The Most Painful Headache Requires Minutes, Not Weeks | The Pained Brain, Chapter 11

Cluster headache screening · Primary care

Screening for Cluster Headache in Primary Care: A Three-Item Workflow

Screen for cluster headache in primary care with the Erwin Test, a validated three-item decision tree on pain intensity, attack duration and same-side autonomic features, with 84 percent sensitivity and 89 percent specificity. Trigger it for recurrent unilateral head or face pain, let clinical staff give it by standing order, and record the answers in a structured field.

The median patient reaches a correct cluster headache diagnosis after a decade of wrong ones. The tool that shortens that interval takes three questions and a structured field in the record.

Screening for cluster headache in primary care is a workflow problem before it is a pharmacology problem. The pooled interval from first attack to correct diagnosis is 10.43 years across 22 studies, only 21 percent of 1,134 American survey respondents were identified correctly at initial presentation, and the instrument that would close much of that gap already exists. Chapter 11 of The Pained Brain, the video above, by Dr. Gurpreet Singh Padda, MD, MBA, MHP, and Dr. KrisJay Fucanan, MD, treats the disorder as time-critical. For a practice the operational questions are narrower: where a validated screen sits in intake, and what a positive result sets in motion.

Why does cluster headache take so long to diagnose?

The 10.43-year mean needs a caveat that sharpens it. It is inflated by older cohorts, and delay has fallen by decade of onset in British and Danish series. First-contact accuracy has not kept pace. In 144 Italian and Eastern European episodic patients, 77 percent were misdiagnosed at first consultation with 2.27 diagnoses each, and 93 percent underwent instrumental or laboratory testing. In a Portuguese series of 64, before the correct label 17.2 percent had received opioids and 14.1 percent dental procedures. In an American chart-validated cohort of 75, mean delay was 12.7 years, and the excess dental, TMJ and septal diagnoses were judged very likely to be misdiagnoses rather than comorbidity.

Three drivers sustain the pattern. Attack timing is hypothalamic, and pain is delivered through the trigeminal-parasympathetic reflex, so the event resolves before examination and is silent on imaging. The third is organizational: among 218 family and emergency physicians, 15 percent rated themselves sufficiently knowledgeable, nearly half did not know autonomic features are mandatory for diagnosis, 72.5 percent of family physicians send an acute attack to the emergency department, and 92.9 percent of emergency respondents manage these patients in the lowest-acuity zone.

How accurate is screening for cluster headache in primary care?

The Erwin Test for Cluster Headache is a three-item decision tree on intensity, duration and autonomic features, validated in 319 participants including 109 with cluster headache. Reported sensitivity was 84 percent, specificity 89 percent, positive predictive value 76 percent and negative predictive value 93 percent. The design limits interpretation: single-center validation with enriched recruitment of trigeminal autonomic cephalalgia means positive predictive value in an unselected primary-care panel will run lower, and no study has yet shown the screen shortens delay. Its strengths are the negative predictive value and its brevity.

Operationally it behaves like any validated instrument: a trigger, a script, a discrete result. A defensible trigger is recurrent unilateral head or face pain, including patients returned from dental or sinus evaluation without a cause. Written as a standing order, it can be administered by clinical staff before the physician enters, and a structured field makes the answer retrievable at the next visit instead of reconstructed from unstructured notes; see getting results into the record.

Can cluster headache be mistaken for migraine?

Migraine-like features are common in cluster headache and should not be used to exclude it. In 1,604 questionnaire respondents meeting criteria, 50.1 percent had photophobia or phonophobia, 31.4 percent pain aggravated by activity and 27.5 percent nausea or vomiting, while 99.0 percent had at least one cranial autonomic feature and 96.6 percent restlessness. The authors concluded that prototypical migrainous features do not separate the two disorders.

Sex compounds the error. In a Swedish biobank of 874 re-verified patients, self-reported migraine was 29.4 percent in women against 12.5 percent in men. In a Chinese registry of 1,206, conjunctival injection was less frequent in women, 38.52 against 56.03 percent, while restlessness was identical. The woman in the exam room is more likely to present as migraine and less likely to show the red eye. Age matters as well: onset before 20 carried a mean 13.8-year delay in 400 Danish patients against 2.1 years after 40, and in the international sample 27.5 percent reported onset before 18, of whom only 15.2 percent were diagnosed before 18. Adolescent visits belong in the screening population.

What changes after a cluster headache diagnosis?

