Headache: The Evidence Behind the Pathway

It is a Monday outpatient clinic. A 34-year-old bank officer sits down and says, a little apologetically, that it is “only a headache”. She has had them since college. They come on one side, throb, make her feel sick and want a dark, quiet room, and last most of a day. Lately there are about eight headache days a month. A doctor near her home started propranolol 40 mg once a day a month ago, and she takes a combination tablet of paracetamol with caffeine “whenever it starts”, which is now about twelve days a month. Her examination is normal. This is an illustrative case, not a real patient.

The Headache Pathway will help you with her: whether this is migraine by the criteria, whether she is overusing her tablets, which triptan to give on top of propranolol, and which preventive suits a woman of her age. What it no longer carries is most of the why. The Indian survey figures, the trial results and the full triptan and prevention tables that used to sit in its Drugs and Evidence tabs have moved here, so that the tool stays quick at the bedside. By the end of this article you will know where the numbers behind those instructions come from, and which of them are firm.

1. “Only a headache” is the commonest neurological disorder in India

Why begin with figures for a whole state, when the patient in front of you is one woman? Because the way she apologised is the way most Indians think about headache, and the figures show how wrong that is. This used to be section 9 of the tool’s Evidence tab.

India has one good population survey of headache, and its own authors say so. In and around Bangalore, 2329 adults were surveyed door to door. The one-year prevalence of any headache was 63.9 per cent. Migraine was 25.2 per cent, against a mean global estimate of 14.7 per cent. Tension-type headache was 35.1 per cent. Headache on 15 or more days a month was 3.0 per cent, and probable medication-overuse headache 1.2 per cent (Kulkarni 2015).

The same survey measured the cost. Headache took 4.3 per cent of all productive time, and 40 per cent of that lost time was paid work. Fewer than a quarter of the people with headache had consulted anybody in a year (Kulkarni 2015; Rao 2015). The Global Burden of Disease estimate is 488 million people in India with a headache disorder in 2019, which makes it the commonest neurological disorder in the country (India State-Level Disease Burden Initiative 2021). That paper also makes a point worth repeating to anybody who calls this a minor complaint: it says headache disorders go unrecognised as a public health problem, and are left out when standards of care are set.

“One in four Indians has migraine? Can that be right?” Hold the figure loosely. When the survey’s questionnaire was checked against an interview by a neurologist, the agreement, measured as kappa, was 0.46 for migraine and 0.39 for tension-type headache. Kappa is agreement beyond what chance alone would give, from 0 for none to 1 for perfect. A kappa of 0.46 is moderate and 0.39 is fair: good enough to say migraine is very common here, not good enough to defend the decimal point.

2. Reading the triptan table

A triptan is the drug for a moderate or severe migraine attack, and none appears anywhere on the national workflow. The tool prints the doses, the repeat interval, the 24-hour maximum and what is on the Indian shelf. On a phone it now shows this as two narrow tables. Here is the whole table, as the tool carried it on 4 November 2026.

Read it across, one row at a time. The dose is what she takes at the start of the attack. “Repeat after” is the shortest gap before a second dose if the headache comes back. The maximum is the ceiling for 24 hours, all doses added together. Notice that naratriptan alone has a four-hour gap; the rest are two hours, and subcutaneous sumatriptan one hour.

“So I just pick a triptan and use the dose in the table?” Not for our patient. She is on propranolol, and propranolol and rizatriptan interact. The tool keeps the rule in its Drugs tab: on propranolol, rizatriptan is 5 mg only, to a maximum of 15 mg a day, half the ceiling in the table. This is the common mistake, and it is an easy one: propranolol is the first preventive the national workflow names, so the pairing is everyday, and the table on its own does not warn you. Sumatriptan has no such adjustment in the tool, and sumatriptan 25 mg and 50 mg are the only triptans on the National List of Essential Medicines 2022.

The brand column comes from Indian drug databases, not from a register of the Central Drugs Standard Control Organisation. Check it against what your own pharmacy stocks.

3. The preventive depends on who the patient is

With eight headache days a month, she qualifies for prevention under the national workflow, which starts it at more than two attacks a month. Here is the tool’s prevention table, whole. On a phone it is now two narrow tables: the doses, then what to watch for.

Why does the last column matter more than the first two? Because every drug here works for some patients, and the choice is made by what each one does to the patient who takes it. Look at the rows with her in mind. She is a woman of childbearing potential. Topiramate is contraindicated for her unless she uses highly effective contraception. Divalproex is contraindicated outside a pregnancy prevention programme, with a 10.3 per cent major malformation rate and no safe low dose. Those two rows are where a prescription can do lasting harm.

