Utilization of antenatal care among adolescent and young mothers in Ghana; analysis of the 2017/2018 multiple indicator cluster survey

We analysed data from the sixth Multiple Indicator Cluster Survey in Ghana. The survey collected data on socio-demographic and health indicators at the national level. The target population for the survey were persons between the ages of 15-49 years, both males and females. A two-stage stratified sampling technique was employed to recruit participants into MICS. Participants were recruited across the former ten regions of Ghana using the 2010 Population and Housing Census as the sampling frame. In the first stage of the sampling, enumeration areas were selected from both rural and urban areas proportional to size. In the second stage, a systematic sampling technique was employed to select households within the enumeration areas. Regarding the women survey, data were collected from participants using Computer Assisted Personal Interviewing (CAPI).

The unweighted total sample size for the women survey (15-49 years) was 14,374 women. However, this study focused on mothers aged 15-24 years. Therefore, women aged 25-49 years (8538), women who did not utilize ANC services in the last 2 years to the survey (4707) and those who responded ‘do not know’ (7) were dropped/deleted before the analysis. The remaining sample for women aged 15-24 years was (unweighted) 1122 women. After adjusting for the sample weight, the sample size was reduced to (weighted) 947 women aged 15-24 years. Participation in MICS was voluntary and informed consent was obtained from adult participants (18-24 years) and caregivers of minors (15-17 years). Assent was obtained from minors. The 2017/18 MICS was approved by the Ghana Health Service Ethics Review Committee. Details about the 2017/18 MICS are provided elsewhere [20].

The main outcome of this study was ANC visits. This was a single continuous variable (i.e how many times you received antenatal care during this pregnancy?). During the analysis, the outcome was recoded into a dichotomous variable (1-3 ANC visits = 0 and 4 or more ANC visits = 1). The categorization was based on WHO’s previous recommendation of 4 or more ANC visits during pregnancy [13, 14]. Between 2002 and 2016, WHO advocated for a four-visit model of focused antenatal care, which prioritised the delivery of evidence-based interventions at each visit. The 2017/18 MICS was conducted barely a year after the WHO latest recommendation of 8 or more ANC visits. Hence, the implementation of the new policy was commencing in Ghana.

The following exposures were identified in the literature: mother’s age, mother’s educational status, marital status, household wealth index, area of residence, region [13, 16, 21] and health insurance membership [22‐24]. Exposure to the media, including frequency of reading newspaper/magazine, frequency of listening to the radio and frequency of watching television, was also identified in the literature [25‐27]. We also assessed the effect of mobile phone ownership and internet use on ANC utilization. Maternal age was coded as (15-19 years = 1 and 20-24 years = 2). Marital status was coded as (currently married/in union = 1, formerly married/in union = 2, and never married/in union = 3). Educational status was coded as (pre-primary or none = 0, primary = 1, junior high = 2, secondary = 3 and higher = 4). Household wealth index was coded as (poorest = 1, second = 2, middle = 3, fourth = 4 and richest = 5). In addition, area of residence was coded as (urban = 1 and rural = 2), while region was coded as (Western = 1, Central = 2, Greater Accra = 3, Volta = 4, Eastern = 5, Ashanti = 6, Brong- Ahafo = 7, Northern = 8, Upper East = 9 and Upper West =10). Items that measured exposure to the media included frequency of listening to the radio, watching television and reading newspaper/magazine, which were coded as (not at all = 0, less than once a week = 1, at least once a week = 2 and almost every day = 3). Mobile phone ownership was coded as (Yes = 1 and No = 2) and internet use was coded as (Yes = 1 and No = 2).

Data were analysed with Stata/SE version 16 (StataCorp, College Station, Texas, USA). We employed both univariate and multivariable analyses. Descriptive statistics, including frequency, percentage and graphs, were computed at the univariate level. At the multivariable level, we computed Binary Logistic Regression. In the crude analysis, we assessed the effect of each exposure variable on the outcome variable. In the adjusted analysis, we assessed the combined effect of all the exposures on the outcome. To offset challenges associated with oversampling, we reported the weighted results. The adjusted model fitted reasonably well (p > 0.05) [28]. All statistical analyses were reported at the 0.05 significance level.

It was found that the majority (70%) of the participants were aged 20-24 years. About half (49%) of the participants had junior high school education and 61% were currently married/in a union. Exactly 9% of the participants were in the richest wealth quintile, while 26% were in the poorest wealth quintile. The majority of the participants resided in rural areas (65%) and 36% of them had enrolled on the National Health Insurance Scheme. Regarding exposure to the mass media, nine in ten participants did not read newspapers, 26% listen to the radio every day and 45% watched television every day. Also, more than half (90%) of the participants had never used the internet and more than five in ten participants owned a mobile phone. Details are provided in Table 1. Concerning ANC visits, the majority (84%) of the participants obtained 4 or more visits, while 16% obtained 1-3 ANC visits.

