If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Violence in interpersonal relationships is a substantial health and social problem in the United States and is associated with a myriad of immediate and long-term physical, behavioral, and neurocognitive impairments. The present study sought to determine the incidence of U.S. emergency department (ED)-attended intimate partner violence (IPV) from 2002 to 2015 and examine the differences in payment sources before and after implementation of the Affordable Care Act.
We analyzed ED visits among female patients aged 15 years or older between 2002 and 2015 from the National Hospital Ambulatory Medical Care Survey. Using International Classification of Disease, Ninth Revision, Clinical Modification, codes from patient visit records, we classified each ED visit to determine the frequency and estimate the relative proportion and national frequency of IPV visits. We explored bivariate and multivariate associations between IPV-related injuries with age, race, ethnicity, method of payment, and region, noting changes over time.
Between 2002 and 2015, female patients visited EDs an estimated 2,576,417 times for IPV-related events, and the proportion of ED visits for IPV increased during that time period. The percentage of ED visits for IPV-related events did not differ significantly by region, race, or ethnicity. Compared with women 25–44 years of age, women aged 65 to 74 (odds ratio, 0.15; 95% confidence interval, 0.05–0.43; p < .001) and 75 years and older (odds ratio, 0.20; 95% confidence interval, 0.08–0.53; p = .001) were less likely to visit an ED for IPV. Women were more likely to pay for IPV-related services out-of-pocket (i.e., self-pay) (odds ratio, 1.85; 95% confidence interval, 1.24–277; p = .003) before the enactment of the Affordable Care Act.
The increase in the percentage of IPV-related ED claims paid by private insurance suggests that the Affordable Care Act may have increased women's willingness and ability to seek medical attention for IPV-related injuries and disclose IPV as the source of injuries.
Intimate partner violence (IPV) is one of the most common forms of violence against women, with 4.0% of women residing in the United States reporting being physically assaulted by a current or former sexual partner in the past 12 months (
collated data from multiple sources (i.e., the 1995–1996 National Violence Against Women Survey, the 1996 Medical Expenditure Panel Survey, and the Medicare 5% Sample Beneficiary Standard Analytic Files) and reported that 28.6% of physical and 32.0% of mental health costs were paid out of pocket by IPV survivors. Similarly, using national survey data (i.e., the 1995 National Violence Against Women Survey and the 1995 Medical Expenditure Panel Survey)
estimated that 48.3% of physical assault costs were borne by the individual's private insurance and/or group plan, 20.1% by public insurance plans (i.e., Medicare and Medicaid), 30.4% by the IPV victim, and 1.2% by other sources.
Although there are a number of reasons why female survivors of IPV pay for services using their own funds rather than using private insurance, recent research has indicated that women often pay out of pocket out of fear that their abuser will find out they have sought medical attention and/or that their insurance provider would use their IPV history to deny, revoke, or increase their health insurance premiums (
). Indeed, before 2010, U.S. insurance companies could classify IPV as a preexisting condition. In practical terms, the insurance company would calculate potential treatment costs for a current/former IPV victim (e.g., mental health treatment, surgery, postinjury treatment, and medications), and require the individual to pay a higher premium (or deny them coverage) if the costs associated with treatment were found to be substantial (
Recent changes in U.S. health policy have improved health care access and coverage for survivors of IPV. Specifically, the Affordable Care Act (ACA) sought to expand coverage to individuals and improve the quality and efficiency of preventive services (
To examine how the ACA might have influenced the use of emergency care for IPV-related injuries, the aim of the current study was to examine ED visits by female patients aged 15 years or older using the National Hospital Ambulatory Medical Care Survey (NHAMCS) data, a nationally representative sample of all ED visits in the United States between 2002 and 2015. We estimate national frequencies and document the secular trends in IPV-related ED visits. We explore time trends and potential differences in the proportion of ED visits for IPV-related injuries by race, ethnicity, payment method, and U.S. region. Finally, we examine the payment sources for IPV-related injury before and after implementation of the ACA.
We conducted a secondary analysis of the NHAMCS ED database for 2002 to 2015. The NHAMCS is a four-stage probability sample collected by the National Center for Health Statistics and the Centers for Disease Control and Prevention of visits to the emergency and outpatient departments during a randomly assigned 4-week data period (
). In the NHAMCS, EDs and outpatients are randomly selected from primary sampling units (i.e., county or other geographical units) from the 50 states and the District of Columbia. The NHAMCS stratifies each primary sampling unit by socioeconomic and demographic variables, and then selects primary sampling units with a probability proportional to their size. NHAMCS collects approximately 234 variables each year, including patient demographics (e.g., race, ethnicity, and age), hospital demographics (e.g., geographical region, metropolitan area, and ownership), administrative visit information (e.g., length of visits, length of stay, waiting time, time of arrival, and payment type), and medical variables (e.g., physician diagnoses, cause of injury, vital signs, and procedures). This study used and merged hospital ED data from 2002 to 2015. We excluded data from males, girls younger than 15 years of age, and incidents with poison-related causes from all analyses. Furthermore, data from 2004 were excluded because cases of IPV were substantially different, suggesting potential coding error or coding changes relative to other years. More information on procedure, coding, and data reporting is available at www.cdc.gov/nchs.
