Motivational Interviewing in Health Coaching

Motivational Interviewing is the part of health coaching that deals with ambivalence. A coaching session normally runs on goals, education, action plans and accountability, and all of that works well with a client who has already decided. With a client who has not decided, the same moves produce agreement in the room and nothing at home. MI is the conversational style that gets past that, and using it in health coaching means noticing the moment ambivalence shows up and suspending the coaching repertoire until the client has argued for change in their own words.

Most guidance for coaches presents MI as a set of skills to add. The harder and more useful framing is that MI asks you to put several coaching habits down for a while.

Where coaching and MI part company

Simmons and Wolever compared integrative health coaching with MI in Global Advances in Health and Medicine (2013;2(4):28-35) and described them as synergistic but distinct. The distinctions they draw are structural. Coaching runs a long arc, several sessions at minimum, across whole-person health and multiple life domains, and stays with the client through maintenance. MI is targeted: it works on ambivalence about one specific behaviour, and its job is finished once commitment is secured.

Read that as a practitioner rather than a researcher and it says something concrete. MI is a phase inside a coaching relationship, not a coat of paint over it. It has a beginning, which is the appearance of ambivalence, and an end, which is the client’s own commitment language. What happens after that is coaching again.

The second thing worth knowing about MI’s mechanism comes from Miller and Rose (American Psychologist, 2009;64(6):527-537), who separated it into a relational component and a technical one. The relational half is accurate empathy. The technical half is evoking and reinforcing the client’s own change talk. Coaches almost always import the relational half, because it is close to what coaching already teaches. The technical half is the one that gets left behind, and it is the half that makes MI different from a warm conversation.

The cue to switch

You do not need a formal assessment to spot this. The pattern is recognisable and it repeats:

The client agrees to the plan, then reports back apologetically, then agrees to a smaller version of the same plan.
Every suggestion you offer meets a reason it will not work, and the client is the one offering the reasons.
The client asks you what they should do, and has already tried what you say.
Enthusiasm at the end of the session, silence between sessions.

That is ambivalence, and no amount of better goal-setting resolves it, because the client is not short of a plan. They are holding two positions at once and your plan only speaks to one of them.

Two things look similar and are not. Low confidence is a client who wants the change and doubts they can manage it, which responds to scaling and past successes. Discord is strain in the working relationship, which responds to repairing it. Ambivalence is wanting two incompatible things at once, and it is the one of the three MI is built for.

What the switch sounds like

A client eight weeks into coaching for blood pressure and weight, who agreed to walk four times a week and has managed it twice in a month.

Coach: “So the walking. Two out of sixteen. What got in the way?”

Client: “Work, mostly. And honestly by the time I get in I just want to sit down.”

Coach: “Shall we make it shorter? Fifteen minutes might be more realistic.”

Client: “Yeah. Yeah, we could try that.” (This is the third time the goal has been cut. The client has agreed each time. The agreement is the problem, and a fourth revision will produce a fourth agreement.)

Coach: “Can I stop us there? I keep offering smaller versions and you keep saying yes, and I don’t think that’s landing. Tell me about the sitting down when you get in.” (Naming the pattern out loud, dropping the plan, and asking about the function of the behaviour rather than the barrier to the goal.)

Client: “It’s the only part of the day nobody wants anything from me. If I go out again that’s gone.”

Coach: “So the hour on the sofa isn’t laziness. It’s the only hour that’s yours.” (A complex reflection. It adds the meaning the client implied and did not say, and it takes the client’s side of the argument rather than the coach’s.)

Client: “That sounds pathetic when you say it out loud.”

Coach: “It sounds like someone with nothing left at seven o’clock.”

Client: “Right. Although I know where that’s heading. My dad had the same job and the same chair and he was on four tablets by sixty.”

Coach: “That’s not a future you want.” (The client produced the first argument for change in the conversation, unprompted. Reflect it. Do not add to it, and do not start planning yet.)

Nothing was agreed and no goal was revised. What changed is which of them is arguing for change, and the coach got there by stopping the thing coaching had trained them to do. The general form of the move is in responding to sustain talk, and the reflection craft behind it is in reflective listening.

Three coaching habits that have to go on hold

Educating in a collaborative voice. Health coaches carry real content knowledge and clients often ask for it. Supplying it during ambivalence is the righting reflex wearing a friendly tone, and it produces the same result as bluntly telling someone what to do: they voice the other side. Elicit-Provide-Elicit exists for exactly this, and the middle step needs permission before it runs.

Action planning before commitment. A plan built while the client is still ambivalent is your plan with their signature on it. In MI terms, planning belongs after evoking, and the whole sequence is set out in the four processes.

Accountability as pressure. Check-ins, tracking and reporting back are useful once someone has decided. Applied to an undecided client they add a reason to avoid the session, and they teach the client to manage your expectations rather than their own behaviour.

Why “well done” does not count

This is where coaching culture and MI coding disagree most sharply, and it is measurable.

