The Economic Burden of Intimate Partner Violence in Colombia: Estimated Health Costs Among Females Aged 13

Abstract

Background

Intimate partner violence (IPV) is the most pervasive form of gender-based violence, often first experienced in adolescence. While the prevalence of IPV is known to be exacerbated in humanitarian settings, little is known in regard to the economic burden of IPV between conflict-affected and non–conflicted-affected groups of women and girls. This top-down costing study examines the total health burden of physical IPV in Colombia, and whether these costs differ by conflict exposure.

Methods

We utilized a nationally representative sample of 13–24-year-old females from the Violence Against Children Surveys (VACS) in Colombia conducted in 2018. Using physical IPV prevalence, the analysis was conducted in four steps: 1) estimate the relative risk of seven IPV-associated health outcomes among the sample and subgroups, 2) estimate the population attributable fraction of IPV for each health outcome, 3) quantify the burden of IPV in disability-adjusted life years (DALYs), and 4) assign health costs in US dollars to the estimated DALYs.

Results

We found that the single year health burden associated with physical IPV was $90.6 million USD. Moreover, nearly 40% of the economic burden of physical IPV among females aged 13–24 in Colombia was from those who were conflict-affected (24%).

Conclusion

Our findings demonstrate that at least 16% of the overall health costs among females 13–24 in Colombia is from the preventable epidemic of physical IPV. In order to prevent and mitigate the costs of gender-based violence, multi-lateral and government investment is critically needed to prevent IPV and support women and girls.

Keywords

Colombia economic burden costs intimate partner violence conflict adolescent girls

Introduction

Intimate partner violence (IPV)—defined as physical, sexual, and psychological violence perpetrated by a current or former partner—is the most pervasive form of gender-based violence across the globe (WHO, 2021). With an estimated 24% of ever-married or partnered girls between the ages of 15–19 having experienced physical and/or sexual violence from an intimate partner (WHO, 2021), IPV is often first experienced during adolescence (Kidman & Kohler, 2020; Stöckl et al., 2014). Factors such as weakened health care infrastructure, diminished rule of law, conflict, and forced migration magnify IPV risk factors in humanitarian settings (IASC, 2019).

Not only is IPV widespread, but it has considerable and interconnected impacts across the ecosystem. Recognizing that the prevalance of IPV is often exacerbated in conflict-affected settings, a recent systematic review was designed to examine the unique consequences of IPV in humanitarian settings (Meinhart et al., 2021). This review found that evidence was robust when examining associations between IPV in humanitarian settings mental health, and substance abuse. Physical health outcomes of the review highlighted sexual and reproductive health outcomes, including pregnancy and sexually transmitted infections (STIs). While the review demonstrated that violence perpetration is less commonly examined in humanitarian settings (Meinhart et al., 2021), there is evidence linking IPV victimization and perpetration in humanitarian settings (Kayibanda & Alary, 2020). These myriad health impacts can result in women and girls’ further diminished participation in society during and post conflict, including in the labor force (Riger & Staggs, 2004; Siddique, 2018). While yet to be examined in humanitarian settings, the financial losses resulting from IPV in low- and middle-income countries (LMICs) are similarly borne across individuals, households, businesses, societies, and governments (Vyas et al., 2021). For example, at the individual level, medical expenses for physical and mental harms are accrued; at the household level, there may be income losses from lost work or property damage; businesses also accrue costs of absenteeism, tardiness, and lost productivity associated with victimization; and governments incur costs of providing violence prevention and response services, potentially crowding out other budgetary considerations (University of Limerick and NUI Galway, 2019; Vyas et al., 2021).

To address IPV and its associated social and economic impacts, it is necessary to understand the prevalence, intersectional risk factors, and root causes of violence perpetration and victimization. The 2018 Colombia Violence Against Children and Youth Survey (VACS) is the first national survey to sample participants from conflict-affected and non–conflict-affected regions, examine the national prevalence of IPV among adolescents and young adults ages 13–24, and measure the multi-level consequences of IPV (Government of Colombia, Ministry of Health and Social Protection, 2019). Thus, the Colombia VACS offers a unique opportunity to conduct novel analyses examining the cost of inaction—or maintaining the status quo—associated with IPV among adolescents and young adults in conflict-affected as well as non-conflict-affected regions of a LMIC.

Over the past decade, there has been an increased demand for providing evidence on the economic burden of IPV to inform investments in prevention and response programming. Such estimates also highlight the burden of IPV, which is has profound national and international costs (societal costs), not simply a private cost. Despite the recognized value of these costings, there is a paucity of peer-reviewed publications examining IPV in LMICs and no peer-reviewed publication examining the unique costs in conflict-affected settings (Vyas et al., 2021). Within the gray literature, one promising recent report found that annual out-of-pocket expenditures averaged $21 USD among women who experienced violence in South Sudan (University of Limerick and NUI Galway, 2019). Such a figure is substantial given that the majority of the South Sudanese population survives on less than $1 USD per day.

Building on the limited evidence, the present study aimed to estimate the economic burden of IPV in Colombia with an acute interest in the costs accrued among women and girls affected by conflict. The study makes an economic case for governments and donors to invest in the elimination of IPV, aligning with Sustainable Development Goal target 5.2 (UNGA, 2015). Women living conflict-affected LMICs face the highest rates of violence (Stark & Ager, 2011), and thus such settings face profound economic burdens as a result of IPV (Matzopoulos et al., 2008). We hypothesized that the health burden of IPV would disproportionately impact women and girls affected by conflict. The present analysis is the first to estimate the health cost of IPV in Colombia between conflict- and non–conflict-affected groups of female youth ages of 13–24 years old. We advance the literature pertaining to IPV costing both methodologically and topically by employing a top-down method and focusing specifically on a LMIC affected by conflict. These estimates provide practitioners, policy makers, and funders a clearer understanding of the economic burden of inaction—that is, the cost associated with maintaining the current state of policies and programming and failing to fully address physical IPV against girls and women ages 13–24 in a single year; these costs could be avoided were IPV to be prevented.

Setting

The longest-running civil war in the Americas was intended to end with the signing of a 2016 peace deal; however, protracted conflict in Colombia persists. A total of 32,569 deaths have occurred as a result of the internal conflict in Colombia, with a nearly 30% increase in violence in the 4 years since signing the peace deal compared to the 4 years prior (UCDP, n.d.). Despite the $30.3 million USD funding for the 2021 humanitarian responses plan, there remains an 82.6% gap in funding (OCHA FTA, 2021). Only $12.4 million have been allocated for the protection cluster/sector, which largely oversees programming to address gender-based violence.

Policies within Colombia have made important steps toward promoting gender equality and addressing gender-based violence including 2012’s Public Policy Guidelines for Women’s Gender Equality which aimed for women to have a life free from violence and 2014’s Law 1719 which improved access to justice for survivors of sexual violence. Moreover, Colombia has integrated considerations to address gender-based violence in its National Development Plan 2018–2022 and Ten-Year Justice Plan. Despite these important achievements, significant gaps remain, as an estimated 30% of women in Colombia have experienced physical or sexual intimate partner violence during their lifetime (WHO, 2021).

At around 10% of the population, females ages 13–24 years old represent 4.83 million out of the total Colombian population of 47.6 million (10%) (Global Burden of Disease Collaborative Network, 2021). Moreover, children represent 2.1 million of the 6.7 million people (30%) in need of humanitarian assistance in Colombia (OCHA, 2021).

Methodology

Data Source

In recognition that IPV is often first experienced during adolescence, this study prioritized data that could enable representative findings among young women and adolescent girls. The primary data for this costing come from the national 2018 Colombia VACS. The Colombia VACS was conducted through a collaboration between the Colombian Ministry of Health and Social Protection, UNICEF, the US Centers for Disease Control and Prevention (CDC), and others. Having been implemented in 24 countries, the VACS collect data among 13–24-year-old females and males on witnessing, perpetrating or experiencing violence (physical, emotional, and sexual), mental health, gender attitudes, as well as individual and household characteristics.

