How not to address maternal mortality

Student midwife examining a pregnant woman in Sierra Leone (Flickr/H6 Partners/Abbie Trayler-Smith CC BY-NC-ND 2.0)
Student midwife examining a pregnant woman in Sierra Leone (Flickr/H6 Partners/Abbie Trayler-Smith CC BY-NC-ND 2.0)

Papua New Guinea is often referred to as the ‘land of the unexpected’. I sometimes wonder whether it might be better referred to as the ‘land of the extreme’.

Extreme is certainly an appropriate adjective to describe the country’s maternal mortality ratio, which is the highest in the Asia-Pacific region: 215 deaths per 100,000 live births. To put this in perspective, in Australia the maternal mortality ratio is 6 per 100,000 nationally, and 14 per 100,000 among Australia’s Aboriginal and Torres Strait Islander women.

Extreme also seems an apt descriptor for PNG Prime Minister Peter O’Neill’s plan to address maternal and infant mortality in his country. As reported by the ABC on Tuesday, O’Neill plans to introduce legislation in early 2017 that will make it mandatory for women to deliver their children in a clinic or hospital. To enable this, women will be paid to attend those clinics or hospitals.

To be fair, few details about O’Neill’s plan have so far been revealed. And it is positive that political will exists to publicly acknowledge and tackle the problem of maternal mortality in PNG. But as the history of public health has shown time and time again, the road to hell – or at least the road to some seriously ineffective interventions – is paved with good intentions.

First, though, to validate those good intentions. What I assume to be the underlying impetus of this hasty policy announcement – the desire to ensure that all women have access to skilled attendance at birth – is a sound one. The most recent data available from the WHO (2013) indicate that less than half of women in PNG deliver their children with the assistance of a trained midwife or doctor. Globally, about 15 percent of women will encounter unexpected complications during or shortly after birth which require emergency obstetric care (EmOC). Skilled attendance and EmOC save lives.

The key question that policymakers must grapple with, then, is how best to enable access to those services in their particular context. A trio of case studies published by UC San Francisco highlights how various countries have been successful by targeting different determinants of why women fail to deliver in health facilities. India, for example, introduced a conditional cash transfer program which incentivized women to deliver at facilities; Malawi pushed women towards facilities by banning traditional birth attendants.

While encouraging women to deliver at health facilities is key, doing so cannot be interpreted as a magic bullet for reducing maternal mortality. Indeed, pushing women to give birth in overcrowded and under-resourced facilities may put them at an increased risk of potentially life-threatening infections, such as sepsis. To that end, it is important to note that the Indian and Malawian interventions described above were not limited to changing mothers’ behaviour, but also included supply-side interventions to improve the availability and quality of maternal health services, such as expanding maternity waiting homes and ambulance services, and training and employing more community-based midwives.

Even in the face of a strong, evidence-based plan for improving the rates of supervised births in PNG, it’s difficult to see how the government will be in a position to implement it. As noted on this blog last month, PNG’s health budget was cut by 21% (K315 million) in the 2017 budget (this followed cuts to health of more than 30% between 2014 and 2015). The number of health facilities that are fully operational in PNG, particularly at the aid post level in rural and remote areas, has dwindled. Health professionals are also few and far between; PNG has just 0.5 nurses/midwives per 1000 people, while the WHO now advocates for a minimum 4.45 skilled health professionals (midwives, nurses and physicians) per 1000. Glen Mola, a professor of obstetrics at UPNG, confirms that the main problem facing the PNG health system is a lack of funding for health staff and facilities; “If we’ve got the money, let’s see it please… we desperately need it,” he told the ABC. In this context, requiring women to travel to facilities to deliver their children simply cannot be automatically equated to their accessing skilled attendance.

Theoretical and practical issues aside, what I personally find most troubling about O’Neill’s outlined proposal is its vaguely punitive tone. Attendance at facilities is to be made “compulsory”. How will this be enforced? Will there be consequences for those women who fail to appear at a health facility, whether that failure is intentional or not? How will the plan ensure that women do not simply show up at facilities, and are not just attended, but that they receive high-quality and respectful care? Abusive and disrespectful care is a serious but underacknowledged problem encountered by many women worldwide, including in PNG, when they attend facilities for antenatal care and delivery.

