Intimate partner violence among prenatal care attendees amidst the COVID-19 crisis: The incidence in Ethiopia.

OBJECTIVE: To assess the incidence and predictors of intimate partner violence (IPV) during pregnancy amidst the coronavirus disease 2019 pandemic.
METHODS: This cross-sectional study was conducted at the prenatal care clinic of St. Paul's Hospital Millennium Medical College (SPHMMC), Addis Ababa, Ethiopia, among pregnant women from 31 August to 2 November 2020. Participants were interviewed using Open Data Kit. Logistic regression was used to assess predictors.
RESULTS: Among the 464 pregnant women, 33 (7.1%) reported IPV during pregnancy, and among these 24 (72.7%) reported emotional violence, 16 (48.5%) reported sexual violence, and 10 (30.3%) reported physical violence. Among the study participants, only 8 (1.7%) were screened for IPV. IPV was reported 3.27 times more often by women who reported that their partner chewed Khat compared with those women whose partner did not (adjusted odds ratio [aOR] 3.27; 95% confidence interval [CI] 1.45-7.38), and 1.52 times more often women who reported that their partner drank alcohol compared with those women whose partner did not (aOR 1.52; 95% CI 1.01-2.28).
CONCLUSION: Very few women were screened for IPV. Partners drinking alcohol and chewing Khat are significantly positively associated with IPV during pregnancy. IPV screening should be included in the national management protocol of obstetric cases of Ethiopia.

INTRODUCTION

Intimate partner violence (IPV) against women is a major public health problem. , According to WHO, “IPV may include physical aggression, sexual coercion, psychological abuse, and/or controlling behaviors by partners”. Purposeful use of physical force is considered “physical violence,” while forcing a woman to engage in a sexual act is referred to as “sexual violence”. “Emotional,” “psychological,” or “verbal” violence include threats, humiliation, control of activities, isolation, name calling, and attempts to frighten. IPV is used to describe what has in the past been referred to as “woman abuse,” “woman battering,” or “domestic violence”.

Worldwide, one‐third of women who have been in a relationship have experienced physical and/or sexual violence by their intimate partner. As many as 38% of all murders of women are committed by intimate partners. Although anyone can be the victim of IPV, most abuse victims are women and in most violent relationships, multiple forms of violence are experienced.

Many pregnant women are abused by their partners and abuse may start or get worse during pregnancy. , Some studies, especially hospital‐ and clinic‐based studies, have found an increase in IPV during pregnancy, but population‐based studies have found that pregnancy is not a risk factor, and might be protective against IPV. There are some who feel that population‐based studies underestimate the proportion of women impacted by IPV because of how IPV is assessed.

The coronavirus disease 2019 (COVID‐19) crisis has worsened situations for many women in abusive relationships ; there was an immediate and marked increase in domestic homicides in several countries. In March 2020, Spain (a country that has been particularly hard hit by the pandemic) witnessed its first domestic violence fatality just 5 days after the lockdown; a woman was murdered by her husband in front of their children in Valencia. There is also emerging evidence of an increased number of domestic homicides in the UK since the lockdown restrictions were established.

This pandemic has created a paradox regarding staying safe by staying at home. While citizens have been called upon to stay at home in order to stop the spread of COVID‐19, a critical analysis of what this means for women in abusive relationships is needed.

According to a study performed in six European countries, pregnant women who experience IPV might benefit from being identified in the prenatal care setting and being given specialized care. Pregnant women perceived prenatal care as a vital time to report IPV. Pregnant women experiencing IPV responded positively to IPV screening and, based on these and other findings, it is recommended that IPV screening should be included in the prenatal care services package. , Studies in Ethiopia point to a high proportion (58%–59%) of pregnant women being victims of at least one form of domestic violence during pregnancy. ,

Although there have only been a few studies in Ethiopia, these studies have focused on the prevalence and associated factors of IPV. , The COVID‐19 crisis is unprecedented, so the impact of the pandemic on IPV during pregnancy has not yet been studied in this setting. The present study aimed to assess the incidence and predictors of IPV during pregnancy amidst the COVID‐19 pandemic.

