Intimate Partner Violence Screening Practices in California After Passage of the Affordable Care Act

Abstract

Under the Affordable Care Act (ACA), insurance coverage should include screening for intimate partner violence (IPV). In this article, we present self-reported IPV screening practices and provider confidence from a post-ACA cross-sectional survey of 137 primary care clinicians in California. Only 14% of the providers reported always screening female patients for IPV and about one third seemed never to screen. Female providers were more likely to screen and use recommended direct questioning. Most providers lacked confidence in screening, referral, and record-keeping. Serving a low-income population predicted more frequent screening and better record-keeping. Overall, IPV screening in primary care was inadequate and needs attention.

Keywords

intimate partner violence screening Affordable Care Act primary care California

Introduction

At least one in four U.S. women experience intimate partner violence (IPV) in their lifetime (National Coalition Against Domestic Violence, 2011). IPV occurs among women of all ethnicities, socio-economic levels, and ages. Estimates of prevalence may vary due to the range of experiences classified as IPV, such as physical violence, emotional abuse, sexual assault, social isolation, and monetary control (Ellis, 1999). The National Coalition Against Domestic Violence estimates that every year, about 1.3 million women are victims of physical assault by an intimate partner, 85% of victims are women, and more than 550,000 require medical attention (National Center for Injury Prevention and Control, 2003). Victims often need intensive physical, emotional, and financial services to recover from their abuse.

Early and careful screening could reduce suffering, disability, and death if appropriate referrals are made. In the United States, the federal government funds a variety of programs to assist IPV victims, including legal aid, counseling, job training, and housing (Rorie, Backes, & Chahal, 2014). Women who take advantage of shelter services are more likely to leave an abusive situation, have shorter overall trajectories of abuse in their lifetime (Panchanadeswaran & McCloskey, 2007), and are less likely to return to their abuser (Hilbert, Kolia, & VanLeeuwen, 1997). Using counseling services increases the self-esteem, assertiveness, and self-efficacy of IPV victims (Mancoske, Standifer, & Cauley, 1994). However, if not addressed, IPV victimization can result in immediate and lifelong negative outcomes for an individual, often leading to hospitalization, disability or death, as well as depression, anxiety, suicidal ideation, substance abuse, and pregnancy complications (National Intimate Partner and Sexual Violence Survey, 2014). Given that violence can be contagious, screening for IPV and referring victims to available services can prevent violence from spreading to families and communities (Slutkin, 2013). Many health care providers, especially in primary care and obstetrics/gynecology, have regular contact with female patients and are in a good position to screen women privately for IPV (Allard, 2013; Ellis, 1999; Tower, 2006; Williamson et al., 2004). Although no consensus exists about which screening procedures work best (Montalvo-Liendo, Wardell, Engebretson, & Reininger, 2009; Saiki & Lobo, 2011; Svavarsdottir, Orlygsdottir, & Gudmundsdottir, 2014), many screening tools are available (Allard, 2013; Chen et al., 2007; Director & Linden, 2004; Family Violence Prevention Fund, 2004; Harwell et al., 1998). From the patient’s perspective, most women—both victims and non-victims—report high levels of comfort in being screened (Allard, 2013; Director & Linden, 2004; Ramsay, Richardson, Carter, Davidson, & Feder, 2002).

Routine screening could identify IPV among patients who are unlikely to disclose IPV unless asked (Beynon, Gutmanis, Tutty, Wathen, & MacMillan, 2012; Hathaway et al., 2000). Asking a specific question in a supportive and non-judgmental environment may identify up to 65% of patients who have been abused (DeBoer, Kothari, Kothari, Koestner, & Rohs, 2013; Feldhaus et al., 1997). One study found that adding a single direct verbal screening question (such as, “During an argument at home, have you ever worried about your safety?”) increased overall IPV detection rates from 0.4% to 7.6% (Morrison, Allan, & Grunfeld, 2000). Another found that female IPV victims prefer screening questions that are direct (e.g., “Do you feel safe at home with your partner?”) to those that are indirect (e.g., “How is everything at home?”) (McNutt, Carlson, Gagen, & Winterbauer, 1999). In recent years, three major organizations have recommended that screening start with a framing statement followed by a direct question (see Table 1).

Table 1.

Recommended Screening Questions on Intimate Partner Violence for Clinicians.

Organization	Recommendations
Family Violence Prevention Fund (2004)	Framing Statement: ● “Because violence is so common in many people’s lives, I’ve begun to ask all my patients about it.” ● “I am concerned that your symptoms may have been caused by someone hurting you.” Sample Direct Verbal Questions: ● “Are you in a relationship with a person who physically hurts or threatens you?” ● “Did someone cause these injuries? Was it your partner/husband?” ● “Is it safe for you to go home?” ● “Has your partner ever forced you to have sex when you didn’t want to?”
American Congress of Obstetricians and Gynecologists (2012)	Framing Statement: ● “We’ve started talking to all of our patients about safe and healthy relationships because it can have such a large impact on your health.” Ensuring Confidentiality: ● “Before we get started, I want you to know that everything here is confidential, meaning that I won’t talk to anyone else about what is said unless you tell me that . . . (insert law by state)” Sample Direct Verbal Questions: ● “Has your current partner ever threatened you or made you feel afraid?” ● “Has your partner ever hit, choked, or physically hurt you?” ● “Has your partner every forced you to do something sexually that you did not want to do, or refused your request to use condoms?”
Center for Relationship Abuse Awareness (2013)	Framing Statement: ● “Because there is help available for individuals who are being abused, I now ask all my patients about domestic violence.” ● “I do not know if this is a problem for you, but because so many patients I see are dealing with abusive relationships, I’ve started asking about it routinely.” Sample Direct Verbal Questions: ● “Do you ever feel afraid of your partner?” ● “Has your partner ever destroyed things that you cared about?” ● “Has your partner ever threatened or abused your children?” ● “Do you have guns in your home? Has your partner ever threatened to use them when he was angry?” ● “Do you feel controlled or isolated by your partner?” Sample Indirect Questions (if patient has denied abuse, but is a possibility) ● “Have you been under any stress lately?” ● “Are you ever afraid?” ● “Have you ever gotten hurt?”

