Before You Ask How Much: The Contextual Interview in Substance Use Encounters

This post contains Amazon affiliate links. If you purchase through them, I earn a small commission at no extra cost to you.

This is the third post in the Addiction and Recovery in Primary Care series. The first post, Most People Get Better Without Us, named the models clinicians inherit about addiction and set them against the natural recovery literature: most people resolve a substance use disorder without ever entering treatment. The second, What the Substance Is Doing, turned that evidence into a clinical stance, one that asks what the substance is doing for the patient before it asks how much of it there is. This post gives that stance a technique.

I remember a patient—unhoused, visibly intoxicated, and in considerable pain—who came into our clinic one afternoon carrying everything he owned. He was dirty, disheveled, and he smelled. He was an established patient, though it had been a long time since he had engaged with primary care.

It was not his first attempt to find help that day. That morning he had gone to the emergency department with a badly infected foot, swollen and red well past the point where primary care could manage it. He was discharged and told, more or less directly, that the infections would keep coming until he stopped using intravenous heroin. The message was clinical in its delivery but moral in its logic: the substance use was the problem, and there was nothing worth treating until it stopped.

He had tried. From the ED he walked to the community mental health center, where a walk-in intake for the chemical dependency program was theoretically available. There were more people in line than there were intake slots that day, and court-ordered patients took priority. He left without being seen. He was in excruciating pain. So he injected heroin—which is, among other things, a powerful opioid analgesic—because it was the only pain relief he could actually obtain, and made his way to us, where he was worked into a late cancellation slot.

Every clinician who touched this man that day—and I include myself and my team—received him through the same frame. The warm handoff to me was about his heroin use, not his foot, not his housing, not the fact that he had spent his entire day trying to get help. This was before office-based opioid treatment had come to our area, though to be frank: even if it had been available, he might not have wanted it. What he needed acutely was treatment for an infected foot. Those are not the same clinical problem, and making one contingent on the other is not medicine—it is a moral judgment wearing a clinical coat.

I want to be careful not to assign blame where there was mostly a broken system and clinicians doing their best within it. The moral model of addiction—use is a choice, consequences are deserved, treatment is a reward for the decision to stop—is not a personal failing. It is a systemic frame, and it shapes behavior even in people who would reject it if you named it directly.

What it produced was a man who had spent not just that day, but many days before it, being told in ways direct and indirect that his pain did not qualify. And yet he kept going—from the ED to the CMHC to our front desk, in pain, carrying everything he owned. In a different clinical frame, that persistence would have been recognized for what it was: evidence of readiness to change. He was not a man without motivation. He was a man without a system willing to meet it.

In Harper Lee‘s To Kill a Mockingbird , Atticus Finch tells Scout that you never really understand a person until you climb into their skin and walk around in it. He is not offering a platitude. He is describing the specific failure of the Maycomb courtroom, where the community had already decided what the evidence meant before anyone examined it—where the diagnostic question had been answered by consensus long before the trial began. Atticus’s method was different: ask careful questions, listen without a predetermined frame, and let context do the work that accusation cannot.

That is also the BHC’s method. The first question is not “how much are you using?”—but the functional question that orients the entire encounter, what is the substance doing?, is one the BHC holds privately until the patient’s context makes it answerable.

The Wrong Question

The infrastructure of substance use screening in primary care is built around a different question entirely. SBIRT—Screening, Brief Intervention, and Referral to Treatment—is the population health framework that most primary care clinics use to identify patients whose substance use warrants clinical attention. The evidence base is substantial. Kaner et al. , in a Cochrane systematic review of 69 trials, found moderate-quality evidence that brief interventions reduce alcohol consumption by approximately 20 grams per week at one year follow-up. The US Preventive Services Task Force assigned screening and brief behavioral counseling for unhealthy alcohol use a Grade B recommendation, finding reductions in both the odds of exceeding recommended drinking limits and heavy use episodes at 6- to 12-month follow-up. The evidence is strongest for risky and hazardous drinking—which is also the range of presentations most commonly seen in primary care.

What a screening score cannot do is tell you anything about context. The AUDIT-C flags frequency and quantity. The DAST-10 flags consequences. The CRAFFT flags high-risk behavior in adolescents. These are useful signals—they tell the BHC that a conversation is warranted and, in some cases, how urgently. They do not explain why the patient drinks, what the drinking is managing, what has already been tried, or what giving it up would actually cost. A screening score is a referral reason, not a clinical picture.

This matters because primary care substance use presentations are distributed across the full range of severity and complexity. Many patients whose use is genuinely problematic will not score high enough on a brief screener to trigger a handoff—their use is episodic, their tolerance has normalized their self-report, or the screener does not ask about the substance they are using in the quantities they are using it. Others will score high and present with straightforward, uncomplicated presentations where a brief intervention is genuinely sufficient. The screening tool cannot tell the difference, and neither can the score.

