Drug overdose among women in intimate relationships: The role of partner violence, adversity and relationship dependencies.

BACKGROUND: This study examines the relationship between experiencing intimate partner violence (IPV), exposure to prior childhood adversity, lifetime adverse experiences, drug-related relationship dependencies with intimate partners and overdose, hospitalization for drug use, friends and family members who overdosed and witnessing overdose.
METHODOLOGY: This paper included a sample of 201 women who use drugs in heterosexual relationships with criminal justice-involved men in New York City. We included measures of experiencing overdose, hospitalization for drug use, witnessing overdose, and having friends and family who overdosed. Intimate partner violence consisted of either 1) none/verbal only, 2) moderate and 3) severe abuse. Dichotomous indicators of drug-related relationship dependencies included financial support, drug procurement, splitting and pooling drugs. A scale measured cumulative exposure to childhood adversity and lifetime exposures to adverse events. This paper hypothesized that experiencing moderate and severe IPV, drug-related dependencies and exposure to prior childhood and lifetime adversity would be associated with a greater risk of experiencing overdose, hospitalization for drug use, witnessing overdose and having friends and family members who overdosed. Generalized linear modeling with robust variance estimated relative risk ratios that accounted for potential bias in confidence intervals and adjusted for race, ethnicity, education and marital status.
RESULTS: We found experiencing moderate or severe IPV was associated with ever being hospitalized for drug use and having a family member who experienced overdose. Experiencing moderate IPV was associated with increased risk of witnessing overdose, Partner drug dependencies were associated with overdose, ever being hospitalized for drug use, witnessing overdose, and having a family member or friend who experienced overdose. Childhood and lifetime adversity exposures were significantly associated with increased risk of overdose, ever being hospitalized for drug use, ever witnessing overdose and having a friend and family member who overdosed.
CONCLUSION: Findings underscore the intersection of experiencing IPV and drug-related relationship dependencies, childhood adversity and lifetime adversity in shaping experiences of and witnessing overdose among women who use drugs. They highlight the urgent need to address IPV, adversity experiences and drug-related relationship dependencies in overdose prevention for women who use drugs.

Introduction

In the United States, drug overdose rates among women have increased exponentially over the past two decades.[1][2][3] Overdose rates among women between 30–65 in the United States increased by 260% from 1999–2017 with rates growing the most due to overdose from opioids.[3] Between 1999–2015, opioid overdose deaths increased by 471% among women compared to 218% among men.[4][5] The greatest increase occurred in deaths involving synthetic opioids, which increased by 850% among women during the same period.[4][5] Racial and ethnic disparities exist in the rates of overdose, in that black women are disproportionately impacted by overdose from opioids particularly fentanyl.[6] Between 2011–2016, overdose deaths from fentanyl increased the most among African Americans with a growth of 140.6% per year compared to Hispanics (118.3%) and whites (108.8%).[6] Despite rapidly rising rates of fatal and non-fatal overdose, research on risk factors associated with overdose specifically among women is limited.

Intimate partner violence, relationship dependencies and overdose

The risk environment framework provides a multi-level conceptualization of mechanisms that explain the association between overdose and experiencing IPV, a history of child adversity and adverse life experiences.[7][8][9][10] The risk environment perspective presumes that intrapersonal factors including intimate partner violence, gender norms and relationship power inequities interacts with individual characteristics to shape the occurrence of drug use, and risk of drug-related harms of overdose.[8] A risk environment that includes being forced to have sex against one’s will or without a condom as well as being kicked, slammed against a wall, beaten, punched, and choked is associated with increased substance use dependence, greater frequency/quantity of use among women and also may lead to increased risk of overdose.[11][12][13][14][15][16]

Prior research has shown that following hospitalization for drug use, people who use opioids are at greater risk of returning to drug use and overdose due to lower tolerance and experiencing symptoms of withdrawal.[17] Emergency department hospitalization is an important yet understudied feature of the overdose risk environment for women who use opioids. Hospitals may be an opportune setting of the risk environment to deliver both overdose prevention through naloxone distribution and IPV services for women who are hospitalized due to drug use.[17] No studies to date have examined the associations among women between IPV and the specific outcomes of overdose, hospitalization from drug use, witnessing overdose, and having friends or family who have overdosed.

One explanatory mechanism for relationships between IPV and overdose is that women may use drugs to cope with negative affective experiences due to prior exposure to adversity such as IPV and childhood abuse (Carbone-Lopez et al., 2006; El-Bassel et al., 2004; El-Bassel et al., 2014; Lipsky et al., 2005).[13][18][19] Prior research by El-Bassel et al[12] and Amaro et al[20] found that women who experienced IPV coped with adverse experiences by using drugs, which could lead to overdose. Testa et al[16] found that use of illicit drugs was associated with an increased risk of IPV compared to women who did not use illicit drugs. Conversely, women who use drugs may be at greater risk of experiencing IPV because partners may perceive them as vulnerable to victimization, an increased risk of relationship conflict, lack of resources to leave violent partners and other factors.[16][21][22][23]

Inequities in drug-related relationship dependencies is another factor of the micro risk environment that may increase the possibility of overdose for women.[23][24][25] Women are more likely to rely on intimate partners to procure illicit substances and to teach them how to administer drugs; thus, shaping the unique microsocial contexts of overdose risk environments.[19][25] New insights regarding relationship dependencies and overdose could inform future research into couple-focused interventions that incorporate both partners’ drug use and sharing behaviors into overdose prevention interventions for women who use drugs.

Childhood and lifetime adverse exposures and overdose

In addition to IPV, literature suggests that coping with prior adverse experiences experienced during childhood and throughout women’s lifetime is an individual-level factor of the risk environment that may heighten risk of overdose.[26][27][28] Childhood adverse experiences may create a vulnerability to using greater quantities and higher frequencies of substance use to cope with the psychological sequelae of trauma; thus, increasing the risk of overdose.[26][27][28] Within the risk environment framework, childhood adversity and exposure to violence in addition to IPV across the life-course are individual factors that predispose women to greater vulnerabilities to using substances and experiencing overdose later in life.[28] Lake et al.[27] found that exposure to physical, sexual, emotional and physical abuse and neglect were associated with non-fatal overdose in a sample of 552 women.

Markers of overdose risk among women who use drugs

Witnessing overdose or having friends and family members who have experienced overdose is a factor of the overdose risk environment that emphasizes the importance of overdose prevention within the micro-social contexts of interpersonal relationships.[29][30] People who use drugs are embedded in familial and social networks with many other drug-using individuals; thus, increasing their risk of knowing others who have overdosed. Research shows that people who use drugs are at greater risk of witnessing overdose compared to people who do not use such substances.[29]30][31] It is estimated that from 58% to 86% of all overdoses occur in the presence of witnesses who could potentially intervene during the observed overdose.[32] A systematic review by Martins et al[33] found lifetime prevalence of witnessing a drug overdose was 73.3% with a range of 50% to 96% in 17 studies of people who use drugs. Having family members and friends who experienced overdose is an understudied social factor of the overdose risk environment.