IPV was officially designated as a public health issue in 1986. Ever since, there has been controversy regarding screening, coverage, and referral (Dagher, Garza, & Kozhimannil, 2014). Universal IPV screening of women in health care settings has been promoted by the American Medical Association, the Joint Commission, the American Congress of Obstetricians and Gynecologists, and other professional associations (Decker et al., 2012). However, some oppose universal screening because evidence is insufficient on whether it leads to improved health outcomes and less abuse of women (Nelson & Johnston, 2004; Ramsay et al., 2002).

Starting in August 2012, the Affordable Care Act (ACA) required insurers to include IPV screening and counseling to all women at no additional cost (Singh, Petersen, & Singh, 2014). The ACA also required Medicare, Medicaid, and new insurance plans to follow the U.S. Preventative Services Task Force (USPSTF) recommendations that all women aged 14-46 be screened for IPV (Singh et al., 2014). Although the USPSTF gave age recommendations, it did not discuss how best to query patients and make referrals (Dagher et al., 2014). Without specific federal or state regulations, individual health care organizations and hospitals often develop their own screening policies and procedures (Chapin, Coleman, & Varner, 2011).

In 1993, California passed a law to mandate IPV screening in health care settings. California’s Health & Safety Code (Sections 1233.5 and 1259.5) specifies that (1) doctors, nurses, and mental health practitioners who apply for California professional licenses must have documented training in how to detect and treat IPV, and (2) licensed clinics and hospitals must have written policies and procedures to (a) routinely screen for IPV, (b) document injuries and illnesses related to IPV, (c) refer IPV victims to crisis intervention services, (d) designate staff to implement policies, and (e) have educated staff on identifying and caring for IPV victims (California Partnership to End Domestic Violence, 2011). However, no enforcement mechanism exists, and providers may not follow written procedures even if they do exist.

Little is known about how often providers screen for IPV, the patient’s age at which they start screening, and the questions they use. Surveys suggest that screening is infrequent. A 1995 survey of California physicians found that only 10% reported routine IPV screening of new patients (Rodriguez et al., 1999). A decade later, in a survey of social workers, family practitioners, and obstetrician/gynecologists in Florida, Tower et al. found that one in five reported that they “nearly always” or “always” routinely screened for IPV—but this was skewed by significantly higher reports by the social workers (Tower, 2006). Meanwhile, a clinic-based survey found that only 7% of female patients reported ever having been screened for IPV in a health care setting (Klap, Tang, Wells, Starks, & Rodriguez, 2007). No study has asked providers to record the screening question they used, to identify referral entities, or to describe how they recorded IPV in medical records.

Given these gaps, this study’s main purpose was to delineate current screening practices of primary health care providers in Southern California after the passage of the ACA. To reduce courtesy bias associated with self-reporting, providers were given a semi-structured questionnaire that asked them to specify how they queried, referred, and recorded patients experiencing IPV. The study’s other purpose was to assess whether provider characteristics predicted reported screening practices. Our hypothesis was that non-physicians would be more likely to screen and to use direct screening questions about IPV, in part because physicians generally have more time constraints and would worry that screening could be time-consuming. We further hypothesized that providers who were confident in their ability to assist victims of IPV would report more frequent screening. The results of this study could assist policy-makers and facility managers to improve the quality and quantity of IPV screening and referrals.

Method

This study was drawn from a cross-sectional survey of health care providers conducted between October 2013 and March 2014 in Southern California. To be eligible for the survey, respondents needed to be clinicians (physicians, nurse practitioners [NPs], physician assistants [PAs], or nurses) who were currently seeing female patients aged 15 or above.

Participants were recruited using convenience and snowball sampling. We first sent emails to the directors of three professional networks of health care providers: University of California, Los Angeles’s (UCLA) Women’s Health Center, UCLA’s Center for Health Improvement of Minority Elderly, and the Westside Domestic Violence Network. We asked the directors to forward the survey link via email to their providers, so they could take the survey online. We also recruited health care providers to take the survey at the beginning of a 1-hour training session in IPV, which we conducted at the headquarters of St. John’s Well Child and Family Centers in Los Angeles. Last, we attended 2 days of the 2014 Annual PriMed Conference West held in Anaheim, California. We recruited most of our providers at this conference, who self-administered the survey on the spot. None was offered any compensation for participating in the study, except for the chance to participate in a raffle for a gift card of nominal value. Altogether, 180 participants self-administered a paper copy of the questionnaire, while 11 completed it online. Of the 191 providers who took the survey, 141 reported working in primary care settings, which was the focus of this analysis. Four questionnaires were only partially completed. Hence, 137 providers were included in the study.

