Methods

We reviewed available literature (Medline, local journals, newsletters, and Ministry of Health reports) on measles vaccination coverage and its impact on measles morbidity and mortality in Ghana. The factors contributing to performance of the immunization program were derived from the same sources and from personal observations.

Results

Vaccination coverage.  Based on routine reports, measles vaccination coverage among infants increased from 24% in 1980 to 84% in 2000 (figure 1). In 1985, the first national measles mass vaccination campaign was conducted, and coverage of 72% was achieved compared with 12% through routine services in 1984. Between 1987 and 1993 the vaccination coverage remained low. During the period of health sector reform in Ghana (1997–2001), measles vaccination coverage increased steadily from 53% in 1996 to 84% in 2000. In 2000, of 110 districts, 60% achieved coverage ⩾80%. In rural northern Ghana between 1989 and 1991, the NDSS found measles vaccination coverage of 48% among 25,443 children aged 0–59 months [9]. More recent data on vaccination coverage were not easily available.

Between 1993 and 1998, the GDHS documented an increase of up to 11 percentage points at the national level for vaccination coverage by age 12 months for each antigen (figure 2) [7]. The proportion of children who had never been vaccinated fell from 18% in 1993 to 9% in 1998. In general, the GDHS coverage was higher than recorded in routine Ministry of Health reports.

Cluster surveys.  In January 1999 in the Upper East Region, mothers and caretakers of 215 children aged 12–23 months were interviewed [10]. Of these, 187 (87%) had vaccination cards (table 1). Measles coverage was 64% by card and 70% by card plus history. Of the children, 40% were fully vaccinated by age 12 months. About 73% of vaccinations were done at a hospital or health center; the remaining 27% were obtained at outreach vaccination posts. In the Cape Coast District in 2000, a survey of 480 mothers or caretakers found a measles vaccination coverage of 76% by card and 89% by card plus history (table 1) [11]. In both settings, survey results were generally higher than shown by the administrative method.

Disease burden.  Reported cases of measles in Ghana decreased by 72% from 82,684 in 1980 to 23,068 in 2000 (figure 1). A large outbreak occurred in 1995 involving 43,177 reported cases and 85 deaths (case‐fatality ratio, 0.2%). During prior epidemics, the case‐fatality ratio was ⩾1% among children [12]. During the first 5 years of health sector reform (1997–2001), reported cases of measles declined by 66% from 36,968 in 1997 to 12,498 in 2001. Measles outbreaks that occurred annually between 1975 and 1978 now occur every 3 to 4 years (figure 1). The last two outbreaks occurred in 1997 and 2000.

The overall estimated measles incidence rate in the Kasena‐Nankana District in rural northern Ghana during 1989–1991 was 24.3 per 1000 child‐years; the acute measles case‐fatality ratio was 15.7% [9]. Incidence rates and case fatality were higher in families with low paternal education, in the dry season, and in unvaccinated children; the case‐fatality ratio was higher in malnourished children. Although there was no sex difference in incidence, acute case fatality was somewhat higher in girls than boys (adjusted odds ratio, 1.3; 95% confidence interval, 0.9–2.1). The incidence was lower in children supplemented with vitamin A than in children without supplementation, but this was not statistically significant.

Based on routine surveillance data for the Central Region, there was a slight shift in the age distribution of measles cases from younger to older children. The proportion of cases aged ⩾48 months increased from 30% in 1990 to 33% in 2000 (figure 3). The proportion of cases in the age group 12–23 months decreased from 20% in 1990 to 16% in 2000. By retrospective review of measles cases attending three district hospitals in the Central Region during 1996–2000, the age distributions among children aged <1, 1–4, and ⩾5 years were 17%, 49%, and 34%, respectively (table 2) [13]. In the Upper East Region, 51% of cases in 2000 were >5 years old [14].

From January to March 2000, an outbreak of measles occurred in Accra. Among 44 cases attending three health centers in the metropolis, the peak incidence was among children aged 6–12 years and 52% of cases were ⩾5 years old (table 2) [15]. The vaccination status could only be ascertained for 16 cases (36%): 12 were previously immunized against measles and 4 were not.

In the prevaccine era, the peak age of 4317 children with measles admitted to the Korle‐Bu Teaching Hospital in Accra during 1973–1982 was 7–12 months [2]. Only 3% of cases were >4 years old. Children hospitalized with measles constituted 8.8% of all pediatric medical admissions over this 10‐year‐period, and the case‐fatality ratio over the period was 17%. In a follow‐up study at the same hospital during 1985–1993, measles accounted for 1.7% of pediatric admissions and the case‐fatality ratio was 7.9% [16]. Most of the children admitted with measles were unvaccinated and resided in areas of high population density in Accra and in periurban slums. Compared with the earlier decade, the number of children aged 3–8 months as well as school‐age children ⩾60 months increased.

