Article Text

Original research
Cross-sectional online survey of clinicians’ knowledge, attitudes and challenges to screening and counselling adolescents and young adults for substance use
  1. Anu L Gorukanti1,
  2. Kim S Kimminau2,
  3. Hilary A Tindle3,4,
  4. Jonathan D Klein5,
  5. Julie Gorzkowski6,
  6. Kristen Kaseeska6,
  7. Raabiah Ali7,
  8. Lavisha Singh8,
  9. Sean P David9,10,
  10. Bonnie Halpern-Felsher1
  1. 1Department of Pediatrics, Stanford University, Stanford, California, USA
  2. 2Department of Family and Community Medicine, University of Missouri Columbia, Columbia, Missouri, USA
  3. 3Department of Medicine, Vanderbilt University Medical Center, Nashville, Tennessee, USA
  4. 4Medicine, Veterans Affairs Tennessee Valley Healthcare System, Nashville, Tennessee, USA
  5. 5Department of Pediatrics, University of Illinois at Chicago, Chicago, Illinois, USA
  6. 6Department of Healthy Resilient Children Youth and Families, American Academy of Pediatrics, Itasca, Illinois, USA
  7. 7Healthcare Settings Program, Chicago Department of Public Health, Chicago, Illinois, USA
  8. 8Department of Biostatistics, NorthShore University HealthSystem, Evanston, Illinois, USA
  9. 9Department of Family Medicine, University of Chicago Pritzker School of Medicine, Chicago, Illinois, USA
  10. 10Department of Family Medicine, NorthShore University HealthSystem, Evanston, Illinois, USA
  1. Correspondence to Dr Sean P David; spdavidmd{at}gmail.com

Abstract

Objective To examine adolescent healthcare clinicians’ self-reported screening practices as well as their knowledge, attitudes, comfort level and challenges with screening and counselling adolescents and young adults (AYA) for cigarette, e-cigarette, alcohol, marijuana, hookah and blunt use.

Design A 2016 cross-sectional survey.

Setting Academic departments and community-based internal medicine, family medicine and paediatrics practices.

Participants Adolescent healthcare clinicians (N=771) from 12 US medical schools and respondents to national surveys. Of the participants, 36% indicated male, 64% female, mean age was 44 years (SD=12.3); 12.3% of participants identified as Asian, 73.7% as white, 4.8% as black, 4.2% as Hispanic and 3.8% as other.

Primary and secondary outcome measures Survey items queried clinicians about knowledge, attitudes, comfort level, self-efficacy and challenges with screening and counselling AYA patients about marijuana, blunts, cigarettes, e-cigarettes, hookah and alcohol.

Results Participants were asked what percentage of their 10–17 years old patients they screened for substance use. The median number of physicians reported screening 100% of their patients for cigarette (1st, 3rd quartiles; 80, 100) and alcohol use (75, 100) and 99.5% for marijuana use (50,100); for e-cigarettes, participants reported screening half of their patients and 0.0% (0, 50), (0, 75)) reported screening for hookah and blunts, respectively. On average (median), clinicians estimated that 15.0% of all 10–17 years old patients smoked cigarettes, 10.0% used e-cigarettes, 20.0% used marijuana, 25.0% drank alcohol and 5.0% used hookah or blunts, respectively; yet they estimated lower than national rates of use of each product for their own patients. Clinicians reported greater comfort discussing cigarettes and alcohol with patients and less comfort discussing e-cigarettes, hookah, marijuana and blunts.

Conclusions This study identified low rates of screening and counselling AYA patients for use of e-cigarettes, hookahs and blunts by adolescent healthcare clinicians and points to potential missed opportunities to improve prevention efforts.