The questionnaire, designed to take about 8-10 minutes to complete, included questions on providers’ demographics, patient type, screening frequency and practices, and their confidence about being able to identify and assist IPV victims. For IPV screening frequency, providers were given the response choices of “always,” “most of the time,” “some of the time,” “rarely,” “only if showing symptoms,” or “other.” Regarding confidence, providers were asked to record whether they felt “very confident,” “confident,” “somewhat confident,” or “not at all confident” on four items: recognizing the signs and symptoms of IPV, knowing services available for IPV victims, communicating effectively with IPV victims, and overall ability to assist IPV victims. Reliability analysis of these four items yielded a Cronbach’s alpha of .874.

To gauge actual practices and knowledge of referral sites, providers were given open-ended questions on (a) what was the first question they normally asked, if any, to screen for IPV, (b) where they would refer IPV victims, and (c) whether they coded for IPV and the specific code they used in medical records. This approach allows some confirmation of the reported frequency of screening, because if a provider could not recall any screening question or provide any referral site, it was unlikely that he or she was doing regular screening. Using a grounded theory approach (Glaser & Strauss, 1967), we developed criteria for coding and jointly agreed on how to categorize the responses. We coded the screening questions given by respondents as “direct,” “indirect,” or “other.” If a question resembled any of the sample direct questions recommended by professional organizations (in Table 1), it was coded “direct.” Typical questions respondents recorded were whether the patient felt safe at home or was being hurt by someone. If a question resembled an indirect question, recommended generally if a patient denied abuse but suspicion lingered, it was coded “indirect.” None of the organizations in Table 1 recommended beginning with an indirect question. Typical questions clinicians provided were asking how things were at home or whether there were problems. Questions focusing on a narrow aspect of abuse or that were overly vague were coded “other.” These included asking whether the patient’s life was under threat or whether she had been touched inappropriately.

Quantitative data were entered into SPSS 22 and analyzed. We did bivariate analysis of screening practices and confidence by two variables: provider occupation and gender. We then performed four logistic regression equations to determine predictors of screening frequency, question used, knowledge of referral site, and specific documentation (Hosmer, Lemeshow, & Sturdivant, 2013).

This study was conducted under the auspices of the UCLA Institutional Review Board (IRB) #13-001626.

Results

Of the 137 providers included in the analysis, about 43% were physicians and the rest were NPs, PAs, or nurses (see Table 2). Two thirds of the respondents were women. Nearly 63% had been in practice for 10 years or more. There were significant differences in occupation and years in practice by gender. About 28% of women were physicians and 60% had worked for 10 years or more, whereas 74% of men were physicians and 72% had worked 10 years or more (not shown). More than 90% of all respondents resided in California, about 75% worked solely in primary care, and 59% reported they served mostly low-income patients.

Table 2.

Demographics of Surveyed Primary Care Providers.

Characteristic	Total (N = 137)
Characteristic	%
Occupation^a
Physician	43.1
NP, PA, or Nurse	56.9
Gender
Male	33.6
Female	66.4
Age in years
30 or below	8.8
31-40	19.7
41-50	21.9
51-60	27.0
61 or more	22.6
Years in practice
5 or fewer	20.4
6-10	16.1
11-20	26.3
21 or more	37.2
Resides in California	91.1
Serves mainly low-income patients	59.1
Works solely in primary care	75.2

Note. ^aOf these, Physician = 59; Nurse Practitioner (NP) = 42; Physician Assistant (PA) = 31; both PA and NP = 1; Nurse = 4.

Table 3 gives examples of typical questions written on the questionnaire in response to an open-ended query concerning the first question a provider usually asked (if any) when screening for IPV. It also shows how coders categorized the queries. Of the screening questions respondents recorded, 42% were categorized as direct, 27% as indirect, 13% as too narrow or symptom-based, and 16% were missing (see Table 4). Female providers were more than twice as likely as male providers to use a direct question (53% vs. 22%) with their female patients. Similarly, NPs and PAs were much more likely to use direct questions than were physicians (53% vs. 29%). When gender and occupation were combined, it was found that female physicians were considerably more likely than male physicians to report engaging in direct questioning (44% vs. 18%), as were female NPs/PAs compared with male NPs/PAs (56% vs. 33%; not shown).

Table 3.

Provider Responses to Open-Ended Response About First Question Used for IPV Screening, and How Coded.

Examples of first question asked	Code used by researchers
Do you feel safe at home?	“Direct”
Have you ever been hit, slapped, kicked, or physically hurt by someone?
Are you currently in a relationship with someone who hurts or threatens you?
Have you been intentionally hurt in any way since our last visit?
Is there violence or problems at home?
Do you feel safe?	“Indirect”
How are things at home?
How are you getting along with your partner?
Are there any problems at home?
What are your concerns? Is there anything you want to tell me privately?
Is your life under threat?	“Other” (too narrow, too vague)
Has anyone touched you in a way you did not like?
Do you have any injury to your body?
Any changes at home?
How are you doing today?

Note. IPV = intimate partner violence.

Table 4.

Providers’ Self-Reported IPV Screening Practices by Occupation and Gender.

