A Digital Intervention to Prevent the Initiation of Opioid Misuse in Adolescents in School-based Health Centers

Perceived risk of harm from opioid misuse was assessed using eight items adapted from the U.S. Monitoring the Future survey. Five items assess perceived risk associated with heroin use and three items assess perceived risk associated with prescription opioid misuse. Participants were asked how much risk people have of harming themselves (physically or in other ways) if they use the drug once or twice, occasionally, or regularly. Response options were "no risk," "slight risk," "moderate risk," "great risk," and "can't say, drug unfamiliar."

Items were scored from 1 to 4 (no risk = 1, slight risk = 2, moderate risk = 3, great risk = 4). Item scores were summed to create a total perceived risk score (range 8-32), with scores <32 indicating "no great perceived risk of harm" and scores=32 indicating "great perceived risk of harm."

The primary outcome was defined as the percentage of participants who reported "great perceived risk of harm" from opioids (score=32) at 3 months.

Intentions to misuse opioids were assessed using 2-4 items adapted from Skenderian et. al. capturing participants' likelihood of using prescription opioids or heroin in the next 12 months. Two primary stem questions assessed likelihood of using (1) prescription opioids and (2) heroin "even once or time." For each substance, if participants endorsed any likelihood of use, a follow-up item assessed likelihood of using the same substance nearly every month for the next 12 months. Due to skip logic, participants who selected "definitely not" on the initial stem questions were not shown follow-up items.

All items were rated on a 4-point Likert scale ranging from 1 (definitely not) to 4 (definitely will). Item responses were summed to create a total intentions score ranging from 2 to 16, with lower scores indicating lower intentions to misuse opioids.

LS mean scores were estimated using mixed-effects models at each timepoint.

Self-efficacy for refusing opioids was assessed using a two-item scale measuring participant's confidence in refusing prescription opioids or heroin if offered by a friend, adult or family member. Reponses were scored on a five-point scale ranging from 1 ("Not at all confident") to 5 ("Extremely confident") and summed across items to create a total score (range = 2-10), with higher scores indicating greater self-efficacy to refuse opioids.

LS mean scores were estimated using mixed-effects models at each timepoint.

Knowledge about opioids was assessed using a 30-item questionnaire measuring participants' understanding of opioid misuse and its risks. Each of the first 27 items was scored 1 point for a correct answer. The final 3 items were multiple-choice 'select all that apply' questions; participants received 1 point for each correct option selected. Total scores ranged from 0 to 38, with higher scores indicating greater knowledge about opioids.

LS mean scores were estimated using mixed-effects models at each timepoint.

Attitudes about opioid use were assessed using a 15-item scale measuring positive and negative expectancies about using prescription opioids or heroin. Participants rated their agreement with statements such as "If I used prescription opioids or heroin, I would feel good" (positive expectancies) or "I would get in trouble with my parents" (negative expectancies) on a five-point scale from 0 ("Strongly disagree") to 4 ("Strongly agree").

Item scores were summed separately for positive and negative expectancies. The first 8 items represented positive expectancies (score range = 0-32). Higher scores indicate stronger expectancies. For positive expectancies, the goal of the intervention is to reduce scores.

LS mean scores were estimated using mixed-effects models at each timepoint.

Attitudes about opioid use were assessed using a 15-item scale measuring positive and negative expectancies about using prescription opioids or heroin. Participants rated their agreement with statements such as "If I used prescription opioids or heroin, I would feel good" (positive expectancies) or "I would get in trouble with my parents" (negative expectancies) on a five-point scale from 0 ("Strongly disagree") to 4 ("Strongly agree").

Item scores were summed separately for positive and negative expectancies. The last 7 items represented negative expectancies (score range = 0-28). Higher scores indicate stronger expectancies. For negative expectancies, the goal is to increase scores.

LS mean scores were estimated using mixed-effects models at each timepoint.

Perceived risk of harm from opioid misuse was assessed using eight items adapted from the U.S. Monitoring the Future survey. Five items assess perceived risk of heroin use and three items assess perceived risk of prescription opioid misuse. Participants were asked how much risk people have of harming themselves if they use the drug once or twice, occasionally, or regularly. Response options were "no risk," "slight risk," "moderate risk," "great risk," and "can't say, drug unfamiliar."

Items were scored from 1 (no risk) to 4 (great risk). Item scores were summed to create a total perceived risk score (range 8-32), with higher scores indicating greater perceived risk of harm.

