Skip to content
MyFreud

Sundowning: Why Evenings Are Different

Confusion and agitation that worsen in late afternoon are common in dementia, and follow a pattern. What that pattern is, and what actually helps.

4 min read

Flat vector illustration of an older woman with her hands clasped, looking thoughtfully to one side.

Key takeaways

  • Sundowning describes worsening confusion, agitation or distress in the late afternoon and evening in someone with dementia. It is a recognised pattern, not something the person is doing on purpose or that a family is imagining.
  • How often it happens varies enormously between studies, and the reason is that researchers have used different definitions of it, some focused on emotion, some on movement, some on sleep.
  • There is no established medication for it. The best-tested non-drug approach is light therapy, and it helps sleep more reliably than it helps agitation or mood.
  • The evening timing is not a coincidence. Circadian rhythm disruption is part of the underlying condition, and fading afternoon light plausibly interacts with a system that is already working less well.
  • What helps most is not a single fix but reducing what the evening asks of an already-tired brain: more light earlier in the day, a predictable routine, and fewer demands as the light fades.

Sundowning is the name for a pattern many families with a person living with dementia recognise instantly: confusion, agitation or distress that worsens as the afternoon turns to evening, often easing again overnight. It is real, it is common, and it is one of the least understood parts of the condition.

Our overview of dementia covers the wider condition; this article is about the evening pattern specifically, why it happens, and what the evidence says helps.

What the pattern actually looks like

Increased confusion, restlessness, anxiety, irritability or suspicion that appears or worsens in the late afternoon or evening, sometimes alongside increased wandering, and usually settling again later at night or by morning. Reviewing the condition, researchers described it as one of the most commonly reported behavioural symptoms of dementia despite being poorly understood mechanistically. [khachiyants-2011-sundowning]

That combination, common and poorly understood, is unusual and worth sitting with. It means families are often managing something real with very little solid guidance about why it is happening on any given evening.

Why the numbers disagree so much

Because researchers have not agreed on what counts as sundowning, and that disagreement produces wildly different rates depending on which study you read. Reported prevalence spans from roughly 2 percent to roughly two-thirds of people with dementia, and the gap is explained by definition rather than by disease severity: some studies count emotional symptoms, others count nighttime movement such as wandering, and others focus specifically on sleep disruption.

Why one number cannot answer this
0 25 50 75 100 Reported prevalence range across studies 2 Lowest reported 25 Midpoint estimates 66 Highest reported

Illustrative range of reported prevalence across the studies discussed in Khachiyants et al. (2011), drawn to show the spread rather than a single pooled estimate.

The practical takeaway is not a number to memorise. It is that “how common is this” is the wrong first question, and “is this what my family is seeing” matters more than matching a percentage.

Why evening, specifically

The leading explanation is circadian: the internal clock that governs sleep, alertness and mood across a 24-hour cycle is disrupted in many forms of dementia, and that disruption does not distribute itself evenly across the day. Late afternoon, when light is fading and the accumulated demands of the day have built up, is a plausible point for an already-strained system to show strain most visibly.

That is a mechanism worth naming rather than a settled fact. It does not rule out other contributors sitting alongside it: fatigue building through the day, low light making a familiar room harder to interpret correctly, or a tired caregiver’s own patience thinning at the same hour the person they are caring for needs more of it.

What actually has evidence behind it

Not a medication. There is no established drug treatment for sundowning specifically, and non-drug approaches are the first line. The best-tested of these is light therapy, and a systematic review and meta-analysis of randomised controlled trials found it had a significant effect on sleep, reducing nighttime awakenings, without a significant effect on agitation or depression. [fong-2023-light-therapy]

That distinction matters for expectations. Light therapy is worth trying, and it is a sleep intervention more than an agitation or mood intervention, so using it and then being disappointed that agitation has not shifted is judging it against a claim it never made.

What does the evening pattern actually look like?

This is for noticing a shape, not for reaching a conclusion. Think about the last two weeks.