	Occupation		Gender		Total (n = 137)
	Physician (n = 59)	NP, PA (n = 78)	Male (n = 46)	Female (n = 91)	Total (n = 137)
	%	%	%	%	%
How often screen females for IPV
Always	6.8	19.2	6.5	17.6	13.9
Most of the time	18.6	25.6	13.0	27.5	22.6
Some of the time	32.2	28.2	34.8	27.5	29.9
Rarely	27.1	14.1	34.8	12.1	19.7
Only if showing symptoms	11.9	7.7	10.9	8.8	9.5
Other or don’t screen	3.4	5.1	0.0	6.6	4.4
Age begin screening females for IPV
10-14 years	10.2	17.9	8.7	17.6	14.6
15-17 years	16.9	16.7	15.2	17.6	16.8
18 years and above	37.3	26.9	41.3	26.4	31.4
No specific age or depends	13.6	17.9	15.2	16.5	16.1
Missing	22.0	20.5	19.6	22.0	21.2
Feel very confident or confident in:
Recognizing signs/symptoms of IPV	33.9	35.9	28.3	38.5	35.1
Knowing services available for IPV	22.0	15.4	19.6	17.6	18.2
Communicating with IPV victims	44.1	41.0	39.1	44.0	42.4
Overall ability to assist IPV victims	35.6	29.5	34.8	30.8	32.1
First question use to screen for IPV
Direct question	28.8	52.6	21.7	52.7	42.3
Indirect question	35.6	21.8	41.3	20.9	27.7
Other (too narrow or technical)	10.2	6.4	13.0	5.5	8.0
Symptom-based	8.5	3.8	8.7	4.4	5.8
Missing	16.9	15.4	15.2	16.5	16.1
Know where to refer IPV victim
Named organization or shelter	22.0	24.4	21.7	24.2	23.4
DV Hotline	5.1	10.3	4.3	9.9	8.0
In-house (e.g., social worker)	22.0	14.1	17.4	17.6	17.5
Police or social services	11.9	19.2	23.9	12.1	16.1
Don’t know/Missing	39.0	32.1	32.6	36.3	35.0
Use specific code to record IPV victim
Yes	28.8	38.5	32.6	35.2	34.3
No	64.4	53.8	63.0	56.0	58.4
Missing	6.8	7.7	4.3	8.8	7.3

Note. IPV = intimate partner violence. Bolded = sig. <.01.

Other screening practices surveyed included how often a provider reported screening for IPV among their female patients, the age at which they began to screen, whether they knew where to refer a victim, and whether they used a specific code to document IPV status. Overall, only 14% reported “always” screening for IPV, and another 23% reported “most of the time.” About one third reported rarely or never screening female patients for IPV, despite being primary care providers. Female providers and non-physicians were significantly more likely to report frequent screening, but even among these groups, fewer than half reported frequent screening.

In general, the providers surveyed did not express much confidence in their ability to identify victims of domestic violence or provide assistance to them. Only 35% felt “very confident” or “confident” in recognizing the signs and symptoms of domestic violence. Even fewer (18%) reported that they knew the services available for victims. About 42% were confident that they could communicate effectively with victims. Overall, only 32% felt “very confident” or “confident” that they could assist victims of domestic violence. There was no significant difference in reported confidence by occupation or gender.

When asked what age screening began, about 31% reported before age 18, which is recommended by professional organizations. Age when screening began did not differ by occupation or gender. About half of the providers could name a place, hotline, or social worker to refer a female patient who reported IPV. About half did not know where to refer or listed something vague such as “police” or “social services.” Specific knowledge of referrals did not differ by occupation or gender. Last, it was found that only about one third used a specific code in their documentation to report an IPV victim.

To test our hypotheses about the predictors of various screening practices, we conducted logistic regression analyses. Table 5 shows the results of four logistic regression equations to determine predictors of (a) self-reported screening frequency, (b) assessment question used, (c) intervention (referral location known), and (d) documentation with specific codes. The five predictors we tested for each outcome were gender of the provider, years in clinical practice, occupation (physician vs. non-physician), patient type (mainly low income vs. not), and providers’ expressed confidence (very confident or confident) in their overall ability to help IPV patients. Because we found no correlation between having knowledge of a referral site and screening frequency or other practices, referral knowledge was not used as a predictor. After controlling for the other variables, we identified three significant predictors of screening always or most of the time: being a female provider (adjusted odds ratio [AOR] = 2.8), having mainly low-income patients (AOR = 4.8), and feeling confident in one’s ability to help (AOR = 2.1). Regarding using a direct screening question, there were two significant predictors: being a female provider (AOR = 3.1) and having confidence in one’s ability (AOR = 1.9). Knowing a specific place to refer had only one significant predictor: one’s overall confidence in ability to assist (AOR = 2.0). Finally, whether one used a specific code to document IPV was predicted only by whether most of one’s patients were low-income (AOR = 3.0).

Table 5.

Predictors of Self-Reported IPV Screening, Assessment, Referral, and Documentation.