Again, for now publicly available details of O’Neill’s plan remain limited. Before they are finalised, PNG’s healthcare leaders and frontline workers should be consulted on where the gaps are, and how they might be best addressed. Implementation of the plan – including not only the proposed payments to women to attend facilities, but also the costs of recruiting and training staff to attend birth at those facilities – needs to be fully and realistically costed before it is presented to parliament.

But most importantly, PNG women themselves must be given an opportunity to express their views and experiences around childbirth, in order to craft a policy solution that responds to the real reasons why many either do not seek out or do not have access to skilled attendance at birth. Just as it is an affront to human rights that so many women in PNG (and around the world) go without access to skilled attendance and emergency obstetric care, so too would it be for them to be deprived of the right to give birth where and in the presence of whom they choose, or to be forced to attend facilities that are underequipped to receive them.

Camilla Burkot is a Research Officer at the Development Policy Centre.

image_pdfDownload PDF

Camilla Burkot

Camilla Burkot was a Research Officer at the Development Policy Centre, and Editor of the Devpolicy Blog, from 2015 to 2017. She has a background in social anthropology and holds a Master of Public Health from Columbia University, and has field experience in Eastern and Southern Africa, and PNG. She now works for the Burnet Institute.


  • Papua New Guinea is not Africa as most would possibly think. Although, issues about health, politics, economy etc are vital for academic debate and discussions I would suggest writers should be very cautious of using the pictorial images that they use for their academic publications. The image as used and portrayed in this article must be taken from somewhere in Africa and inserting it in an article that talks about Papua New Guinea is indeed an intellectual fallacy. I suggest the author of this article to withdraw the image inserted in this article. Thank you.

    • Hello Gabriel, thank you for your comment. You are quite right. I’d like to note that at the time this post was published, the photo caption clearly stating the origin of the photo was visible, but it was lost in the intervening years when this blog was re-designed. The editors have restored the caption now. Nevertheless, I appreciate your concerns about the perceived misrepresentation and I apologise for that. Thank you,

  • All over the world in the hospitals Nursing care ( bathing patients, changing their bed linens, sitting them up for meals, turning bed patients and cleaning them up changing their pads, rubbing their backs for good circulations) is done by nurses who are trained to do these jobs. But nursing care is done by poor relatives and wantoks of the patients. Theses people have got no idea, they know nothing about nursing care and they are not train to do NC of the patients. My beloved big sist fasted away 3 wks ago with stroke, she was lack of NC. My niece did everything for her mum. No nurse was there to help my niece in the medical ward. Nurses just come to throw medicines check vital signs and gone. PNG Government should employ more nursing staff or more staff to assist nurses with NC. Thousands of PNGans dying lack of nursing care. Very vital, very important for patients!

  • I think there is no guarantee in funds commitment to such abrupt statements, until done otherwise. In fact, aid posts here have not dwindled but all shutdown long time ago, because of funding. If state facilities are hard to manage, church-run facilities operate better in rural settings, we just need the DSIP and PSIP monies diverted to them. In stead, the devolution of powers through the DDA Act gave open checkbooks to District Administrators and District treasurers at their disposal. Some of them leave no traces… while rural clinics remain closed.

  • Well said the commentaries

    What the current situation at the health centres and the numerous difficulties faced by mothers and sisters in rural and urban centres are explained well by Dr Mola, the PM should read this.

    The idea on compulsory push for mothers to deliver babies at a health centre will not improve on maternal and child mortality rate of this country. Many socio economic policies initiated by current government already proved to fail (free health and education policy, compulsory basic education for school age kids, SME) so this proposed policy to push all pregnant mothers to a health centre for delivery is an unrealistic option.