The argument for screening rests on what follows a correct label. The American Headache Society lists subcutaneous sumatriptan, zolmitriptan nasal spray and high-flow oxygen as Level A acute treatments and suboccipital steroid injection as the only Level A preventive; opioids are absent, partly because they were never trialed. Practice diverges sharply. Among 7,589 newly diagnosed commercially insured patients, the leading prescription class over the next twelve months was opioids, at 41 percent, and oxygen was claimed by 16.2 percent within seven days of diagnosis. Patients rated opioids completely or very effective in 6 percent against 54 percent for oxygen. Access is the other constraint: of 566 patients who eventually obtained oxygen, 13 percent waited two to five years, and the mean commercial wait for a new neurology visit is 49.7 days. A documented positive screen lets the plan begin during the bout rather than after it.

Documentation after a positive result

The three screen answers, dated, in a discrete field
Episodic or chronic classification, because response splits: galcanezumab reduced weekly attacks in episodic disease and missed its primary endpoint in chronic disease, and non-invasive vagus nerve stimulation separated the same way in two sham-controlled trials
Prior diagnoses and procedures, dental and sinus interventions included
For patients on verapamil, the electrocardiogram schedule: in an audit, 19 percent of 108 patients with tracings on file had an arrhythmia, and 41 percent of 217 treated patients had no tracing at all
Refractory status against the consensus definition of at least three severe attacks weekly despite three adequate preventive trials

Where Measura fits, and where it does not

Measura [Cardiometabolic and Autonomic Health Analysis] does not diagnose primary headache disorders, and nothing in its library substitutes for the screen or the criteria. It also does not perform the electrocardiogram that verapamil monitoring calls for; that rhythm check remains with the prescriber. Cardiac autonomic reflex tests and heart rate variability quantify systemic autonomic regulation of the heart, a different physiological question from the cranial trigeminal-parasympathetic reflex, and a result should not be read as evidence for or against cluster headache. These measurements belong to the rest of the patient, when your own selection criteria call for an autonomic or cardiometabolic baseline. The circadian physiology behind attack timing is covered in the clock and pain, and the incentive structure that keeps fast, low-intensity care in the slow lane is taken up in the repeat-procedure pathway. Every study cited here, with its limits, is in the Chapter 11 technical companion.

Frequently asked questions

What is the Erwin Test for Cluster Headache?

It is a three-item decision tree covering headache intensity, attack duration and cranial autonomic features. In its validation study of 319 participants it showed 84 percent sensitivity and 89 percent specificity, with a negative predictive value of 93 percent. Recruitment was enriched, so expect lower positive predictive value in general practice. Building it into intake works best when the trigger and result field are fixed; see making screening reproducible.

Which patients should be screened?

A practical trigger is recurrent, strictly unilateral head or face pain, particularly in patients carrying a migraine, sinus or dental explanation that has not held up. Adolescents belong in scope, since 27.5 percent of one international sample reported onset before 18 and few were diagnosed then. Women presenting with nausea and light sensitivity deserve the questions rather than a default migraine label. More on applying criteria by specialty is in specialty applications.

Do nausea and light sensitivity exclude cluster headache?

No. In 1,604 respondents meeting criteria, half reported photophobia or phonophobia and more than a quarter nausea or vomiting, while nearly all had at least one cranial autonomic feature and most were restless. The authors concluded that migrainous features do not differentiate the disorders. Restlessness during the attack and ipsilateral autonomic signs are the discriminating findings. Common workflow questions are collected in the physician questions page.

Can autonomic testing confirm or exclude cluster headache?

No. Autonomic nervous system testing, heart rate variability and cardiac autonomic reflex tests measure systemic autonomic regulation of the heart and circulation. Cluster headache is a clinical diagnosis based on attack pattern and cranial autonomic features, and a normal or abnormal systemic result does not change that. Order autonomic testing when your criteria call for it on its own merits. Details are on the autonomic nervous system testing page.

How should a positive screen be documented?

Record the three answers with a date in a discrete field, the episodic or chronic classification, prior diagnoses and procedures, and the monitoring plan for any preventive that needs an electrocardiogram. Structured entries keep the finding visible at the next encounter and support the same documentation discipline used for quality reporting. The MIPS and quality reporting page describes how structured results are handled.

How does a doctor diagnose cluster headaches?

