What helps a tension headache? Ordinary painkillers, and only modestly

The dull, pressing headache that affects about one person in five worldwide has a plain answer to the question of what helps: an ordinary over-the-counter painkiller works, but only modestly, and for a minority of the people who take it [s1]. That is the consistent verdict of three separate Cochrane reviews of paracetamol, ibuprofen and aspirin for the acute treatment of frequent episodic tension-type headache — the most rigorous evidence available, and a useful corrective to the idea that any single tablet reliably switches the pain off.

How the benefit is measured

The reviews report a "number needed to treat," or NNT: how many people must take the drug for one extra person to get a defined benefit that they would not have got from a placebo. A low NNT means a drug that helps many; a high one means a drug that helps few. Because headaches often ease on their own, this is the honest way to size a painkiller's effect — it strips out the people who would have improved anyway.

Paracetamol: a small benefit

The paracetamol review pooled 23 studies in adults with frequent episodic tension-type headache, with 8079 participants in total [s1]. For the headache society's preferred outcome — being pain free two hours after dosing — paracetamol 1000 mg gave an NNT of 22 (95% confidence interval 15 to 40) against placebo, with no significant difference from placebo at one hour [s1]. On a softer measure, being pain free or in only mild pain at two hours, the NNT was 10 (7.9 to 14) [s1]. Fewer people needed rescue medication on the drug than on placebo, with a number needed to treat to prevent one such event of 7.8 (6.0 to 11) [s1]. Paracetamol 1000 mg was no different from ibuprofen 400 mg or ketoprofen 25 mg, and side effects matched placebo (relative risk 1.1) with no serious events reported [s1]. The reviewers called the benefit "small."

Ibuprofen: a similar picture

The ibuprofen review included 12 studies and 3094 participants [s2]. For being pain free at two hours, ibuprofen 400 mg had an NNT of 14 (8.4 to 47) versus placebo, again with no advantage at one hour [s2]. For a global rating of "very good" or "excellent," the NNT was 5.9 (4.2 to 9.5) [s2]. As with paracetamol, rescue-medication use was lower than with placebo, giving a number needed to treat to prevent one event of 8.9 (5.6 to 21) [s2]. Adverse events did not differ from placebo (relative risk 1.1), and no serious events were reported [s2]. The authors described "an important benefit… for a small number of people" — a phrase that captures the whole field.

Aspirin: thinner evidence still

The aspirin review found only five eligible studies [s3]. No trial reported the pain-free-at-two-hours outcome at all, so the strongest claims rest on secondary measures [s3]. Fewer people needed rescue medication on aspirin 1000 mg than on placebo — 14% versus 31% — and more rated themselves satisfied, 55% versus 37% [s3]. Side effects again matched placebo [s3]. The reviewers judged the evidence low to very low quality and said it "should be interpreted with caution" [s3].

What the pattern means

Read together, the three reviews say something clearer than any one of them. A standard dose of a common analgesic genuinely outperforms a dummy pill for acute tension-type headache, the effect is modest and reaches only a fraction of users, and none of the three drugs stands out as markedly better than the others on the head-to-head data that exist [s1] [s2] [s3]. Reassuringly, at these single doses the side-effect rate was no higher than placebo across all three reviews [s1] [s2] [s3].

The reviews studied single doses for occasional headaches, not daily use — and that boundary matters, because frequent reliance on painkillers can itself drive medication-overuse headache, a problem these trials were not designed to detect. They also excluded migraine, which is treated differently. A headache that is new, unusually severe, changing in pattern, or accompanied by other neurological symptoms sits outside this evidence entirely and is a reason to seek care. Which drug suits a given person, and whether a recurring headache is ordinary tension-type at all, is a clinical judgement this article does not make.