Introduction

Mechanisms that enable primary health care (PHC) to effectively reach and engage with communities can have important benefits for population health [1]. In many settings—primarily in low and lower middle-income countries (LMICs) but also in many higher income countries—“community health workers” (CHW) have played a central role in this effort. The CHW concept has been around for many decades. They are known by many names and have played a wide range of roles. Over the past 10–15 years there has been a resurgence of interest in their potential contribution, most recently during the acute COVID response (with new duties assigned to them), and this has been accompanied by calls for greater formalization of their roles and for better support and working conditions [2,3].

With increasing attention to the potential contribution of CHWs and increasing numbers of countries seeking to incorporate CHW cadres within the PHC workforce, one of the challenges in the discourse and policy debates around CHWs is that there are very different understandings on the definition of CHWs. Approximately 45 different descriptors have been used to describe CHW cadres around the world, but they remain conceptually poorly defined [4]. As pointed out in the 2018 CHW guidance from WHO [5],

“Unclear nomenclature and classification complicate the policy discourse on CHWs: the term “community health workers” is often used in a non-specific way, referring to a diverse typology of lay and educated, formal and informal, paid and unpaid health workers.”

Given that CHW programs have emerged largely in response to local needs, and these needs and hence roles vary and have changed over time, this diversity in how CHWs are defined and classified is understandable. Authors have constructed CHW taxonomies based on several considerations:

• the roles CHWs may play, specifically in immunization programs [6],
• education level required and duration of pre-service training [7,8],
• relative emphasis on promotive vs. preventive vs. curative tasks [8], and
• relative strength of provisions for supervision and other health system support functions [8].

These proposed taxonomies draw attention to important differentiating characteristics of CHW programs; however, we believe there is still imprecision in where we draw the boundaries around those we are calling CHWs and how we differentiate across different types, under the broad heading of “CHW”. With further clarification of terminology and taxonomy, expected benefits include:

• clearer communication [9,10],
• more standardized reporting [11–14],
• clearer framing for synthesizing evidence from program experience [12,15], and
• better transferability of lessons learned across different settings and types of programs.

There are also other occupational groups whose work aligns closely with what is understood to be the role of a CHW (and how this role is described by the International Labor Organization (ILO), as discussed below) but who are not typically considered CHWs. For example, in Uganda, there are “nursing assistants”, “enrolled nurses”, and “enrolled midwives” staffing “Health Center IIs” who play a role quite similar to that of Health Extension Workers associated with “health posts” in Ethiopia, who are widely seen as CHWs. However, because these Ugandan cadres are not normally thought of as CHWs, they were passed over when the country began to roll out “integrated community case-management” of childhood illness (iCCM), in contrast to implementation of iCCM in Ethiopia, where this was done primarily through Health Extension Workers.

And then there is the “worker” portion of this label. For paid full-time or regular part-time staff, “worker” seems an appropriate term. But there are diverse arrangements for how PHC engages with members of the community. Often there are community members who are regularly or intermittently engaged in some way with government PHC services and this may or may not include a service delivery role. In some programs, such individuals may receive honoraria, stipends or other monetary incentives. Does that make them “workers”? As we will discuss below, use of the term “volunteer” as blanket term to describe those who are not formally employed may also be problematic.

Within actual country programs, it may not matter if, for a particular role, a different job title is used than that used in another jurisdiction. However, if we want to draw potentially transferable lessons from one country’s experience to apply in another or if we are engaged in developing global or regional program guidance, we need to be comparing apples with apples, and oranges with oranges.

One important factor adding complexity to efforts to develop definitions and terminology for use across diverse settings is the diverse range of formally-employed (usually full-time) workers and various less formally-engaged, less-than-full-time community actors who are connected in some way with government PHC. In this paper, we present a taxonomy that segments across a small number of dimensions, the most important of which is the distinction between—on the one hand—formally-employed workers and—on the other—community members who are actively engaged with government PHC on at least an intermittent basis, but who are not formally employed (we acknowledge the important role, in some settings, of NGO-associated CHWs; in future work, it may be worthwhile to better characterize a taxonomy of such programs. However this paper largely focuses on large-scale government programs).