Collected through a three-stage split-sampling approach, the Colombia VACS first identified 619 enumeration areas (EAs) using the national sampling frame. EAs were split, so that approximately half of the EAs were used to collect data from female respondents and the other half from males in order to protect confidentiality and promote safety. Next, a total of 24 households in each EA were selected. Finally, one female or male aged 13–24 per household was randomly selected to complete the interview. Informed consent was provided by all participants 18 and older, while informed assent was obtained by each participant under age 18 with written consent provided by a caregiver. Ethics approval was provided by the Ethics and Research Methods Committee of the National Institute of Health of Colombia and the CDC’s Institutional Review Board (Government of Colombia, Ministry of Health and Social Protection, 2019).

With 6900 females screened for eligibility, the VACS sample included 1406 females for the national dataset. Sub-group analysis based on conflict exposure (n = 291) or no conflict exposure (n = 1115), to examine if the magnitude and significance of the associations between IPV and health outcomes varied based on conflict exposure. All analyses were conducted using Stata SE version 15.

Measurement

Our analysis examined the impacts of lifetime physical IPV as the primary independent variable. Physical IPV was a binary variable based on the responses to a probe asking whether a boyfriend/romantic partner, ex-boyfriend/romantic partner, or husband ever perpetrated physical violence against the responded. Four questions asked about various types of physical violence: slapping, pushing, shoving, shaking, or intentionally throwing something; punching, kicking, whipping, or beating with an object; choking, smothering, trying to drown, or intentional burning; and threatening with a knife, gun, or other weapon. Respondents who answered “yes” to one or more of these forms of physical IPV were coded as experiencing physical IPV; otherwise, respondents were coded as not experiencing physical IPV.

The seven health outcome variables were selected based on available VACS and Global Burden of Disease (GBD) data and supported by existing IPV (Grose et al., 2019, 2020; Meinhart et al., 2021) and costing (Fang et al., 2017; UNICEF, 2019) literature. Lifetime health variables included diagnosis of a sexually transmitted infection (STI/STD), and violence perpetration against another person, self-directed thoughts or attempts to hurt. Past 30-day variables included disordered mental health based on the Kessler-6 thresholds for (Kessler et al., 2003), alcohol abuse based on drinking four or more drinks in a row, and drug use including marijuana, pills, ecstasy, or huffing chemicals. Tobacco use was measured based on daily frequency of smoking tobacco.

Individual-level covariates included whether the respondent had ever been married or lived with a partner as if married, whether the respondent was currently enrolled in school, and whether the respondent had worked within the past 12 months. Age was also included as a continuous independent variable.

Conflict exposure was dichotomized based on one household-level question and two individual-level questions. Respondents who lived in a household listed in the national Victim’s Unique Registry for internal violence were classified as conflict-affected, as were respondents who witnessed internal conflict and/or experienced forced migration.

Estimating costs

Aligning with previously published costing estimates of violence against children in Nigeria (UNICEF, 2019) and South Africa (Fang et al., 2017), this costing followed a four step procedure to estimate the health costs of IPV among adolescent females and young women in Colombia.

Step one: estimate the national prevalence

This analysis identified the overall prevalence of physical IPV among adolescent females and young women in Colombia, as well as whether the prevalence differed for females who had experienced conflict and those who had not. Emotional and sexual IPV were excluded from the prevalence estimate because clear questions and data were not available for consistent time periods within the measured outcomes (step 2). The full sample and conflict-affected or non–conflict-affected sub-groups were used to examine how physical IPV impacted females at the national level and based on conflict exposure.

Step two: estimate impact of intimate partner violence

Based on established outcomes or impacts of IPV (Grose et al., 2019, 2020; Meinhart et al., 2021), the overall prevalence of nine common health outcomes of IPV were identified within the VACS dataset. Relative risk (RR) calculations were conducted to assess the impact of IPV by estimating the probability of adverse health outcomes among females who had experienced IPV and those who had not. This RR analysis was also replicated for the conflict-affected and non–conflict-affected sub-groups. Using generalized linear models with Poisson-distributed errors, multivariable regression analyses were conducted to examine the adjusted RR between IPV and health consequences. While any strata with a singleton PSU were included in the adjusted RR results, singleton PSUs were dropped from contribution to standard errors.

Step three: establish the population attributable fractions

We used a “top-down” costing approach (Brown et al., 2008), in which the inputs from steps 1 and 2 are combined to estimate population attributable fractions (PAFs). PAFs are a standard epidemiologic concept which represents the share of an outcome, or total burden, which is estimated to result from a risk factor, in this case, IPV (Greenland, 2015). For example, an estimated PAF for IPV of 0.4 for a mental health condition implies that IPV is responsible for a 40% of the total estimated burden of that condition. PAFs may be applied to health outcomes such as mortality or morbidity, or also to economic outcomes, such as costs.

Validity of the PAF relies on the basic assumption of a counterfactual—namely, that the impacts in step 2 are estimated correctly—as well as specific details of the types of RRs and prevalence that are used. When adjusted RRs are used, as here (adjusted for age and marital status), an adjusted PAF formula is required (Benichou, 2001; Rockhill et al., 1998).

Step four: develop a costing model

We used a prevalence-based costing approach (Brown et al., 2008)) to estimate costs of IPV for health conditions attributable to IPV. For each health condition above, the associated PAF is multiplied by the estimated number of disability-adjusted life years (DALYs) for females in Colombia ages 13–24 in 2018 from the GBD series (Global Burden of Disease Collaborative Network, 2021). The PAF for interpersonal violence was combined with GBD estimates of DALYs for “interpersonal violence,” the PAF for STDs with “sexually transmitted infections excluding HIV,” the PAF for lifetime self-harm or suicidal thoughts/attempts with “self-harm,” the PAF for disordered mental health with “depressive disorders,” the PAF for alcohol abuse with “alcohol use disorders,” the PAF for tobacco use with “smoking,” and the PAF for drug use with “drug use disorders.” DALYs for ages 13–24 were found by summing estimates for females in Colombia ages 15–19, 20–24, and an adjusted 41.1% of the estimates for girls 10–14 (reflecting the population share ages 13–14 out of the 10–14 interval).

In the GBD classification system, all of these are considered as “Level 3 causes,” except for smoking (a Level 3 risk factor).¹ For interpersonal violence, we included only the non-fatal portion (years lived with disability, YLD) of the GBD DALY estimates, since the VACS question captures the incidence and impacts of perpetration on respondents, not estimated fatalities. Pregnancy and abortion were not included because of a poor match between the VACS survey and the adverse impacts captured in similar GBD categories and limited response rates to these questions. Finally, because smoking is a “risk factor” in the GBD, we also reviewed all “causes” associated with smoking and removed any overlap with the aforementioned causes to avoid double counting; in this case, “self-harm and interpersonal violence” attributable to smoking were removed from the GBD estimates before the PAF was applied. Double counting did not affect any other health impacts since no other GBD “risk factors” were used in the estimates.

Next, the level of DALYs that were estimated as attributable to IPV are valued as economic burden of both direct and indirect intangibles. Similar to other authors (D. W. Brown, 2008; Fang et al., 2017; WHO, 2001), we took a human capital approach and assumed that one DALY was equal to a country’s per capita gross domestic product (GDP). That is, a year of lost productive capacity from a country’s economic output (GDP) was equal to the DALY, a year lost due to living with disability or premature mortality. GDP per capita in Colombia in 2018 was $6729 USD, measured using the average foreign exchange rate during 2018 (World Bank, n.d.World Bankn), so one DALY was valued at $6729 USD.

