Introduction

Identifying substance use and domestic violence during pregnancy is critical in assuring the health of mothers and babies and is a priority of the Washington State Department of Health (DOH). The Washington State DOH conducted a study to increase understanding of strategies that may be effective in promoting universal screening for substance use and domestic violence during pregnancy. Findings from the study will inform decisions about the best strategies to continue prenatal care quality improvement activities.

Background

The use of alcohol, tobacco, and illegal substances during pregnancy is a major risk factor for low birth weight, premature delivery, mental retardation, developmental disabilities (such as cerebral palsy, autism, mental retardation, vision and hearing impairments), physical abnormalities and fetal death [1]. Research supports the strong association between tobacco use during pregnancy and adverse perinatal outcomes [2]. In addition to substance use, domestic violence is emerging as one of the most serious public health problems affecting women's health in this country [3, 4]. Those experiencing domestic violence just prior to or during pregnancy are more likely to have caesarean birth, kidney infection, abruptio placentae, premature labor/birth, low birth weight and trauma directly resulting from the abuse [5]. Substance use and domestic violence often occur in tandem, although neither specifically causes the other [6–9].

Doctors with appropriate training are more likely to screen for alcohol use than those who have not had training [10]. According to an American College of Obstetricians and Gynecologists (ACOG) survey, most physicians asked about smoking but wanted to learn more about how to intervene [11]. Educating physicians and providing them with a process for implementing successful screening practices, documenting high risk patients, and ensuring that the intervention is started early and maintained during pregnancy is critical in reducing health risks to the unborn child.

Why screen?

Routine screening during pregnancy takes advantage of the frequent contact between clients and health care providers to identify problems and engage women in treatment. Identification and referral are particularly important for domestic violence victims and substance users, who are disproportionately isolated from the health care system due to fear, stigma, or shame associated with their situations. In addition, treatment for substance use during pregnancy can be more effective than at other times in a woman's life because the woman is often more motivated to change her behavior for the sake of her child [12]. Current research on prenatal smoking cessation shows that a brief intervention by a trained provider will increase smoking cessation rates [13, 14].

According to ACOG, obstetrician-gynecologists have an ethical obligation to use a protocol for universal screening questions, brief intervention and referral to treatment in order to provide patients and their families with comprehensive and effective medical care [15]. To promote screening for alcohol, drug, and tobacco use, ACOG has published several documents to encourage providers to learn established techniques for screening and intervention, to create a team approach to deal with barriers and to use external resources. ACOG also encourages obstetric providers to screen for psychosocial risk factors, such as violence and substance use and refer for essential services to address these concerns [9]. In September 2000 and again in October 2005, ACOG published guidance on smoking cessation during pregnancy, recommending that all providers identify pregnant smokers and integrate the brief intervention into their office practices [16, 17].

Washington State's effort to promote screening

Washington State recognizes the importance of universal screening for substance use and domestic violence during prenatal visits. Based on data from the 2003 Pregnancy Risk Assessment Monitoring System (PRAMS) survey, an estimated 91% of women in Washington State reported being asked by their provider about smoking during prenatal care, 83% reported being asked about alcohol use, 73% reported being asked about illegal drug use, and 61% reported being asked about domestic violence. The DOH has tried several initiatives to improve identification and screening rates for substance use and domestic violence.

In 1998, DOH initiated the Perinatal Partnership Against Domestic Violence (PPADV), a training curriculum for prenatal care providers on screening for domestic violence. A perinatal outreach nurse was paired with a domestic violence advocate in her community and together, they attended a training session on how to train health care providers to screen pregnant women for domestic violence. This collaboration resulted in an opportunity for hospital-based nurse educators and community advocates to share networking strategies and experiences from their workplaces. The PPADV, while not completely successful in its goal to train large numbers of physicians, did reach and train numerous health care professionals, including nurses, midwives, social workers, community health workers, and outreach workers.

