Table 2.

Characteristics of included studies

Authors	Context	Study type	Population and sampling	Primary outcome	Description
Bishop49	Private, non-profit, community homeless shelter	Description of Charlottesville Health Access initiative to enhance access to care	Homeless and near-homeless people, without a healthcare provider, attending health fair at shelter or soup kitchen (no sample reported)	People connected to permanent healthcare provider	Volunteer navigator (student or community member) completed a training course, engaged person by building relationships, assessed needs, guided to providers, translated confusing information, coordinated follow-up and empowered people to understand health system and self-care.
Chan et al 42	ED in low-income, urban area served by three community clinics	Non-randomised, non-blinded interventional trial to improve primary care access for underserved patients	Patients with no primary care provider assessed by ED physician to benefit from clinic follow-up (n=326)	Patients follow-up at community clinic within 14 days	Internet-based secure referral system between ED medical record and clinic appointment systems. System accessed clinic availability and allowed ED physicians to give patients follow-up appointments at clinics.
Doran et al 57	Urban, public, safety-net hospital ED with primary care clinic in same building complex	Quasiexperimental trial to navigate willing patients from ED to clinic	Adults with no primary care provider, presenting with low-acuity problems, assigned to intervention or usual care based on where care expected to result in least delay (n=965)	Patients follow-up at primary care clinic within 1 year	Trained patient navigator escorted patients from ED waiting room to clinic. Patients assigned physician who addressed current problems, established care plan and gave card with name and clinic telephone number.
Elliott et al 43	Urban ED, serving high proportion of vulnerable patients	Retrospective study using full electronic medical record abstraction, randomly sampled	Patients with no primary care provider, discharged and referred to transitional care clinic (n=660)	Patient completed follow-up visit in transitional care clinic as scheduled	Transitional care clinic staff worked with patients to determine preferences and locate convenient, appropriate provider and made new appointment with chosen provider.
Gany et al 51	Unused parking lot adjacent to JFK International Airport’s taxi holding lot	Description of Step On It! workplace intervention to increase healthcare access	Convenience sample of taxi drivers waiting in airport holding lot (n=466)	Driver completed follow-up visit with linked provider within 6 months	Healthcare access and case management to link drivers to providers, including referrals to low-cost (or free) culturally appropriate clinics or hospitals.
Griswold et al 39–41	Urban Comprehensive Psychiatric Emergency Program (psychiatric assessment and management, targeted therapeutic approaches, links to community mental health services) as usual care	Randomised controlled trial comparing linkage with primary care with usual care after psychiatric emergency visit	Adults presenting with psychiatric disorder, with no primary care provider or have not seen one within 6 months (n=101–175)	Patients connected to and visited primary care within 3 and 12 months	Care navigator trained in interviewing and case management provided information about low-cost care, facilitated access and reinforced patient education; information to providers about patient’s history, follow-up, peer connections to access community and social services.
Horwitz et al 58	Level 1 urban trauma centre	Randomised study of intensive case management intervention to improve primary care use	Uninsured adults presenting to ED, excluding substance abuse or mental health issues only (n=230)	Patients visited one of four participating primary care clinics within 2 months	Health Promotion Advocates in ED assisted patients to choose provider, gave brochure, faxed information to case worker at selected clinic. Clinic case worker contacted patient to make appointment.
Kahn et al 61	Medicaid managed care organisation for people with mental health and/or substance abuse diagnoses	Evaluation to assess effectiveness of case management in linking new members with primary care providers	New members with behavioural health diagnosis and no primary care provider completing mailed survey, referred to case management (n=368)	Member visited primary care provider within 12 months	Telephone case managers made at least three contact attempts to ensure linkage to provider.
Kangovi et al 45	Two urban, academically affiliated hospitals	Two-armed, single-blind, randomised clinical trial to improve primary care follow-up postdischarge	Newly admitted low-income, uninsured or Medicaid adult inpatients randomly numbered, approached until three per day enrolled (n=446)	Patient completed follow-up visit with primary care provider within 14 days	Community health workers (trained lay people of similar backgrounds to patients, selected for personality traits patients identified as important) set goals, supported goal achievement, connected to provider.