Methodologic note

This review can be understood as the synthesis of a consensus process. It is not intended as a research article or systematic review. The process in developing the paper includes efforts stretching back several years. This has entailed the development, by various of the co-authors of this paper, of a set of comprehensive program case studies of CHW cadres from 29 countries [16]. It also included work done in 2021, commissioned by the Bill & Melinda Gates Fund, to derive a CHW “typology”, based on information on large-scale government CHW programs and a modified Delphi process seeking perspectives from a diverse range of global and country-level CHW program stakeholders. This process continued in 2023 with multiple rounds of consultation, including a Delphi-type survey, webinars, and an iterative process of circulating drafts to stakeholders, seeking their perspectives, and revising accordingly. This is reflected in our author list, which includes 33 individuals from 19 countries.

Community-based primary health care

In this paper, we are concerned with “community-based primary health care” (CB-PHC). We recognize, however, that there are categories of workers otherwise meeting definitions for CHW whose work is not community-based, e.g., peer counsellors working with people living with HIV but based in clinics and hospitals rather than the community. Furthermore, in the future we can expect development of PHC services, in which our definitions of “community” will not be restricted to physical localities such as villages or neighborhoods; “community” may be understood to include groups of people not necessarily living close together but sharing other characteristics.

For purposes of this article, we are using the term CB-PHC mainly to refer to public-sector PHC services provided below the level of health center; we are not looking at services at the level of “health centers” or “primary health care centers”. In this paper we restrict our use of the term “health center” to health facilities with at least some staffing by professional-level health workers (as defined by ILO) and normally serving populations of 20,000 or more, recognizing that in some countries (e.g., Uganda) the term, “health center”, is also used at a more peripheral level, more or less equivalent to Ethiopia’s Health Posts.

Service delivery modalities or platforms at the CB-PHC level—as we are using this term—may include health posts or dispensaries, outreach/mobile and home-based services. There may be some professional-level providers working at this level, particularly in higher-income countries, but typically services are delivered primarily by “associate professional” workers (using the classification from the ILO [17], discussed in more detail, below). In many countries, this is an important component of rural government PHC and is commonly organized in service delivery units covering populations of about 3,000 to 10,000. Where such a service delivery level exists in urban settings, the populations covered may be larger. Furthermore, we recognize that the way we are defining CB-PHC for the purposes of this paper offers a rather narrow perspective, focusing largely on provision of PHC services at the most peripheral level of government health services. Other entirely legitimate definitions of CB-PHC [18] focus attention less exclusively on services provided by the government and more on actions by members of the community.

Definitions for CHWs

We will begin by briefly reviewing recent definitions from the World Health Organization (WHO) and ILO (see Table 1, below). In the following sections, for the portion of the CHW spectrum comprising formally employed workers, we will draw particularly on classifications and definitions from the ILO’s International Standard Classification of Occupations: ISCO-08 [17]. We believe that the WHO and ILO definitions serve as a good starting point for characterizing CHW programs, although further clarification and elaboration is needed.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Comparison of “Community Health Worker” definitions.

https://doi.org/10.1371/journal.pgph.0004156.t001

To be clear, we are not making the case for standardizing CHW programs; instead, we believe that some further standardization of terminology can be helpful for drawing potentially transferable lessons between programs (positive and negative) and for informing development of global strategy and program guidance. Standardization of terminology may also help national governments that are establishing or institutionalizing CHW programs to clarify scopes and roles within CB-PHC. We are concerned, here, with the language used for evidence synthesis, for global guidance, and perhaps most importantly for exchanges of information and insights between countries. Within a country, however, there may be very good reasons to continue to employ the designations and categories currently in use.

The ILO, in ISCO-08 [17], classifies “community health workers” (3253) at the level of associate professional/ technician, and specifies that CHWs:

“provide health education and referrals for a wide range of services, and provide support and assistance to communities, families and individuals with preventive health measures and gaining access to appropriate curative health and social services. They create a bridge between providers of health, social and community services and communities that may have difficulty in accessing these services”.

Scott et al. [19], in a review conducted for WHO, define CHWs as:

“health workers based in communities (i.e., conducting outreach from their homes and beyond primary health care facilities or based at peripheral health posts that are not staffed by doctors or nurse), who are either paid or volunteer, who are not professionals, and who have fewer than 2 years training but at least some training, if only for a few hours.”

This definition is used in the 2018 WHO guidance [5] for which the review by Scott et al. was commissioned. As this definition comprises “volunteers” as well as formally-employed, paid CHWs, it is broader in scope than the ISCO-08 definition (Table 1).

