At three in the morning the house has its own way of becoming bigger. The refrigerator makes noise, the phone stays on even when it would be better to turn it face down, an ordinary thought slips into the room and starts walking back and forth with its shoes on. A discussion, a fear, a poorly said phrase, a deadline, a loneliness that during the day sat composed in a corner: at that time it takes up space, turns up the volume, demands an audience.
Paolo Crepet often uses an image attributed to Napoleon: “courage is that of three in the morning”. The expression has a long and somewhat slippery history, because already Thoreau, in Waldenspoke of the “three-o’clock-in-the-morning courage” which Bonaparte considered very rare. In normal life, without horses, armies and statue-like poses, it means something much more everyday: the courage you need when the world sleeps, no one distracts you, no one absolves you, no one tells you “come on, tomorrow will pass”. You stay. And often you, at that time, are the least suitable person to judge your life.
Now a study published on BMJ Mental Health gives measurable form to this intuition. The researchers analyzed data from the UCL COVID-19 Social Study, following 49,218 adults in England between March 2020 and March 2022, with almost a million responses collected over time. The result does not promise miracles, fortunately. It says something more useful: on average, people report better mental health and well-being in the morning, while around midnight the numbers worsen.
Nearly a million responses show that mood changes with the time
The study’s question seems simple, one that we could ask ourselves without lab coats and graphs: does mental health change during the day? To answer, the research team looked at six different dimensions: depressive symptoms, anxiety symptoms, happiness, satisfaction with one’s life, feeling that what one does has value and loneliness. Sensible choice, because the mind rarely acts like a switch. You can feel anxious and still have a purpose. You can be alone and still function. You can be satisfied with your life and, at the same time, feel like your head is frayed like an old sweater.
For depression and anxiety, well-known clinical instruments, the PHQ-9 and the GAD-7, were used. Loneliness was measured with the three-question UCLA scale. Happiness, satisfaction and sense of value in life were measured with direct questions, adapted from British measures of personal well-being. Each response had a time. This seemingly bureaucratic detail made it possible to compare what people declared from six in the morning until midnight.
The resulting drawing has a rough clarity. In the morning, people reported fewer depressive symptoms, less anxiety, less loneliness, and higher levels of happiness, satisfaction, and sense of purpose. Towards midnight the picture became darker. No theatrical cliffs, no gothic novel transformations. The differences during the day were small, but constant. And a small thing, when it comes back almost a million times, stops being a detail.
Caution is needed. The study is observational, so it shows an association and does not prove that time alone improves or worsens mental health. The authors say it openly: it may be that those who feel better in the morning are also more inclined to answer the questionnaires in those hours, while those who feel worse end up filling them out later. There is also a lack of information on sleep cycles, latitude and weather, all factors capable of significantly shifting mood.
The strength of the work lies in the quantity of observations and in the fact that the pattern remains even after several statistical adjustments. The initial sample was unbalanced: before weighting, women, white people and people with a degree were over-represented. The data was then weighted to better reflect the English population by age, gender, ethnicity and education; the models also considered employment, urban or rural residence, diagnosed physical conditions, and existing mental health conditions. The time-related oscillation remained there.
The six measures, however, did not all move in the same way. Depression and anxiety appeared lighter in the early hours of the day and heavier in the late evening. Happiness and life satisfaction tended to decline as midnight approached. The sense that one’s life is worth something showed the most marked fluctuation: high in the early morning, lower towards the middle of the day, again a little more present in the evening, then falling towards the night. Loneliness changed less, almost sitting in the same spot, like certain presences who know the clock and don’t care.
This detail matters. Loneliness can behave more like a stable condition than a passing state. The individual well-being measures fluctuated more also for a technical reason: depression and anxiety were assessed with scales made up of multiple questions, while happiness, satisfaction and purpose with single questions from 0 to 10. A single question leaves more room for fluctuations. Even in research, the container changes the shape of the liquid.
