Abstract
No country is immune from intimate partner violence (IPV). The objectives of this study were to analyze the prevalence, correlates and trends of this global public health menace perpetrated by men against women using the most recent nationally representative Gambia Demographic and Health Survey (GDHS) 2019-20; with analysis of levels and trends of IPV perpetrated by current/former husbands/partners from ever-married women, from GDHS conducted in 2013, at the 8 subnational regions. IPV association with 12 covariates entailing socio-demographic, experiential and attitudinal characteristics was analyzed in bivariate and multivariable models, using simple and multiple logistic regression. Physical, emotional, and sexual IPV was reported by 29.09%, 24.03%, and 5.52% respectively. While prevalence of having experienced any type of IPV was39.23%. Statistically significant association of IPV with various covariates computed in univariate analysis were used in the multivariable logistic regression model. Women’s and husband’s educational attainment, wealth status, having witnessed father physically beat mother, and marital control exhibited by the husband were statistically significantly associated with IPV, in the final model. Physical, emotional, and sexual IPV increased from 2023 to 2019-20 in all 8 regions, with the exception of sexual IPV in Kanifing region. However, not all these changes were not statistically significant. Physical and sexual IPV prevalence in Gambia was slightly lower compared with the African region. The disturbing finding of increase in all 3 types of violence in all regions—with one exception—paints a dire scenario that augurs the imperative of women empowerment and revisiting of cultural norms for protecting women.
The last nationally and regionally representative Demographic and Health Survey (DHS) on intimate partner violence (IPV) prevalence and its correlates was conducted in 2013 in Gambia, and studies have been published using this data on the correlates of IPV in a multivariable analysis.
This paper used the latest available that is, 2019-20 nationally and regionally representative DHS data on IPV in Gambia to compute the correlates of IPV at the national level using multivariable model, and the linear trend analysis at the regional level from 2013 to 2019-20.
The findings from this secondary analysis will potentially help the health and social policymakers to address the IPV burden more effectively in Gambia.
Introduction
Violence is the price some women pay for intimacy with men, which could manifest as emotional, physical, and/or sexual violence perpetrated by their male intimate partner. Intimate partner violence (IPV) is a universal phenomenon, as no country or culture is immune from it.1-6 Based on the review of data from 2000 to 2018, the World Health Organization (WHO) reported that globally 27% and in the African region 33% of ever married/partnered women aged 15 to 49 years experienced physical and/or sexual violence at least once in their lifetime. 7 Intimate partner violence (IPV) against women perpetrated by men is a global health problem and a human rights issue. The United Nations Sustainable Development Goal 5 envisage gender equality, and women empowerment. 8 IPV saddles its victims with plethora of health and economic problems ranging from the psychiatric sequalae, sexually transmitted diseases, poor pregnancy outcomes, various pain syndromes, suicide, homicide, loss of wages and compromised labor force participation.1,7
Four recent meta-analysis, systematic reviews, and multiple country studies highlighted important correlates of IPV against women. Using nationally representative cross-sectional Demographic and Health Surveys (DHS) data from 44 sub-Saharan African countries, a meta-analysis identified the crucial associations of rural residency status and women’s low educational attainment with IPV. 3 Another review using the “WHO Global Database on Prevalence of Violence Against Women” data highlighted the association of high IPV prevalence in in low-income countries, rural residency, and women’s low educational attainment. 2 While poverty as a metric of poor living conditions in terms of unimproved water access, sanitation facilities, insufficient space, and unfinished materials in the house was identified as a stout correlate of IPV in a study based on DHS data 25 sub-Saharan African countries. 4 The association of infertility and IPV was underscored by another meta-analysis based on systematic literature review of several online academic databases. 5