The BHC who walks into that room carrying only a screening result is asking Maycomb’s question. The Contextual Interview asks something else: who is this person, and what is their life actually like?

The Contextual Interview

The Contextual Interview is a semi-structured clinical interviewing framework developed within the Focused Acceptance and Commitment Therapy (FACT) tradition and elaborated most recently for primary care settings by Cahill et al. . Its organizing logic is straightforward: before a clinician can understand a patient’s problem, they need to understand the patient’s life. The CI builds that understanding through four domains—love, work, play, and health behaviors—followed by a structured problem context inquiry organized around time, trigger, trajectory, and workability. Life context comes first. The problem comes last. For readers new to the model, a fuller introduction appears in the Trauma in Primary Care series; what follows here focuses on how the CI functions specifically in substance use encounters.

When substance use appears in the CI, it appears where health behaviors appear—after sleep, exercise, and diet have already established a conversational tone in which the clinician is curious about the patient’s daily life rather than building a case against it. From there the sequence moves deliberately: caffeine, tobacco, alcohol, cannabis, other drugs. Least stigmatized to most. A patient who might shut down entirely if the first question out of a clinician’s mouth were “are you still using heroin?” will often answer the same question candidly after twenty minutes of conversation about their living situation, their kids, what they do to decompress, and whether they have been sleeping—and after the clinician has already asked, without any particular weight, whether they drink coffee and how much.

The quantity clarification follows the same logic. The BHC asks in the patient’s language and converts internally for documentation. “A couple of drinks” is a starting point, not an answer—”so like two, or like a hundred?”—with the upper bound set high enough that there is no reason to minimize. The same matter-of-factness governs the safer use questions for patients who inject: do you inject? Have you shared or reused a needle, even once? Asked in the same clinical register as the sleep schedule question, these questions communicate that the BHC is interested in the patient’s actual practices, not their adherence to an expected narrative.

What happens after the disclosure matters as much as the disclosure itself. The BHC gets the information, acknowledges it briefly—”okay, so three drinks a day, two or three days a week, thank you”—and moves on. We’ll come back around to that, but let me just review your health history for a moment. That sentence signals that the substance use is one piece of information among many, heard without alarm, to be returned to when the time is right. The patient who has just disclosed something they expected to be judged for watches the clinician write it down and keep going, and something in the room shifts.

The problem context section—the Three T’s and the workability question—is where the CI explicitly surfaces what the substance is doing. Time, trigger, and trajectory map the pattern. The workability question—what have you tried, and how has that worked?—opens the door to ambivalence without demanding it, and reframes the patient’s history of attempts not as an inventory of failures but as a map of what they already know about their own recovery. By the time the BHC arrives here, they are not asking a stranger to account for their use. They are asking someone whose context they now understand to help them make sense of something that already makes a certain kind of sense. It is, as Robinson and Reiter describe it, a conversation about function rather than a deposition about behavior.

The Same Tool, A Different Revelation

Readers of this blog will note the similarities between this post and my thoughts in the Trauma in Primary Care series—it seems I keep coming back to the Contextual Interview. There are worse habits. A tool that works tends to show up wherever the work is being done.

What changes between a trauma encounter and a substance use encounter is not the instrument—it is what the instrument reveals. And in practice, it often reveals both at once, because trauma and substance use co-occur at rates that make their clean separation something of a clinical fiction.

The BHC conducting a CI for a patient referred for alcohol use is not simply listening for substance use cues. They are holding the whole person in view—tracking relational safety, adverse history, affect regulation, and the functional role the substance may be playing in a life shaped by experiences that were never adequately processed. The CI does not ask the BHC to choose between frames. It asks them to hold both, letting the patient’s context determine which becomes the organizing concern for this encounter.

What each domain surfaces looks different depending on which concern is primary. In substance use encounters specifically: the love domain tells you whether the patient’s relationships are implicated in the use—a partner who also uses, a family system organized around shame, a social world where sobriety would mean isolation. The work domain tells you whether the substance is managing something specific about occupational life—the night-shift worker who drinks to sleep, the laborer whose chronic pain is undertreated, the person whose job disappeared and whose use escalated in the same month. The play domain, which clinicians are most likely to underweight, often reveals the most: what the patient has stopped doing, what the substance has replaced, and what recovery would need to give back before it could compete.

This concurrent listening matters especially when the referral had nothing to do with substance use at all. A BHC handed off for diabetes management or a positive PHQ-9 may find—through the ordinary course of a well-conducted CI—that significant substance use or trauma history surfaces. What to do with that information is not reducible to a clinical algorithm. It is a shared decision, made with the patient, grounded in the BHC’s contextually informed conceptualization of what is most pressing and the patient’s own account of what they are willing and ready to address. The clinician may set aside what has surfaced and return to it at follow-up, flag it for the PCP, or gently reorient the encounter—always in conversation with the patient rather than over their head.