Discussion

Since 1995, there has been a steady increase in vaccination coverage in Ghana (measles and other vaccines) measured both by administrative methods and by surveys, resulting in a substantial reduction in the burden of disease due to measles. However, over 20,000 cases of measles still occurred in 2000 and the BCG‐measles dropout rate was high (25% in the 1998 GDHS), indicating the potential for further improvement in coverage.

Two key events contributed to increases in routine vaccination coverage. First, in 1995, a medium‐term health strategy was developed based on a national development agenda, Vision 2020. The main objective of this strategy was to maximize the health and productive life of people living in Ghana. Second, in 1996, health sector reform began to be implemented in Ghana based on a working document called the Five Year Program of Work, 1997–2001. The objectives of this program were to increase funding, improve access to care, increase the quality and efficiency of health services, and to develop public‐private partnerships. A major achievement of these reforms has been increased district autonomy with increased financial and managerial capacity (table 3). The proportion of the total recurrent budget (including donor funds) allocated to districts increased from 20% in 1996 to 42% in 1999 at the expense of the central level. This meant that more funds were available to districts for implementation of the EPI. In addition, the overall percentage of the government budget spent on health increased from 7% in 1996 to 8.4% in 1997 and 8.6% in 1998 [17].

Immunization services access improved after the number of outreach posts was increased and mothers visited the sites more frequently. The number of outreach posts per health facility in Ghana increased from 7 in 1997 to 8.3 in 1999 [18]. With a view to improving the quality of service, various studies have examined the factors associated with immunization coverage from both the supply and demand sides [19, 20]. Analyses of the 1988 GDHS data indicated the following factors were significantly associated with vaccination coverage: the child’s age, place of residence, maternal education, father’s occupation, region, and type of prenatal care [20]. The improved motivation of staff, availability of transport, prompt payment of travel claims, community involvement, home visits, intersectoral collaboration, and reduced waiting times all positively affect vaccination coverage [19]. Providing district health teams with regular feedback on the performance of their immunization services has led to a healthy competition between operational units, and weak districts have embarked on mini‐mass campaigns to increase their coverage.

Rising vaccination coverage has led to an overall decline in measles cases, widening of the interepidemic interval, and a shift in the age distribution of cases to older ages. Recently, one‐third to one‐half of cases were ⩾5 years old compared with <5% in the prevaccine era. However, every 3 or 4 years widespread measles outbreaks still occur with substantial morbidity, although mortality has been lower in recent years.

The demographic effects of improved vaccination in Ghana have been modeled by using different mortality reduction scenarios [20]. These scenarios reflect a mortality risk from measles in a vaccinated child that was worse than, similar to, or better than that in an unvaccinated child. Based on the best case scenario (i.e., the immunized child is protected from death due to measles and other associated conditions such as malnutrition) and data on existing child mortality rates, cause‐specific proportional mortality, measles vaccination coverage of 100% by 2000, and projections about growth rates from 1975 to 2025, it was estimated that a measles vaccination campaign in Ghana could save 459,256 lives from 1985 to 2025. More than one‐fifth of the lives saved would be children living in the less‐developed northern areas of Ghana.

Based on these analyses and recommendations from the WHO Regional Office for Africa, Ghana developed a 5‐year strategic plan to reduce measles mortality to near zero [3]. Strategies include increasing first‐dose coverage at age 9 months to 90%; a mass vaccination or catch‐up campaign targeting all children aged 9 months to 14 years, regardless of immunization status, in one region in December 2001 and in the nine remaining regions by December 2002; introduction of case‐based measles surveillance with laboratory confirmation; improved case management; and integration of vitamin A into the routine EPI schedule. The catch‐up campaign would rapidly increase population immunity by providing a second opportunity for immunization to children who missed their first dose and would also protect children who did not respond to the first dose.

Future challenges for the EPI in Ghana include achieving and maintaining the high routine measles coverage in all districts, implementing a quality catch‐up campaign in 2002, using funds provided by the Global Alliance for Vaccines and Immunization to improve immunization services in general (e.g., staff training, education materials, technical guidelines, data systems, and vaccine management), strengthening surveillance, finalizing the national EPI policy, and improving the quality of immunization services and community involvement in the delivery of child care services.