Intergenerational experiences of IPV has also been identified as a disturbing and perpetuating factor, with those women who had witnessed their father physically beat up their mothers while growing up, being more likely to report IPV themselves.9,10 Plethora of IPV correlates have been reported using DHS and other data: these include gender-based power differentials with women experiencing limited empowerment and autonomy in terms of making or contributing to personal or household decisions, acceptance of IPV, low educational attainment by women or her partner, alcohol use by the partner, age of women, age of partner, and marital control displayed by the partner.11-18
Hence, the mediating framework underpinning IPV perpetrated by men against women include women’s economic dependence on their male intimate partner; physical power imbalance with resultant limited ability to defend and respond; household poverty leading to family disputes and threatening traditional male role as a provider; low educational attainment in both partners with no or limited understanding the effects of IPV on the victim and acceptance of such behavior, alcohol use by male partner with resultant low inhibitions to violence, intergenerational witnessing and acceptance of IPV; limited autonomy in terms of making household and personal decisions; number of children: more children impacting family’s economic status, and infertility contributing to family and societal expectations to procreate; age of women: with older age accentuating economic dependence on male partner and younger age with limited ability to manage family affairs; and patriarchal societies dictating and perpetuating sociocultural norms reinforcing gender inequality, low empowerment and autonomy of women.1-7,9-18
Situated in the West Africa, the Republic of the Gambia is the smallest country in terms of land size in the mainland Africa. The first Demographic and Health Survey (DHS) was conducted in 2013, and the second in 2019-20. The “DHS Program” is part of the United States Agency for International Development (USAID) that has conducted over 400 Demographic and Health Surveys in over 90 countries around the world. 19 The 2 DHS surveys conducted in Gambia are the only nationally and regionally representative surveys on IPV in the country.
Few studies have been published on IPV and its correlates in Gambia. Using the Gambia DHS 2013 data, no statistically significant relationship was determined between a women’s experience of IPV and her having undergone female genital cutting or support for such cutting. 20 While another study using the same data reported that over 40% of women had reported either emotional, physical, and or sexual IPV. 21 Using the same data another study examined the association between sexual autonomy (determined in terms of capacity for asking partner to use condom or to refuse sex) and IPV found positive associations. 22 The association between IPV and pregnancy termination—defined as either stillbirths, miscarriages, or induced abortions—was found to be high in Gambia using 2013 DHS, in addition to 24 other sub-Sharan African countries among adolescent girls and young women. 23 Another study using the Gambia DHS 2019-20 data, in addition to 26 other sub-Saharan African countries, examined the association between women in sexual unions having ever witnessed their father physically beat their mother and their belief in IPV justification concluded that this justification was more pronounced in women who had such exposure 24 ; however, Gambia was an exception, with 3 other countries. A health facility-based cross-sectional study conducted in rural Gambia of 373 pregnant women reported the IPV prevalence of 67%. 25 Another cross-sectional study done in a teaching hospital in Gambia of 136 pregnant women reported the IPV prevalence of 61.8%. 26
The 2 objectives of this study were to analyze the prevalence and correlates of IPV perpetrated by men against women using recent nationally representative GDHS2019-20, and to provide levels and linear trends of IPV perpetrated by current/former husbands from GDHS20-13 to GDHS2019-20 at the regional level, for the 8 regions.
Materials and Methods
Study Area and Data Source
Using the cross-sectional demographic and health surveys (GDHS) data from Gambia, conducted in 2013 and 2019-20, a secondary analysis was conducted. For GDHS 2019-20, the data was collected from November 2019 to March 2020. Administratively, Gambia is subdivided into 8 regions; 6 Local Government Areas (LGAs) and 2 urban municipalities, named Brikama, Mansakonko, Kerewan, Kuntaur, Janjanbureh, Basse, and Banjul, Kanifing, respectively. Both GDHSs are representative at the national and the 8 subnational regions. For the conduct of 2019-20 GDHS, the sampling frame was based on 2013 Gambia Population and Housing Census, which was updated using the 2015-16 Integrated Household Survey (IHS). The sampling frame comprised of all the enumeration areas (EAs) in the country; with each EA entailing all households in a defined geographical area, with an average of 68 households. In total, there are 4098 EAs in the country.
