Why this matters
The conversation about safety, consent, and exploitation is the provider's job. Not just the parent's. Not just the school's. Yours. Adolescents who do not have a trusted adult to talk to about their bodies, their relationships, and their safety end up in the clinical record of someone like me decades later, when the harm has long since happened and what they want from me is not prevention but reconstruction. The girls in the Epstein files were seen by clinicians during the period of their exploitation. None of those clinicians appear to have asked them what was happening. The conversation was structurally available and was not had.
I want to say something clinicians often resist hearing: the reason the conversation does not happen is rarely that the provider does not care. It is usually that the provider does not have a script. Adolescent visits are short, the chief complaint is usually something else, the developmental work of the visit is invisible to the EMR, and the conversation about safety and exploitation feels like an unfunded mandate the system has put on top of an already overstretched encounter. That is real. It is also why I am writing this guide as a set of scripts. The structural problem of fifteen-minute visits is not going to be solved by this document; the problem of "I do not know what words to use" can be.
Developmental framing: what to assume at what age
The HEADSS conversation adapts to developmental stage; it does not change in structure. Here is the working frame I use for what an adolescent at each age can engage with substantively, drawn from the AAP Bright Futures adolescent framework and from a couple of decades of clinical encounters.
- Ages 11–13 (early adolescence). Concrete-operational thinking is dominant. Capacity to discuss bodies and safety is present but limited by abstract reasoning constraints — they can answer "has anyone touched you in a way that made you uncomfortable" much more easily than "do you feel your relationships are equitable." Use concrete, behaviorally specific language. Do not assume they have sexual experience; do assume some of them do and have not told anyone. Privacy from parents is critical even at this age — the disclosure rate goes up materially when the adolescent is seen alone.
- Ages 14–16 (middle adolescence). Abstract reasoning is emerging. Romantic and sexual relationships are increasingly part of life, including for adolescents whose parents do not know. Power dynamics in relationships are accessible to discussion but often not yet integrated into self-understanding. This is the age group most likely to be in an exploitative relationship and not recognize it as exploitation; the "older boyfriend" frame is most active here. Use developmentally explicit language about power and age gaps.
- Ages 17–18 (late adolescence). Approaching adult cognitive capacity but still inside the legal age structure for minors in most contexts. More likely to recognize exploitation retrospectively and to disclose historical events; less likely to disclose current ones. Use the late-adolescence visit to revisit any prior screening that was negative — the answer changes.
The HEADSS framework — what to actually ask
The HEADSS framework (Home, Education, Activities, Drugs, Sex, Safety) was originally articulated by Goldenring and Cohen in 1988 and has been refined since (the HEEADSSS variant adds Eating and Suicidality). The structure is durable because it walks from low-stakes to high-stakes — building rapport before asking the hardest questions. The order matters.
| Letter | Domain | What to ask (open-ended) | What you are listening for |
|---|---|---|---|
| H | Home | "Who lives with you? Tell me about the people at home. How do you get along?" | Who has access. Family instability. Foster placement. Adults the patient cannot place. |
| E | Education / Employment | "How is school? What are you good at? What is hard? Are you working?" | Attendance drops. Sudden disengagement. Unexplained income from "work." |
| A | Activities | "What do you do for fun? Who do you spend time with? Online too — what are you on?" | Older friends. Adults presenting as friends. Unsupervised travel. Apps you do not recognize. |
| (E) | Eating | "How do you feel about your body? Tell me what you eat in a typical day." | Disordered eating, often comorbid with CSA. Body-image distress. |
| D | Drugs | "Has anyone you know used alcohol, weed, pills, anything? What about you?" | Substance use, including as a control mechanism by a trafficker. |
| S | Sexuality | "Are you in any romantic or sexual relationships? Tell me about that person. How old are they?" | Age gap. Power asymmetry. Patient describing a relationship in language that does not match what she is reporting. |
| (S) | Suicidality | "Have you been thinking about hurting yourself or not wanting to be here anymore?" | Self-harm. Suicidal ideation, often elevated in exploitation contexts. |
| S | Safety | "Is there anyone in your life right now who scares you? Has anyone ever touched you in a way you did not want?" | The direct disclosure question. End on it. |
The opening: how to start the conversation
The opening of the HEADSS visit is the moment where the patient decides whether you are a person she can talk to. The wrong opening shuts the conversation down before it starts. The right opening makes it possible. Three scripts I have used across thousands of adolescent visits.