  • Child & adolescent psychiatry
  • PAEDIATRICS
  • PUBLIC HEALTH
  • PRIMARY CARE

Data availability statement

Further data resulting from the survey are available from the authors on request. The data used for these analyses were collected as a survey that was administered through multiple sources, including a main Qualtrics survey link and through the academic societies and clinical departments participating in the study. The authors can provide results upon request for specific survey items if contacted by others who wish to extend the research further or for collaboration. Requests for data would need to be approved by the participating entities that surveyed their memberships. The data are not available in a public registry.

http://creativecommons.org/licenses/by-nc/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

STRENGTHS AND LIMITATIONS OF THIS STUDY

  • This study uses cross-sectional data from adolescent healthcare clinicians (physicians and other adolescent health professionals) from a survey of faculty from 12 US medical schools and the membership of three national organisations representing adolescent health, family medicine and paediatric providers, representing the perspectives of 771 clinicians on adolescent and young adult substance use including alcohol, cigarette smoking, e-cigarettes and use of marijuana, hookah or blunts.

  • The survey did not directly query adolescent and young adult patients, and therefore, cannot compare adolescent health clinicians’ perspectives of adolescents and young adults substance use with self-reported use of their patients.

  • The cross-sectional nature of this survey limits the ability to capture the full complexity of the rapidly changing landscape of tobacco products such as e-cigarettes and heated tobacco.

Introduction

Substance use and abuse remain among the most concerning and challenging public health issues in America, affecting health across the lifespan.1 While cigarette smoking has declined among adolescents and young adults (AYA), use of electronic cigarettes (e-cigarettes) and marijuana continue to rise, and rates of hookah and alcohol use remain high.2–5 Substantial data exist on the health impact in the USA associated with cigarettes, e-cigarettes, other tobacco products, marijuana and alcohol.6–10 Ultimately, substance use remains a critically relevant issue in paediatrics and family medicine, given that the majority of adult users initiate their substance use during adolescence or young adulthood.11

Clinicians play an important role in the identification, prevention and treatment of AYA substance use.12 13 There are published guidelines providing recommendations regarding screening AYA for substance use. The US Preventive Services Task Force recommends screening all adolescents for tobacco use including e-cigarettes (evidence level B) and screening adults for ‘unhealthy drug use’, not including alcohol or tobacco use (evidence level B).14 15 The American Academy of Family Physicians (AAFP) and the American Academy of Pediatrics both recommend screening and cessation support for adolescents for these substances as well as marijuana.12 16–19

Despite these guidelines, studies show that while clinicians often screen adolescent and young adult patients for cigarette and alcohol use, they are less likely to screen and counsel for other substances such as e-cigarettes or hookah.20–23 In addition, the existing literature on clinician screening practices for adolescent and young adult substance use have focused on examining a limited number of products such as e-cigarettes, cigarettes, marijuana, or alcohol use, rather than studying screening practices for multiple substances. Studies on screening practices for e-cigarettes have shown that paediatric clinicians feel less comfortable than adult clinicians in discussing such products with patients.24 In a 2021 study of paediatric tobacco screening practices, 67% of clinicians reported counselling adolescents who smoke cigarettes about the dangers of e-cigarettes while 57% counselled on the dangers of e-cigarettes to those who were non-smoking.25 Studies looking at barriers to screening and counselling for alcohol and marijuana use have identified the impact of insufficient time as well as lack of training and treatment resources.25–27 To our knowledge, there are few studies that have comprehensively examined clinicians’ self-reported screening behaviours, attitudes and perceived challenges towards screening AYA across different substances as well as more novel products.

The aim of this study was to survey adolescent healthcare clinicians about their knowledge, attitudes, comfort level, self-efficacy, self-reported screening behaviours and challenges with screening and counselling AYA across a range of tobacco products (cigarettes, e-cigarettes and hookah), alcohol and marijuana (including blunts). The goals of this study were to: (1) assess how frequently clinicians self-report screening for and counselling about these substances during clinic visits, and (2) assess healthcare clinicians’ knowledge about, attitudes towards and comfort level with screening and counselling AYA about these products. We expect these findings will generate hypotheses on how to best support clinician education on adolescent and young adult substance use to enable further screening and counselling.

