Culture, status and context in community health worker pay: Pitfalls and opportunities for policy research. A commentary on

Introduction

The sustainability and fairness of drawing on low-income women and to a lesser extent men for volunteer community health labor are hotly-contested and increasingly studied (Akintola, 2008, Maes, in press, Maes et al., 2010, Maes et al., 2010, Rödlach, 2009, Swidler and Watkins, 2009). In some nations, delivery of health care services, goods, and technologies relies heavily upon volunteer labor. At a societal level, large volunteer programs depend upon (i) expansive “pools” of people who are available (often because of underemployment) to serve as unpaid labor and (ii) the legal status of organizing and using unpaid labor. The promotion of volunteerism also depends on the perceived health needs of communities and the empowerment of communities to advocate for public and private (and mixed) services that might meet their needs. At an individual level, volunteering is tied to livelihood security, gender, social status, and shared understandings of service. Indeed high stakes and potential controversy are involved in questions of global health volunteerism in low-income settings. Research that examines volunteerism can thus have major impacts on the design of models for delivering health services and goods, and on the livelihoods of community health workers (CHW) and those they serve around the world.

In their Social Science & Medicine article “The female community health volunteer programme in Nepal: Decision makers’ perceptions of volunteerism, payment and other incentives” Glenton et al. (2010) make the apparently sound recommendation that, because volunteer CHW operate in various cultural and social contexts, a one-size-fits-all scheme for remunerating and sustaining “high-performance” CHW around the world is inappropriate. And yet the authors’ assertion that wages might be inappropriate in some contexts (particularly in the Nepal Female Community Health Volunteer [FCHV] Programme that they studied) is conceptually oversimplified and based on tenuous evidence. Glenton et al. make claims that have significant implications for development in a variety of contexts, and especially for women’s roles in healthcare in Nepal. We therefore aim in this commentary to (i) point out major shortcomings of the research reported by Glenton et al. and (ii) outline more rigorous research designs that are better positioned to make policy recommendations on the important issue of volunteer labor in global health.

Methodological pitfalls and opportunities

The Glenton et al. (2010) article reports on semi-structured interviews conducted with 19 purposively-selected government officials (“non-volunteer stakeholders”), 4 volunteers, and 2 activists affiliated with the FCHV program in Nepal. The article claims that, among officials and volunteers alike, regular wages are seen as financially unfeasible and also as a “potential threat to the Volunteers’ social respect, and thereby to their motivation” (Glenton et al., p.1920). The authors then suggest that “it may not be useful to promote a genericrange of incentives, such as wages, to improve community health worker programme sustainability” (p.1920).

Unfortunately, the implementation and analysis of the study demonstrate inadequate attempts to avoid, mitigate, and/or address through in-depth interviewing potential biases in the sample related to gender and status differences. Only at the end of the article do the authors note that the key informants they interviewed in the Nepal Ministry of Health and Population (MoHP) were mostly “men in senior positions” (p. 1926). Because the key informants were from the MoHP and other high status positions, the main conclusion that can be drawn from this study is that male officials who help or have helped organize the FCHV program are in favor of using female volunteers for unpaid healthcare labor.

Furthermore, it is highly problematic to select and interview only 6 volunteers and activists and assume that their views are representative of the majority of volunteers. Besides the problem of representation, those who did participate apparently had little opportunity to contest the FCHV program. The presence during the interviews of an employee from one of the agencies (the Nepal Family Health Program) supporting the FCHV program may have discouraged respondents from making statements critiquing the program, even though the authors dismiss this as “unlikely” (p. 1926). Similarly, the 2007 New Era report on the Nepal FCHV program, which the authors use to contextualize much of their findings, used a survey team composed entirely of men (New Era, 2007: 49). This demonstrates a major missed opportunity to allow female participants to openly express their views to other women.

Making recommendations based on such potentially biased data is extremely problematic. The researchers would therefore have done well to attend to and address the biases and internal inconsistencies in their informants’ verbal reports, a necessary task in qualitative research.