METHODS

This cross‐sectional study was conducted at the prenatal care clinic of Saint Paul's Hospital Millennium Medical College (SPHMMC), in Addis Ababa, Ethiopia, among pregnant women from 31 August to 2 November 2020. Ethical approval of and clearance for the study was obtained from the Ethical Review Board of SPHMMC before the start of data collection. Informed written consent was obtained from all women who were interviewed. To protect those women who were found to be victims of violence during our interview, women who screened positive for IPV were given counseling using a modified version of the American College of Obstetricians and Gynecologists’ patient information document and those women who also had symptoms of depression and stress were linked to the psychiatric clinic for counseling and treatment. We linked one pregnant woman to a non‐governmental organization, which gives safe house and financial support to victims of IPV.

SPHMMC is the second‐largest teaching hospital in Addis Ababa, Ethiopia. On average, 100 women visit the prenatal care clinic on working days and the clinic is open starting from 2:30 am to 6:30 am and 1:30 pm to 5:30 pm. Prenatal care is given by obstetrics and gynecology residents.

The sample was calculated using a single population formula, n = Z 2 p(1–p)/w 2, where Z is Z score, p is percentage, and w is width (precision). According to a study in Ethiopia, 59% of pregnant women report IPV. This proportion was taken to calculate the sample size. Assuming a precision of 5% and 95% confidence limits, a sample size of 372 was calculated. A total of 464 women were interviewed. Data were collected by Bachelor of Science nurses and every other prenatal patient who visited the regular prenatal clinic in their third trimester (at or beyond 7 months) was invited to participate in exit interviews. Data were collected using open data kit (ODK) exported to and analyzed using Stata Statistical Software: release 14 (StataCorp LP, College Station, TX, USA). Pregnant women in their third trimester were included because they were pregnant for 1–3 months before the COVID‐19 outbreak and it is assumed that their perceived level of IPV before and after the COVID‐19 outbreak could be compared. Women in their first or second trimester of pregnancy, women who had obstetric emergency conditions (women in their third trimester with vaginal bleeding, rupture of fetal membranes, severe pregnancy‐related hypertension, and those in labor) who needed transfer to the emergency obstetric outpatient clinic were excluded.

The WHO multi‐country study on women's health and domestic violence against women questionnaire was used to assess IPV. The questionnaire was adapted, translated into Amharic (the local language), and back‐translated into English to ensure consistency and conceptual equivalence. The response to each item is either “Yes” or “No”. The survey asks six yes/no questions about physical IPV, two yes/no questions about sexual IPV, and four yes/no questions about emotional IPV. Accordingly, participants who respond “Yes” to any of the questions are considered as incident cases of IPV victimization. This is used as the outcome variable in all analyses.

Participants who reported IPV were further asked if this violence had changed since the beginning of the COVID‐19 pandemic. The possible responses were “Got worse”, “Got less” and “Remain about the same”. Those women who responded “Got worse” were categorized as “Yes” and the rest of the responses were categorized as “No” to perceived increase in IPV.

The predictor variables are based on a “Yes” or “No” response to whether or not the woman's partner drinks alcohol, smokes cigarettes, chews Khat (a locally grown stimulant leaf), and whether the woman or her partner lost a job after COVID‐19. The authors added these predictor variables to the questionnaire, which was adapted from the WHO multi‐country study on women's health and domestic violence against women because they have been found to be predictors of IPV in other studies. Questions related to COVID‐19 were included by the authors to determine the impact of the pandemic on IPV in pregnancy in this population.

RESULTS

During the study period, 464 prenatal clients were interviewed. The sociodemographic characteristics of the participants are shown in Table 1. The mean (±SD) age is 28.1 (4.8) years and 219 (47.2%) women were in the age range 26–31 years. In all, 418 (90.1%) women had attended school and 219 (52.4%) had primary education. A total of 309 (66.6%) women were Christians, 319 (68.7%) had one or no children, and the mean (±SD) number of children that the women had was 1.1 (1.0). Most (328; 70.7%) were unemployed and the mean family income was US$ 28.6 per month. The majority of the study participants (325; 70.0%) lived in the capital city (Addis Ababa).

