Barriers Faced by Physicians in Screening for Substance Use Disorders Among Adolescents

We used a sequential exploratory mixed-methods design to examine provision of substance use disorder screening to adolescents and young adults (ages 12–21). First, we conducted key informant interviews with 12 experts identified through a snowball strategy. Eight of these informants were SBIRT researchers (some of them were also practicing medical doctors) affiliated with a research institution. Four were affiliated with a government entity or health plan, providing technical assistance to PCPs. Our semistructured interview protocol was constructed on the basis of a literature review. The instrument was pilot tested, and we added probes for clarifications when needed. Key informants were asked about PCP familiarity with SBIRT guidelines and tools, barriers to substance use disorder screening among adolescents and young adults, and practices for overcoming barriers. Interviews took place between July and October 2015 and were conducted by telephone. Thematic saturation was achieved; no new barriers surfaced during the final interviews. Information regarding each barrier to SBIRT was analyzed separately.

We used the information from the interviews to construct a 25-item survey, which was fielded to 75 PCPs across the United States. The survey was designed to determine whether PCPs addressed potential substance use with patients, the tools and methods used to assess risk, and barriers encountered. Survey respondents were able to select all barriers that they encountered; they were also provided space to comment on additional barriers that were not included. We also asked respondents to identify interventions that would help them increase their screening rates. Respondents were able to select multiple interventions but were not able to add additional interventions.

We used M3 Global Research panel to conduct the surveys. This panel included health care professionals who opted to participate in research, and it was designed to elicit 75 responses. The survey was fielded in 2015 from November 6 to December 1. Fifty-six responses were obtained in the first week, and the remainder were obtained on December 1. PCP respondents were mostly pediatricians, nurse practitioners, and other medical doctors (N=65, 87%); however, health educators and case managers were also surveyed (N=10, 13%). The methodology ensured variation in practice type, practice setting, size of practice, and geographic diversity, but it was not intended to be nationally representative. This research was exempt from institutional review board evaluation, and written consent was not obtained from key informants because the information collected was not private.

The majority of PCPs (at least 84%) indicated that they spoke with adolescents and young adults about risky behavior associated with substance use disorders. However, 27% never used a validated or unvalidated tool to provide systematic screening to all patients (Table 1). Of the respondents, 28% used a standard screening instrument only among patients whom they deemed to be at risk. Of PCPs who used a screening tool (N=55), 38 (69%) used a validated tool, such as CRAFFT or CAGE. The remaining 17 (31%) PCPs crafted their own tool or amended a broader health risk assessment tool.

Of the 75 surveyed PCPs, the 41 who indicated that they did not screen every patient by using a validated tool were asked about barriers to providing this service. The most frequent answers were as follows: “difficulty with tenacious parents who will not allow confidential consultation with adolescents” (51%), “process is too time consuming” (41%), “sensitivity around whether and how to involve parents” (22%), “need dedicated person or technology to screen patients” (17%), and “screening for substance use disorders is not reimbursable” (12%). Comments elaborated on screening policies without surfacing new barriers.

Responses converged with the themes heard from key informants in all areas but one. Notably, nine of the 12 key informants suggested that PCP discomfort with SBIRT and doubt about its effectiveness were barriers to its uptake. However, only 2% of survey respondents selected these issues as barriers.

All survey respondents were asked what interventions they thought would be helpful in increasing SBIRT uptake, and the majority (51%) wanted increased reimbursement. Survey respondents also indicated that dedicated technology (47%) or staff (44%) to conduct the screening and having more specialized staff (40%) would be helpful. This finding was consistent with key informants’ perceptions that PCPs may need to hire trained staff and use dedicated technology to improve their workflow to allow for a systematic approach to SBIRT.

Respondents also had doubts about the effectiveness of brief interventions and reported problems with referring patients to treatment. Although 69% of respondents said that they provided brief intervention or consultation, only 36% indicated that they thought the brief intervention was effective. Only 53% of respondents reported receiving training on providing brief interventions. Half of key informants thought that PCPs needed more guidance to successfully implement a brief intervention. Similarly, survey respondents indicated that more training to help them know how to better respond to a positive screen would help increase SBIRT uptake (N=28, 37%) .

More than two-thirds (68%) of survey respondents expressed that they usually or often perceived barriers to successful referral to treatment, such as lack of high-quality treatment options available for referral as well as lack of patient motivation and patient ability to seek treatment.

Our study indicates that most PCPs discuss substance use disorder issues with their adolescent and young adult patients, although there are limitations in how systematically they approach these conversations. These limitations included lack of time, the sensitivities of addressing substance use disorder issues with adolescents and young adults, lack of PCP training, concerns about the potential effectiveness of a brief intervention, and perceived barriers to successful referral to treatment.

This study points to several potential avenues for increasing SBIRT uptake. The time burden of systematically fielding a screening instrument, such as the CRAFFT or the CAGE, to all adolescent and young adult patients might be alleviated by changes in workflow, such as dedicated technology or specialized staff to conduct the screening. However, PCPs may need dedicated funding to implement these changes. Providing more continuing education opportunities to train PCPs on how to provide brief intervention after screening may also improve SBIRT uptake and remedy provider discomfort and perceptions of ineffectiveness. Finally, identifying ways in which PCPs can protect their adolescent patients’ confidentiality while promoting the appropriate involvement of parents in their child’s treatment may be another important avenue for increasing SBIRT use.

One limitation of this study was that it is not a representative sample. Survey respondents opted in to participate in this type of research. However, we have no reason to believe that our sampling method would systematically bias these results. Furthermore, studies that use more standard sampling strategies usually receive a low response rate, which may have the same bias.

The authors acknowledge their colleagues who provided project management and research support.