Medication-overuse headache: when painkillers fuel the pain they treat

Medication-overuse headache, a secondary headache disorder caused by the regular overuse of acute headache medication, affects an estimated 60 million people worldwide, yet it remains poorly recognized by health care professionals and incompletely understood even among specialists. A new narrative review published in Frontiers in Pain Research by researchers at the Danish Headache Center in Copenhagen synthesizes what is currently known about the condition’s epidemiology, diagnosis, underlying mechanisms and treatment, and argues that greater clinical and public awareness is essential because the disorder is, in the authors’ words, entirely preventable.

According to the International Classification of Headache Disorders, third edition, medication-overuse headache is diagnosed in individuals with a pre-existing headache disorder when headache occurs on 15 or more days per month and acute medication has been overused regularly for more than three months. The criteria are drug-specific: triptans, ergotamines, opioids and combination analgesics qualify as overuse when taken on 10 or more days per month, while simple analgesics require 15 or more days per month. Crucially, the current criteria no longer require proof that the headache improves after withdrawal, a change introduced in 2006 that allows earlier diagnosis and intervention but also complicates prevalence research by blurring the line between medication overuse itself and the headache disorder it may drive.

The scale of the problem is difficult to pin down precisely. The global prevalence is suggested to be around 1 to 2 percent, but estimates vary enormously by region, with reported rates of 7.6 percent in Russia, 4.6 percent in Iran and 7.1 percent in Zambia. A large survey in the Danish Capital Region estimated a prevalence of 2.0 percent in the general population, covering adolescents from age 16 upward. In clinical settings the picture is starker: between 30 and 70 percent of chronic headache patients meet the diagnostic criteria. A longitudinal Norwegian study reported an incidence of 0.72 cases per 1000 person-years, and registry data from the same country showed that 10.3 percent of 33,790 triptan users met criteria for the disorder over an 11-year period, with overuse consistently more common in women than in men.

The public health burden is substantial. New data from the Global Burden of Disease, Injuries, and Risk Factors study published in 2025 indicate that over 15 percent of the health loss attributed to migraine, and more than half of the health loss attributed to tension-type headache, amounting to 20.8 percent overall, can be traced back to medication-overuse headache as a sequela of these disorders. At the population level, the condition produces losses in productivity and socioeconomic costs that exceed those of migraine or tension-type headache alone. The review’s authors caution, however, that epidemiological estimates should be interpreted carefully, because registry-based studies may miss over-the-counter analgesics while clinic-based studies may inflate frequency by including selected patients with more severe headache.

The pathophysiology remains only partly elucidated, but preclinical work in rodents offers a detailed hypothesis. Sustained administration of sumatriptan produced prolonged increases in CGRP- and substance P-positive neuronal profiles in the trigeminal ganglion, and caused cephalic and extracephalic allodynia that could be reversed by naproxen and by a CGRP antagonist but not by a substance P receptor blocker. Long-term sumatriptan exposure also increased the number of trigeminal ganglion cells expressing neuronal nitric oxide synthase, with the largest increase in nerves supplying the dura, and blocking this enzyme with the selective inhibitor NXN-323 reversed and prevented the allodynia. Sumatriptan exposure further lowered the threshold for cortical spreading depression and increased Fos expression in the trigeminal nucleus caudalis. The review stresses that direct human evidence for these mechanisms remains limited, since the findings come from rodent models.

Human studies nonetheless support the idea that pain processing is altered in the disorder. Neuroimaging research has reported functional and structural changes in regions involved in pain modulation, reward processing and cognitive control. A PET study found hypometabolism in the orbitofrontal cortex in individuals with medication-overuse headache, and these changes were partly normalized after medication withdrawal, suggesting reversibility, although causality cannot be definitively established. A structural imaging study reported reduced grey matter volume in the anterior cingulate cortex, insula and thalamus compared with healthy controls, though that work was cross-sectional and used a definition of medication overuse rather than strict diagnostic criteria. Psychophysical studies have documented reduced pain thresholds and exaggerated responses to suprahreshold stimuli that diminished after withdrawal, and exploratory biomarker and proteomic studies have identified differences in serum profiles, findings the authors describe as preliminary and in need of replication.

