We tend to talk about depression as one illness. But two people with the same diagnosis can have very different experiences, and respond very differently to treatment.
A large Australian study suggests one way to tell some of them apart: the body clock. Nearly a quarter of people with depression showed a “circadian” pattern of late, irregular or excessive sleep, and that group looked different in several ways.
The body clock times sleep, hormones and appetite
Your body runs on an internal circadian rhythm, a roughly 24-hour cycle that times sleep, hormones, body temperature and appetite. Daylight keeps it in step with the outside world.
When the clock drifts, for example when you go to bed very late or sleep at different times on weekdays and weekends, mood, energy and metabolism can be affected.
Researchers analysed 11 380 Australians with depression
Emiliana Tonini, Ian Hickie and colleagues at the University of Sydney’s Brain and Mind Centre, QIMR Berghofer and other institutions analysed data from the Australian Genetics of Depression Study, a large nationwide project. Their paper was published in the British Journal of Psychiatry on 20 October 2026.
They looked at 11 380 people with depression, most with severe or recurring illness. Each was classed as having a circadian profile if they had at least three of six features:
- Social jetlag: sleeping at very different times on workdays and free days
- Seasonality: mood that changes with the seasons
- Late sleep timing: a sleep midpoint well into the morning
- Evening chronotype: being a natural “night owl”
- Morning tiredness: feeling groggy and slow to wake
- Hypersomnia: sleeping much more than usual
They then compared the two groups on symptoms, self-reported medication response and polygenic scores, which add up many small genetic variants to estimate someone’s inherited tendency towards a trait.
Nearly 1 in 4 had a body-clock profile, with earlier and more severe illness
About 23% had the circadian profile: 2 604 people, roughly 1 in 4.
Compared with the rest, they:
- Became ill younger: on average at about 19 rather than 23
- Had more severe features, including more suicidal thoughts, more distress, mania-like and psychotic-like experiences, and weight gain during episodes
- Reported slightly weaker responses to the two most common types of antidepressants, SSRIs and SNRIs, and more side effects
That medication difference was modest. The odds of reporting a good response were about 11 to 12% lower in the circadian group. Two other antidepressants, mirtazapine and amitriptyline, showed no clear difference.
Their genes leaned toward ADHD, bipolar disorder and metabolic traits
The circadian group carried slightly higher genetic scores for a striking mix of traits:
- Mental health: ADHD, depression, anxiety and bipolar disorder
- Metabolism: body mass index, blood fats and insulin resistance
- Inflammation: interleukin-6, a signalling molecule linked to inflammation
- Sleep: later sleep timing, insomnia and eveningness
Each genetic difference was small, around 8 to 15% higher odds. But together they sketch a picture of depression linked to sleep, metabolism and inflammation rather than mood alone.
A body-clock subtype could point to different treatments
Psychiatry has long tried to split depression into subtypes that point to different causes and treatments. Many proposed subtypes describe symptoms but haven’t led to better care.
A body-clock subtype is appealing because it suggests a different target for treatment. If someone’s depression is tangled up with a disrupted rhythm, treatments that reset the clock, such as morning light, regular sleep and wake times, or carefully timed medication, might help.
Think of it like a car that keeps stalling. One mechanic replaces the battery; another notices the timing belt has slipped. If the timing is the real problem, a new battery won’t fix it.
This study didn’t test any of those treatments. But it suggests a group of people for whom they might be worth studying.
What the study doesn’t show
- It’s a snapshot. The data are cross-sectional, so they can’t show whether body-clock problems cause depression or the other way round.
- Treatment response was self-reported, looking back, not measured in a trial.
- The definition is provisional. The “three of six” cut-off was chosen by the researchers and hasn’t been validated.
- The sample was specific: mostly women (about 80%), of European ancestry, with severe or recurring depression.
- The effects are modest. The differences in medication response and genetic scores are real but small.
- It needs replicating. The authors say independent studies must confirm the subtype before it’s used in practice.
- Peer reviewed, following a preprint in February 2026. The senior author declares industry links, including a stake in a mental health technology company.
Sleep and rhythm problems are worth raising with a doctor
If your depression comes with late, irregular or excessive sleep, it’s reasonable to mention this to a doctor and ask whether treatments aimed at sleep and daily rhythms might help.
That’s an inference. Light therapy, melatonin and sleep-timing treatments weren’t tested in this study, and nobody should stop or change medication without medical advice.
Depression may be partly a matter of timing
Treatment for depression often involves trial and error. Finding groups of people who respond differently could help match them to the right treatment sooner.
For some people, depression may be as much about when they sleep as about how they feel.