	Outcomes
	Screening (Always or most of the time)	Assessment (Use direct question)	Referral (Know specific referral place)	Documentation (Use specific code for IPV)
	AOR (CI 95%)	AOR (CI 95%)	AOR (CI 95%)	AOR (CI 95%)
Predictors
Gender: Female provider	2.88 [1.05, 7.99]	3.17 [1.25, 7.79]	1.64 [.71, 3.81]	0.77 [.31, 1.88]
Years: In clinical practice	0.93 [.68, 1,26]	0.78 [.59, 1.05]	1.19 [.90,1.57]	0.85 [.64, 1.13]
Occupation: NP, PA, or Nurse (not MD)	1.77 [.69, 4.38]	1.80 [.78, 4.14]	0.90 [.40, 2.00]	1.52 [.64, 3.60]
Patients: Mainly low-income	4.82 [1.89,12.32]	1.07 [.48, 2.40]	0.94 [.44,2.02]	2.98 [1.26, 7.03]
Confidence: In overall ability to help IPV patients	2.06 [1.19, 3.56]	1.92 [1.16, 3.17]	1.99 [1.25, 3.16]	1.46 [.91, 2.34]

Note. Bolded items are significant predictors of the outcomes. AOR = adjusted odds ratio; CI = confidence interval; NP = nurse practitioner; PA = physician assistant.

Discussion

Despite the ACA and existing California law, as well as a 2013 poll in which 81% of Californians approved of providers’ screening for IPV (Tulchin & O’Neil, 2013), this study found that screening of female patients by primary care providers was inadequate. Not only did it occur infrequently even by self-reports, but screening seemed to target low-income patients. Contrary to our hypothesis, gender was a significant predictor of screening frequency, but not provider occupation type. Female providers (physicians and non-physicians) appeared more likely to screen and use the recommended approach of direct questions, but their knowledge of specific referral agencies was limited, and their use of codes to record IPV was rare. We also found that most providers (both male and female) reported lacking confidence in their ability to screen and assist women who might be experiencing IPV.

Other studies similarly have found that female providers are more likely to screen for IPV than their male counterparts (Elliott, Nerney, Jones, & Friedmann, 2002; Jonassen & Mazor, 2003; Tower, 2006). With increasing numbers of women entering the medical field, more screening may occur over time. It is also promising that female providers in this study were more likely to use direct questions. Curiously, although female providers engaged in more and better screening, they did not feel more confident, nor did they have greater referral knowledge. Perhaps female providers simply feel more at ease in speaking about sensitive issues such as IPV with female patients, and do not fear the emotionality that might ensue after a positive screen. Female providers may also be more willing to take the time to screen because they have greater empathy with other women or have experienced IPV themselves.

Interestingly, in this study, a main predictor of more frequent screening was serving a predominantly low-income population. It is not known whether this was due to providers’ implicit bias that higher socio-economic status (SES) patients are not at risk of IPV, whether providers had more institutional pressure to screen when they served a low-income clientele, or whether they found it awkward to broach the topic with women from their same social strata. Although researchers have found that lower SES is a risk factor for IPV (Bonomi et al., 2009; Center for Disease Control and Prevention, 2013), higher income women still need to be screened.

Serving a low-income population also resulted in three times the odds of using a specific code to document IPV in their patients’ medical record. Increased documentation among low-income populations could contribute to the perception that prevalence is greater within that population. Without standardized documentation, regardless of patient income level, it is nearly impossible to measure and track rates of IPV revealed in the clinical setting.

As we had hypothesized, the study found that providers’ confidence in their overall ability to help IPV victims predicted the frequency of screening and also knowing where to refer. It seems logical that providers who feel more confident are more likely to discuss IPV with their patients. Alternatively, engaging in routine screening and referral may create feelings of self-efficacy and confidence among providers. Lacking confidence in one’s ability to make referrals is a well-documented barrier to screening (Jonassen & Mazor, 2003). Studies suggest that patients often presume that providers possess the requisite knowledge in IPV screening and referral, even when they do not (Morse, Lafleur, Fogarty, Mittal, & Cerulli, 2012). Offering in-house or online training focused on how to screen patients, where to refer, and how to document abuse might build confidence and increase screening frequency.

Although this study found that a provider’s occupation was not a significant predictor of screening practices when gender was controlled, most previous studies have reported that nurses and NPs are more likely to screen for IPV than physicians (Center for Disease Control and Prevention, 1998; Goff, Byrd, Shelton, & Parcel, 2001; Stayton & Duncan, 2005), with the exception of a Canadian study that found physicians to be more likely to initiate the topic of IPV than nurses (Falsetti, 2007). Because nurses and NPs are disproportionately female, it may be that gender underlies previous findings. From the patients’ standpoint, women report no difference in comfort when being screened by a physician, nurse, or NP (Thackeray, Stelzner, Downs, & Miller, 2007).

This study had several limitations. As a convenience sample, it is possible that it did not represent accurately the screening practices and confidence of primary care providers in California. However, the researchers were not aware of any inherent biases in the recruitment of providers. Moreover, the providers’ gender and occupational mix appear to mirror the proportion in California (Jones & Danish, 2011). It should be noted that 9% of the primary care providers in the sample recorded that they were practicing outside of California. However, because statistical tests did not reveal any differences with their Californian counterparts, they were not excluded from analysis.

Another limitation was that the study did not collect data on the race/ethnicity of either the providers or the patients served. It is possible that providers’ screening practices were influenced by the race/ethnicity of their patients as well as their income level. We also do not know whether providers were more or less likely to screen when their patients were of their same race/ethnicity. These issues require further scrutiny.