Gaps in the literature

Several gaps persist in research examining the overdose risk environment of women who use drugs in intimate partnerships. First, women, particularly racial and ethnic minorities, are underrepresented in extant research on factors of the risk environment that are associated with overdose in the United States despite mounting literature suggesting exposure to adversities are associated with greater severity of substance use. Second, research on overdose prevention neglects risk factors that disproportionately impact women, namely IPV and childhood adversity. Third, the association between drug-related relationship inequities and overdose among women remains unknown. Finally, little is known about how exposure to IPV and drug-related relationship inequities shape social factors of the overdose risk environment including witnessing overdose and having friends and family members who have overdosed. Investigating the association between IPV and overdose risk factors could inform future overdose prevention interventions to address the unique risk factors facing women who use drugs in the United States.

The following paper described the frequency of overdose, witnessing overdose of friends and family members, and hospitalization due to drug use in a sample of women in New York City. We examined the association between the experience of IPV and several indicators of overdose risk that included 1) experiencing overdose, 2) being hospitalized because of drug use, 3) witnessing overdose, and 4) having family and 5) friends who experienced overdose after adjusting for covariates of race, ethnicity, less than high school education and marital status. This paper also examined the association between drug-related relationship inequities (financial, splitting drugs, pooling drugs) and indicators of overdose risk after adjusting for potential confounders. Finally, this paper investigated the association between exposure to childhood and other lifetime adversities and overdose risks after adjusting for potential confounders. We hypothesized that 1) experiencing intimate partner violence and 2) drug-related relationship dependencies as well as lifetime exposure to 3) childhood adversity and 4) lifetime adversity would be associated with indicators of overdose risk after adjusting for potential confounders.

Methods

Data and procedures

Data consisted of a subset of 201 women who participated in Project PACT, a couple-focused randomized clinical trial of an HIV prevention intervention for men undergoing community corrections and their female intimate partners.[34] Following sentencing, participants were recruited from community correction provider sites by research assistants who provided informational fliers to male clients.[34] Because data was from a couples-focused intervention, women were recruited through first screening their male partners. All male partners were involved in community corrections at the time of recruitment and identified their female partners for inclusion in the study. Women who consented were administered a screening instrument that determined eligibility. The sample was restricted to include only women who reported lifetime use of illicit drugs from a starting sample of 239 women who were partners of men in community corrections. The Columbia University Institutional Review Board approved this study protocol and written consent was obtained from all participants.

Inclusion/Exclusion criteria

Couples were eligible to participate in the study if 1) they were at least 18 years of age, 2) identified each other as their primary partner, 3) the length of the relationship was greater than 3 months 4) they had unprotected vaginal or anal sex within the past 90 days 5) at least one partner had exposure to an outside HIV risk in the past year including unprotected sex with another partner, shared syringes, tested positive for an STI or HIV, 6) the male partner had the male partner had been mandated to some form of community supervision in the past 90 days (e.g. probation, parole, ATI, drug court, community court).

Measures

Participants were interviewed by trained research staff and administered a structured questionnaire. All participants were reimbursed 265$ for participation in all research-related activities including baseline questionnaire including screening, baseline, biological specimen and follow up assessments.

Dependent variables

Variables related to overdose included 1) experiencing overdose, 2) hospitalization due to drug use, 3) witnessing an overdose, and 3) having friends and family members who had experienced overdose. Overdose consisted of a dichotomous variable based on self-reported lifetime and past year experience of overdose (losing consciousness) while using drugs and whether the drugs were heroin, opiate pain relievers or tranquilizers. Hospitalization because of drug use included having ever received emergency room treatment for drug or alcohol use-related problems (overdose, loss of consciousness, physical problems or injuries caused by use of drugs or alcohol). Witnessing an overdose included a dichotomous variable indicating ever witnessing an overdose (loss of consciousness) while using drugs. Dichotomous variables measured having one or more 1) family members or 2) friends experience an overdose from using drugs.

Independent variables

Intimate partner violence consisted of a categorical variable with 21 questions measuring intimate partner violence based on the Conflict Tactics Scale with high reliability and validity.[35] A 4-level categorical variable was created with categories measuring exposure to 0) no physical/sexual, 1) verbal aggression only (no moderate /severe physical or sexual abuse) 2) only moderate and 3) severe sexual/physical abuse.

Lifetime adverse experiences was based on the Stressful Life Events Screening Questionnaire.[35] The scale summed question items consisting of dichotomous variables of ever experiencing 1) physical assault, 2) being threatened with a weapon, 3) witnessing death/injury, 4) being in an extremely frightening situation, 5) having a loved one who died because of accident, suicide or homicide, 6) experiencing physical force in a robbery and 7) experiencing serious injury and 8) other situation where life was in danger (Chronbach’s alpha .87) (range 0–8).

Childhood adverse experiences consisted of a scale based on the Revised Inventory of Adverse Childhood Experiences summing 6 dichotomous variables.[36] (Questions measured 1) verbal (‘Before the age of 18, has a parent or caregiver repeatedly ridiculed you, put you down, ignored you, or told you were no good?’) 2) sexual (Before the age of 17, did anyone ever touch private parts of your body, made you touch their body, or made you have sex against your wishes?), and 3) physical abuse, (‘Before the age of 17, did a parent, caregiver or other person ever slap you repeatedly, beat you, or otherwise attack or harm you?’), 4) witnessing intimate partner violence (‘Before the age of 17, did you ever witness your parents hitting, slapping, beating or physically hurting each other?’), 5) having been placed in foster care (‘Before the age of 17, were you ever removed from your home and placed in foster care?’) and 6) running away from home (‘Before the age of 17, did you run away from home?’) (Chronbach’s alpha = .89) (range 0–6).[37]

Drug-related relationship dependencies included dichotomous variables measuring if more than half the time participants 1) purchased drugs for intimate partners, 2) relied on intimate partners to purchase their drugs, and 3) pooled and 4) split drugs with their intimate partner. Question items were summed to produce a scale reflecting cumulative dependencies on partners for drugs (Chronbach’s alpha = .84) (range 0–4).

Socioeconomic factors consisted of a categorical variable measuring race (African American, Non-Hispanic White, Asian), dichotomous indicators of Hispanic ethnicity, less than high school education and marital status, and a continuous variable measuring age.

Drug and alcohol use included dichotomous variables indicating lifetime use of binge drinking (4 or more drinks within 5 hours) and use of heroin, prescription pain relievers, cocaine, crack, stimulants, tranquilizers and other drugs.