Methods

Participants

The study population included adolescent healthcare clinicians including general internal medicine physicians, paediatricians and family medicine physicians who care for adolescent and young adult patients. Clinicians from several sample frames were recruited and surveyed: 1203 members of the Society for Adolescent Health and Medicine (SAHM); 335 participants in a prior randomised control trial (ClinicalTrials.gov NCT01312480; post-results) of adolescent tobacco cessation conducted through Paediatric Research in Office Settings, the primary care practice-based research network of the American Academy of Paediatrics,28 called the “Adolescent Health in Pediatric Practice” (AHIPP) sample; and 700 physicians from the AAFP’s Member Insight Group (currently called the Member Insight Exchange). In addition to these professional societies, substance and tobacco use researchers were identified using convenience sampling from the investigators’ research networks at 12 institutions (Stanford; University of California San Francicso (UCSF); University of Texas (UT) Health Sciences Center; University of Florida; Vanderbilt University Medical Center; University of Massachusetts, Worchester; University of California Los Angeles (UCLA); Dartmouth; Children’s Hospital of Philadelphia; Johns Hopkins; University of Michigan; Columbia University) and were asked to distribute the survey to all clinical departments serving adolescent and young adult patients. The only inclusion criterion was for the respondents to be clinicians serving adolescents and/or young adult patients. A limited number of survey participants who were members of these associations were in social work and public health, as indicated by table 1. No patients were involved in this study. Electronic informed consent was required before participation in the survey. At the start of the survey, participants reviewed a consent form that included the purpose of the study, a statement that the study was voluntary, that individual participant results would be confidential and anonymous, and that participants could discontinue at any time. Contact information for the senior principal investigator (BH-F) was provided should any questions arise. All participants had to confirm they consented to participate before moving onto the survey.

Table 1

Participant demographics and practice characteristics (N=771)

Patient and public involvement

There was no public consultation or patient involvement in the planning of this study.

Procedures

The survey was fielded between February and June 2016. Participants received an email that contained a link to complete the survey. The survey for this study was conducted using Qualtrics software (Qualtrics, Provo, Utah, USA).

A staff member or representative from each sample frame sent out an introductory e-mail which included a brief description of the study and the link to the survey. These emails were sent to the divisions of adolescent medicine and to the departments of paediatrics, internal medicine and family medicine to ask clinicians who provided clinical care for adolescent and young adult patients to complete the survey. The survey was open for a 6-week period for each group of participants, with reminders 2 weeks and 4 weeks postsurvey launch.

Participants from SAHM, university health centres and AHIPP all received the full survey. Members of the AAFP who self-select an option to receive routine member surveys through the AAFP marketing department received an abbreviated version of the survey due to standard protocol which limits surveys to 10 questions.

Measures

The survey items were adapted from previous preventive service studies focusing on healthcare clinicians’ attitudes towards and practices concerning screening and counselling AYA regarding health risk behaviours.13 29 30 Outcomes were measured using ordinal multiple choice responses and continuous numeric variables, as described below. We then pilot tested the survey with over 10 clinicians similar to the participants in our sample, and revised the survey based on the responses and their feedback.

Demographics and practice characteristics

Participants were asked the following questions, all of which were categorical variables unless otherwise specified: gender, age (continuous, quantitative variable), race/ethnicity, year they graduated from medical school/graduate school (categorical, ordinal variable), specialty in which they completed their primary training, whether or not they completed an adolescent medicine fellowship, in which state they practice medicine, whether the state they live in had legalised medical or recreational marijuana, their professional society memberships, setting for their clinic practice (private practice, hospital, clinic, etc), what percentage of time they engage in patient care (continuous, quantitative variable), and what percentage of the time they work with each age group (continuous, quantitative variable).

Knowledge of tobacco products and perceptions of use

Participants were asked whether they had ever heard of e-cigarettes, hookah or blunts (yes/no). Participants were also asked to estimate what percentage of all 10–17 years old and 18–26 years old they believed were using each of the following products: cigarettes, e-cigarettes, hookah, marijuana, blunts or alcohol. They were then asked to estimate what percentage of their patients between 10–17 and 18–26 years they thought were using each of these products. These two sets of questions were treated as continuous, quantitative variables.