Last, the self-report nature of the survey is a limitation. Due to social desirability bias, some providers may have exaggerated their frequency of screening, so the amount actually occurring may be less. Although there is no way to verify providers’ reports, the open-ended questions allowed for some confirmation of findings. For instance, we found that 98% of the providers self-reporting high screening frequency wrote an IPV screening question, as compared with only 66% of the providers self-reporting low screening frequency. In combining the various open-ended questions, we estimate that about one third of the respondents probably never screened for IPV.

This study has several policy implications. Given that male providers were significantly less likely to screen frequently and appropriately, more attention may need to be given in medical schools and professional associations to building men’s skills and confidence in broaching sensitive topics such as IPV with female patients. Academic detailing may be a promising strategy to increase male physicians’ comfort in identifying patients experiencing IPV and learning about local referral options (Edwardsen, Horwitz, Pless, le Roux, & Fiscella, 2011).

To promote more IPV screening in general, providers need to believe that the organization they work for favors universal screening. This is more likely to occur if written screening policies exist, standardized procedures have been created for screening, posters are displayed in the workplace, and managers periodically check on providers’ performance (Allen, Lehrner, Mattison, Miles, & Russell, 2007). Also, providers may need specific encouragement to screen for IPV among higher income populations. Management should verbally remind providers of the importance of screening and conduct random patient exit interviews to confirm that screening occurred. Audiotaping encounters could yield more accurate information to managers, but may not be feasible on a regular basis or may not be acceptable to patients. With the rise of electronic medical records and the recognized importance of “big data,” it is increasingly important for IPV to be coded in medical records to enable tracking and improving care.

Finally, better communication needs to occur between health care providers and IPV referral/service agencies. Agencies should advertise their services to all health facilities within their catchment area, and management needs to make more effort to have this information readily available to providers. These approaches could lead to better assistance to patients experiencing or at risk of IPV.

Footnotes

Acknowledgements

We wish to acknowledge Dr. Chandra Ford and Stephanie Albert for their advice and suggestions. Naira Setrakian, Karina Pimentel, and Rachel Erickson assisted with data collection.

Declaration of Conflicting Interests

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

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The Blue Shield Foundation of California provided partial financial support under Grant 6408746. The Foundation had no role in the study design, data analysis, writing, or decision to submit this article for publication.

Author Biographies

Paula Tavrow is an associate adjunct professor in the Department of Community Health Sciences at University of California, Los Angeles’ (UCLA) Fielding School of Public Health. She is also the director of UCLA’s Bixby Program in Population and Reproductive Health, and co-director of the University of California Global Health Institute’s Center of Expertise on Women’s Health and Empowerment. She received a PhD in health services organization and policy and an MSc in population planning from the University of Michigan, Ann Arbor. Her main research interests are in sexual and reproductive health, violence, adolescent health, and quality of health services in low-income countries.

Brittnie E. Bloom received her MS in public health from UCLA’s Fielding School of Public Health in the Community Health Sciences Department in 2014. She also received a BA in psychology and English with a minor in counseling and social change from San Diego State University (SDSU) as a Ronald E. McNair Scholar. She currently works as the program manager for SDSU’s Initiative for Maximizing Student Development (funded by the National Institutes of Health) and as an adjunct faculty member at Grossmont Community College. She also serves as an advocate for survivors of sexual assault with the Center for Community Solutions in San Diego, California.

Mellissa H. Withers is an assistant professor at the University of Southern California Keck School of Medicine in the Institute for Global Health. She also leads the Global Health Program of the Association of Pacific Rim Universities. She received a PhD in community health sciences from UCLA’s Fielding School of Public Health. She also holds a Master’s degree in international health from the Johns Hopkins Bloomberg School of Public Health and a BA in international development from University of California (UC) Berkeley. Her primary research interests lie in community participatory research, maternal and child health, women’s empowerment, violence, and global reproductive health.

References

Allard

(2013). Caring for people who experience domestic abuse. Emergency Nurse, 21, 12-16.

Allen

N. E.

Lehrner

Mattison

Miles

Russell

(2007). Promoting systems change in the health care response to domestic violence. Journal of Community Psychology, 36, 103-120. doi:10.1002/jcop.20137

American Congress of Obstetricians and Gynecologists (2012, February) Committee Opinion, Interpersonal Violence (No. 518). Retrieved from http://www.acog.org/Resources-And-Publications/Committee-Opinions/Committee-on-Health-Care-for-Underserved-Women/Intimate-Partner-Violence

Beynon

C. E.

Gutmanis

I. A.

Tutty

L. M.

Wathen

C. N.

MacMillan

H. L.

(2012). Why physicians and nurses ask (or don’t) about partner violence: A qualitative analysis. BMC Public Health, 12, 473. doi:10.1186/1471-2458-12-473

Bonomi

A. E.

Anderson

M. L.

Reid

R. J.

Rivara

F. P.

Carrell

Thompson

R. S.

(2009). Medical and psychosocial diagnoses in women with a history of intimate partner violence. Archives of Internal Medicine, 169, 1692-1697. doi:10.1001/archinternmed.2009.292

California Partnership to End Domestic Violence. (2011). California Domestic Violence Law Compendium. Berkeley: University of California.

Centers for Disease Control and Prevention. (1998). Rural health-care providers’ attitudes, practices, and training experience regarding intimate partner violence—West Virginia, March 1997. Morbidity and Mortality Weekly Report, 47, 670-673.