Largely in line with the definition from Scott and the WHO, we will now offer a synthesis which includes both formally-employed (generally full-time) staff as well as others not formally employed who have commonly been referred to in the literature as “volunteers”, although this term (normally understood as meaning “someone who does work without being paid for it”) can be misleading given the variety, across programs, in time commitment expected and remuneration provided, and in the extent to which the role is primarily defined in relation to the community vs. the health system. Communities may bring significant material support to CHWs (paid or unpaid) who they see as bringing value to the community, in the same way those communities might support others who contribute to the community, such as teachers or religious leaders.

For formally employed workers, we are proposing adherence to the schema proposed by ISCO-08, including the principle of classification based primarily on the tasks involved rather than the credential or duration of training required. As discussed below, we also note that in some programs, CHWs who are involved in service delivery on a regular basis, but not formally employed, may have tasks otherwise corresponding with the ISCO-08 CHW designation.

Details in this paper on specific cadres are drawn mainly from a set of country program case studies by Perry et al. [16], supplemented by further documentation on these programs done by several of the authors in the context of a contract with the Bill and Melinda Gates Foundation, on CB-PHC.

Differentiating across CHW programs

Formally-employed workers in CB-PHC

Terms of service for formally-employed CB-PHC staff may be permanent—typically with benefits such as pension and health insurance—or time-bound, e.g., an annual contract, potentially renewable, generally with limited or no supplementary benefits. The hiring authority is commonly a national or sub-national civil service or local government. In some settings, non-governmental organizations (NGO) engage CHWs under service agreements with public sector agencies, with various arrangements for remuneration.

Low health professional density is common in LMICs, particularly in rural areas. In the Africa region, for example, there are 2·9 medical doctors and 12·9 nursing and midwifery personnel per 10,000 population (ratios would be considerably lower in rural areas); in the Europe region, the corresponding figures are 36·6 and 83·4 (WHO Global Health Observatory [20]). A common strategy to address this gap has been delegating tasks usually performed by health professionals to lower-level occupational groups. Indeed, such “task shifting” or “task sharing” has also been effectively deployed for improving service access in some settings in high-income countries, e.g., engaging physician assistants and nurse-practitioners. There is some evidence that this modality of service provision has benefits even when there is no shortage of clinicians, for example because of improved access, reduced opportunity costs, and greater cultural and linguistic appropriateness.

Many workers engaged in service delivery in CB-PHC fall under what ILO categorizes—in ISCO-08—as the associate professional/technician level (3000). Under ISCO-08, at this level the entry requirement is typically secondary school graduation and the duration of pre-service training is usually at least a year. However, as explained below (Table 2), categorization under ISCO-08 is primarily based on the content of the work rather than on the credential or duration of training.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. ISCO “skill levels” and designation as “professional” vs. “associate professional”.

https://doi.org/10.1371/journal.pgph.0004156.t002

In this paper, we largely restrict our use of the term “professional” to its meaning as used in ISCO-08 (see Table 3), rather than how the terms “professional” and “professionalized” have been used in recent global discourse on CHW programs, referring to salaries, supervision, accreditation, etc.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 3. Task descriptions of relevant “associate professional” occupational groups from ISCO-08 [17].

https://doi.org/10.1371/journal.pgph.0004156.t003

CHWs in many programs do not neatly fit within any one of the 4 skill levels in the ILO scheme, outlined above. Some of their tasks are of a complexity consistent with skill level 3. However, typically the level of education and duration of preservice training required are lower than what is common for level 3. ILO specifies, however, that “formal education and training requirements are only one element used for determining skill level” and, further, that for some occupations, formal education and training requirements differ between countries. Specifically, they note this is the case for primary school teachers and nurses; in some countries, university degrees are required for these occupations, in others, they are not. If categorization by level were based primarily on training requirements, this could result in school teachers and nurses in some countries being classified under the “professional” category and, in others, under “associate professional”. In the interest of improving international comparability, ILO has resolved this inconsistency by ruling that, in assigning formal skill level, job content (tasks and duties) be given priority over education and training requirements. As such, school teachers and registered nurses are classified under “professional,” rather than “associate professional,” even for countries where university-level credentials are not required for these occupations. By this logic, CHWs whose tasks align well with those specified by ILO for 3253 (see Table 3) would be classified under skill level 3, i.e., “associate professional”, even in programs where the required level of education and training may be lower than typical for this level. In its current work revising ISCO, ILO may want to consider recent evidence suggesting that training and experience, rather than formal education, are better predictors of CHW performance [21].

