What do we know about community-based health worker programs? A systematic review of existing reviews on community health workers

1. Deliver diagnostic, treatment, and other clinical services

• Identify and assess sick community members: Use rapid tests for malaria [21, 61, 76] and HIV [32, 77]; determine if a child’s breathing is dangerously rapid [78], identify high-risk pregnancies [42]; monitor clinical symptoms and signs of drug toxicity in people living with HIV and refer when appropriate [52], monitor the effects of mental-health-related medications [54]; conduct breast-cancer screening exams [79], measure and monitor blood pressure [43]

• Provide medicines and other pharmaceuticals: Dispense contraceptives [80]: administer injectable contraceptives [81]; distribute antiretroviral drugs [52], iron folic acid tablets [82], vitamin A [82] or antimalarials [33, 82]

• Directly provide care and treatment: Perform home deliveries [15, 22, 83]; vaccinate children [16]; provide community-level diagnosis and treatment for pneumonia, malaria, and other infectious diseases [84]

2. Assist with appropriate utilization of health services, make referrals

Help ethnic minorities in the USA make and keep medical appointments for cancer screening [87] or for diabetes management [88], help people with hypertension in the USA access health insurance [43], help pregnant women with birth planning and preparedness to facilitate institutional delivery [91, 92], mobilize communities around maternal and neonatal health practices [93], refer women to health facilities for delivery [82, 94], encourage access and adherence to HIV care [32, 39, 95, 96], or find underserved groups and encourage them to have their children immunized [20]

3. Provide health education and behavior change motivation to community members

Provide education to reduce HIV stigma [50] or promote behaviors that reduce risk of acquiring HIV [74]; assist with family planning [80], depression, or assessment of child mental development [97]; encourage physical activity among those with non-communicable disease [98]; promote exclusive breastfeeding [82], antenatal and postnatal care and family planning [82]; advise on tetanus vaccination [82] or family planning [82]; provide education on cancer [87, 99], hypertension [43] and diabetes [89, 100]; reduce childhood asthma-triggering behaviors and environmental pathogens that provoke asthma [56, 57]

4. Collect and record data

Perform general clerical duties [70] and data collection (including using mHealth tools [101, 102]), identify and report on malaria outbreaks [44], monitor medicine stocks and notify government agencies when stocks are low [44, 76]

5. Improve relationships between health services and communities

Act as mediators between individuals and health services (e.g., to improve provider responsiveness to patient needs) [43], act as cultural mediators [51] (e.g., between Aboriginals and non-Aboriginals in Australia [45]), serve as patient advocates (e.g., for those with diabetes [89, 90] or cancer [87] in the USA, or for mental healthcare in LMICs [51, 103]); serve as community advocates (e.g. for Latino communities in the USA [104])

6. Provide psychosocial support

Form support groups for people with HIV [14, 50] or women [93, 105]; provide anti-retroviral treatment adherence reminders [50]; provide one-to-one psychosocial support to reduce maternal depression [73, 106], for people with hypertension [43], or for USA Latino parents of youth with mental health issues [106]; support adherence to drug regimens by sending short messages to cell phones to remind people living with HIV to take their medication [107]

 Multiple primary health care interventions

Most CHW programs focused on underserved populations in HICs (such as ethnic/racial minorities, economically marginalized, rural populations or immigrant groups) [25, 45, 90, 104, 106, 108, 109]. CHW interventions, such as through peer-support telephone calls [69] or home visits [110], can be effective for a wide range of health issues, including increasing knowledge about parenting [110], disease prevention (moderate strength of evidence) [25], influenza prevention [110], promotion of home safety [110], increasing parenting self efficacy [110], patient enrollment in research [99], uptake of early intervention services [99], increasing access to primary health care for screening [108], improving workplace safety (low strength of evidence) [25] and disease prevention (mixed evidence) [25], and reducing urgent care visits [110]. CHWs can reduce obesity among postpartum teens [110], improve nutritional eating habits [99]; and increase physical activity [98].

CHW programs can promote equity of healthcare access and utilization by reducing inequities relating to place of residence, gender, education and socio-economic position, and supporting more equitable uptake of referrals [111] (low-quality evidence from Brazil [112]). Deploying lay refugees/internally displaced persons as CHWs to provide basic health services to women, children, and families in camps can increase service coverage, knowledge about disease symptoms and prevention, uptake of treatment and protective behaviors, and access to reproductive health information (some evidence, weak quality) [113]. There was no clear evidence for equitable quality of services provided by CHWs, and there was limited information regarding the role of CHWs in generating community empowerment to respond to social determinants of health [111]. There is some evidence (moderate quality) that CHWs are effective in providing health education [114] and psychosocial support [114]. There is an absence of evidence on CHW potential to support community-based palliative care [75].

