Trends in Male Contraceptive Use in Ghana: An Analysis of Self-Reported Contraceptive Use from GDHS 2003, 2008 and 2014

Background: Male participation in contraception is proven to have positive outcomes on reproductive health and remains a critical issue of public health importance. However, there is a dearth of evidence on men’s share of the contraceptive burden among couples in Ghana. Knowledge of the trends and trajectories of men’s contraceptive practice could inform male focused programmatic interventions to boost contraceptive use for effective fertility regulation. This study examined the trends in contraceptive use as reported by men. Methods: This paper examines trends in self-reported contraceptive use among sexually active Ghanaian males, using data from the three most recent national Demographic and Health Surveys (2003, 2008 and 2014). Frequency distributions of socio-demographic and background characteristics were used to describe the sample by each survey year. The Z test of proportions was used to identify signicant differences in the estimated proportions and subgroups of men using contraceptives by survey year. The Combined data from the three nationally representative surveys were analyzed, accruing a sample of 9,008, comprising (GDHS 2003=3,104; GDHS 2008=3,007; and GDHS 2014=2,897) male respondents who met the inclusion criteria. Results: Men’s contraceptive use appears to have declined over the last three demographic surveys despite an increase in the proportion of sexually active males. However, a signicant proportion of men who used contraception (44.5% - 41.5%) still relied on the male condom as their primary choice of contraception. Place of residence, educational attainment, occupation, region of residence and religious aliation were all found to be signicantly associated with contraceptive use consistency. Conclusion: There has been a signicant decrease in contraceptive use by Ghanaian men between 2003 and 2014. Although the male condom appears to be the most popular modern male contraceptive method, especially during adolescence, along the life course the phenomenon changes, with a shift towards modern female methods. There is therefore, the need for family planning service providers, policymakers and all relevant stakeholders to consciously target men with contraceptive products and services and community engagement to improve FP uptake residence, Educational attainment, occupation, region of residence and religious aliation were all statistically signicant with men’s contraceptive use consistency in the two surveys (p<0.001). an increasing male adolescent population, who without specialised service centres to address their needs will report diculties in accessing contraceptives, including condoms. Pattern of men’s contraceptive method preference at last sex among users: 2003-2014


Sample design and sampling
The country's general population as projected by the Population and Housing Censuses (PHCs) serves as the sampling frames for the Ghana Demographic and Health Survey (GDHS) data collection with the exclusion of institutional (hotels, barracks, and prisons) and nomadic populations. The GDHS employs a two-stage sampling design; the rst stage selects sample points or clusters consisting of enumeration areas (EAs) as demarcated by the GSS for the PHCs. The second stage involved systematic sampling of approximately 30 households per selected EA. Weighting factors were applied to the data to ensure nationally representative results.

Questionnaires
The GDHS uses three standard DHS questionnaires: Household, Woman's, and Man's Questionnaires tailored to mirror population and health issues in Ghana. Based on this paper's interest, the authors used data from the Man's Questionnaires of the three GDHS datasets, involving men aged  years.

Sample
The inclusion criterion was all male respondents aged 15-59 years at the time of each GDHS. Men who had not been sexually active in the past six months, or who needed a child within one (1) year of the survey were excluded. Men with an infecund or currently pregnant partner were also excluded. The variables that were used for this analysis were informed by relevant literature which detailed various determinants of contraceptive use among men. The outcome of interest, male contraception (men's contraceptive use), was de ned as the use of contraception by male respondent or partner during last sexual intercourse at the time of each survey. On the other hand, the predictor variables for the analysis were the socio-demographic, and sexual and reproductive health characteristic of respondents.

Data analysis
The data were analyzed using STATA version 15. Descriptive statistics showing frequency distributions and proportions of respondents by sociodemographic and other background factors were carried out, summarized and presented in tables, graphs and charts. Bivariate analysis using the Ztest of proportions was used to identify signi cant differences in identi ed proportions. Logistic regression was used to identify signi cant predictors of men's contraceptive use for each GDHS survey year. Based on the outcomes, contingency tables were generated, and results were reported at 95% con dence interval.

