Bulletin of the World Health Organization

Birth registration and access to health care: an assessment of Ghana’s campaign success

Sonja Fagernäs a & Joyce Odame b

a. University of Sussex, Department of Economics, Jubilee Building, Brighton BN1 9SL, England.
b. International Needs Ghana, Accra, Ghana.

Correspondence to Sonja Fagernäs (e-mail: s.a.e.fagernas@sussex.ac.uk)

(Submitted: 14 August 2012 – Revised version received: 28 February 2013 – Accepted: 07 March 2013 – Published online: 25 April 2013.)

Bulletin of the World Health Organization 2013;91:459-464. doi: http://dx.doi.org/10.2471/BLT.12.111351


In many developing countries today, the births of a substantial share of children go unregistered. According to survey-based estimates, between 2005 and 2008, the share of children younger than 5 years whose births were registered was 7%, 41%, 60% and 88% in Ethiopia, India, Kenya and Viet Nam, respectively.1

Registration of vital events is essential for accurately calculating birth and death rates and for assessing the level of infant mortality. It has other advantages. For example, when linked with medical records, birth registration systems can alert health-care providers to the presence of children needing vaccination.2 Accurate information on births and deaths has been stressed as important for tracking progress towards the health-related Millennium Development Goals.3,4 The importance of birth registration has also been emphasized from a child rights perspective.2,5

For slightly over a decade, children’s organizations, in particular the United Nations Children’s Fund (UNICEF) and Plan International, have been involved in campaigns promoting the registration of births in developing countries. In the context of sub-Saharan Africa, national action plans for registration were developed in 24 countries in central and western Africa in 2004.6 This article reports on the experience of Ghana in raising birth registration rates from 2004 to 2008. Survey data, namely from the Demographic and Health Surveys (DHS)7,8 and the Multiple Indicator Cluster Surveys (MICS),9 are combined with observations from the field.

According to DHS data, registration rates for children younger than 5 years in Ghana increased from 44 to 71% between 2003 and 2008.10 In 30 other sub-Saharan African countries with survey data for a similar period, progress in the registration of children younger than 5 years was slow. In these countries, the average registration rate was 53% in 1999–200311 and 49% in 2004–2010,1 with only a few countries making notable progress. Therefore, Ghana stands out as a success story. This article discusses the different approaches taken to increase registration rates and focuses on the role played by the health system.

Birth registration and campaign strategies

In Ghana, Birth registration is compulsory under the Registration of Births and Deaths Act (1965). Ghana has 10 administrative regions and each of the country’s 170 registration districts has at least one registry office. However, the absence of registration offices in rural areas and a shortage of registration staff have hampered registration.12 The registration of births that occur in health facilities begins with the issuance of a medical certificate or a health card.13 Formally, parents are required to present the health card when they visit a registry to register a birth. Birth registration offices are often located within the premises or in the proximity of public health facilities, although not all health facilities have a registration office. There has also been an expectation that births take place in health facilities, but according to DHS data, only approximately half do.

Registering a child generally involves both direct costs (fees) and indirect costs (time off from work, travel expenses). The indirect costs in particular affect poorer areas disproportionately. According to the 2006 MICS, the most common reasons for not registering a child were the high cost of registration (31.9%), distance to registration locations (21%) and a lack of awareness that children should be registered (20%). Birth registration campaign activity in Ghana has focused on such factors.

To incentivize people to register a child, beginning in mid-2003 the legal period for free registration of infants was extended from 21 days to 1 year.12,13 Late registration carries a fee (equivalent to about 1.1 United States dollars).14

In 2004–2005 other campaign activities began, including intensive public education. The first annual Birth Registration Day was held in September 2004 and 10 000 children were registered across the country. Since 2004, the Births and Deaths Registry has participated in annual child health promotion weeks, organized by the Ghana Health Service in May and November of each year. Community health workers were trained to register births.12,13 These workers offer services in community health clinics and also on a mobile basis.10 Mobile community registration volunteers were introduced to register births, especially in remote areas.12,13,15 Community population registers, which in the long term are considered key to raising registration coverage and reducing the hidden costs of registration, were piloted in 21 remote communities in four regions.12

Table 1 shows the trend in the rate of birth registration among children younger than 5 years in Ghana based on data from the DHS for 2003 and 2008 and from the MICS for 2006. These surveys show different registration rates for children born in the same year perhaps because of age differences and overlaps. Differences in the precise questions posed may also explain some of the differences. Nevertheless, it is clear that birth registration rates increased significantly for children born in, or after 2003–2004, in tandem with the intensification of the campaign activities. The figures also suggest that campaign activities became more effective from 2006 to 2008 and that delayed registration took place over that period.