Reproductive, maternal, neonatal and child health

 Neonatal/child health

CHW interventions can be effective in increasing infant-stimulating home environment scores [110], reducing psychiatric diagnoses among children [110], improving child development [99], and improving child well-being (mixed evidence) [25].

CHWs providing community-based care for infants and children in resource-limited settings can reduce neonatal, infant and child mortality and morbidity (e.g., from malaria, pneumonia and diarrhea) [35, 42, 46, 84, 85, 91, 93, 115‐118]. While there is high-quality evidence that home-based neonatal care reduces neonatal and perinatal mortality in South Asian settings with high neonatal mortality rates and poor access to health facility-based care [91, 116] other reviews reported mixed results, with some individual empirical studies included in reviews not showing improvements in CHW intervention areas [85]. Evidence of the impact of CHW interventions on neonatal outcomes is promising but of moderate quality [46] and on CHW capacity to provide skilled birth care is of low quality [46]. Antenatal and neonatal practice indicators significantly improved [116]. Compared to physicians, trained CHWs may screen for possible bacterial infection in young infants with relatively high sensitivity but somewhat lower specificity [119]. There is some evidence of moderate quality that CHWs are effective in the promotion of essential newborn care [114], including skin-to-skin care for newborns [114]. CHWs can perform effective case management of child pneumonia [76], although pneumonia management performance is mixed when pneumonia management is integrated with malaria diagnosis and treatment [33]. The use of CHWs, compared to usual healthcare services, may increase the number of parents who seek help for their sick child [118]. Women’s groups (facilitated by CHWs) practicing participatory learning and action, compared with usual care, have a positive impact on reducing neonatal mortality in low-resource settings (but no evidence of impact on reducing stillbirths) [105]. Trained traditional birth attendants (TBAs) compared to untrained TBAs showed significant increases in safe delivery practices and appropriate referral knowledge and practice [94] and are associated with small but significant decreases in perinatal mortality and neonatal mortality due to birth asphyxia and pneumonia [94]. However, another review [82] concludes that there is insufficient evidence to establish the potential of TBA training to improve perinatal and neonatal mortality. CHWs in Brazil have demonstrated effectiveness in increasing the frequency of child weighings [112].

 Maternal health

Peer-support can be effective for reducing depressive symptoms in mothers with postnatal depression [69] and can positively impact women’s perinatal mental health [72]. One study on addressing stress and mental health among pregnant women on Medicaid in the USA found that adding a CHW to a nurse home visit program increased the number of at-risk women reached [106].

One review reported that almost all of the intervention studies involving CHWs showed a significant impact on reducing maternal mortality and on improving perinatal and postpartum service utilization indicators [35]. Another found that community-based intervention packages, which almost always involved CHWs, may have a possible effect on reducing maternal mortality, although the pooled result just crossed the line of no effect [93]. Women’s groups (facilitated by CHWs) practicing participatory learning and action, compared with usual care, have a positive impact on reducing maternal mortality in low-resource settings [105]. In settings characterized by high mortality and weak health systems, trained TBAs can contribute to reducing mortality through participation in key evidence-based interventions [94]. There is some evidence of moderate quality that CHWs are effective in providing psychosocial support [114]. CHWs were effective in delivering psychosocial and educational interventions to reduce maternal depression [73]. Non-specialist providers (a classification that includes CHWs) may be effective in reducing perinatal depression [54].

 Immunization

CHW programs increase the number of children whose vaccinations were up to date (moderate quality) [16].

There is evidence, but low quality or inconsistent, that CHWs can increase immunization coverage through promoting vaccination [16, 94, 118, 120] and providing vaccination themselves [16]. There is low-quality evidence that health professionals are confident that CHWs can deliver vaccines or other medicines using compact pre-filled auto-disposal devices [121].

 Contraception

CHW interventions have been found to reduce unplanned repeat births among adolescents [110, 122] but there was no significant association detected in terms of repeated pregnancies [122].

CHWs were able to deliver injectable contraception safely and effectively, with high quality and with high levels of patient satisfaction [81, 123], and initiate their use (which involves screening women and counseling them on side effects), with no difference in the quality of counseling on side effects between CHWs and clinic-based providers [81]. Most (93%) studies indicated that CHW family planning programs increased the use of modern contraception and most (83%) reported an improvement in knowledge and attitudes concerning contraceptives [80]. CHWs can provide counseling on contraceptives, provide contraceptives, and refer to health facilities for more specialized care [80].

