Executive Summary Rouselle Flores Lavado (ID03P001) The dissertation analyzes barriers to health care utilization in the Philippines. It starts with a review of the Philippine health sector and an analysis of patterns of utilization in the maternal and child care services. Based on these two chapters, the framework for analyzing barriers to health care utilization was laid out. The succeeding three essays empirically analyze each barrier to health care utilization. The main findings of the sectoral review highlight that the Philippines has been underperforming in its health outcome attainments in the past two decades. There were rapid achievements until the 1980s; however, improvements had tapered off in the recent years. At present, it has worse health indicators compared to other ASEAN countries. A closer look at the national health accounts shows that the share of social insurance in total health spending is even less than ten percent. Most of the expenditures are out of pocket payments, contributing to catastrophic expenditures on health for a lot of households. Very little is also spent on public health or preventive care a major sum of the budget is spent on curative care. Health care from the private sector is very expensive. Government facilities, although cheaper, are viewed as giving lower quality of care and are sparsely available. These trends suggest that major barriers to health care utilization are income, access, and perceived quality of care. The next chapter sets out to examine the validity of these trends at the household level. Using the 2003 National Demographic and Health Survey, a national survey conducted every 5 years, this chapter examines patterns of exclusion in maternal and child care by calculating concentration indices for prenatal care, iron supplementation, skilled birth attendance, delivery in a medical facility, and complete immunization for children under 5. The survey identified about 1,500 pregnant women and 3,000 children below 5. Results show inequality in maternal and child care services utilization across economic classes and across regions. Pregnant women in the richest quintile are two times more likely to have prenatal check-up than those in the poorest quintile, four times more likely to have skilled attendance during child birth, and nine times more likely to deliver in a medical facility. However, there is not much difference between the richest and poorest quintiles in accessing iron supplementation during pregnancy and full immunization for children under 5. Patterns of exclusion within each region are also
examined. For regions where highly urbanized cities are located, high utilization rates and high concentration indices are observed. The two poorest regions have low utilization rates but different patterns of inequality is discerned one has massive deprivation for all income levels in health care access leading to low levels of coverage while the other has only the richest quintile accessing health care. While the uptake of maternal and child care services has been reasonably satisfactory, the focus on the poor remains low as evidenced by vast differences in patterns of utilization across economic classes and across regions. It is possible that as the national average for health services utilization shows improvements, it is likely that only those in the richest quintile are showing improvement while the poor remain to be marginalized. Given these findings, the succeeding chapters of the dissertation examine each barrier to health care utilization. Chapter 4 looks at the mechanisms as to which decisions to seek health care are made at the household level. It analyzes the determinants of maternal and child care utilization by examining the role of intra- and inter-household dynamics in the decision to seek care. In most health-education literature, the analysis done at the household level assumes that there is a single decision maker. In reality, however, not all members of the household share the same preferences. This underlines the importance of departing from the unitary model of household in evaluating health decisions. Regression analyses utilize the six outcome indicators as dependent variables. Prenatal care takes a value of 1 when the pregnant woman had at least 3 antenatal care check-ups, had her first check-up during the first trimester, and took iron supplements. Skilled birth attendance is equal to 1 when the medical attendant during delivery was a doctor, nurse, or midwife. A woman is considered to have given birth in a medical facility if she gave birth in a hospital, clinic, or public health center. The immunization indicator for child care takes a value of 1 when the child had 3 dosages of DPT and polio, and one dose each of measles and BCG. The impact of other household members on the decision to utilize care is proxied by their education level while the impact of the neighbors is proxied by the average of their education level. Decisions on seeking maternal and child care appear to be not dependent on a woman s decision alone. Intra-household dynamics matter in the decision to seek antenatal care,
delivery care, and child immunization. For women living in extended families, education level of the household head (woman s father or father-in-law), wife of the household head (woman s mother or mother-in-law), and other educated members of the household appears to influence a woman s decision in seeking maternal and child care. The education level of the neighborhood also contributes to the woman s health knowledge, implying the role of social networks in influencing decisions. This kind of analysis can provide signals to policy makers whether there is a need for interventions to be targeted to specific groups or individuals within the household. Chapter 5 examines financial barriers to seeking health care in the last 20 years. Using the Family Income and Expenditure Survey, a micro dataset at the national level, this chapter constructs pseudo panels to examine the trends in medical expenditures on different groups of population. It is found that certain groups of populations are more vulnerable, such as female-headed households, households with low educational attainment, and those living in rural areas. Medical expenditure also appears to be greatly influenced by economic fluctuations. When times are hard (particularly during the Asian crisis), households reduced medical expenditure. Health care in the Philippines has a skewed tendency for curative personal care but with exorbitant out of pocket payments, and illnesses become catastrophic incidences for the poor. This finding necessitate the need to shift burden of financing personal curative care to social risk pools such as the Philhealth so that health services can be accessible even to the poor. One of the recent programs of the government, the Philhealth Indigent Program targets the poorest 25 percent of the population where insurance premium is subsidized by the local and national government. The main impediments to this program are the inability of the poorer local government units to pay for their indigents and the frame of thought of most Filipinos that paying premium is a waste of money. This highlights the need for health education, particularly for main decision makers as discussed in Chapter 4. The local government also needs to be aware of the importance of allocating expenditures for social services. The last chapter examines the efficiency of the provision of health services by local governments. The Philippine government budget was around 17 percent of GDP in 2004. A huge part of this expenditure, however, goes to debt servicing. At present, total public debt is 109.8 percent of GDP. With what is left, only one percent is spent for health and three percent for education. Given such budget constraint, it is important to examine the efficiency of spending on social services since small changes can have a
major impact in achieving the Millennium Development Goals. The relationship between spending and outcomes is difficult to ascertain. For instance, experience of developed countries shows that health spending is poorly correlated with health outcomes. The United States has one of the highest health expenditure in the world yet its health outcome is below most OECD countries. For developing countries, however, such correlation between health outcomes and expenditure is higher. Until a country reaches a level of health care provision that is accessible for everyone, such correlation will persist. When there are major inequalities in income, the poorer segment of the population only has the government facilities as the last resort. And when the government cuts its expenditure on health, the poor are gravely affected. In 2005, Herrera and Pang published a paper on efficiency of public spending in developing countries. This chapter follows the steps undertaken in their paper and goes down to the level of provinces and health care centers to serve as guide to providing incentives to improve outcomes. The first level of analysis looks at provinces in utilizing their social expenditure budget and medical resources. The second level goes down the frontline by comparing the efficiencies of rural health centers. Efficiency is defined as the deviation from the frontier which represents the maximum output attainable from each input level. This efficiency frontier is estimated using the Data Envelopment Analysis (DEA) and Free Disposal Hull (FDH). For inputs, the first study uses expenditures for social services. Outputs are life expectancy, primary and secondary enrollment, and literacy rate. The results show that the provinces that receive higher fiscal grants per capita are among the least efficient. On the other hand, provinces with higher real per capita income are more efficient. The implications of this study are of particular policy importance for the allocation and rationalization of budget among provinces. When services are publicly provided, performance measurement becomes an inevitable management tool. The second study considers only the technical efficiency of provinces. While many studies in the past analyzed the efficiency of hospitals using various performance indicators, studies on efficiency of primary health care provision is not so common. In a report by the Commission on Macroeconomics and Health, very few studies look at primary health care efficiencies although outcomes indicators such are prenatal care access are readily available in most countries.
Four programs of the Department of Health are evaluated: Maternal Care Program, Expanded Program on Immunization, Control of Diarrheal Diseases Program, and Vitamin-A Supplementation Program. Inputs are quantifiable indicators such as number of doctors, nurses, midwives, and health stations while outputs are immunization rate, pre-natal consultancy rate, tuberculosis cure rate and Vitamin A usage. Results show that 17 out of 75 provinces are efficient in providing Maternal Care Programs given their medical staff and facilities. Input efficiency score is 70 percent while output efficiency score is 79 percent. Fifteen provinces were efficient in the immunization program. On average, input usage can be reduced by 30 percent while immunization outcomes can be increased further by 20 percent. Provinces are least efficient in providing the Control of Diarrheal Diseases Program with only ten efficient provinces. The Vitamin-A Supplementation Program is most efficient, with input and output efficiency scores of 75 and 82, respectively. The last study of this chapter looks at the efficiency of public health units which serve as a backbone of the Philippine health system, making health care universally accessible to all individuals and families in their respective communities. A subsample of 30 rural health units and city health offices is examined using a dataset from a baseline survey conducted by the Department of Health. The study finds that expenditure efficiency score ranges from 31 to 51 percent, depending on the program. This implies that there is much room for increasing outcomes given the current budget. Output-oriented technical efficiency score ranges from 76 to 91 percent, also implying that with their given level of medical staff, health units can increase their outcome achievements by 9 to 24 percent. To illustrate the possible use of benchmarking exercises, targets for input reduction and outcome increases are also outlined. DEA is a methodology that lies between crude ratio analysis and complex statistical techniques. The methodology allows for categorization of provinces and health units according to their efficiency. Since there are no data which allow for the estimation of quality and suitability of care, the results might not be complete. A caveat that needs to be underlined is that the results are indicative, rather than absolute measures of efficiency. Despite the limitations, DEA still offers a useful means of assessing performance in the health sector. Knowledge of who the best performers are is critical to understanding policy options that may allow for improvements in outcomes without additional resource outlays.