Health beliefs concerning each substance

Participants were asked about their health beliefs concerning each product. They were first asked to rate their perception of how addictive cigarettes, e-cigarettes, hookah, marijuana, blunts or alcohol, using five ordinal response options, from ‘not at all addictive’ to ‘extremely addictive’. Next, participants were given a scenario in which they were asked to imagine an adolescent and young adult who used each product 2–3 times a day, every day and then asked to rate how harmful each of these products would be for their health using a 5-point ordinal response set from ‘not at all harmful’ to ‘extremely harmful’. Participants were asked how strongly they agreed or disagreed with the statements ‘E-cigarettes help addicted adult smokers quit using cigarettes’ and ‘adolescents/young adults who use e-cigarettes are more likely to later go on and use cigarettes’. Participants were then asked how likely they were to recommend an adolescent/young adult patient who is smoking 2–3 cigarettes a day to switch to e-cigarettes (very likely to very unlikely). They were also asked if they would recommend parents switch to e-cigarettes if they learnt that parents of an adolescent/young adult patients were smoking (very likely to very unlikely). These were all treated as ordinal measures.

Self-efficacy

Participants were asked about their self-efficacy with screening and counselling AYA about the six products in this study. They were first asked to indicate how comfortable they feel talking to adolescents about each of the six products using an ordinal measure with four response options, from ‘very uncomfortable’ to ‘very comfortable’. They were then asked to indicate how confident they felt explaining the health effects of these products to these patients, using an ordinal measure with four response options from ‘very unconfident’ to ‘very confident’.

Screening and counselling practices

Participants were asked to estimate what percentage of their patients between the ages of 10–17 and 18–26 years they screened, counselled and/or referred to counselling/substance use treatment for each of the six products, separately. This was analysed as a quantitative variable.

Barriers and resources

Participants were asked to indicate which of the following were their biggest challenges for screening and counselling patients about tobacco products (check all that apply): lack of support from administration; lack of knowledge about different tobacco products; do not believe screening/counselling will have any impact on their substance use; uncomfortable talking to adolescents about substance use; lack of training in screening/counselling adolescents for tobacco products; inadequate time in office visit; and lack of place for referral/lack of cessation resources. Participants were asked about their biggest challenges for screening and counselling for marijuana products (check all that apply) with the same questions as above and the unique question of ‘difficult to screen and counsel about marijuana as it is now legal in our state’. Finally, participants were asked about the resources/support they would need to increase their rates of screening/counselling patients for tobacco and/or marijuana products. These data were treated as categorical responses.

Analysis

We analysed (using χ2 and analysis of variance (ANOVA)) whether there were significant differences among the three main sample groups (AAFP, AHIPP, SAHM) in survey responses; data not shown. Since there were no significant differences in patterns of results, we collapsed across the samples and reported findings for all participants together.

A preliminary profile of the data examined means with SD, medians with first (25th percentile) and third (75th percentile) quartiles (for quantitative data only), and frequencies with percentages (table 1). Shapiro-Wilk tests were used to determine the normality of continuous response variables. As none of the variables appeared to satisfy normal distribution requirements, we used the non-parametric Kruskal-Wallis test to compare providers’ estimated prevalence of different substance use (table 2), providers’ perception of harm and addiction of different substances (table 3), providers’ perceptions of e-cigarette safety and self-efficacy (table 4), and providers’ comfort level with screening and referral practices for different substances (table 5). Following Kruskal-Wallis test, for all possible pair comparison of different substances, we used Dwass-Steel-Critchlow-Fligner (DSCF) test (detailed in online supplemental tables 1–3).31–33 As opposed to the all-pairs comparison procedures that depend on Kruskal ranks, the DSCF test is an extension of the U-test as reranking is conducted for each pairwise test. All participant responses in tables are presented as medians. We also report first and third quartiles ranges for quantitative but not ordinal measures. P values less than 0.05 were considered statistically significant. SPSS V.27.0 and SAS V.9.4 (SAS Institute) were used for all statistical analyses. The p values were not adjusted for multiple comparisons.

Table 2

Providers’ estimated prevalence of substance use among all adolescents and young adults and among their patients (medians and IQR)

Table 3

Providers’ perception of harm and addiction of various substances

Table 4

Providers’ perceptions of e-cigarette safety and self-efficacy with substance use

Table 5

Provider comfort level with and screening and referral practices for substances (median and quartiles)

Results

Demographics

While 828 participants started the survey, 57 respondents answered fewer than half of the questions and were therefore excluded from the sample, resulting in a final N of 771 participants. For the demographic data, participants from the AAFP group were not asked to complete most of the demographic questions due to a restriction on the number of questions asked.