 Breastfeeding

CHW interventions can be effective for increasing breastfeeding continuation [58, 69], attempts and duration [110], initiation, duration, and exclusivity [124].

The use of lay health workers, compared to usual healthcare services, probably increases breastfeeding [118] and there is some evidence of moderate quality that CHWs are effective in exclusive breastfeeding promotion [114]. CHWs in Brazil have demonstrated effectiveness in increasing the prevalence of breastfeeding [112] and delaying the introduction of bottle feeding [112].

Non-communicable diseases (NCDs)

 Diabetes

There is weak evidence that CHW interventions improve knowledge of medication-label reading among diabetics [25]; improve self-management [60] (low strength of evidence) [25]; decrease glycaemia [60] (mixed evidence) [90] (modest reduction) [125]. There is no evidence that telephone interventions provided by lay and peer-support workers improve mental health or quality of life among diabetics [60]. For children with type 1 diabetes, CHWs improved glycemic control and decreased hospitalizations [56].

CHW capacity in addressing diabetes in LMICs was not reported in the systematic review literature.

 Cancer

CHW interventions (peer support phone calls [69], home visits [110]) can be effective in increasing cancer screening rates [69, 99, 108, 110, 126]; knowledge about prostate cancer (but not screening) [110]; cancer screening (moderate evidence) [25]; planned use of cancer screening tests (mixed evidence) [25]; breast self-examination (mixed evidence) [25].

Only one non-systematic review [79] discussed the potential of CHW to address cancer in LMICs, and did not provide evidence on CHW capacity.

 Mental health

CHW interventions can reduce depression [110] and stigma toward depression treatment (one study) [106], improve depression knowledge and efficacy to seek treatment [106], and produce beneficial changes in health status measures in many, but not all, studies [127]. CHW interventions in children with chronic conditions may lead to modest improvements in parental psychosocial outcomes [56] and parental quality of life [56].

CHW-led interventions can reduce the burden of mental, neurological and substance-use disorders, including depression and post-traumatic stress disorder among adults (evidence from 3 studies) [97]; and can also improve child mental health outcomes (evidence from four studies) [97]. Non-specialist providers, usually CHWs, are more effective than usual care or delayed treatment (waitlisted) groups in the provision of mental health treatments, generally for depression or post-traumatic stress [128]. Non-specialist health workers, which in this review [54] included both professionals (e.g., doctors, nurses, and social workers) and CHWs (22 of the 38 studies), compared with usual healthcare services, have some promising benefits in improving outcomes for general and perinatal depression, post-traumatic stress disorder and alcohol-use disorders, and outcome for patients with dementia and their caretakers (evidence mostly of low or very low quality) [54].

 Asthma

Peer-support telephone calls can be effective for increasing the number of asthma-free days [110] as well as the use of bedding encasements for asthma patients (moderate strength of evidence) [25]. While some CHW interventions for children with asthma decreased rapid breathing episodes, activity limitation, and asthma exacerbations, and increased the number of symptom-free days, results were inconsistent and risk of bias was often unclear [56]. Lay and peer interventions for adolescents with asthma could lead to small improvements in asthma-related quality of life (weak evidence) but there was insufficient evidence on asthma control, exacerbations and medication adherence [129].

CHW capacity in addressing asthma in LMICs was not reported in the systematic review literature.

 Other NCDs (chronic disease, hypertension)

Peer-support telephone calls can be effective for diet change in post-myocardial infarction patients [69]. CHW interventions may improve chronic disease management among children (modest improvements in reduced urgent care use [56], decreased symptoms [56], and fewer missed work and school days [56]) and adults [108], including improvements in blood pressure among adults with hypertension [43, 99], in self-management behaviors (including appointment keeping and adherence to antihypertensive medications [43]), and in healthcare utilization (e.g., fewer emergency visits and an increased proportion of patients having a nurse or physician) [43].

CHW capacity in addressing other NCDs in LMICs was not reported in the systematic review literature.

Infectious diseases

 HIV

Task shifting to CHWs may enhance emotional support and increase retention in care, and better link people with HIV to care (one qualitative study) [39, 95, 96].

Task shifting from higher-level providers and clinic-based care to CHWs was generally acceptable to individuals living with HIV [39, 95]. This may enhance dignity and quality of life [50] and increase retention in care [50, 95], without decreasing the quality of care [52] or patient outcomes (such as virologic failure and mortality) [50, 53, 107]. Task shifting and community-based outreach involving CHWs effectively links people living with HIV to care [96].

 Malaria

CHW capacity in addressing malaria in HICs was not reported in the systematic review literature.