Centers for Disease Control and Prevention. (2013, December 24). Intimate Partner Violence: Risk and protective factors. Retrieved from http://www.cdc.gov/violenceprevention/intimatepartnerviolence/riskprotectivefactors.html

Center for Relationship Abuse Awareness (2013). Screening and Interview Strategies for Intimate Partner Violence. Retrieved from http://stoprelationshipabuse.org/pdfs/Screening%20and%20Interviewing%20Strategies%20for%20Intimate%20Partner%20Violence.pdf

10.

Chapin

J. R.

Coleman

Varner

(2011). Yes we can! Improving medical screening for intimate partner violence through self-efficacy. Journal of Injury and Violence Research, 3, 19-23. doi:10.5249/jivr.v3i1.62

11.

Chen

P. H.

Rovi

Washington

Jacobs

Vega

Pan

K. Y.

Johnson

M. S.

(2007). Randomized comparison of 3 methods to screen for domestic violence in family practice. Annals of Family Medicine, 5, 430-435. doi:10.1370/afm.716

12.

Dagher

R. K.

Garza

M. A.

Kozhimannil

K. B.

(2014). Policymaking under uncertainty: Routine screening for intimate partner violence. Violence Against Women, 20, 730-749. doi:10.1177/1077801214540540

13.

DeBoer

M. I.

Kothari

Koestner

A. L.

Rohs

Jr. (2013). What are barriers to nurses screening for intimate partner violence? Journal of Trauma Nursing, 20, 155-160; quiz 161-152. doi:10.1097/JTN.0b013e3182a171b1

14.

Decker

M. R.

Frattaroli

McCaw

Coker

A. L.

Miller

Sharps

. (2012). Transforming the healthcare response to intimate partner violence and taking best practices to scale. Journal of Women’s Health (Larchmt), 21, 1222-1229. doi:10.1089/jwh.2012.4058

15.

Director

T. D.

Linden

J. A.

(2004). Domestic violence: An approach to identification and intervention. Emergency Medicine Clinics of North America, 22, 1117-1132. doi:10.1016/j.emc.2004.05.008

16.

Edwardsen

E. A.

Horwitz

S. H.

Pless

N. A.

le Roux

H. D.

Fiscella

K. A.

(2011). Improving identification and management of partner violence: Examining the process of academic detailing: A qualitative study. BMC Medical Education, 11, 1-9.

17.

Elliott

Nerney

Jones

Friedmann

P. D.

(2002). Barriers to screening for domestic violence. Journal of General Internal Medicine, 17, 112-116.

18.

Ellis

J. M.

(1999). Barriers to effective screening for domestic violence by registered nurses in the emergency department. Critical Care Nursing Quarterly, 22, 27-41.

19.

Falsetti

S. A.

(2007). Screening and responding to family and intimate partner violence in the primary care setting. Primary Care, 34, 641-657. doi:10.1016/j.pop.2007.05.004

20.

Family Violence Prevention Fund. (2004). National Consensus Guidelines: On identifying and responding to domestic violence victimization in health care settings. San Francisco, CA: Author.

21.

Feldhaus

K. M.

Koziol-McLain

Amsbury

H. L.

Norton

I. M.

Lowenstein

S. R.

Abbott

J. T.

(1997). Accuracy of 3 brief screening questions for detecting partner violence in the emergency department. Journal of the American Medical Association, 277, 1357-1361.

22.

Glaser

Strauss

A. L.

(1967). The discovery of grounded theory: Strategies for qualitative research. Chicago, IL: Aldine.

23.

Goff

H. W.

Byrd

T. L.

Shelton

A. J.

Parcel

G. S.

(2001). Health care professionals’ skills, beliefs, and expectations about screening for domestic violence in a border community. Family & Community Health, 24, 39-54.

24.

Harwell

T. S.

Casten

R. J.

Armstrong

K. A.

Dempsey

Coons

H. L.

Davis

(1998). Results of a domestic violence training program offered to the staff of urban community health centers. Evaluation Committee of the Philadelphia Family Violence Working Group. American Journal of Preventive Medicine, 15, 235-242.

25.

Hathaway

J. E.

Mucci

L. A.

Silverman

J. G.

Brooks

D. R.

Mathews

Pavlos

C. A.

(2000). Health status and health care use of Massachusetts women reporting partner abuse. American Journal of Preventive Medicine, 19, 302-307.

26.

Hilbert

J. C.

Kolia

VanLeeuwen

D. M.

(1997). Abused women in New Mexican shelters: Factors that influence independence on discharge. Affilia, 12, 391-407. doi:10.1177/088610999701200403

27.

Hosmer

D. W.

Jr. Lemeshow

Sturdivant

R. X.

(2013). Applied logistic regression (Vol. 398). Hoboken, New Jersey: John Wiley & Sons, Inc.

28.

Jonassen

J. A.

Mazor

K. M.

(2003). Identification of physician and patient attributes that influence the likelihood of screening for intimate partner violence. Academic Medicine, 78(10 Suppl.), S20-S23.

29.

Jones

Danish

(2011). The 2011 State Physician Workforce Data Book. Washington, DC: Association of American Medical Colleges.

30.

