Administrative prevalence and incidence, characteristics and prescription patterns of patients with migraine in Germany: a retrospective claims data analysis

Migraine is a disabling primary headache disorder [1] characterised by headaches and associated symptoms including aura, photophobia, phonophobia and nausea/vomiting [2, 3], which can lead to considerable disruption of the professional and private lives of affected individuals [4, 5].

The diagnosis of migraine is based on clinical history and a structured exclusion of secondary headache, using the International Classification of Headache Disorders (ICHD) criteria [1]. Migraine is also accompanied by a number of comorbidities, such as psychiatric disorders (e.g. depression and anxiety), allergy, asthma, sleep disorders, fatigue and cardiovascular diseases [3, 4, 6].

The treatment of patients with migraine aims to relieve pain or limit the attack, reduce disability and restore function, improve health-related quality of life and manage comorbidities [2, 7]. Pharmacological therapy comprises both acute and preventive treatments [8]. Acute treatment options include analgesics (foremost non-steroidal anti-inflammatory drugs [NSAIDs]), antiemetics (e.g. metoclopramide), triptans (e.g. almotriptan, oral or subcutaneous sumatriptan) and emergency medications (e.g. intravenous metoclopramide, subcutaneous sumatriptan) [8]. Triptans, as a class of specific migraine medications, are the first choice for patients who have had an inadequate response to non-migraine specific analgesics [8]. Opioids and ergotamines take a special position in the acute treatment algorithm, and they are generally not recommended by the guideline of the German Migraine and Headache Society and the German Society of Neurology (hereafter referred to as the German guideline for migraine) for reasons of toxicity and other adverse effects [8].

Preventive medications are primarily used to reduce the attack frequency [8]. According to the German guideline, the decision to administer preventive medications should be driven by the frequency of monthly migraine attacks, significant impairment in health-related quality of life and a high risk of medication overuse [8]. In Germany, approved conventional preventive medications (medication classes) are propranolol and metoprolol (beta-blockers), flunarizine (calcium channel blocker), topiramate (anticonvulsant), amitriptyline (antidepressant) and onabotulinum toxin A (muscle relaxant; in chronic migraine only) [8]. Most recently – and outside the scope of these analyses – erenumab, galcanezumab and fremanezumab (calcitonin gene-related peptide [CGRP] inhibitors) were approved in 2018 or 2019 [9‐11]. Valproic acid, although off-label, may be prescribed at the expense of the German Statutory Health Insurance (SHI) pursuant to Annex VI, Section K of the German Medicinal Product Directive. In addition to the approved preventive medications mentioned here, a variety of other medicines are listed in the German guideline for migraine, although they are not approved for migraine in Germany. These include bisoprolol (good evidence for use according to the German guideline for migraine) and agents with less scientific evidence to support their use (opipramol, acetylsalicylic acid, magnesium, magnesium plus vitamin B2 plus coenzyme Q10, angiotensin-converting enzyme inhibitors [ACEis], specifically lisinopril, and the angiotensin II receptor antagonists [ARAs; ‘sartans’] candesartan and telmisartan) (Fig. 1) [8].

Prevalence data for migraine have been obtained using population-based studies [12, 13]. Migraine was reported to have a crude global prevalence of 11.6% (11.4% in Europe) and occurred more frequently in women than men (globally: 13.8% vs 6.9%) [14]. In Germany, the prevalence of migraine was estimated to be 10.6% (women: 15.6% vs men: 5.3%), according to a telephone survey (n = 7341) conducted in 2004 [12], and 13.4% (19.1% vs 7.1%), according to a population-based, longitudinal cohort survey (n = 9944) conducted from 2003 to 2005 [13].

It has been reported that the majority of patients with migraine do not consult a healthcare professional [15]. It is therefore of interest to ascertain how many patients consult healthcare professionals because of their migraine and to further characterise these patients with regard to the treatments received and their healthcare system interaction. Analyses based on claims data are important and powerful sources of information to assist decision making for healthcare stakeholders, researchers and policy makers, and are supported by the German government [16].

The aims of this study – based on administrative claims data – were therefore multifaceted: to describe the administrative prevalence and incidence of migraine in Germany; to provide information on the characteristics of and prescription patterns in patients with migraine; and to determine which healthcare professionals were involved in the diagnosis and continued treatment of these patients. Additionally, the study aimed to describe whether patients prescribed more preventive treatments differ in utilisation patterns and healthcare professional contacts from those prescribed fewer or no preventive treatments.