Statistical analyses

Descriptive analyses

Descriptive statistics provided proportions and counts for dichotomous and categorical data as well as median and interquartile range (IQR) estimates for continuous variables. Bivariate analyses consisted of tests for significant differences between adversity exposure variables and experiencing and witnessing overdose as well as having friends and family members who experienced an overdose using chi-square tests comparing categorical variables and Wilcoxon Rank Sum tests for significant differences between categorical and continuous variables.[38][39][40]

Hypothesis testing in unadjusted and adjusted models

Generalized linear modeling tested the hypotheses that intimate partner violence, relationship dependencies, childhood and lifetime adversity would be associated with 5 indicators of overdose risk (experiencing overdose, hospitalization due to drug use, witnessing an overdose, having friends and family who experienced overdose).[41][42][43][44][45] Hypothesis tests of IPV, childhood sexual abuse, lifetime experiences of adversity and drug-related relationship dependencies were performed in separate models to avoid potential issues of collinearity.[39][42]

Parameter estimates of relative risk ratios (RR) tested hypotheses by measuring the likelihood of overdose risk and other overdose indicators based on exposure to intimate partner violence, drug-related relationship dependencies, childhood and lifetime adverse compared to individuals who were not exposed to adverse experiences.[43][45] Sensitivity analysis with other variables were performed with age, binge drinking and several types substances (i.e cocaine, crack, heroin). None of these variables were significant and due to the small sample size were not included in the final regression model estimating the association between IPV and overdose. Final models adjusted for race, ethnicity, education, and marital status. All participants reported lifetime illicit drug use prohibiting covariance adjustment for illicit drug use. The error variance correction method provided estimates that were robust to bias in confidence intervals.[46] All analyses were performed in STATA 15.[47]

Results

Descriptive findings

Indicators of overdose risk. Overall 12.4% of the women reported experiencing overdose in their lifetime (n = 25) (Table 1). Of those who experienced overdose, 6.0% (n = 12) reported overdose on heroin, 5.0% (n = 10) overdosed on benzodiazepines and 3.0% (n = 6) overdosed on opiates. Hospitalization due to drug or alcohol use was reported by 16.9% (n = 34) of the sample. More than a quarter reported having a friend (29.3%, n = 59) and a fifth reported a family member (20.9%, n = 42) who overdosed. More than a quarter of the sample reported witnessing an overdose at some point in their life (27.4%, n = 55).

Table 1

Descriptive characteristics of women who ever used illicit drugs (n = 201).

			Female partners
			% or median	(n or IQR)
Overdose
	Lifetime		12.4	(25)
	Type of drug
		Heroin	6.0	(12)
		Other opiates	3.0	(6)
		Benzodiazepine	5.0	(10)
Ever hospitalized from drug/alcohol use			16.9	(34)
Social networks
	Friend overdose		29.3	(59)
	Family member overdose		20.9	(42)
Witnessing overdose
	Ever witness overdose		27.4	(55)
Access to naloxone
		Ever heard of naloxone	13.4	(27)
		Ever talked with others about naloxone	8.7	(18)
		Ever used naloxone to reverse overdose	2.0	(4)
Interpersonal factors
	Intimate partner violence IPV)
		None	50.3	(101)
		Verbal only	11.0	(22)
		Moderate sexual/physical abuse only	17.4	(35)
		Severe sexual/physical abuse	21.4	(43)
Relationship dependencies
		Drug dependency	30.9	(62)
		Drug provider	33.3	(67)
		Pool money to buy drugs	11.0	(22)
		Do drugs or split drugs	16.9	(34)
	Childhood adversity		2.1	(0, 6)
	Lifetime adversity		2.3	(0, 8)
Illicit drug use
		Binge drinking	46.8	(94)
		Heroin	24.4	(49)
		Prescription pain relievers	19.4	(39)
		Cocaine	40.3	(81)
		Crack	29.4	(59)
		Stimulants	8.5	(17)
		Tranquilizers	15.4	(31)
		Other drugs	30.9	(62)
Black race			70.7	(142)
Hispanic ethnicity			30.4	(61)
Income (<850$/month)			53.2	(107)
Education			33.3	(67)
Married			31.8	(64)
Age			34.8	(23.09, 64.85)

Intimate partner violence. More than a fifth of the sample reported history of experiencing severe (21.4%, n = 43), nearly a fifth reported moderate abuse (17.4%, n = 35) and 11.0% (n = 22) reported only verbal abuse by current intimate partners.

Lifetime and childhood adversity. The median exposures to adversity was 2.1 for childhood adversity (IQR = 0, 6.0) and 2.3 (IQR = 0,8) for lifetime adversity.

Relationship dependencies. The most prevalent relationship dependency was providing money to partners to purchase drugs (33.3%, n = 67) followed by depending on partners to buy them drugs (30.9%, n = 62), pooling drugs with partners (11.0%, n = 22) and splitting drugs with partners (16.9%, n = 34). The mean score for relationship dependencies was 2.00 (0–4).

Overdose and hospitalization

Socioeconomic factors. A majority of the sample were black (70.7%, n = 142) and 30.4% (n = 61) were of Hispanic ethnicity. A third of the sample reported less than a high school education (33.3%, n = 67) and 31.8 (n = 64) were married. The mean age of the sample was 34.8 years (IQR = 23.09, 64.85).

Nearly a quarter of women reported using heroin (24.4%, n = 49) in their lifetime and nearly a fifth reported using prescription pain relievers (19.4%, n = 39). Tranquilizer use was reported by 15.4% (n = 31) of the sample. Cocaine use was reported by 40.3% (n = 81) and crack cocaine was reported by 29.4% (n = 59) of the women.

Hypothesis 1: Intimate partner violence and indicators of overdose risk

Multivariable analysis did not rule out the null hypothesis regarding an association between severe IPV (3) or moderate IPV (2) compared to those reporting no physical/sexual IPV (0) and experiencing overdose (Table 2). Experiencing moderate (RRadjusted = 2.1, 95% CI = 1.0, 4.6) and severe IPV (RRadjusted = 2.0, 95% CI = .9, 4.2) were associated with increased risk of reporting prior hospitalization for drug use. Experiencing moderate IPV was significantly associated with increased risk of ever witnessing an overdose compared to participants who were not exposed to IPV (RRadjusted = 1.9, 95% CI = 1.1, 3.2) (Table 3). Exposure to moderate IPV (RRadjusted = 2.0, 95% CI = .9, 4.1) and severe IPV (RRadjusted = 2.1, 95% CI = .9, 4.1) was associated with increased risk of having a family member who experienced an overdose (Table 4).

Table 2

Bivariate and multivariate associations between exposures to intimate partner violence, adversities, relationship factors, and overdose and hospitalization for women who reported ever using illicit drugs (n = 201).