TABLE 1

Sociodemographic characteristics of prenatal care attendees at St. Paul's Hospital Millennium Medical College

Characteristics a	n	%
Age, years
14–19	11.0	2.4
20–25	135.0	29.1
26–31	219.0	47.2
≥32	99.0	21.3
Mean ± SD	28.1 ± 4.8
Educated
No	46.0	9.9
Yes	418.0	90.1
Educational level (n = 418)
Primary (up to Grade 8)	219.0	52.4
High school	131.0	31.3
Diploma and above	68.0	16.3
Religion
Muslims	155.0	33.4
Christians	309.0	66.6
Parity (mean ± SD)	1.1 ± 1.0
Low parity (≤1)	319	68.7
Multipara (≥2)	145	31.2
Occupation
Not employed	328.0	70.7
Daily laborer	11	2.7
Employed	125.0	26.9
Monthly income, US$
Mean ± SD	28.6 ± 99.8
Living area
Addis Ababa	325.0	70.0
Urban	129.0	27.8
Rural	10.0	2.2

n = 464 unless mentioned otherwise.

Sociodemographic characteristics of the partners of the study participants are shown in Table 2. The mean age of partners was 34.0 years and 242 (52.2%) were between 31 and 40 years of age. Most (421; 90.7%) had attended school and 164 (39.0%) completed high school up to grade 12. The majority (343; 73.9%) were employed, and their mean income was US$ 103.4 per month.

TABLE 2

Sociodemographic characteristics of partners of prenatal care attendees at St. Paul's Hospital Millennium Medical College

Characteristics a	n	%
Partner age, years
20–30	168.0	36.2
31–40	242.0	52.2
≥41	54.0	11.6
Mean ± SD	34.0 ± 6.2	100.0
Educated
No	43.0	9.3
Yes	421.0	90.7
Educational level
Elementary education	160.0	38.0
Grade 10 and Technic school	164.0	39.0
Diploma and above	97.0	23.0
Occupation
Unemployed	6.0	1.3
Employed	343.0	73.9
Daily laborer	115.0	24.8
Income, US$ (mean ± SD)	103.4 ± 112.3

n = 464 unless mentioned otherwise.

Among the 464 pregnant women, 69 (14.9%) had life‐time IPV exposure, and 44 (9.5%) had IPV exposure within the year of the interview. Furthermore, 33 (7.1%) reported IPV during the pregnancy under study. Among these 33 women, 24 (72.7%) experienced emotional violence, 16 (48.5%) experienced sexual violence, and 10 (30.3%) experienced physical violence (Table 3).

TABLE 3

Intimate partner violence among prenatal care attendees at St. Paul's Hospital Millennium Medical College

Characteristics a	n	%
Lifetime
No	395.0	85.1
Yes	69.0	14.9
Physical violence (n = 69)
No	37	53.6
Yes	32.0	46.4
Sexual violence (n = 69)
No	49	71
Yes	20.0	29.0
Emotional violence (n = 69)
No	17	24.6
Yes	52.0	75.4
In this year
No	420.0	90.5
Yes	44.0	9.5
Physical violence (n = 44)
No	30.0	68.2
Yes	14.0	31.8
Sexual violence (n = 44)
No	27.0	61.4
Yes	17.0	38.6
Emotional violence (n = 44)
No	10.0	22.7
Yes	34.0	77.3
During current pregnancy
No	431.0	92.9
Yes	33.0	7.1
Physical violence (n = 33)
No	23.0	69.7
Yes	10.0	30.3
Sexual violence (n = 33)
No	17.0	51.5
Yes	16.0	48.5
Emotional violence (n = 33)
No	9.0	27.3
Yes	24	72.7

n = 464 unless mentioned otherwise.

Among the 44 pregnant women who experienced IPV in year of the interview, 9 (20.4%) reported an increase in IPV after pregnancy (Table 4) and 8 (18.2%) perceived increased IPV after COVID‐19 outbreak (Table 5).