In 1999, the Washington State Legislature provided funding to educate health care providers in the use of substance use assessment tools and interventions for pregnant women. Shortly after, the DOH Office of Maternal and Child Health established the Maternal Substance Abuse Screening Initiative for Providers. Resources for prenatal care providers included three best practice booklets, fact sheets, safety cards, clinician pocket reference cards, and a professional website. Changes were made to the uniform prenatal medical record (used by over 90% of prenatal care providers in the state) to improve the assessment and documentation of screening for substance use and violence. From January 2000 through June 2004, training was conducted in the four regional perinatal programs across the state through individualized practice training, presentations and exhibits at professional meetings.

This project required regional program trainers to develop strategies for engaging physicians and office staff in the outpatient setting instead of in hospitals and at conferences and had limited success. Providers were reluctant to attend this training and office staff had difficulty scheduling time on a clinic day. Many health care providers stated that they were already informed and adequately screening, and did not consider this topic a priority. Perceived lack of treatment options and difficulty with intervention steps following identification also affected provider motivation to learn about and change screening practice.

To address the need to increase smoking cessation efforts within maternity care, the DOH developed a best practice booklet that provides guidelines and tools for doing a brief intervention with pregnant women who smoke. In addition, Washington Medicaid has added a smoking cessation counseling benefit for pregnant women in its fee-for-service coverage. The benefit covers pregnant women up to two months postpartum and when appropriate, payment for Bupropion SR (Zyban). DOH developed a provider reference card with information on the benefit and guidelines for prescribing Bupropion SR (Zyban). All of these materials were mailed to obstetrical providers statewide and distributed at state medical meetings and conferences and through regional perinatal programs.

Despite improvement in screening rates and numerous available referral resources, Washington has not achieved universal screening for substance use and domestic violence. Therefore in 2003, DOH conducted a study to learn what physicians thought might be more effective strategies for promoting universal screening for substance use and domestic violence during pregnancy. The purpose of the study was to identity physician perceptions on the importance of screening, barriers to effective prenatal screening, awareness of resources from the Washington State Department of Health, and the effectiveness of various provider training strategies.

Methods

This qualitative study consisted of both telephone interviews and focus groups with Washington State physician providers of obstetric care. Participant recruitment, interviews and the focus groups were conducted by a professional research organization. Their summary report identified common themes, patterns, and trends. This work was screened by the Washington State Institutional Review Board (IRB) and was classified as public health practice and not research. Therefore, review and approval by the Washington State IRB was not required.

Obstetric providers were selected using systematic sampling of the Washington State Integrated Provider Network Database, a comprehensive list of providers maintained by the Washington Department of Social and Health Services (DSHS). Medicaid, the Basic Health Plan and all insurance carriers who participate in the state's Public Employees Benefits Board contribute to this database, which was originally intended as a means for the general public to find medical providers. Before sampling, the list was sorted geographically by perinatal region, zip code, city and street name. A systematic sample of physicians was selected to provide geographic diversity across the state and ensure that no two providers were selected from the same practice. A separate sample was selected for the semi-structured interviews and for each of the four focus groups. Eligibility was limited to physicians who provide obstetric care.

Semi-structured interviews

Individual semi-structured telephone interviews were conducted with 8 physicians, recruited from a systematic sample of 30 physicians who practice obstetric care in Washington. The interviews lasted 60–90 min and physicians were compensated $150. Common themes were identified and guided development of the focus group protocol.

Focus groups

Efforts were made to ensure diverse participation by physicians. While all participants were physicians who provided obstetric care, half were also family practitioners. Participants included male and female physicians from large urban practices as well as solo practitioners in rural areas with fewer annual deliveries. Almost all practices served a proportion of high-risk patients and many estimated that approximately 50% of the women they served were Medicaid eligible. Three participants worked in clinics for low income, high-risk patients only, while one belonged to a group practice serving affluent women. Prior to each focus group, respondents were asked to complete a short written questionnaire that asked for general background about their screening activities and the size and nature of their medical practices.

Two in-person focus groups were held in both eastern and western Washington and two were held by telephone to facilitate inclusion of physicians serving rural communities. Phone participants received $200 while in person participants received a $250 honorarium. The groups consisted of 28 out of 38 randomly selected physicians who practice obstetric care.

