Globally, the proportion of unsafe abortions in developing countries is 49.5% [
Despite the 1985 Ghanaian law accommodating safe abortion practice on certain medicosocial grounds, there have been observed delays in the formulation and implementation of policies to integrate the law into health services [
These challenges associated with the delays in the implementation of comprehensive abortion care in Ghana make the use of modern contraceptives imperative and a reliable option towards reducing the high rates of unwanted pregnancies [
The use of modern contraceptives and its associated factors among women age (15–49 years) have been documented in Ghana [
This study, therefore, used a nationally representative survey [
This study is a retrospective cross-sectional design using the Ghana Maternal Health Survey (GMHS) conducted in 2017. The GMHS is a nationally representative special survey, which collects data relating to women, children, birth, household, and other feminine variables covering all regions and districts of the country. The survey was first conducted in 2007 and also in 2017.
For this current Ghana Maternal Health Survey (GMHS) 2017, 27,001 households were selected for the sample, of which 26,500 were occupied at the time of fieldwork. Data were collected from participants with a questionnaire through a multistage probability sampling technique. During the first stage, proportional sampling was used to select 900 enumeration areas (EAs) (including 466 EAs in urban and 434 EAs in rural areas, respectfully) to reflect the sizes of EAs with independent sampling in each stratum. Secondly, a household listing was carried out from 25 January to 9 April 2017 in all the selected EAs to serve as a sampling frame for the selection of households in the second stage. The interviewed households were 26,324, resulting in a response rate of 99%. With the interviewed households, 25,304 eligible women were recruited for individual interviews. However, interviews were completed with 25,062 women. The response rate of the completed interview was 99%. The sampling procedure is extensively detailed in the complete final report of the Ghana Maternal Health Survey 2017 (ICF, Ghana Health Service (GHS), 2018) [
The study included women in their reproductive ages between 15 and 49 years who have ever been in a situation in which she or someone else had to do something to end her pregnancy.
The exclusion criteria were those who met the inclusion criteria above but had missing data.
The outcome variable was “current usage of any method of contraceptives (modern or traditional) to delay or avoid getting pregnant.” The responses were coded as either “Yes” or “No.”
We included several theoretical pertinent sociodemographic variables. Women age group (14–19, 20–24, 25–29, 30–34,35–39, 40–44, and 45–49), religious status (Christianity, Islam, traditional, and others), age at first intercourse (<16 and 16 or above), health insurance (yes or no), knowledge of fertility (yes or no), current union (yes or no), region (Eastern, Greater Accra, Northern, Upper East, Upper West, Volta, Eastern, Central, Ashanti, Western, and Brong Ahafo), education (primary, JHS, SHS or above, and no education), place of residence (rural or urban), and gravidae (1–3, 4-5, and 6 or more). Furthermore, variables were recoded where appropriate to produce a meaningful sample for the analysis.
This analysis used both descriptive and inferential methods. Descriptive statistics used are mainly frequencies and percentages. Chi-square tests, univariate and multivariable techniques were used to assess statistical associations between the outcome variable and the predictors.
The chi-square test of independence was used to statistically identify factors associated with the outcome variable looking at their estimated confidence intervals (CI), and
This statistical approach was implemented in STATA (Stata Statistical Software: Release 16. College Station, TX: StataCorp LP) software. The complex survey command “svy” was used to estimate means, proportions, and confidence intervals (CI). The final model goodness of fits was checked using the “svylogitgof” command [
The Ghana Health Service Institutional Review Board (IRB) approves the Demographic Health Surveys study protocol, survey instruments, and materials before commencement of the surveys. Individual consent was also obtained during the data collection process. Data were then used after approval was obtained. Furthermore, this study concept was presented to ICF for permission to use the datasets in this study, and approval was duly granted before its use. All terms of use have been observed.
A total of 3,039 women met the inclusion criteria for the study. The mean and standard deviation age in years of the women was 32.93 (8.32). Almost 37% (95% CI: 34.6, 38.84) of the participants used contraceptives. The mean and standard deviation age in years of respondents using contraceptives was 33.11 (8.09). The participants who had education above primary school level were 73.43%. Approximately, 67% of participants were in union (either married or cohabitating with a man). Majority of participants were Christians (91.16%), and 68.35% of participants lived in urban areas. Greater Accra and Ashanti regions cumulatively constituted 51.1% of all participants (Table
Background characteristics of respondents (
Variable | Frequency | Percentage |
---|---|---|
Age | ||
15–19 | 115 | 3.516 |
20–24 | 446 | 14.45 |
25–29 | 629 | 20.85 |
30–34 | 576 | 18.74 |
35–39 | 515 | 17.36 |
40–44 | 399 | 13.54 |
45–49 | 359 | 11.53 |
Educational level | ||
Primary | 503 | 16.68 |
JHS | 1222 | 41.29 |
SHS and above | 1011 | 32.14 |
No education | 303 | 9.891 |
Religious distribution | ||
Christianity | 2751 | 92.16 |
Islam | 244 | 6.276 |
Traditional | 8 | 0.2365 |
Others | 36 | 1.325 |
Contraceptives use | ||
Yes | 1130 | 36.69 |
No | 1909 | 63.31 |
Residence | ||
Rural | 925 | 31.65 |
Urban | 2114 | 68.35 |
Union | ||
Yes | 2002 | 66.51 |
No | 1037 | 33.49 |
Region | ||
Western | 388 | 13 |
Central | 248 | 8.125 |
Greater Accra | 567 | 26.05 |
Volta | 212 | 6.005 |
Eastern | 364 | 10.62 |
Ashanti | 644 | 25.05 |
Brong Ahafo | 383 | 9.493 |
Northern | 86 | 0.7622 |
Upper East | 69 | 0.4839 |
Upper West | 78 | 0.4042 |
In Table
Distribution and chi-square analysis of outcome (contraceptives use) across respondents’ characteristics (sociodemographic, socioeconomic, and abortion-related characteristics).
