INTRODUCTION

Worldwide, non-communicable diseases account for more than 70% of the 41 million annual deaths1 and 80% of total disabilities,2 and pose a huge demand for health systems.3

Due to fragmentation in health system and complex needs of the patients, many patients find difficulties navigating the increasingly complex health system.4, 5 This is particularly true for patients from culturally and linguistically diverse backgrounds including immigrants and refugees with poor health literacy which often results from culture and language barriers.6, 7

Primary care is the first level of contact into the health system and often the principal avenue of care for chronic disease management.8 A strong primary care system which provides coordinated, comprehensive, affordable, community-focused, and family-oriented care results in better management of chronic disease.9, 10 Alongside the treatment and long-term care for the patients with chronic diseases, primary care also carries out activities such as health promotion, community education, patient advocacy, and illness prevention.11, 12 Furthermore, primary care is also responsible for the coordination of specialist services.9

While there is increasing recognition of the importance of primary care in management of chronic illnesses, there is often lack of coordination among different health providers.13 Studies carried out in different countries point out that lack of time during the clinical visit is one of the most important barriers to the provision of quality primary care for chronic diseases.14 In a recent study carried out in Australian general practice, Song and colleagues15 reported that a major barrier to access care among patients with chronic conditions was providers’ inability to tailor care to patients’ expectations and preferences. Culturally and linguistically diverse patients also experience barriers due to inadequate English language proficiency, lack of understanding of health system, poor health literacy, and inability to pay high out-of-pocket costs for referral services.16,17,18

Globally, community health workers (CHWs) are considered an integral part of the primary care.19,20,21 CHWs are defined as “frontline public health workers who serve as a liaison/link/intermediary between health/social services and the community to facilitate access to services and improve the quality and cultural competence of service delivery”.22 Community health workers can serve several roles, including care coordination, health assessments,23, 24 health education,25,26,27 and psychosocial support.28,29,30

Meanwhile, patient navigation is an important health service delivery tool to promote timely movement of an individual patient across the complex healthcare continuum.31 The concept of patient navigation was first developed by Professor Harold Freeman, in 1990. Freeman introduced the term “patient navigator” whose role was to address the barriers in breast cancer diagnosis and treatment among poor, uninsured, and underserved women in Harlem.31, 32 In recent years, the patient navigation role has been expanded to different aspects of healthcare.33

However, there is a lack of systematic reviews focusing on the role of CHWs as navigators. Therefore, the present systematic review aimed to evaluate the effectiveness of the CHW role as navigator in the context of chronic disease management in primary care.

METHODS

We followed PRISMA (preferred reporting items for systematic reviews and meta-analyses) guidelines34 in conducting this systematic review and the review protocol was registered in PROSPERO (CRD42020153921).

Search Strategy

We systematically searched the electronic databases Medline, Embase, Emcare, PubMed, Psych Info, CINAHL, Scopus, and Medline Epub ahead of print to identify potential research studies. We used Google and Google scholar in scoping the review and searching for possible gray literature. A combination of Medical Subject Headings (MeSH) and free-text keywords categorized into three groups (CHW, navigation, and primary care) was used to search for relevant articles in selected databases which are published in English between January 1990 and March 2020. Detailed search strategy is presented in Annex 1.

Study Selection

The articles identified through electronic database searches were assessed against predefined inclusion and exclusion criteria (Box 1) by two independent reviewers. First, we screened the title and abstract of the articles and full text was obtained for articles passing this screen. More in-depth assessment was carried among the full-text articles. The study selection was performed using Covidence (https://www.covidence.org/)35 in which reviewers could vote independently. Differences were resolved by discussion.

Box 1 Inclusion and exclusion criteria

Inclusion criteria

• Published in English

• Published between 1990 and March 2020

• Studies evaluating navigation intervention

• Carried out in primary care setting or in other community settings

• Navigation services provided by CHWs

• Navigation services provided for chronic disease management

• Primary studies

Exclusion criteria

• Published in language other than English

• Published before 1990 and March 2020

• Studies not evaluating navigation intervention

• Carried out in secondary and tertiary care setting

• Navigation services provided by professionals other than CHWs

• Navigation services provided for infectious disease management

• Article is review/commentary/protocol

Data Extraction and Quality Assessment

Relevant data were extracted using a template developed by the authors. Two of the authors assessed the methodological quality of each of the selected studies using Effective Public Health Practice Project (EPHPP) quality assessment tool.36 Each study was scored (weak, moderate, or strong) against six methodological criteria: selection bias, study design, confounders, blinding, data collection methods, withdrawals, and drops. A study was categorized as weak if it had two or more weak rating, moderate if had one weak rating, and strong if had no weak rating. Differences were resolved by discussion. We also assessed the quality of reporting of the intervention in the selected studies using TIDieR checklist.37

