Societal Burden of the Headache

Jenny Berg

Nabih M. Ramadan

Headache is almost a universal experience in that it affects most people at some stage in their lives. Nevertheless, the impact of a headache attack or the headache disorder on the individual and society vary significantly from one person to the next. The individual burden of headache is measured by the degree of pain and suffering, and by the health consequences of the disorder. On the other hand, the burden of headache on society is assessed through cost-of-illness studies, which quantify resources and costs resulting from a disease, which can be complemented by cost-effectiveness evaluations that assess the economic impact of different treatment strategies. Understanding the economic costs of headache to society facilitates informed resource allocation in the overall research and management of the condition. Furthermore, cost-effectiveness analyses allow decision making on the differential, and potentially preferential, funding of specific therapies.

This chapter focuses on the economic costs of headache to society, including medical resource use, lost productivity, and other direct and indirect costs of illness. To date, most published cost-of-illness studies have focused solely on migraine as the prototype primary headache, although tension-type headache is almost five times more prevalent (10). Therefore, the cost estimates presented in this chapter apply mostly to migraine, but we also describe the limited information that is available on resource use and productivity impact of other headaches.

DEFINITION AND METHODS

The costs of an illness are generally defined as follows:

Direct costs relate to the management and treatment of the disease and cover costs incurred to the healthcare system, social services, patients, or their families.
Indirect costs are those resulting from the loss of productivity caused by the disease. These can be a consequence of lost working days (absenteeism), reduced productivity while at work (presenteeism), unemployment because of the disease, early retirement, or premature death. Indirect costs may also include social drift; however, to our knowledge, no such data exist for headache.
Intangible costs relate to patients’ suffering and reduced quality of life from the disease.

To assess the burden of migraine to society, cost-of-illness studies using a societal perspective should be used. In terms of specific costs, direct medical costs and indirect costs due to absenteeism and presenteeism are most relevant for this condition. Impaired quality of life from headache is covered in detail elsewhere in this book.

There are several different methodologic approaches to cost-of-illness studies. Concerning the patient population, a study can be prevalence- or incidence-based. In the prevalence-based approach, all costs for patients with the condition incurred during a given time period, generally 1 year, are included. In the incidence-based approach, in contrast, the lifetime costs for patients first diagnosed in a given year are calculated. The first type of studies are useful for budget and planning decisions, and the latter are more suited for estimating the effect of treatment on future costs. Because of the recurrent nature of most headaches, prevalence-based studies are more relevant for this condition.

Depending on the data sources used, a cost-of-illness study can be described as top-down or bottom-up. Top-down studies draw on statistical databases and registries, and bottom-up studies collect costs directly from a sample of the patient population. The latter approach can be applied either prospectively by following the sample for a given time period or retrospectively by gathering information about resource use through patient charts and questionnaires.

For this overview, published journal articles covering costs of headache and migraine, using the International Headache Society (IHS) criteria, as well as productivity and employment aspects were reviewed. Although a fair number of studies has been conducted regarding the resource use relating to and the general impact of migraine
and other headaches, only a limited number of studies have reported direct and indirect costs for migraine. As far as possible, studies with a societal perspective and based on the IHS criteria were used. One of the cost studies listed in Table 2, however, does not entirely fulfill these criteria. The direct costs in the study by Hu et al. (8) were based on medical claims data from an insurance database of over 40 employers, which may not be representative of the whole migraine population in the U.S. Although the U.S. study by Osterhaus et al. (18) is often quoted in cost-of-illness reviews, the patient population was drawn from a clinical trial, which is likely to lead to an overrepresentation of moderate and severe patients. Indeed, the estimated costs per patient in this study are very high and lie well above the range of other estimates available for the United States and other countries. Therefore, for the United States, only the costs derived by Hu et al. are used in this overview.

HEALTHCARE RESOURCE UTILIZATION CAUSED BY MIGRAINE AND OTHER HEADACHES (DIRECT COSTS)

The direct costs of an illness are those incurred in diagnosing and treating the condition (Table 5-1). Migraine has a significant impact on direct costs, which is illustrated by the fact that migraineurs are more likely to consult general practitioners than nonmigraine age- and gender-matched controls. Based on an analysis of U.S. medical claims data from 1989 and 1990, migraineurs had 1.7 times the number of medical claims compared to controls, 4 times the number of emergency department (ED) visits, and nearly 2.5 times as many pharmacy claims (4). A recent hospital survey in the United States indicates that 2.4 million of 90 million ED visits are headache related, representing 2.6% of total ED visits (National Hospital Ambulatory Medical Care Survey 1999; available at: www.cdc.gov/nchs). In other words, headache is the fourth most common cause of ED visits in the United States.