The MITI 4.2.1 manual, the field’s standard fidelity instrument, defines an affirmation as an utterance that accentuates something positive about the client and is explicitly tied to their strengths, efforts, intentions or worth. It then rules out what most coaches would call encouragement. Agreement, approval, cheerleading and non-specific praise are not coded as affirmations. In the manual’s own examples, “You did great!” and “Way to go!” earn nothing at all, and neither does telling a client you are proud of them, because none of them names anything specific about the person.

Two further rules catch habits that coaching actively rewards. Generic statements of support, the sympathetic “this part is always hard” variety, were dropped from the affirmation code in MITI 4 entirely. And there is a three strikes rule: a coach who repeats the same affirmation through a session gets credit for the first two or three, then stops being counted, on the reasoning that repetition drains the credibility out of it.

The practical version is short. An affirmation has to contain a fact about this client that you could not have said to the previous one.

Encouragement	Affirmation
”Well done, that’s brilliant progress."	"You rearranged two shifts to get here. That took some doing."
"I’m so proud of you."	"You said no to it twice at a family meal, which is the hardest place to."
"You’re doing amazingly."	"You’ve kept the food diary for three weeks and you hate writing things down.”

Checking that you actually did MI

MI is easy to believe you are doing. A session can feel collaborative, warm and client-led and contain almost no evoking. Three numbers settle it, and each can be counted by hand from a recording the client has consented to.

MITI 4.2.1 measure	Fair	Good
Reflection-to-question ratio	1:1	2:1
Complex reflections, as a share of all reflections	40%	50%
MI non-adherent utterances	as few as possible	as few as possible

The reflection-to-question ratio is usually the one that exposes a coaching session, because coaching runs on questions and the ratio inverts. Only two behaviours count as MI non-adherent in MITI 4: persuading without permission, and confronting. Both are easy to slip into when a client is not moving.

One caveat, which the manual makes itself: these thresholds are expert opinion and have no normative validity data behind them yet. They are a direction of travel, not a pass mark. The rest of the picture, including the global ratings that behaviour counts miss, is in MI fidelity and MITI coding, and choosing an instrument for a whole cohort is covered in MI assessment tools.

Practising the switch

The switch is the skill, and workshops rehearse it badly, because their role-plays are usually built around a client who is willing. The version that matters is the client who agrees and does not move. It takes repetition to stop answering that with a smaller goal.

The MI Practice Lab gives you voice conversations with AI clients written to be ambivalent rather than cooperative, then returns the numbers above after each session: reflection and question counts, the simple and complex split, a talk ratio, and where change talk appeared and what you said next. It sits alongside supervision rather than replacing it, and it fills the gap between training days that most CPD or CEU schedules leave open.

Frequently Asked Questions

Is motivational interviewing the same as health coaching?

No. Simmons and Wolever (2013) described them as synergistic but distinct: health coaching runs a long arc across whole-person health and stays with the client through maintenance, while MI is a targeted method for resolving ambivalence about one behaviour and finishes once the client commits. MI is best understood as a phase a coach moves into when ambivalence appears, not as a style the whole relationship runs on.

Can a health coach use motivational interviewing without clinical training?

Yes. MI is a conversational style rather than a treatment, and it is used across nursing, dietetics, physiotherapy, social work and coaching. What it does require is feedback on your own sessions, because the parts that make MI work are hard to self-assess: practitioners consistently believe they reflect more and persuade less than a recording shows.

When should a coach stop coaching and switch to MI?

When the client keeps agreeing and keeps not doing it. Repeated agreement followed by no change is the signature of ambivalence, and revising the goal downwards a third time will produce a third agreement. That is the point to drop the plan and ask what the current behaviour is doing for the client.

Which MI skill should a health coach learn first?

Complex reflection. Coaching training already covers open questions and summarising well, so the reflection-to-question ratio is where most coaches fall furthest below MITI thresholds, and reflections are what carry the technical half of MI. A useful starting target is one reflection for every question, then working on the share of those reflections that add meaning rather than repeating it.

Does praising a client count as an affirmation in MI?

Usually not. MITI 4.2.1 excludes agreement, approval, cheerleading and non-specific praise from the affirmation code, so phrases like “well done” and “you’re doing great” score nothing. An affirmation has to name something specific about the client’s strengths, efforts, intentions or worth, and the manual also stops counting the same affirmation after it has been repeated two or three times.

Is motivational interviewing evidence-based for health coaching?

MI has a substantial evidence base across health behaviour change, and it is the ambivalence-focused component that carries it. The honest limit is that the evidence sits mainly on MI as a defined method delivered with measured fidelity, not on coaching sessions that borrow its vocabulary. If you want the results the research describes, the fidelity measurement is part of the intervention rather than an optional extra.

Want to practise the client who agrees and never moves? The MI Practice Lab puts you in voice conversations with ambivalent AI clients and shows you your reflection-to-question ratio, your complex reflection share, and the moment change talk appeared. Start a free trial: 5 minutes, no card required.