A recent report from the Burnet Institute, Compass and World Vision argues for a significant expansion of the PNG Village Health Volunteer (VHV) program (a form of family and community care) to deliver a range of cost-effective interventions to improve the parlous state of maternal and child health in Papua New Guinea.  The report claims that full coverage of family and community care in PNG could prevent up to 32% of maternal deaths, 70% of newborn deaths, and 50% of child deaths.

Underpinning this claim is a cogent and accessible mixture of international and more limited PNG evidence to support the content of cost-effective packages, and to make the case that these packages can be delivered by non-professional community health workers (Village Health Volunteers in PNG).

There is much to recommend in the report.  In particular it does a good job in assembling a vast array of technical information, and prioritising these into sensible, graduated packages for potential delivery to the women and children of PNG.

However it is the question of how the recommended interventions are to be delivered in the PNG context that requires further consideration.  The report recognises there is “limited evidence on the impact, cost and effective implementation strategies of many maternal, newborn and child health interventions in PNG” and a stronger evidence base is required to support scale-up (p. 31).  This is certainly true.  There is an urgent need to conduct an evaluation of the performance (and reasons for good/poor performance) of the existing VHV schemes in PNG – particularly those that have been long running.

However, even without such an evaluation, there is evidence from both PNG and internationally to suggest that considerable thought is required as to the best way to expand community-based access to maternal and child health services in PNG.

WHO has conducted a comprehensive review of the international evidence of the effectiveness of community health worker (CHW) programs.  This report concludes:

• CHWs can contribute to improved access and coverage of basic health services, particularly in child health, and some disease control programs (although there is limited evidence of maternal health improvements). However, CHW schemes do not consistently provide services likely to have substantial health impact and the quality of services they provide is sometimes poor.
• For CHWs to be able to make an effective contribution, they must be carefully selected, appropriately trained, and adequately and continuously supported. Large-scale CHW systems require substantial increases in support for training, management, supervision and logistics.
• CHW programs are therefore neither the panacea for weak health systems nor a cheap option to provide access to health care for underserved populations. Numerous programmes have failed in the past because of unrealistic expectations, poor planning and an underestimation of the effort and input required to make them work.
• CHW programs are vulnerable unless they are driven, owned by and firmly embedded in communities themselves. Where this is not the case, they exist on the geographical and organizational periphery of the formal health system, exposed to the moods of policy swings without the wherewithal to lobby and advocate their cause, and thus are often fragile and unsustainable.  Evidence suggests that CHW programmes thrive in mobilized communities but struggle where they are given the responsibility of galvanizing and mobilizing communities.

Without pre-empting the findings of an in-depth evaluation of the PNG VHV experience, it is likely that most, if not all of the above issues will be present in PNG.

• The question of health impact of the schemes in PNG is critical but to date is unknown.
• Even if a heath impact can be demonstrated, it is likely there will be critical institutional questions of if/how existing or expanded schemes are being supported and sustained – both from within the formal system and from communities.  Insights from the Burnet/World Vision report suggest these issues are substantial in PNG. The report points to “challenges in providing refresher training, and supervision and supply shortages” (p. 19), and concerns from volunteers around payment, workload, and lack of support from village leadership (p. 32).
• There is also the important issue that despite a number of intensive efforts (largely supported by donors) to strengthen the management of VHV programs at the national and provincial levels, these efforts appear not to have taken hold (p. 31).  The reasons for this apparent lack of commitment to the approach need to be understood.
• It is also relevant that the major historical effort in PNG to produce and support a community-based health worker (the aid post system) is in decline (at best only 70% of aid posts are functioning).  Prima facie, this decline is driven by the same issues that have bedevilled international community health programs, and appear to be facing the various PNG Village Health Volunteer schemes – ie. lack of support, supervision and supply from the formal health system, and lack of acceptance and ownership from communities.

Thus in short, the key question in PNG is not knowing what interventions need to be provided to improve maternal and child health. The Burnet/Compass/World Vision report does an excellent job in synthesising and increasing the accessibility of this information. However, the risk is that the conclusions based on this information alone will lead to simplistic recommendations and claims that small amounts of money will produce dramatic results. This is the path to disappointment. Rather, the key question is how these services can be effectively and sustainably provided in order to reach an increased number of Papua New Guineans. It is important that the urgent need to increase the accessibility of cost effective, lifesaving maternal and child interventions does not override the need to think carefully and strategically about the best way to achieve this.

It is clear that VHV programs require significant commitment and capacity in term of training, supervision, and logistics. In the context of limited health systems capacity in PNG (which has been unable to consistently provide support to health centres and aid posts) an early strategic question is whether it is realistic to expect the system to be able to effectively support current or expanded VHV programs? If not, are there alternative ways this support can be provided – which are both cost-effective and acceptable to PNG stakeholders?

It is also clear that VHV programs need to be anchored in some form of community ownership. The limited evidence from PNG would suggest this has not been achieved – for VHV’s or aid posts.  If this is the case, the key question is why previous efforts have failed, and are there other approaches which may be more effective?

Given the above, the key challenge appears to be institutional not technical.  In essence, it is how to situate Village Health Volunteer programs so that they garner both the necessary health system and community support, and thus have the best chance of improving maternal and child health outcomes in PNG.

Fortunately, there is a rich reservoir of experience internationally and in PNG of efforts to deliver community-based health services to inform responses to this challenge.

First and foremost, the reasons for success and failure of past community-based delivery experiences in PNG need to be understood, and where relevant lessons learnt, incorporated into future strategies.

However at the same time, these reviews should not only look at what has been tried before.   An in-depth understanding of the aid post and VHV experience to date in PNG will undoubtedly produce stories of success and failure. However, it may also reveal pockets of innovation that work in a different way than was intended – where managers, providers and communities in effect have created ‘practical hybrids’ that combine traditional and formal ways of working to solve problems in the local contexts. It is also likely there will be other providers and resources at a community level – private practitioners, traditional healers, pharmacies, retired/off duty public providers, community groups/associations – who are potential allies in the expansion of cost-effective packages.

These unexpected successes and untapped resources may contain the clues for possible new institutional arrangements that span the formal/informal, and as such provide for different, more sustainable ways to provide community-based health services in PNG.  At the very least, such approaches need to be considered alongside the lessons learnt from traditional community/village health worker schemes in order to assess the full array of possible ways to improve the accessibility of community-based maternal and child health services in Papua New Guinea.

Chris Morgan and Abby Byrne have provided a response to this post on our blog here.

Andrew McNee is a Visiting Fellow at the Development Policy Centre at the ANU’s Crawford School.