TABLE 4

Perceived increment in intimate partner violence during pregnancy among prenatal care attendees at St. Paul's Hospital Millennium Medical College

Character	n	%
Intimate partner violence a (n = 44)
No	35.0	79.5
Yes	9.0	20.5
Physical violence (n = 9)
No	7.0	77.8
Yes	2.0	22.2
Sexual violence (n = 9)
No	7.0	77.8
Yes	2.0	22.2
Emotional violence (n = 9)
No	4.0	44.4
Yes	5.0	55.6

In 2020.

TABLE 5

Perceived increment in intimate partner violence after COVID‐19 outbreak among prenatal care attendees at St. Paul's Hospital Millennium Medical College

	n	%
Intimate partner violence a (n = 44)
No	36.0	81.8
Yes	8.0	18.1
Physical violence (n = 8)
No	6.0	75.0
Yes	2.0	25.0
Sexual violence (n = 8)
No	6.0	75.0
Yes	2.0	25.0
Emotional violence (n = 8)
No	4.0	50.0
Yes	4.0	50.0

In 2020.

Among the 464 participants, only 8 (1.7%) were screened for IPV at the prenatal clinic.

In bivariate analysis, those women who made decisions with their husband on household issues had an 81% reduced likelihood of IPV during pregnancy than women who made decisions by themselves (crude odds ratio [cOR] 0.19; 95% confidence interval [CI] 0.07–0.59; P = 0.004). IPV during pregnancy was 3.36 times higher among women who reported that their partner drank alcohol compared with those women whose partner did not (cOR 3.36; 95% CI 1.64–6.91; P = 0.001), and IPV was 3.22 times higher among women who reported that their partner chewed Khat compared with those women whose partner did not (cOR 3.22; 95% CI 1.51–6.86; P = 0.002) (Table 6). Those variables with a P value below 0.2, as well as variables deemed important to control for by the study team, were included in the multivariable analysis.

TABLE 6

Bivariate logistic regression analysis of predictors’ of intimate partner violence among prenatal care attendees

Variables	Crude odds ratio (95% CI)	P value
Decision made together a	0.19 (0.07–0.59)	0.004
Partner drinks alcohol	3.36 (1.64–6.91)	0.001
Partner chews Khat	3.22 (1.51–6.86)	0.002
Woman lost job b	1.18 (0.39–3.52)	0.762
Planned pregnancy	0.44 (0.18–1.08)	0.073
Wanted pregnancy	0.33 (0.91–1.23	0.100
Source of income c	0.23 (0.04–1.35)	0.104
Partner lost job b	0.75 (0.33–1.72)	0.508
Partner stays at home b	0.55 (0.25–1.21)	0.136
Partner smokes cigarettes	3.33 (0.36–30.7)	0.288
Parity	1.93 (0.94–3.94)	0.072

On household issues.

After COVID‐19 outbreak.

From both partners.

In multivariable analysis, the factors associated bivariately continued to show significant association; women who made decisions with their husbands on household issues were 81% less likely to report IPV during pregnancy compared with women who made decisions by themselves (adjusted OR [aOR] 0.19; 95% CI 0.06–0.62; P = 0.006). IPV was 3.27 times more likely among women chewed Khat (aOR 3.27; 95% CI 1.45–7.38; P = 0.004), and 1.52 times higher among women who reported that their partner drank alcohol (aOR 1.52; 95% CI 1.01–2.28; P = 0.046) (Table 7).

TABLE 7

Multivariable logistic regression analysis of predictors’ of intimate partner violence among prenatal care attendees

Variables	Adjusted odds ratio (95% CI)	P value
Decision made together a	0.19 (0.06–0.62)	0.006
Partner chews khat	3.27 (1.45–7.38)	0.004
Partner drinks alcohol	1.52 (1.01–2.28)	0.046
Partner lost job b	0.55 (0.23–1.32)	0.184
Partner smokes cigarettes	1.39 (0.13–14.95)	0.781
Woman lost job b	0.83 (0.26–2.67)	0.759

On household issues.

After COVID‐19 outbreak.

DISCUSSION

Among the 464 prenatal care attendees, only 8 (1.7%) were screened for IPV at the prenatal clinic. Thirty‐three (7.1%) had experienced IPV during pregnancy. Pregnant women whose partners drank alcohol and chewed Khat were significantly more likely to report experiencing IPV, whereas women who made decisions on household issues with their husbands had a reduced risk of experiencing IPV.