The primary cost estimates (Table 4 & Figure 1) utilized the PAF estimates for all outcome variables from the full sample, due to their comparative robustness, and the IPV prevalence per sub-group, due to their specificity. We conducted three sensitivity analyses to compare cost estimates. First, cost estimates in Table 5 utilized the PAF estimates and IPV prevalence from each sub-group and only included findings from outcomes with statistically significant relative risks ratios (p < 0.05). Second, Monte Carlo simulations of 1000 draws were conducted to provide simulated confidence intervals to examine the range of estimates that may arise from the original estimates in Table 4. For each of the PAFs, we drew new random values from the statistical distribution associated with those PAFs. For the GBD estimates that are multiplied by the PAFs, we used a uniform normal distribution so that any value between the GBD “low” and “high” values was equally likely to occur. When multiplied together, they provided a range of DALY and cost estimates for the study, and the 5^th and 95^th percentiles were approximate estimates of the uncertainty bounds for our estimates. (This reflects uncertainty in the data and parameters but is not a true statistical confidence interval since a formal statistical distribution with correlations is not defined for the overall costing approach.) Third, for context, we conducted a final sensitivity analysis by estimating the total health costs from interpersonal violence, as well as conflict and terrorism, per the GBD estimates for “intimate partner violence” (a level 2 risk in the current GBD classification system) and “conflict and terrorism” (a level 3 cause in the GBD) for females 13–24.

Figure 1.

Estimated annual health burden of intimate partner violence among females 13–24 in Colombia based on conflict exposure, in 2018 US dollars.

Results

Demographics

Approximately 14% of females ages 13–24 experienced physical IPV during their lifetime, ranging from 22% in the conflict-affected sub-group and 11% in the non-conflict-affected sub-group (see Table 1). Nearly a quarter (24%) of the full sample were conflict-affected. There were no significant differences between other demographic characteristics in the conflict-affected and non–conflicted-affected sub-groups. However, there were differences between females who experienced IPV and those who did not. Females who experienced IPV were more likely to be older (20.55 vs. 18.18 years old), married (65% vs. 27%), not currently enrolled in school (27% vs. 59%), and have a past 12-month work history (63% vs. 33%) than females who did not experience IPV

Table 1.

Demographic and Costing Outcome Characteristics Among Females 13–24 in Colombia.

	Full Sample (n = 1406)	No Lifetime History of IPV (n = 1213)	Lifetime History of IPV (n = 189)	Full Conflict Affected Sub-group (n = 291)	No Lifetime History of IPV (n = 231)	Lifetime History of IPV (n = 60)	Full Non-Conflict-Affected Sub-group (n = 1115)	No Lifetime History of IPV (n = 982)	Lifetime History of IPV (n = 129)
	Mean [95% CI]	Mean [95% CI]	Mean [95% CI]	Mean [95% CI]	Mean [95% CI]	Mean [95% CI]	Mean [95% CI]	Mean [95% CI]	Mean [95% CI]
Intimate partner violence
Ever	0.14	-	-	0.22	-	-	0.11	-	-
	(0.10–0.17)	-	-	(0.12–0.32)	-	-	(0.08–0.15)	-	-
Conflict exposure
Ever	0.24	0.22	0.39	-	-	-	-	-	-
	(0.17–0.31)	(0.16–0.29)	(0.22–0.56)	-	-	-	-	-	-
Demographics
Age	18.51	18.18	20.55	19.12	18.42	21.6	18.31	18.12	19.89
	(18.17–18.84)	(17.84–18.53)	(20.55–20.55)	(19.12–19.12)	(18.42–18.42)	(21.60–21.60)	(17.94–18.67)	(17.70–18.53)	(19.89–19.89)
Ever married or cohabitating as married	0.32	0.27	0.65	0.44	0.4	0.61	0.28	0.23	0.67
	(0.27–0.37)	(0.22–0.31)	(0.65–0.65)	(0.44–0.44)	(0.40–0.40)	(0.61–0.61)	(0.22–0.33)	(0.17–0.28)	(0.67–0.67)
Currently in school	0.54	0.59	0.27	0.48	0.58	0.14	0.56	0.59	0.35
	(0.49–0.59)	(0.53–0.64)	(0.27–0.27)	(0.48–0.48)	(0.58–0.58)	(0.14–0.14)	(0.49–0.63)	(0.52–0.66)	(0.35–0.35)
Work history (past 12-mo experience)	0.37	0.33	0.63	0.44	0.4	0.62	0.35	0.31	0.63
	(0.32–0.42)	(0.27–0.39)	(0.63–0.63)	(0.44–0.44)	(0.40–0.40)	(0.62–0.62)	(0.28–0.41)	(0.24–0.38)	(0.63–0.63)
Costing outcomes
Lifetime STD	0.02	0.02	0.06	0.05	0.06	0.03	0.02	0.01	0.07
	(0.01–0.04)	(-0.00–0.04)	(0.06–0.06)	(0.05–0.05)	(0.06–0.06)	(0.03–0.03)	(0.00–0.03)	(0.00–0.02)	(0.07–0.07)
Lifetime violence perpetration	0.15	0.09	0.55	0.22	0.09	0.7	0.13	0.09	0.45
	(0.11–0.19)	(0.05–0.12)	(0.55–0.55)	(0.22–0.22)	(0.09–0.09)	(0.70–0.70)	(0.08–0.17)	(0.04–0.13)	(0.45–0.45)
Lifetime self-harm	0.23	0.21	0.37	0.33	0.33	0.31	0.2	0.18	0.4
	(0.18–0.29)	(0.16–0.26)	(0.37–0.37)	(0.33–0.33)	(0.33–0.33)	(0.31–0.31)	(0.15–0.25)	(0.12–0.23)	(0.40–0.40)
Past 30-day disordered mental health	0.52	0.48	0.72	0.61	0.57	0.77	0.49	0.46	0.69
	(0.45–0.58)	(0.42–0.55)	(0.72–0.72)	(0.61–0.61)	(0.57–0.57)	(0.77–0.77)	(0.41–0.56)	(0.38–0.54)	(0.69–0.69)
Past 30-day alcohol abuse	0.28	0.25	0.46	0.29	0.22	0.66	0.28	0.27	0.37
	(0.22–0.35)	(0.19–0.31)	(0.46–0.46)	(0.29–0.29)	(0.22–0.22)	(0.66–0.66)	(0.21–0.35)	(0.20–0.33)	(0.37–0.37)
Past 30-day drug use	0.01	0.01	0.06	0.01	0.01	0.01	0.02	0.01	0.09
	(0.00–0.03)	(0.00–0.01)	(0.06–0.06)	(0.01–0.01)	(0.01–0.01)	(0.01–0.01)	(0.00–0.03)	(-0.00–0.02)	(0.09–0.09)
Currently smoking	0.06	0.04	0.15	0.1	0.06	0.22	0.05	0.04	0.1
	(0.03–0.09)	(0.02–0.07)	(0.15–0.15)	(0.10–0.10)	(0.06–0.06)	(0.22–0.22)	(0.02–0.07)	(0.02–0.06)	(0.10–0.10)

Notes: Differences between cells were determined by examining if confidence intervals overlapped. 95% CI = 95% confidence interval.

Health Outcomes

Table 2 presents results from the relative risk analyses. When controlling for covariates, findings from the full sample reveal that those who experienced IPV were significantly more likely to experience six of the seven health outcomes: lifetime violence perpetration (RR = 6.03; 95% CI = 3.92, 9.26), lifetime self-harm (RR = 1.86; 95% CI = 1.23, 2.82), past 30-day disordered mental health (RR = 1.46; 95% CI = 1.17, 1.83), past 30-day alcohol abuse (RR = 1.54; 95% CI = 1.02, 2.31), past 30-day drug use (RR = 19.56; 95% CI = 5.22, 73.32), and currently smoking (RR = 3.06; 95% CI = 1.06, 8.82). Lifetime STD diagnosis was not significantly associated with IPV experience.

Table 2.

Relative Risks of Health Outcomes Associated With Intimate Partner Violence Among Females Ages 13–24 in Colombia.

