but the difference in ED visits was not statistically significant (31 participants [11.2%] vs 46 participants [16.8%]; OR, 0.62; 95% CI, 0.38-1.02; P = .06) (Figure 2A). Similar effects were observed in the subgroup analysis among intervention vs control participants discharged to rehabilitation (missed appointments: 13 participants [21.7%] vs 24 participants [35.8%]; OR, 0.5; 95% CI, 0.22-1.09; P = .08; ED visits: 5 participants [8.3%] vs 14 participants [20.9%]; OR = 0.34; 95% CI, 0.12-1.02; P = .05) (Figure 2B) or to home (missed appointments: 48 participants [22.1%] vs 68 participants [33.0%]; OR,  = 0.58; 95% CI, 0.37-0.89; P = .01; ED visits: 26 participants [12.0%] vs 32 participants [15.5%]; OR, 0.74; 95% CI, 0.42-1.29; P = .30) (Figure 2C). In multivariate analyses of all participants controlling for demographic and clinical covariates (ie, age, race/ethnicity, sex, number of hospitalizations, insurance, living alone status, and discharge to home vs rehabilitation), the adjusted ORs for clinical outcomes in intervention vs control participants were similar to the unadjusted ORs (30-day hospital readmissions: adjusted OR, 0.45; 95% CI, 0.29-0.72; missed appointments: adjusted OR, 0.56; 95% CI, 0.38-0.82; ED visits: adjusted OR, 0.62; 95% CI, 0.38-1.02) (eTable 1 in the Supplement). A similar analysis was performed for participants discharged home and to short-stay rehabilitation (eTable 3 and eTable 4 in the Supplement).

    More than 80% of all participants indicated they were very or somewhat confident in caring for themselves at the time of enrollment, including 234 participants (84.5%) in the intervention group and 240 participants (87.9%) in the control group, and this did not change in their poststudy questionnaire (Table 2). Intervention participants demonstrated a decrease in their perceived likelihood of 30-day readmission in the poststudy questionnaire compared with the enrollment questionnaire (24 participants [12.5%] vs 53 participants [19.1%]; P = .04); this difference was not statistically different from the control group.

    Intervention participants had a mean (SD) of 3.2 (2.4) contacts with CHWs after hospital discharge during the 30-day study period. Of 277 intervention participants, 247 (88.9%) communicated with CHW staff during at least 1 phone call. In addition, 198 participants (71.4%) in the intervention group received at least 1 home, rehabilitation, or field visit during the study interval (Table 3). CHWs completed different types of interactions focused on medical, social, or basic needs and coaching or education. CHWs most commonly provided participants with counseling to reinforce adherence to their clinical care plans (239 participants [86.3%]) and with psychosocial support (229 participants [82.7%]). Other CHW-patient activities ranged widely across supportive clinical and social domains (Table 3), including making and confirming clinical appointments (128 participants [46.2%]), assistance with securing basic needs like food or housing (108 participants [39.1%]), helping with access to medications (97 participants [35.0%]), creating plans for reliable transportation (91 participants [32.9%]), initiating elder care services (80 participants [28.9%]), engaging case management support (76 participants [27.4%]), and assisting with completing insurance forms or obtaining benefits (75 participants [27.0%]).

    Discussion

    In this randomized clinical trial at 1 academic medical center, a CHW intervention reduced 30-day hospital readmissions in adult general medicine inpatients by nearly 50%. However, subgroup analyses revealed that most of the effect occurred for participants initially discharged to short-term rehabilitation. Intervention participants also were less likely to miss clinic appointments, but no significant reductions in ED visits were noted. These results indicate that CHW interventions may help reduce hospital readmissions and improve preventive care among some clinically complex patients within an ACO.