The DHSs use a stratified 2-stage cluster design. For GDHS 2019-20, in the first stage, 281 EAs were selected, and household listing was carried out. Which served as the selection of 25 households from each EA, in the second stage using equal probability systematic sampling approach. Women aged 15 to 49 in the selected households were deemed eligible for the GDHS 2019-20. However, for the administration of “Domestic Violence” (DV) module that includes questions on intimate partner violence (IPV), every second household was selected with woman from each household selected for the administration of DV module. However, the IPV questions were asked from ever-married women only; for currently married women, the questions pertained to IPV committed by either the current husband or partner. While, for formerly married women, IPV committed by the most recent husband or partner. Hence, ever-married women were asked questions like: “Did your (last) (husband/partner) ever do any of the following with you: slap you?.”
The GDHS 2019-20 was conducted by the Gambia Bureau of Statistics in collaboration with the Gambian Ministry of Health; with technical support provided by the ICF International Inc., United States. The institutional review board of the Gambian Government/Medical Research Council Joint Ethics Committee in Gambia gave the approval for the conduct of survey; additionally, the ICF institutional review board also provided the ethical approval. The DHS program (www.dhsprogram.com) was requested and provided the approval for this secondary analysis. Anonymized data were used for this secondary analysis, as such no ethical approval was required. Furthermore, author has no conflicts of interest.
The data files for both GDHSs were downloaded in the Stata format. For the administration of domestic violence module, there were 2470 women selected and interviewed for the GDHS2019-20, while 4525 women were selected and interviewed for the GDHS2013. Owing to lack of privacy, interruptions during the conduct of interview, and/or inability to find the selected woman for interview despite repeated attempts, 49 women for GDHS2019-20 and 69 women for GDHS2013, albeit selected for the domestic violence module, could not be interviewed. Details of both GDHSs, including generation of sampling weights, adjustment for non-response sampling design, survey methodology, and survey questionnaires are freely downloadable from the country reports available at the DHS website.
Study Variables
The modified version of the Conflict Tactics Scale, a standardized domestic violence module was used for the GDHSs.27,28 The methodology of computing composite, various types of IPV, and their correlates has been described in detail previously, 29 a brief explanation is provided below.
Outcome Variable
Intimate partner violence (IPV) was defined as ever having experienced any form of either emotional, physical, and/or sexual violence perpetrated by the husband/partner and computed as a dichotomous outcome variable. In the GDHSs the husband/partner referred to the current husband/partner for currently married women and the most recent husband/partner for the divorced, separated, or widowed women. Hence, only ever-married women were asked questions pertaining to IPV.
Explanatory Variables
Previous studies from several African, Asian, and Latin American countries have reported various IPV correlates1-5,9-22,24,26; in this secondary analysis 12 correlates at the level of individual respondent, husband/partner, and familial levels were studied in terms of association with respondent having ever experienced any form of intimate partner violence. These included women’s’ age, women’s and her husband’s/partner’s educational attainment, women’s occupation, household’s wealth index, residential status (urban or rural), number of living children, decision making, IPV acceptance, husband’s/partner’s alcohol use, having ever witnessed one’s father physically beating up one’s mother, and marital control. Derivation of these explanatory variable has been previously described. 29 In brief, all variables were dichotomized, while “don’t know” answers were inscribed as “no.” Acceptance of any type of IPV, participation in any decision-making event, or any form of marital control demonstrated by the husband/partner, were coded as “present/yes” if answered in the affirmative. The age, household wealth, number of living children, educational attainment of respondent and her husband/partner, and occupation were coded as recorded in the Table 1.
Counts and Proportions of Study Variables—Gambia Demographic and Health Survey (DHS) 2019-20.
Asked from currently married women only.
Statistical Analysis
All reported P values are 2-sided, and P < .05 was considered statistically significant. Analyzes were performed using Stata, version 17.1 (StataCorp, 2021, Texas, USA), while accounting for the complex sample design of the GDHSs.