	Lifetime STD	Lifetime Violence Perpetration	Lifetime Self-Harm	Past 30-day Disordered Mental Health	Past 30-day Alcohol Abuse	Past 30-day Drug Use	Currently Smoking
	Adjusted Relative Risk [95% CI]	Adjusted Relative Risk [95% CI]	Adjusted Relative Risk [95% CI]	Adjusted Relative Risk [95% CI]	Adjusted Relative Risk [95% CI]	Adjusted Relative Risk [95% CI]	Adjusted Relative Risk [95% CI]
Full sample	1.35	6.03***	1.86**	1.46***	1.54*	19.56***	3.06*
	(0.21–8.47)	(3.92–9.26)	(1.23–2.82)	(1.17–1.83)	(1.02–2.31)	(5.22–73.32)	(1.06–8.82)
Conflict-affected sub-group	0.37	8.09***	0.92	1.42	3.07***	21.65***	1.94
	(0.06–2.34)	(3.94 - 16.62)	(0.48–1.79)	(0.98–2.05)	(1.64–5.74)	(3.90–120.32)	(0.70–5.39)
Non–conflict-affected sub-group	8.41*	5.14***	2.73***	1.46*	1.18	18.77***	2.62*
	(1.23–57.63)	(2.95 - 8.96)	(1.70–4.37)	(1.09–1.95)	(0.67–2.09)	(5.28–66.80)	(1.04–6.63)

Notes: Findings show relative risk of physical intimate partner violence and column variables. Covariates include age, marital status, school status, and working status. ***p < 0.001, **p < 0.01, **p < 0.05.95% CI = 95% confidence interval.

Findings were less consistent at the sub-group level and, overall, CIs were much wider. In the conflict-affected sub-group, females who experienced IPV were significantly more likely to experience only three of the seven health outcomes: lifetime violence perpetration (RR = 8.09; 95% CI = (3.94, 16.62), past 30-day alcohol abuse (RR = 3.07; 95% CI = 1.64, 5.74), and past 30-day drug use (RR = 21.65; 95% CI = 3.90, 120.32). Findings from the non–conflict-affected sub-group were more consistent; only past 30-day alcohol use was not significantly associated with IPV experience (p > 0.05).

Table 3 presents results from the PAF analyses, and only statistically significant associations from Table 2 were considered in our costing models. The highest PAFs were found when examining to what extent IPV experience attributed to past 30-day drug use. PAFs for past 30-day drug use ranged from 50.71% in the full sample (95% CI = 12.60%, 72.20%) to 52.54% in the conflict-affected sub-group (95% CI = −59.78%, 85.90%). Lifetime violence perpetration (36.90%) and currently smoking (18.97%) had the second and third highest PAFs in the full sample, respectively. IPV experience attributed less than 10% to the remaining three outcomes: lifetime self-harm (8.54%), past 30-day disordered mental health (5.26%), and past 30-day alcohol abuse (6.29%).

Table 3.

Population Attributable Fractions for Health Outcomes Associated With Intimate Partner Violence Among Females Aged 13–24 in Colombia.

	Lifetime STD	Lifetime Violence Perpetration	Lifetime Self-Harm	Past 30-day Disordered Mental Health	Past 30-day Alcohol Abuse	Past 30-day Drug Use	Currently Smoking
	Adjusted Relative Risk [95% CI]	Adjusted Relative Risk [95% CI]	Adjusted Relative Risk [95% CI]	Adjusted Relative Risk [95% CI]	Adjusted Relative Risk [95% CI]	Adjusted Relative Risk [95% CI]	Adjusted Relative Risk [95% CI]
Full sample	7.24%	36.90%	8.54%	5.26%	6.29%	50.71%	18.97%
	(−51.24% − 43.10%)	(24.33% − 47.39%)	(1.63% − 14.95%)	(1.75% − 8.65%)	(−0.59% − 12.70%)	(12.60% − 72.20%)	(−6.81% − 38.53%)
Conflict-affected sub-group	−23.30%	46.77%	−0.90%	4.79%	19.20%	52.54%	9.80%
	(−60.58% − 5.32%)	(20.69% − 64.27%)	(−8.51% − 6.18%	(−1.10% − 10.34%)	(4.54% − 31.61%)	(−59.78% − 85.90%)	(−10.25% − 26.21%)
Non–conflict-affected sub-group	49.07%	31.41%	14.38%	5.10%	2.50%	51.81%	15.43%
	(−14.97% − 77.44%)	(15.59% − 44.27%)	(4.80% − 23.00%)	(0.59% − 9.41%)	(−6.84% − 11.02%)	(13.23% − 73.24%)	(−7.59% − 33.51%)

Notes: Findings show point estimates and confidence intervals of the population attributable fraction of column variables based on physical intimate partner violence experience. Covariates include age, marital status, school status, and working status. Bolded outcomes represent relative risk findings that were significant (p < 0.05). 95% CI = 95% confidence interval.

Economic Burden

Primary results for the costing analysis are shown in Figure 1 and Table 4. Figure 1 provides a comparative summary of the overall health burden borne by each outcome among the two sub-groups, while Table 4 provides a detailed description of the DALY and cost estimates. The first panel (first rows) of Table 4 contains results for the entire population, with results for conflict-affected and non–conflict-affected in the two panels below. In each panel, results are shown in both DALYs and USD estimates for non-fatal costs (YLD), premature mortality (YLL), and the combined total. Columns in the tables show each health outcome along with the total for all health consequences.

Table 4.

Estimated Economic Value of DALYs Lost to Intimate Partner Violence Among Females 13–24 in Colombia (Per Full Sample PAFs).

		Lifetime STD	Lifetime Violence Perpetration	Lifetime Self-Harm	Past 30-day Disordered Mental Health	Past 30-day Alcohol Abuse	Past 30-day Drug Use	Currently Smoking	Total
		DALY (2018 USD Cost in Millions)	DALY (2018 USD Cost in Millions)	DALY (2018 USD Cost in Millions)	DALY (2018 USD Cost in Millions)	DALY (2018 USD Cost in Millions)	DALY (2018 USD Cost in Millions)	DALY (2018 USD Cost in Millions)	DALY (2018 USD Cost in Millions)
Entire population	YLD $	$701,379	$5,785,349	$187,516	$13,366,813	$4,823,663	$55,008,201	$97,588	$79,968,663
	YLD (DALY)	104.2	860	27.9	1986.3	716.8	8174.1	14.5	11,883.4
	YLL $	$97,378	N/A	$8,007,868	$0	$33,332	$2,160,276	$331,604	$10,630,215
	YLL (DALY)	14.5	N/A	1190.0	0.0	5.0	321.0	49.3	1579.7
	Total $	$798,757	$5,785,349	$8,195,388	$13,366,813	$4,856,995	$57,168,477	$429,192	$90,600,971
	Total DALY	118.7	860.0	1217.8	1986.3	721.7	8495.1	63.8	13,463.4
Conflict-affected population	YLD $	$271,503	$2,239,509	$72,587	$5,174,293	$1,867,240	$21,293,679	$37,776	$30,956,587
	YLD (DALY)	40.3	333	10.8	768.9	277.5	3164.2	5.6	4600.1
	YLL $	$37,695	N/A	$3,099,848	$0	$12,903	$836,243	$128,364	$4,115,052
	YLL (DALY)	5.6	N/A	460.6	0.0	1.9	124.3	19.1	611.5
	Total $	$309,198	$2,239,509	$3,172,435	$5,174,293	$1,880,143	$22,129,917	$166,140	$35,071,635
	Total DALY	45.9	332.8	471.4	768.9	279.4	3288.5	24.7	5211.6
Non-conflict-affected population	YLD $	$429,875	$3,545,840	$114,929	$8,192,519	$2,956,424	$33,714,528	$59,811	$49,013,925
	YLD (DALY)	63.9	527	17.1	1217.4	439.3	5009.9	8.9	7283.4
	YLL $	$59,683	N/A	$4,908,025	$0	$20,429	$1,324,033	$203,240	$6,515,410
	YLL (DALY)	8.9	N/A	729.3	0.0	3.0	196.7	30.2	968.2
	Total $	$489,558	$3,545,840	$5,022,953	$8,192,519	$2,976,852	$35,038,559	$263,051	$55,529,332
	Total DALY	72.7	526.9	746.4	1217.4	442.4	5206.6	39.1	8251.5

Notes: Findings show point estimates of the share of DALYs and dollar costs attributable to physical intimate partner violence experience. Bolded outcomes represent relative risk (RR) findings that were statistically significant in PAF and RR calculations (p<0.05). Years of life lost (YLLs) were not estimated for violence perpetration, and violence perpetration YLLs were excluded from total estimates. YLD = Years of healthy life lost due to disability; $ = dollar amount: USD = United States dollars.