    A subgroup analysis of participants who were discharged directly home compared with those discharged to short-term rehabilitation before going home demonstrated that the CHW intervention effect on 30-day readmissions was large for participants who went to rehabilitation. Prior studies have shown that deconditioned patients with complex comorbidities who are discharged to rehabilitation facilities have elevated rates of readmission, ranging from 28% to 75%.35,36 The findings of our study suggest that CHWs had a significant effect on preventing readmissions during or after short rehabilitation stays. Potential reasons for this effect may be that CHWs addressed unmet medical and social needs that occurred during the transition from rehabilitation to home and that CHWs improved communication among the patient, rehabilitation staff, and primary physician prior to return to home. While inpatients discharged to rehabilitation prior to transitioning home have been identified as high-risk for 30-day readmission, little research has been done to determine effective interventions to reduce readmissions. Results from our post hoc analyses demonstrated that CHWs are a promising intervention for this population, but future studies are needed to confirm these findings.

    Most CHW-focused trials have not demonstrated significant reductions in hospital readmissions. However, a 2020 pooled analysis of 3 CHW randomized clinical trials37 found a combined significance in reduced hospital readmissions. The Community Care Transitions intervention used in this study differed from prior studies in that all participants were affiliated with a hospital ACO insurance benefit and had a PCP, a working phone number, and a residential address within a specific radius. Preexisting participant connections to primary care and the ACO network were essential for effective communication between CHWs and clinical teams and for assisting participants with access to postdischarge care. This was key to CHWs connecting participants with resources and programs when they needed them. As in most prior published trials, CHWs in our study received unique training focused on reaching patients after discharge and identifying resources for patient health-related social needs. Each of these elements likely contributed to improved clinical outcomes.

    The results of this randomized clinical trial also emphasized that even in an ACO where there are more resources than in prior published CHW-intervention focused trials, readmission rates for control participants were relatively high. Although an equal proportion of intervention and control participants were enrolled in comprehensive case management or nursing programs as a part of usual care in an ACO, adding CHW care significantly reduced readmission rates. This difference highlights CHW capacity to address gaps in care related to unmet psychosocial needs (eg, securing meal delivery, transportation, access to medications, elder care services, accompaniment to clinic visits). CHWs can add value to the ACO model in a large health care organization by better connecting patients to community resources and programs as well as primary and subspecialty clinical homes.

    Overall, there was a reduction of more than 30 percentage points in missed appointments among intervention participants 30 days after discharge compared with control participants. This effect was sustained in the subgroup analysis for participants discharged to rehabilitation and home. This difference has been demonstrated in prior care transitions studies38,39 and underlines the efficacy of CHWs in improving postdischarge follow-up. Baseline and postintervention period surveys demonstrated a reduction in perceived likelihood of hospital readmission for intervention participants, although this reduction was not statistically significantly different from the control group. This suggests that CHWs may promote increased competence, knowledge, and engagement influencing perceptions about readmission.40,41

    The most frequent CHW activities were psychosocial support and reinforcement of adherence to care plans. These activities were similar to those described in studies by Kangovi et al21 and Wells et al,23 but most studies to date have not established a standard of care for CHW interventions, making direct comparisons between studies challenging. This represents an area of opportunity to more completely categorize and describe CHW outreach as part of an evidence base.42-45

    Limitations

    This study has limitations. Despite use of validated self-reported measures of health care utilization in our 30-day postdischarge survey, we may not have identified all encounters occurring outside our hospital system. However, the fact that all participants were within the ACO network helped ensure that enrolled participants received most, if not all, of their care within designated ACO coverage. Also, healthy user bias, with patients who were the most ill being unable to enroll, may have resulted in underrepresentation of patients with even higher rates of medical complexity. While we considered individuals discharged to rehabilitation to be a legitimate group to study, study participants were enrolled and randomized while hospitalized, and it was not possible to determine who was going to rehabilitation vs home until the time of discharge. Therefore, we were unable to stratify on this characteristic at the time of randomization. Approximately 32% of 1280 patients identified as eligible for the study were unable to be enrolled owing to being busy with clinical care or discharged prior to meeting with study staff. In the future, this could be addressed by having CHWs contact patients via phone after discharge. We were unable to enroll non–English-speaking participants because of limited funding for bilingual study materials and staff. The study was conducted at a single urban hospital and enrolled ACO-insured inpatients with a working telephone and who lived within a 20-mile hospital radius. Therefore, study findings may not be generalizable to patients who have non-ACO health insurance, are uninsured, do not speak English, do not have a working telephone, or live in rural settings.