Variable | Contraceptives use | |
---|---|---|
No (%) | Yes (%) | |
Age | ||
| ||
15–19 | 61 (2.84) | 54 (4.674) |
20–24 | 224 (11.66) | 222 (19.28) |
25–29 | 347 (18.31) | 282 (25.23) |
30–34 | 363 (18.75) | 18.73 (18.73) |
35–39 | 334 (18.42) | 181 (15.54) |
40–44 | 298 (15.84) | 101 (9.59) |
45–49 | 284 (14.18) | 77 (6.96) |
Union | ||
| ||
Yes | 1232 (64.29) | 770 (70.33) |
No | 677 (35.71) | 360 (29.67) |
Educational level | ||
| ||
Primary | 320 (16.22) | 183 (17.45) |
JHS | 727 (39.73) | 495 (44) |
SHS | 683 (34.81) | 328 (27.54) |
No education | 179 (9.24) | 124 (11.01) |
Religion | ||
| ||
Christianity | 1720 (92.33) | 1031 (91.87) |
Islam | 162 (6.23) | 82 (6.35) |
Traditional | 4 (1.87) | 4 (0.32) |
Others | 23 (1.25) | 13 (1.45) |
Residence | ||
| ||
Rural | 532 (28.14) | 393 (37.71) |
Urban | 1377 (71.86) | 737 (62.29) |
Age at first intercourse | ||
| ||
<16 | 589 (30.49) | 347 (31.86) |
16+ | 1320 (69.51) | 783 (68.14) |
Knowledge of fertile period | ||
| ||
Yes | 1710 (90.25) | 1058 (94.23) |
No | 199 (9.75) | 72 (5.77) |
Registered with health insurance | ||
| ||
Yes | 1543 (80.52) | 941 (83.19) |
No | 366 (19.48) | 189 (16.81) |
Gravidae | ||
| ||
1–3 | 760 (39) | 516 (44.67) |
4–5 | 590 (32.56) | 29 (25.24) |
6+ | 559 (28.44) | 323 (30.1) |
Chi-square tests revealed respondents’ current age (95% CI:
Univariate and multivariate logistic analyses of predictors of contraceptives use among women who ever had induced abortion (15–49 years).
Variable category | Univariate OR (95% CI), | Multivariate aOR (95% CI), |
---|---|---|
Sociodemographic | ||
Age of respondents | ||
15–19 | 1 | 1 |
20–24 | 1.01 (0.62–1.63), 0.979 | 0.97 (0.60–1.57), 0.903 |
25–29 | 0.84 (0.52–1.35), 0.467 | 0.76 (0.46–1.25), 0.279 |
30–34 | 0.61 (0.38–0.98), 0.043 | 0.50 (0.30–0.85), 0.010 |
35–39 | 0.51 (0.31–0.83), 0.008 | 0.35 (0.20–0.62), 0.000 |
40–44 | 0.37 (0.22–0.616), 0.000 | 0.25 (0.14–0.45), 0.000 |
45–49 | 0.30 (0.18–0.48), 0.000 | 0.22 (0.12–0.41), 0.000 |
Age at first intercourse | ||
<16 | 1 | |
16 or above | 0.94 (0.77–1.14), 0.519 | |
Union | ||
No | 1 | 1 |
Yes | 1.32 (1.09–1.59), 0.004 | 1.45 (1.20–1.75), 0.000 |
Religious affiliation | ||
Christianity | 1 | |
Islam | 1.02 (0.69–1.52), 0.906 | |
Traditional | 1.734 (0.40–7.56), 0.463 | |
Others | 1.17 (0.54–2.53), 0.693 | |
Region | ||
Central | 1 | 1 |
Western | 1.75 (1.12–2.75), 0.015 | 1.58 (1.00–2.49), 0.049 |
Ashanti | 1.820 (1.21–2.74), 0.004 | 1.87 (1.25–2.80), 0.002 |
Northern | 2.20 (1.19–4.08), 0.012 | 1.98 (1.05–3.72), 0.034 |
Brong ahafo | 2.06 (1.28–3.31), 0.003 | 2.07 (1.31–3.26), 0.002 |
Upper West | 1.06 (0.53–2.13), 0.876 | 0.92 (0.44–1.90), 0.819 |
Upper East | 1.61 (0.80–3.27), 0.184 | 1.50 (0.72–3.13), 0.284 |
Greater accra | 1.12 (0.72–1.73), 0.612 | 1.37 (0.87–2.15), 0.175 |
Volta | 1.52 (0.93–2.48), 0.093 | 1.60 (0.97–2.62), 0.175 |
Eastern | 2.27 (1.47–3.49), 0.000 | 2.32 (1.50–3.60), 0.000 |
Socioeconomic characteristics | ||
Educational level | ||
Primary | 1 | 1 |
JHS | 1.03 (0.80–1.54), 0.821 | 0.87 (0.66–1.14), 0.322 |
SHS or above | 0.74 (0.55–0.99), 0.043 | 0.79 (0.56–1.11), 0.173 |
No education | 1.11 (0.80–1.54), 0.542 | 1.04 (0.73–1.48), 0.831 |
Place of residence | ||
Urban | 1 | 1 |
Rural | 1.55 (1.26–1.90), 0.000 | 1.31 (1.04–1.65), 0.022 |
Registered with health insurance | ||
Yes | 1 | |
No | 0.84 (0.68–1.03), 0.086 | |
Abortion-related characteristics | ||
Gravidae | 1 | 1 |
1–3 | 1 | 1 |
4–5 | 0.68 (0.55–0.84), 0.000 | 0.95 (0.75–1.23), 0.733 |
6 or more | 0.92 (0.75–1.14), 0.454 | 1.74 (1.35–2.23), 0.000 |
Knowledge of the fertile period | ||
No | 1 | 1 |
Yes | 1.63 (1.08–2.46), 0.020 | 1.84 (1.30–2.60), 0.001 |