The analysis entailed 4 steps: unweighted counts, missing records, and cumulative weighted percentages were calculated for outcome and explanatory variables using GDHS2019-20 in the first step. In the second step, simple binary logistic regression models were developed to determine the statistical significance of each explanatory variable’s association with the outcome variable that is, women who reported having ever experienced any type of IPV. And the crude odds ratios based on simple logistic regression models, their statistical significance, and 95% confidence intervals were calculated. All explanatory variables found to be statistically significantly associated with the IPV in this step were used in the binary multiple logistic regression model in the third step to compute adjusted odds ratios, their statistical significance, and 95% confidence intervals. Finally, the 3 types of IPV (physical, emotional, and sexual) prevalence were computed using the GDHS2013 data, as well as for GDHS2019-20 for each geographical region. Followed by comparison of linear trend between the 2 surveys, by merging both data files together and using year as a dichotomous variable in the binary simple logistic regression model, with IPV as the explanatory variable. All proportions, expressed as percentages are weighted.
Results
The domestic violence module was administered to 2470 women, out of which 1953 were ever-married and were either currently or formerly in a union or living with a man. Hence, the IPV questions were asked from these ever-women only. The husband/partner referred to the current husband/partner for currently married women and the most recent husband/partner for divorced, separated, or widowed women. In total, 784 women reported one or more types of IPV perpetrated by their male partner. While emotional, physical, and sexual violence were reported by 505, 591, and 120 women, respectively. Seventy-two (2.91%, 95% CI: 2.04-4.16) women reported all 3 types of IPV; physical and emotional violence were reported by 327 (14.90%, 95% CI: 12.56-17.58) women; physical and sexual violence were reported 85 (3.61%, 95% CI: 2.59-5.02) women; while emotional and sexual violence were reported by 92 (3.82%, 95% CI: 2.79-5.20) women. There were 626 (31.00%, 95% CI: 27.52-34.70) women who reported physical and/or sexual IPV.
Table 1 shows results in terms of outcome and explanatory variables’ unweighted counts, and cumulative weighted percentages, based on the 1953 women aged 15 to 49 years who answered IPV questions. However, for 129 women, information on husband/partner’s educational attainment, and decision-making pertaining to healthcare seeking for herself, large household purchases, and visits to relatives, were not available as these questions were only asked from women were currently married.
The prevalence of having ever experienced emotional, physical, and/or sexual spousal violence perpetrated by either current or most recent husband/partner was 39.23% (95% CI: 35.67-42.92). The emotional, physical, or sexual IPV were reported by 24.03% (95% CI: 21.30-26.99), 29.09% (95% CI: 25.67-32.76), and 5.52% (95% CI: 4.09-7.43) women respectively. The most common type of physical IPV reported was ever having been slapped (21.17%, 95% CI: 18.06-24.66); most common type of emotional violence reported was ever having been insulted or made to feel bad (19.00% − 95% CI: 16.43-21.88); and the most common type of sexual violence reported was ever having been physically forced into unwanted sex (5.23% − 95% CI: 3.82-7.13).
Cumulatively among the survey respondents, over half (58.63%) of respondents were aged 30 years or over; 55.14% had formal education while 42.17% respondent’s husband/partner had some formal education; most (70.83%) did some work; 20.14% belonged to the poorest wealth index; most (70.33%) were urban denizens; 28.41% had 5 to 11 children; 69.92% did not take part in making major decisions; 54.06% believed that IPV was justified; husband/partner’s use of alcohol was reported by only 1.81%; 10.92% had witnessed their father had ever physically beaten their mother; and marital control was shown by 56.89% of women’s husband/partner.
Table 2 shows the results of simple and multiple logistic regression models in terms of crude Odds Ratios (OR), adjusted odds ratios (aOR), their statistical significance, and the associated 95% confidence intervals. Out of the 12 explanatory variables used in the bivariate analysis, 8 were statistically significantly associated with having ever experienced any type of intimate partner violence. The remaining 4 explanatory variables that is, occupation, residential status, participation in the decision-making, and alcohol use by the husband/partner were excluded in the final, multiple logistic regression model. Five out of 8 explanatory variables were found to be statistically significantly associated with having experienced IPV in the multiple logistic regression model.
Crude Odds Ratios and Adjusted Odds Ratios for all Statistically Significant Associations Between Intimate Partner Violence and the Selected Variables—Gambia Demographic and Health Surveys (DHS) 2019-20.