    Conclusions

    This randomized clinical trial found that integration of CHWs into clinical care improved preventive care and reduced 30-day readmissions for patients discharged to rehabilitation. Reducing preventable readmissions is a central priority for the Centers for Medicare & Medicaid Services and principal health care stakeholders. As US health care organizations continue to adopt ACO models with the goal of achieving higher quality care at lower costs, policies supporting insurance-based reimbursement for CHW care and investment in comprehensive training and integration of CHWs as valued members of patient care teams will continue to be critical.46 Future research is needed to identify which patients benefit most from certain CHW intervention activities.

    Back to top
    Article Information

    Accepted for Publication: March 30, 2021.

    Published: May 20, 2021. doi:10.1001/jamanetworkopen.2021.10936

    Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2021 Carter J et al. JAMA Network Open.

    Corresponding Author: Jocelyn Carter, MD, MPH, Department of Medicine, Massachusetts General Hospital, 55 Fruit St, Blake 15, Boston, MA 02114 (jcarter0@partners.org).

    Author Contributions: Dr Carter had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

    Concept and design: Carter, Walton, Donelan.

    Acquisition, analysis, or interpretation of data: Carter, Hassan, Yu, Donelan, Thorndike.

    Drafting of the manuscript: Carter, Yu, Thorndike.

    Critical revision of the manuscript for important intellectual content: Carter, Hassan, Walton, Donelan, Thorndike.

    Statistical analysis: Carter, Yu, Donelan, Thorndike.

    Obtained funding: Carter.

    Administrative, technical, or material support: Carter, Donelan.

    Supervision: Carter, Donelan, Thorndike.

    Conflict of Interest Disclosures: Dr Donelan reported receiving grants from the Patient-Centered Outcomes Research Institute, Gordon and Betty Moore Foundation, American Cancer Society, National Institutes of Health, and Physicians Foundation and serving as an unpaid reviewer for the National Academies of Sciences, Engineering, and Medicine outside the submitted work. No other disclosures were reported.

    Funding/Support: This study was supported by grants from the Partners Healthcare Center for Population Health (Dr Carter) and the Healthcare Transformation Lab at Mass General (Dr Carter).

    Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

    Data Sharing Statement: See Supplement 3.

    Additional Contributions: The Boston Public Health Association and the Massachusetts Association of Community Health Workers provided strategic project planning support. Both were compensated for their contributions.

