Sage Journals: Discover world-class research

Abstract

Background

Case definitions of medication-overuse headache (MOH) in population-based research have changed over time. This study aims to review MOH prevalence reports with respect to these changes, and to propose a practical case definition for future studies based on the ICHD-3 beta.

Methods

A systematic literature search was conducted to identify MOH prevalence studies. Findings were summarized according to diagnostic criteria.

Results

Twenty-seven studies were included. The commonly used case definition for MOH was headache ≥15 days/month with concurrent medication overuse ≥3 months. There were varying definitions for what was considered as overuse. Studies that all used ICHD-2 criteria showed a wide range of prevalence among adults: 0.5%–7.2%.

Conclusions

There are limits to comparing prevalence of MOH across studies and over time. The wide range of reported prevalence might not only be due to changing criteria, but also the diversity of countries now publishing data. The criterion “headache occurring on ≥15 days per month” with concurrent medication overuse can be applied in population-based studies. However, the new requirement that a respondent must have “a preexisting headache disorder” has not been previously validated. Exclusion of other headache diagnoses by expert evaluation and ancillary examinations is not feasible in large population-based studies.

Keywords

Chronic headache medication-overuse headache epidemiology prevalence classification International Classification of Headache Disorders

Introduction

Medication-overuse headache (MOH) is a condition in which the cure for a headache becomes the cause. A person with headache or predisposition to headache takes pain medication frequently but without relief. On the contrary, the headache worsens, leading to a cycle of increasing medication use and increasing pain. The adverse effect of medication overuse as a possible cause for the headache is commonly unacknowledged both by headache sufferers and health care providers. The result is chronic headache (CH) that is refractory to treatment (1 –4).

“Medication-overuse headache” as a diagnosis is very new. A classification scheme for headache was first published by an ad hoc committee of the United States (US) National Institutes of Health in 1962 but there was no mention of medication overuse as cause or associated factor (5). More than 20 years later, in 1986, an expert group agreed on a definition of “drug-related headache” (6). The first International Classification of Headache Disorders (ICHD-1) was published in 1988, and described “headache induced by chronic substance use or exposure” (7). In 1994, Silberstein et al. proposed criteria classifying four types of primary headaches as being with or without medication overuse (8) (in this paper referred to as CDH-1994). ICHD-2 was published in 2004 with the new diagnosis “medication-overuse headache” (9). Revisions to the definition were published in 2005, 2006 and 2009, collectively referred to in this paper as ICHD-2 R (10 –12). ICHD-3 beta was published in mid-2013 (13). A summary and comparison of the definitions since 1988 are shown in Table 1.

Table 1.

Definitions of medication-overuse headache from 1988 to 2013.

ICHD-1 (7)	CDH-1994 (8)	ICHD-2 (9)	ICHD-2 revised (10,11,12)	ICHD-3 beta (13)
Year	1988	1994	2004	2005, 2006, 2009	2013
Diagnosis	Headache induced by chronic substance use or exposure	(Primary headache) with medication overuse	Medication-overuse headache	Medication-overuse headache	Medication-overuse headache
Headache frequency	≥15 d/mth	>15 d/mth (>180 d/yr) lasting 4 h/d	>15 d/mth	≥15 d/mth	≥15 d/mth
Symptoms	Diffuse, pulsating, distinguished from migraine by absent attack pattern and/or absent associated symptoms	Symptoms of transformed migraine, chronic TTH, NDPH or HC	Bilateral, pressing/ tightening quality, mild to moderate intensity With triptans: aggravated by routine physical activity, associated with nausea/vomiting or photo-/phonophobia Developed or markedly worsened during medication overuse	Headache developed or markedly worsened during medication overuse	Preexisting headache disorder
Drugs/substances and frequency of use	Ergotamine oral ≥2 mg or rectal ≥1 mg daily Aspirin ≥ 50 g/mth or equivalent of other mild analgesics Combination analgesics ≥100 tablets/mth One or more narcotic analgesics daily	Simple analgesics >1 g ASA/ acetaminophen, >5 d/w for migraine, NDPH and HC; > 6 d/w for TTH Combination analgesics >3 tablets/d, >3 d/w Narcotics >1 tablets/d, >2 d/w for migraine, NDPH and HC; >3 d/w for TTH Ergotamine oral 1 mg or rectal 0.5 mg >2 d/w	≥10 d/mth for ergotamines, triptans, opioids or combination preparations ≥15 d/mth for simple analgesics	≥10 d/mth for ergotamines, triptans, opioids or combination preparations ≥15 d/mth for simple analgesics or a combination of any class of analgesics without overuse of any single class alone	≥10 d/mth for ergotamines, triptans, opioids, combination preparations, or a combination of drugs without overuse of any single drug or drug class alone, or if the identity/ quantity/ pattern of overuse cannot be reliably established ≥15 d/mth for simple analgesics
Duration of overuse	≥3 months	>1 mth for transformed migraine, NDPH and HC; >6 mth for TTH	>3 months	>3 months	>3 months
Other criteria	Disappears within one month after withdrawal of the substance		Headache resolves or reverts to previous pattern within two months after discontinuation of overused medication	Improvement after detox not required for diagnosis Not better accounted for by another headache diagnosis.	Not better accounted for by another headache diagnosis

h: hours; d: day; w: week; mth: month; yr: year; ASA: acetylsalicylic acid; HC: hemicrania continua; NDPH: new daily persistent headache; TTH: tension-type headache; detox: detoxification.