“Then why not flunarizine for everyone? It is cheap, and it is on the essential list.” It does work. The tool gave responder rates of 46 and 53 per cent at 5 and 10 mg, against 48 per cent for propranolol 160 mg, in 810 patients. The tool names no paper for that trial on its reference list, and neither can this article. But look at its row: weight gain in 11 per cent, drowsiness in 9 per cent, depression in 5 per cent, no use in depression or Parkinsonism, two drug-free days every week, and a stop at six months even when it is working. A drug that is easy to start is not always easy to take.

For our patient, the propranolol she is already on is a reasonable preventive at a dose that has not been reached: 40 mg once a day against a usual effective range of 80 to 160 mg a day. Give any preventive eight weeks at a therapeutic dose before calling it a failure.

4. The tablets that keep the headache going

Count her tablet days before you add anything. Twelve days a month of a combination analgesic with caffeine is over the tool’s threshold of 10 or more days a month for that class. She is not yet at 15 headache days a month, so she does not meet the criteria for medication-overuse headache, but she is heading there.

“How common is medication overuse in Indian headache clinics?” It depends on who is counting, and how. In an Indian series of 202 patients with chronic daily headache, 12.3 per cent met the older criteria for medication-overuse headache and 46.5 per cent the newer ones (Laskar 2019). The patients were the same. Only the definition changed, and it changed the answer nearly fourfold. So when you read an Indian prevalence figure for medication overuse, ask which criteria stand behind it before you quote it.

The practical advice is simpler than the epidemiology. Keep acute medication to two headache days a week, and ask her to write the date of every headache and every tablet on the back of her prescription for a month.

5. The sudden headache, and the six hours

Now suppose she comes to casualty one evening, years later, with a headache that reached its worst within a minute, two hours ago. Here the question is not which headache it is. It is whether she has bled. Why does the time since onset matter so much?

Because the scan is at its best early. A non-contrast computed tomogram within six hours of onset had a sensitivity of 100 per cent in 953 patients, of whom 121 had bled (Perry 2011). Sensitivity is the share of patients with the disease whom the test picks up. Across five studies the pooled figure is 98.7 per cent, with 13 misses in about 8907 patients. That pooled analysis is named in the tool’s source notes but is not on its reference list. Past six hours, the sensitivity in the whole cohort falls to 92.9 per cent, which is not an exclusion.

Hold those figures together with the conditions the tool keeps beside them: a modern multi-slice scanner, and a report by somebody who routinely reports head scans. Where either is uncertain, the six-hour exclusion does not apply, and the lumbar puncture is still needed. A figure from a Canadian emergency department is a figure about that scanner and that reporter.

6. Why the oxygen has to come from a cylinder

One more patient, briefly: a man in his thirties who comes in clutching one eye, which is red and watering, in a pain that will be gone within three hours. This is cluster headache, and the national workflow says only to refer. The tool gives oxygen 100 per cent at 12 L/min through a non-rebreathing mask with a reservoir bag, for 15 minutes.

“Is oxygen really a treatment, or just something to do while we wait?” It is a treatment. In the randomised trial, 78 per cent were pain free or adequately relieved at 15 minutes, against 20 per cent on air (Cohen 2009). The flow and the mask are part of the treatment: a 10 L/min concentrator cannot deliver it, and does not deliver 100 per cent oxygen. It needs a cylinder and a reservoir mask.

Back to the clinic

Our bank officer meets the criteria for migraine without aura. She is told that the combination tablet, used on twelve days a month, is part of the problem, and is asked to keep it to two days a week and to keep a diary. For her attacks she gets sumatriptan 50 mg with an antiemetic, or, if rizatriptan is preferred, 5 mg only, because of the propranolol. Her propranolol is increased towards the 80 to 160 mg range, with a pulse check, and reviewed after eight weeks at a therapeutic dose. Topiramate and divalproex are not used unless highly effective contraception is in place. Each of those steps is a line in the tool. Now you know what stands behind each line.

Try this

A 40-year-old man with migraine takes rizatriptan 10 mg for his attacks. He is started on propranolol for prevention. He asks whether he can carry on as before, with a second 10 mg tablet if the headache returns. What do you tell him? Illustrative case.

Answer. No. On propranolol, rizatriptan is 5 mg only, to a maximum of 15 mg in a day. That is a 5 mg tablet, repeated after two hours if needed, no more than three in a day. Alternatively, switch him to sumatriptan, which needs no such change. And ask how many days a month he takes it: 10 or more days a month of a triptan is overuse.