Cluster headache is a clinical diagnosis built on the attack pattern: severe one-sided head or face pain, how long each attack lasts, restlessness during the attack and same-side autonomic features such as a red eye. Imaging stays silent because the attack is over before the exam. In primary care, the three-item Erwin Test gives clinical staff a validated way to ask those questions and record the answers in a structured field.

What treatments are recommended for cluster headache?

The American Headache Society lists subcutaneous sumatriptan, zolmitriptan nasal spray and high-flow oxygen as Level A acute treatments, and suboccipital steroid injection as the only Level A preventive. Opioids are not on the list. Practice runs the other way: opioids were the leading prescription class after diagnosis, while patients rated oxygen completely or very effective 54 percent of the time and opioids 6 percent.

See how the protocol fits a practice

Learn how Measura testing is ordered, performed and returned to the chart when your selection criteria call for an autonomic or cardiometabolic baseline.

Send a message

4477 Woodson Rd, Suite 201, St. Louis, MO 63134. Monday to Friday, 9:00 a.m. to 5:00 p.m. Please do not send symptoms, diagnoses or images through a web form — a website form is not a secure medical channel. Send your name and number and we will call you back.

References

Van Obberghen, E. K., Fabre, R., & Lanteri-Minet, M. (2025). Cluster headache diagnostic delay and its predictors: a systematic review with a meta-analysis. The Journal of Headache and Pain, 26(1), 71. https://doi.org/10.1186/s10194-025-02001-7
Parakramaweera, R., Evans, R. W., Schor, L. I., Pearson, S. M., Martinez, R., Cammarata, J. S., Amin, A. J., Yoo, S. H., Zhang, W., Yan, Y., & Burish, M. J. (2021). A brief diagnostic screen for cluster headache: Creation and initial validation of the Erwin Test for Cluster Headache. Cephalalgia, 41(13), 1298–1309. https://doi.org/10.1177/03331024211018138
Schor, L. I., Pearson, S. M., Shapiro, R. E., Zhang, W., Miao, H., & Burish, M. J. (2021). Cluster headache epidemiology including pediatric onset, sex, and ICHD criteria: Results from the International Cluster Headache Questionnaire. Headache, 61(10), 1511–1520. https://doi.org/10.1111/head.14237
Hasirci Bayir, B. R., Nazli, E., & Ulutas, C. (2025). Cluster Headache Management: Evaluating Diagnostic and Treatment Approaches Among Family and Emergency Medicine Physicians. Medicina (Kaunas, Lithuania), 61(3), 437. https://doi.org/10.3390/medicina61030437
Joshi, S., Rizzoli, P., & Loder, E. (2017). The comorbidity burden of patients with cluster headache: a population-based study. The Journal of Headache and Pain, 18(1), 76. https://doi.org/10.1186/s10194-017-0785-3
Silva, L., Millner, T., Cabral, M., Costa, A., Almeida, P., Barreto, B., Pinto, M., Dias, R., Moreira, S., Bonifácio, G. V., Pinto, S. M., Castro, R., Alves, I., Rocha, A. L., Varanda, S., Costa, A., & Andrade, C. (2026). Delayed diagnosis of cluster headache in Portugal. Arquivos de Neuro-Psiquiatria, 84(9), 1–6. https://doi.org/10.1055/s-0046-1827040
Robbins, M. S., Starling, A. J., Pringsheim, T. M., Becker, W. J., & Schwedt, T. J. (2016). Treatment of Cluster Headache: The American Headache Society Evidence-Based Guidelines. Headache, 56(7), 1093–1106. https://doi.org/10.1111/head.12866
Choong, C. K., Ford, J. H., Nyhuis, A. W., Joshi, S. G., Robinson, R. L., Aurora, S. K., & Martinez, J. M. (2017). Clinical Characteristics and Treatment Patterns Among Patients Diagnosed With Cluster Headache in U.S. Healthcare Claims Data. Headache, 57(9), 1359–1374. https://doi.org/10.1111/head.13127
Cohen, A. S., Matharu, M. S., & Goadsby, P. J. (2007). Electrocardiographic abnormalities in patients with cluster headache on verapamil therapy. Neurology, 69(7), 668–675. https://doi.org/10.1212/01.wnl.0000267319.18123.d3
Fourier, C., Ran, C., Steinberg, A., Sjöstrand, C., Waldenlind, E., & Belin, A. C. (2022). Sex Differences in Clinical Features, Treatment, and Lifestyle Factors in Patients With Cluster Headache. Neurology, 100(12), e1207–e1220. https://doi.org/10.1212/WNL.0000000000201688