    The real problem is meeting the needs of the health centres to competently deliver services and improve the health indicators envisioned by PM. Basically the PM should not think hard but categorise them into three basic approaches; (1) build health centres with adequate space & facilities, (2) upscale skills and manpower of staffing and (3) procure & supply appropriate equipment/drugs for health centre, then the PM’s dream of improving PNG maternal and child health will bear some fruits.

    The answer to the problem is not with Cuban doctors, do not take the fight too far into the Caribbean but back to Waigani, Treasury, Finance and to specialist like Dr Mola for advice.

  • Glen Mola has provided an excellent outline the complex challenges ahead if PNG is to reduce maternal mortality. What he accurately describes is the weaknesses in the maternity health system, that is about to become considerably weaker with a planned 60% reduction in revenue (USD equivalent) for the foreseeable future. There has been only a muted response to the magnitude of this humanitarian crisis so far, as outside donors and observers move to protect their slice of the health development program, the wider health system crisis that is under way is being largely ignored. No health system can sustain revenue reductions of this magnitude, and hope to maintain service levels or make improvements . As PNG is home to 77% of the Pacific Islands’ population, the chances of this sub-region reaching the health SDG by 2030 is already slipping away.

    The PNG PM has provided very necessary focus to this ongoing challenge. However, solutions that have been developed within the PNG context are more likely to be successful than those form outside. In this respect, Glen draws attention to the Milne Bay experience.

    Milne Bay is worth a closer look.

    Milne Bay is a high performing and consistently improving province in the PNG health system. It also has geographical and demographic features that are typical of the health service challenges in Pacific Island health services – with both remote hill country, a town (population 12,000), and populated, widely dispersed islands. The total population is 280,000.

    The reasons behind Milne Bay’s success include a combination of factors. Consistent with the complex nature of health systems, it is unlikely that any one feature is “the magic bullet” – more likely the combination of factors and their positive synergistic interactions.

    The wider context includes Milne Bay provincial administration as one of four consistently higher performing provinces in PNG. ( NEFC 2013) It achieves this despite relatively low levels of provincial revenue:
    It has a level of service closer to the cost of service estimate for health than many other provinces, including those with more resources.

    Milne Bay has had consistent, skilled health leadership over an extended period, demonstrating exemplary management of the key relationships, and through this facilitated resources and support flowing to the province from both donor and government sources. Interviews I undertook in 2015 revealed skilled relationship management of important provincial and political stakeholders.

    Within the Milne Bay health sector, there was impressive clarity of vision across the different health cadres. The 2015 interviews revealed a shared culture and understanding of the role of the health system’s Provincial Health Authority across all layers of the organisation. This included a high level of commitment from senior clinicians to the overall health objectives of the province. To quote a senior hospital specialist: “I am responsible right down to the aid post”.

    This leadership has also enabled / encouraged partnerships with donors and other actors. This includes the churches, independent NGOs (Dr Kirby being the prime example ), major donor initiatives (such as Rural Primary Health Service Delivery Program where implementation barriers have been overcome that slowed progress in other provinces). It has also received considerable financial and technical assistance from DFAT and its predecessor AusAID over an extended period.

    The 2015 interviews with the CEO also revealed his focus on managing the relationship with the NDoH, especially around issues of funding and funding flow. The latter plagues the PNG government system, exacerbating the shrotage of the former. An analysis of the level of hospital spending reflects the success of this focus. Alotau hospital grew its share of national and provincial hospital funding from 4.79% to 5.9% over a three-year period 2011-2013, a greater shift than in any other province.

    An important indicator of health system access is the number of outpatient visits occurring. An analysis of PNG’s 89 districts shows that 3 Milne Bay districts (Alotau, Esa’ala, Kirwini Goodenough) were in the top 6, with the greatest improvement in this indicator.

    There have been major barriers to family planning services in PNG. Recently, Milne Bay has piloted the use of contraceptive implants (particularly in Kiriwina Goodenough) and between 2013 and 2015 this has doubled the use of contraception in this district. This far exceeds the improvement in this indicator nationally, and will have profound downstream impacts such as reduced maternal and child mortality, and significant economic and wider social benefits in the medium term.