		Overdose ever				Ever hospitalized for drug use
		Unadjusted		Adjusted		Unadjusted			Adjusted
		RR	95%CI	RR	95%CI	RR	95%CI	RR	95%CI
Interpersonal trauma
Intimate partner violence
	None	ref	ref	ref	ref	ref	ref		ref
	Verbal	.8	(.2, 3.5)	-	-	1.1	(.4, 3.8)	1.2	(.4, 3.9)
	Moderate	1.3	(.5, 3.5)	-	-	2.2	(1.0, 4.7)	2.1	(1.0, 4.6)
	Severe	1.5	(.6, 3.6)	-	-	2.0	(1.0, 4.2)	2.0	(.9, 4.2)
Relationship dependencies
	Drug financial dependency	2.4	(1.2 5.0)	2.0	(1.0, 4.0)	1.8	(1.0, 3.3)	1.7	(.9, 3.2)
	Drug financial provider	3.6	(1.7, 7.6)	2.9	(1.3, 6.2)	3.2	(1.7, 6.1)	3.1	(1.7, 5.8)
	Pool money to buy drugs	5.4	(2.8, 10.6)	4.4	(2.0, 9.7)	2.5	(1.3, 4.8)	2.4	(1.2, 4.8)
	Do drugs or split drugs	4.5	(2.3, 9.1)	3.9	(1.9, 8.0)	3.0	(1.7, 5.5)	3.0	(1.7, 5.5)
	Scale	1.7	(1.3, 2.0)	1.6	(1.2, 2.0)	1.4	(1.2, 1.7)	1.4	(1.2, 1.7)
Childhood adversity		1.3	(1.1, 1.6)	1.3	(1.1, 1.6)	1.2	(1.0, 1.4)	1.2	(1.0, 1.4)
Lifetime adversity		1.2	(1.1, 1.4)	1.2	(1.0, 1.3)	1.2	(1.1, 1.3)	1.2	(1.1, 1.3)

Adjusted models included covariates of race, ethnicity, education, marital status; RR: Relative Risk

Table 3

Bivariate and multivariate associations between exposures to intimate partner violence, adversities, relationship dependencies and witnessing overdose for women who reported ever using illicit drugs (n = 201).

		Ever witness overdose
		Unadjusted		Adjusted
		RR	95% C.I	RR	95% C.I
Interpersonal trauma
Intimate partner violence
	None	ref	ref	ref	ref
	Verbal	1.0	(.5, 2.4)	1.0	(.4, 2.3)
	Moderate	1.9	(1.0, 3.0)	1.9	(1.1, 3.2)
	Severe	1.3	(.7, 2.3)	1.2	(.7, 2.2)
Relationship dependencies
	Drug financial dependency	1.5	(1.0, 2.3)	1.4	(.9, 2.2)
	Drug financial provider	1.5	(1.0, 2.4)	1.5	(1.0, 2.4)
	Pool money to buy drugs	2.8	(1.8, 4.2)	2.7	(1.7, 4.2)
	Do drugs or split drugs	2.0	(1.3, 3.2)	1.9	(1.2, 3.0)
	Scale	1.3	(1.1, 1.5)	1.2	(1.1, 1.4)
Childhood adversity		1.3	(1.1, 1.4)	1.3	(1.1, 1.4)
Lifetime adversity		1.3	(1.2, 1.4)	1.3	(1.2, 1.4)

Adjusted models included covariates of race, ethnicity, education, marital status; RR: Relative Risk

Table 4

Bivariate and multivariate associations between exposures to intimate partner violence, adversities, relationship dependencies and family and friends experiencing overdose for women who reported ever using illicit drugs (n = 201).

		Friends experience overdose				Family experience overdose
		Unadjusted		Adjusted		Unadjusted		Adjusted
		RR	95% C.I	RR	95% C.I	RR	95% C.I	RR	95% C.I
Intimate partner violence
	None	ref	ref	ref	ref	ref	ref	ref	ref
	Verbal	.8	(.3, 1.8)	-	-	2.0	(.9, 4.6)	1.9	(.84, 4.4)
	Moderate	1.1	(.6, 1.9)	-	-	2.1	(1.0, 4.2)	2.0	(1.0, 4.1)
	Severe	1.0	(.6, 1.8)	-	-	2.0	(1.0, 4.0)	2.1	(1.1, 4.2)
Relationship dependencies
	Drug financial dependency	1.5	(1.0, 2.4)	1.3	(.9, 2.0)	1.9	(1.1, 3.2)	1.9	(1.1, 3.2)
	Drug financial provider	2.2	(1.5, 3.4)	1.9	(1.2, 2.9)	2.2	(1.3, 3.7)	2.2	(1.3, 3.8)
	Pool money to buy drugs	3.6	(2.6, 5.0)	2.9	(2.0, 4.2)	2.9	(1.7, 4.9	2.8	(1.6, 4.8)
	Do drugs or split drugs	2.5	(1.7, 3.7)	2.2	(1.5, 3.3)	2.2	(1.3, 3.8)	2.1	(1.2, 3.7)
	Scale	1.4	(1.2, 1.5)	1.3	(1.1, 1.5)	1.4	(1.2, 1.6)	1.3	(1.1, 1.6)
Childhood adversity		1.1	(1.0, 1.3)	1.1	(1.0, 1.3)	1.2	(1.1, 1.4)	1.2	(1.1, 1.4)
Lifetime adversity		1.2	(1.1, 1.3)	1.2	(1.0, 1.3)	1.2	(1.1, 1.3)	1.2	(1.1, 1.3)

Adjusted models included covariates of race, ethnicity, education, marital status; RR: Relative Risk

Hypothesis 2: Childhood adverse events and indicators of overdose risk

Each additional exposure to childhood adverse events was associated with an increase in the risk of experiencing overdose (RRadjusted = 1.3 95% CI = 1.1, 1.6). Exposure to childhood adversity was associated with an increase in risk of prior hospitalization due to drug use (RRadjusted = 1.2, 95% CI = 1.0, 1.4). Each additional exposure to childhood adversity was associated with greater risk of having witnessed an overdose (RRadjusted = 1.3, 95% CI = 1.1, 1.4). Each additional exposure to childhood adversity (RRadjusted = 1.2, 95% CI = 1.1, 1.4) was associated with an increase in risk of a family member experiencing an overdose. Each additional exposure to childhood adversity was associated with greater risk of having a friend who overdosed (RRadjusted = 1.1, 95% CI = 1.0, 1.3).

