10 November 2025
“Sleepe is that golden chaine that ties health and our bodies together. Who complains of want ? of woundes ? of cares ? of great mens oppressions, of captivity ? whilest he sleepeth ? Beggers in their beds take as much pleasure as Kings : can we therefore surfet on this delicate Ambrosia ? can we drink too much of that whereof to tast too little tumbles us into a church-yard, and to use it but indifferently, throwes us into Bedlam ?”
Thomas Dekker, The Guls Horn-Booke
Thomas Dekker penned this paean of praise for the physical and mental health benefits of sleep in 1609. I guess the field of sleep medicine isn’t that new, after all. Yet for far too many people in the modern world, that ‘golden chain’ is stretched to – and often well past – breaking point. According to a 2019 report on chronic insomnia,
“Around 60% of people report at least one sleep symptom occurring 3 or more times per week, and this is consistent across age groups.”
Chronic Insomnia Disorder in Australia: A Report to the Sleep Health Foundation
‘Sleep symptoms’ include difficulty in falling asleep, waking up multiple times during the night, and waking up too early and not being able to get back to sleep. Younger people are more likely to report difficulty in getting off to sleep, while waking up overnight or early in the morning is more common in older people.
With so few people regularly enjoying a good night’s sleep, it’s no wonder that so many resort to using medications to try to fast-track their nightly trip to the land of Nod. In a survey conducted in 2022, 23 per cent of Australian adults admitted that they had become dependent on sleeping pills in the past, or were currently dependent. Another 14 per cent reported that they had been, or were currently, dependent on supplemental melatonin to help them sleep.
More younger than older people had resorted to some kind of chemical solution to help them sleep: 44 per cent of respondents aged 18-30 years reported dependence on sleeping medication (26 per cent using sleeping pills, 18 per cent using melatonin), compared with 32 per cent of over-50s (21 per cent sleeping pills, 11 per cent melatonin).
The principal reasons that respondents gave for taking sleeping pills or melatonin were getting fast results (32 per cent) and finding that nothing else they tried had worked (31 per cent).
On that last point, the 2019 report cited above found that less than 9 per cent of insomniacs had used cognitive behavioural therapy for insomnia (CBTi), which is considered the gold standard treatment. Unfortunately, respondents were not asked whether they had been offered CBTi but declined to use it, or had never heard of it. My hunch, though, is that sadly, most people would prefer to pop a pill than to engage with a CBTi program which actually requires them to do some work.
But what are the risks of using common prescription sleep medications, and of melatonin – a supposedly natural alternative?
Hazard #1. Sleeping pills increase the risk of death, even when taken fairly irregularly.
There are several classes of drugs that are prescribed as ‘hypnotics’, or sleep-inducing agents, including
- Benzodiazepines such as temazepam and diazepam, sold as Restoril, Normison, Xanax, Librium and Valium;
- Non-benzodiazepine ‘Z-drugs’ such as zolpidem and zopiclone, sold as Ambien, Stilnox, Imovane and Zimovane;
- Barbiturates such as amobarbitol, phenobarbital, methohexital, butalbital and pentobarbital; and
- Sedative antihistamines such as triprolidine, diphenhydramine and promethazine, sold as Actifed, Benadryl, Phenergan and Tixylix.
Multiple studies have found that people who take drugs of any class to induce sleep have higher mortality than those who don’t. For example:
- In a study that used data from the electronic health records of over 34,000 people enrolled in a large integrated health system in the US, people who received a prescription for a sleeping pill were over three times more likely to die during the 2.5 year follow-up period compared to individuals matched for sex, age and smoking status who had not been prescribed sleep medication. Separate analyses for several common hypnotics (sleep-inducing drugs), including zolpidem, eszopiclone, zaleplon, temazepam, other benzodiazepines besides temazepam, barbiturates and sedative antihistamines, found that all drug classes were associated with increased mortality.
There was a dose-dependent relationship between sleep medication prescription and risk of death, with a 3.6 times higher risk of dying in those prescribed just 0.4-18 pills a year; 4.4 times higher risk in those taking 18-132 pills a year, and 5.3 times higher risk in those using more than 132 pills a year, compared to non-users.