Data Analysis

Descriptive analysis was performed to summarize the interventions, roles, training, and outcomes of the studies in relation to the objectives of the research. Considering the heterogeneous design and outcomes, we did not perform meta-analysis.

RESULTS

Search Result

A total of 539 articles were identified through initial searching. After screening, we included 29 articles in the final analysis (Fig. 1). Study characteristics are summarized in Table 1. Six of the 29 studies had strong methodological quality,38,39,40,41,42,43,44 12 moderate, and 11 weak 45,46,47,48,49,50,51,52,53,54,55 (Annex 2). When assessed according to the TIDieR checklist, 24 studies reported procedures, who provided, how, where, and when.37 Few studies11 reported the frequency and time period of the intervention sessions and only one study55 reported how the intervention had been tailored to local needs. No studies reported on modifications to the intervention or fidelity of its implementation (Annex 3).

Figure 1
figure 1

PRISMA diagram of study selection.

Full size image
Table 1 Characteristics of Included Studies
Full size table

Study Settings

All but two studies were carried out in the USA. One study was carried out in France39 and one in Australia.56 All included studies were carried out in primary care38, 40,41,42,43,44, 46, 47, 49,50,51,52,53,54,55, 57,58,59,60,61,62,63,64 or in other community settings.39, 45, 48, 56, 65, 66 A range of primary care settings were reported including primary care practices, community health centers, and primary care units of hospitals. Community settings included rural community areas as well as civic organizations such as senior centers, churches, and a barbershop.

Study Designs

A variety of research designs were reported. Fifteen of the included studies were randomized controlled trials (RCT),38,39,40,41,42,43,44, 47, 48, 50, 54, 56, 59, 62, 64 six were quasi-experimental studies,49, 55, 58, 61, 63, 65 four were pre-post,45, 51, 57, 66 one was an interventional prospective cohort study,52 one was a retrospective comparative study,60 and two were mixed method studies.46, 53

Type of Participants

The interventions predominantly focused on the medically underserved populations such as racial/ethnic minorities or immigrants40, 43, 44, 46, 47, 52, 56,57,58, 64 including African Americans,41, 42, 48, 50, 54, 60, 65 Hispanic/Latinos,45, 49, 55, 59, 63, 66 and Asian and Pacific Islanders.38 In eight studies, participants were economically disadvantaged,39, 43, 46, 47, 58, 61, 62, 64 and in four, they were underinsured or uninsured.43, 44, 58, 61

Role Titles of CHWs

Healthcare navigation roles of CHWs were described through different role titles based on the activities they performed, including patient navigators,41, 43, 44, 46, 47, 52, 54, 55, 59, 62, 63 CHWs,48, 53, 57, 60, 64, 65 lay health workers,40, 51 bilingual CHWs,58, 66 Promotoras,49 care navigators,50 health navigators,38, 61 screening navigators,39 indigenous health workers,56 and peer navigators.42, 45

Recruitment, Training, and Supervision of CHWs

In most studies, CHWs were recruited from the communities they served with or without having any previous healthcare experience. CHWs were mostly female lay community members who spoke English in addition to their own community languages.

There was considerable variability in the breadth and extent of the description of training. Twenty-two studies38,39,40,41,42,43,44,45, 50,51,52,53,54,55,56, 58,59,60, 62,63,64, 66 documented the training of CHWs as navigators. In 16 studies,39, 40, 42, 43, 45, 50,51,52,53,54, 56, 59, 60, 63, 64, 66 CHWs received training as part of the intervention while six studies38, 41, 44, 55, 58, 62 reported that CHWs were trained before their recruitment. The most commonly reported components of the training included clinical education on specific diseases,42, 43, 45, 46, 53, 56, 60 screening related information,42, 45, 46, 53, 64, 66 motivational interviewing,50, 52, 53, 62 and communication skills.42, 52, 53, 59, 62 Other training addressed the navigation processes and responsibilities,42, 64 case management,50, 60 conducting community education,45, 59 patient centeredness,52, 56 risk factors of specific diseases,45, 66 patients readiness to change,55, 62 and identification and use of local resources.60