For the full population of girls and women ages 13–24 in Colombia in 2018, we estimate that the societal burden of physical IPV totaled $90.6 million USD from adverse health outcomes alone. This total was driven by an estimated 13,463 DALYs, with nearly $80 million of the burden coming from nonfatal health outcomes (YLDs). The greatest contributor to the total burden was from drug abuse, representing approximately 63% of the estimated health and economic burden ($57.2 million), with nearly all derived from nonfatal health outcomes. Considering the other health outcomes, the next largest cause was mental health (15% of the total), followed by self-harm (9%), violence perpetration (6%), alcohol abuse (5%), STDs (1%), and smoking (<1%). The costs of self-harm resulted largely from premature mortality (97% of its total).

Considering conflict-affected and non–conflict-affected populations result parallel, the story for the full population since for our main estimates the PAF was adjusted for population size and incidence of IPV, but not differing relative risks (RRs). Overall, the economic burden of physical IPV among conflict-affected girls and women ages 13–24 were an estimated $35,071,635 from 5212 DALYs in 2018. Among those not conflict-affected, IPV attributable costs were estimated at $55,529,332 from 8252 DALYs.

Sensitivity analyses

In Table 5, results are recomputed using PAFs which are specific to each population, conflict and non-conflict-affected, and include only the adverse health outcomes with statistically significant relative risk ratios (p < 0.05) for that population. If relative risks were non-significant (Table 2) for a given health outcome, we excluded that category, setting the PAF for that category to zero. For the entire population, all outcomes except STDs were included. Because STDs accounted for less than 1% of the total burden of health, the costs were almost indistinguishable between Tables 4 and 5 for the full population of females ages 13–24 in Colombia ($89.8 million vs. $90.6 million). For the conflict-affected population, four health outcomes were excluded: STDs, self-harm, mental health, and smoking. (In this sub-group, the RRs for these health outcomes were likely non-significant because of small sample sizes—a lack of statistical power in the VACS data.) Again, these causes also represented a small share of the total burden and total health costs for women in conflict-affected areas are estimated at $31.5 million (compared to $35.1 million in Table 4). For the non-conflict-affected population, only alcohol abuse was excluded, which is about 5% of the share of the total burden, and total costs were estimated at $59.3 million (compared to $55.5 million in Table 4). The increased total for this subpopulation results from larger PAFs than in Table 4.

Table 5.

Estimated Economic Value of DALYs Lost to Intimate Partner Violence Among Females 13–24 in Colombia (Per Sub-Group PAFs).

		Lifetime STD	Lifetime Violence Perpetration	Lifetime Self-Harm	Past 30-day Disordered Mental Health	Past 30-day Alcohol Abuse	Past 30-day Drug Use	Currently Smoking	Total
		DALY (2018 USD Cost in Millions)	DALY (2018 USD Cost in Millions)	DALY (2018 USD Cost in Millions)	DALY (2018 USD Cost in Millions)	DALY (2018 USD Cost in Millions)	DALY (2018 USD Cost in Millions)	DALY (2018 USD Cost in Millions)	DALY (2018 USD Cost in Millions)
Entire population	YLD $	-	$5,785,349	$187,516	$13,366,813	$4,823,663	$55,008,201	$97,588	$79,269,130
	YLD (DALY)	-	860	27.9	1986.3	716.8	8174.1	14.5	11,779.2
	YLL $	-	N/A	$8,007,868	$0	$33,332	$2,160,276	$331,604	$10,533,080
	YLL (DALY)	-	N/A	1190.0	0.0	5.0	321.0	49.3	1565.2
	Total $	-	$5,785,349	$8,195,388	$13,366,813	$4,856,995	$57,168,477	$429,192	$89,802,214
	Total DALY	-	860.0	1217.8	1986.3	721.7	8495.1	63.8	13,344.7
Conflict-affected population	YLD $	-	$2,838,174	-	-	$5,702,128	$22,064,559	-	$30,604,861
	YLD (DALY)	-	421.7	-	-	847.3	3278.7	-	4547.8
	YLL $	-	N/A	-	-	$39,402	$866,514	-	$905,916
	YLL (DALY)	-	N/A	-	-	5.9	128.8	-	134.6
	Total $	-	$2,838,174	-	-	$5,741,530	$22,931,073	-	$31,510,777
	Total DALY	-	421.7	-	-	853.2	3407.5	-	4682.4
Non-conflict-affected population	YLD $	$701,379	$3,545,840	$193,616	$7,946,620	-	$34,447,311	$48,642	$46,883,409
	YLD (DALY)	104.2	526.9	28.8	1180.8	-	5118.8	7.2	6966.8
	YLL $	$2,618,659	N/A	$8,268,387	$0	-	$1,352,814	$165,287	$12,405,146
	YLL (DALY)	389.1	N/A	1228.7	0.0	-	201.0	24.6	1843.4
	Total $	$3,320,038	$3,545,840	$8,462,003	$7,946,620	-	$35,800,125	$213,929	$59,288,555
	Total DALY	493.3	526.9	1257.4	1180.8	-	5319.8	31.8	8810.1

Notes: Findings show point estimates of the share of DALYs and dollar costs attributable to physical intimate partner violence experience. Bolded outcomes represent relative risk (RR) findings that were statistically significant in PAF and RR calculations (p < 0.05). Years of life lost (YLLs) were not estimated for violence perpetration, and violence perpetration YLLs were excluded from total estimates. YLD = Years of healthy life lost due to disability; $ = dollar amount: USD = United States dollars.

Simulated confidence intervals are shown in Table 6 to provide a gauge on the potential range of estimates that may result from the underlying data and steps used in our costing model, including low and high estimates reported by the GBD. For the overall estimates, the likely bounds for the $90.6 million total range from $74.8 to $147.3 million. Within conflict-affected populations, likely bounds for the $35.1 million total range from $28.9 to $57.8 million, and within non–conflict-affected populations, likely bounds for the $55.5 million total range from $45.8 to $89.5 million. (Simulated bounds for DALYs and YLLs were also calculated but are not shown.) These simulated confidence intervals provide further evidence that the health cost burden of physical IPV among females 13–24 is substantial and may be as much as 50% higher than our totals here, or possibly somewhat smaller. Note that the bounds are somewhat larger on the high end and do not take the symmetric form that is common with classical statistical confidence bounds because they are based on multiple non-normal statistical distributions underlying the PAFs (RRs and prevalence) and GBD.

Table 6.

Monte Carlo Simulations of Estimated Economic Value of DALYs Lost to Intimate Partner Violence Among Females 13–24 in Colombia.