The focus groups lasted 60 to 90 minutes and were moderated by a senior researcher with extensive experience in this activity. Topics covered during the interviews and focus groups included patient demographics and characteristics of the physician's practice as well as specifics on their office procedures for screening for substance use and domestic violence during pregnancy and postpartum (Table 1). The moderator used a semi-structured guide that focused on the three major areas of interest: physician opinions and attitudes about substance use and domestic violence screening and intervention for pregnant women; physician awareness of existing DOH materials about screening; and barriers and effective strategies for communication and education of physicians on screening. After each session, audio or videotapes and extensive field notes were analyzed to identify dominant themes and key points.

Table 1 Topic areas for interviews and focus groups
Full size table
Table 2 Key findings/current screening behaviors
Full size table

Results

Provider attitudes toward screening practices were consistent (Table 2). Providers felt that screening for drug use, alcohol use, and domestic violence during pregnancy is good practice; however, most screen only once. Physicians reported they were more likely to screen for tobacco and alcohol use than for drug use and domestic violence. Physicians mentioned that establishing a bond with the patient and conducting the screening interview after that bond had been formed was the best means for identifying patients with these issues. Cultural and language barriers make it more difficult to screen effectively.

Table 3 What physicians want in education materials and training
Full size table

While respondents appreciated efforts to help educate them about screening, they saw the provision of reliable and updated referral information as even more important. Many physicians said that if there was one foreseeable barrier to their ongoing screening efforts it was the fact that they often felt discouraged after identifying a potential problem then being unsure of where to refer the patient for continued assistance.

Physicians identified four themes necessary to encourage their participation in screening education (Table 3):

  • Access to current intervention and referral resources

  • Time-saving tools for screening and intervention

  • Emphasis on avoidance of legal risk through good screening

  • Patient education materials that encourage and support provider efforts.

Physicians also suggested information about access to referral, short and scientific-looking training materials, and on-site trainings for all office staff. Ineffective training strategies cited included e-mail alerts, legislative mandates, direct mailings/flyers, and telephone conferences.

Discussion

Overall, physicians in this study felt that screening pregnant women for substance use and domestic violence is an integral and worthwhile part of prenatal care. These physicians saw all screening topics as equally important, but stated they had the most difficulty screening for drugs and domestic violence. Domestic violence was discussed as especially challenging in part because of the difficulty in affecting patient behavioral change or getting the patient to use a referral. Frequency and methods for screening varied according to physician attitude, experience and preference. DOH best practice materials, while appreciated by physicians, were not considered the best strategy to encourage universal screening.

A variety of factors may explain the varying practice levels of patient screening by obstetric providers. Studies have shown that while providers generally support prenatal screening, they struggle with barriers to implementing screening which include time limitations, lack of training, and lack of resources [11, 14, 18]. Overall the physicians participating in our focus groups felt that screening should occur for substance use and domestic violence during prenatal visits. The following is a discussion of factors that influenced or presented barriers to quality screening practices.

Physicians may be influenced by the potential legal implications of screening pregnant women for alcohol, drug use, and domestic violence. Mandated reporting requirements, if present, may inhibit patient-provider communication and lead to significant ethical issues. These issues impact what the woman is willing to tell her provider and likewise, may influence the types of questions the provider is willing to ask. In South Carolina, for example, one hospital's policy was to turn positive drug testing results over to law enforcement. Though this policy was struck down by the Supreme Court in Ferguson vs. City of Charleston, it demonstrates the potential legal implications of substance use screening [19]. Washington State does not have a mandatory reporting law for health care providers for domestic violence involving pregnant women, nor is there a mandate to report substance use during pregnancy. Children's Protective Services encourages reporting positive maternal or newborn drug tests. However, the provider must decide if the woman's situation potentially endangers her children because mandatory reporting laws do exist for the abuse of children and vulnerable adults.