0 of 5 ticked

What this does not establish

Neither citation here settles the mechanism. The circadian explanation is plausible and widely cited, not proven beyond dispute, and multiple factors likely combine in any one evening. The light therapy evidence is specifically about sleep outcomes, and generalising it to a claim that light therapy treats sundowning broadly goes further than the trials support.

When to seek help

Speak to a doctor if evening confusion or agitation is new, is worsening, or is putting the person or people around them at risk, since a change in this pattern can also signal an unrelated problem such as infection or pain that is easy to miss in someone who cannot describe it clearly. Ask specifically about non-drug approaches first, since none of the medications sometimes used for agitation in dementia are approved specifically for sundowning and each carries its own risks worth weighing with a clinician.

Seek help urgently if the person becomes a danger to themselves or others, or if a sudden, marked change in confusion appears rather than a gradual pattern, since sudden confusion has different, sometimes treatable, causes.

If you are having thoughts of harming yourself, treat that as urgent and contact your local emergency services or a crisis helpline.

How MyFreud can help

MyFreud is useful here for the caregiver’s own experience across a long stretch of exhausting evenings, which is easy to lose track of from inside it. Logging your own mood and sleep, day by day, makes visible whether you are approaching the point where you need more support yourself, which is a question worth answering honestly rather than after the fact.

Download MyFreud and start today: App Store or Google Play.

Frequently asked questions

What is sundowning?

Sundowning is the name given to a worsening of confusion, agitation, anxiety or restlessness that occurs in the late afternoon or evening in someone with dementia, often followed by a calmer stretch overnight or the next morning. It is not a diagnosis in its own right, and it is best understood as a pattern that shows up within the wider condition rather than a separate illness. Reviewing the research, one summary described it as one of the least understood behavioural symptoms of dementia despite being one of the most commonly reported by families.

How common is sundowning?

That depends heavily on how it is defined, which is itself part of the problem with this area of research. Reported rates across studies span an extremely wide range, from as low as around 2 percent to as high as roughly two-thirds of people with dementia, and that spread reflects real differences in what researchers counted rather than a single settled number. Some definitions focus on emotional symptoms, others on increased nighttime movement such as wandering, and others on disrupted sleep, and a study using one definition will not find what a study using another one finds.

Why does it happen in the evening specifically?

The leading explanation involves the circadian system, the internal clock that governs sleep, wakefulness and mood across a 24-hour day, which is disrupted in many forms of dementia. That disruption does not announce itself uniformly across the day; late afternoon and evening, when light is fading and the demands of the day have accumulated, appear to be when an already-strained system is most likely to show it. This is a plausible mechanism rather than a fully proven one, and it does not rule out other contributors such as fatigue, low lighting making the environment harder to interpret, or a caregiver's own tiredness changing the interaction.

Does light therapy help?

It helps some things more than others, which is worth knowing before trying it. A systematic review and meta-analysis of randomised controlled trials found light therapy had a significant effect on reducing nighttime awakenings, improving sleep, but did not find a significant effect on agitation or depression specifically. That is a meaningfully different result from claiming light therapy fixes sundowning broadly, and it is why the honest answer is that it is worth trying for sleep in particular, with more modest expectations for mood and agitation.

What actually helps day to day?

Reducing what the evening asks of a brain that is already working harder by then. Getting bright light exposure earlier in the day, keeping a predictable and unhurried routine as evening approaches, minimising unnecessary decisions and stimulation in the late afternoon, and not scheduling demanding tasks such as appointments or visits for that window all follow from the same idea: the problem tends to be capacity running low, not defiance. There is no established medication for sundowning itself, so these environmental and routine changes are the first line rather than a fallback.

References

  1. 1.Khachiyants N, Trinkle D, Son SJ, Kim KY ( 2011). Sundown syndrome in persons with dementia: an update. Psychiatry Investigation. doi:10.4306/pi.2011.8.4.275
  2. 2.Fong KNK, Ge X, Ting KH, Wei M, Cheung H ( 2023). The effects of light therapy on sleep, agitation and depression in people with dementia: a systematic review and meta-analysis of randomized controlled trials. American Journal of Alzheimer's Disease and Other Dementias. doi:10.1177/15333175231160682