For example, it would have been very helpful to evaluate the internal consistency of informants’ claims about how wages would jeopardize religious merit-seeking among volunteers. The researchers could have asked the non-volunteer informants to explain how they achieved religious merit (dharma) given the fact that they were government officials paid well for their labor. Traditionally, a major form of social status acquisition in Nepal comes from making donations to build temples and stupas, or to throw large weddings and other celebrations. The opportunity to gain status through these activities is traditionally monopolized by men because they control the income necessary for these endeavors. Women deprived of wages are deprived of opportunities to participate in such merit- and social status-earning activities. One informant, an FCHV program evaluator, claimed that paying wages to women volunteers would “kill” the philosophy of receiving a seat in heaven for serving people. This should raise the question: how do men achieve dharma and gain a seat in heaven if they get paid for their work? The authors also fail to mention that dharma is the center of a major public relations campaign used to motivate and maintain the FCHV program, which includes weekly radio broadcasts called Sewa Nai Dharma Ho (“Service is Reward”). Thus, when informants in the study invoked dharma as a motivation for volunteering, they may have been reiterating radio and other PR messages.

Similarly, respondents in Glenton et al.s’ (2010) study claimed that there exists among community members a “widespread negative attitude toward paid health workers and so-called ‘salary men,’ particularly government employees” (p. 1923). Given that the majority of informants were salaried government employees, it would have been revealing to follow up such claims by encouraging respondents to address their own status in the communities they inhabit and serve. The article states that early in the FCHV program, volunteers were paid a monthly stipend of 100 Nepali Rupees, but that this was discontinued because it was viewed as “financially unsustainable” (p. 1923). Roughly speaking, then, the cost of paying 50,000 health workers would be on par with the cost of employing a dozen expatriate development officials. In other words, the “enormous” financial implications of paying volunteers could easily be met if a handful of high level stakeholders were to volunteer their time instead of being paid. Posed carefully to the informants in the study, this point could have generated very interesting commentaries and led the analysis to different conclusions. The article in fact indicates that “demands [by volunteers] for salaries have been made in the past few years” (p. 1923). And yet the authors appear to accept their informants’ doubts as to the representativeness of the individuals and organizations making these demands, despite the fact that the informants, by nature of their sociopolitical status, have economic reasons to dismiss demands for salaries among unpaid health workers.

Because Glenton et al. (2010) develop a policy recommendation discouraging the payment of wages to workers without contextualizing and critically examining the comments of their informants, these are more than just methodological concerns. They are pitfalls. Understanding the views of the officials interviewed in the study is a worthy endeavor – for one thing, this could shed light on why volunteers continue to be underpaid. But the endeavor requires more critical interviewing and analysis. If one aims to make policy recommendations with significant and potentially long-term impacts on the empowerment of women in development, then it is necessary to conduct a much more rigorous study of the views and experiences of volunteers.

Such a study would benefit from participant observation, a powerful method that is central to anthropological qualitative research. Other methods can also be relied upon, including in-depth life history interviews with a representative sample of CHW. A recent study of women’s empowerment (Hadley, Brewis, & Pike, 2010) demonstrates that representative survey data can also be used to investigate relationships among women’s autonomy, economic insecurity, employment status, reproductive history, and wellbeing. One could apply these more in-depth methods to important questions of (i) the individual- and group-level impacts of receiving wages for one’s labor, and (ii) volunteer attrition or “drop-out”. Glenton et al. fail to describe the tremendous heterogeneity of the FCHV program in Nepal. Although the average attrition rate in the program is roughly 4%, there is significant variation by district, with seven districts reporting turnover rates of 40–55% (New Era, 2007: 11).

In sum, methods should be mixed and biases avoided or mitigated in an effort to understand the experiences and values of more marginalized groups or “stakeholders” affected by a certain policy question. This is particularly important if one aims to develop recommendations or interventions that benefit and/or minimize harmful outcomes for all interested parties, particularly for those who are marginalized and economically insecure. These caveats alone should call into question Glenton et al.’s (2010) conclusion that wages are not compatible with local traditions in rural Nepal. Yet there are a number of conceptual problems that further challenge the assertion that features of the local moral context are an excuse for not paying wages.

The importance of context – gender, globalization, and health systems

Glenton et al. (2010) make the claim that cultural context justifies not paying women in Nepal for their healthcare labor. Thus the study falls into the trap of invoking “culture” to justify differential treatment of already disadvantaged groups (Hruschka & Hadley, 2008). Several studies have described how health professionals – some in positions similar to those of the stakeholders interviewed in the Glenton et al. study – inappropriately attribute high rates of disease in impoverished populations to the cultural beliefs and practices of those populations. In the process, they obscure political-economic and social determinants of health disparities and the need to improve service delivery to marginalized populations (Briggs and Mantini-Briggs, 2003, Farmer, 1990, Packard and Epstein, 1991). These studies illustrate the need to be skeptical whenever cultural beliefs of impoverished groups are used to justify providing these populations with fewer resources.