Note. Not Applicable: owing to the fact that a given explanatory variable was not statistically significant in the simple logistic regression model (unadjusted odds ratios), these were not included in the multiple logistic regression model that provided “adjusted odds rations”.
OR = Odds Ratio; CI = Confidence Interval.
In the multiple logistic regression model, compared to women with no education, women with education higher than secondary school experienced much less IPV with adjusted odds ratios (aOR) of 0.293 (95% CI: 0.131-0.654). While compared to women whose husband/partner had no education, women whose husband/partner had higher than secondary school education experienced almost twice as much IPV with the aOR of 1.969 (1.026-4.780). Compared to women in the poorest quantile of household wealth, the women in the richest quantile experienced less IPV with the aOR of 0.537 (95% CI: 0.294-0.981). Regarding having ever witnessed one’s father physically beat mother; the odds were 2.552 times (95% CI: 1.572-4.145) higher for women experiencing IPV in their own relationship, who witnessed such IPV, compared to women who did not witness such violence. Finally, While the odds of IPV were 4.340 times (95% CI: 3.124-6.030) higher in women who experienced marital control, compared to those women whose husband/partner did not engage in such controlling behavior.
Table 3 shows the emotional, physical, and sexual violence proportions and statistical significance of linear trend analysis across the 2 GDHSs. The lowest proportion of emotional IPV was reported by women in Kerewan (7.71%) in the GDHS 2013 and in Basse (17.46%) in the GDHS 2019-20. While the highest proportions were reported from Mansakonko (24.94%) in the GDHS 2013, and in Janjanbureh (36.00%) in the GDHS 2019-20. For physical IPV, the lowest and highest proportions were reported from Kerewan (11.09%) and Mansakonko (27.83%), respectively in the GDHS 2013. While in the GDHS 2019-20, the highest and lowest proportions were reported from Janjanbureh (42.82%) and Kerewan (22.77%), respectively. For sexual IPV, the highest and lowest proportions in GDHS 2013 reported from Kerewan (0.54%) and Janjanbureh (6.10%), respectively. In GDHS 2019-20, the highest and lowest proportions were reported from Janjanbureh (9.68%) and Kanifing (1.75%), respectively.
Proportions and Trend Analyzes for Emotional, Physical, and Sexual Violence by Local Government Area—Gambia Demographic and Health Surveys (DHS) 2013 and 2019-20.
CI = Confidence Interval.
Trend analysis was done using logistic regression models.
With the exception of sexual IPV in Kanifing, all 3 types of IPV increased from GDHS 2013 to GDHS 2019-20 in all 8 regions. However, the statistically significant linear increase in prevalence of various types of IPV were across the period of 2013 to 2019-20 in the 8 regions were for: emotional IPV in Banjul, Kerewan, Kuntaur, and Janjanbureh; for physical IPV in Brikma, Kerewan, Janjanbureh, and Basse; for sexual IPV in Brikma, Kerewan, and Basse.
Discussion
The objectives of this secondary analysis of Gambia DHSs 2013 and 2019-20 were to glean better appreciation of the various social, demographic, and attitudinal, beliefs, and experiences that were associated with IPV against ever-married women perpetrated by men with whom they were having intimate relationship. An associated objective was to determine the trend and statistical significance of IPV experiences between the 2 DHSs from 2013 and 2019-20 for each of the 8 regions in terms of emotional, physical, and sexual IPV. This is the first study computing the IPV linear trends by region in Gambia.
Over one-third of women (39.23%) reported having experienced one or more types of IPV committed by their current husband/partner for currently married women and the most recent husband/partner for the divorced, separated, or widowed women, based on the most recent nationally representative GDHS2019-20. The most common type of IPV reported was physical, with 29.09%, followed by emotional (24.03%) and sexual (5.52%) IPV. Physical IPV being the most common type of IPV has also been previously reported in other studies using the nationally representative DHSs data.29,30 The prevalence of sexual and/or physical IPV was 31.00%. This is slightly lower than the 33% that was reported by the WHO for the African region, and 4% points higher than the 27% globally. 7 However, the WHO figures are based on ever married/ partnered women aged 15 to 49 years who experienced either or both of these 2 types of IPV at least once in their lifetime. 7 Contrasting with this analysis that pertained to IPV perpetrated by the current or most recent husband/partner in the GDHS2019-20.