In this investigation, the incidence of IPV during pregnancy was lower than the incidence reported in other studies conducted in Ethiopia. , This could be because these other studies were performed in rural parts of Ethiopia where IPV is more common compared with the capital city, and the majority of our study participants were residents in the capital city. According to a systematic review of IPV during pregnancy in Ethiopia, women from rural regions were found to have a higher rate of IPV than women from Addis Ababa.

However, our finding of a lifetime IPV victimization rate of 14.9% is much lower than the 29% rate of IPV reported in Addis Ababa. The possible reason for our observed lower incidence of IPV is that there has been 10 years between our study and the previous study conducted in Addis Ababa. Additionally, even if the study participants were interviewed in Addis Ababa, many may have lived in a rural part of the country, as residence was not reported in the previous study.

We attempted to measure the increase in IPV after pregnancy and after the COVID‐19 outbreak by including women who were pregnant for 1–3 months before COVID‐19 so that they could compare the rate of IPV before and after the outbreak as well as before and after pregnancy. At the time of the COVID‐19 outbreak, reports showed an increase in domestic homicides in several affected countries since the lockdown restrictions were put into place. , Similarly, in the present study, 18.2% of pregnant women reported perceived increase in IPV after the COVID‐19 outbreak.

Prenatal care is an ideal time to screen women for IPV, as women can be connected with services to protect them and their babies. All women should be screened during prenatal care for IPV. , , , However, in the present study very few women were screened for IPV, which may be because IPV screening is not included in the prenatal care section of the national management protocol for obstetrics cases.

In the present study among pregnant women who experienced IPV during pregnancy, the most common form of violence was emotional violence, impacting 72.7% of the victims of IPV. Similarly, in another study in Ethiopia, emotional violence was the commonest form of violence during pregnancy.

Pregnant women who have a partner who drinks alcohol have a significantly increased risk of IPV during pregnancy. This finding is similar to others studies conducted in Ethiopia where the partner's alcohol consumption was associated with an increased odds of IPV during pregnancy. , Further, pregnant women whose partner chewed Khat were more likely to experience IPV during pregnancy. This finding was similar to the finding of another study in Ethiopia. Both of these findings suggest an association between substance use and violent behavior.

Women who make decisions together with their husband have reduced odds of experiencing IPV during pregnancy. This might be a sign of a healthier, violence‐free, relationship, where women feel free to discuss household matters with their partners. For partners who work together to make decisions, there are probably other aspects of their relationship that are healthier.

Surprisingly, neither the woman nor the husband's loss of job after the COVID‐19 outbreak was significantly associated with IPV during pregnancy.

A limitation of the present study is that there was overlap between COVID‐19 and pregnancy, and it may be difficult to differentiate whether the perceived increase in IPV is due to COVID‐19 or pregnancy, or a combination of the two. Both may be stresses on a relationship that could increase the risk of IPV. We attempted to address this by including those women who had been pregnant for 1–3 months before the COVID‐19 outbreak so that they were able to compare the IPV before and after the COVID‐19 outbreak.

In conclusion, very few women were screened for IPV during pregnancy, a finding that has implications for policy moving forward. Given the relatively high prevalence of IPV in this population, we recommend that all clients are screened for IPV so that they can be directed towards services to protect women and their babies from violence. Having partners who drink alcohol and chew Khat is significantly associated with increased risk of IPV during pregnancy. IPV screening should be included in the national management protocol of obstetric cases of Ethiopia.

CONFLICTS OF INTEREST

The authors have no conflicts of interest.

AUTHOR CONTRIBUTIONS

AT contributed to the study planning, design, data collection and analysis, and writing of the manuscript. WG contributed to the study planning, data analysis, and editing of the manuscript. DB contributed to the study planning and editing of the manuscript. MA contributed to study planning and editing of the manuscript. RE contributed to study planning and supervision of the data collection process and training of data collectors. SDC contributed to data analysis and editing of the manuscript.