Hypothesis 3: Lifetime adverse experiences and indicators of overdose risk

Each increase in lifetime exposure to adversity was associated with greater risk of experiencing an overdose (RRadjusted = 1.2, 95% CI = 1.0, 1.3). Each additional lifetime exposure to adversity was associated with greater risk of witnessing an overdose after adjusting for potential confounders (RRadjusted = 1.2, 95% CI = 1.1, 1.3). Each additional increase in lifetime exposures to adversity was associated with greater risk of a family member experiencing an overdose (RRadjusted = 1.2, 95% CI = 1.1, 1.3). Each additional exposure to lifetime adverse experiences was associated with greater risk of friends experiencing an overdose (RRadjusted = 1.2, 95% CI = 1.1, 1.3).

Hypothesis 2: Relationship dependencies and indicators of overdose risk

Participants who reported relying on their partners to buy them drugs (RRadjusted = 2.0, 95% CI = 1.0, 4.0), paying for their partners’ drugs (RRadjusted = 2.9, 95% CI = 1.3, 6.2), pooling money to buy drugs (RRadjusted = 4.4, 95% CI = 2.0, 9.7), and doing drugs or splitting drugs with their partners (RRadjusted = 3.9, 95% CI = 1.9, 8.0) were associated with increased relative risk of ever experiencing an overdose. Each additional relationship dependency was associated with an increase of 60% in the risk of experiencing overdose (RRadjusted = 1.6, 95% CI = 1.2, 2.0). Participants who paid for their partners’ drugs (RRadjusted = 3.1, 95% CI = 1.7, 5.8), pooled money to buy drugs (RRadjusted = 2.4, 95% CI = 1.2, 4.8) and did drugs or split drugs with their intimate partners (RRadjusted = 3.0, 95% CI = 1.7, 5.1) were more likely to report having a prior hospitalization due to substance use. Each additional relationship dependency was associated with a 42% increase in risk of prior hospitalization due to substance use (RRadjusted = 1.4, 95% CI = 1.2, 1.7).

Participants who paid for their partners’ drugs (RRadjusted = 1.5, 95% CI = 1.0, 2.4), pooled money to buy drugs (RRadjusted = 2.7, 95% CI = 1.7, 4.2), and did drugs or split drugs with their partners (RRadjusted = 1.9, 95% CI = 1.2, 3.0) were more likely to report ever witnessing an overdose. Each additional relationship dependency was associated with greater risk of ever witnessing an overdose (RRadjusted = 1.2, 95% CI = 1.08, 1.43). Participants who were dependent on their partners to purchase drugs (RRadjusted = 1.9, 95% CI = 1.1, 3.2), paid for their partners’ drugs (RRadjusted = 2.2, 95% CI = 1.3, 3.8), pooled money to buy drugs (RRadjusted = 2.8, 95% CI = 1.6, 4.8), and did or split drugs with their partners (RRadjusted = 2.1, 95% CI = 1.2, 3.7) were more likely to report having a friend who experienced an overdose. Each additional relationship dependency was associated with an increase in the relative risk of having a family member who overdosed by 30% (RRadjusted = 1.3, 95% CI = 1.1, 1.6). Participants who paid for their partners’ drugs (RRadjusted = 1.9, 95% CI = 1.2, 2.9), pooled money to buy drugs (RRadjusted = 2.9, 95% CI = 2.0, 4.2), and did or split drugs with their partners (RRadjusted = 2.2, 95% CI = 1.5, 3.3) were more likely to report having a friend who experienced an overdose. Each additional relationship dependency was associated with an increase in the risk of having a friend who experienced an overdose by 30% (RRadjusted = 1.3, 95% CI = 1.1, 1.5).

Discussion

This study examined the association between several social factors of the risk environment including 1) exposures to 4 levels of IPV (none, verbal, moderate, severe), 2) drug-related relationship dependencies, and 3) history of childhood adversities and relative risk of experiencing an overdose, hospitalization because of drug use, witnessing an overdose and having friends or family who experienced an overdose among women who use drugs in New York City. Exposure to IPV was not associated with experiencing overdose. Findings from this paper supported hypotheses that exposure to moderate and severe IPV would be associated with ever being hospitalized for drug use. Hospitalization due to losing consciousness, injury or drug poisoning may provide a marker for experiencing non-fatal overdose. In addition to IPV, women who paid for their partners’ drugs, were dependent on their partners for money to purchase drugs, split drugs with their partners and pooled money to buy drugs were more likely to experience overdose, be hospitalized for drug use, witness an overdose, and have friends and family who had overdosed. Findings supported hypotheses that each cumulative exposure to childhood adversity was associated with increased risk of experiencing overdose, and ever being hospitalized for drug use. Future research must include multiple markers of overdose and overdose risk in assessments of overdose and drug use history.

Findings from this study are consistent with prior literature suggesting that exposures to childhood adversities are associated with increased risk of overdose and substance use severity among people who use illicit drugs.[27][48] Lake et al[27] found that exposures to childhood adversity was associated with increased risk of non-fatal overdose among people who inject drugs. To our knowledge, this is the first study to examine the association between IPV, relationship dependencies and overdose among women who use illicit drugs. However, findings from this study are congruent with previous literature that suggests exposure to IPV is significantly associated with substance use severity among women who use drugs.[16][49] Additional research is needed that further elucidates the relationships between IPV and overdose among women who use drugs.

Implications for substance misuse treatment and overdose prevention

This paper generated several implications for substance use treatment and overdose prevention interventions as well as services for women exposed to intimate partner violence, drug-related relationship dependencies and childhood and lifetime adverse experiences. Research is needed that investigates if exposure to IPV may increase attitudinal barriers to treatment resulting in greater vulnerabilities to overdose. Future research must investigate whether women who have experienced IPV have lower self-efficacy, fewer positive attitudes toward treatment, and feelings of hopelessness creating barriers to accessing treatment. In addition to IPV, prior research suggests that exposure to childhood adversity is associated with lower health seeking behaviors among women who use drugs.[49]

Moreover, prior research suggests that trauma-focused interventions are needed in substance use treatment that acknowledge the vulnerabilities to substance use and relapse for people exposed to IPV and childhood adversity.[50] However thus far, IPV and childhood adversity have not been thoroughly integrated into overdose prevention interventions for women who use drugs despite prior literature suggesting that adversity-exposed women use drugs in greater quantities and higher frequencies.[15][16][22] Future studies must investigate whether providing interventions to reduce intimate partner violence may attenuate rapidly rising rates of overdose among women who use drugs.

Findings from this study are consistent with prior literature emphasizing the importance of assessing for the co-occurrence of prior IPV exposure and substance use in emergency departments and shelters for women.[11][51][52] Moreover, overdose risk assessments must consider exposure to IPV, childhood adversity and substance use in the distribution of naloxone and identifying individuals who could benefit from overdose education. Exposure to IPV and other forms of adversity may inhibit access to naloxone, naloxone training programs and education about overdose from non-opiates and other important overdose prevention resources. Future research must evaluate the potential benefits of distributing naloxone within shelters, case management and family justice centers on reducing the risk of overdose among women who experience IPV. To our knowledge there are no studies to date that evaluate interventions to integrate harm reduction interventions into responses for women who experience intimate partner violence.