ISCO-08 also has a lower skill-level occupational group (at level 2) with functions somewhat analogous to those of CHWs: “home-based personal care workers” (5322), defined as: “workers who provide personal care [such as assistance with feeding, bathing, grooming, personal hygiene, transfers, e.g., from bed to walker] to residents of independent living units, generally without permanent medical or nursing supervision.” In future revisions to ISCO, potentially the task description for 5322 could be expanded such that his category could be used for some CHW cadres.

In some country programs there are workers engaged in governmental PHC below the health center level (i.e., CB-PHC) who would be categorized under ISCO-08 as professional, not associate professional. Indeed, over time—with rising educational levels and associated “credential creep”—in many countries there has been a shift to an increasing number of professional staff at this level (see Table 4). Examples of professional level staff working in CB-PHC include community-based professional nurses (ISCO-08 code: 2221) and midwives (2222), whose work is PHC/prevention-focused (although it may also involve curative services), serving as public health nurses. In some country programs there are also health professionals at CB-PHC level who can be classified as “paramedical practitioners” (2240), such as Clinical Officers in some African countries (e.g., Zambia, Malawi, Mozambique), Health Assistants in Nepal, and Sub-Assistant Community Medical Officers in Bangladesh. However, in most countries those with such designations are more commonly deployed at health center or hospital levels. These occupational groups, like professional nurses, generally have at least 3 years of pre-service training. Nigeria’s Community Health Extension Workers (CHEWs)—often considered a type of CHW—could also be classified as professionals, falling in the category of “paramedical practitioners” (ISCO-08:2240), given their 3-year duration of pre-service training and a scope of practice spanning that of CHW and paramedical worker. Similarly, WHO [22] has identified Mid-Level Health Workers, a category in which Nigeria’s CHEWs would fit well. Note that we are not proposing use of the “CHW” designation for these ISCO-08 2000-level “professional” workers (though we would include them under a broader designation such as “Frontline PHC Workers”).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 4. “Task-shifting” and “credential creep”.

https://doi.org/10.1371/journal.pgph.0004156.t004

The associate professional/ technician occupational level is where ISCO-08 places “community health workers” (ISCO-08 code: 3253). There are other ISCO-08 occupational categories which, in principle, could be applied to at least some of those working in CB-PHC, notably “nursing associate professionals” (3221) and “midwifery associate professionals” (3222). Indeed, in many countries there are categories of workers with titles such as enrolled, auxiliary or assistant nurse or midwife working in CB-PHC (most with 2 academic years of pre-service training). The tasks listed in Table 3, for these two occupational categories, are most closely aligned with the role of such workers based in hospitals or health centers and working under the direct supervision of professional nurses or other health professionals. However, commonly their counterparts working in CB-PHC have tasks more closely approximating those of the ISCO-08 category “community health worker” (3253), as outlined in Table 3, albeit with more task-shifted clinical functions than we see in this outline of CHW tasks. The actual scope of practice of auxiliary-nurse-type workers (3221, 3222) engaged at health post level and below can be considered to fall in an overlap area spanning 3253 and 3221/3222. However, currently, for none of these 3 categories do the ISCO-08 task descriptions fully reflect the actual scope of practice and level of autonomy of most such workers.

Examples of such occupational groups with roles in this overlap area include Public Health Midwives in Sri Lanka, Auxiliary Nurse-Midwives (ANMs) in India based in Health and Wellness Centers, ANMs in Nepal based in Health Posts, and Enrolled Nurses working at CB-PHC level in many African countries. Conceptually (and for planning purposes), we propose that it is appropriate to consider such cadres together with others more conventionally classified as CHWs. However, how the various occupational groups are designated will be determined by history and other contextual factors; we are not trying to prescribe how they should be officially labeled at this level, within a country.