    References
    1.
    Centers for Medicare & Medicaid Services. National Health Expenditure Data. Accessed December 18, 2017. https://www.cms.gov/Research-Statistics-Data-and-Systems/Statistics-Trends-and-Reports/NationalHealthExpendData
    2.
    Freeman  WJ, Weiss  AJ, Heslin  KC. Overview of U.S. hospital stays in 2016: variation by geographic region: statistical brief #246. In:  Healthcare Cost and Utilization Project (HCUP) Statistical Briefs. Agency for Healthcare Research and Quality; 2018.
    3.
    Auerbach  AD, Kripalani  S, Vasilevskis  EE,  et al.  Preventability and causes of readmissions in a national cohort of general medicine patients.   JAMA Intern Med. 2016;176(4):484-493. doi:10.1001/jamainternmed.2015.7863 PubMedGoogle ScholarCrossref
    4.
    McWilliams  JM, Hatfield  LA, Landon  BE, Hamed  P, Chernew  ME.  Medicare spending after 3 years of the Medicare shared savings program.   N Engl J Med. 2018;379(12):1139-1149. doi:10.1056/NEJMsa1803388 PubMedGoogle ScholarCrossref
    5.
    Hannan  EL, Racz  MJ, Walford  G,  et al.  Predictors of readmission for complications of coronary artery bypass graft surgery.   JAMA. 2003;290(6):773-780. doi:10.1001/jama.290.6.773 PubMedGoogle ScholarCrossref
    6.
    Aujesky  D, Mor  MK, Geng  M, Stone  RA, Fine  MJ, Ibrahim  SA.  Predictors of early hospital readmission after acute pulmonary embolism.   Arch Intern Med. 2009;169(3):287-293. doi:10.1001/archinternmed.2008.546 PubMedGoogle ScholarCrossref
    7.
    Harjai  KJ, Thompson  HW, Turgut  T, Shah  M.  Simple clinical variables are markers of the propensity for readmission in patients hospitalized with heart failure.   Am J Cardiol. 2001;87(2):234-237, A9. doi:10.1016/S0002-9149(00)01328-X PubMedGoogle ScholarCrossref
    8.
    Calvillo-King  L, Arnold  D, Eubank  KJ,  et al.  Impact of social factors on risk of readmission or mortality in pneumonia and heart failure: systematic review.   J Gen Intern Med. 2013;28(2):269-282. doi:10.1007/s11606-012-2235-x PubMedGoogle ScholarCrossref
    9.
    Hu  J, Gonsahn  MD, Nerenz  DR.  Socioeconomic status and readmissions: evidence from an urban teaching hospital.   Health Aff (Millwood). 2014;33(5):778-785. doi:10.1377/hlthaff.2013.0816 PubMedGoogle ScholarCrossref
    10.
    Goldfield  NI, McCullough  EC, Hughes  JS,  et al.  Identifying potentially preventable readmissions.   Health Care Financ Rev. 2008;30(1):75-91.PubMedGoogle Scholar
    11.
    Jack  HE, Arabadjis  SD, Sun  L, Sullivan  EE, Phillips  RS.  Impact of community health workers on use of healthcare services in the United States: a systematic review.   J Gen Intern Med. 2017;32(3):325-344. doi:10.1007/s11606-016-3922-9 PubMedGoogle ScholarCrossref
    12.
    We analyzed 25 population health interventions—and these 2 give the best ‘bang for your buck.’ Advisory Board. March 22, 2018. Accessed July 19, 2018. https://www.advisory.com/blog/2018/03/population-health-interventions
    13.
    Kangovi  S, Mitra  N, Grande  D, Long  JA, Asch  DA.  Evidence-based community health worker program addresses unmet social needs and generates positive return on investment.   Health Aff (Millwood). 2020;39(2):207-213. doi:10.1377/hlthaff.2019.00981 PubMedGoogle ScholarCrossref
    14.
    American Public Health Association. Support for community health worker leadership in determining workforce standards for training and credentialing. Accessed July 19, 2018. https://www.apha.org/policies-and-advocacy/public-health-policy-statements/policy-database/2015/01/28/14/15/support-for-community-health-worker-leadership
    15.
    Lohr  AM, Ingram  M, Nuñez  AV, Reinschmidt  KM, Carvajal  SC.  Community-clinical linkages with community health workers in the United States: a scoping review.   Health Promot Pract. 2018;19(3):349-360. doi:10.1177/1524839918754868 PubMedGoogle ScholarCrossref
    16.
    Malcarney  M-B, Pittman  P, Quigley  L, Horton  K, Seiler  N.  The changing roles of community health workers.   Health Serv Res. 2017;52(suppl 1):360-382. doi:10.1111/1475-6773.12657 PubMedGoogle ScholarCrossref
    17.