In this study, administrative claims data were analysed, revealing that the overall administrative prevalence of migraine, extrapolated to the SHI, was 4.0% in 2016. The extrapolated administrative prevalence of migraine in Germany ranged from 2.9% in 2008 to 4.1% in 2015, possibly indicating greater awareness – and hence more diagnoses – of migraine over the 9-year study period. These administrative prevalence rates are considerably lower than the migraine prevalence rates reported in German population-based analyses conducted in the early 2000s (10.6% in 2004 [12] and 13.4% in 2003–2005 [13]). The comparably lower administrative prevalence than population-based prevalence could be due to several factors. A possible explanation is that many patients with migraine experience a low frequency of attacks (≤4 headache days per month) [19] and consequently are likely to have only mild headache-associated disability, or mild impairment of health-related quality of life, work or daily activity [20]. In Germany for example, 45.2% of patients with migraine reported experiencing less than 4 headache days per month [12]. A low frequency of attacks could lead to self-medication and the patient not seeing a healthcare professional at all or after an initial diagnosis. According to an online survey of patients with migraine in Germany, the majority felt able to treat their migraine symptoms with over-the-counter medications after being diagnosed [21]. In addition, not all patients receive a migraine diagnosis. In the American Migraine Prevalence and Prevention Study, only 56.2% of participants with an ICHD 2nd edition diagnosis of migraine reported that they had ever received a medical diagnosis of the disease [22]. An analysis of data from six countries that used population-based sampling in the Eurolight survey revealed that only a minority of those with migraine consulted a healthcare professional (15.8%–33.0% across countries) [15], which might be the reason for many patients fulfilling diagnostic criteria but not receiving a medical diagnosis of migraine.

The increase in the administrative prevalence of migraine in the study period from 2008 to 2016 was observed in both men and women (from 1.0% to 1.6% in men and 4.5% to 6.1% in women). During the period 2011 to 2015, we observed little variation in the administrative incidence of migraine (between 0.19% and 0.22% in men and 0.58% and 0.64% in women). The years 2011 to 2015 were unlikely to have been affected by censorship of the data (our restriction of 2008 to 2016). Some patients diagnosed before 2008 could have been misidentified as incident cases in the first years of the study period, and some patients diagnosed late in 2016 were probably not considered as incident cases due to insufficient time to fulfil the M2Q criterion, which might explain the higher incidence rates in 2009 and 2010 and the lower rate in 2016. The opposing trend in the development of administrative prevalence and incidence over time in this study may be due to the chronic nature of migraine: while more patients with a new migraine diagnosis might enter the sample, fewer patients might leave the sample.

In the current study, a male to female ratio of close to 1:4 was observed (78.0% of the population was female). More than 40% of patients were aged 35 to 54 years. Previous population-based studies show results of a similar order of magnitude, and support the findings of the current study, despite being potentially affected by participation bias. In a 2008 review, Obermann and Katsarava [23] reported a male:female ratio of between 1:2 and 1:6 for migraine across studies; the most frequent onset of this disease occurred during the second and third decades of life, and a peak prevalence during the fourth decade of life. Lipton et al. [19] reported the highest prevalence of migraine in patients aged 30 to 39 years for both men and women, using data from the American Migraine Prevalence and Prevention study; from the published data it can be concluded that 77% of patients were women in that study. Yoon et al. [13] found most patients (both men and women) with migraine were aged between 36 and 45 years, with approximately 62% of patients being women, and the Eurolight project found a similar gender distribution in 15 European countries, with 58% of patients being women [5].

To our knowledge, this is the first study that has used claims data to evaluate changes over time in the type of healthcare professional visited by patients with migraine in Germany. In most instances, general practitioners diagnosed migraine in the index year and most frequently treated patients throughout each follow-up year. These findings are corroborated by another study that analysed patient visits during a 12-month period prior to presenting to a German specialist headache and facial pain clinic: the healthcare professionals most commonly consulted due to migraine were also general practitioners (89.5% of patients), followed by neurologists (74.9%) [24].

The most commonly prescribed acute medications in the overall study population were analgesics/NSAIDs (74.2% of acute medication prescriptions, including 23.7% ibuprofen, 19.7% metamizole and 19.1% opioids), with only 21.2% of prescriptions being for migraine-specific triptans. Patients with complicated migraine (G43.3, including chronic migraine) had a slightly higher number of acute and emergency medication prescriptions, and a higher percentage of all acute prescriptions were for triptans than was observed in the overall study population (37.9% vs 21.2%). However, analgesics/NSAIDs were still the most commonly prescribed acute medications in this patient subgroup (57.4% of all acute prescriptions). These findings suggest that migraine-specific medication is underutilised in Germany. An analysis of Eurolight data indicated that the majority of patients with migraine do not receive adequate treatment in a number of European countries, including Germany [15]. A recent report from a single specialist headache and facial pain centre in Germany showed that more than one-third of patients were not treated according to the German guideline for migraine [24]. Another study reported that approximately one-third of German patients did not receive any migraine treatment (acute or preventive) as revealed by a database analysis of 56,823 patients in 2015 [25].