Results
Socio-demographic characteristics Table 1 presents socio-demographic characteristics of respondents. In all the three surveys, a higher proportion of males were aged 25-39 years. The mean age of participants between 2003 and 2014 did not change: 35±11.1 in 2003 and 35±11.6 in 2014. There was a decline in the proportion of men who had never been to school -from 17.6% in 2003 to 10.8% in 2014 The proportion of rural dwellers dipped from 54.0% to 52.8% to 46.9%, characterizing growing urbanisation. The proportion of men living below the middle class (poorest and poorer quintiles) decreased from 34.0% in 2003 to 32.2% in 2014. Table 2 presents demographic and sexual and reproductive health characteristics of participants. The proportion of married men reduced by approximately 8 percentage points, while the proportion of men who had never married or been in a recognised union increased by about 7 percentage points. Men's knowledge of any contraceptive method appeared almost universal over the period (99.6% -99.7%) although a small proportion (4.4% -5.6%) reported uneasy access to contraceptives like condoms. Table 1 >>   << Insert Table 2 Table 3. Table 3  Difference in inter-regional method preference by users was signi cant for each survey (p<0.001). By religion, modern methods use increased for both male (43.6% -52.5%) and female methods (23.3% -29.5%) among catholic men, with a 13.  Table 4.

<< Insert Table 4 >>
The results (Table 5) shows that over a third of male adolescents (35.3%-34.5%) [2003 -2008] had never used any contraception, and over the same period, an increasing proportion (23.1%-30.6%) reported inconsistent contraceptive use. Conversely, majority of men aged 40 years and above were inconsistent contraceptive users in 2003(35.4%) and 2008(44.6%). This difference in contraceptive use consistency between age groups was statistically signi cant (p<0.001) in the two surveys.
Between 2003 and 2008, over a third of rural dwelling men had never used contraceptives (33.9%-33.7%), whilst consistent rural male contraceptive users decreased from 36.4% to 28.4%. The majority of men with no formal education had never used contraceptives (58.0%-61.5%) compared to their counterparts with tertiary education (8.6% -8.8%).. Place of residence, Educational attainment, occupation, region of residence and religious a liation were all statistically signi cant with men's contraceptive use consistency in the two surveys (p<0.001).
<< Insert Table 5  Geographically, men from rural Ghana and those without formal education are less likely to use modern male contraceptives than their urban and educated counterparts. The results of this study, mirrors ndings in a Nigerian study which also identi ed differences in contraceptive use by geographical location [23], and [26,27] who separately reported that educational attainment was signi cantly associated with contraceptive choices. This may be because many rural dwellers have no formal education and consequently have poor knowledge about contraception variants. Additionally, rural socio-cultural norms that promote fertility and mitigate contraceptive uptake, as well as issues of access may be accountable for these differences.
This suggests a need for locally-adapted interventions for men with no formal education who work in informal sectors, such as the use of behavior change communication in local languages and with community engagement.
Geographically, more than half of men in 2003 and 2008 from the Savanna-North had never used contraceptives than men from the other regions of the country. These ndings are consistent with other studies in Ghana which reported signi cant socio-demographic variations in contraceptive use [26,[28][29][30][31]. The ndings in this study may be due to the men in the Savanna-North region having more Islamic, Traditional and other religions as well as be rural dwellers, all of which are known to in uence fertility and contraceptive decisions [32,33].
The Ghanaian male adolescent population is steadily increasing, and a greater proportion of males initiated sexual intercourse during adolescence in 2014 than did in 2003. This may be attributable to the fact that adolescents probably becoming more exposed to growing urbanisation and peer pressure from school and other social groupings. This study however discovered that between 2003 and 2008, more than a third of adolescents had never used contraception, whilst a smaller proportion were inconsistent contraceptive users. This suggests a need for special attention towards male adolescent sexual reproductive health needs particularly, on contraceptive use.
The analyses also discovered a decline in ease of contraceptive access as reported by respondents. The proportion of men who said it was easy to access condoms if they needed one dropped from 2003 to 2014. This phenomenon may be attributable to an increasing male adolescent population, who without specialised service centres to address their needs will report di culties in accessing contraceptives, including condoms.
Pattern of men's contraceptive method preference at last sex among users: [2003][2004][2005][2006][2007][2008][2009][2010][2011][2012][2013][2014] Results for the period between 2003 and 2014 show that among contraceptive users, there was an overall increase in the reliance on modern female methods (the pill, IUD, injectable, implant, female condom, female sterilization), although a signi cant proportion of users still relied on the male condom at last sex. This correlated with ndings by [34] in the Volta region of Ghana, which reported condom as the most (90%) accessed method.
[35]also reported that almost a third of secondary school male students consistently used the condom. These ndings are in contrast to other studies that report the underutilization of male condoms [10,36]. This increasing reliance may be attributable to the male condom being the only available, wellknown, easily accessible and reversible modern male-dependent method. Literature establishes that men will be more likely to use a method they know about; which they consider safe and effective; available; and grants autonomy of use [37][38][39].