		Lifetime STD	Lifetime Violence Perpetration	Lifetime Self-Harm	Past 30-day Disordered Mental Health	Past 30-day Alcohol Abuse	Past 30-day Drug Use	Currently Smoking	Total
		DALY (2018 USD Cost in Millions)	DALY (2018 USD Cost in Millions)	DALY (2018 USD Cost in Millions)	DALY (2018 USD Cost in Millions)	DALY (2018 USD Cost in Millions)	DALY (2018 USD Cost in Millions)	DALY (2018 USD Cost in Millions)	DALY (2018 USD Cost in Millions)
Entire population	Total $	$798,757	$5,785,349	$8,195,388	$13,366,813	$4,856,995	$57,168,477	$429,192	$90,600,971
	Low $	$635,470	$4,041,381	$3,802,431	$12,348,780	$3,598,060	$49,950,688	$376,110	$74,752,920
	High $	$2,890,531	$7,754,154	$10,117,227	$24,942,244	$8,280,288	$92,196,059	$1,158,342	$147,338,845
Conflict-affected population	Total $	$309,198	$2,239,509	$3,172,435	$5,174,293	$1,880,143	$22,129,917	$166,140	$35,071,635
	Low $	$223,411	$1,564,419	$1,471,921	$4,780,213	$1,392,808	$19,335,911	$145,593	$28,914,276
	High $	$1,118,924	$3,001,632	$4,755,079	$9,655,144	3,205,300	$35,689,094	$448,394	$57,873,567
Non-conflict-affected population	Total $	$489,558	$3,545,840	$5,022,953	$8,192,519	$2,976,852	$35,038,559	$263,051	$55,529,332
	Low $	$412,059	$2,476,962	$2,330,510	$7,568,567	$2,205,252	$30,614,777	$230,517	$45,838,644
	High $	$1,771,607	$4,752,522	$5,362,148	$15,287,100	$5,074,988	$56,506,965	$709,948	$89,465,278

Notes: Findings show point estimates of the share of dollar costs attributable to physical intimate partner violence experience. The low and high values represent the 5th and 95th percentiles from 1000 simulated draws, based on the distribution of study parameters (PAFs) and Global Burden of Disease (GBD) costs. Bolded outcomes represent relative risk (RR) findings that were statistically significant in PAF and RR calculations (p < 0.05). Years of life lost (YLLs) were not estimated for violence perpetration, and violence perpetration YLLs were excluded from total estimates. YLD = Years of healthy life lost due to disability; $ = dollar amount: USD = United States dollars.

Discussion

This study aimed to estimate the health cost burden of inaction—or maintaining the status quo—resulting from physical IPV experienced by females ages 13–24 in Colombia in 2018. GBD estimates are $569 million (Global Burden of Disease Collaborative Network, 2021), indicating that at least 16% of that burden results from physical IPV and is preventable. Moreover, since IPV affects the broader population of females, IPV-related societal costs are considerably larger than the estimates in this analysis. We use a top-down, prevalence-based approach, which permits costing using a variety of different data sources: VACS estimates of the prevalence and risk of IPV and of associated health problems computed by the authors, previously-reported GBD estimates of the burden of various adverse health outcomes and risk factors, GDP per capita economic data, and daily currency exchange rates. Our estimates from these data provide a variety of stakeholders with a snapshot of the economic burden of inaction in a single year—that is, the costs lost to society because IPV occurs despite existing policies and programs. Put another way, if IPV were fully prevented, these costs could be recouped by society and used for other purposes, the economic notion of opportunity cost. Furthermore, to the extent that prevention and mitigation can be implemented, and the costs of those activities measured, then such efforts are beneficial to society up to the level of burden measured here (net benefits are costs are avoided minus costs spent to reduce them).

Conservative estimates with high costs

The estimates in this study are likely conservative and understate the financial burden of IPV among females 13–24 in Colombia for several reasons. First, for costing purposes, we needed a consistent time period in reporting, and the VACS data only consistently measured physical IPV with the same recall period overall ages. Emotional and sexual IPV were excluded because of measurement difficulties in VACS, even though both are common and known to have damaging effects (Aye et al., 2020; Falb et al., 2019; Gibbs et al., 2018; Seff & Stark, 2019; Shamu et al., 2018)—and thus, financial costs—on the health and well-being of girls and women. Second, the VACS data also only capture survivors between the ages of 13–24 using items feasible for costing, so the health costs of IPV for adolescents younger than 13 and females older than 24 are not included in the estimates. Women and girls who would have otherwise participated in the VACS if not fatally impacted by physical IPV were not included, resulting in lower prevalence estimates than if including IPV fatalities. Third, the costing approach also requires that each item be linked to an associated adverse health outcome in order to estimate the RR and PAF. We included seven health impacts (violence perpetration, STDs, self-harm, mental health, alcohol abuse, tobacco use, and drug use) which were also measurable in both the VACS as associated with IPV and quantified for overall burden the GBD reports. Other outcomes are known to be associated with IPV include, particularly other health conditions and economic strains (Grose et al., 2019, 2020; Kayibanda & Alary, 2020; Meinhart et al., 2021; Riger & Staggs, 2004; Siddique, 2018). Yet, we were not able to include these outcomes in our models as they were not captured in the VACs dataset. Thus, our scope and costs of IPV are lower than would be the case if additional impacts were measured. Fourth, the prevalence-based costing approach provides a 1 year, point-in-time snapshot of costs, but it does not reflect lifetime costs per victim. This is important as costs change and accumulate across the stages of life, particularly if not sufficiently addressed at the onset of violence victimization. Multi-year impacts from chronic or lasting health problems are measured only during the associated 12-month period. Although these limitations are necessary to implement the costing approach, we emphasize to readers that the “true” costs of IPV are likely to be far higher our estimate.

Another important reason why our estimate is likely to be under-measured is that we were unable to include any productivity losses for the impacts of IPV on educational or labor market outcomes. Girls and women who are victims of IPV are more likely to be absent from school and work, resulting in reduced educational outcomes (lowered attainment, reduced performance, graduation rates), and eventually, reduced earnings and lower labor force participation ( Brown et al., 2008; Robst & Smith, 2008). In related studies using a similar approach to estimate the cost of violence against children in Nigeria (UNICEF, 2019) and South Africa (Fang et al., 2017), labor market costs were substantial. In some studies (Fang et al., 2012; Letourneau et al., 2018), productivity costs far exceed the health impacts. Future studies could usefully seek data sources and use approaches that allow estimation of the educational and productivity impacts of IPV, especially if they include adult populations.

Conflict-specific considerations

Despite only representing 24% of the sample, females in the conflict-affected sub-group who experienced IPV accounted for 39% of the overall economic burden. This finding reflects the disproportionate risk of IPV experienced by women and girls living in conflict-affected environments described by other studies; (Catani, 2010; Kelly et al., 2018; Parcesepe et al., 2016; Stark et al., 2013; Stark & Ager, 2011). Our findings reveal that females ages 13–24 affected by conflict were twice as likely to experience physical IPV within the previous year (22%) compared to females not affected by conflict (11%). Multi-lateral and national policies that address gender-based violence need to consider how conflict exacerbates IPV and compounds the economic burden of IPV. Future population-based research could help further elucidate these relationships by oversampling conflict-affected populations to facilitate more robust analysis at the sub-group level.

Moreover, our findings highlight that the estimated health costs of physical IPV for conflict-affected females ages 13–24 in Colombia ($35.1 million) is more than two times the health costs of the GBD defined level 3 cause of conflict and terrorism in Colombia ($15.6 million) (Global Burden of Disease Collaborative Network, 2021). In other words, health costs from IPV in humanitarian settings may account for twice as many costs as those directly related to conflict. This is a critical insight given the asymmetry between the estimated cost of physical IPV in Colombia among conflict-affected adolescent females and the amount of funding allocated per the Humanitarian Response Plan (HRP) to potentially address gender-based violence. No HRP funding was allocated to Colombia in 2018, and only $12.8 million (in 2021 USD) has been allocated for protection purposes in Colombia (OCHA FTA, 2021). Thus, the health costs incurred by this population in a single year would be substantially higher than allocated HRP spending, even if all of the 2021 HRP protection funding for Colombia was allocated to address physical IPV among adolescent females. While gaps in funding for the violence faced by women and girls is ubiquitous, the limited HRP funding allocated to protection suggests the need for increased investment, including by the Colombian government, to support women and girls impacted by conflict.