A dichotomous outcome was derived such that a total score of 32 indicated "great perceived risk of harm," and scores less than 32 indicated "no great perceived risk of harm."

Outcome=% of respondents who changed status from "No great perceived risk" at baseline to "Great perceived risk" at 3-months.

Lifetime use of prescription opioids was assessed by asking participants, "Have you ever used a prescription opioid without a doctor's prescription or differently than how a doctor or medical provider told you to use it?" at each timepoint. Response options were "Yes" or "No." Only participants who reported no prior misuse at baseline were included in analyses for this outcome.

The outcome was defined as the percent of participants reporting any prescription opioid misuse (i.e. 'yes') at 6 weeks post-gameplay and at 3, 6, and 12 months following baseline.

Lifetime use of prescription opioids was assessed by asking participants, "Have you ever used heroin?" at each timepoint. Response options were "Yes" or "No." Only participants who reported no prior use at baseline were included in analyses for this outcome.

The outcome was defined as the percent of participants reporting any use (i.e. 'yes') at 6 weeks post-gameplay and at 3, 6, and 12 months following baseline.

Gameplay engagement was assessed using four self-report items evaluating participants' experiences while playing the PlaySmart game, including frustration during gameplay, sensory appeal, perceived reward, and interest in playing the game. Participants rated their agreement with each statement using a five-point Likert scale ranging from "Strongly disagree" to "Strongly agree."

For positively worded items (e.g. "Playsmart was rewarding"), results are reported as the percentage of participants who responded 'Agree' or 'Strongly Agree.' For negatively worded items (e.g. "I felt frustrated"), results are reported as the percentage who responded 'Disagree' or 'Strongly Disagree.' Denominators include all participants who provided a response to the item.

Participants' experience with the PlaySmart game was assessed using six self-report items evaluating perceived learning, frustration, usability, character connection, control, and enjoyment. Example items include "This game helped me learn important things," "I felt frustrated while playing PlaySmart," and "I enjoyed playing the game." Participants rated their agreement on a five-point scale: "Strongly disagree," "Disagree," "Agree," "Strongly agree," or "Not sure."

For positively worded items, results are reported as the percentage of participants who responded 'Agree' or 'Strongly Agree.' For negatively worded items, results are reported as the percentage who responded 'Disagree' or 'Strongly Disagree.' Denominators include all participants who provided a response to the item.

Participants' perceived usability of the PlaySmart game system was assessed using four self-report items evaluating perceived complexity, learnability, and confidence in using the game. Example items include "I found the game system to be complex," "I think that most people would learn to use this game system very quickly," and "I felt very confident using the game system." Participants rated their agreement using a five-point Likert scale: "Strongly disagree," "Disagree," "Neither agree nor disagree," "Agree," and "Strongly agree."

Results are summarized as follows: for positively worded items (e.g., "I felt very confident using the game system"), the percentage of participants who responded Agree or Strongly Agree is reported; for negatively worded items (e.g., "I found the game system to be complex"), the percentage of participants who responded Disagree or Strongly Disagree is reported. Denominators include all participants who provided a response to the item.

Lifetime use of alcohol was assessed by asking participants, "Have you ever had at least one drink of alcohol?" at each timepoint. Response options were "Yes" or "No."

The outcome was defined as the percent of participants reporting any alcohol use (i.e. 'yes') at 6 weeks post-gameplay and at 3-, 6-, and 12-months following baseline.

Analysis forthcoming.

Lifetime use of marijuana was assessed by asking participants, "Have you ever used marijuana? Marijuana also is called pot, weed, or cannabis." at each timepoint. Response options were "Yes" or "No."

The outcome was defined as the percent of participants reporting any marijuana use (i.e. 'yes') at 6 weeks post-gameplay and at 3-, 6-, and 12-months following baseline.

Analysis forthcoming.

Depressive symptoms were assessed using the 8-item Patient Health Questionnaire (PHQ-8), which measures the frequency of depressive symptoms experienced over the past two weeks. Participants rated how often they were bothered by symptoms such as little interest or pleasure in doing things, feeling down or hopeless, sleep difficulties, fatigue, poor appetite or overeating, feelings of failure, difficulty concentrating, and psychomotor changes.

Each item was scored from 0 ("Not at all") to 3 ("Nearly every day") and summed to create a total score ranging from 0 to 24, with higher scores indicating greater depressive symptom severity. Standard cut points were used to characterize symptom severity (0-4 = no significant symptoms; 5-9 = mild; 10-14 = moderate; 15-19 = moderately severe; 20-24 = severe).