In most cases, training was provided face-to-face using both didactic classroom lectures and interactive role play. The length of the training period varied considerably, ranging from 3 h52 to 3 weeks.56 However, only a few studies reported details of the trainers and/or the institution or programs through which CHWs were trained. No studies evaluated the skills and competencies gained through CHW training. Supervision of the CHWs was reported in only 5 studies53, 55, 60, 62, 66 mostly by the program managers, in weekly or bi-weekly supervision meetings where case allocation,60 problem solving6053,62 and CHW audit53, 55, 62, 66 were performed.

Types of CHW Roles as Navigator

The CHWs provided navigation assistance to the participants through a wide range of activities involving the prevention and management of different chronic illnesses (Table 3). The most frequently reported roles were education and counselling, addressing barriers to accessing screening or health services, and supporting patients to attend appointments (through scheduling, reminders, assistance with transport, or accompanying patients) and provide patient follow-up.

CHWs communicated with the participants both in-person and via telephone in 15 studies.38,39,40, 43, 45, 50, 54, 55, 58, 60,61,62,63,64, 66 However, in some studies, navigation consultations were provided either via telephone42, 44, 46,47,48, 51,52,53, 59 or in-person.41, 49, 56, 57, 65 Online media was also used in some studies40, 43, 45, 59 to complement in-person and telephone communication.

In 26 studies, CHWs were solely responsible for carrying out the intervention and the effect of the CHW role was evaluated independently. In three studies,42, 49, 57 CHWs were engaged as part of a team with other trained staff and CHW role was assessed as part of the team’s achievements.

Effects of CHW Navigation Intervention (Table 2)

Cancer Prevention and Treatment

Among the 17 studies38, 39, 41, 42, 44,45,46,47,48, 51, 53, 54, 59, 61, 63, 64, 66 that investigated the effect of CHWs in improving cancer screening outcomes through providing navigation supports, increase in cancer screening rates was noted in 15 studies38, 39, 41, 44, 46,47,48, 51, 53, 54, 59, 61, 63, 64, 66 (3 of them were of strong methodological quality, 6 moderate, and 6 weak). Significant positive changes in adherence to breast cancer screening were reported in 5 studies38, 44, 46, 54, 66 of the 6 that targeted breast cancer. Of the 13 studies that focused on colorectal cancer, 12 studies noted significant increase in adherence to colorectal cancer screening.38, 39, 41, 47, 48, 51, 53, 59, 61, 63, 64, 66 Positive outcomes in screening for multiple cancers such as breast, colorectal, cervical, and prostate were also noted in all the three studies38, 45, 66 (with varied mythological qualities) that focused on multiple cancers. Two interventions, one with strong methodological quality43 and other weak,49 focused on the time for cancer diagnostic resolution following an abnormal screening result and both reported a significantly shorter time among the navigated participants compared to those without navigation. Another study with strong methodological quality40 reported no significant improvement in the time to completion of primary cancer treatment among the navigated participants compared to the non-navigated.

Table 2 Impact/Outcome of Healthcare Navigation Delivered by CHWs
Full size table

Effective Primary Care Services

Effective primary care services included improved access and continuity of care through regular visits, appropriate follow-up, and linking to long-term resources for care. Of the five studies40, 46, 50, 57, 65 that focused on primary care, all but one study40 noted a significant improvement in use of primary care services among the navigated patients. In two of the four studies with positive outcomes, the methods were of moderate quality,57, 65 and in others, the methods were weak.46, 50 Battaglia et al.46 found that 71% of the participants scheduled primary care visit to follow-up on their health priorities, and after 30 days, 54% of participants visited their primary care provider. Chukwudozie and colleagues57 noted that CHWs were effective in increasing the use of patient-centered medical home for kidney disease primary care physician (PCMH-KD PCP) among the patients. A moderate--quality study65 that focused on identifying the patients with long-term health needs and linking them to local resources found that CHWs were effective in identifying and helping patients access long-term care services (HCBS). However, Fiscella et al.,40 in a study of strong methodological quality, did not find any significant effect of navigators in time to completion of cancer treatment or psychological distress in primary care.