Case Definition of IPV
Although this study focuses on violence against women inflicted by an intimate partner, we included causes directly related to IPV and incidents in which women were abused or assaulted by another adult. IPV cases were identified using the International Classification of Disease, Ninth Revision, Clinical Modification (ICD-9-CM) codes found in the cause of injury (rape [E960.1], spouse abuse [E967.3]) and diagnosis (adult abuse [995.80–995.83, 995.85], history of violence [V15.41, V15.42], and counseling [V61.10, V61.11]) fields. These IPV ICD-9-CM codes were selected based on prior research (
We classified each ED visit to identify one outcome: an IPV event. For the outcome of interest (i.e., IPV), we calculated the frequency (number) of diagnoses and/or causes by year, as well as the proportion of all ED visits during which the outcome of interest was diagnosed by year. Furthermore, we used weights included in the data to calculate national estimates of IPV-ED visit frequencies by year. Secular trends were assessed with the weighted χ2 test for trend. We then examined bivariate relationships between patient demographic and visit characteristics (i.e., age, race, ethnicity, geographic region, and method of payment) and visit rates. Pearson's χ2 was used to evaluate the significance of comparison.
Subsequently, multivariable logistic regression was used to assess differences in IPV-related ED visits by patient demographics and payment method before and after the ACA went into effect. Bivariate group difference analysis included the following payment methods: private, Medicare, Medicaid, workers’ compensation, self-pay, charity/no charge, other, unknown, and blank. Self-pay and charity/no charge were combined to form a single group for the interacted model presented. A p value of less than .05 was considered to be statistically significant.
Data management and analysis were performed using the survey [svy] commands in Stata software (version 13.1, Stata Corporation, College Station, TX). Institutional review board exemption was granted by the institutional review board at San Francisco State University because data were de-identified and publicly available.
In a total sample of 188,448 ED visits, there were 652 cases of ED visits for IPV (0.35%) in female ED visits from 2002 to 2015. After appropriate weighting, this represents 2,576,417 national ED visits for IPV-related injuries from a total of 749,418,720 national ED visits. Figure 1 presents the annual rates of IPV-related ED visits. Between 2002 and 2015, the annual rate of IPV-related ED visits ranged from 24 to 55 visits per 10,000 ED visits, with an average of 34 IPV-related ED visits per 10,000 ED visits (0.34% [Supplemental Table 1]). Using linear regression, we estimate that the annual rate of increase for IPV-related ED visits was 4.27%.
As seen in Table 1 column 1, there were significant differences by age, χ2 (4, N = 188,443) = 253.37, p < .0001; payment method, χ2 (8, N = 188,443) = 282.59, p < .0001; and timing, χ2 (12, N = 188,443) = 96.10, p = .0365; in IPV-related ED visits in univariate analyses. In Table 1 column 2, we present multivariate logistic regression models including indicator variables for year, race, ethnicity, age, geographical region, and payment method. Based on a sample of 150,396 ED visits, women were more likely to visit an ED for an IPV-related injury in 2010 (odds ratio [OR], 2.04; 95% confidence interval [CI], 1.19–3.47; p = .009), 2013 (OR, 2.37; 95% CI, 1.14–4.02; p = .021), and 2014 (OR, 2.63; 95% CI, 1.33–5.23; p = .006), compared with 2002. In addition, we found that women were more likely to self-pay for IPV-related care (OR, 1.46; 95% CI, 1.02–2.08; p = .039) than to use insurance. In comparison with women aged 25–44 years, women aged 65 to 74 (OR, 0.15; 95% CI, 0.05–0.43; p < .001) and 75 years and older (OR, 0.20; 95% CI, 0.08–0.53; p = .001) were less likely to visit an ED owing to IPV. We did not find significant differences by race, ethnicity, or census region in the proportion of ED visits for IPV-related injuries.
Table 1Demographic Characteristics, Percent of ED Visits for IPV (N = 2,576,417), and OR (95% CI) for IPV-related ED Visits (N = 150,396), United States, 2002–2015
Ethnicity percentage denote the proportion of women who reported their ethnicity; observations of those who did not report their ethnicity were excluded (N = 2,094,798). Column 1 presents estimated number and percentage of national IPV ED visits and Column 2 presents results from logistic regression based on the sample of IPV ED visits after controlling for year, race, ethnicity, age, region, and payment method.