Role of the health system

One campaign strategy has been to tie registration more closely to the provision of health care. For instance, midwives and health workers were instructed to register children during child health campaigns.12 According to UNICEF,2 “immunization efforts provide an opportunity for health-care workers to be alerted to the absence of a health card or birth certificate, leading vaccination to be viewed as a potential point of entry to registration for a child”. Another study suggests16 that in Ghana “the collaboration between the civil registration office and Ghana Health Service, where volunteers and registration officers accompany community health nurses to the maternal and child welfare clinics in the communities to register infants, has the most direct impact on birth registration coverage”. The fact that registration offices are often located within health facilities or close to them implies a direct connection between health care and registration.

As discussed by Addo,13 a functioning interface from registration offices to health-care providers was still a task for the future in Ghana in 2009. Therefore, the registration system may not have been used yet for the planning of health services between 2004 and 2008. Fig. 1 plots the association between different indicators of access to health-care and birth registration, as well as between registration and household wealth quintile and urban residence. Through a logistic regression model we tried to isolate the effect of specific factors by controlling for each health-care access indicator and a range of family characteristics. We conducted separate estimates using DHS data for all children who were younger than 5 years in 2003 (those born in the pre-campaign period) and in 2008 (those born during the campaign). The health access variables we employed depict access to health care at birth, access to institutionalized health care, access to immunizations and participation in vaccination campaigns.

Fig. 1. Association between access to health care, household wealth quintile and urban/rural residence status on the probability of birth registration in 2003 (n = 3212) and 2008 (n = 2490), Ghana

The analysis indicates that between 2003 and 2008, access to health care at birth (i.e. birth in a health facility) became less important as a determinant for registration. In 2003, the likelihood of having been registered was 12 percentage points lower for a child born at home than for one born in a health facility; in 2008, this likelihood was only 5 percentage points lower. In 2003, children who received polio vaccine at birth were 6 percentage points more likely to have been registered than those who did not, a difference that was statistically significant; in 2008, the difference was no longer significant. Regarding institutionalized care, in 2003 children whose mothers had visited a health facility within the last year were 7 percentage points more likely to have been registered than those whose mothers had not done so. This may be because of the ease of accessing registration facilities during health centre visits due to their proximity, or because the visit alerted health officials to the absence of a birth certificate. However, in 2008 those children whose mothers had accessed a health facility were no longer more likely to have been registered. On the other hand, having been vaccinated showed an association with registration in both years, and this association was even stronger in 2008. A significant positive association between registration and the receipt of all doses of the diphtheria, tetanus and pertussis (DPT) vaccine and vitamin A was present both in 2003 and 2008. The connection with vitamin A became stronger in 2008 and that with the receipt of measles also was statistically significantly positive. Birth registration has been incorporated into vaccination campaigns in Ghana since 2004. In 2003, children who participated in a vaccination campaign were 8 percentage points less likely to have been registered than those who did not. However, by 2008 the difference is no longer statistically significant, which suggests that children participating in vaccination campaigns were no longer disadvantaged with respect to registration. By 2008, rural children were no longer at a disadvantage, as they had been in 2003, with respect to urban children in terms of birth registration, and socioeconomic status played a smaller role in the likelihood of registration in 2008.


This study shows that the birth registration campaign initiated in Ghana in 2004 substantially increased registration rates among children younger than 5 years. It reduced inequalities in registration as a function of socioeconomic status and place of residence (urban versus rural) and weakened the association between birth registration and access to health care at birth or subsequent access to health centres. However, vaccinated children were more likely to have been registered both before and during the campaign period. Vaccination in turn could take place not only during health centre visits, but also through community health workers and through mobile services and outreach health activities.

The key policy lessons are summarized in Box 1. The findings of this study show that the incorporation of birth registration into community health care and child health campaigns, together with mobile registration activities in remote areas, succeeded in raising registration rates. However, full registration coverage has not been reached and progress has slowed down, with an estimated 65% of births registered in 2011, a rate similar to the 2008 rate for children younger than one year.15 Hence, efforts should be made to target the poorest households, which are less likely than more prosperous households to have access to vaccination and health centres. It may not be possible to rely on mobile strategies and outreach activities as permanent, long-term solutions. In more remote areas, the promotion of community population registers is seen as a key strategy. Additionally, health facilities could be even more strongly connected to birth registration by including registration facilities in all health centres and mandating that health workers register births. Not all health facilities in Ghana have incorporated registration facilities, but a large proportion of children still come into contact with health facilities for basic health-care needs. Thus, more remains to be done to connect health facilities with the registration process.

Box 1. Summary of main lessons learnt

  • The incorporation of birth registration into community health care, health campaigns and mobile registration activities have increased birth registration in Ghana by reducing the indirect costs of birth registration, especially in poorer communities.
  • The links between the health sector and birth registration should be strengthened further, ideally by locating registration facilities within all health facilities.
  • In more remote areas, local community population registers should be actively encouraged to expand registration coverage.


The authors thank Emelia Allen from UNICEF Ghana and Simon Heap (formerly at Plan International) for helpful background information and discussion.

Competing interests:

None declared.