    Improving supervised deliveries, the PMs current focus, is a complex task in health system strengthening and Milne Bay has outperformed the rest of PNG’s provinces in this regard, with a 23% improvement over the period 2011 to 2014. As noted earlier, this excellent result is due to a combination of core government, NGO, MP, staff and community engagement. The latter being engaged in the demand side with the provision of ‘baby bundles’ supplied by an NGO.

    A high level of success is also seen in other areas (see 2015 PNG SPAR report). Milne Bay leads the country in outreach clinics for children, immunisation levels (measles and DPT3) and TB treatment success levels. It is also in the top three provinces for supervisory visits, and functioning telecommunications. It has low levels of diarrhoeal illness and low rates of pneumonia deaths of children in hospitals.

    Despite the successes outlined, Milne Bay also faces very big challenges. It has the nation’s highest levels of malnutrition in children (40% and likely to have deteriorated further with last year’s drought conditions), malaria (despite improvements), and it ranks 16th among provinces in availability of supplies in clinics.

    In conclusion the evidence above suggests there is considerable value in describing and sharing the experience of Milne Bay province more widely across PNG. The evidence suggests dynamic improvement and leading performance within the health sector, enabled by strong performance in the province. This has enabled the effective mobilisation and utilisation of resources and adoption of innovations. It is likely these positive management and leadership conditions have been operating over an extended period (10years+).
    There are also unmet challenges in conditions such as malnutrition, malaria, and NCDs where effective an intersectoral approach is paramount. The potential to address these deeper issues is readily visible in the existing relationships, and will provide valuable lessons on how intersectoral approaches can be tackled in PNG more generally.

    The Milne Bay experience has much to offer at the political, governance, management, clinical, and program levels.

    However, even Milne Bay will struggle to survive the level of reduced health expenditure that is occurring.

    Don Matheson.
    (reference material related to this post available from

  • Fabulous analysis Camilla and yes, one positive is that this is being discussed by high-level leaders.
    This policy will certainly be doomed to failure unless there are huge increase in capacity building and funding for education of more human resources for health -. for PNG midwives, nurses and doctors in line with population increases, there needs to be a dedicated workforce for reproductive health.
    Funding for infrastructure, funding for heath supplies and drugs, funding for good health management in the LLGs, districts and provinces, funding for maternity waiting homes and emergency transport to referral hospitals when needed.
    It is so much more complex than making facility birth compulsory and bringing in Cuban doctors.
    How will women travel to health services? Who will provide them with skilled birth attendance when significant health system capacity building is needed?
    Why would women engage with the health system when there is not even gloves available in the national referral centre in Port Moresby?
    Thanks again for your blog on this vital development issue.

  • Hi Camilla,
    I agree, particularly, the idea of giving PNG women the ‘opportunity to express their views and experiences around childbirth’. There are several processes to follow, including consultations with women and various stakeholders before introducing any legislation. PNG continues to have persistent woeful maternal health outcomes despite huge investments by development partners in the health sector, growth over a decade before the current economic conditions, several major policy and political reforms and changes, and numerous studies on health service delivery. However, there has been limited attention given to significance of user (demand)-side factors that shape health beliefs of users, which ultimately influences health care-seeking behaviours (HSB), health service utilisation and outcomes.

    Maternal health outcomes and indicators cannot be improved if the focus of improving this area is provider (supply)-driven. The use of legislation to make it mandatory for all women in PNG to deliver babies in formal, modern biomedical health facilities, is like using a whip to force someone against their will. Without addressing barriers at health facility levels (supply-side) and not having an understanding of the constraints faced by mothers in accessing and utilising health services (demand-side), and introducing legislation to influence women to deliver at health facilities is insensitive. It does more harm than good. This is symbolic violence where the State uses legislation to legitimate its power and imposition over other groups and classes. There are deeper socio-cultural, economic, and physical factors to address on both supply-side and demand-side of health services. Any intervention to address maternal health outcomes need to understand these factors for the interventions to be practical and useful. Interventions aimed at improving maternal health service delivery are doomed to fail without considering barriers on the demand-side that prevent health service utilisation.