OR, odds ratio. aOR, adjusted odds ratio. CI, confidence interval. Goodness of fittest. F-adjusted test statistic = F (9,680) = 0.84. Prob > F = 0.5748.
A univariate and multivariate logistic regression was used to identify further significant predictors of contraceptives use (Table
Also, the probability of using contraceptives among participants with SHS level of education was 0.74 (95% CI: 0.55–0.99,
This study was carried out to estimate the proportion and identify predictors of contraceptives use among women with induced abortion history. We estimated 36.69% (95% CI: 34.6, 38.84) contraceptives use in this population. The estimated proportion is lower compared to similar studies carried out by Makenzius et al., Mekuria et al., and Opoku [
The findings also revealed that older women aged 30–34, 35–39, 40–44, and 45–49 are less likely to use contraceptives relative to younger ones (15–19 years); this corroborates a study conducted by Abebe [
This study finding also observed that participants in union (married or cohabitating) have higher odds of using contraceptives than those not in a union, which is similar to other studies [
Women living in rural areas had higher odds of contraception than those in urban areas. However, rural women access to maternal and child health services are lowest compared to those in urban areas [
While acknowledging the importance of all these predictors highlighted, there is the need for broader methods to satisfy clients varieties and differences regarding contraception [
This study is from a nationally representative sample and makes room for generalizability of study findings across Ghana. Also, Demographic and Health Surveys (DHS) are planned properly and well-executed; hence, the data are of high quality. Furthermore, observations with complete dataset meeting the study criteria was large.
Firstly, the data used for this study were obtained through a cross-sectional study design, hence preventing causations from being inferred. The survey [
We explored the proportion of contraceptives use and its associated significant predictors among women aged 15–49 years with a history of induced abortion. The proportion was low at 37%. Women age, union, place of residence, knowledge of fertile period, gravidae, and region were identified as significant predictors of contraceptives use. We recommend that the Ministry of Health and Ghana Health Service facilitate widespread and easy access to modern contraceptives, provide postabortion contraception services, including guidance and counselling to women who need these services. In addition, we recommend efforts to be enhanced in the formulation and implementation of policies by the Government of Ghana to effectively and efficiently integrate the abortion law into services, and lastly, we recommend that the National Health Insurance Scheme (NHIS) should absorb contraception services as a way of curbing high cost as a potential predictor of its low uptake
The datasets generated and/or analyzed during the current study are available in the Ghana demographic and health repository,
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
The authors declare that there are no conflicts of interest regarding the publication of this paper.
MGS conceptualized the present study. MGS led the data extraction and analysis and wrote the first draft of the manuscript. MGS and MK contributed to the write-up of the different sections of the manuscript. MK and MGS reviewed the draft manuscript and contributed to the final version of the manuscript. All authors read and approved the final manuscript before submission.