In addition to IPV and childhood adversity, mechanisms that link inequities in drug-related relationship dependencies and overdose have not been studied in prior literature. Women who depend on their intimate partners to pay for substances may be at greater risk of overdose when partners withhold payment for drugs or use financial support as a means of coercion. Prior literature suggests that financial and drug dependencies on intimate partners increases engagement in sex trading among women who use drugs who are in committed intimate relationships.[53] No prior studies have investigated the relationship between relationship dependencies and overdose among women who use drugs. Future research must examine whether financial dependency on partners to pay for drugs increases risk of overdose through mechanisms of lowered tolerance to substances, inconsistent quality of substances and increased engagement in overdose risk behaviors including injecting drugs. Additional research is needed that examines if women who depend on partners to procure their drugs may not know the quality of the substances resulting in greater risk of overdose. Studies must investigate whether women who rely on partners to provide drugs or split drugs are not aware of the quality of substances and consume doses that are too large, heightening their risk of overdose. Findings from this study suggests that additional research is necessary that investigates the benefits of providing couple-focused overdose prevention to drug-using couples that specifically addresses inequities in drug-related dependencies as a major risk factor for overdose in women.

Limitations

There are several limitations of this study worth noting. The analysis used cross-sectional data thus precluding any causal inference, rather all statistical associations are associational. Participants were recruited by their male intimate partners who were in community corrections, thus generating a non-random sample selection limiting generalizability to the general population of women. Women who have died from overdose were not included in the sample, limiting generalization of findings to women who experienced non-fatal overdose. A number of variables were measured as lifetime experiences including experiencing overdose and hospitalizations due to drug use, witnessing an overdose and having friends and family who have experienced overdose. The cross-sectional design of the study and measurement of lifetime overdose tempers the strength of the findings in this manuscript. Overdose due to substance use may increase risk of experiencing intimate partner violence due to drug and financial relationship dependencies. However, our findings are consistent with prior longitudinal research that found exposure to partner violence at baseline increased engagement substance use.[54]

The small sample size used in this study precluded inclusion of other confounders that may potentially explain the association between IPV and overdose. The small number of overdoses that were reported by this sample also restricted the number of variables that could be included. This study did not include types of substances and injection drug use, in the multivariable regressions because too few participants overdosed across the types of substances. Nonetheless, these are important variables to consider in future analyses with larger sample sizes to take into account the role of substance use and other factors into the analyses of factors that may drive overdose among women who use illicit drugs. It is possible that underreporting occurred given that the research interviews occurred at probation and may have resulted in participants not disclosing overdose and drug use history.

Finally, male partner substance use behaviors were not included in this study and provide a fruitful avenue of future research. Prior studies suggest that male partner substance use is associated with IPV severity which may exacerbate risk of overdose.[24][55][56] Future research must investigate how male-partner substance use interacts with male-partner perpetrated IPV to shape overdose risk environments of women who use drugs.

Nonetheless, findings from this study identify a population of women who are exposed to intimate partner violence at heightened risk of overdose. Future research is needed within a more recent timeframe to allow for correlational inference about the relationship between adversity and recent overdose experiences. Another limitation is this study did not investigate the potential underlying mechanisms that may be explaining associations between IPV and risk of overdose. Findings from this study call for future longitudinal epidemiological research to study the emergence of substance use and overdose following repeated exposures to adverse events. Due to issues of collinearity and sample size, all of the models examined risk factors separately that adjusted for several potential confounders. Future research with large sample sizes must examine how exposures to childhood adverse events interact with IPV to heighten overdose risk among women who use drugs.

Conclusion

Limitations notwithstanding, identifying groups at greatest risk of overdose for prevention interventions including naloxone distribution, testing drug samples, and education could inform future research into strategies to attenuate rising rates of drug overdose among women. This paper addressed a significant gap in existing literature by elucidating several relationships between experience IPV, history of childhood and other lifetime adversity and several overdose-related factors including, experience of an overdose, hospitalization because of drug use, witnessing an overdose and having friends and family members who have experienced overdose. Future research must address the unique needs of women by expanding the incorporation of IPV, drug-related relationship inequities and childhood adversity into overdose prevention research.

(DTA)

Click here for additional data file.

Transfer Alert

This paper was transferred from another journal. As a result, its full editorial history (including decision letters, peer reviews and author responses) may not be present.

12 Sep 2019

PONE-D-19-22408

Drug overdose among women in intimate relationships: The role of partner violence, trauma and relationship dependencies

PLOS ONE

Dear Dr. Marotta,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

As indicated by the Reviewers, required changes to the manuscript include addressing concerns about reverse causality, adjusting for known confounders, and developing the Discussion section for a more cohesive interpretation of the findings.

We would appreciate receiving your revised manuscript by 10/11/19. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Javier Cepeda

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We note that you have stated that you will provide repository information for your data at acceptance. Should your manuscript be accepted for publication, we will hold it until you provide the relevant accession numbers or DOIs necessary to access your data. If you wish to make changes to your Data Availability statement, please describe these changes in your cover letter and we will update your Data Availability statement to reflect the information you provide.

3. We note you have included tables to which you do not refer in the text of your manuscript. Please ensure that you refer to Tables 1 - 4 in your text; if accepted, production will need this reference to link the reader to these Tables.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: No

Reviewer #2: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This paper seeks to understand the relationship between IPV, experiences of traumatic events, relationship dependency and markers of drug overdose. Overall, it’s a very clear and concise paper, there are just a series of relatively minor issues I have with it, that need to be attended to.

First, I think the coding for IPV needs to be described in greater detail, and some of the terminology needs to be revised. In the abstract, IPV is wrongly described as a three level variables. Can the authors clarify their coding for minor versus severe phys/sex IPV experience please. In addition, could the authors reframe their language away from minor to maybe moderate IPV? I think the politics of describing violence against women as minor is worrying, given we know even small amounts of violence have negative health impacts. So this needs to be corrected throughout the paper.

P-values versus confidence intervals. Throughout the paper they have some pretty clear statistical language and are testing the null hypothesis, yet at times the authors then use p>0.05 to show association. Looking at the confidence intervals, it seems it’s because of it including 1. Given the small sample and the push away from p-values, I wonder if it would be better to drop p-values and just used 95%CI for the tables and paper. If the authors don’t then they need to revise the paper to be strict on the p<0.05 statistical rules etc.

In terms of language, I wonder if it’s right to describe witnessing violence etc. as trauma, or traumatic events, as trauma would be the psychological outcomes, e.g. PTSD.

In the discussion section, there is no previous literature cited – is this truly the case? The only time literature comes in is around implications. I would have imagined there would be a little bit of previous work on this etc.