Looking at preventive treatments, 22.3% of patients in the overall study population received at least one prescription for any in-label preventive medication (plus valproic acid) (Group 1 preventive medication). When both in- and off-label preventive medications (Group 2 preventive treatments) listed in the German guideline for migraine [8] were considered, the percentage was slightly higher (29.1%). Compared with the overall study population, a higher percentage of patients who had been diagnosed with complicated migraine (G43.3, including chronic migraine) received at least one prescription for in-label preventive medications plus valproic acid (38.0%; Group 1 preventive medication). Altogether, the doctors’ prescribing behaviour generally adhered to the approval and refunding status of the preventive medications in Germany.

The proportion of patients ever receiving any preventive medication in the current study is at the lower end of the range suggested as adequate by publications based on disease characteristics. The American Migraine Prevalence and Prevention study identified that 25.7% of patients met the criteria for being offered preventive medication, and an additional 13.1% should consider optionally using it [19]. More recently, the Eurolight study identified that 38.5% of patients with migraine in Germany had an attack on ≥5 days/month and would be expected to require preventive medication [15]. The potential differences between patients who might benefit from preventive medications and patients ever receiving such treatment in the current study are smaller than those reported by Ziegler et al. [24], who reported that half of the patients considered eligible for preventive medication had never received such medications.

In the current analyses, the most frequently prescribed preventive medications included beta-blockers and ARAs, which – according to the German guideline for migraine – have good and lower evidence, respectively [8]. Of those patients who received either in-label preventives (plus valproic acid) or any preventive medication (i.e. both in- and off-label preventives per German guideline for migraine), most had been prescribed only one type of drug (i.e. one drug class) during the analysis period. As few as 7.9% of patients received only one type of preventive medication lasting more than 1 year; most discontinued medications during the 9 years of follow-up, receiving a preventive medication for only a short time or only once. This finding is in general agreement with two studies conducted in the United States [26, 27]. An analysis of US administrative claims data (2005–2014) revealed that 90.3% of patients receiving first-line preventive medications were non-persistent; of these, 38.9% switched, 30.1% restarted and 31.0% discontinued treatment [26]. In another US administrative claims database analysis (2008–2015) – which included only patients with chronic migraine – oral preventive medication adherence at 12 months was higher than the rate identified in the current analyses, but still low, ranging from 17% to 20% [27]. Although an analysis of observational studies confirmed that 12-month adherence (35%–56%) and persistence (7%–55%) with preventive medications were low in patients with migraine [28], rates were usually higher than those reported in the current study. The most common reason for discontinuation of preventive medication in randomised controlled trials was reported to be adverse events [28]. Reasons for the short duration of preventive medication treatment in the current analyses are unknown.

In the current study, patients who received at least two in-label preventive medications from different classes were more likely to have reports of in- or outpatient hospitalisation. We were unable to ascertain whether hospitalisation increased the chance of receiving preventive medication or whether patients receiving multiple types of preventive medication had more severe disease that resulted in an increase in the risk of hospitalisation. Higher acute and emergency medication consumption was recorded in patients having received at least two preventive medications from different treatment classes. Healthcare resource use (migraine-related outpatient, emergency room and hospital visits, and testing) was high among patients who had discontinued at least two preventive medications because of lack of efficacy and/or tolerability in a retrospective chart review performed among neurologists, headache specialists and pain specialists in France, Germany, Italy and Spain [29]. These recent findings led the authors to conclude that there is a need for more effective prophylactic treatments to appropriately manage migraine and to reduce the associated healthcare resource use.

When deciding on a preventive treatment, a number of factors need to be considered, including existing comorbidities, as these have the potential to result in contraindications to preventive medications. As most patients in this analysis had at least one concurrent diagnosis between 2008 and 2016, it is possible that the available preventive medications were not suitable for some of the patients. However, this cannot be confirmed as the usage of preventive medication was not analysed according to concurrent diagnoses. Nevertheless, some patients may have benefitted from alternative therapeutic options. Since the end of the analysis period (2016), monoclonal antibodies targeting CGRP or its receptor have expanded the preventive treatment options for patients with migraine in Germany. Further studies are needed to determine the impact of this expanded portfolio.

The administrative prevalence of migraine in these analyses was lower than historically reported population-based prevalence rates, suggesting that many patients with migraine do not seek medical care. Preventive medication was received by less than one-third of patients, with most prescribed only one such medication and few continuing preventive treatment beyond 1 year. Patients who did have prescriptions for at least two different preventive therapy classes during the study period were the most likely to use other healthcare resources (healthcare professional visits and acute and emergency medications). These outcomes suggest that there is scope for improvement in migraine management in Germany.