    The Rural Health Information Hub, Health Resources and Services Administration. Community Health Workers Toolkit. Accessed July 20, 2018. https://www.ruralhealthinfo.org/toolkits/community-health-workers
    18.
    Balcazar  H, Rosenthal  EL, Brownstein  JN, Rush  CH, Matos  S, Hernandez  L.  Community health workers can be a public health force for change in the United States: three actions for a new paradigm.   Am J Public Health. 2011;101(12):2199-2203. doi:10.2105/AJPH.2011.300386 PubMedGoogle ScholarCrossref
    19.
    Commonwealth of Massachusetts. Core Competencies for Community Health Workers. Revised May 2014. Accessed July 19, 2018. https://www.mass.gov/service-details/core-competencies-for-community-health-workers
    20.
    Reinschmidt  KM, Ingram  M, Morales  S,  et al.  Documenting community health worker roles in primary care: contributions to evidence-based integration into health care teams, 2015.   J Ambul Care Manage. 2017;40(4):305-315. doi:10.1097/JAC.0000000000000178 PubMedGoogle ScholarCrossref
    21.
    Brownstein  JN, Chowdhury  FM, Norris  SL,  et al.  Effectiveness of community health workers in the care of people with hypertension.   Am J Prev Med. 2007;32(5):435-447. doi:10.1016/j.amepre.2007.01.011 PubMedGoogle ScholarCrossref
    22.
    Rothschild  SK, Martin  MA, Swider  SM,  et al.  Mexican American trial of community health workers: a randomized controlled trial of a community health worker intervention for Mexican Americans with type 2 diabetes mellitus.   Am J Public Health. 2014;104(8):1540-1548. doi:10.2105/AJPH.2013.301439 PubMedGoogle ScholarCrossref
    23.
    Wells  KJ, Luque  JS, Miladinovic  B,  et al.  Do community health worker interventions improve rates of screening mammography in the United States: a systematic review.   Cancer Epidemiol Biomarkers Prev. 2011;20(8):1580-1598. doi:10.1158/1055-9965.EPI-11-0276 PubMedGoogle ScholarCrossref
    24.
    Percac-Lima  S, Grant  RW, Green  AR,  et al.  A culturally tailored navigator program for colorectal cancer screening in a community health center: a randomized, controlled trial.   J Gen Intern Med. 2009;24(2):211-217. doi:10.1007/s11606-008-0864-x PubMedGoogle ScholarCrossref
    25.
    Krieger  JW, Takaro  TK, Song  L, Weaver  M.  The Seattle-King County Healthy Homes Project: a randomized, controlled trial of a community health worker intervention to decrease exposure to indoor asthma triggers.   Am J Public Health. 2005;95(4):652-659. doi:10.2105/AJPH.2004.042994 PubMedGoogle ScholarCrossref
    26.
    Venditti  EM, Zgibor  JC, Vander Bilt  J,  et al.  Mobility and Vitality Lifestyle Program (MOVE UP): a community health worker intervention for older adults with obesity to improve weight, health, and physical function.   Innov Aging. 2018;2(2):igy012. doi:10.1093/geroni/igy012 PubMedGoogle Scholar
    27.
    Kangovi  S, Mitra  N, Grande  D,  et al.  Patient-centered community health worker intervention to improve posthospital outcomes: a randomized clinical trial.   JAMA Intern Med. 2014;174(4):535-543. doi:10.1001/jamainternmed.2013.14327 PubMedGoogle ScholarCrossref
    28.
    Balaban  RB, Galbraith  AA, Burns  ME, Vialle-Valentin  CE, Larochelle  MR, Ross-Degnan  D.  A patient navigator intervention to reduce hospital readmissions among high-risk safety-net patients: a randomized controlled trial.   J Gen Intern Med. 2015;30(7):907-915. doi:10.1007/s11606-015-3185-x PubMedGoogle ScholarCrossref
    29.
    Kangovi  S, Mitra  N, Norton  L,  et al.  Effect of community health worker support on clinical outcomes of low-income patients across primary care facilities: a randomized clinical trial.   JAMA Intern Med. 2018;178(12):1635-1643. doi:10.1001/jamainternmed.2018.4630 PubMedGoogle ScholarCrossref
    30.
    Kangovi  S, Mitra  N, Grande  D, Huo  H, Smith  RA, Long  JA.  Community health worker support for disadvantaged patients with multiple chronic diseases: a randomized clinical trial.   Am J Public Health. 2017;107(10):1660-1667. doi:10.2105/AJPH.2017.303985 PubMedGoogle ScholarCrossref
    31.
    Finkelstein  A, Zhou  A, Taubman  S, Doyle  J.  Health care hotspotting—a randomized controlled trial.   N Engl J Med. 2020;382(2):152-162. doi:10.1056/NEJMsa1906848 PubMedGoogle ScholarCrossref