Minor issues

In the childhood trauma measure section, the first question about verbal trauma, was this before the age 17 or ever? If ever, I wonder if this needs dropping, as it’s quite different to the other questions – or did it just get missed out in the typing.

In the description of the statistical analyses, they say they are going to give standard error estimates – surely inter-quartile range looking at the tables?

In descriptive stats section, “2.00 for lifetime traumas (0-6)” just a few more words needed to make this interpretable.

In the discussion the authors state:”A greater number of women reported experiencing hospitalization from drug use suggesting that hospitalization from drug use may provide a more accurate marker of self-reported non-fatal overdose than asking about overdose.” I think this needs unpacking further, as it’s not clearly established by the analyses presented here.

I could not see clear evidence that the paper included access to the data set as per PLoS guidelines.

Other than this I think the paper is very well written.

Reviewer #2: This study examines the trauma- and drug-related correlates of five outcomes (overdose, hospitalization for drug use, and witnessing overdose) using a convenience sample of partnered women recruited through a randomized HIV prevention intervention.

Though the authors have identified a neglected area of research, there were notable methodological and presentation issues with the manuscript. The use of lifetime outcomes (rather than recent events) tempered my enthusiasm for this exploratory analysis as was difficult to rule out reverse causality. Key confounders were missing in the models (e.g., age, types of drugs used, injection drug use, incarceration history). Given that the eligibility criteria for drug use was broad, the analysis likely includes a heterogeneous sample of drug-using women with varying levels of overdose-related risk, which was a concern.

The introduction was unusually long and did not cohesively frame the research hypotheses. For example, the exploration of hospitalization for drug use was not justified in the risk environment conceptual framework used. In fact, the authors cited multiple papers conceptualizing the risk environment – given the evolution of this framework in the fields of HIV and opioid research, better articulation will be needed to guide the reader. Conversely, the discussion was surprisingly underdeveloped and not grounded in the aforementioned literature. This manuscript needs significant revision.

Minor comment:

There were several typographical errors.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Andrew Gibbs

Reviewer #2: Yes: Ju Nyeong Park

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

28 Oct 2019

Dear Dr. Cepeda,

We are delighted to provide the following revised manuscript for consideration in the special issue “Substance Use, Misuse and Dependence: Prevention and Treatment.” We greatly appreciate the feedback provided by you and the two peer reviewers. We agree with all of the feedback and comments raised by the reviewers. We revised the manuscript and addressed the comments raised by the reviewers. We provided a marked-up copy of our manuscript that highlights changes made to the original version and is uploaded as a separate file and labeled as ‘revised manuscript with tracked changes.’ We also included an unmarked version of our revised paper without tracked changes and is labeled ‘manuscript.’ To enhance the reproducibility and transparency of our results and to meet the requirements of PLOS one, we deposited the data underlying findings from this study in Columbia University academic commons. We will provide repository information with the DOI number for the dataset. The revised manuscript meets PLOS ONE’s style requirements.

We provide a point-by-point response below to all of the feedback raised by the editor and peer reviewers.

Associate Editor

“We note that you have stated that you will provide repository information for your data at acceptance. Should your manuscript be accepted for publication, we will hold it until you provide the relevant accession numbers or DOIs necessary to access your data. If you wish to make changes to your Data Availability statement, please describe these changes in your cover letter and we will update your Data Availability statement to reflect the information you provide.”

We are committed to complying with providing repository information for our data at acceptance of this article.

There are no ethical or legal restrictions on sharing our de-identified data set. We deposited our data in Columbia Academic Commons and will share a DOI with the available data, as well as contact information for the data access committee at Columbia University if the manuscript is accepted. We would be delighted to provide any additional information.

“We note you have included tables to which you do not refer in the text of your manuscript. Please ensure that you refer to Tables 1 - 4 in your text; if accepted, production will need this reference to link the reader to these Tables.”

The authors now refer to all the tables in the text of the manuscript (Tables 1-4) so that production will have this to link the readers to the tables.

“Have the authors made all data underlying the findings in their manuscript fully available?”

We deposited our data in Columbia Academic Commons and will share a DOI with the available data, as well as contact information for the data access committee at Columbia University if the manuscript is accepted. We would be delighted to provide any additional information.

Reviewer 1:

“The coding for IPV needs to be described in greater detail, and some of the terminology needs to be revised. In the abstract, IPV is wrongly described as a three-level variables. Could the authors clarify their coding for minor versus severe phys./sex IPV experience please?”

We agree with the reviewers and clarified the description of coding for minor versus severe phys./sex IPV to accurately reflect what was done to the variable. The IPV variable was created as none (0), verbal only (1), moderate only (2) and major abuse (3).

The following revisions to the methods section of the manuscript:

“A 4-level categorical variable was created with categories measuring exposure to: 0) no physical/sexual, 1) verbal aggression only (no moderate /severe physical or sexual abuse 2) only moderate and 3) severe sexual/physical abuse.”

“In addition, could the authors reframe their language away from minor to maybe moderate IPV? I think the politics of describing violence against women as minor is worrying, given we know even small amounts of violence have negative health impacts. So, this needs to be corrected throughout the paper.”

We revised the manuscript throughout to replace the use of “minor” with “moderate.” The revised manuscript reflects these changes.

“P-values versus confidence intervals. Throughout the paper they have some pretty clear statistical language and are testing the null hypothesis, yet at times the authors then use p>0.05 to show association. Looking at the confidence intervals, it seems it’s because of it including 1. Given the small sample and the push away from p-values, I wonder if it would be better to drop p-values and just used 95%CI for the tables and paper. If the authors don’t then they need to revise the paper to be strict on the p<0.05 statistical rules etc.”

Initially we presented findings with p<.10 as the threshold for statistical significance. We agree with the push away from p-values. Based on recommendations from the reviewers we decided to only include confidence intervals. We deleted all reference to p-values and only use confidence intervals throughout.

“In terms of language, I wonder if it’s right to describe witnessing violence etc. as trauma, or traumatic events, as trauma would be the psychological outcomes, e.g. PTSD.”

We agree with this feedback and revised the manuscript to include only use of “adverse events” or “adversity” to replace use of Trauma.

“In the discussion section, there is no previous literature cited – is this truly the case? The only time literature comes in is around implications. I would have imagined there would be a little bit of previous work on this etc.”

We agree with this feedback and significantly restructured the discussion section to include more citations and we contextualize our research findings in existing research. To our knowledge there are no previously published studies that examine the relationship between intimate partner violence and overdose. There is however additional research on the relationship between exposure to childhood adverse events and overdose. In addition to childhood adversity, a robust body of literature exists connecting IPV to substance use severity. As suggested we endeavored to incorporate this into the discussion section.