In describing the earliest field testing of ICHD-1, Rasmussen et al. wrote about the fundamental requirements of a headache classification system: that it should be generalizable, exhaustive, reliable and valid. Generalizable means that “it should be applicable in diverse settings, e.g. headache clinics, general practices, general populations” (14).

Arriving at a working definition of MOH that is “valid, reliable, (and) agreed-upon” has been challenging (15) and as Ferrari et al. pointed out in 2008, “in such a complex disorder, it is difficult to obtain a classification which is both easily applicable and unambiguous” (16).

Nonetheless, we cannot underestimate the positive effect of efforts to develop diagnostic criteria in order to foster greater commonality in understanding the condition across various settings. The alternative would have been for clinicians and researchers to deal with about 40 concurrent diagnoses (Table 2) with the attendant difficulties in comparing results across settings and countries, and a general confusion as to the extent of the problem.

Table 2.

Synonyms and related terms used to describe medication-overuse headache.

• Analgesic headache (17)

• Analgesic abuse headache (18)

• Analgesic-induced headache (19–21)

• Analgesic-induced chronic headache (22,23)

• Analgesic rebound headache (24)

• Chronic analgesic-induced headache (25)

• Chronic analgesic-overuse headache (26)

• Chronic headache with analgesic overuse (27,28)

• Chronic daily headache with (substantial) medication overuse (8,29)

• Chronic daily headache associated with long-term misuse of headache medication (30)

• Drug abuse headache (31)

• Drug dependence headache (32)

• Drug-associated headache (33)

• Drug-induced headache^a (34–37)

• Drug-induced daily headache (38) or chronic daily headache (39)

• Drug-induced refractory headache (40)

• Drug-rebound headache (41)

• Drug-related headache (42)

• Headache associated with chronic use of substances (43,44)

• Headache caused by analgesic or ergotamine abuse (45–47)

• Headache caused by ergotamine addiction (48)

• Headache induced by chronic substance use or exposure^b

• Headache: with special reference to the excessive use of ergotamine (49)

• Medication abuse-related headache (50)

• Medication-induced headache (51,52)

• Medication-misuse headache^a (53)

• Medication rebound headache (54,55)

• Painkiller headache (56)

• (Primary headache diagnosis) with medication overuse (8)

• Rebound headache^a (57)

• Rebound-withdrawal headache (58)

• Substance name + headache; e.g. ergotamine headache (59)

• Substance name + induced (daily) headache; e.g. ergotamine-induced headache (60), caffeine-induced headache (61), sumatriptan-induced daily headache (62)

• Substance name + withdrawal headache; e.g. ergotamine withdrawal headache (17), caffeine-withdrawal headache (63)

• Withdrawal headache (64)

References illustrate use of the term in a journal article or book, and do not necessarily represent first use by the authors.

^aListed as “previously used terms” in the International Classification of Headache Disorders (ICHD), second edition (2004) and third edition (2013).

^bDiagnosis used in the ICHD, first edition (1988).

To our knowledge, there has not been a study looking at how the different definitions and diagnostic criteria for MOH relate to the estimates of MOH prevalence worldwide.

The objectives of this paper are to review the estimates of MOH prevalence in population-based studies with respect to diagnostic criteria, and to explore a practical case definition that will be useful for future prevalence studies based on the new ICHD-3 beta criteria. The purpose is to contribute to improving the generalizability of the new classification scheme so that it can be applied both in clinical settings and in population-based studies.

Methods

The flowchart for selection of papers included in this review is shown in Figure 1, following PRISMA guidelines (65).

Figure 1.

Flow diagram for identification of papers with data on the prevalence of medication-overuse headache or chronic headache with/without medication overuse.

An electronic database search was conducted in April–May 2013 to identify population-based studies (community, town or country) with representative data on the prevalence of MOH or CH with medication overuse. The search was performed online using PubMed (www.ncbi.nlm.nih.gov/pubmed) and limited to December 31, 2012, with no start date (default 0001-01-01). Language was not delimited at the outset.

Given the many synonyms for MOH (Table 2), the database search was performed using very general Medical Subject Headings (MeSH) Terms “headache disorders/epidemiology” or “headache/epidemiology” and “prevalence.” “Medication overuse headache” did not exist as a MeSH term, but the equivalent terms “analgesic overuse headache” and “analgesic rebound headache” were classified under “headache disorders, secondary.” In addition, a free text search using “medication overuse headache” and “prevalence” was conducted.

All hits were downloaded as citation.nbib files in batches of 200 and entered into Reference Manager 12. Journal articles labeled as case report, review, lecture, comment, guideline, editorial, historical article or randomized controlled trial were excluded. Citations retrieved from the different search methods were combined and duplicates were removed.

At the next screening step, titles and abstracts were reviewed. Records that appeared potentially relevant were retrieved as full papers. A standard for acceptable yield was not predefined, although a yield of one relevant item per 100 references scanned was used as a guide.