Currently, Nepal is one of the three lowest-ranked countries in Asia on the UN Human Development Index (HDI). Nepal’s Gender-related Development Index (GDI), which attempts to quantify labor burden differences between women and men, is on the low end globally (UNDP, 2009). This is in part because only 49% of adult women in Nepal receive compensation for their labor whereas 68% of men are involved in compensation-based activities (CBS/HMG, 2004). Relative to men in Nepal, women report greater mental health problems (Kohrt et al., 2005, Tausig et al., 2004, Thapa and Hauff, 2005) in both low and high caste groups (Kohrt et al., 2009). Worldwide, Nepal has one of the highest gender disparities in anxiety disorders, with economic insecurity emerging as one of the strongest mediators of this difference (Kohrt & Worthman, 2009).

The claim that wages may be a culturally incongruent form of incentive in rural Nepal also disregards global-level historical and political-economic contexts in which the FCHV program operates. CHW in various contexts throughout the world exist within global sociopolitical contexts that are rapidly changing, thus altering and being altered by the local moral systems in which volunteers live. Beyond the local context of rural Nepal, it is important to understand the dynamic distribution of global financial and material resources amongst community health stakeholders. Since the Alma Ata declaration the world has witnessed enormous increases in the amount of money invested in global health (often for specific diseases such as HIV/AIDS), with donor dollars increasing from both public institutions and private philanthropic corporations (Ravishankar et al., 2009). Funds are increasingly channeled from wealthy donor countries to low- and middle-income countries through NGOs as well as the Global Fund to Fight AIDS, Tuberculosis and Malaria and the Global Alliance for Vaccines and Immunization.

And yet there is reluctance among wealthy donors to allow funds to be used for creating health care jobs in low-income countries and for paying steady wages and salaries (Ooms, Van Damme, & Temmerman, 2007). This reluctance is contingent upon perceptions of economic parameters, not of potential cultural incongruence in local communities. This is despite the fact that shortages of health workers in developing countries have been identified by the WHO (2006) as a critical bottleneck in turning donor funds into positive population health outcomes in the face of multiple challenges. It is within this changing context that cheap, volunteer labor has emerged as a major force in global health. The inherent inequality in community health programs in developing countries that rely simultaneously on international donor funding and local volunteer labor reminds us that forms of volunteerism must be situated within their broader sociopolitical contexts.

Furthermore, recent food and financial crises likely have negative effects on local experiences of economic insecurity among CHW in various contexts. Thus it is unfortunate that Glenton et al. (2010) fail to engage with a crucial part of the WHO recommendation (WHO, 2008) they criticize: that “stipends, travel allowances and other non-financial incentives are not enough to ensure the livelihood of health workers” (Glenton et al., 2010: 1921). Acknowledging the social determinants of poverty, poor health, primary health care system failure, and human resource scarcity in poor countries, Paul Farmer and others have argued that CHW who serve on a volunteer basis should be incorporated into the public sector and remunerated (see the U.S. PBS-produced documentary available online: http://www.pbs.org/now/shows/537/index.html Accessed 17.05.10). This policy measure is meant to be just one aspect of wider efforts to strengthen public health systems and ensure human rights to health and economic security in poor countries.

In order to arrive at sound policy recommendations regarding remuneration for volunteer CHWs, one should assess the experiences of economic insecurity among low-income volunteers, as well as the evolving goals of international donors, policy makers, and implementing organizations with regard to financing health systems strengthening. Addressing these issues leads to the following policy questions that call out for debate, but are absent from Glenton et al.s’ article:

• What is the value of the labor of community health workers?

• What is fair compensation for this labor?

• Who can and/or should pay for such an investment?

In debating answers to these questions, one must give due attention to historical international relationships that in many ways have influenced labor relations and economic inequalities in the developing world today. In sum, the promotion of volunteerism in these contexts does not take place in a political-economic vacuum. One must account for larger political-economic contexts in order to fully understand i) what it means to pay economically insecure people modest but predictable wages for their health care labor, and ii) the possible socioeconomic outcomes of paying wages to health workers in various low-income contexts.