Data analysis forthcoming. Analyses will include participants with complete responses to all PHQ-8 items at each timepoint.

Anxiety symptoms were assessed using the 7-item Generalized Anxiety Disorder scale (GAD-7), which measures the frequency of anxiety symptoms experienced over the past two weeks. Participants rated how often they were bothered by symptoms such as feeling nervous or on edge, difficulty controlling worry, excessive worry, trouble relaxing, restlessness, irritability, and fear that something awful might happen.

Each item was scored from 0 ("Not at all") to 3 ("Nearly every day") and summed to create a total score ranging from 0 to 21, with higher scores indicating greater anxiety symptom severity. Standard cut points were used to characterize symptom severity (0-4 = minimal; 5-9 = mild; 10-14 = moderate; 15-21 = severe).

Data analysis forthcoming. Analyses will include participants with complete responses to all GAD-7 items at each timepoint.

Participants were presented with a list of potential sources of support (e.g., friends, family members, adults, professionals) and were asked to indicate which individuals they sought help or advice from for a personal or emotional problem during the past 6 months.

For each selected person, participants reported the number of times they met with that individual for advice or help during the past 6 months, using whole numbers.

Analysis forthcoming.

Participants reported on their problem-solving and decision-making behaviors using nine self-report items. Examples include: "I think about difficult situations that I might find myself in and make plans to deal with them." and "When there is a problem, I try to find out what caused it."

Response options for each item were: "Never," "Seldom," "About Half the Time," "Often," and "Always," with higher responses indicating more frequent use of decision-making strategies.

Analysis forthcoming.

Participants completed a ten-item self-report measure assessing attitudes toward seeking help for personal or emotional problems. For example, "I would feel inadequate if I went to a therapist for psychological help." Respondents were asked to use a 5-point Likert scale ranging from 'Strongly Disagree' to 'Strongly Agree' to rate the degree to which each statement describes how they would react in the situation.

Analysis forthcoming.

Participants reported their perceptions of substance use among peers at their school. The following questions were asked: "How many kids in your school would you guess have misused prescription opioids at least once?" and "How many kids in your school would you guess have used heroin at least once?"

Response options for each item were categorical: "All or Most," "Half," "Some," and "Hardly Any or None." Higher responses reflect perceptions that substance use is more common among peers.

Analysis forthcoming.

Participants were asked to indicate how they have been feeling emotionally during the past week in response to ten statements which assess emotional regulation, emotional awareness, and perceived control over emotions and behavior.

Responses were rated on a 5-point Likert scale: (a) Not at all (b) Somewhat (c) Moderately (d) Very much (e) Completely.

Analysis forthcoming.

Participants were asked how likely they would be to seek help for a personal or emotional problem from a range of potential help sources, including informal supports (e.g., intimate partner, friends, family), formal supports (e.g., mental health professionals, doctor/GP, phone helpline), and other sources (e.g., religious leaders). Participants also indicated whether they would not seek help from anyone. Responses were provided on a 7-point Likert-type scale ranging from Extremely unlikely to Extremely likely.

Analysis forthcoming.

Participants completed 18 self-report true/false questions assessing their beliefs about seeking psychological services. Example item includes "If a good friend asked my advice about a serious problem, I would recommend that he/she see a psychologist."

Analysis forthcoming.

Self-efficacy for refusing opioids was assessed using a two-item scale measuring participant's confidence in refusing prescription opioids or heroin if they had access to it. Reponses ranged from 1 ("Not at all confident") to 5 ("Extremely confident").

Analysis forthcoming.

Participants reported how pain affected their daily functioning and emotions over the past 7 days, using four self-report items adapted from Cunningham et al. (2017) and Varni et al. (2010).

Response options for each item were: "Never," "Almost Never," "Sometimes," "Often," and "Almost Always," with higher responses indicating greater impact of pain on function or mood.

Analysis forthcoming.

Participants reported their perceptions of substance use among people who live with them. The following questions were asked: "What is the most often that one person who lives with you drinks alcohol?"; "What is the most often that one person who lives with you uses marijuana?"; and "What is the most often that one person who lives with you uses heroin?"

Response options for each item were: "Never," "Less than once a week," "1-3 days a week," and "4-7 days a week," with higher responses indicating more frequent substance use by household members.

Analysis forthcoming.