Ashar Malik

Pakistan is considered a classic case of growth without development as its performance on social development including health is not very encouraging. The private sector dominates in healthcare finance: private households contribute 60% of total health expenditure, and in healthcare delivery more than 50% of the population seeks care from private providers. The government of Pakistan (GoP) has adopted a series of strategic attempts to improve access to and use of public health facilities and thus reduce the financial burden of health-seeking on households. These include, for example, the district-based health infrastructure development (in the eighties), the sector-wide approach (in the nineties), and the prioritization of Maternal and Child Health (MCH) (since 2000). The latter strategy was part of the efforts to achieve the health-related Millennium Development Goals.

Despite large-scale investments in the health sector, by the end of the MDG era Pakistan had missed the targets of health related MDGs. Moreover, little was known about the relative effectiveness of most of these investments. In the field of healthcare financing, neither resource-tracking was routinely carried out, nor were the contents and format of the existing health accounts useful for health policy analysis. The evidence, before 2015, on the effectiveness of the significant programmes and reforms was limited to some project completion reports of the Pakistan government to a few studies which employed simple before and after comparison research designs.

The research aim of this thesis is to support evidence-based policymaking in Pakistan. Specifically, we will answer two research questions:

1. What were the patterns of healthcare financing in Pakistan and their relationship with health outcomes, particularly in the MCH sub-sector?
2. Are the claims of increased utilization of health services causally related to areas exposed to the health reforms/programmes?

This thesis contains two sections. Section 1 addresses question 1. First, a review of trends in healthcare financing and health outcomes in Pakistan compared with the Countdown-to-15 countries is followed by an estimation and review of the expenditure in the Reproductive, Maternal, Neonatal and Child Health (RMNCH) sub-sector and the magnitude and the determinants of the out-of-pocket health expenditure. Section 2 addressed the second question, first by reviewing the trends of maternal and child health outcomes and health services delivery in the South Asian region and followed by two programme evaluation studies of (1) contracting of primary healthcare units and (2) the Norway Pakistan Partnership Initiatives (NPPI).

In chapter 2, we find that maternal and child health expenditure increased in all countries included in the review, namely Afghanistan, Ethiopia, Malawi, Pakistan, Peru, and Tanzania. However, only a few countries managed to achieve the targets of health related MDGs, e.g., Ethiopia and Peru. In Ethiopia- a low-income country, progress on MDGs was associated with accelerated economic growth coupled with a high-level commitment and an integrated approach for investments in MCH sector, while in Peru, it was associated with investments in poverty alleviation and decentralized health policymaking. Pakistan’s low performance on MDG targets were suspected to be due to the fact that investments in the MCH sector were channelled through centralized vertical programs of the federal government, with many examples of fragmented efforts and often overlapping functions.

In chapter 3, we find a two-fold increase in expenditure on RMNCH over the period 2000-2010. Pakistan was among the countries with the highest growth in RMNCH expenditure and one of the countries that received record levels of foreign development assistance to their MCH sector. However, these efforts did not improve the utilization of MCH services by the poor. For example, births at public facilities increased by 3% points but remained pro-rich over 2000-2010. These findings lend some credibility to our conclusion that does not resource scarcity, but the quality of health bureaucracy, good governance, and socio-economic and behavioural aspects of health and health care seeking were the missing elements in these health policies.

In chapter 4, we explored the extent and the determinants of out-of-pocket health expenditure (OPHE) - the major source of financing in Pakistan. Our findings indicate that poor sanitation and unsafe drinking water contribute to higher OPHE. On the aspects of health, pregnancy-related healthcare needs at the household and distance of more than 5 kilometres to the nearest health facility were the factors which showed a strong association with higher OPHE by households in Pakistan.

From Part One of this thesis, we conclude that good governance and accountability through a decentralized and integrated approach to investment in the MCH sector demonstrated better health outcomes albeit the role of political and economic development cannot be ignored.

In chapter 5, we tracked progress on health related MDGs in the South Asian region including Afghanistan, Bangladesh, Bhutan, India, Maldives, Nepal, Pakistan, and Sri Lanka. In the region, Pakistan was among the worst performers both in terms of health outcomes and in services delivery-related indicators. Pakistan did improve the skilled (24% to 59%) and institution-based care (19% to 68%) use by mothers and their new-borns, but these gains were mostly in the private sector where the rural-poor were left-out. Countries that performed better in the MCH sector, such as Bangladesh and Nepal, focused on community (especially women) empowerment and female literacy as the primary means to improve health outcomes, while Pakistan was ranked second-lowest (only above Afghanistan) on literacy and women's improvement.

In chapter 6, we evaluated a large-scale experiment of contracting Basic Health Units (BHUs) in 73 districts (out of 113). Using a difference-in-difference approach we find no effect of this experiment on seeking care from BHUs for either unknown common illnesses or for childhood diarrhoea in treated versus control districts over 2001-2012. A literature review indicated multiple issues in the design of contracts like, for example, lack of accountability of the contracted agencies towards their public sector counterpart and the fact that the contracts were limited in scope to such factors that hamper population access to BHUs such as their location.

In chapter 7, we evaluated a large-scale MCH project i.e., the Norway Pakistan Partnership Initiatives (NPPI) that was implemented in the MCH sector in ten rural districts (out of 26 districts) of the province of Sindh. Using a similar methodology as in chapter 6, we did not find any improvement in the skilled and institutional care for childbirths in the intervention districts of NPPI. Surprisingly, we did find a significant improvement in the home-based skilled births in NPPI districts that were further exposed to either the contracting (13 percentage points) or the vouchers (7 percentage points) schemes- designed to improve institution-based care for pregnant women.

Our findings from part 2 suggest that the observed progress made in MCH was primarily associated with the general socio-economic development since the intervention areas did not show any significant difference from their respective controls as far health-seeking of the general population or pregnant women was concerned.

In general, the findings in this thesis indicate that the strategy to improve the use of government health facilities by investing in the quality of care and community awareness campaigns did not sufficiently consider the barriers on the demand and supply sides that contributed to the underutilization of government health facilities. For example, the popular perception of the size of the health facility as a proxy of better quality, or the informal payments made at public -- particularly primary -- healthcare facilities are a few of the demand-side factors that require particular attention of policymakers. On the supply side, geographical access to health facilities and an effective referral system can improve the use of district-based health systems and contain its costs.

Historic underperformance (also in the MDG era) does call for improved credibility of health investment. Since 2015, a few promising developments have taken place in the health sector such as financial risk protection plans for the poor, devolution of the health sector to the provinces, and a sector-wide approach in the health sector in a few provinces. Together with the continuation of democracy, an active judiciary, and vibrant electronic media, the findings of this thesis may contribute to better underpin efforts to improve health policymaking with an evidence base in Pakistan.