Reducing Use of Secondary and Tertiary Care

Three moderate quality studies58, 60, 65 investigated the effect of CHWs as navigators in reducing the use of secondary and tertiary services and all reported positive outcomes. Of the two studies that focused on reducing use of emergency department, one study58 noted that navigated patients had significantly fewer primary care-related emergency department visits during the intervention period. The other study60 also reported that emergency room visits decreased by 38% among the navigated patients and hospital admissions were reduced by 53% in 1 year post-intervention. One study65 tested the effect of CHWs as navigators in use of nursing homes and found that visits by navigated participants to nursing homes were five times fewer than those who were not navigated.

Intermediate Outcomes

All four studies that focused on assessing the effect of CHWs as navigators on patient reported outcomes or intermediate outcomes52, 55, 56, 63 reported significant positive outcomes. However, their methodological quality varied from weak52, 55 to moderate.56, 63 Two studies52, 56 reported significant reductions in HBA1c level (p<0.05) and another55 a reduction in total cholesterol (183 mg/dL vs 197 mg/dL). Another study63 found a reduction in the severity of cervical abnormalities in Pap smears among intervention compared control patients.

Addressing Risk Factors for Chronic Disease

Two moderate62 to weak55 methodological quality studies tested the effect of a CHW navigation intervention on how effectively the risk factors of chronic disease were addressed. One62 found no difference in adherence to smoking cessation (47.4% vs 42.9%, p=0.80). Another found that CHW navigation had a greater impact on reducing cardiovascular risk68 after 12 months of follow-up.

Patient Satisfaction

Patient levels of satisfaction with the navigation services were assessed in six studies38, 40, 46, 47, 52, 62 and all but one study reported that majority of the navigated participants were satisfied with the navigation services. For example, Loskutova et al.52 reported that levels of satisfaction were high among the participants receiving the navigation services and they would recommend the services to their friend and family, and 90% of the patients reported that they would use the program in future. A study by Lasser et al.62 found that most of the patients were satisfied with the navigation services and they did not feel any pressure to make changes. However, Fiscella et al.40 found no significant difference between navigated and non-navigated participants in terms of satisfaction with cancer-related primary care. However, they noted that socially disadvantaged patients who received patient navigation services were more satisfied than those who received usual care (p<0.05). Three of the studies with positive outcomes were of weak,46, 47, 52 one moderate62, and two strong 38, 40 methodological quality.

Cost Outcomes

Three moderate methodological quality studies58, 60, 65 measured cost outcomes of the CHW navigation intervention. In one study,58 it was found that the total savings associated with reduction of emergency room visit among the navigated participants was higher than the cost of implementing the intervention. Fedder et al.60 also noted that their CHW program resulted in an annual average savings of $2245 per year. The other study65 found that mean annual spending per beneficiary was significantly higher among the navigated participants than non-navigated ($6769 vs $3687, p<0.0001), but this was not assessed in terms of the benefit of the intervention.

DISCUSSION

This review identified that CHW navigation interventions were effective in increasing adherence to cancer screening, particularly for breast, cervical, and colorectal cancers, and improving use of primary care for effective chronic disease management. However, there was insufficient evidence of their impact on use of secondary and tertiary care, risk factors or intermediary outcomes in chronic disease management, or cost-effectiveness. The impact on patient satisfaction was uncertain due to lack of high-quality studies. These findings are consistent with other reviews evaluating CHW interventions, although in these reviews, navigation was only one component of a more complicated intervention33, 67, 68 or navigation was performed by other providers other than CHWs.5, 68, 69 Kim et al.67 reported in a systematic review that CHWs were effective in improving access to cancer screening. They also commented on the lack of strong evidence for the cost-effectiveness of CHW interventions. In another review,68 it was also reported that CHWs and patient navigators were effective in improving adherence to and timely completion of breast, cervical, and colorectal cancer screening in medically underserved populations.

The most frequently reported functional roles for community health navigators are listed in Table 3. Most interventions with these roles reported positive outcomes in terms of either increased cancer screening and diagnosis or more appropriate use of primary care. These roles also map against the patient education and care coordination roles defined in the CHW Core Consensus (C3) project70 and similar roles for CHW in primary care were defined in a review by Hartzler et al..71 Based on these findings, we have suggested four core functional roles for CHW as navigators for patients with chronic disease in primary care (Box 2). There was limited evidence for other roles such as social support and self-management.

Table 3 Roles of CHW as Navigators in Primary Care
Full size table

Box 2 Functional roles for CHW as navigators in primary care

 

Domain

Sub-roles

Core

1

Health Education

Provision of education and counselling about health care

2

Addressing barriers

Identification and addressing of barriers to health care

3

Care navigation

Helping schedule appointments, reminders, assisting with transport and accompanying patients to appointments

4

Patients follow up

Following patients up after healthcare

Other potential

 

Coordination with other providers

Communication and coordination with other healthcare providers

 

Community resources/social support

Linking patients with community resources

 

Monitoring

Supporting patients to monitor their condition

 

Self-management

Motivation and support for self-management

Similar to another study,67 most of the studies in our review reported that CHWs were recruited from the communities they served and this has been previously noted to help ensure cultural sensitivity and wider community acceptance.33 The settings in which the CHW was based were primary care or the community. However, neither of these settings was more likely to be associated with positive screening or access outcomes.

There was a wide variability in terms of the criteria for recruitment of the CHWs, duration, frequency and mode of training, qualification of the trainers, and supervision arrangements in the included studies. Many of the studies provided insufficient information on these critical aspects of CHW programs and we were unable to assess their effectiveness. Other systematic reviews have also reported variability in recruitment, training, and supervision of navigators.67, 68 Previous reviews33, 72, 73 pointed out the importance of rigorous training and optimum level of competencies gained through the training to successfully complete the assigned tasks. Yet, no study in the present review reported the competencies gained through the training. In the future, studies should clearly describe the training including training materials, frequency, and duration of training as well as competencies gained through the training.

Policy Implications

Previous research has highlight the important contribution which system navigation can make to healthcare access and care coordination.4 The findings of the present review have significant implications for incorporating CHWs as part of the patient-centered primary care team. Carter and colleagues4 also mentioned that while navigator roles can add an additional complexity to the health system, they can substantially contribute to improve care coordination and facilitation of care and services. While discussing the navigation role, McBrien et al.5 recommended that patient navigators should focus on identification and addressing barriers to receipt of care rather than providing any clinical support. It is also very important to develop clear selection criteria for navigators, a well-developed training procedure, competency assessment, and proper supervision arrangements before recommending that CHWs as navigator become integrated into the healthcare team.68 The level of reimbursement for the CHWs is also important. While we did not find any information pertaining to this, previous studies74, 75 have pointed out that CHWs were more flexible and productive when they were adequately reimbursed.

As in previous studies,67, 68 we found a lack of sufficient information on the cost-effectiveness of CHW interventions. This can be challenging, as it is often difficult to separate out the cost-effectiveness of CHW interventions themselves from the medical interventions that they support (e.g., cancer screening). Future research should focus on rigorous cost-effectiveness analysis of CHW navigation interventions before integrating them into the healthcare system.

Limitations

Our review had several limitations. As there are several role titles for CHWs in the literature, it is possible that some, with titles outside those used in our search, may have been missed. To address this, we exploded both MeSH terms and keywords in our database searches. We restricted our searches to peer-reviewed articles only, and therefore, we might have missed some non-peer-reviewed gray literature published outside of academic journals.

There are also methodological issues that need to be considered when interpreting the study results as only 6 of the included studies were of strong methodological quality. Several of those with a strong research design (RCT)40, 47, 50, 59 were of weak to moderate methodological quality. There was also significant variability in the reporting of the intervention characteristics (Annex 3) that need to be considered when interpreting the impact of specific roles for CHW as navigators.

We restricted our searches to English papers and thus may have missed publications written in non-English languages. In some of the included studies, the intervention involved CHWs as part of a team and the effect of CHWs was not assessed separately. Also, we could not conduct a meta-analysis due to heterogeneity in the study designs and outcomes of the included studies.

CONCLUSION

Our review suggests that CHW navigation interventions were effective in supporting screening for cancer and access to primary care. Our findings suggest that core roles include health education, addressing barriers to care, providing navigation to care, and patient follow-up. However, there was insufficient evidence of cost-effectiveness, clinical outcomes, and patient experience resulting from the CHW navigation roles. Future research needs to be of strong methodological quality and should focus on determining the relative effectiveness of elements of CHW navigation roles in improving health outcomes and patient experience in primary care.