Klap

Tang

Wells

Starks

S. L.

Rodriguez

(2007). Screening for domestic violence among adult women in the United States. Journal of General Internal Medicine, 22, 579-584. doi:10.1007/s11606-006-0097-9

31.

Mancoske

R. J.

Standifer

Cauley

(1994). The effectiveness of brief counseling services for battered women. Research on Social Work Practice, 4, 53-63.

32.

McNutt

L. A.

Carlson

B. E.

Gagen

Winterbauer

(1999). Reproductive violence screening in primary care: Perspectives and experiences of patients and battered women. Journal of the American Medical Women’s Association, 54, 85-90.

33.

Montalvo-Liendo

Wardell

D. W.

Engebretson

Reininger

B. M.

(2009). Factors influencing disclosure of abuse by women of Mexican descent. Journal of Nursing Scholarship, 41, 359-367. doi:10.1111/j.1547-5069.2009.01304.x

34.

Morrison

L. J.

Allan

Grunfeld

(2000). Improving the emergency department detection rate of domestic violence using direct questioning. Journal of Emergency Medicine, 19, 117-124.

35.

Morse

D. S.

Lafleur

Fogarty

C. T.

Mittal

Cerulli

(2012). “They told me to leave”: How health care providers address intimate partner violence. Journal of the American Board of Family Medicine, 25, 333-342. doi:10.3122/jabfm.2012.03.110193

36.

National Center for Injury Prevention and Control. (2003, March). Costs of intimate partner violence against women in the United States. Atlanta, Georgia: Centers for Disease Control and Prevention.

37.

National Coalition Against Domestic Violence. (2011). Fact Sheets. 2014. Retrieved from http://ncadv.org/files/National%20Statistics%20Domestic%20Violence%20NCADV.pdf

38.

The National Intimate Partner and Sexual Violence Survey. (2014). Injury prevention and control. Retrieved from http://www.cdc.gov/violencePrevention/NISVS/index.html

39.

Nelson

J. C.

Johnston

(2004). Screening for family and intimate partner violence. Annals of Internal Medicine, 141, 81; discussion 82.

40.

Panchanadeswaran

McCloskey

L. A.

(2007). Predicting the timing of women’s departure from abusive relationships. Journal of Interpersonal Violence, 22, 50-65. doi:10.1177/0886260506294996

41.

Ramsay

Richardson

Carter

Y. H.

Davidson

L. L.

Feder

(2002). Should health professionals screen women for domestic violence? Systematic review. British Medical Journal, 325, 314.

42.

Rodriguez

M. A.

Bauer

H. M.

McLoughlin

Grumbach

(1999). Screening and intervention for intimate partner abuse: Practices and attitudes of primary care physicians. Journal of the American Medical Association, 282, 468-474.

43.

Rorie

Backes

Chahal

(2014). Services for IPV victims: Encouraging stronger research methods to produce more valid results (National Institute of Justice Journal No. 274). Washington, DC: National Institute of Justice.

44.

Saiki

L. S.

Lobo

M. L.

(2011). Disclosure: A concept analysis. Journal of Advanced Nursing, 67, 2713-2722. doi:10.1111/j.1365-2648.2011.05741.x

45.

Singh

Petersen

Singh

S. R.

(2014). Intimate partner violence victimization: Identification and response in primary care. Primary Care, 41, 261-281. doi:10.1016/j.pop.2014.02.005

46.

Slutkin

(2013, February 6). Forum on Global Violence Prevention; Board on Global Health; Institute of Medicine; National Research Council. Contagion of Violence: Workshop Summary. Washington, DC: National Academies Press (US); II.9, Violence is a Contagious Disease. Retrieved from http://www.ncbi.nlm.nih.gov/books/NBK207245

47.

Stayton

C. D.

Duncan

M. M.

(2005). Mutable influences on intimate partner abuse screening in health care settings: A synthesis of the literature. Trauma Violence Abuse, 6, 271-285. doi:10.1177/1524838005277439

48.

Svavarsdottir

E. K.

Orlygsdottir

Gudmundsdottir

(2015). Reaching out to women who are victims of intimate partner violence. Perspectives in Psychiatric Care, 51, 190-201. doi:10.1111/ppc.12080

49.

Thackeray

Stelzner

Downs

S. M.

Miller

(2007). Screening for intimate partner violence: The impact of screener and screening environment on victim comfort. Journal of Interpersonal Violence, 22, 659-670. doi:10.1177/0886260507300206

50.

Tower

L. E.

(2006). Barriers in screening women for domestic violence: A survey of social workers, family practitioners, and obstetrician-gynecologists. Journal of Family Violence, 21, 245-257. doi:10.1007/s10896-006-9024-4

51.

Tulchin

O’Neil

(2013). New statewide poll on Californians’ attitudes toward domestic violence: Strong support for treating domestic violence as a health issue. Tulchin Research: Polling & Strategic Consulting. Retrieved from http://www.blueshieldcafoundation.org/sites/default/files/u14/Blue%20Shield%20Domestic%20Violence%20Statewide%20Public%20Memo%20FINAL.pdf

52.

Williamson

K. J.

Coonrod

D. V.

Bay

R. C.

Brady

M. J.

Partap

Wolf

W. L.

(2004). Screening for domestic violence: Practice patterns, knowledge, and attitudes of physicians in Arizona. Southern Medical Journal, 97, 1049-1054.