Pre-service training for formally-employed CHWs and other front-line associate professional health workers with similar scopes of work typically ranges from a few months up to about 2 academic years, in many cases in some kind of a nursing or technical college or training institution. Training may be done before hire (as it is for ANMs in India and Nepal) or after hire (e.g., for Health Extension Workers in Ethiopia and for Agentes Comunitários de Saúde in Brazil). Given their work responsibilities, based on ISCO-08, it can be argued that even those workers in programs in which entry may not require having completed secondary education and pre-service training may be as little as 2 or 3 months (i.e., substantially shorter than what is typical for the ISCO-08 associate professional/ technician category) would still appropriately be designated as “associate-professional”.

A proposed taxonomy

The full spectrum of health workers and community members involved with CB-PHC is reflected below in Fig 1. All of these players, where present, need to be adequately taken into account in government human resource planning and management for PHC, including program monitoring, evaluation, and formal reporting.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Spectrum of PHC players.

https://doi.org/10.1371/journal.pgph.0004156.g001

The dotted section of the line is intended to suggest that, depending on the setting, salaried, associate-professional-level workers are sometimes but not always members of the communities where they serve.

In Table 5 below, we propose 5 categories:

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 5. A proposed taxonomy.

https://doi.org/10.1371/journal.pgph.0004156.t005

1. (1). formally-employed CHWs (or auxiliary workers; they may have designations such as auxiliary nurse-midwife, but their tasks generally align relatively well with ISCO-08:3253; this category corresponds fairly closely to what Olaniran [7] designates “L2 paraprofessionals”);

and 4 categories not formally employed, classified by level of integration within government PHC:

1. (2). regular service-delivery CHWs, normally spending 10+ hours a week on CHW-related tasks (corresponds fairly closely to “L1 paraprofessionals” in Olaniran’s taxonomy),
2. (3). an intermediate category, engaged in a service delivery role, typically spending several hours a week, but fewer than 10,
3. (4). episodic/ occasional service delivery involvement, but no CHW-related activity most weeks, and
4. (5). governance or community action role, e.g., as members of a local health committee.

The last 3 categories roughly correspond to what Olaniran [7] has labeled “lay health workers”.

Note that the table below does not include professional health workers (what WHO refers to as “mid-level workers”) involved in CB-PHC, such as professional public health nurses or clinical officers (at the far left in Fig 1).

Programs with community members actively engaged with government CB-PHC are arrayed along a fluid continuum. In a common scenario, programs may have predominantly episodically/ occasionally-involved CHWs in districts without donor-funded partners but predominantly intermediate-level involvement (i.e., many CHWs involved most weeks) in partner-supported districts. This is a common phenomenon and points to complexities in the relationships between host government and external partners, raising such questions as: Who “owns” these programs? Who is making the decisions? [23]

Furthermore, programs change over time; where we have located specific CHW programs across this spectrum may not, in every case, reflect the current situation. As we have stated earlier, this taxonomy and the criteria used to differentiate categories are intended to be descriptive, not normative.

Other dimensions

Scope of work

Community-based PHC services provided by CHWs vary considerably across programs. Fig 2 presents a spectrum of functions often seen, ranging from service extension, such as clinical tasks and dispensing of health commodities, to promotion activities, such as advertising upcoming campaigns or providing educational materials, and attempting to influence care-seeking and other health-related practices in their communities [25]. The work of formally-employed CHWs typically focuses more on the service extension end of this spectrum. Regular service-delivery CHWs (not formally employed) can be involved in activities across the spectrum, although in most programs they are less involved in what could be characterized as “clinical functions” than are formally-employed CHWs (as illustrated in Fig 2, below). An important exception is sick child care, which has been introduced as part of the scope of work of RSD-CHWs in several dozen countries, under the rubric of Integrated Community Case Management (iCCM). Although they are an important component of the CHW role in many programs, such task-shifted, clinical functions (i.e., individual preventive and curative patient care) are not prominent in the ISCO-08 task description for CHWs, as outlined in Table 4. We propose that this should be addressed in subsequent revisions of ISCO.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 2. Spectrum of CHW functions.

