Six more quits per 100 patients, and that’s the confident number
Counselling plus nicotine patches in psychiatric wards helps 6 more per 100 quit — with low confidence and thin data.
By The Weekend · · 5 min read
Offer someone counselling and nicotine replacement while they're admitted to a psychiatric ward, keep supporting them after discharge, and about 6 more people out of 100 will still not be smoking six months later, compared with usual care. That's the headline finding from a Cochrane review of 10 trials and 2,262 patients. The reviewers rate their own confidence in it as low.
The gap this fills isn't medical — it's logistical
Anti-smoking medicines already work for people with mental illness. Varenicline has shown benefit in outpatient, community settings for years. What's been missing is evidence on what to do with the captive, smoke-free window of a hospital stay — because most psychiatric wards in most countries don't allow smoking on the premises, whether or not the patient wants to quit permanently.
That gap matters more than it sounds. People treated for mental illness die 15 to 20 years earlier than the general population, mostly from the ordinary killers — heart disease, lung disease, diabetes, cancer — that tobacco accelerates. They smoke at higher rates than the general population and quit less successfully, despite wanting to stop as often as anyone else. A locked, smoke-free ward is, in theory, the best window a health system gets: forced abstinence, clinical staff already present, weeks of contact time. The review asked whether anyone has used that window well, and with what.
Ten trials, five countries, one intervention that clears the bar
The reviewers searched for any smoking-cessation programme started inside a psychiatric ward and found 10 qualifying trials from Australia, the US, Taiwan, Israel and Iran. Most enrolled patients with a mix of diagnoses — mood disorders, anxiety, schizophrenia; three trials enrolled only people with schizophrenia-spectrum illness. They tested four different approaches: counselling plus nicotine patches or gum, continued after discharge; group behavioural support without medication; drug trials of bupropion against placebo and cytisine against nicotine replacement; and comparisons between different doses and forms of nicotine replacement itself.
Only one of those four cleared the evidence bar with anything close to confidence: counselling combined with nicotine replacement therapy, extended past the hospital doors. Five studies and 1,611 people fed that comparison. The result — 6 more quitters per 100 at six months, versus usual care — is real enough to report, but the review explicitly flags low confidence in it, the standard Cochrane language for "plausible, underpowered, and worth funding properly before anyone treats it as settled." Everything else — the behavioural groups, the drug comparisons, the nicotine-dose questions — didn't have enough data to conclude anything at all.
The number nobody should skip: 1 fewer death per 100, "very uncertain"
Buried in the same comparison is a mortality signal: 1 fewer death per 100 people in the counselling-plus-nicotine-replacement group. The review's own language for this is "very uncertain," because there were simply very few deaths across the trials to measure. That's not a rounding caveat — it's the difference between a real safety signal and a statistical accident produced by small numbers. Nobody funded these ten trials to answer a mortality question with the power it needs, and until someone does, the honest reading is: probably not harmful, quite possibly helpful, not proven either way.
Who pays for the six-month follow-up, and who doesn't
The intervention that worked required something none of the failed comparisons did: continuity after discharge. Counselling on the ward is comparatively cheap — staff time, patches, a script. Following a patient for months after they leave, when they've lost the structure of a locked ward, insurance coverage may lapse, and the smoke-free environment that enforced abstinence disappears, costs more and is harder to fund inside standard psychiatric budgets built around acute stays, not post-discharge case management.
That's the quiet finding hiding in the numbers: the version of the intervention that worked is also the version hospitals are least structured to pay for. A ward can hand out patches during a two-week admission without touching its budget model. Committing a case manager to call a discharged patient for months afterward is a different line item, and it's the line item the data says actually moves the needle.
The mechanism: a smoke-free ward is a forced trial, not a cure
Here is the part worth keeping. A psychiatric admission removes access to cigarettes by policy, not by choice. That produces a period of enforced withdrawal that looks, on paper, like the ideal moment to start a quit attempt — motivation is irrelevant because the behaviour is already interrupted. But forced abstinence during a stay and a durable quit six months later are two different physiological and behavioural events, separated by discharge, stress, symptom relapse and the return of every cue that made someone reach for a cigarette in the first place. Counselling plus nicotine replacement worked, in this review's read of the data, because it was the only strategy tested that explicitly bridged that gap — treating the ward stay as the start of a quit attempt rather than a temporary suspension of smoking that ends at the front door. The drug trials and behavioural-group trials that didn't clear the evidence bar mostly stopped at discharge, or close to it.
What would actually settle this
The review is a map of what hasn't been tested properly, more than a verdict on what works. Nobody has run a hospital-ward trial of varenicline, despite it being an established outpatient option — the ten trials tested bupropion and cytisine instead, with too few participants to draw conclusions. Nobody has compared nicotine replacement doses inside this specific population with enough power to say which dose matters. And the behavioural-group approach, tested against a waiting list rather than active care, was set up to answer a narrower question than the one clinicians actually need answered.
The falsifiable version of this story is straightforward: if a larger, better-powered trial of counselling-plus-nicotine-replacement-with-follow-up reproduces something close to a 6-per-100 quit advantage and narrows the mortality estimate away from "very uncertain," the case for funding post-discharge continuity becomes hard to argue against on cost-effectiveness grounds alone, given the 15-to-20-year mortality gap it's chipping at. If the effect shrinks or vanishes under a bigger sample, the honest conclusion is that the smoke-free ward window is real but the follow-up model tested so far isn't yet the right tool for using it. Either way, the answer needs trials that don't yet exist, in a population that has waited longer than most for anyone to run them.