Progressive policies, often enabled through the recurrent National Development Plans, have placed Colombia at the global forefront of addressing gender-based violence and violence against children. The government’s concentrated interest to address to the known issue of IPV can further position Colombia as a leader in transformative programming, especially if policies better address the critical period of youth and address the exacerbating impact of IPV in post-conflict settings. Our findings enable the government to have rigorous estimates of IPV costs to compliment the robust knowledge gained from the VACS work. The timing of our estimates aligns well with the development of Colombia’s forthcoming National Action Plan, where the government will outline a roadmap to invest in violence prevention and response services. Our hope is that these estimates may enable future comparison of the costs to intervene and the costs of maintaining the status quo to provide as a beacon for other country’s interested in investing to eliminate IPV.

Limitations

This study is subject to additional limitations which have not been discussed above. First, our estimates are directly influenced by the GBD estimates of the burden of health for the seven adverse health outcomes which are included. To the extent that GBD estimates vary, this variance would also affect our findings. There is limited potential for double counting between the “risk” category of smoking and the other “causes” included in our study although we reviewed the corresponding GBD methodology documents and removed the portion attributable to a shared cause (“self-harm and interpersonal violence” attributable to smoking were removed from the GBD estimates before the PAF was applied). Also, since GBD estimates for ages 13–14 were not available, we used the population share of 13–14 and applied this to the GBD estimates for 10–14 to reduce the DALY estimates by the corresponding share. This decision assumes an equal impact across all years of age, which is likely conservative adjustments since the incidence of IPV—while low among all of ages 10–13—increases nonlinearly with age (WHO, 2021). Next, our estimates are specific to 2018 since that was the most recent year of VACS data available. (GBD estimates for 2019 are available.) Fluctuations in GDP, inflation, and exchange rates between 2018 and the present would affect our results in an unknown manner. Finally, we excluded from our formal costing analysis the impacts of abortion and pregnancy even though these are measured somewhat in the VACS and the GBD. Specifically, these outcomes were excluded because the definition between VACS and GBD did not align well, and response rates to these questions on the VACS were low and potentially unreliable.

Conclusion

Findings from this study emphasize the ubiquity of IPV among females ages 13–24, particularly among those who are conflict-affected. We demonstrate that the overall economic burden of physical IPV in 2018 among this population to be $90.6 million. These estimates provide a clear and compelling picture of the minimum health costs of IPV in a single year. To the extent that IPV continues, readers should expect that these same costs will be incurred annually for this population and, of course, would increase as the rate of IPV increases. Bearing in mind the staggering health burden of IPV, effective prevention programming is likely to be more cost effective than maintaining the status quo.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the US State Department Bureau of Population, Refugees and Migration.

ORCID iDs

Derek Brown

Melissa Meinhart

Lindsay Stark

Note

Author Biographies

Derek S. Brown, PhD, is a health economist and an associate professor of public health and social work at the Washington University in St. Louis. His research includes methods in economic evaluation, studies of the impacts of child abuse and neglect on economic outcomes, health preference research, and the impacts of economic supports on the incidence of child maltreatment and intimate partner violence. His goal is to use quantitative and economic evaluation methods to improve policies and decision making at both domestic and global levels. Dr Brown’s work has been funded by CDC, NIMH, AHRQ, UNICEF, and other funders.

Melissa Meinhart has a PhD in Social Policy from Columbia University’s School of Social Work with a social science concentration in economics. Her dissertation integrated structural equation modeling to explore how latent classes of social support may be integrated within resiliency models among migrant populations at risk of depression. Her research focuses on the measurement and examination of underlying constructs that influence gendered inequities and social discordance in low- and middle-income countries, particularly within humanitarian emergencies.

Catherine Leila Poulton has more than 20 years of experience responding to emergencies around the world. She holds a BscEcon from the University of Cardiff and a Master’s degree from the European Centre for Peace Studies, and has been at the forefront of global efforts to ensure Gender-Based Violence response and prevention are an integral part of humanitarian responses. She has worked as a direct responder on the ground with several organisations, as well as been involved in significant research and measurement projects. She is also a member of COFEM. She currently leads UNICEF’s global GBViE programmes based in New York.

Lindsay Stark is a social epidemiologist and associate professor of public health and social work at Washington University in St. Louis. Dr Stark is an internationally recognized expert on the protection and well-being of women and children in situations of extreme adversity, with more than a decade of experience leading applied research with operational agencies such as UNICEF, UNHCR, International Rescue Committee and the Women’s Refugee Commission. Dr Stark co-directs the Center on Violence and Injury Prevention and the International Center for Child Health and Development, and holds affiliate appointments with the department of Women, Gender, and Sexuality Studies and the Institute for Public Health.

References

Aye

W. T.

Lien

Stigum

Schei

Sundby

Bjertness

(2020). Domestic violence victimisation and its association with mental distress: A cross-sectional study of the Yangon region, Myanmar. BMJ Open, 10(9), e037936. https://doi.org/10.1136/bmjopen-2020-037936

Benichou

(2001). A review of adjusted estimators of attributable risk. Statistical Methods in Medical Research, 10(3), 195-216. https://doi.org/10.1177/096228020101000303

Brown

D. W.

(2008). Economic value of disability-adjusted life years lost to violence: Estimates for WHO Member States. Revista Panamericana de Salud Pública, 24(3), 203-209. https://doi.org/10.1590/s1020-49892008000900007

Brown

D. S.

Finkelstein

E. A.

Mercy

J. A.

(2008). Methods for estimating medical expenditures attributable to intimate partner violence. Journal of Interpersonal Violence, 23(12), 1747-1766. https://doi.org/10.1177/0886260508314338

Catani

(2010). War in the home- an overview of the connection between war traumatization and domestic violence. Behavioral Therapy, 20(1), 19-27. https://doi.org/10.1159/000261994

Falb

K. L.

Blackwell

Stennes

Hussein

Annan

(2019). Depressive symptoms among women in Raqqa Governorate, Syria: Associations with intimate partner violence, food insecurity, and perceived needs. Global Mental Health, 6, e22. https://doi.org/10.1017/gmh.2019.20

Fang

Brown

D. S.

Florence

C. S.

Mercy

J. A.

(2012). The economic burden of child maltreatment in the United States and implications for prevention. Child Abuse & Neglect, 36(2), 156-165. https://doi.org/10.1016/j.chiabu.2011.10.006

Fang

Zheng

Fry

D. A.

Ganz

Casey

Hsiao

Ward

C. L.

(2017). The economic burden of violence against children in South Africa. International Journal of Environmental Research and Public Health, 14(11), 1431. https://doi.org/10.3390/ijerph14111431

Gibbs

Corboz

Jewkes

(2018). Factors associated with recent intimate partner violence experience amongst currently married women in Afghanistan and health impacts of IPV: A cross sectional study. BMC Public Health, 18(1), 593. https://doi.org/10.1186/s12889-018-5507-5

10.

Global Burden of Disease Collaborative Network (2021). GBD Results Tool . Global Burden of Disease Study 2019 (GBD 2019) Results http://ghdx.healthdata.org/gbd-results-tool

11.

Government of Colombia, Ministry of Health and Social Protection . (2019). Colombia Violence Against Children and Youth Survey, 2018.

12.

Greenland

(2015). Concepts and pitfalls in measuring and interpreting attributable fractions, prevented fractions, and causation probabilities. Annals of Epidemiology, 25(3), 155-161. https://doi.org/10.1016/j.annepidem.2014.11.005

13.

Grose

R. G.

Chen

J. S.

Roof

K. A.

Rachel

Yount

K. M.

(2020). Sexual and reproductive health outcomes of violence against women and girls in lower-income countries: A review of reviews. Journal of Sex Research, 58, 1-20. https://doi.org/10.1080/00224499.2019.1707466

14.