Concerning the respondents social, demographic and other studied attributes, over one-third (41.37%) of respondents were under the age of 30 years; almost 45% had no formal education while about 58% respondents husband/partner also did not have any formal education; almost 30% did not work; 38.35% belonged to the poorer or poorest wealth quantile; almost 30% were rural residents; about 13% did not have living child; only 30% or so participated in the decision making; about 46% believed that IPV was not justified; less than 2% women reported that their husband/partner used alcohol; almost 90% had never witnessed their father physically beat their mother; and about 43% reported that their husband/partner displayed marital control.
Bivariate analyzes results, in terms of the association between IPV and the 12 explanatory variables revealed statistical significance of 8 such covariates. While occupational status, residential status, participation in decision making, and partner’s alcohol use were not found to be statistically significantly associated with IPV; and were not included in the final multivariable model. Lack of IPV association with occupational status was also reported in a recent study from Liberia using the 2019-20 DHS data. 29 Lack of association with residency status was unexpected, as several studies report higher association of rural residency status with IPV.3,4 While other studies in Liberia and Afghanistan also did not report statistically significant association of IPV with urban/rural residency status.29,30 These findings of IPV’s lack of associations with either participation in decision making or occupational status of women, suggest deeply entrenched sociocultural norms and mores acting as conduit for the acceptance and practices of IPV transcending urban/rural residency status. 29 Against the backdrop of merely 1.81% prevalence of alcohol use by male partners, it is unsurprising that this attribute was not found to be statistically significant. However, the association of IPV with use of alcohol by partner is one of the strongest associations that has been reported.3,16,29,31
The final multivariable model included the remaining 8 covariates; with 3: respondents age, number of living children, and acceptance of IPV were not statistically significantly associated with IPV. Contradictory evidence has been reported for the association of age with IPV; younger women were found to have higher odds of IPV in several other countries.2,23,30 While other studies have reported higher IPV association with older age or no significant association with age, as in this study.29,32 Plausible explanations include older women being more economically dependent on their husband, thus even less empowered, while younger women being less able to negotiate marital disputes. Association between children and IPV is also contradictory; higher IPV prevalence rates have been reported in infertile women 5 ; low IPV rates in women with children have been reported from Liberia 29 ; while higher rates have been reported from Afghanistan Peru in women with children.30,31 Possible explanations include infertile women not being able to meet familial and social expectations, while women with more children and resultant increase in household workload leading to marital arguments. Finally, regarding acceptance of IPV; increased IPV has been associated with acceptance of IPV and wife-beating attitudes in several countries.2,29,30 Acceptance of IPV by women could be construed as a barometer of the societal acceptance of IPV in general, and low empowerment of women in particular. IPV acceptance might also as a self-fulling prophecy in such social milieus.