Of the participants, 36.0% identified male and 64.0% identified female as their gender identity. Participants had a mean age of 44 years (SD=12.3); 12.3% of participants identified as Asian, 73.7% as white, 4.8% as black, 4.2% as Hispanic and 5.0% as other. There was a difference in the distribution for gender identity (60% female for AHIPP, 72% female SAHM and 44.9% female AAFP). However, race/ethnicity demographics distribution was similar among the three samples.

Among the respondents, 73.8% identified paediatrics as their specialty, with 11.0% in family medicine and 15.2% in other specialties such as internal medicine, public health and social work. Ninety-two per cent of clinicians provided care for 10–17 years old patients and 87.2% provided care for 18–26 years old patients. Clinicians worked in a variety of practice settings, with 30.2% working in a hospital-based clinic and 37.8% working in a university-based clinic (table 1).

Clinicians’ estimated prevalence of substance use among all AYA and their patients

When asked to estimate the rates of substance use among all 10–17 years old patients, based on median response by participants, participants perceived that 25.0% (Q1/Q3=17, 45) of all 10–17 years old patients used alcohol, 20.0% (Q1/Q3=10, 30) used marijuana, 15.0% (Q1/Q3=10, 20) used cigarettes, 10.0% (Q1/Q3=5, 15) used e-cigarettes and 5.0% ((Q1/Q3=2, 10), (Q1/Q3=3, 10)) used hookah and blunts, respectively. For their 10–17 years old patients, they estimated that 20.0% (Q1/Q3=10, 33) of their patients used alcohol, 15.0% (Q1/Q3=5, 25) used marijuana, 10.0% (Q1/Q3=5, 15) used cigarettes, 2.0% (Q1/Q3=1, 5) used hookah and 5.0% ((Q1/Q3=1, 10), (Q1/Q3=2, 10)) used e-cigarettes and blunts, respectively (see table 2 for more details on rates across ages). As detailed in table 2, in general, within each age group, participants estimated higher rates of alcohol use, followed by marijuana and then cigarettes.

We conducted non-parametric Kruskal-Wallis test followed by DSCF test for pairwise comparison to determine whether participants estimated different rates of substance use among all 10–17 and 18–26 years old individuals in the general population versus their own patients (online supplemental table 1). Participants estimated significantly more use of all substances except blunts among all 10–17 years old patients compared with their own patients. In addition, there was a statistically significant difference in participants’ estimates of use between all 18–26 years old and their 18–26 years old patients for all substances except for marijuana (p<0.0001). Participants estimated higher use rates across all substances except blunts for 18–26 years old patients in comparison to 10–17 years old patients.

Harm and addiction

The providers’ response on the perceived addictiveness across the different substances was a statistically significantly difference based on Kruskal-Wallis test. Each question response was measured using an ordinal measure with five response options (1—extremely addictive to 5—not all addictive). Pairwise comparison test (DSCF) showed statistically significant difference in the median response by clinicians for the perceived addictiveness of cigarettes versus e-cigarette (median 1 vs 2, p<0.0001), and cigarette versus marijuana (median 1 vs 3, p<0.0001).

Similarly, there was a statistically significant difference in the perceived harmfulness of different products (using an ordinal measure with five response options, from 1—extremely harmful to 5—not at all harmful). Pairwise comparison showed that there was a statistically significant difference in providers’ perceived harmfulness for cigarette use compared with e-cigarette use (median 1 vs 2, p<0.0001), and cigarette use compared with marijuana use (median 1 vs 2, p<0.001) (see table 3 and online supplemental table 1 for more details).

Utility of e-cigarettes to quit smoking

Based on an ordinal measure with four response options ((1—very unlikely to 4—very likely), the median response of providers on the likelihood that they would recommend an adolescent/young adult who was smoking 2–3 cigarettes a day switch to e-cigarettes was 1 (very unlikely). Median response of providers on likelihood that they would recommend a parent who was smoking cigarettes to switch to e-cigarettes was 1 (very unlikely).