Incentivization and innovation

Glenton et al. (2010) assert that the consequences of paying wages to the Nepali CHW in question would be predictable and negative. Based on the qualitative data they collected, the authors reason that receiving wages would cause the female workers to (i) lose respect in the eyes of their fellow community members, (ii) lose spiritual merit, and (iii) lose the intrinsic pro-social motivations that the authors theorize to exist among individual volunteers until wage-earning motivations move in to “crowd” out their altruistic tendencies. Introducing wages will certainly have impacts on the psychosocial and spiritual experiences of volunteers. But one cannot easily predict those impacts. The recommendation that wages are problematic in the local setting fails to appreciate the possibility for naked market-based interventions such as paying wages to be implemented in ways that seek to promote both livelihood security and harmony among community members (including, perhaps, divine entities).

Glenton et al. (2010) assert that non-wage incentives that celebrate the role of volunteers are likely to maintain the social respect enjoyed by volunteers and the intrinsic rewards they experience through service. What if it were possible to pay community health workers modestly and celebrate their roles, thereby seeking to support their livelihoods and their pro-social motivations and respect in the community? It should be possible, for wages and other forms of incentives are not mutually exclusive. There are many examples of paid health professionals in Nepal, as in other places, who receive wages or salaries and are respected and cherished by men and women in their communities because they are recognized as sensitive, competent caregivers.

Ritualized social events, for example, play a crucial role in motivating CHW and in fostering their acceptance and respect by fellow community members. The ritual reinforcement of pro-social values among volunteers occurs in situations ranging from appreciation and initiation ceremonies to every-day interactions between supervisors, patients, and volunteers (Maes, 2010). Glenton et al. (2010) mention an annual FCHV Day (p. 1921), but surely there are other ceremonies and more mundane interactions between volunteers and others in their communities that serve to promote pro-social motivations and respect for volunteers. If one were to begin paying wages, such an intervention would meet the ritual environment already in place, in other words the set of repetitive and more or less sacred social interactions that communicate shared understandings and expectations (Collins, 2004). Meanings and relationships will indeed change, but outcomes like the moral status of volunteers will be difficult to predict. Introducing wages will certainly not be like flipping a switch changing the moral status of volunteers from positive to negative. Rituals to motivate and define the moral status of volunteers are in fact always evolving, often because local and foreign stakeholders realize that rituals can be ‘tweaked’ and innovated in order to improve morale and shared understandings.

Glenton et al. (2010) claim that all the stakeholders they interviewed for their study “expressed a fear that salaries [for volunteers] would undermine the Programme by threatening what they referred to as the FCHV’s volunteer spirit” (p. 1923). The assumption that local communities in low-income countries are full of “untapped” moral and social energy (i.e. “volunteer spirit”) is in fact apparent in discourses of major international health NGOs (Maes, forthcoming). The Secretary General of the United Nations noted in his statement on International Volunteer Day 2008 that “The altruistic spirit of volunteerism is immense and renewable. On this International Volunteer Day, I urge all members of our global community to tap this great reserve of energy and initiative” (see http://www.worldvolunteerweb.org/fileadmin/docdb/pdf/2008/World_Volunteer_Web_stuff/IVD_reports_2008/Ethiopia_IVD_2008_report.pdf Accessed 02.03.10). This statement nicely sums up the interest that community health programmers have in both promoting (i.e. “renewing”) and “tapping” the pro-social spirit of volunteers – in conjunction with their physical labor. Health programmers like the high-ranking officials interviewed in the study by Glenton et al. (2010) most likely recognize – and certainly benefit from – this ritualized economy of labor and psychosocial capital, but it has gone largely unexamined by social scientists. In any setting, it is problematic to ignore these processes, just as it is unacceptable to disregard gender and status inequalities. Ignoring these processes means missing opportunities to develop ways to pay CHW for their labor, thereby potentially supporting their livelihoods, while promoting their pro-social motivations and social standing in the community.

Conclusion

Biased and decontextualized qualitative studies like that reported by Glenton et al. (2010)unfortunately miss the larger context and come up short on ethnographic insights. Thus they are not well positioned to speak in terms of an “overemphasis” on the payment of wages to CHWs. In-depth ethnographic research coupled with well-designed surveys that include large samples of volunteers are necessary to clarify what is at stake for the multiple public, private, and state parties involved in volunteer-based community health programs in various cultural and socioeconomic contexts. Analysis, too, should take into account the political-economic significance of community health volunteerism. Arguably, such research will lead to better policy-making – for the volunteers, beneficiaries, and wider public health systems.