This table uses 2002–2015 National Hospital Ambulatory Medical Care Survey data. Observations from 2004 were excluded owing to coding changes and/or potential survey administrator coding error.
∗ p < .01.
† p < .05.
‡ Ethnicity percentage denote the proportion of women who reported their ethnicity; observations of those who did not report their ethnicity were excluded (N = 2,094,798). Column 1 presents estimated number and percentage of national IPV ED visits and Column 2 presents results from logistic regression based on the sample of IPV ED visits after controlling for year, race, ethnicity, age, region, and payment method.
We then examined differences in payment methods for IPV-related ED visits before (i.e., 2002–2009, N = 93,336) and after the ACA (i.e., 2010–2015, N = 55,527). In Table 2, we present stratified multivariate analyses comparing payment methods before and after the enactment of the ACA. Women were more likely to use self-pay/charity for IPV-related ED visits relative to other ED visits before ACA enactment. Self-pay/charity for IPV-related ED visits was almost two times higher compared with private insurance before ACA enactment (OR, 1.85; 95% CI, 1.24–277; p = .003). In contrast, there were no statistically significant difference in payment method type post-ACA implementation.
Table 2OR (95% CI) for IPV-related U.S. ED Visits before and after ACA Enactment
This table uses 2002–2015 National Hospital Ambulatory Medical Care Survey. Observations from 2004 were excluded owing to coding changes and/or potential survey administrator coding error. Workers' compensation, unknown, and blank observations were excluded from payment method category (N = 244,892). Colum 1 presents results from 2002 to 2009 and column 2 presents results from 2010 to 2015 logistic regression based on the sample of intimate partner violence ED visits after controlling for year, race, ethnicity, age, region, and payment method.
In this study, we examined ED visits by female patients aged 15 years or older using NHAMCS data between 2002 and 2015, exploring time trends and potential differences in the proportion of ED visits for IPV-related injuries by race, ethnicity, payment method, and U.S. census region. We observed the following main findings. First, the analysis of secular trends indicated that the number of IPV-related ED visits increased between 2002 and 2015. Second, there were significant differences in ED visits for IPV by age, but not race or ethnicity. Third, before the ACA, IPV-related ED visits were more likely to be paid for by the survivors relative to non–IPV-related ED visits. In contrast, after the ACA took effect, there was no systematic difference in the method that women used to pay for IPV-related ED visits compared with non–IPV-related ED visits. This discussion focuses primarily on these new findings.
The data from the present study indicate that from 2002 to 2015, 34 of every 10,000 ED visits were by women seeking care for IPV-related injuries. Although differences in methodology make comparisons of prevalence estimates difficult, the number of ED visits obtained in the present study is slightly higher than
, who derived a national estimate of 21 per 10,000 visits between 1997 and 2001 using the NHAMCS dataset, but lower than that reported by prior studies that used population-based random telephone surveys (
An analysis of secular trends indicated that there was a 4.3% increase in the proportion of IPV-related injuries across the 13-year period (Figure 1). Although it is possible that the rates of severe IPV (e.g., broken bones, internal injuries, loss of consciousness) have been increasing since the early 2000s, it is more plausible that these findings are due to the changing social and cultural norms regarding gender-based violence (
). This finding is incongruent with prior studies using survey-based methodologies (i.e., National Crime Victimization Survey) that have reported a decrease in the prevalence of IPV between 1993 and 2010 (
), it is possible that the current findings are a result of financial instability during the Great Recession (starting in December 2007) and the subsequent slow recovery that continued to affect household resources well into 2014. For example,
reported that women who experienced economic hardship in the Great Recession were twice as likely to experience violent or controlling behavior by their male intimate partners, with an increase in violent and controlling behavior toward women in situations where both adults were unemployed.
An analysis of NHAMCS data revealed significant differences for age, but not race or ethnicity. Consistent with prior studies (
), we found that women 65 years of age and over were less likely to visit an ED for IPV, compared with women ages 25–44 years of age. This finding may be due to several factors related to aging, such as age-related decreases in violent behavior and/or the death of the abuser (
). Given that our methods were comparing rates of IPV-related ED visits with non–IPV-related visits, it could be that Black and Hispanic women have higher rates of ED visits overall compared with non-Hispanic White women, which would mask differences in IPV incidents. Although the risk of IPV victimization may be similar across racial and ethnic groups, it is more likely that the nonsignificant results obtained in the present study are due to historical medical mistrust, perceived discrimination, and structural inequities experienced by minoritized groups (
We also found that IPV-related ED visits were higher in 2010, 2013, and 2014. It is possible that some of the increases in the number of visits in 2013 and 2014 were related to changes in reimbursement and coverage policies owing to the enactment of key provisions of the ACA that affected IPV survivors in particular (
Women’s coverage, utilization, affordability, and health after the ACA: A review of the literature: A literature review of evidence relating to the Affordable Care Act’s impact on women’s health care and health.