    32.
    Carter  J, Walton  A, Donelan  K, Thorndike  A.  Implementing community health worker-patient pairings at the time of hospital discharge: a randomized control trial.   Contemp Clin Trials. 2018;74:32-37. doi:10.1016/j.cct.2018.09.013 PubMedGoogle ScholarCrossref
    33.
    Carter  J, Ward  C, Wexler  D, Donelan  K.  The association between patient experience factors and likelihood of 30-day readmission: a prospective cohort study.   BMJ Qual Saf. 2018;27(9):683-690. doi:10.1136/bmjqs-2017-007184 PubMedGoogle ScholarCrossref
    34.
    Carter  J, Ward  C, Thorndike  A, Donelan  K, Wexler  DJ.  Social Factors and Patient Perceptions Associated With Preventable Hospital Readmissions.   J Patient Exp. 2020;7(1):19-26. doi:10.1177/2374373518825143PubMedGoogle ScholarCrossref
    35.
    Fisher  SR, Graham  JE, Krishnan  S, Ottenbacher  KJ.  Predictors of 30-day readmission following inpatient rehabilitation for patients at high risk for hospital readmission.   Phys Ther. 2016;96(1):62-70. doi:10.2522/ptj.20150034 PubMedGoogle ScholarCrossref
    36.
    Ottenbacher  KJ, Karmarkar  A, Graham  JE,  et al.  Thirty-day hospital readmission following discharge from postacute rehabilitation in fee-for-service Medicare patients.   JAMA. 2014;311(6):604-614. doi:10.1001/jama.2014.8 PubMedGoogle ScholarCrossref
    37.
    Vasan  A, Morgan  JW, Mitra  N,  et al.  Effects of a standardized community health worker intervention on hospitalization among disadvantaged patients with multiple chronic conditions: a pooled analysis of three clinical trials.   Health Serv Res. 2020;55(suppl 2):894-901. doi:10.1111/1475-6773.13321 PubMedGoogle ScholarCrossref
    38.
    Nuti  LA, Lawley  M, Turkcan  A,  et al.  No-shows to primary care appointments: subsequent acute care utilization among diabetic patients.   BMC Health Serv Res. 2012;12(1):304. doi:10.1186/1472-6963-12-304 PubMedGoogle ScholarCrossref
    39.
    Ellis  DA, McQueenie  R, McConnachie  A, Wilson  P, Williamson  AE.  Demographic and practice factors predicting repeated non-attendance in primary care: a national retrospective cohort analysis.   Lancet Public Health. 2017;2(12):e551-e559. doi:10.1016/S2468-2667(17)30217-7 PubMedGoogle ScholarCrossref
    40.
    Gimpel  N, Marcee  A, Kennedy  K, Walton  J, Lee  S, DeHaven  MJ.  Patient perceptions of a community-based care coordination system.   Health Promot Pract. 2010;11(2):173-181. doi:10.1177/1524839908320360 PubMedGoogle ScholarCrossref
    41.
    Brooks  BA, Davis  S, Frank-Lightfoot  L, Kulbok  PA, Poree  S, Sgarlata  L.  Building a Community Health Worker Program: The Key to Better Care, Better Outcomes, & Lower Costs. CommunityHealth Works; 2018.
    42.
    Covert  H, Sherman  M, Miner  K, Lichtveld  M.  Core competencies and a workforce framework for community health workers: a model for advancing the profession.   Am J Public Health. 2019;109(2):320-327. doi:10.2105/AJPH.2018.304737 PubMedGoogle ScholarCrossref
    43.
    Rosenthal  EL, Rush  CH, Allen  C. Understanding scope and competencies: a contemporary look at the United States community health worker field: progress report of the Community Health Worker (CHW) Core Consensus Project. Accessed October 5, 2020. https://sph.uth.edu/dotAsset/28044e61-fb10-41a2-bf3b-07efa4fe56ae.pdf
    44.
    Ingram  M, Reinschmidt  KM, Schachter  KA,  et al.  Establishing a professional profile of community health workers: results from a national study of roles, activities and training.   J Community Health. 2012;37(2):529-537. doi:10.1007/s10900-011-9475-2 PubMedGoogle ScholarCrossref
    45.
    Arvey  SR, Fernandez  ME.  Identifying the core elements of effective community health worker programs: a research agenda.   Am J Public Health. 2012;102(9):1633-1637. doi:10.2105/AJPH.2012.300649 PubMedGoogle ScholarCrossref
    46.
    Lapidos  A, Lapedis  J, Heisler  M.  Realizing the value of community health workers—new opportunities for sustainable financing.   N Engl J Med. 2019;380(21):1990-1992. doi:10.1056/NEJMp1815382 PubMedGoogle ScholarCrossref
    ×