"Before we get into anything, I want to tell you how this works. Most of what you tell me stays between us — your parent does not get to read your chart, your school does not get a copy. There are a couple of exceptions: if I think you are being hurt by someone, if I think you are going to hurt yourself, if I think someone else is in danger. Those I have to do something about. Everything else stays between us. Is that okay with you?"
"We see all of our adolescent patients alone for part of the visit. It is the way we make sure you can ask me anything you want without having to think about who else is in the room. So I am going to ask your [parent / aunt / accompanying adult] to step out for the next bit, and then they can come back in when we talk about the plan."
"I am going to ask you some questions that I ask all of my patients your age. None of these are because I think something is wrong. I am asking because I want to know you. Some of them might feel awkward. You can answer them, you can skip them, you can answer them halfway. You can also ask me something instead."
The harder questions and how to ask them
The HEADSS table gives the domain-level prompts. Inside those prompts, the questions that actually surface exploitation tend to share a few features: they are behavioral, not categorical; they assume nothing; and they invite description rather than yes/no answers.
"You mentioned [name]. Tell me about him. How old is he? How did you meet? How long have you been together? What kinds of things do you do together?"
Notice: no judgment, no editorializing on the age. You are gathering information. The age difference and the meeting context will surface naturally if you give them room.
"Are you in a sexual relationship right now? Are you having sex by choice?"
Notice: "by choice" is the operative phrase. It surfaces coercion without using the word coercion. Patients in coercive relationships frequently hesitate or qualify on this question even when they have not consciously named the relationship as exploitative.
"Has anyone ever touched you in a way that made you uncomfortable? Has anyone ever asked you to do something with your body that you did not want to do? Has anyone ever traded something — money, a place to stay, anything — for you doing something with your body?"
The third sentence is the trafficking-specific question. Ask it explicitly. Many adolescents in transactional sexual relationships do not recognize them as trafficking because no one has used the word "trafficking" with them; the third sentence asks the underlying behavior.
When the patient is in an exploitative relationship and does not see it that way
This is the hardest version of the encounter. The patient describes a man significantly older than she is, the relationship has obvious power-asymmetry markers — he buys her things, he picks her up from school, he decides where she goes — and she is invested in framing it as love. She is fifteen. He is twenty-eight. She tells you he is "different" and "treats me really well" and "actually listens to me." You are looking at the textbook architecture of grooming and she is looking at the first relationship of her life.
The mistake providers make here is the cold confrontation. "That is not a relationship. That is exploitation. He is grooming you." That intervention closes the door. The patient will agree with you in the moment, leave the visit, and never come back, and you will have lost the only person who was asking her these questions. The work in this visit is different: it is to stay engaged, to plant a few specific questions in her mind, and to make sure she leaves knowing your door is open.
"I am glad you have someone in your life who you feel cared for by. I want to ask you a few things about him — not because I am trying to argue with you about whether you should be with him, but because I want to make sure you are safe. Is that okay?"
"How did you meet?"
"What was the first thing he gave you?"
"Does he know you are here today?"
"What does he say when you want to spend time with your friends instead of him?"
"Has he ever said anything that scared you, even a little?"
"If you wanted to end things, what do you think he would do?"
These questions get the patient describing the actual mechanics of the relationship in her own words. Many of them are the same questions she will eventually ask herself in retrospect when she leaves the relationship; you are planting them now.
"I want you to know that I am not going to tell you what to do about this. You are the person who knows your own life. I want you to know two things. First, if anything ever feels off with him — if he ever scares you, hurts you, or asks you to do something you do not want to do — I am a person you can come to. You can come to me even if you are still with him. I am not going anywhere. Second, I am going to put my number and the number of an organization I trust on a piece of paper. You do not have to use them. You can put them in your phone or throw them away. I just want you to have them."