The risk was even higher in obese participants; compared to obese nonusers, obese individuals who took 18 or fewer pills annually had an 8.1 times higher risk of death while those on 132 or more annually had a 9.3 times higher mortality rate. And while the excess risk of death was seen in every age group, people aged 18–55 years had a higher relative risk if they took sleeping pills, than more elderly people.
An obvious potential confound is that people in poor health are both more likely to experience insomnia and more likely to die than those in good health. In other words, it may be that there is an association, but not a causal relationship, between sleeping pill prescription and death. To tease out the nature of this relationship, the researchers separated out those in poor health – and the results still held. In their words,
“Control of selective prescription of hypnotics for patients in poor health did not explain the observed excess mortality.”
Hypnotics’ association with mortality or cancer: a matched cohort study
- A retrospective cohort study of over 500,000 Koreans over the age of 40 used data from the National Health Insurance Service (NHIS) which collects all claims data for medical treatment including disease diagnosis, prescriptions provided, and procedures performed. It found that people who took 30 or more doses per year of a sedative-hypnotic drug (benzodiazepine, zolpidem, antidepressant, or low-dose antipsychotic) had a 14 per cent higher mortality risk than nonusers, after adjusting for age, sex, medical and psychiatric comorbidities and socioeconomic status. Zolpidem was the drug most strongly associated with increased chances of death; users had a 59 per cent higher risk.
- A systematic review and meta-analysis of 25 studies enrolling a total of 2,350,093 patients, concluded that both hypnotics and benzodiazepine anxiolytics increased the risk of dying by 43 per cent.
Is it possible that insomnia itself might be a confound? That is, could difficulties in going to sleep or staying asleep, rather than the hypnotic drugs that people are taking in an attempt to get more sleep, be increasing the risk of premature death?
Actually, no. A meta-analysis of 17 studies, including a total of 36,938,981 individuals followed up for an average of 11.6 years, found no increased risk of mortality in people suffering from frequent and ongoing insomnia (defined as insomnia occurring on three or more nights per week, for at least one month).
And a Taiwanese cohort study found that in people who got 6-8 hours of sleep per night (the sleeping duration associated with the lowest mortality risk), people who used sleeping pills had a 55 per cent higher risk of dying than nonusers.
Possible causes of excess mortality among sleeping pill users include:
- ‘Hangover’ effects – that is, reduced day-time brain cell activity, which makes people sleepy, less alert, confused, and weak during the day, in turn increasing the risk of falls and accidents;
- Exacerbation of breathing pauses in sleep apnoea, which increases blood pressure and the risk of heart attacks, heart failure and stroke – especially in patients with pulmonary and cardiac diseases;
- An increased risk of suicide due to amnesia, confusion, hallucination, parasomnias, and exacerbation of depression (see point # 5 below);
- Increased risks of infection, pneumonia, and cancer associated with suppression of the immune system;
- Accidental overdose; and
- Cancer (see next point).
Hazard #2. Frequent use of sleeping pills is associated with an increased risk of cancer.
In the US electronic health records study mentioned above, people in the highest bracket of sleeping pill use had a 35 per cent higher risk of developing cancer. Once again, the researchers used statistical adjustment to factor out the effect of pre-existing disease, and found that “death and cancer hazards associated with hypnotic drugs were not attributable to pre-existing disease.”
And a 14-year follow-up study in Taiwan found that patients with insomnia who used sedative-hypnotics had a 49 per cent higher risk of being diagnosed with cancer, compared to insomniacs who did not use any such drugs. Among patients without insomnia who used sedative-hypnotics (remembering that some of these drugs are used for other purposes, such as relief of anxiety), there was a 68 per cent higher risk of cancer compared with patients without insomnia who did not use any drugs in these classes. And finally, people who used sedative-hypnotics had a higher risk of dying of cancer.
Finally, a meta-analysis of 28 studies involving a total of 340,614 hypnotics users and 1,828,057 non-users found that both benzodiazepines and Z-drugs were significantly associated with a 17 per cent increased risk of cancer. The risk jumped up to 26 per cent higher when only Z-drugs and sedative benzodiazepines were included, and anxiolytic benzos were excluded. And people in the highest dosage bracket had more than double the risk of cancer as non-users of hypnotics. The cancer types with the strongest association with hypnotic use were brain, esophagus, liver, lung, stomach, pancreatic, colon, renal and prostate cancer.