Based on an ordinal measure with four response options (1—strongly disagree to 4—strongly agree), participant responses indicated a moderate level of disagreement (median=2, disagree) with the statement that ‘E-cigarettes help addicted smokers quit cigarettes,’ and a moderate level of agreement (median=3, agree) with the statement that ‘Adolescents/young adults who use e-cigarettes are more likely to later go on and use cigarettes’ (see table 4).

Self-efficacy and screening

The providers’ response to the question, ‘How comfortable do you feel talking to your adolescent/young adult patients about the following products?’ was measured using an ordinal measure with four response options (1—very uncomfortable to 4—very comfortable). The median response reported by participants was 4 (very comfortable) for cigarettes, e-cigarettes, marijuana and alcohol, whereas the median response for hookah and blunts was 3 (somewhat comfortable). There was statistically significant difference in provider’s reported comfort for different substances (p<0.0001, see table 5). Detailed pairwise comparison results are reported in online supplemental table 1.

When asked how confident they felt discussing the health impact of different substances, participants responded using an ordinal measure with four response options (1—very unconfident to 4—very confident). There was a statistically significant difference in provider confidence for discussing the health impact of different products (p<0.0001, table 5). The median response was 4 (very confident) for discussing cigarettes, marijuana and alcohol, whereas, median response was 3 (somewhat confident) for e-cigarettes, hookah and blunts. Notably, pairwise comparisons showed participants were significantly more confident in discussing the health impact of marijuana compared with e-cigarettes (median 4 vs 3, p<0.0001).

When asked to estimate what percentage of their 10–17 years old patients they screen for substance use, the median response was that participants in this study screened 100.0% of their patients for cigarette use (median Q1/Q3=80, 100), and for alcohol use (Q1/Q3=75, 100) and 99.5% (Q1/Q3=50, 100) for marijuana use, while participants screened 50.0% (Q1/Q3=0, 100) of patients for e-cigarette use, 0.0% (Q1/Q3=0, 50) for hookah use and 0.0% (Q1/Q3=0, 75) for blunt use (p<0.0001). For 18–26 years old patients, the median number of clinicians screened 100.0% (Q1/Q3=100, 100) of patients for cigarette use and 100.0% (Q1/Q3=95, 100) for alcohol use. In comparison, the clinicians screened 75.0% (Q1/Q3=0, 100) of patients for e-cigarette use, 0.0% (Q1/Q3=0, 50) for hookah use, 100.0% (Q1/Q3=90, 100) for marijuana use and 0.0% (Q1/Q3=0, 100) for blunt use (p<0.0001). There was a significant difference in clinician-estimated screening rates between products (see table 5).

When asked what percentage of their 10–17 years old they counsel about substance use, the median numbed of clinicians counselled 90.0% (Q1/Q3=50, 100) of patients on cigarette use, 80.0% (Q1/Q3=50, 100) on alcohol use and 75.0% (Q1/Q3=25, 100) on marijuana use. In comparison, participants counselled 20.0% (Q1/Q3=0, 80) of patients on e-cigarette use, 0.0% (Q1/Q3=0, 10) on hookah use and 0.0% (Q1/Q3=0, 40) on blunt use (p<0.001). For 18–26 year-old patients, the median number of clinicians counselled 100.0% of patients (Q1/Q3=50, 100) on cigarette use and 100.0% (Q1/Q3=50, 100) on alcohol use, respectively, and 90.0% (Q1/Q3=25, 100) of patients on marijuana use (p<0.0001). Participants estimated counselling 30.0% (Q1/Q3=0, 100) of patients on e-cigarette use; in contrast, they reported counselling 0.0% (Q1/Q3=0, 20) of patients on hookah use and 0.0% (Q1/Q3=0, 50) on blunt use (p<0.0001). Of note, there was a significant difference in clinician counselling and referral rates between products. P values to compare providers’ estimated prevalence by substance type were derived from Kruskal-Wallis test (table 5, online supplemental table 3).

Barriers to screening and counselling

The most cited challenges for screening and counselling patients about tobacco and marijuana were time (N=375/771 (48.6%); N=366/771 (47.4%) for tobacco and marijuana, respectively), lack of places to refer for treatment for tobacco (N=288/771 (37.3%)) and marijuana (N=326/771 (42.2%)), lack of training in screening/counselling on tobacco and marijuana (N=102/771 (13.2%); N=139/771 (18%), respectively), and lack of knowledge of different tobacco products (N=116/771 (15%)). When asked what resources or supports might increase screening and counselling, participants reported a need for substance use treatment referral resources (N=303/637 (47.5%)), increased time with patients during clinic visits (N=270/637 (42.3%)), increased training in screening/counselling adolescents for substance use (N=163/637 (25.6%)) and resources for learning more about different tobacco and marijuana products (N=175/637 (26.0%)).

Discussion

Our study assessed healthcare clinicians’ knowledge, attitudes, self-efficacy, screening and counselling practices with their adolescent and young adult patients across a range of substances, including cigarettes, e-cigarettes, hookah, marijuana (including blunts) and alcohol.

In general, clinicians believed that the most used substances among their 10–17 years old were marijuana and alcohol. In contrast, the median number of physicians believed 5% of their 10–17 years old patients used e-cigarettes (IQR=2–10). This reported prevalence by clinicians is an underestimation when compared with national prevalence data collected the same year (2016), which showed that 44.5% of 12th graders reported trying marijuana, 61.2% of 12th graders had used alcohol and 17.5% of 8th graders and 33.8% of 12th graders reported trying e-cigarettes.34 35 Even more striking is the fact that most clinicians believed their 10–17 years old patients were less likely to use any of the reported substances when compared with all 10–17 years old. Similar trends held true for clinicians’ perceived prevalence of substance use between their and all 18 and 26 years old patients (except for marijuana). While there is limited literature documenting this phenomenon, this difference in perceive prevalence may be important for future study.

Most adolescent healthcare clinicians in our sample reported that they screen and counsel 10–17 years old patients for cigarettes, marijuana and alcohol, even as use of these substances has decreased in the last decade.36 In contrast, only 50.0% report that they screen their patients for e-cigarette use and fewer for hookah and blunt use. Even fewer clinicians report counselling regarding these products. This trend is reflected in recent studies on tobacco screening where only two-thirds of clinicians reported counselling on e-cigarette use, even though e-cigarette use has continued to rise among adolescents in the ensuring years,25 37 For 18–26 years old patients, many clinicians in our sample screen and counsel for cigarettes, alcohol and marijuana, though rates of screening and counselling are lowest for marijuana among these three substances. Greater than half of clinicians in this cohort screen for e-cigarette use, but fewer subsequently counsel about these products.

Discrepancies between screening and counselling could reflect a variety of barriers such as decreased clinician comfort and knowledge about these products, appointment time or counselling skills as well as beliefs about utility of counselling, or whether the clinicians provided confidential, private time for these discussions to be able to proceed.25 38 Clinicians reported being more comfortable discussing cigarettes and alcohol use and least comfortable discussing hookah and blunts with their patients. Only half of respondents felt comfortable discussing e-cigarettes and less than two-thirds felt comfortable discussing marijuana. Similarly, participants felt most confident discussing the health impact of cigarette smoking and alcohol use and felt least confident discussing the health impact of hookah use. Clinicians in our sample did not feel as comfortable or confident discussing products that adolescents are most commonly using such as blunts, hookah, marijuana and e-cigarettes and their possible health effects. This lack of comfort addressing substances other than alcohol and cigarettes is consistent with literature documenting the need for substance use training in medical schools and may indicate the need for educational interventions to increase clinician comfort with addressing substance use.39 40 In addition, it is necessary to address structural barriers, such as limited appointment time in clinician visits, including limited private time with AYA patients, to ensure clinicians are able to adequately incorporate substance use screening in their clinical encounters.25

This study has several strengths. These include the survey’s sample size and range of primary care and adolescent clinician specialties, its national scope, mix of academic and community clinicians across the USA. Compared with the many other published surveys on adolescent and young adult substance use, our study may be unique in its querying of primary care clinicians’ perspectives on a national scale and a wide range of alcohol, tobacco and marijuana products.