Grose

R. G.

Roof

K. A.

Semenza

D. C.

Leroux

Yount

K. M.

(2019). Mental health, empowerment, and violence against young women in lower-income countries: A review of reviews. Aggression and Violent Behavior, 46, 25-36. https://doi.org/10.1016/j.avb.2019.01.007

15.

IASC (2019). The inter-agency minimum standards for gender-based violence in emergencies programming Gender-based violence area of responsibility (p. 159). UNFPA. https://www.unfpa.org/sites/default/files/pub-pdf/19-200_Minimun_Standards_Report_ENGLISH-Nov.FINAL_.pdf https://www.unfpa.org/minimum-standards

16.

Kayibanda

J. F.

Alary

(2020). Trends and risk factors associated with the perpetration of physical intimate partner violence by women in Haiti, 2000-2012. Women & Health, 60(1), 12-25. https://doi.org/10.1080/03630242.2019.1607800

17.

Kelly

J. T. D.

Colantuoni

Robinson

Decker

M. R.

(2018). From the battlefield to the bedroom: A multilevel analysis of the links between political conflict and intimate partner violence in Liberia. BMJ Global Health, 3(2), e000668. https://doi.org/10.1136/bmjgh-2017-000668

18.

Kessler

R. C.

Barker

P. R.

Colpe

L. J.

Epstein

J. F.

Gfroerer

J. C.

Hiripi

Howes

M. J.

Normand

S.-L. T.

Manderscheid

R. W.

Walters

E. E.

Zaslavsky

A. M.

(2003). Screening for serious mental illness in the general population. Archives of General Psychiatry, 60(2), 184-189. https://doi.org/10.1001/archpsyc.60.2.184

19.

Kidman

Kohler

H.-P.

(2020). Emerging partner violence among young adolescents in a low-income country: Perpetration, victimization and adversity. Plos One, 15(3), e0230085. https://doi.org/10.1371/journal.pone.0230085

20.

Letourneau

E. J.

Brown

D. S.

Fang

Hassan

Mercy

J. A.

(2018). The economic burden of child sexual abuse in the United States, 79, 413-422. https://doi.org/10.1016/j.chiabu.2018.02.020

21.

Matzopoulos

Bowman

Butchart

Mercy

J. A.

(2008). The impact of violence on health in low- to middle-income countries. International Journal of Injury Control and Safety Promotion, 15(4), 177-187. https://doi.org/10.1080/17457300802396487

22.

Meinhart

Seff

Troy

Mcnelly

Vahedi

Poulton

Stark

(2021). Identifying the Impact of Intimate Partner Violence in Humanitarian Settings: Using an Ecological Framework to Review 15 Years of Evidence. International Journal of Environmental Research and Public Health, 18, 6963. https://doi.org/10.3390/ijerph18136963

23.

OCHA (2021). Colombia: Humanitarian Response Plan Summary 2021 . UN Office for the Coordination of Humanitarian Affairs https://reliefweb.int/report/colombia/colombia-humanitarian-response-plan-summary-2021

24.

OCHA FTA (2021). Colombia humanitarian response plan 2021. United Nations Office for the Coordination of Humanitarian Affairs Financial Tracking Service. https://fts.unocha.org/appeals/1016/summary

25.

Parcesepe

Stark

Roberts

Boothby

(2016). Measuring Physical Violence and Rape Against Somali Women Using the Neighborhood Method. Violence Against Women, 22(7), 798-816. https://doi.org/10.1177/1077801215613852

26.

Riger

Staggs

(2004). The Impact of Intimate Partner Violence on Women’s Labor Force Participation. National Institute of Justice. Final Report (No. 207143; p. 134) https://nij.ojp.gov/library/publications/impact-intimate-partner-violence-womens-labor-force-participation-final-report

27.

Robst

Smith

(2008). The effect of childhood sexual victimization on Women’s income. Eastern Economic Journal, 34(1), 27-40. https://doi.org/10.1057/palgrave.eej.9050017

28.

Rockhill

Newman

Weinberg

(1998). Use and misuse of population attributable fractions. American Journal of Public Health, 88(1), 15-19. https://doi.org/10.2105/ajph.88.1.15

29.

Seff

Stark

(2019). A sex-disaggregated analysis of how emotional violence relates to suicide ideation in low- and middle-income countries. Child Abuse & Neglect, 93, 222-227. https://doi.org/10.1016/j.chiabu.2019.05.008

30.

Shamu

Munjanja

Zarowsky

Shamu

Temmerman

Abrahams

(2018). Intimate partner violence, forced first sex and adverse pregnancy outcomes in a sample of Zimbabwean women accessing maternal and child health care. BMC Public Health, 18(1), 595. https://doi.org/10.1186/s12889-018-5464-z

31.

Siddique

(2018). Violence and Female Labor Supply. http://ftp.iza.org/dp11874.pdf

32.

Stark

Ager

(2011). A systematic review of prevalence studies of gender-based violence in complex emergencies. Trauma, Violence, & Abuse, 12(3), 127-134. https://doi.org/10.1177/1524838011404252

33.

Stark

Warner

Lehmann

Boothby

Ager

(2013). Measuring the incidence and reporting of violence against women and girls in Liberia using the “neighborhood method. Conflict and Health, 7(1), 20. https://doi.org/10.1186/1752-1505-7-20

34.

Stöckl

March

Pallitto

Garcia-Moreno

WHO Multi-country Study Team (2014). Intimate partner violence among adolescents and young women: Prevalence and associated factors in nine countries: A cross-sectional study. BMC Public Health, 14(1), 751-751. PubMed https://doi.org/10.1186/1471-2458-14-751

35.

UCDP . n.d. Colombia: Number of deaths 1889-2020 [Map]. Uppsala Conflict Data Program (UCDP) https://ucdp.uu.se/country/100

36.

UNGA (2015). Transforming our world: The 2030 agenda for sustainable development. United Nations General Assembly.

37.

UNICEF (2019). The economic burden of violence against children: Nigeria study. https://www.unicef.org/nigeria/media/2451/file/Economic%20burden%20of%20violence%20against%20children.pdf

38.

University of Limerick and NUI Galway (2019). Economic and social costs of violence against women in South Sudan: Summary report. NUI Galway. North (p. 36).

39.

Vyas

Meinhart

Troy

Brumbaum

Poulton

Stark

(2021). The Economic Cost of Violence Against Women and Girls in Low- and Middle-Income Countries: A Systematic Review of the Evidence. Trauma, Violence, & Abuse, 152483802110160. https://doi.org/10.1177/15248380211016018

40.

WHO (2001). Macroeconomics and health (investing in health for economic development). World Health Organization.

41.

WHO (2021). Violence against women prevalence EstimatesWorld health organization: Sexual and reproductive health and research, 2018. https://www.who.int/publications/i/item/9789240022256

42.

World Bank . GDP per capita. (n.d.). Work Bank. Retrieved July 22, 2021, from https://data.worldbank.org/indicator/NY.GDP.PCAP.CD?locations=CO https://data.worldbank.org/country/colombia?view=cha

The Economic Burden of Intimate Partner Violence in Colombia: Estimated Health Costs Among Females Aged 13–24

Abstract

Background

Methods

Results

Conclusion

Keywords

Introduction

Setting

Methodology

Data Source

Measurement

Estimating costs

Step one: estimate the national prevalence

Step two: estimate impact of intimate partner violence

Step three: establish the population attributable fractions

Step four: develop a costing model

Results

Demographics

Health Outcomes

Economic Burden

Sensitivity analyses

Discussion

Conservative estimates with high costs

Conflict-specific considerations

Limitations

Conclusion

Footnotes

Declaration of Conflicting Interests

Funding

ORCID iDs

Note

Author Biographies

References