The 5 covariates found to be statistically significantly associated with IPV in this study were respondents and their partners educational attainment, household wealth, having witnessed one’s father beat mother, and marital control. The strongest association with IPV was for the covariate of marital control that is, odds were 4.34 times higher for IPV when husband exhibited marital control; this association of controlling behavior by husband/partner with IPV has been previously reported from 9 African, Latin American, and Asian countries. 33 Predilection for control expressed in terms of toxic possessiveness and dominance to assert marital power is a morbid segues to IPV. Educational attainment of women and their partners were both statistically significantly associated with IPV in the multivariable model. Compared with women with no education, those who had higher than secondary education had 70% lower odds of reporting IPV. Ostensibly, education bestows more sense of caring and fairness in one’s interactions, respect for other people’s rights, and freedoms; the attributes catalyzing women empowerment. While those women whose partners had primary and higher than secondary education were more likely to report IPV. Several studies have reported the association of low or no educational attainment of women or their partners, with higher IPV levels.3,30,31 Association of higher educational attainment in partners with higher reporting of IPV in women is rather perplexing, and perhaps points toward pervasiveness of societal norms averse to women empowerment. In general, IPV rates are higher in low-income countries, and in women from poorest wealth quantiles.2,4 Similarly, compared to the lowest/poorest wealth quantile, women in the richest quantile reported lower IPV. Plausible explanations of this association would be more autonomy, economic independence, and empowerment of women in well off households with resultant less reliance on their husbands. Finally, the most robust association of IPV has been reported with having witnessed one’s father physically beat mother24,29-31,33; and the same was found in this study. Growing up in a household that normalized physical violence by witnessing one’s own father physically beating one’s mother, ostensibly makes it an element of intimate relationships. Thus the misguided attitudes of women pertaining to IPV, coupled with having witnessed IPV in their own families, while growing up; perhaps preps these women into acceptance of IPV and mire them into the morass of intergenerational IPV.9,10
Hence, the pernicious recipe of illiteracy/low educational attainment and poverty in women ultimately results in economic dependence on their intimate partner, with resultant economic power imbalance; coupled with physical power imbalance renders some women susceptible to IPV. This is probably even truer when the societal norms do not frown up IPV against women. A recent meta-analysis on the effectiveness of housing/shelter interventions for the women victims of IPV demonstrated its effectiveness. 34 While better educated women tend to be more averse to passively accepting IPV. 35
Regarding linear trend analysis in the 8 regions in Gambia, from 2013 to 2019-20, a dismal picture emerged showing a statistically significant linear increase in the prevalence of all 3 forms that is, emotional, physical, and sexual IPV. The only 2 exceptions were decrease in emotional IPV in the Brikama region and decrease in sexual IPV in Kanifing region, across the period from 2013 to 2019-20; albeit none was statistically significant. However, statistically significant increase was found in emotional and physical IPV in 4 regions each, out of the total of 8 regions. While sexual IPV was statistically significantly associated with 3 regions. Lack of statistical significance for increase in various types of violence in other regions need to be interpreted with the caveat of weighing statistical significance versus practical significance.
All studies are fraught with strengths and limitations and this secondary analysis is no exception. The primary strength of this study is its national and regional representativeness; use of a validated questionnaire; and the generalizability of the results to the target population of the GDHSs. However, the major limitations include the fact that women older than 49 were not included as well as women who were never married. Hence results do not necessarily reflect the true burden of IPV in Gambia. Secondly, the self-reported survey instrument for IPV and its correlates was used, this might have resulted in the under reporting of IPV burden due to cultural sensitivities, feelings of shame or guilt that is, social desirability bias; including susceptibility to recall bias. Finally, the inherent limitation of every cross-sectional survey like GDHS 2019-20 is the inability to determine causality, as only associations between IPV and the covariates could be studied.
Conclusion
The most disturbing finding of this study is that with one regional exception, the 3 types of IPV prevalence increased over the course of 5 years from 2013 to 2019-20 in all regions of Gambia; albeit this increase was not statistically significant in each and every instance of IPV. In Gambia, the prevalence of physical and/or sexual IPV in the ever-married women perpetrated by the current or most recent husband was higher compared to global prevalence but slightly lower from the WHO AFRO region. Findings underscore the need for women empowerment by addressing the socio-cultural norms and crafting of social, legal, and health policies to improve the health and human rights status of women in Gambia in every region of the country. The regional disparities augur the need for targeted health education, promotion, and women empowerment interventions that address the health, social and economic wellbeing of women in Gambia.
Footnotes
Declaration of Conflicting Interests
The author declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author received no financial support for the research, authorship, and/or publication of this article.
Ethical Approval
The GDHS 2019-20 was conducted by the Gambia Bureau of Statistics in collaboration with the Gambian Ministry of Health; with technical support provided by the ICF International Inc., United States. The institutional review board of the Gambian Government/Medical Research Council Joint Ethics Committee in Gambia gave the approval for the conduct of survey; additionally, the ICF institutional review board also provided the ethical approval.
My study did not require an ethical board approval because the results of secondary analysis are reported in this paper using anonymized data, as such no ethical approval was required.