Monday comes out better than expected
The day of the week enters into the matter with less malice than office mythology would have it. Monday, the eternal target of memes, yawns and subway faces, fared even better than Sunday on some measures in this study. Happiness, satisfaction, and feeling that life has value were slightly higher on Monday and Friday than on Sunday; happiness also increased on Tuesdays. Loneliness, however, varied little from one day to the next.
Depressive symptoms were higher on Wednesday and Thursday than on Sunday, while anxiety was higher on all days except Friday, again compared to Sunday. The most interesting data concerns the shape of the day: at the weekend the mood seemed to move more between morning, midday, evening and night. During weekdays, the routine flattens out. Work, school, travel, meals, schedules, tasks. Everything pushes the body into a track. The weekend, with more freedom and more emptiness, allows more nervous curves to emerge.
The seasons, however, make themselves felt more decisively. Compared to winter, spring, summer and autumn were associated with fewer depressive symptoms, less anxiety, less loneliness and more happiness, satisfaction and sense of purpose. Summer was the best season on all six measures. It seems almost obvious, if you think about how much research has already worked on seasonal depression and on that gray area made up of short light, cold, isolation, days that seem to end before they even begin.
Yet, the season changes the general level, without really distorting the curve of the day. The morning remains better. Midnight remains worse. This makes the data more interesting, because if everything depended only on available light, the morning-night pattern should change much more between summer and winter. The authors therefore suggest a more complex interweaving: light, temperature, weather, social habits, work calendar, holidays, cultural routines. The head lives in the body, but it also lives in other people’s timetables.
The most immediate biological explanation comes from the circadian rhythm, the internal clock that regulates sleep, wakefulness, body temperature, metabolism, attention, hormones and mood. Among the prime suspects is cortisol, involved in the stress response, energy and alertness: it tends to peak shortly after waking up and fall towards bedtime. Serotonin, dopamine, inflammation, exposure to light, fatigue and social life can also participate in that very concrete sensation whereby in the morning, at least sometimes, the mind has more margin.
The social part weighs as much as the biological one. The body ignores the concept of Wednesday, the calendar knows it very well. If the patterns change between weekdays and weekends, it means that obligations, free time, relationships and daily activities participate in building mood. A free night can become rest, or rumination. A busy morning can provide direction, or pressure. It depends on what you find inside.
Then there is the enormous shadow of the period observed. The data comes from the years of the pandemic: lockdown, fear of contagion, bereavement, isolation, remote working, closed schools, fierce discussions on public health, social relationships reduced to the bare bones. From 2020 to 2022, mental health and well-being gradually improved in the sample. The first pandemic year was the hardest; then more information arrived, vaccines, reopenings, tools for orientation. This makes the study both powerful and situated: it speaks of England in an exceptional historical phase. In other countries, climates, cultures and post-pandemic routines, the same curve may have different contours.
From here it is best to avoid the shortcut of “sleep on it and everything will go away”. Sometimes sleeping really helps. Sleep can lower emotional reactivity, restore order to attention, give the brain a minimum amount of fuel to face things with less panic. Morning light can help your internal clock get back on track. However, depression, anxiety and loneliness pass through the next day together with those who bear them. Debts remain debts. Diagnoses remain diagnoses. Losses, conflicts, fears, empty homes and broken relationships wait even after coffee.
The useful part of the study lies right here: it shows that mental distress also has a rhythm. A person who answers a questionnaire at midnight can tell a darker version of himself than he would tell at eight in the morning. For research, this means considering the time at which you collect the data. For support services, it may mean looking more carefully at the late hours, when the need risks becoming more intense and less visible. The authors indicate the possibility of organizing resources and availability taking into account fluctuations during the day, for example by strengthening access to support at night.
Crepet, when he talks about the courage of three in the morning, intercepts a scene that we all know without having to dignify it too much. True heroism, at that time, often consists in not making definitive decisions while the mind is working with the lights off. Wait. Drink a glass of water. Leave your phone on the bedside table. Remember that the night doesn’t invent everything, but it amplifies. It takes away proportions, shortens breathing, puts the finger where it already hurts. The morning, sometimes, returns at least a table on which to place the pieces. Not three in the morning. At three in the morning they hold everything in their hands and squeeze.
You might also be interested in:
