The provision of social health insurance has been an increasingly
popular mechanism for addressing financial barriers to health care in
developing countries. In the Philippines, the social health insurance
program known as PhilHealth has been expanding its breadth of
coverage since its promulgation in 1995. This study looks at the
regional- and provincial-level coverage of two different PhilHealth
programs: the Individually Paying Program (voluntary) and the
Employed Program (government and private), and identifies the
possible reasons for the variation between provinces. Coverage levels
for the Individually Paying Program were found to be considerably
below the government’s full coverage target. The regional average was
found to be at 57 percent while provincial average was at 53 percent.
Variation was massive, ranging from 2.4 percent to 166 percent. For
the voluntary program, provinces with higher bed-population and
health professionals-population ratios were more likely to have higher
coverage rates. The number of private hospitals in the province was
also found to be a positive indicator for coverage levels. That is,
the more the number of private hospitals, the higher the PhilHealth
coverage level in a province. Curiously, incomes did not appear to
be a factor in the level of insurance coverage in the province. For the
Employed Program, both the private and government sectors have not
achieved full coverage but the figures observed are promising. The
private sector Employed Program’s regional average is at 71 percent while the provincial average is at 75 percent. For the government
Employed Program, the regional average lies at 74 percent while the
provincial average is at 80 percent. For the private sector, the size of
the establishment (in terms of number of employees) proves to be a
significant factor. That is, as more employees are hired by mediumsized
establishments, the higher the province is likely to experience
PhilHealth undercoverage. Certain sectors are also found to be more
prone to undercoverage. These findings are possible propositions on
how targeting should be implemented. For the government sector, no
clear pattern was found based on the model presented.