Authors MW and RJ identified four definitive review papers on the prevalence of CH. MW is a medical doctor who has finished the first year of a neurology residency, and works full time as a researcher on MOH. RJ is a professor of neurology and head of a national headache center. The four review papers were used as high-yield sources to identify papers on MOH prevalence. Furthermore, the review papers acted as a gold standard against which the results of the PubMed search were compared to see what proportion of references were retrieved by the keyword search strategies used (66). These reviews listed epidemiological studies worldwide from 1999 to 2005 (Stovner et al.) (67), worldwide up to 2006 (Scher et al.) (68), in Europe up to 2007 (Stovner and Andree) (69), and in Asia-Pacific up to 2012 (Stark et al.) (70).

Two measures of prevalence were used, either: (1) the percentage of people with MOH in a sample of the population with age- and sex-specific data, if available; or (2) the percentage of people with MOH among those with CH. Prevalence data were summarized according to year and method of data collection, case definitions, population groups (adolescents, adults or elderly) and country.

Uncertainty about the summary data from one study was discussed with its author (Chu Min Kyung, personal communication June 8, 2013).

In this paper, CH is defined as headache ≥15 days/month over a prolonged (not necessarily specified) period. Some of the reviewed papers used the terms chronic daily headache (CDH), chronic frequent headache, or daily headache. These terms also denoted headache ≥15 days/month but specified other qualities of the headache such as duration per episode.

Results

Twenty-seven papers were included in this review (Table 3). The three search methods complemented each other. The initial screening of electronic database records yielded about seven papers per 100 records retrieved. The final screening of papers for eligibility yielded two papers per 100 initially retrieved.

Table 3.

Population-based studies on medication-overuse headache or chronic headache with medication overuse identified through a systematic literature search.

Year	First author	MeSH keyword search^a	Free text “MOH”^a	Review papers	Country	Year of data collection	Method of data collection	Validation methods and case ascertainment	Recall period^b
1999	Castillo (71)	+		+	ES	1995–1997	Q, I, E, D	Different data collection methods; all with frequent headache examined by neurologist	Six months^c One month diary
2000	Wang (72)	+		+	TW	1993–1994	Q, I, E	Validated questionnaire; all with frequent headache examined by physician	One year
2001	Lu (73)	+	+	+	TW	1997–1998	I	Responses to interview confirmed by physicians by telephone	One year
2001	Prencipe (74)		+	+	IT	1992	Q, I, E	Validated questionnaire; all with frequent headache examined by neurologist	One year
2001	Scher 2001 (75)			+	US	1997–2000	I	Responses to general health questionnaire confirmed by structured telephone interview	One year
	Scher 2010 (76)	+	+		US
2004	Colás (77)	+		+	ES	1998–1999	Q, I, E, D	Different data collection methods; all with frequent headache examined by family physician and neurologist	Three months One month diary
2004	Zwart (78)	+		+	NO	1995–1997	Q, E	Two sets of validated questionnaires; general health examination (98)	One year
2006	Wiendels (79)	+	+	+	NL	2002–2003	Q	Two sets of questionnaires	Six months
2006	Dyb (80)	+	+	+	NO	1995–1997	Q, I	Validated questionnaire and structured interview conducted by nurses	One year
2006	Wang (81)	+	+	+	TW	2000	Q, I, E	Validated questionnaire; all with frequent headache examined by neurologist	One year
2008	Aaseth 2008 (82)	+	+		NO	2005	Q, I, E	Validated questionnaire; all with frequent headache examined by neurologist with video or interviewed by telephone	One year
	Aaseth 2009 (83)	+	+		NO		Q, I, E
2008	Rueda-Sánchez (84)	+			CO	n.r.	Q	Validated questionnaire and prior pilot study	Six months
2009	Straube (85)	+	+		DE	2002–2006	Q, I	Validated questionnaire and structured interview by trained interviewers	Six months headache One month medication
2009	Katsarava (86)	+		+	GE	n.r.	I, E	Validated questionnaire and structured interview conducted by medical residents (99)	One year
2010	Da Silva (87)		+		BR	2005–2006	I, E	Structured interviews; all with frequent headache examined by neurologist; reassessment of MOH after two months	One year
2011	Jonsson 2011 (88)	+	+		SE	2009–2010	I	Questionnaire based on ICHD-2; national telephone survey conducted by experienced interviewers	Three months
	Jonsson 2012 (89)	+	+		SE		I
2011	Lipton (90)	+	+		US	n.r.	Q, I, D	Different data collection methods; validated questionnaire; structured telephone interview	One year One month diary
2011	Linde (91)	+	+		NO	2006–2008	Q	Validated questionnaires and follow-up of cohort (100)	One year headache One month medication
2011	Chu (92)			+	KR	2009	I	Semi-structured interview based on ICHD-2	Three months
2012–2013	Shahbeigi (93)		+		IR	2010	I	Structured interview based on ICHD-2	One year
2012	Ertas (94)	+	+		TR	2008	I	Questionnaire based on ICHD-2; structured interview conducted by physicians	One year
2012	Yu (95)	+		+	CN	2008–2009	I	Validated questionnaire and structured interview conducted by neurologists; prior pilot study	One year
2012	Ayzenberg (96)	+	+		RU	2008	Q, I	Validated questionnaire and structured interview conducted by medical residents (101)	One year
2012	Hagen (97)	+	+		NO	1995–1997 2006–2008	Q, I, E, NorPD	Validated questionnaires and follow-up of cohort (100)	One year recall; 11 years’ follow-up

Medical Subject Headings (MeSH) keyword search refers to “headache disorders/epidemiology” or “headache/epidemiology” combined with the keyword “prevalence.” Free text “MOH” refers to “medication overuse headache” and “prevalence.”