Estimates of the direct cost of headache vary widely, and tend to be based on migraine data (Table 5-2). In population-based European studies, the 2003 scaled direct annual cost of migraine per patient ranged from $33 in Germany (17) to $243 in Spain (2). In the United States, the comprehensive analysis of a large medical claims database yielded an annual cost of $59 per patient, with women contributing over 80% of the total direct costs (8). By comparison, studies that have evaluated the direct cost of migraine in clinical trial participants result in significantly higher figures (e.g., $817 per patient per year [18]). There are several reasons that account for the reported discrepancies in the estimated direct costs of migraine, including the following:

TABLE 5-1 Contributing Factors to the Direct Cost of Headaches

Outpatient visits

Inpatient visits and hospitalizations

ED visits

Prescription and nonprescription medications

Diagnostic testing (e.g., CT scan, MRI, LP)

Complementary and alternative treatments (e.g., herbal therapy, physical therapy, biofeedback)

Abbreviations: CT, computed tomography; ED, emergency department; LP, lumbar puncture; MRI, magnetic resonance imaging.

Patient population. Some studies have estimated the direct cost based on data obtained from participants in clinical trials, clearly leading to an overestimate of costs, because these patients are likely to represent the more severe migraine segment.
Cost ascertainment. For example, the estimate from The Netherlands (28) based the number of outpatient visits on a pilot study where hospital representatives and neurologists were interviewed, which might introduce an element of recall bias. Furthermore, most direct cost estimates are based on top-down calculations, which in turn may lead to underestimation. The only exception to this approach is the French bottom-up study by Michel et al. (14).
Time period of study. Estimates of direct yearly cost of migraine are not constant over time. For example, IMS Health data suggest that U.S. migraine medication sales were almost 10 times higher in 1996 and 1997 ($700 million) than in 1993 ($86 million).
Definition of direct cost of illness. For example, Hu et al. (8) included acute migraine medications, in- and outpatient visits, and diagnostic testing for migraine in their calculation of direct costs. Over-the-counter medication use, alternative treatments such as physical therapy or acupuncture, and preventative medications were not taken into account. On the other hand, Michel et al. (14) and van Roijen et al. (28) included the cost of complementary/alternative therapies in their estimates of the total direct cost of migraine.
Confounding costs. For example, Clouse and Osterhaus (4) did not differentiate migraine from nonmigraine medical costs. Because comorbidity is significant in migraineurs, the high direct cost of migraine that Clouse and Osterhaus reported may be an overestimation.
Geo-cultural differences. The use of healthcare resources may vary among countries. For example, complementary therapies may be more commonly used in Europe than in the United States.

The effectiveness of headache management strategies with respect to the medical cost of migraine has been the focus of several recent publications (e.g., Silberstein et al. [23], Adelman et al. [1], Williams [30], and Goldfarb
et al. [6]). Silberstein et al. (23) retrospectively reviewed information stored in a U.S. health use database. A period prior to the use of preventive drugs was compared to a period when patients were prescribed a prophylactic therapy. In addition, patients were eligible only if they were using sumatriptan for acute treatment. The authors’ results suggested that the cost of acute medications and total healthcare costs are reduced when preventative therapies are used. These data, when recalculated to account for the cost of preventative therapy, indicate that cost reduction on prophylactic agents is highest with generic treatments such as amitriptyline and propranolol (1). Cost improvement with valproate is observed beyond 6 months of treatment in high-end sumatriptan users. Williams et al. (30) used data from a clinical trial to project the differential effect of stepped versus stratified care on total direct migraine cost in the United Kingdom. Stratified care was estimated to reduce the medical cost of migraine by approximately £5 per patient per year, which is a substantial number given the high prevalence of the condition. Last, Goldfarb et al. (6) reviewed medical claims in a health maintenance organization in the United States and found that limiting the monthly use of sumatriptan reduced the total prescription drug cost, but did not have any significant influence on the monthly total direct cost of migraine.

TABLE 5-2 Total Direct Medical Costs of Migraine per Patient and Year^a

*Country*	*Total Direct Medical Costs*	*Hospitalization*	*Drugs*	*Outpatient Care*	*Medical Procedures and Devices*	*Reference*
France	76	21	15	24	16	Michel et al. (14)
Germany	33	5	15	13	n/a	Neubauer and Ujlaky (17)
The Netherlands	78	2	7	67	1	Van Roijen et al. (28)
Spain^b	243	50	28	165	n/a	Badia et al. (2)
U.S.^b	59	8	17	33	n/a	Hu et al. (8)
^aUS$, scaled to 2003 Prices. ^bNot including OTC drug costs; ED visits included under hospitalization costs.