Take-home points

• Headache is the commonest neurological disorder in India, and fewer than a quarter of people with it have seen anybody in a year.
• Read a triptan table across: dose, repeat interval, 24-hour ceiling. On propranolol, rizatriptan is 5 mg, at most 15 mg a day.
• Choose the preventive by the patient: no topiramate or divalproex for a woman who may become pregnant without highly effective contraception.
• Count tablet days at every visit. Combination analgesics and triptans become overuse at 10 days a month.
• A scan within six hours excludes a bleed only with a modern scanner and an experienced reporter. Cluster headache needs a cylinder, not a concentrator.

Declaration: the author built and maintains the tool. It is free to use. The figures in this article are those the tool carried in its Drugs and Evidence tabs on 4 November 2026, each from the source named beside it; where the tool names no source on its reference list, the article says so. The Headache Pathway is still in administrator preview, pending clinical review.

References

1. Indian Council of Medical Research, Department of Health Research. Standard treatment workflow (STW) for the management of headache, ICD-10 G43-44. New Delhi: Ministry of Health and Family Welfare; 2019.
2. Headache Classification Committee of the International Headache Society. The International Classification of Headache Disorders, 3rd edition. Cephalalgia. 2018;38(1):1-211.
3. Do TP, Remmers A, Schytz HW, Schankin C, Nelson SE, Obermann M, et al. Red and orange flags for secondary headaches in clinical practice: SNNOOP10 list. Neurology. 2019;92(3):134-44.
4. Perry JJ, Stiell IG, Sivilotti MLA, Bullard MJ, Hohl CM, Sutherland J, et al. Clinical decision rules to rule out subarachnoid hemorrhage for acute headache. JAMA. 2013;310(12):1248-55.
5. Perry JJ, Sivilotti MLA, Sutherland J, Hohl CM, Emond M, Calder LA, et al. Validation of the Ottawa Subarachnoid Hemorrhage Rule in patients with acute headache. CMAJ. 2017;189(45):E1379-85.
6. Perry JJ, Stiell IG, Sivilotti MLA, Bullard MJ, Emond M, Symington C, et al. Sensitivity of computed tomography performed within six hours of onset of headache for diagnosis of subarachnoid haemorrhage: prospective cohort study. BMJ. 2011;343:d4277.
7. Perry JJ, Alyahya B, Sivilotti MLA, Bullard MJ, Emond M, Sutherland J, et al. Differentiation between traumatic tap and aneurysmal subarachnoid hemorrhage: prospective cohort study. BMJ. 2015;350:h568.
8. Ailani J, Burch RC, Robbins MS; American Headache Society Board of Directors. The American Headache Society consensus statement: update on integrating new migraine treatments into clinical practice. Headache. 2021;61(7):1021-39.
9. National Institute for Health and Care Excellence. Headaches in over 12s: diagnosis and management. Clinical guideline CG150. London: NICE; 2012, updated 2021.
10. Cohen AS, Burns B, Goadsby PJ. High-flow oxygen for treatment of cluster headache: a randomized trial. JAMA. 2009;302(22):2451-7.
11. Ravishankar K, Chakravarty A, Chowdhury D, Shukla R, Singh S. Guidelines on the diagnosis and the current management of headache and related disorders. Ann Indian Acad Neurol. 2011;14(Suppl 1):S40-59.
12. Kulkarni GB, Rao GN, Gururaj G, Stovner LJ, Steiner TJ. Headache disorders and public ill-health in India: prevalence estimates in Karnataka State. J Headache Pain. 2015;16:67.
13. Rao GN, Kulkarni GB, Gururaj G, Stovner LJ, Steiner TJ. The burden attributable to headache disorders in India: estimates from a community-based study in Karnataka State. J Headache Pain. 2015;16:94.
14. Laskar S, Kalita J, Misra UK. Comparison of chronic daily headache with and without medication overuse headache using ICHD II R and ICHD 3 beta criteria. Clin Neurol Neurosurg. 2019;183:105382.
15. India State-Level Disease Burden Initiative Neurological Disorders Collaborators. The burden of neurological disorders across the states of India: the Global Burden of Disease Study 1990-2019. Lancet Glob Health. 2021;9(8):e1129-44.
16. Robbins MS, Farmakidis C, Dayal AK, Lipton RB. Acute headache diagnosis in pregnant women: a hospital-based study. Neurology. 2015;85(12):1024-30.
17. Marchenko A, Etwel F, Olutunfese O, Nickel C, Koren G, Nulman I. Pregnancy outcome following prenatal exposure to triptan medications: a meta-analysis. Headache. 2015;55(4):490-501.
18. Ministry of Health and Family Welfare, Government of India. National List of Essential Medicines 2022. New Delhi: Government of India; 2022.