In this figure, moving from left to right: “depot holders” are community members having supplies of certain basic health commodities available to dispense to community members; “campaigns” refers to periodic mass distribution events, e.g., twice-yearly Child Health Days, during which commodities such as vitamin A, deworming medication, LLINs may be distributed and vaccines may be administered. Note that these specific activities are only illustrative; this is not meant to be an exhaustive list. NCD, non-communicable diseases (notably hypertension); FP, family planning; ANC, antenatal care; PNC, postnatal care; ORS, oral rehydration solution; LLINs, long-lasting insecticide-impregnated nets; IFA, iron-folate supplementation; ART, antiretroviral therapy.

https://doi.org/10.1371/journal.pgph.0004156.g002

Discussion/Conclusions

Health systems around the world have different histories, face different challenges, have elaborated different solutions, and are continually evolving. Nevertheless, if we are trying to generalize and learn from program experience in other settings, using more standardized terminology can help us avoid misinterpreting what is reported. In that spirit, we are proposing to the global PHC community a framework and terminology for talking about those providing services in government primary health care at community level.

There is a relatively clean distinction that can be made between those who are formally employed and working full-time and those who are neither. Among formally employed workers in CB-PHC, some can be classified—using ISCO-08 criteria—as professional. They would include professional nurses and clinical officers (typically with 3 or more years of preservice training, as indicated in ISCO-08). Although workers in this category are found at the most peripheral level of government PHC in urban settings in LMICs, they are less common in rural areas. The remaining formally-employed health workers in rural CB-PHC can generally be categorized by ISCO-08 criteria as associate professional. Many of these have tasks that align fairly closely with the ISCO-08 “community health worker” designation (3253), as outlined in Table 4. A subset of this group have “nurse” or “midwife” in their job-titles. We are proposing that, for purposes of between-country comparisons and lesson learning, it is clearer to group such occupational groups together with others having job titles more traditionally recognized as “community health workers”, focusing on actual tasks rather than formal designations.

We are also proposing not to restrict the CHW designation to those with only a few weeks or months of pre-service training (as would be the case with some earlier definitions) but also, consistent with ISCO-08 and WHO, to include workers with 1 to 2 years of pre-service training.

We have chosen not to use the term “volunteer” to designate those not formally employed. This group includes a subset who are engaged as what we are characterizing as “regular service delivery” CHWs. Our use of a cutoff of 10 hours or more per week typically involved in CB-PHC-related activities as a threshold for classifying this group as RSD-CHWs may appear somewhat arbitrary. However, this cut-off has allowed us to sort a number of programs, cleanly, between the “regular” and “intermediate” categories (e.g., India’s ASHAs typically exceed this threshold whereas most of Nepal’s FCHVs and Rwanda’s Binômes normally do not). As we have noted, currently RSD-CHWs—as we have defined this category—are not formally employed although generally they are remunerated in some way and in some cases their remuneration may approach the level of a low-level salaried worker (e.g., ASHAs in some parts of India). It is also worth noting that in essentially all LMIC economies, only a minority of the working-age population is engaged in formal-sector salaried employment [26].

Further to the right on our spectrum of CHWs (Fig 1 and Table 5), we have those engaged in “episodic or occasional” service, defined—again somewhat arbitrarily—as playing such a role less frequently than every week, through most of the year. In between, we have an intermediate group. An important reason for making these distinctions is that expectations are likely to vary considerably across these groups, for example with regard to remuneration. Although, in most settings, it may be unreasonable (and unfair) to expect someone to consistently put in 10 hours or more of work every week largely on an uncompensated basis, such arrangements may be judged, in the local context, to be acceptable and fair when the time commitment is substantially lower (than 10 hours/week) and the involved community members have more discretion with regard to when they’re engaged in such activities, especially if they derive other valued non-monetary benefits from their role and are supported in other ways by their communities.

It is important to recognize that the term “professional”, in addition to the ISCO designation, carries a colloquial meaning of someone engaged in skilled work for compensation (e.g., a professional athlete). This colloquial understanding is particularly relevant in the context of CHWs, many of whom have organized and demanded recognition as professionals, seeking salaries and acknowledgment of their expertise [27]. Indeed, multiple ministries of health have expressed a commitment to “professional CHWs”, in this sense, including accreditation and fair compensation—as advocated in the Monrovia Call to Action [28]. There remain significant challenges: governments having expressed a strong commitment to professionalize CHWs still have to count the cost per worker multiplied by the scale of the program and consider anticipated client reach and financial viability. Furthermore, even community health programs staffed by formally-employed CHWs often also have recourse to Women’s Clubs, Papa-Champions, Women Leaders, Youth Peer Educators, and miscellaneous Community Leaders serving a complementary function as CHWs, of a kind, without an expectation of formalized salaried employment. Some such programs would fall in the category we are proposing of “primarily engaged in governance or community action”, at the right end of the spectrum in Fig 1. Such community agents play an important role and we are not suggesting that they should be formalized.