    • Very well said, Ellen, I couldn’t agree more. There are certainly supply-side issues to take care of, but demand also needs to be looked at very closely. And, any proposed policy needs to recognise that those demand-side constraints will likely vary considerably, particularly in a country as diverse (culturally, economically, geographically) as PNG.

    • It is great idea and very commendable that the PM has decided that he is going to do something about the horrible rate of maternal and newborn death in PNG by encouraging (making it compulsory) women to have professionally supervised births in health facilities.
      However, this plan needs to be very carefully thought through, as it could end up as an enormous waste of money without very much good effect.
      There are about 260,000 births every year in PNG. Of these about 104,000 women come to health facilities to have a professionally supervised birth, but this also means that 156,000 women deliver their babies in their villages without professional assistance. This is a huge number, and while it is absolutely critical that women must have a supervised birth if we are to reduce risk of maternal and newborn death, the parts of the health system that the PM says he is planning to bring the women to for supervised birth (ie urban areas), hospitals are already overstretched and many are unable to cope with the numbers.
      For example at PMGH 15,000 women came to the public labor ward to deliver in 2015, – ie. over 40 per day, but we actually only have 24 delivery beds in the labor ward. This means that many women cannot be allocated a bed to have their labors supervised when they arrive, and some end up delivering on the floor or the in reception area. These circumstances, of course do not produce optimal outcomes, and if extra women are to be brought in from rural areas of Central Province (and the Eastern end of Gulf) to deliver at PMGH, bed space will be quite insufficient and staff will not be able to cope.
      In fact all over PNG provincial hospitals in the urban areas are being inundated already by women from rural areas coming to have a supervised birth in town. This is because rural health facilities are either not accessible to village women, or women do not trust them (because they are run down, ill-equipped and poorly staffed), or lack of supportive management in the health system over many years means that many rural facilities do not remain open 24/7 and thereby give women and their families confidence that by turning up at a rural facility in the middle of the night or over the week-end, that there will be a skilled person available to receive the woman in labor and look after during her birthing process. The figures paint a stark picture; 40% of women in PNG currently have a supervised birth, but rural facilities typically only supervise less than 10% of the births occurring in their catchment areas, – this is because women are either staying at home or are bypassing their local rural facility and travelling all the way to town for their supervised birth.
      I personally don’t think that it is sensible to ‘fund women to come to urban areas to have their babies’ as suggested by the PM; the hospitals in towns are already overburdened and having difficulty coping with the women who are already coming. What needs to be done is
      – Building capacity in the rural sector; upgrade birthing units in rural health facilities and equipping them properly, but also
      – Upskilling CHWs and nurses to provide quality birthing care for women in rural health facilities, and
      – Training more doctors, nurses, midwives and CHWs to provide quality health care for the expanded rural health service – so that the majority of women can access life saving supervised birth close to their villages.
      All this will require massive capital expenditure in health manpower training and upgrading, and also building, upgrading and equipping health facilities to cope with an extra 150,000+ women per year who will need birthing care in the PM’s plan.
      It can be done. In fact a pilot incentive program in Esa Ala district of Milne Bay province has been able to increase the supervised births from less than 40% to more than 70% in just 3 years by incentivizing both supply side (facilities and staffing) and the demand side (helping women to want to come to rural facilities by direct incentives to them and their families, to do so.
      Bringing women to urban areas to give birth will not only overburden the already overburdened city hospitals, and possibly thereby cause deterioration in birthing outcomes, but it will also lead to massive rorting of transport and accommodation monies. Calculation of the due date for the birth is at best plus or minus 2 weeks. Therefore women would have to receive expert antenatal care in their home areas in order to work out their due dates (not available at the present time), and then be transported to town at least 2-3 weeks ahead of the due date to await onset of labor. Most women would therefore end up spending 3-4 weeks in town. If we calculated that they could be accommodated for about K100-150 per week in purpose built hostels (yet to be constructed) or in the homes of wantoks with accommodation support from government, this would amount to about K100x4x150,000women per year = K60-90m per year in accommodation costs, in addition to the transport costs and the medical costs of the urban maternity care. Total costs could be several hundreds of millions of kina; money that would be much better spent on building capacity for maternity care closer to their village homes and incentivizing their use of rural services as has been done in Milne Bay province.
      The PM also needs to know that Cuban generalist doctors do not have O&G training or skills in maternity care. This is not how the Cuban medical training systems work. And, specialist obstetrician Cuban doctors are very narrowly trained to operate in advanced hospital environments. Therefore I don’t think
      “Our aim is to bring them (PNG women) into towns and cities and hospitals where there is adequate care that we are going to build up because of the Cubans and the sup¬port of the doctors that we are going to roll out,”
      Is going to be a viable strategy to achieve what the PM has in mind. Mr O’Neill needs to speak more with his local maternity care experts to help strategize his very laudable goal of supervised birth for all to reduce maternal and infant deaths in PNG. But above all, he needs to find many hundreds of millions of Kina to do it. The money is there in the DSIP and PSIP accounts. Let’s use the money for good effect rather than wasting it on misdirected strategies and overseas doctors.