As for how hypnotic drugs elevate the risk of developing cancer, no one is quite sure. Some studies have reported evidence that hypnotics disrupt immune function, thereby interfering with immune surveillance and destruction of nascent tumours. Hypnotics work by upregulating gamma-aminobutyric acid (GABA) activity; unfortunately this may also disrupt GABA-regulated cell proliferation and differentiation of brain cells, leading to brain cancer.
Hazard #3. Sleeping pills are addictive, and the withdrawal symptoms include insomnia!
While sleeping pills almost always help you go to sleep the first couple of times you take them, tolerance (where the dose you were taking no longer works to get you off to sleep) develops quite rapidly. It only takes between three and fourteen days of continued use to become tolerant to a benzodiazepine sleeping tablet, while tolerance may take somewhat longer to develop to other types of medication. An astonishing 15.7 per cent of Australians over 65 years of age regularly take benzodiazepines, despite their limited window of efficacy, rapid development of tolerance and high potential for dependence.
After several weeks of use, sleeping pills work no better than a placebo – but when you stop taking them, you are likely to suffer rebound insomnia, a drug withdrawal-induced insomnia which is often worse than the insomnia that you started taking the drug for in the first place. Contrary to claims often made by drug companies and doctors, both benzodiazepines and Z-drugs can cause rebound insomnia.
Another common withdrawal symptom is anxiety. Since anxiety makes it hard to get to sleep, it is a cruel irony that the drug you take to ‘turn your mind off’ so you can sleep, makes it even harder to turn your mind off when you stop taking it! One of the primary reasons people become hooked on sleeping pills is because they experience such anxiety and poor sleep if they try to stop. But if they stayed off the drug for a few days, they might sleep just as well without it!1
Hazard #4. Using sleeping pills impairs daytime performance.
People often resort to sleeping pills because they are afraid that their inability to sleep at night will impair their concentration, memory, mood and other aspects of performance the following day. They figure it MUST be better to take the pill and at least get some sleep. But they’re wrong, on two counts. Firstly – and this may shock you –
“No documented study has demonstrated a clear relationship between amount of sleep actually obtained by insomniacs and daytime performance.”
Sedative-Hypnotics and Human Performance
And secondly, the overwhelming majority of controlled studies show that even when a person sleeps somewhat longer after taking a sleeping pill, their performance is either worse on the following day, or they function no better in their daily life than they would if they had just missed a few hours of sleep. Ability to process information, make mental calculations, remember important facts, and most worryingly, drive a car safely, are all impaired by sleeping pill use.
Daytime impairment is obviously more severe with the longer-acting drugs including diazepam (Valium) and chlordiazepoxide (Librium), which accumulate in the bloodstream over the course of 10-20 days, reaching much higher concentrations than after the initial dose. But even drugs with a short half-life, such as zolpidem (Ambien), may still be in the bloodstream by morning, at doses capable of impairing performance, if taken in the middle of the night. Furthermore, there is some evidence that even after these short-acting drugs are fully eliminated, they may still impair performance of daily tasks.
Hazard #5. Sleeping pills increase the risk of depression, which is a major cause of insomnia.
Since at least 80 per cent of depressed people experience insomnia, it might be reasonable to think that insomnia medications may be helpful in treating depression. However, an analysis of data of clinical trials on sedative hypnotic drugs found that they more than doubled the risk of developing depression compared to placebo pills. What this means is that sleeping pills are more likely to cause depression than to help it.
One of the unfortunate consequences of severe depression is an increased risk of suicide, and long-term users of sleeping pills have a markedly raised suicide risk.
Is melatonin the answer?
OK, so pharmaceutical hypnotics are not the way to get a good night’s sleep. What about melatonin? Melatonin was made available over the counter to over-55s in Australia in June 2021, because it was seen as a safer option than prescription drugs. But research presented at the American Heart Association’s 2025 Scientific Sessions uncovered some extremely worrying findings. Compared to people with insomnia who did not take melatonin, long-term melatonin supplementation in people with insomnia was associated with
- An 89 per cent higher risk of developing heart failure;
- A three-fold increase in heart failure-related hospitalisations, and
- A doubling of all-cause mortality over five years of follow-up.