In addition to changes throughout the discussion section we included the following:

“Findings from this study are consistent with prior literature suggesting that exposures to childhood adversities are associated with increased risk of overdose and substance use severity among people who use illicit drugs (Lake et al., 2015; Sacks et al., 2008). Lake et al., (2015) found that exposures to childhood trauma was associated with increased risk of non-fatal overdose among people who inject drugs. To our knowledge ,this is the first study to examine the association between IPV, relationship dependencies and overdose among women who use illicit drugs. However, findings from this study are congruent with previous literature that suggests exposure to IPV is significantly associated with substance use severity among women who use drugs ; Salomon et al., 2002; Testa et al., 2003). Additional research is needed that further elucidates the relationships between IPV and overdose among women who use drugs.”

We also included several additional citations to address this suggestion.

Minor issues

“In the childhood trauma measure section, the first question about verbal trauma, was this before the age 17 or ever? If ever, I wonder if this needs dropping, as it’s quite different to the other questions – or did it just get missed out in the typing.”

The childhood trauma measure (now referenced as adversity) asked about verbal trauma before the age of 17. The authors added “before the age of 18” to accurately reflect how the question was worded

“In the description of the statistical analyses, they say they are going to give standard error estimates – surely inter-quartile range looking at the tables?”

Yes, we provided IQR for the descriptive tables. The manuscript now reflects the provision of this descriptive statistic (IQR)

“In descriptive stats section, “2.00 for lifetime traumas (0-6)” just a few more words needed to make this interpretable.”

The following was included to make this interpretable:

“The mean score for relationship dependencies was 2.00 experiences (0-6).”

“In the discussion the authors state:” A greater number of women reported experiencing hospitalization from drug use suggesting that hospitalization from drug use may provide a more accurate marker of self-reported non-fatal overdose than asking about overdose.” I think this needs unpacking further, as it’s not clearly established by the analyses presented here.”

The authors decided to omit this sentence from the manuscript

“I could not see clear evidence that the paper included access to the data set as per PLoS guidelines.”

We deposited this dataset in academic commons at Columbia University and will provide the DOI upon acceptance of this manuscript.

“Other than this I think the paper is very well written.”

Thank you

Reviewer #2:

“The use of lifetime outcomes (rather than recent events) tempered my enthusiasm for this exploratory analysis as was difficult to rule out reverse causality.

We agree that lifetime outcomes and cross-sectional data temper the strength of the findings in this manuscript. We believe that our exploratory analysis presents a compelling case for future research. Nonetheless, we emphasize this limitation in the revised manuscript. We included the following in the manuscript:

“The cross-sectional design of the study and measurement of lifetime overdose tempers the strength of the findings in this manuscript. Overdose due to substance use may increase risk of experiencing intimate partner violence due to drug and financial relationship dependencies. However, our findings are consistent with prior longitudinal research that found exposure to partner violence at baseline increased engagement substance use (Salomon et al., 2002).

“Key confounders were missing in the models (e.g., age, types of drugs used, injection drug use, incarceration history). Given that the eligibility criteria for drug use was broad, the analysis likely includes a heterogeneous sample of drug-using women with varying levels of overdose-related risk, which was a concern.”

We agree with the importance of confounders that were not included in the regression models. The sample included a heterogeneous sample of women who use drugs with varying levels of overdose-related risk. We were limited by the small sample size and thus were conservative with the number of covariates included in the models. We have noted this limitation in the discussion section. For several of the variables the responses were too low to include in the models including types of drugs used and injection drug use. We included this information descriptively for the readers’ benefit. To address these concerns we conducted a sensitivity analysis by including age in the multivariable model which was insignificantly associated with all of the Overdose outcomes and did not significantly change the results of the associations between IPV, childhood adversity and overdose outcomes. We included the following in the limitation section:

The following was added to the methods section:

“We performed sensitivity analysis with other variables with age, binge drinking and several types substances. None of these variables were significant and due to the small sample size were not included in the final regression model estimating where we examine the association between IPV and overdose.”

The following was added to the discussion section:

“The small sample size used in this study precluded inclusion of other confounders that may potentially explain the association between IPV and overdose. The small number of overdoses that were reported by this sample also restricted the number of variables that could be included. This study did not include types of substances and injection drug use, in the multivariable regressions because too few participants overdosed across the types of substances. Nonetheless, these are important variables to consider in future analyses with larger sample sizes to take into account the role of substance use and other factors into the analyses of factors that may drive overdose among women who use illicit drugs. Additionally, it is possible that underreporting occurred given that the research interviews occurred at probation and may have resulted in participants not disclosing overdose and drug use history.”

“The introduction was unusually long and did not cohesively frame the research hypotheses. For example, the exploration of hospitalization for drug use was not justified in the risk environment conceptual framework used. In fact, the authors cited multiple papers conceptualizing the risk environment – given the evolution of this framework in the fields of HIV and opioid research, better articulation will be needed to guide the reader.”

We shortened the introduction. We also restructured the introduction throughout to better frame the research hypotheses. We agree that hospitalization as a dimension of the risk environment needed more development. To address these concerns, we added the following:

“Prior research has shown that following hospitalization for drug use, people who use opioids are at greater risk of returning to drug use and overdosing due to lower tolerance and experiencing symptoms of withdrawal (Papp & Schrock, 2017). Emergency department hospitalization is an important yet understudied feature of the overdose risk environment for women who use opioids. Hospitals may be an opportune setting to deliver both overdose prevention through naloxone distribution and IPV services for women who are hospitalized due to drug use (Papp & Schrock, 2017). No studies to date have examined the associations among women between IPV and the specific outcomes of overdose, hospitalization from drug use, witnessing overdose, and having friends or family who have overdosed.”

“Conversely, the discussion was surprisingly underdeveloped and not grounded in the aforementioned literature.”

We agree with this feedback and significantly developed the discussion section. We added more citations from the aforementioned literature as well as additional text contextualizing how our research findings fit with extant literature on IPV and substance use among women. Specifically ,we provide citations and note where our findings are consistent with prior literature. The conclusion is now grounded in the aforementioned literature.

Minor comment:

There were several typographical errors.

The authors extensively edited this manuscript for typographical errors.

Submitted filename: Plos1ReviewerCommentsnepmFINALRR.docx

Click here for additional data file.

14 Nov 2019

Drug overdose among women in intimate relationships: The role of partner violence, adversity and relationship dependencies

PONE-D-19-22408R1

Dear Dr. Marotta,

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

With kind regards,

Giuseppe Carrà, PhD

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

3 Dec 2019

PONE-D-19-22408R1

Drug overdose among women in intimate relationships: The role of partner violence, adversity and relationship dependencies

Dear Dr. Marotta:

I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

For any other questions or concerns, please email plosone@plos.org.

Thank you for submitting your work to PLOS ONE.

With kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Giuseppe Carrà

Academic Editor

PLOS ONE