Recall period refers to frequent headache or medication intake concurrent with frequent headache to fulfill diagnostic criteria for MOH. In several studies, data were also gathered about lifetime headache prevalence and age of onset of frequent headaches.

Recall duration was not reported but following CDH-1994 criteria: >1 month for transformed migraine, new daily persistent headache and hemicrania continua; >6 months for tension-type headache.

+ indicates that the article was found using this search method.

Q: questionnaire; I: interview (telephone or face-to-face); E: examination by physician; D: headache diary; NorPD: Norwegian Prescription Drug Registry; n.r.: not reported; ICHD-2: International Classification of Headache Disorders, second edition; MOH: medication-overuse headache; CDH: chronic daily headache; BR: Brazil; CN: China; CO: Colombia; DE: Germany; ES: Spain; GE: Georgia; IR: Iran; IT: Italy; KR: Republic of Korea; NL: Netherlands; NO: Norway, RU: Russian Federation; SE: Sweden; TR: Turkey; TW: Taiwan; US: United States.

The highest level MeSH terms “headache disorders/epidemiology” (2703 citations) and “headache/epidemiology” (1665 citations) did not produce the same results. The MeSH term “headache disorders, secondary” (1866 citations) yielded duplicates of the records found using the other two search terms and was thus not used in the next step. The first two keyword searches were then combined with the MeSH term “prevalence” resulting in 1280 unique citations. There were 1025 records left after exclusion of journal articles labeled as case reports, reviews, lectures, comments, guidelines, editorials, historical articles and randomized controlled trials.

The free text search for “medication overuse headache” initially identified 232 citations that yielded 133 papers after delimiting by time and exclusion of nonrelevant article categories. The results of the MeSH term search and the free text search were then combined to yield 1117 unique citations.

The MeSH term search found 22 papers. The free text search identified two unique papers (by Da Silva et al. (87) and Shahbeigi et al. (93)) that had no associated MeSH terms as of April 2013. The paper by Prencipe et al. (74) was found using this free text search method because this study was surprisingly indexed under MeSH terms “headache/drug therapy” and “headache/physiopathology” and therefore not found in the search for “headache/epidemiology.”

The four review papers contributed two additional studies by Scher et al. (75) and Chu et al. (92) These were poster abstracts presented at conferences and thus not indexed in PubMed. Aside from these two posters, an unpublished report and a foreign-language thesis, all other papers cited by the reviews were also found in the various keyword searches. Using only previous reviews as source material would have been inadequate because these identified only 13 out of the 27 papers that were finally included. The other 14 papers were published more recently or were outside the region of interest in the specific reviews (Europe or Asia Pacific).

Several papers used the same data sets: Scher et al. 2001 and 2010 (75,76); Aaseth et al. 2008 and 2009 (82,83); and Jonsson et al. 2011 and 2012 (88,89). The studies described findings from 16 countries. Only one study by Hagen et al. (97) presented an incidence estimate (Table 3).

The studies from Norway and Sweden had 20,000 to 50,000 participants. The study of US adults started with a large baseline survey of 55,000 people then analyzed data from 206 cases and 507 controls. The three-phase study of US adolescents had baseline data from 21,000 households. There were data from 16,000 people in the study from the Netherlands. Eight studies used data from 3000 to 8000 people and eight studies reached 2000 people or less.

Diagnostic criteria

The diagnostic criteria used by the authors closely followed the consensus of their time. The studies on the elderly by Wang et al. (72) and Prencipe et al. (74) used structured questionnaires based on ICHD-1 but all those with headache more than 15 days a month were assessed for medication overuse following CDH-1994 criteria.

The study on adolescents by Wang et al. (81) collected data in 2000, before the publication of ICHD-2. The reported prevalence of 0.3% in this group of 12 - to 14-year-olds nonetheless was based on ICHD-2 criteria, but the authors commented that the prevalence would have been lower if CDH-1994 criteria had been used.

Among reports that used ICHD-2, only the study by Da Silva et al. (87) attempted a trial of two-month detoxification before making a firm diagnosis of MOH. Their reported prevalence proportions were 1.6% (all probable MOH), and 0.9% (including only those who improved with drug withdrawal).

In terms of headache frequency, the studies were very consistent in the use of 15 days per month as the cut-off point. “Headache days ≥180 per year” was used by some studies, such as Scher et al. (75,76) and Aaseth et al. (82) as an alternative to “≥15 days per month.” The longer recall period probably introduced a bias, and the case definition assumed that the duration of symptoms should at least be one year. The study on adolescents by Dyb et al. (80) used “>5 days a week.”

Duration of frequent headache was rather variable: from one month to one year. In some reports, it was not clear whether the duration of symptoms was specified at all during data gathering, or this was somehow inferred from subsequent questions about treatment (i.e. one, three or six months of medication overuse).