This study also has limitations. We acknowledge that variables used for collecting participants’ gender and racial/ethnic identities have dated language and do not provide participants with sufficient options for describing their identity. While this study included a diverse sample of clinicians, data were not systematically sampled to be nationally representative and there were missing data that could together skew its representativeness, and thus limit the generalisability of the results. Other limitations are that the study was completed in 2016, so it does not capture subsequent changes in product innovation, availability, range and co-use. Consequently, clinicians’ knowledge, beliefs and behaviour may differ now from the results reported here. In addition, marijuana has continued to become more broadly legalised since study data were collected, which could also influence clinician perspectives, norms, and screening and counselling practices. Another limitation is that Bonferroni correction was not applied to the reported statistics. Therefore, it is possible that reported differences in estimates of substance use were prone to type I error. Despite these limitations, the findings provide important information on clinician knowledge, practices and concerns precisely at the time when the use of e-cigarettes escalated dramatically.

Conclusions

In our nationally distributed study, adolescent healthcare clinicians tended to underestimate use of multiple products (cigarette, e-cigarette, marijuana, hookah, blunt, alcohol) in the general population and estimate lower than population rates of use for their own patients. In addition, our study identified low rates of screening and counselling for use of e-cigarettes, hookahs and blunts by adolescent healthcare clinicians with their AYA patients as well as decreased confidence with discussing the health impacts of substances such as marijuana, hookah or blunts. As national rates of polysubstance use continue to increase, clinicians who care for adolescent and young adult patients are in a unique position to help prevent and reduce substance use by screening and counselling.

Data availability statement

Further data resulting from the survey are available from the authors on request. The data used for these analyses were collected as a survey that was administered through multiple sources, including a main Qualtrics survey link and through the academic societies and clinical departments participating in the study. The authors can provide results upon request for specific survey items if contacted by others who wish to extend the research further or for collaboration. Requests for data would need to be approved by the participating entities that surveyed their memberships. The data are not available in a public registry.

Ethics statements

Patient consent for publication

Ethics approval

This study underwent research ethics review by the Institutional Review Board at Stanford University (protocol #36727; registration #4947) and was approved to proceed with a designation of ‘exempt’ status. This study also received approval after research ethics review by the Institutional Review Board of the American Academy of Pediatrics (The AAFP does not assign a protocol number). Participants gave informed consent to participate in the study before taking part.

References

Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

Footnotes

  • SPD and BH-F are joint senior authors.

  • Twitter @spdavid

  • Contributors ALG, SPD and BH-F contributed to the study design and interpretation of data. ALG and BH-F had full access to the data. HAT, KK, JG, KK contributed to study participant recruitment. ALG, LS and RA performed the data extraction and analysis. ALG prepared the initial draft of the manuscript with additional input from SPD, BH-F and JDK. ALG and LS designed the tables with additional input from RA, JDK, SPD and BH-F. All authors contributed to the drafts and final version of the manuscript. BH-F is the guarantor of the study, has access to the data, and controlled the decision to publish.

  • Funding The research reported in this article was supported in part by the Taube Research Faculty Scholar Endowment (no award number) (Halpern-Felsher), American Academy of Pediatrics Friends of Children Fund (no award number) (Klein, Gorzkowski, Kaseeska) and NIH Clinical and Translational Science Award (award number UL1TR002366 (Kimminau) and award number UL1TR002389 (David)), and the Health Resources and Services Administration (HRSA) of the US Department of Health and Human Services (HHS) award numberUA6MC15585 (American Academy of Pediatrics (AAP) Pediatric Research in Office Settings). In addition, the American Academy of Pediatrics Julius B. Richmond Center of Excellence receives support from the Flight Attendant Medical Research Institute (FAMRI) (award number 052302_CoE (Klein, Gorzkowski, Kaseeska)).

  • Disclaimer The contents are those of the author(s) and do not necessarily represent the official views of, nor endorsement by, HRSA/HHS, the US Government, or the AAP.

  • Competing interests BH-F is a paid expert scientist in some litigation against e-cigarette companies and is an unpaid scientific advisor and expert witness regarding some tobacco-related policies. None of the other authors have any conflicting interests.

  • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.