Physician attitudes and expectations impact screening practices. Because domestic violence interventions are often perceived to be less successful than interventions for tobacco, alcohol, or drug use, providers may feel that domestic violence screening is futile [20, 21]. Physician expectations for screening sometimes differ from domestic violence advocates’ and substance abuse providers’ expectations of screening. The medical model expects that when a positive screening occurs the physician will intervene with services that are effective and the patient is “cured.” Professionals in substance use and domestic violence consider screening the first intervention in a process or series of interventions over time that leads to a positive outcome for the patient, so expectations from a medical model perspective differ substantially from the domestic violence and substance abuse “process” perspective. Physicians may feel a sense of failure when they cannot “cure” the domestic violence survivor or the addicted victim with one quick and effective intervention. This difference requires long term attitudinal change.

Lack of time and timing of screening plays a role. Most screening for substance use and domestic violence during pregnancy occurs at the first visit with their provider [10]. For those patients who use their primary care practitioner as their obstetric provider, the first visit may pre-date prenatal care visits, and the screening questions may not be asked again during pregnancy. In addition, many pregnant women with substance use and violence issues will not disclose at the first visit. This makes the development of a trusting relationship and multiple screening points important in order to identify women who are ready to make behavioral changes and accept referrals [22].

Last, many physicians believe that they have inadequate resources to intervene; therefore, they may screen less often or screen only for those behaviors for which resources and referrals, (such as drug and alcohol treatment programs and women's shelters) are readily available and effective [10, 20]. In one study on physician beliefs about domestic violence, the majority of physicians surveyed (70%) do not believe that they have the resources available to them to assist victims of domestic violence [21]. Another study found that 95% of obstetrician-gynecologists surveyed were unsure about what patient education materials on smoking cessation were appropriate for pregnant women [11]. Increasing linkages between providers and the interventionists for each area may increase provider confidence and referrals to available resources.

Generally, continuing medical education programs are developed by planning committees who decide what medical providers need to know. While this method may ensure that critical information is provided, it may fall short of success in terms of engaging providers and in effecting provider behavioral change. Focus groups of physicians can be a useful means of identifying the issues for structuring professional education programs [23, 24]. Getting input from the target audience enables the education sponsor to understand the current care standards and how best to improve practice. Target audience input about preferred education methods and best practice tools may improve quality of care and outcomes [24].

Physicians look to their professional association for practice guidance; however, lack of awareness or disagreement about a clinical guideline have been identified as barriers to provider adherence. Other reasons providers don't adopt a best practice include lack of self efficacy, lack of outcome expectancy, and the inertia of previous practice patterns [18]. Perhaps more focus on ways to overcome all barriers would help to increase provider implementation of proven screening methods.

Limitations of this study must be acknowledged, and include those that are inherent with focus groups, self-reporting, and small sample size. Limitations of focus groups include facilitator bias and the risk of dominant members taking over and sidetracking discussions. We attempted to avoid these limitations by using professional contractors with experience in facilitating groups of physicians. The primary limitation for this study is that the results cannot be generalized to the whole provider population. However, the focus groups provided us with access to our target audience and allowed us to solicit recommendations and suggestions for future consideration.

Findings from this study will help guide efforts to educate obstetric providers about the importance of screening during pregnancy. Providers made it clear that they are more motivated to screen if they have good support, materials and interventions that help their patients. These come in part from professional groups (for example, ACOG), but this does not ensure that all providers are compliant. Standardized tools might prove helpful if administered by physicians on a voluntary basis. Legislative mandates and small monetary incentives were clearly denounced as a motivator for screening.

Although this study was small, it could be easily replicated on a larger scale to gain a broader perspective. Replication of this qualitative study with a larger sample size might provide important data to validate the findings from these focus groups. This would help to determine if differences exist in screening practices and attitudes between family practitioners and obstetricians, between urban and rural providers, between providers serving Medicaid or non-Medicaid clients, between providers under age 40 and over age 40, and between physician providers and office staff. Expansion of the study might include information on methods to train providers during internships and residency.

Findings from this study provided rich data to further explore strategies for improving universal screening of pregnant women for substance use and domestic violence. Suggestions to improve physician education in this area did not point to any one strategy that would be effective for all physicians. However, results are valuable because physicians’ attitudes about screening, barriers to universal screening, and recommendations to improve education and outreach will stimulate discussion with stakeholders and other state agencies. These qualitative data are an important component for future planning of communication and education strategies with obstetric providers.