The greatest variation was in the definition of medication overuse. The CDH-1994 criteria included dose and frequency whereas ICHD-2 criteria specified only frequency (Table 1).

Data collected before the publication of ICHD-2 (by Castillo et al. (71), Wang et al. (72), Lu et al. (73), Prencipe et al. (74), and Colás et al. (77)) used the CDH-1994 criteria for overuse. The study by Colás et al. reacted to increasing reports of triptan-overuse headache in the mid-1990s, and revised the CDH-1994 criteria by adding triptans ≥ 2 days a week.

The HUNT-2 questionnaire from Norway (91) used the simpler categorical response of “daily or almost daily” medication use over a period of one month and ≥3 months. The studies by Scher et al. in the US asked about number of medication days, and used “daily intake” to describe a subset of those with chronic headache.

The study by Wiendels et al. (79) in the Netherlands listed specific substances with cut-off points generally lower than the CDH-1994 criteria. This was the only study that included caffeine (>5 units a day) in the list of overused substances. This might account for the relatively higher prevalence compared to other European studies, and possibly also for the larger proportion of people with frequent headache associated with overuse (63%).

Hagen et al. (97) were unique in exploring the use of data retrieved from the Norwegian Prescription Drug registry (NorPD). They established a set of cut-off points for different medications according to defined daily doses. These cut-off points were estimated from an earlier study on critical doses by Limmroth et al. (102).

Prevalence in adults and the elderly

Prevalence was reported in two ways: (1) crude prevalence (number of cases divided by total respondents) overall, sometimes by gender and age, and sometimes adjusted by population census parameters; and (2) as a fraction of cases of CH.

Reports of 19 studies on adults and the elderly are summarized in Figure 2. The prevalence proportions for adults mostly ranged from 0.5% to 2.6%, plus several outliers: 7.2% reported from Russia by Ayzenberg et al. (96) and 4.9% from Iran by Shahbeigi et al. (93). The prevalence of 4.6% is estimated for Colombia based on the finding of Rueda-Sánchez and Díaz-Martínez (84) that 54.9% of 122 people with CDH overused medication according to ICHD-2 criteria. The authors reported an adjusted CDH prevalence of 8.4% (95% confidence interval (CI) 6.9–10.0). They did not report the prevalence of MOH, although it can be estimated that more than half of those with CDH (8.4 × 54.9%) may have MOH.

Figure 2.

Prevalence of medication-overuse headache in adults and elderly by year of data collection, case definition and country (19 studies).

The two studies that focused on the elderly (65 years or older) reported 1.0% prevalence in Taiwan (72) and 1.7% in Italy (74).

The reported proportion of people with CH who overused medication ranged from 11% in Georgia (86) to 68% in Russia (96). The highest proportion of 70% was derived from raw data in the report from Iran (178 with MOH among 253 with CH) (93).

The large studies that were able to show sub-analyses across age groups showed peak prevalence among adults in their 40 s and 50 s (Table 4). Predominance among females was shown in all the studies that reported sex-specific data, except the study from Iran where prevalence was 4.8% among females and 5.1% among males (p = 0.712) despite CH being slightly more prevalent among females (7.2% vs 6.5%, p = 0.446) (93) (Table 4).

Table 4.

Summary of results of population-based studies on medication-overuse headache or chronic headache with medication overuse.

Year	First author	Country	Criteria	Age group (years)	Total prevalence, % (raw data)		Prevalence in females, %	Prevalence in males, %	Age group with highest prevalence (years)	% with medication overuse among those with chronic headache (raw data)
1999	Castillo (71)	ES	CDH-1994	A (≥14)	1.2	(22 of 1883)	25	(22 of 89)
2000	Wang (72)	TW	CDH-1994	E (≥65)	1.0	(15 of 1533)	25	(15 of 60)
2001	Lu (73)	TW	CDH-1994	A (≥15)	1.1	(37 of 3377)	1.3^e	0.9^e	34	(37 of 108)
2001	Prencipe (74)	IT	CDH-1994	E (≥65)	1.7	(14 of 833)	38	(14 of 37)
2001	Scher 2001 (75)	US	Other	A (18–65)	32	(Of 206 cases)
Scher 2010 (76)	US
2004	Coláas (77)	ES	CDH-1994	A (≥14)	1.4	(74 of 4855)	2.6	0.2	50–59	30	See reference (77)
2004	Zwart (78)	NO	Other	A (≥20)	1.0^a 0.9^b	(508 of 49064)^e (447 of 49064)^e	1.3^a 1.2^b	0.7^a 0.6^b	F: 40–49 M:50–59
2006	Wiendels (79)	NL	Other	A (25–55)	2.6^e	(430 of 16232)	63	(430 of 679)
2006	Dyb (80)	NO	Other	Y (13–18)	0.2	(10 of 5847)^e	0.3	0.1	36	(10 of 28)^e
2006	Wang (81)	TW	ICHD-2 R	Y (12–14)	0.3	(24 of 7900)	0.5^e	0.1^e	20	(24 of 122)
2008	Aaseth 2008 (82)	NO	ICHD-2 R	A (30–44)	1.7^f	(240 of 20598)	2.2	1.1	40–44
Aaseth 2009 (83)	NO	About half
2008	Rueda-Sánchez (84)	CO	ICHD-2	A (18–65)	4.6^e	(Of 1505) See text.	55^f	(Of 122 cases)
2009	Straube (85)	DE	ICHD-2 R	A (35–75)	1.1^e	(73 of 6536)
2009	Katsarava (86)	GE	ICHD-2 R	A (≥16)	0.9	(10 of 1145)	11	(10 of 87)
2010	Da Silva (87)	BR	ICHD-2	A (10–93)	1.6^c 0.9^d	(25 of 1605)^e (15 of 1605)^e	44^c	(25 of 57)^e
2011	Jonsson 2011 (88)	SE	ICHD-2 R	A (≥15)	1.8	(799 of 44300)	2.5	0.9	40–49	56	(799 of 1428)
Jonsson 2012 (89)	SE
2011	Lipton (90)	US	ICHD-2 R	Y (12–17)	See text.	See text
2011	Linde (91)	NO	Other	A (≥20)	1.0^f	(409 of 39690)	1.3	0.6	30–49	44^e 42^e	(524 of 1189) in 1995 (409 of 973) in 2006
2011	Chu (92)	KR	ICHD-2 R	A (>19)	0.5^e	(7 of 1506)	28^f	(7 of 27)
2012	Shahbeigi (93)	IR	ICHD-2 R	A (>10)	4.9	(178 of 3655)	4.8	5.1	45–64	70^e	(178 of 253)
2012	Ertas (94)	TR	ICHD-2 R	A (18–65)	2.1	(114 of 5323)
2012	Yu (95)	CN	ICHD-2 R	A (18–65)	0.6	(Of 5041)	0.9	0.3
2012	Ayzenberg (96)	RU	ICHD-2 R	A (18–65)	7.2	(145 of 2025)	68	(145 of 213)
2012	Hagen (97)	NO	Other	A (≥20)	Incidence: 0.72 per 1000 person years

CDH-1994: Criteria for chronic daily headache with medication overuse developed by Silberstein, Lipton et al. (1994);

ICHD-2 R: International Classification of Headache Disorders, 2nd edition, revised;

A: adults; Y: youth/adolescents; E: elderly; n.r.: not reported; F: female; M: male; BR: Brazil; CN: China; CO: Colombia; DE: Germany; ES: Spain; GE: Georgia; IR: Iran; IT: Italy; KR: Republic of Korea; NL: Netherlands; NO: Norway, RU: Russian Federation; SE: Sweden; TR: Turkey; TW: Taiwan; US: United States.

Country codes are listed under Figure 2.

≥1 month of medication overuse; ^b≥3 months of medication overuse; ^call MOH prior to detoxification; ^dMOH diagnosed two months after drug withdrawal;

Extracted from published data; ^fAdjusted according to demographic data.

Prevalence in young people

There were only three studies in the list dealing exclusively with young people, but these were geographically diverse: from Norway, Taiwan and the US. Dyb et al. (80) used a simplified case definition of > 5 days per week and “almost daily” use of analgesics. Lipton et al. (90) tried to minimize recall bias using headache diaries, although only half of the respondents managed to report information on at least 15 out of 30 days.

The prevalence reports from Norway and Taiwan were 0.2% and 0.3%, respectively (80,81). Both countries had equivalent studies in adult populations that reported higher proportions (1.0% in Nord-Trøndelag (91) and 1.1% in Taiwan) (73).

The study on adolescents from the US did not estimate prevalence of medication overuse associated with all types of CH. There were data specific for chronic migraine with overuse (0.96%) to which can be added overuse due to chronic TTH and other headaches (90).

Reports within countries

There were few attempts at repeat studies within a country. The studies from Spain looked at distinct populations that were closely situated (Camargo and Santoña) in Cantabria, northern Spain (71,77).

Norwegian researchers first collected data in 1995–1997 from all residents of Nord-Trøndelag County aged 20 years or older and reported a prevalence of 1%. Follow-up data were collected in 2006–2008. They were thus able to estimate the incidence of MOH at 0.72 per 1000 person-years (97). Another set of reports from Akershus County (82,83) were based on data collected in 2005 from among a narrower age group of 30- to 44-year-olds. The authors reported a prevalence of 1.7%.

The three studies from Taiwan focused on different age groups: adolescents 12 to 14 years, adults (mean 37 ± 15 years), and the elderly ≥65 years. Their prevalence reports were 0.3%, 1.1% and 1.0%, respectively (72,73,81).

The two studies from the US also looked at different age groups: 12–17 years (90) and 18–65 years (76).

Discussion

Case ascertainment in population-based studies

Diagnostic criteria that are useful in the clinic are not always applicable in population-based research. It has been pointed out that people with medication overuse associated with frequent headache should not be labeled as having MOH without careful evaluation. In a specialist clinic, the association can be further investigated: whether the medication overuse is the cause or the effect of the headache (15,103,104). In population-based studies, case ascertainment relies on self-reports based on validated questionnaires (98 –101,105,106) often with no examination by a neurologist or headache specialist.

Researchers investigating MOH epidemiology have had to contend with this challenge by adapting clinical criteria to community-based settings. We make no attempt to evaluate which case definition used in the past is more valid but only show the different ways by which diagnostic criteria have been applied in prevalence studies in various settings. As far as the authors of the studies were concerned, these were the most valid definitions at the time, despite limitations.

For example, the earlier criterion specifying improvement after detoxification required clinical intervention and a careful follow-up study. Da Silva et al. (87) discussed how following the original criteria was not an easy task (requiring in-person assessments, intervention and prospective follow-up), and concluded that not requiring detoxification was “more reasonable when diagnosing MOH.”

Another example is worsening of the headache during medication overuse (105). This is often interpreted as concurrent chronic headache and drug overuse over a recall period (107) without making the respondent decide on which came first, the worsening headache or the increase in drug intake.

The past experience of adapting existing criteria is instructive because it serves as a guide for how future studies can be conducted using ICHD-3 beta. Research teams undertaking epidemiologic studies are again faced with the challenge of interpreting the criteria for the community setting—how these can be used in questionnaires, telephone interviews, and face-to-face interviews by nonspecialists. It is important that the criteria be applied in similar ways in future epidemiologic studies so that results from different countries and settings can be comparable.

Criterion A in ICHD-3 beta reads as “Headache occurring on ≥15 days per month in a patient with a pre-existing headache disorder.” Translating this to a standard questionnaire is challenging because one must operationally define “pre-existing headache disorder” in a way that is uniformly understood by respondents and interviewers who are often not specialists.

Criterion C requires that other causes of headache be ruled out. This makes sense in the clinics but cannot be implemented in population-based studies unless the study is designed in two parts: first to identify cases with frequent headache and then to arrange for their evaluation by a specialist. Such an evaluation would involve history-taking, neurological examination and possibly laboratory tests or imaging studies. None of the studies reviewed described a protocol for radiologic examination or other laboratory tests for the purpose of excluding other headache diagnoses.

On comparisons over time

Changes in drug use profiles over time can be expected to affect the overall prevalence of MOH. A US study spanning 1990 to 2005 observed a decrease in overuse of ergot compounds, acetylsalicylic acid, paracetamol, butalbital and opioids, but an increase in overuse of triptans and nonsteroidal anti-inflammatory drugs (NSAIDs) (108). The authors attributed these developments to changes in the acute treatment of migraine (introduction of triptans in the early 1990s) and the increasing use of multiple medications.

In the present paper, we focused only on how changes in criteria over time might influence prevalence. Retrospective comparisons of diagnoses using ICHD-2 and CDH-1994 criteria have been performed by reviewing clinic records (109 –111) as well as population data (112) suggesting that ICHD criteria tend to give a higher prevalence compared to CDH-1994. With the removal of the detoxification requirement in ICHD-2R, it was expected that higher prevalence would be reported because “it would be assumed that all CDH with medication overuse is due to the medication overuse” (113).

Figure 2 shows that the wide range of prevalence data might not be as much an effect of the changing criteria, as it is an outcome of the increasing diversity of countries (towns, communities) that are now publishing their reports. Excluding the incidence/prevalence studies from Norway, the studies from 2000 onward came from countries that had not reported prevalence before. Even with the many changes to the definition of overuse, the results were still clustered around 1% to 2%, but the reports from China and South Korea were lower than 1%, and there were three reports that were much higher than 2%.

On global prevalence and comparing prevalence between countries

Several studies (86,95,96) referred to the questionnaire developed by Lifting The Burden: The Global Campaign against Headache (www.l-t-b.org). This effort to conduct prevalence studies in areas of the world with no (or little) data on headache prevalence is just beginning to bear fruit. The methodologies and validation studies described in separate papers (99,101,106,114) could well be replicated in other regions of the world with very little data on MOH.

There are still many unknowns even for the most basic epidemiologic questions on MOH. There were only three studies focused on adolescents. Studies with small sample sizes did not report prevalence for specific age groups, although the trend was a peak toward middle age. The usual observation is that MOH is much more common among females but this was not true in one study (Table 4).

Our analysis documents the paucity of studies in several regions. Of the 15 countries with prevalence reports for adults, six are in Western Europe. There were no MOH prevalence studies from Africa, South Asia, Southeast Asia, and the Western Pacific. The US studies did not present overall prevalence.

The studies are too few to allow generalizations on regional differences, and the effect of ethnicity and culture. In addition, there are commercial and economic conditions in different countries that produce specific medication-abuse profiles (115). In MOH, racial differences in headache susceptibility interact with differences in medication access and use. For example, Yu et al. postulated that the low usage of analgesics in China compared to Taiwan probably made MOH less likely even though race, environment or habitation were similar (95).

There appears to be an increasing interest in investigating MOH in general populations. One-third of the papers in this review were published in 2011–2012.

A new working definition

Based on the methods used in past MOH prevalence studies, we find that the new ICHD-3 beta criteria for MOH will be very useful for the conduct of future population-based studies. However, the criterion for headache frequency (Criterion A) would have to be modified.