Kim et al 53	Five hospital EDs in an affluent area with large and poor immigrant population	Analysis of Emergency Department-Primary Care Connect initiative to link patients to four local primary care clinics	Merged data set (hospital discharge, clinic and navigator referral data) of low-income or uninsured patients with no primary care provider (n=10 761)	Patients completed two or more visits to same clinic across 33-month period	Patient navigators of various backgrounds (most unlicenced, selected for communication skills) based in clinics (three sites) or hospitals (two sites) spoke face to face or telephoned patients referred by ED providers.
Marr et al 50	Urban ED with high rates of potentially avoidable hospitalisations and lack of community-based care	Evaluation of programme to connect patients with community-based, primary care providers	Patients with no primary care provider approached by navigator (n=7185)	Patients completed three or more visits to same clinic across 18-month period	Patient navigator (advocate) recruited from community, trained in ED, visited patients waiting for medical care or before discharge, offered referral within 18-clinic system.
Overholser et al 52	Specialist outpatient clinics of urban tertiary teaching hospital	Description of patient navigation programme to overcome barriers to finding primary care	Adults with sickle cell disease with no primary care provider or not seen regularly by provider, referred by specialist physicians (n=21)	Patients attended initial visit with new primary care provider	Patient navigators of various backgrounds trained in navigation proactively sought local providers and established network through outreach, made appointments with patients, sent reminders and educated on importance of primary care.
Treadwell et al 54	African-American community centre	Evaluation of Save Our Sons group health education and intervention model to reduce incidence of diabetes and obesity, improve regular access to care and build community networks	African-American men at risk for or diagnosed with diabetes and/or in poor health related to obesity and/or other health concerns; recruited at community event (n=42)	Participants connected to medical home	Six-week community-based, culturally responsive, gender-specific health prevention programme delivered by community health workers, and trusted community members provided links between health system and community.
Wang et al 37	Community health centre providing comprehensive services to ethnically diverse population with low incomes or uninsured	Evaluation of patient navigation programme to optimise healthcare utilisation	Patients with diabetes and/or hypertension not seen by provider in last 6 months (n=215)	Patient visited primary care provider and/or chronic disease nurse within 6 months	Patient navigator trained in chronic illness education, motivational interviewing and appointment scheduling. Telephoned patients, built rapport, educated patients, made appointment with provider, assessed need for specialist referrals, identified barriers to access and assisted to overcome barriers.
Wexler et al 44	ED within urban academic medical centre and affiliated primary care practices	Randomised controlled trial comparing health information technology intervention to improve access to primary care, with usual care	Medicaid enrollees who did not have usual source of care, ED physician confirmed visit non-urgent, completed baseline survey, randomly assigned (n=148)	Patients attend primary care provider office after discharge at 3, 6 and 12 months	ED electronic medical record to make appointment at clinic based on patient location and preference. Patient given appointment reminder card and directions to clinic. Electronic message to clinic with information about patient and appointment.
Emergency department navigators connect patients to better venues of care55	EDs of eight-hospital system	News article on use of ED navigators to redirect patients with non-emergency issues to most appropriate care setting	Health plan members with non-urgent problems (no sample reported)	Patient scheduled to be seen by another provider	Navigator with customer service background assigned members to provider and made appointments.
Navigator reduces readmissions, inappropriate ED visits56	Urban ED	News article on community health outreach worker helping patients find a primary care provider	Patients with non-urgent problems who are uninsured and do not have a primary care provider, insured but do not have a provider or have a provider but cannot access him or her (n=1500)	Self-pay patients find medical home; other patients identify primary care provider and set up follow-up appointment	Community health outreach coordinator/navigator of varying cultures representing patients served. Met patient in ED, coordinated appointments and set patients up in medical homes.
ED navigators help patients find a PCP59	Urban ED	News article on a pilot project to reduce 30-day readmissions and number of self-pay patients who visit ED for non-emergent care	Patients without insurance and primary care provider admitted to hospital through ED and/or not admitted (no sample reported)	Patients directed to primary care provider and set up in medical home	Navigator worked with patients to discuss discharge and help facilitate follow-up appointments.

ED, emergency department; PCP, primary care provider.