Cluster headaches affect roughly 1 in 1,000 adults, making them far less common than migraines or tension-type headaches. The pooled lifetime prevalence sits at about 0.12%, though individual studies have reported rates anywhere from 3 to 150 per 100,000 people depending on how the population was surveyed.
Prevalence by the Numbers
A meta-analysis of 16 population-based studies found that one-year prevalence ranged widely, from 3 to 150 per 100,000 people. The variation largely comes down to methodology. In one well-known Norwegian study, a neurologist personally interviewed and examined every participant, catching cases that surveys alone would miss. That study found a prevalence of 381 per 100,000, more than double the next highest estimate. Population surveys in the Republic of Georgia and rural Ethiopia produced figures of 87 per 100,000 and 1.3% respectively, reinforcing that detection depends heavily on how hard you look.
To put this in perspective, migraine affects 12 to 15% of the general population, with lifetime prevalence reaching 25% in women. Cluster headache prevalence, at 0.5 to 1 per 1,000, is roughly 100 to 150 times lower. If migraine is a common cold of headache disorders, cluster headache is closer to an uncommon chronic illness that most primary care doctors see only a handful of times in their careers.
Who Gets Cluster Headaches
Cluster headaches have long been considered a predominantly male condition, with a frequently cited ratio of about 4 to 1, men to women. The WHO puts the ratio even higher at 6 to 1. But those numbers are shifting, not because the actual prevalence is changing, but because more women are now being correctly diagnosed. For decades, women with cluster headaches were commonly told they had migraines, which skewed the epidemiological data.
The average age of onset is around 30, with a peak in the 20s for both sexes. There’s a secondary, smaller peak in the 40s and 50s. Onset before age 15 is uncommon but not unheard of, accounting for about 7% of cases in one large clinical series of over 800 patients. Late onset, after age 50, occurred in about 11% of cases and was somewhat more common in women (nearly 17%) than in men (9%).
Most Cases Are Episodic, Not Chronic
About 79% of people with cluster headaches have the episodic form. This means attacks come in “clusters” lasting weeks to months, separated by remission periods that can stretch for months or even years. The remaining 21% have chronic cluster headache, where remission periods either don’t occur or last less than three months. The chronic form can develop from the episodic type over time, or it can be chronic from the start.
During an active cluster period, attacks happen frequently, anywhere from once every other day to eight times in a single day. Each individual attack lasts between 15 minutes and 3 hours, with severe or very severe pain on one side of the head, typically around or behind the eye. The pain is accompanied by visible signs on the same side: a red or watering eye, a drooping eyelid, nasal congestion, or facial sweating. Most people feel intensely restless or agitated during an attack, pacing or rocking rather than lying still (which is a key distinction from migraine).
Why It Takes So Long to Get Diagnosed
Despite having one of the most recognizable patterns among headache disorders, cluster headache takes an average of five years to diagnose. That delay often involves visits to multiple clinicians before someone connects the dots. The core problem is rarity: most doctors have limited firsthand experience with the condition, and its symptoms overlap with more common diagnoses.
Migraine is the single most common misdiagnosis. Both conditions cause severe head pain and can involve tearing or nasal congestion. But the pattern is different: cluster attacks are shorter, more frequent during active periods, and one-sided with clockwork regularity. They also tend to occur at the same time of day, often waking people from sleep. Other common misdiagnoses include sinusitis, dental problems, tension-type headache, eye disorders, and trigeminal neuralgia. Each of these shares a surface-level symptom with cluster headache but differs significantly in timing, severity, and associated features.
The diagnostic delay matters because effective treatments exist for both stopping individual attacks and preventing cluster periods, but they’re different from what works for migraines or sinus issues. People who are misdiagnosed often spend years on medications that do nothing for their actual condition.
Genetics and Family Risk
Most cluster headache cases appear sporadically, but there is a genetic component. Across 22 large cohort studies, the percentage of patients reporting a family history of cluster headache ranged from 0% to 22%, with a median of about 8%. When researchers traced inheritance patterns through family trees, roughly 73% of familial cases followed a dominant pattern, meaning the trait could be passed on by a single parent carrying the gene variant. About 24% followed a recessive pattern, requiring both parents to carry the variant.
Having a first-degree relative with cluster headaches does increase your risk, but the condition doesn’t follow a clean, predictable inheritance pattern like some genetic diseases. Environmental and lifestyle factors likely interact with genetic susceptibility. Smoking and alcohol use during active cluster periods are well-established triggers, though neither fully explains who develops the condition in the first place.
Likely Underdiagnosed Worldwide
The wide variation in reported prevalence, from 3 per 100,000 in some surveys to over 1,000 per 100,000 in studies with hands-on neurological exams, strongly suggests that cluster headaches are underdiagnosed in many populations. The studies that find higher rates tend to be the ones where trained specialists directly evaluate participants rather than relying on questionnaires or medical records. In regions with limited access to neurologists, many people with cluster headaches may never receive a correct diagnosis at all, living through repeated attack cycles without knowing what they have or that treatments are available.