There is some evidence of moderate quality that CHWs are effective in malaria prevention [35, 114]. CHWs can perform rapid diagnostic tests with high sensitivity and specificity, and display high levels of adherence to treatment guidelines [21, 33, 61, 76, 86]. There was insufficient research to enable an effect on morbidity or mortality to be estimated [21].

 Other infections

Home visits from CHW can be effective in increasing hepatitis B testing [110] and increasing hepatitis B virus testing uptake (moderate quality evidence) [109].

CHW interventions have helped decrease the incidence of tuberculosis [35] and contributed to the control of neglected tropical diseases [130]. They can support the control of Buruli ulcer in sub-Saharan Africa [47]. CHWs probably increase the number of people with tuberculosis who are cured, though they do not appear to affect the number of people who complete preventive therapy [118].

Link between CHW training and performance (knowledge, skills, and motivation)

All nine studies in one review that described CHW training reported improvements in CHWs knowledge or skills [25]. TBA training was found to be associated with significant increases in TBA knowledge, improved attitude, behavior and advice for antenatal care, and improved pregnancy outcomes [22, 82]. Training and supervision are vital for high-quality performance in integrated community case management programs [27]. Although no studies included in Kok et al.’s review examined the impact of CHW training on health outcomes, training was found to influence CHW motivation, job satisfaction, and performance in the following ways [15]:

• Training generally resulted in expanded CHW knowledge and performance

• Training linked to allowances and favoritism could lead to demotivation

• Continuous training increased job satisfaction/motivation

• Training should include counseling and communication skills

• Training can increase community confidence in CHWs

Beneficial approaches to training (e.g., continuous education and mixing of training components)

For CHW training to improve CHW performance it must include a mix of approaches (knowledge- and skills-based) [15, 21, 48], be complemented by ongoing field-based mentoring and back-up support, [15, 20, 21] and enable CHWs to have an increased sense of self-efficacy, mastery of the tasks, and self-esteem [15, 48]. In CHW programs for common peripartum mental disorders in women in LMICs, continuous supervision was found to be more effective than one-off training [73]. However, the frequency of refresher training had no effect on guideline adherence [15] and training duration had no consistent effect on the effectiveness of the intervention [24, 42]. CHW technical competency tends to drop after training, necessitating follow-up and regular supervised practice opportunities [40, 131].

Supervision appears to be effective in combination with other supports

• Supervision is critical to maintain quality and motivation [19, 23, 33, 35, 76, 132].

• In integrated community case management programs, supervision and on-site training of CHWs improved clinical practices, with providers showing increased knowledge, increased effectiveness in promoting care-seeking behaviors, or improved basic disease management [27].

• Frequent supervision and continuous training led to better CHW performance in certain settings, but the evidence is mixed [15].

Many unknowns and need more research

• There is some evidence of benefit for health care performance, but evidence quality is low [30] and follow-up is limited [29].

• Supervision and training were often mentioned as facilitating factors, but few studies have tested which approaches work best or how these were best implemented [15].

What might work?

• Supervision that focuses on supportive approaches, quality assurance and problem solving may be most effective at improving CHW performance (as opposed to more bureaucratic and punitive approaches) [15, 29, 31].

• Enhanced supervision of CHWs was only superior to routine supervision in two low quality-studies, which examined the effect of regular, supportive supervision and the use of checklists on workforce performance [30].

• Less-intensive supervision of CHWs in one study of low quality did not show any adverse effect on the quality of care or health workers attrition [30].

• Improving supervision quality has a greater impact than increasing frequency of supervision alone [31].

Regular supplies enable effectiveness

Directly: Equipping CHWs with the medicines (e.g., drug kits) and supplies (e.g., rapid diagnostic tests, job aids such as checklist and patient forms) that they are trained to use and mandated to have enables them to perform their related roles [27, 35, 38].

Indirectly: Community trust and respect can be eroded if CHWs experience frequent stock outs or do not have access to the supplies needed to perform their role [35, 38].

Need for travel support in remote areas

• Travel can be a barrier to effectiveness as CHWs are dependent on road infrastructure and transportation options (e.g., availability of busses); bicycles or a transportation allowance can support CHW access in remote areas [15].

mHealth tools are being explored

• mHealth (mobile technology: phones, personal digital assistants) is being explored as a tool to support CHW work through assisting with diagnostics and enabling communication, reminders, and reporting between the periphery with the center [15, 44, 101, 102].

Low-tech job aids support CHW activities

• Counting beads can be designed to support assessment of rapid breathing [78].

• Treatment cards that remind CHWs how to prescribe drugs [15] and pictorial instructions for rapid diagnostic tests for malaria [61] can improve adherence to guidelines.

• Checklists and standard record forms are considered “best practice” for some HIV CHW programs [53].

Financial incentives

Financial incentives increased motivation: one study in Kok et al.’s review found that CHWs getting financial incentives performed better than CHWs receiving in-kind incentives [15]. However, performance-based incentives focus CHW efforts toward remunerated tasks [15].

Other incentives

Other important incentives are community respect, trust, and recognition (discussed in “Community embeddedness”); personal growth and learning; and access to career progression and other future opportunities [15].

CHW rights

Performance-based incentives, linked to CHWs’ volunteer status and flexible tasks and timings, do not provide financial security and ultimately impede CHW rights [41].

Of central importance

Community embeddedness is associated with CHW retention, motivation, performance, accountability, support, and ultimately the acceptability and uptake of CHWs’ health-related work [14, 15, 19, 23, 34, 35, 37, 40, 43‐48, 107].

Mechanisms to foster community embeddedness

Community embeddedness can be fostered through [15, 19, 48]:

• Community members being involved in CHW selection and selecting a locally admired and trusted person

• Community having a clear understanding of and reasonable expectations for their CHW

• Community monitoring of CHWs

• Community ownership of the CHW program

• Community involvement in selection of activities and priority-setting of CHW work

• Health system backs up the CHWs with supervision, supplies and support, which in turn helps to maintain community trust in CHWs

Evidence that CHWs are cost-effective

• CHWs in LMICs are cost effective when compared to standard care for tuberculosis; weaker evidence of cost effectiveness is present for other areas (malaria programs and reproductive, maternal, newborn, and child health) [55].

• Task shifting to CHWs from higher-level staff for HIV care in LMICs is cost effective [50, 52, 53].

• There is a cost savings of 24% when CHWs collect data using personal digital assistants compared to when they use traditional manual methods of data collection and transmission [44].

• Women’s groups (which were almost always facilitated by CHWs) practicing participatory learning and action to improve maternal and newborn health in LMICs were cost-effective as defined by WHO standards [105].

• Pediatric asthma care in HICs by CHWs may be cost-effective [56, 57].

• Diabetes care in HICs by CHWs could save US$2000 annually per Medicaid participant (according to one study) [125]; yield a return on investment of $2.28 per dollar invested (one study) [125], and reduce inappropriate health care utilization [100].

• Community case management of malaria by CHWs using rapic diagnostic tests  is cost-effective in areas with low-to-medium prevalence [21].

• Potential cost savings are present by using CHWs for mental, neurological, and substance-abuse disorders in LMICs [54].

Some cost-effectiveness analyses found no evidence

• The evidence regarding the cost effectiveness of vaccination promotion by CHWs in LMICs is inconclusive [59].

• There are no studies of the cost effectiveness of CHWs for the support of HIC populations with vascular disease [60].

• There are insufficient data to assess the cost-effectiveness of CHWs in the USA underserved groups compared to other types of community health interventions [25].

Integration with the health system is essential for having strong programs

• Integration and cooperation with the broader health system and existing healthcare providers was the most frequently cited enabling factor for CHW programs in one review [23] and discussed as a vital enabler in many other reviews [15, 19, 20, 35, 38, 47, 48].

• The lack of a national CHW policy has been linked to:

  • Inadequate support and recognition for CHWs, which limits their ability to function effectively in the community;

  • Issues around role definition (e.g., whether CHWs should treat illnesses and prescribe medications) [24].

Scaling up and integrating CHW programs with health systems has risks and pitfalls

• A national CHW policy by itself is insufficient; the health system needs to be equipped to supervise, support, and incentivize CHWs [24].

• Scaled-up, integrated CHW programs are often less effective than small, NGO CHW programs because insufficient attention is given to maintaining the quality of the training, supervision, and motivation of CHWs in scaled-up programs [42].

• Integration with a dysfunctional health system can erode CHW programs [35].

Integration with health systems should be built on collaborative, respectful relationships

• Integration must foster respectful collaboration and trust between CHWs and the health system, and it can be facilitated by role clarity and effective two-way communication [15] (potentially supported by mHealth [44]).

• The less hierarchical and the more collaborative are relationships between CHWs and the health system, the greater is the likelihood of benefitting from the unique, practical knowledge that CHWs have [19]; moreover, these collaborative relationships can support CHW retention [19, 24, 38].

• Engagement with stakeholders (policymakers, government officials, civil society and communities) fosters integration by enhancing acceptability and credibility of the CHW program [38].