10 min read Treatment

You mention the sleep again, near the end of the appointment. Your GP pauses, then signs the renewal. Neither of you says the thing you are both thinking, which is that this was supposed to be temporary.

Zopiclone, usually. Sometimes lorazepam. The drug varies. The situation does not.

Zopiclone Is Approved in Canada for Seven to Ten Days

Health Canada's summary safety review of zopiclone states that Imovane and its generics are approved for short-term use by adults with insomnia, up to seven to ten consecutive days. That is the approved window. Not a few months, not a year.

The Canadian product monograph goes further. Long-term use should be avoided, including in older patients. Abrupt discontinuation should be avoided, and treatment should be ended by gradually tapering the dose under close monitoring, even when the course has been short.

Choosing Wisely Canada, working through the Canadian Geriatrics Society, puts the harm-benefit ratio in numbers. The number needed to treat with a sedative-hypnotic for improved sleep is 13. The number needed to harm is 6.

The Canadian Psychiatric Association makes the same recommendation, and the underlying evidence is a 2005 meta-analysis in the BMJ by Glass and colleagues, which found that in adults over sixty the adverse effects of sedative-hypnotics outweighed the modest improvement in sleep.

Meanwhile, a joint report from Choosing Wisely Canada and the Canadian Institute for Health Information found that roughly one in twelve older adults uses a benzodiazepine or Z-drug regularly. The guidance and the prescribing have been pointing in opposite directions for years.

The Head-to-Head Trials Diverge at Follow-Up

The comparison that shaped the field was published in JAMA in 1999 by Charles Morin's group at Université Laval. Older adults with chronic insomnia were randomly assigned to CBT-I, to temazepam, to both, or to placebo.

At the end of the eight-week treatment, the three active conditions all beat placebo, and CBT-I and medication looked broadly similar. If the trial had stopped there, the honest summary would have been that both work.

It did not stop there. Patients were followed to three months, twelve months and twenty-four months. Over that period the sleep improvements were better sustained in the groups that had received CBT-I, alone or in combination, than in the group treated with medication alone.

The 2024 Canadian Delphi consensus on chronic insomnia management, developed by sixteen sleep specialists across the country, reaches the same conclusion for Canadian practice: CBT-I is first-line, and pharmacotherapy is positioned as short-term or adjunctive. CADTH's review of reviews on insomnia interventions describes the same pattern in the evidence base. So does the American College of Physicians guideline, which recommends CBT-I as the initial treatment for all adults with chronic insomnia.

Adding a Pill to CBT-I Does Not Improve the Long-Term Result

Morin's group tested this directly in a second JAMA trial in 2009. One hundred and sixty adults with persistent insomnia received six weeks of either CBT-I alone or CBT-I combined with zolpidem, followed by a six-month extended phase.

Combined therapy produced a small extra gain in total sleep time during the acute phase. Over the longer run, the highest remission rates were in the group that had started on combined therapy and then came off the medication while continuing CBT-I.

Adding a drug to a behavioural programme did not make the behavioural programme work better. Taking it away, once the behavioural work was underway, is where the outcomes improved.

The Decision to Take the Pill Is Part of What Keeps You Awake

Most sleeping pills are prescribed PRN, from the Latin for as the need arises. That instruction sounds cautious. In practice it hands you a decision to make every night, at the exact hour when deciding things is least useful.

What counts as needing it? After one hour awake, or two? After three bad nights, or four? What if you already took half a pill last night? Jade Wu, a behavioural sleep medicine specialist at Duke, calls this the PRN paradox in her 2023 book Hello Sleep. The bargaining is itself a form of sleep effort, and sleep effort raises the arousal that is keeping you awake. The pill is being asked to solve a problem that the deliberation about the pill has just made worse.

Her test for psychological dependence is not the dose and not the frequency. It is whether you find yourself negotiating.

The second mechanism she describes is misattribution. Someone stops a hypnotic abruptly, sleeps badly for three nights, gives in on the fourth and sleeps heavily. The pill gets the credit. What did the work was three nights of accumulated sleep drive, which would have produced the same heavy sleep with or without the tablet. The person now has personal proof of something that is not true, and the dependence is stronger than it was before the attempt.

This is the same arousal system described in middle-of-the-night waking, arriving by a different route.

Prescribing Patterns Reflect Access, Not Evidence

None of this means your GP made a poor decision. They were responding to real distress with the tool that was available inside a fifteen-minute appointment.

The Canadian Delphi panel names the reasons pharmacotherapy remains the most widely used treatment despite the guidelines: limited access to CBT-I, financial constraints, and provider awareness. There are very few trained behavioural sleep medicine clinicians in this country, and referral pathways to them are thin. A prescription takes three minutes. CBT-I takes six to eight weekly sessions and a therapist who knows the protocol.

Where those constraints bite hardest is cost and waiting. Both are worth checking before assuming the door is closed, since many extended health plans in BC cover sessions with a Registered Clinical Counsellor. There is a breakdown of that in what MSP and extended benefits cover for CBT-I.

A Taper Works Better Alongside CBT-I Than After It

Before anything else: medication changes belong to your prescribing physician. Benzodiazepines in particular should never be stopped abruptly. Zopiclone tapers are usually more manageable, and still benefit from a written plan agreed with your GP.

What the evidence supports is a concurrent approach. Begin CBT-I, and begin a gradual predetermined taper alongside it, rather than waiting until one is finished to start the other. The behavioural work gives you something to stand on as the medication comes down.

The shape of the schedule matters as much as the pace. A fixed written plan, with the reduced nights chosen in advance, tends to work better than reducing by feel. The reason is the PRN paradox again. When the schedule is set, there is no negotiation at eleven at night about whether tonight qualifies. The decision was made weeks ago, in daylight, by someone who was not tired. The pill shrinks from a nightly referendum to an instruction you follow, and then stop following.

1. Steady the baseline. Move from irregular use to a consistent dose on a consistent schedule, with no rescue doses during the night.
2. Reduce on predetermined nights. Lower the dose on specific nights of the week, chosen in advance and written down.
3. Reduce across all nights. Bring the lower dose to every night once the partial weeks are settled.
4. Widen the gaps. Drop predetermined nights entirely until the medication is no longer part of the week.

The first week or two is often the hardest part of the whole programme. Sleep can worsen before it improves. That is rebound insomnia, it is documented, and it is temporary. The nervous system takes time to recalibrate after months or years of chemical suppression. Knowing that this is a phase rather than a verdict changes how people move through it.

Sleep restriction, which runs in parallel during this period, is doing the useful work of rebuilding sleep drive while the medication comes down. There is more on how that piece functions in the ranked comparison of insomnia treatments, and the practice runs a supported sleeping pill taper alongside standard CBT-I for people in exactly this position.

Medication Still Has a Place in Short-Term Insomnia

A hypnotic prescribed for a fortnight after a bereavement, a surgery, or an acute crisis is being used the way it was designed to be used. That is what the seven-to-ten-day approval window is for. Used that way, medication can stop an acute sleep disruption from consolidating into something chronic.

The problem is the drift. Two weeks becomes two months, tolerance builds, and stopping produces a rebound that reads as proof the insomnia was there all along. By that point the medication is managing a problem it is also helping to maintain.

Chronic insomnia has been running for three months or more, several nights a week. If that is where you are, the research points in one direction. Treating what drives the insomnia produces better outcomes than continuing to suppress it. If you are unsure which category you fall into, the sleep and insomnia assessment is a reasonable place to start, and what happens in a first session covers what the work looks like from the inside.

You do not need to have the taper figured out before you begin. Working it out is part of the treatment.