The case that taught me how to do this conversation was a fifteen-year-old patient I will call M. M was being seen for an annual exam. Her sexual history was significant for a single partner, a twenty-six-year-old whom she described as her boyfriend and who she said had picked her up from school that day. He was in the waiting room. M did not see this as exploitation. She told me he had bought her a phone, she was the only one of her friends who had this phone, and he was the first person who had ever really listened to her.
I knew the trap I was walking into. If I told her this was grooming, I was going to lose her. If I did not say anything, I was going to also lose her, in a different way. What I did instead was ask her how she would know if she wanted to end the relationship. I asked her what he had said the last time she wanted to do something he did not want her to do. I asked her if she had ever felt scared, even for a second.
She did not disclose anything in that visit. She came back four months later. She told me she had thought about my questions a lot. She had broken up with him. She had not told her mother, because her mother did not know about him, and she did not want to. She wanted to know if she could be tested.
I tested her. We talked about birth control. We did the whole follow-up visit and at the end she asked me if she was stupid for having been with him as long as she was. I told her she was not stupid. She was a fifteen-year-old who had been groomed by a man who knew what he was doing, and the fact that she had asked herself the question she asked herself was not stupidity, it was clarity. I told her that the questions I had asked her in the prior visit were the questions I ask everyone, and the answer she had given herself had been hers, not mine.
That is what I tell residents about how to do this conversation. The leverage you have is curiosity, not confrontation. The patient is not going to leave the relationship because you told her she should. She is going to leave it because she started asking herself questions she had not asked before, and you put those questions in front of her.
What to document, what to follow up on
- The HEADSS domains you covered, in the chart, in her words. Not "HEADSS reviewed, no concerns." That phrase is what every chart in the Epstein record looked like. Document specifically: "patient endorses sexual relationship with [age] partner; denies coercion; states relationship is consensual; describes [specific concerning details patient stated]; declined further questioning at this time."
- What you offered and what she took. Resources given (paper printout from R1, your callback number, the trafficking hotline). Specific follow-up offered (return visit within 4–8 weeks). Whether she accepted.
- The threshold question for mandatory reporting. If you did not report, document why — what specifically made the threshold not met. If you did report, document the call and the report number. See PG3 for the full reporting framework.
For your own care
The hardest part of this work is the patients you ask the questions of, and ask well, and the answers do not surface anything actionable, and you do not know whether they were not surfacing because there was nothing to surface or because she was not ready to tell you. You will sometimes spend an hour with a patient, ask all the right questions, document carefully, and walk out of the room not knowing whether you missed something or whether nothing was there to find. Both of those experiences feel the same in your body. That ambiguity is the work.
What I tell residents is that the conversation itself is the intervention. Even when nothing surfaces, you have done something that matters. You have communicated to an adolescent that an adult will ask her behaviorally specific questions about her safety, will not editorialize on her answers, will hold her confidence, and will be available when she is ready. The next person who tries to harm her now has to do it past the memory of you. That is not a small thing. That is a structural change in her experience of adults, and it is the work you can actually do in the room.
The way I make peace with the not-knowing is the same way I make peace with the not-knowing in every other part of clinical work. You do the work in front of you. You ask the questions you know to ask. You document what you saw. You make the follow-up. And then you let go of the part you cannot control, because the alternative is to take it home and grind on it at three in the morning, which does not help the patient and ends your capacity to do the next visit.
What helps is a peer who does the same work. The conversation about a difficult adolescent encounter belongs in the curbside between attendings, in the residency case conference, in the morning huddle, in the text to a friend who is also a clinician. It does not belong inside my own head at midnight. The reason I can do this work for as long as I have done it is that I do not do it alone. The patient cannot be held by a clinician who is not being held. The conversation cannot keep happening if the clinician doing it has burned out from the weight of every patient she did not know how to save.
Find your person. Use them. Do not do this alone.