Study limitations
Like any other survey, the data used in this study was subject to recall bias and evasion or misreporting on sensitive issues. These issues are particularly encountered when surveys retrospectively collect data spanning a period of ve (5) years preceding it. Also, it may be di cult for men to know and report their partners' actual contraception status. This analysis, however, utilizes the most reliable nationally representative data, believed to have captured the true reports, of sexually active Ghanaian men who are effectively involved in contraceptive decision-making with their partners.

Conclusions And Recommendations
Conclusions This paper makes an important contribution to literature by demonstrating the need for men to be factored into sexual and reproductive health issues and for proposing improvements required for effective contraception inclusiveness as an intervention for improving male involvement in FP and overall sexual and reproductive health for men and their partners. The study reveals that there has been a signi cant decrease in contraceptive use by men from 2003 to 2014. The male condom was the most common modern contraceptive method men use, especially during adolescence; and along the life course it shifts towards modern female methods such as the pill, IUD, injectable, implant, female condom, female sterilization. Whereas the proportion of male adolescents is increasing, more than a third of it has never used contraception of any sort. Finally, signi cant geographical and educational differences in contraceptive utilization exist among Ghanaian men.

Recommendations
From the ndings made in this paper, the following policy interventions were recommended: Government in collaboration with FP Service providers and program implementers should continuously focus programs and policies towards improving men's contraceptive use; FP interventions such as regular male-focused outreach FP services including counselling, and male-directed advocacy drives within catchment areas, should be implemented to get men informed, and, to understand the bene ts of limited family sizes to the family and country.
FP Service providers and policymakers should consciously engage local authorities, men, traditional and religious groups and other relevant stakeholders in all-inclusive approaches such as community engagement to improve FP uptake.
Interventions should be tailored and speci c to the needs of sub-population groups including the use of variable behavior change communications to ensure improved satisfaction and acceptability.

Availability of data and materials
The datasets analysed for this study are available to the public freely at MEASURE DHS, www.measuredhs.com. The MEASURE DHS standard questionnaires for the three surveys were used and can be found at http://dhsprogram.com/publications/publications-FR307-DHS-Final-Reports.cfm.

Competing interests
The authors declare no competing interests.
Funding CA received funding from the HRP Alliance, part of the UNDP-UNFPA-UNICEF-WHO-World Bank Special Program of Research, Development and Research Training in Human Reproduction (HRP), a cosponsored program executed by the World Health Organization (WHO), to complete his studies. This article represents the views of the named authors only and does not represent the views of the World Health Organization.
Authors' contributions CA conceived the study and AM, EM and AA contributed to its design. CA, AM, EM, HH AA and KT, made substantial contributions to data analysis plan, and CA performed all statistical analyses. All authors contributed to interpretation of data. CA drafted the initial version of the manuscript, it was revised thoroughly by HH, AM, EM and AA. AM, HH, EM, KT, RA and AA critically reviewed subsequent versions of the manuscript for intellectual content. All authors read and approved the nal manuscript. The total sample size (n) = 9,008 (2003=3,104; 2008=3,007; 2014=2,897), but it varies for some variables due to missing values. The total sample size (n) = 9,008 (2003=3,104; 2008=3,007; 2014=2,897), but varies for some variables due to missing values and rounding up effects. DK = Don't know answer, as having been returned by respondent -Missing entries-question on age of most recent sexual partner was neither asked nor imputed for the 2003 survey.