). For example, the preventive health coverage funds increased in 2013, which by extension increased federal funding and resources for IPV services (i.e., mandatory IPV screenings, supplemental physician training on IPV screenings, and administrator training on IPV reporting [
) and Medicaid coverage was expanded in several states for previously ineligible adults. Factors outside the scope of this study are likely related to the higher IPV-related ED visits in 2010.
Interestingly, multivariate analysis using stratified models, separating the pre-ACA and post-ACA period (as shown in Table 2), and models that interact the post-ACA period with payment type (Supplemental Table 2) show similar results and indicate that women were more likely to self-pay for IPV-related services before the enactment of the ACA. One hypothesis for this finding is that insurance companies could deny, cancel, exclude, and increase health insurance premiums for preexisting conditions such as IPV before the ACA (
). Furthermore, many women gained insurance coverage under the expansion of Medicaid in certain states in 2014. Obtaining insurance could have made women more likely to seek care for physical injuries.
Implications for Practice and/or Policy
The results of the present study have implications for future research and policy. First, the increasing incidence of ED visits for IPV between 2002 and 2015 observed in the present study highlights that IPV is a significant health problem in the United States. However, it is not clear what changes have caused the increase. Future research should explore whether these changes are driven by increases in the severity of IPV incidence that require immediate attention from health care systems and professionals, such as changes in attitudes that lead to an increase in patient disclosures, and/or changes in reimbursement and coding practices. Second, our finding that there was a differential financial burden on IPV survivors for payment of ED-related visits relative to non-ED related visits before, but not after, ACA enactment highlights the importance of the ACA for women generally and IPV survivors in particular. Current presidential proposals and legal challenges to the ACA may lead to an environment where women and IPV survivors will have to disproportionately shoulder the cost of their victimization once again. Any future changes to health insurance policy must consider its effects on IPV survivors.
Study Strengths and Limitations
The present study's main strength is its use of hospital-level data that allowed for the examination of IPV-related injury secular trends between 2002 and 2015. The NHAMCS stratifies each primary sampling unit by socioeconomic and demographic variables, and via the use of a four-stage probability sample, enabled for the extrapolation of IPV-related injury national estimates by applying appropriate weights.
As with all research, this study has some limitations. First, the NHAMCS codes patient visits, rather than individual patients. Women who visited the ED on more than one occasion were counted independently. Furthermore, the NHAMCS is an administrative dataset, and the determination of race and ethnicity is sometimes self-reported by the patient, inferred by staff observation, transferred from another hospital record, or inferred from the patient's surname. We cannot rule out the possibility that some patients' race and ethnicity were misclassified.
Second, we acknowledge that the use of a limited set of ICD-9-CM codes to define IPV may have underestimated the magnitude of IPV-related ED visits, given that IPV is associated with a multitude of health conditions (
). Furthermore, the present study examined cases of IPV that were severe enough to require immediate medical attention. Our data set likely does not include injuries treated in doctors' offices, walk-in clinics, and urgent care (e.g., sprains, strains, and headaches). Because women seek medical attention from a variety of providers (
), future research should examine care utilization across medical settings with varying levels of charity care provision and Medicaid acceptance.
Third, IPV injuries are rare events among all ED visits. Our analysis yielded 652 weighted cases out of 188,448 weighted ED visits. Logistics regression can underestimate the probability of such rare events, which may cause biases in estimated associations. However, our results were similar when we examined linear probability models that are not affected by the rarity of the events. Thus, we believe the results are valid.
This study examined IPV-related injuries that were severe enough to require immediate medical attention and found a 4.27% increase in the proportion of ED visits for IPV-related injuries between 2002 and 2015. Before the implementation of the ACA, women were more likely to pay out of pocket for IPV-related services. Our findings suggest that the implementation of the ACA may have reduced barriers to receiving medical care for IPV-related injuries.
Women’s coverage, utilization, affordability, and health after the ACA: A review of the literature: A literature review of evidence relating to the Affordable Care Act’s impact on women’s health care and health.
Tatiana L. Mariscal, MS, is a Lecturer who uses her academic background in movement science and public health to explore the health determinants associated with intimate partner violence among the underserved female population in the U.S and developing countries.
Charmayne M.L. Hughes, PhD, uses her background in motor control neuroscience and engineering to elucidate the impact of violence and trauma on underlying neurocognitive and neuromotor functioning, and to develop technological solutions that benefit resource-constrained and vulnerable populations.
Sepideh Modrek, PhD, is an assistant professor of Health Economics at the Health Equity Institute at San Francisco State University. Her research examines social determinants and structural inequities that impact health outcomes.
Published online: June 28, 2020
Received in revised form:
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.