How to hold concurrent presentations, weigh competing clinical priorities, and build a conceptualization that is genuinely contextual rather than diagnosis-first is a clinical skill that deserves fuller treatment than a single post can provide. A future post in this series will take up contextual conceptualization directly: what it looks like to build a working clinical picture across co-occurring concerns, how to communicate that picture to the team, and how shared goal-setting with the patient shapes what actually gets addressed in the time available.

When the Functional Question Surfaces

The Three T’s and the workability question are where the CI shifts from life context to problem context, and where the functional question the BHC has been holding finally becomes speakable.

Time orients the inquiry chronologically. When did the use become a concern—to the patient, to the people around them, or to the clinical record? The answer is rarely a clean origin story—more often a pattern of escalation that maps onto something else: a loss, a transition, a point at which the substance stopped working the way it once did. The BHC who has already heard the love, work, and play domains will often recognize the chronology before the patient names it. That recognition, offered carefully, is itself a clinical intervention.

Trigger asks what sets the pattern in motion. A patient may identify a specific cue—a person, a place, a time of day—but the trigger is usually a portal to something deeper: an emotion the patient does not have another way to manage, a physiological state the substance reliably resolves. The BHC is listening for the shape of the problem—what it responds to, what it is organized around—so that the workability question can land in the right place.

Trajectory asks how the pattern has changed over time, and does more clinical work in substance use encounters than in almost any other presenting concern because it surfaces natural recovery directly. Most people whose use has been genuinely problematic have also had periods when it was less so—those periods are data. They tell the BHC what conditions support change for this particular patient, what resources have worked before, and what was different about the times when things went better. Trajectory, asked without judgment, is an implicit argument against hopelessness.

Consider what a well-conducted CI would have produced with the patient from our opening. His heroin use had a history and a logic. It had begun years earlier with pain pills—self-medication for depression and the accumulated shame of a bad breakup and not finishing school. When those became harder to obtain, he moved to smoking heroin. By the time he walked into our clinic, he was also managing withdrawal, because the substance that had started as a solution had become a physiological necessity. And underneath all of it: the humiliation of homelessness, the guilt of being, in his own accounting and his family’s, a failure. The heroin was doing a great deal of work. It was managing pain—acutely, the infection in his foot; chronically, everything else. None of that makes it a good long-term solution. But it makes it an understandable one, and understanding it is the precondition for offering anything better.

This is the moment the functional question surfaces. Not as an interrogation—what are you getting out of this?—but as a collaborative observation, grounded in everything the CI has already produced: it sounds like the substance has been doing something specific for you, and I want to make sure I understand what that is before we talk about where to go from here. That sentence is what the entire interview has been building toward. It is Atticus’s quiet question after the cross-examination has already established the frame—the one that lets the witness tell the truth because the truth has finally become safe to tell.

Workflow, Shared Goals, and the Limits of a Single Encounter

SBIRT works, and the CI is what happens after the handoff.

A patient who screens positive for risky alcohol use, a patient who screens positive for opioid use disorder, and a patient whose PCP flagged cannabis use that did not technically meet screening threshold are not the same clinical picture—and the score alone cannot distinguish between them. The CI can, not because it produces a cleaner diagnosis, but because it produces a person whose relationships, work life, daily pleasures, health behaviors, and problem history are now legible enough to ask the question that actually matters (4 Cs: Comprehensive—the CI’s four domains make the whole person, not just the substance, legible to the encounter .).

The CI also catches patients who do not get caught by screening at all. Someone who has already told the BHC about their work stress, their sleep, their isolation, and the things they have stopped enjoying is far more likely to mention, almost in passing, that they have been drinking more than usual—and to mean it when they say so. What to do with that disclosure returns us to the shared decision-making framework described above: the BHC’s contextual conceptualization and the patient’s goals together determine what gets addressed today and what gets carried forward.

A score is a flag, and a flag is only as useful as the conversation it opens. The CI is that conversation—structured enough to be reliable, flexible enough to follow the patient, and sequenced in a way that makes honesty possible before it makes honesty necessary.

Atticus Finch won his argument and lost his case. Maycomb had decided, and no amount of careful questioning was going to change what the community already knew. Primary care is not a courtroom, and the BHC’s job is not to win an argument. It is to understand a person well enough to be genuinely useful to them—to climb into their skin, as Lee put it, and see the problem from inside the life that contains it. The CI is how that happens in twenty-five minutes, in a room down the hall from where the patient came to talk about something else entirely.

References