      • Professor Mola, many thanks for taking the time to share your detailed knowledge on this topic. It’s exceptionally valuable. I was particularly interested to learn that many women are already seeking out care at provincial hospitals, indicating that it’s not demand for supervised births that is lacking but rather the supply of quality services close to where women live. Thanks also for highlighting the pilot project in Milne Bay (has that been published or more information available? I would love to read about it in more detail).

      • Thank you Professor Mola, your insights should be one of consultation by the PM. Many of your colleagues, knowledgeable in the issue seems to be left in the dark as well. PM may have a great vision but it needs to be founded in reality.

  • Hi Camilla,

    Addressing demand side issues, such as compensating for the cost of travelling to a health clinic, is a good idea. Making it compulsory is a recipe for disaster, although one would need to seriously question any enforcement of the policy. An important question to be answered, which also applies to basic education, is what are the main demand side constraints. As with schooling, issues around cost (fees, travel costs, etc.) is important but not necessarily the main demand side constraint. I would argue, as you also mention, that service quality concerns are probably just as important. The Cuban doctors idea is great, but unless they bring their own water, electricity, drugs, equipment, and support staff then quality is not likely to improve. An RCT is needed to test how quality improvements, incentive payments to women for supervised births, and a combination of both treatments, impact on levels of supervised births and maternal mortality. Of course, this evidence is needed before important policy decisions are made.

  • Paraphrasing the IEG World Bank re: cash incentives increase facility deliveries but they also increased the fertility rates and without investment in health service strengthening, they did not alter the maternal mortality rates.

    My notes – not a quote ….Increasing utilisation of services alone does not necessarily lead to better health outcomes within the services. Janani Suraksha Yojana (JSY), a government initiative in India, gave cash to women who delivered in a hospital or health facility. The initiative also gave health workers incentives with a cash transfer for every delivery they attended. This combined effort led to a highly significant increase in both facility births and skilled birth attendance in general. Note JSY found a significant increase in SBA but also finds that while the cash incentives do increase facility deliveries, they also increased fertility rates and had no discernible effects on neonatal or maternal mortality. Note other successful interventions of this type included health system strengthening (e.g Progresa and CSR).

    • Hi Neil,
      Thanks for your useful notes. I’ve also read some critiques elsewhere of the JSY initiative, but didn’t want to get too bogged down into the details in this post (not least because India and PNG are such vastly different contexts, and I think there can be real dangers in applying a cut-and-paste approach to interventions). Clearly any approach has to be comprehensive and evidence-based, taking these kinds of lessons into account but also being aware of the potential for unintended consequences.
      Thanks again,

Leave a Comment