Clinically, the condition usually arises in people with an underlying primary headache disorder, most often migraine, tension-type headache or a combination of the two, and it has also been described in cluster headache and in secondary disorders such as post-traumatic headache and idiopathic intracranial hypertension. The headache phenotype typically reflects the underlying disorder but often blurs into a more diffuse, holocranial, dull or pressing daily pain that lacks the characteristic migraine features of nausea, vomiting and light or sound sensitivity. The trajectory toward the disorder is substance-specific: overuse of triptans, opioids and combination analgesics appears to accelerate its development faster than overuse of simple analgesics, which are nonetheless the most commonly overused drugs globally. In one cluster headache study, 16 percent of patients fulfilled the criteria for probable medication-overuse headache, falling to 4 percent under a more conservative definition, while a South Korean study found the condition in 7.8 percent of idiopathic intracranial hypertension patients and a Danish study in 13 percent of those with persistent post-traumatic headache.

Risk factors are reasonably well mapped. The most consistent is high baseline headache frequency, in the range of 10 or more days per month, which substantially increases the likelihood of progression. Elevated body mass index, smoking, sedentary lifestyle, low socioeconomic status and psychological comorbidities such as stress, anxiety and mood disorders have also been associated with increased susceptibility. A population-based case-control study found opioid use associated with chronic daily headache after adjustment for confounders, while aspirin and ibuprofen were inversely associated, though the authors cautioned that confounding by indication could not be excluded. The review also notes that the condition overlaps conceptually with substance-use disorders in showing dependency-like behaviors, but should not be considered an addiction disorder, since medication overuse often occurs in the context of frequent or undertreated headache.

Treatment centers on withdrawal from the overused medication, either by complete discontinuation or by restricting acute medication to a maximum of two days per week, often combined with prophylactic treatment. Most individuals can be managed as outpatients with education, follow-up and preventive therapy, while inpatient withdrawal may be considered for severe overuse or opioid and barbiturate use. Relapse occurs most often within the first year and is associated with high baseline headache frequency, psychiatric comorbidity, poor adherence and continued frequent acute medication use. Preventive options with demonstrated efficacy in chronic migraine populations that include patients with medication overuse include topiramate, valproic acid and amitriptyline, though results are limited by side effects, and onabotulinumtoxinA showed benefit only in subgroup analyses. Newer monoclonal antibodies targeting CGRP reduced monthly migraine and headache days regardless of medication-overuse status at baseline, and post-hoc analyses showed efficacy was not blunted by overuse, with low relapse rates during follow-up. Indirect evidence suggests gepants such as rimegepant may also aid management, but no randomized trial has yet tested them specifically for the condition.

Prevention may be the most powerful tool of all. A Norwegian population-based study with a 1.5-year follow-up showed that simple advice about the condition from general practitioners reduced analgesic use days from 22 to 6 per month, underlining the potential of primary care intervention, while public awareness campaigns in Russia and Denmark have sought to educate the wider population. The review’s authors frame the disorder as a state of maladaptive but reversible brain changes rather than an irreversible condition, a message they argue can motivate patients to adhere to withdrawal strategies. Because the mechanisms by which altered pain modulation operates remain incompletely established, they conclude that management should combine withdrawal, preventive treatment, patient education and relapse prevention, and that directing resources toward prevention is ultimately more cost-effective than allowing poorly managed pain to progress into a chronic condition requiring complex interventions.

Subject of Research: Medication-overuse headache, a secondary headache disorder caused by overuse of acute headache medication

Article Title: Medication-overuse headache: a review

Article References: Melchior, A. G., Carlsen, L. N., & Jensen, R. H. (2026). Medication-overuse headache: a review. Frontiers in Pain Research, 7, Article 1846017. https://doi.org/10.3389/fpain.2026.1846017

Image Credits: AI Generated

DOI: 10.3389/fpain.2026.1846017

Keywords: medication-overuse headache, headache disorders, ICHD-3, migraine, CGRP, medication withdrawal, triptans, pain modulation, neuroimaging, prevention, epidemiology, Frontiers in Pain Research