Unlike certain schemes that have been proposed [8,9], we have not used duration of pre-service training as a primary criterion for classifying programs and, indeed, ISCO-08 gives priority attention to tasks rather than training duration in its categorization of occupational groups. In the programs we have looked at that engage formally-employed CHWs and other workers with equivalent tasks, training duration ranges from as little as a few weeks, for Health Assistants and Family Welfare Assistants in Bangladesh, up to 2 academic years (indeed, up to 3 academic years, if we include Community Health Extension Workers in Nigeria; however, as we have pointed out, CHEWs may fit better at the professional level, under ISCO-08: 2240 “paramedical practitioners”, along with such as cadres as Clinical Officers). Work currently underway at WHO, developing guidance on CHW training, should direct attention to important considerations in developing CHW training programs based on the skills required for their tasks, including issues such as needed duration of training.

There are implications arising from our proposal that will be relevant to ILO for the revision of the ISCO classification, currently underway. The task description for CHWs, under ISCO-08, captures most of the functions outlined in Fig 2, but it does not reflect the “task-shifted” clinical tasks that are common for formally-employed CHWs working in CB-PHC in LMICs (and, indeed, also for many RSD-CHWs), e.g., family planning, treatment of childhood illnesses, antenatal care, or administering vaccines. As we have suggested, the task description for CHWs (3253) should be revised accordingly in the next edition of ISCO. Furthermore, if ISCO codes 3221 (nursing associate professional) and 3222 (midwifery associate professional) are to be applied for cadres working at community level (i.e., rather than hospital or primary healthcare center), task descriptions should reflect their more autonomous roles—in contrast to their roles when working in higher-level health facilities where they are more directly supervised by health professionals.

There are also implications for governments. Although we are not proposing that countries change the formal designation of occupational groups, such as enrolled nurses, who are providing CB-PHC services, we do encourage governments to consider such cadres together with other categories of CHWs in their PHC human resource planning and management (indeed, they should also include professional-level workers at this level). Within the logic of the scheme we are proposing, they are grouped together with “formally employed (associate-professional) CHWs”.

As we have pointed out, many large-scale government programs falling in our category of RSD-CHWs have functions aligning well with the task descriptions for ISCO-08:3253. For such programs, governments and their development partners may want to consider whether further formalization of these CHWs would be appropriate (and feasible), including improved remuneration, supervisory support, supply of program commodities, career trajectories, and integration within health sector human resource planning and management (including related data systems). Such review and reflection may also be appropriate for programs working with cadres we are classifying as “intermediate”.

Also, we should not assume that all these issues are necessarily adequately addressed for formally-employed CHWs. For them to be effective and satisfied in their work, they also need fair (and reliable) remuneration, appropriate training, good supervisory support, reliable supply of program commodities, and attractive opportunities for career progression—dimensions of “professionalization”, as this word has been used in recent CHW discussion and debate, in the Monrovia Call to Action [28], and in WHO guidance [5]. Given the larger magnitude of potential lives saved investing in community-based PHC—vs. at facility-based PHC and hospital levels [29]— it is certainly appropriate that governments commit adequate resources and give serious attention to sound, context-appropriate planning for work at this level. Partners can play a supportive role, helping transition away from partner-driven to fully government-led, context-appropriate, community-based PHC.

Although the primary focus of this paper has been on CHWs who are integrated, at least to some degree, within government PHC services, NGOs working at community level also work with CHWs of various kinds and may also find the proposed taxonomy useful.

Health systems continue to evolve, driven by multiple forces including changing disease burdens, urbanization, professionalization of the workforce, and new technological solutions for service delivery. Nevertheless, we are confident that community-level service delivery and actors playing a boundary-spanning or bridging role between health services and the community will remain important for universal access and equity, and for individual and population health. Furthermore, it will be important to ensure dignified, decent and humanized work for the community health workforce. To build on many decades of accumulated program learning, extending back to Alma-Ata in 1978 [18] and earlier, and to draw on and apply lessons from current programs, we encourage the use of a standardized terminology, allowing us to compare apples with apples, and oranges with oranges.