At the crude analysis level, it was found that mothers aged 20-24 years were 1.5 times (COR = 1.55; 95% CI: 1.00-2.38) more likely to obtain 4 or more ANC visits compared with those aged 15-19 years (reference category). Participants with senior high school education were 3 times (COR = 3.07; 95% CI: 1.13-8.33) more likely to obtain 4 or more ANC visits compared with those with pre-primary or no education (reference category). In addition, participants in the fourth wealth quintile (COR = 3.04; 95% CI: 1.39-6.66) had a higher likelihood of obtaining 4 or more ANC visits compared with those in the poorest wealth quintile (reference category). Participants residing in rural areas had decreased odds (COR = 0.51; 95% CI: 0.29-0.88) of obtaining 4 or more ANC visits compared with those residing in urban areas (reference category). Further, adolescents and young mothers who have ever used the internet were 6 times (COR = 6.72; 95% CI: 2.22-20.36) more likely to obtain 4 or more ANC visits compared with those who have never used the internet (reference category).

At the adjusted analysis level, the likelihood of obtaining 4 or more ANC visits was associated with higher educational status, socio-economic status, geographical region and exposure to the internet. For instance, adolescents and young mothers with junior high school education were 2 times (AOR = 2.19; 95% CI: 1.03-4.66) more likely to obtain 4 or more ANC visits compared with those with pre-primary or no education (reference category). Also, participants in the second wealth quintile were 2 times (AOR = 2.23; 95% CI: 1.20-4.16) more likely to obtain 4 or more ANC visits compared with those in the poorest wealth quintile (reference category). Further, adolescent and young mothers residing in the Upper East region were 6 times (AOR = 6.30; 95% CI: 1.03-38.74) more likely to obtain 4 or more ANC visits compared with those in the Greater Accra region (reference category). Adolescents and young mothers who had ever used the internet were 4 times (AOR = 4.38; 95% CI: 1.09-17.64) more likely to obtain 4 or more ANC visits compared with those who had never used (reference category) Table 2.

The findings of this study provide relevant information for maternal health policy and programming. For instance, a substantial proportion of adolescent and young mothers did not obtain the recommended ANC visits. These young mothers are at a higher risk of pregnancy and childbirth complications as well as negative birth outcomes. This may delay the achieving of Sustainable Development Goals 3, which aims to ensure healthy lives and promote wellbeing at all ages, including adolescent and young mothers. It is, therefore, necessary for stakeholders, including the Ministry of Health and Ghana Health Services, to invest resources that will help increase ANC coverage among young mothers. Stakeholders can leverage existing youth-friendly initiatives such as the Adolescent Health and Development (ADHD) programme. Currently, the ADHD programme does not cover ANC services, hence, stakeholders should consider incorporating these services. Also, the Ghana Health Service should consider separating young mothers from adult mothers during ANC visits to help encourage ANC utilization among adolescent mothers. Owolabi and colleagues [4] revealed that adult mothers stigmatize pregnant adolescent girls. Moreover, stakeholders should strengthen efforts towards providing Focused Antenatal Care. This may help increase ANC utilization among adolescent and young mothers in Ghana.

Further, adolescent and young mothers from poor households, those with lower education and no internet exposure were less likely to obtain optimal ANC visits. This suggests that poor mothers still face financial barriers to accessing ANC services despite the Free Maternal Health Care Policy. This has the potential to delay progress towards improving maternal and child health outcomes as well as achieving SDGs 3. Therefore, the National Health Insurance Authority needs to liaise with health service providers to eliminate all forms of unapproved charges (fees for medication, ultrasound scan, urine and blood test) on maternal healthcare services as reported by Ziblim and colleagues [14] in Northern Ghana. In addition, stakeholders must invest resources in promoting girl child education. Hence, Ghana’s free basic and secondary education policy is commendable. The internet may provide a golden opportunity for stakeholders to increase ANC coverage among young mothers. With the rapid increase in internet penetration in the country, stakeholders can leverage the internet to raise awareness and educate adolescent and young mothers about the importance of ANC and danger signs in pregnancy and childbirth.

This is the maiden study in Ghana to estimate the prevalence of ANC visits among adolescent and young mothers using national representative data. Although this study provides invaluable information for maternal health policy, it is not devoid of limitations. This study focused on socio-demographic factors and ANC utilization; hence the interpretation of the findings must be done with caution. Another limitation of this study is the small sample size. Also, quantitative surveys are unable to expose the many intricate views of participants regarding a subject matter. Future studies should, therefore, consider adopting qualitative designs as well as assessing more exposures. For instance, the 2017/18 MICS did not collect data on some potential exposures of ANC utilization among adolescent and young mothers. These exposures include school attendance, pregnancy history and violence against women, hence they should be included in future surveys.