ICHD-3 beta Criterion A reads: “Headache occurring on ≥15 days per month in a patient with a pre-existing headache disorder.” The phrase “pre-existing headache disorder” has not been used before, and was not part of previous validation studies. It would be easier to use the existing description of frequency and duration of CH (which now make up the new Criterion A for chronic migraine and chronic TTH) and establishes MOH as being a subset of CH. This also follows the convention of reporting MOH as a proportion of those with CH. Furthermore, the word “patient” should be removed because it is inappropriate for population-based studies. Even those respondents with the most frequent headaches are not necessarily headache “patients” if they self-medicate and have not consulted health care professionals about their headache.

Criterion B in ICHD-3 beta can be easily applied in population-based studies. It has some important changes compared to ICHD-2 R. The lower cut-off point for multiple drug intake (≥10 days/month) could probably increase prevalence. Criterion B now allows for the possibility of diagnosing MOH even though the names, types and doses of medications cannot be readily recalled or verified. This kind of recall problem is common in population-based studies.

Criterion C, on exclusion of other headache diagnoses, requires a two-phase design (screening then clinical examination). This is difficult to implement in population-based studies that do not have an appropriate complement of expertise or a set-up for ancillary examinations. Expert diagnosis is the gold standard, but given limited resources for epidemiologic studies, adhering to Criterion C will likely lead to use of smaller sample sizes, which might then influence generalizability of the findings.

With respect to ICHD-3 beta, we suggest that the following criteria be used to estimate MOH prevalence in population-based studies:

Headache occurring on ≥15 days/month, on average, over at least three months.

Regular overuse for at least three months of one or more medicines taken for acute treatment of headache:

Intake of ergotamine, triptans, opioids or combination analgesics on ≥10 days/month; or

Intake of simple analgesics on ≥15 days/month; or

Intake of any combination of multiple drugs without overuse of any single drug or drug class alone on a total of ≥10 days/month; or

Intake of any combination of drugs from multiple drug classes on ≥10 days/month even though the identity, quantity and/or pattern of use or overuse of these classes of drug cannot be reliably established.

Several studies used a one month medication recall period, although three months is ideal. We propose that Criterion C not be used for case ascertainment in population-based studies. Exclusion of other headache diagnoses cannot be easily implemented and should not hinder efforts to estimate prevalence especially in large-scale studies.

Methodological considerations and limitations

The search for literature in this review considered the trade-offs between rigor and relevance (66). The MeSH search was conducted using top-level keywords because of the many possible synonyms for MOH. This method increased sensitivity but also decreased precision (116). The hundreds of returned entries increased time requirements for review of abstracts and full papers. The process of sorting these papers was tedious, and we may have missed studies because of human error. Data on MOH prevalence could have been missed in the broad screening of titles and abstracts if papers appeared to be focused on only one headache type, and it was not the main objective of the study to estimate MOH prevalence overall.

The two posters from neurology conferences were identified through review papers (Stark et al. and Scher et al.). We may have missed other conference abstracts because these would not be indexed in PubMed or MeSH.

In order to test the reproducibility of the search method, we repeated the free text keyword search “medication overuse headache” AND “prevalence” six months after the initial search. This produced 130 instead of 133 citations because of improvements performed by PubMed: A double entry was removed and two articles were correctly labeled as reviews.

Our search conducted in April–May 2013 included e-publications ahead of print in 2012. These articles may have been later indexed as publications printed in 2013.

It was not the objective of the study to compute an overall estimate of the global prevalence of MOH, nor to assign weights to the prevalence reports. However, we included raw data on the number of respondents in Table 4 as an indicator of the statistical power of the studies.

It is unlikely that publication bias (selective reporting of completed studies) was a major factor in this systematic review. The greater challenge in this field lies in encouraging headache epidemiology studies in countries with no previous reports on headache prevalence. There are several other ongoing national studies (Timothy J. Steiner, personal communication, 11 March 2013) and results are eagerly anticipated.

There were studies on CH that unfortunately did not go the extra step of investigating the possibility of concurrent medication overuse. This was not routinely performed in the early epidemiologic studies when MOH was not yet a well-defined diagnostic entity. It is unlikely that this lack of information was due to a selective reporting of outcomes.

Conclusions

Major changes in diagnostic criteria pose limitations to the comparison of MOH prevalence estimates over time. Among studies that used ICHD-2 R, reports ranged from 0.5% to 7.2%. There are still many knowledge gaps in estimating the global prevalence of MOH. The often-quoted global prevalence of 1% to 2% may be a gross underestimate in some regions and a slight overestimate in others. The observation that MOH is more predominant in females was not true in one study.

A common case definition of MOH for population-based prevalence studies allows greater comparability of results between countries and settings. This also sets the stage for future combined analyses and longitudinal studies, and ensures greater accuracy of estimates of global prevalence.

ICHD-3 beta Criteria A and B for the diagnosis of MOH can be easily applied in future prevalence studies but questionnaires would have to incorporate a definition for “pre-existing headache disorder.” A modified version of Criterion A that is consistent with the general description of CH (occurring on 15 or more days per month, on average, for at least three months) would be easier to implement.

Clinical implications

Diagnostic criteria for medication-overuse headache (MOH) that are useful in the clinic are not always applicable in population-based research.

Case definitions for MOH have changed markedly over time, posing limitations in comparing prevalence estimates.

Among studies that all used ICHD-2, there was a wide range of prevalence across countries (0.5%–7.2%). The often-quoted global prevalence of 1% to 2% may be a gross underestimate in some regions and an overestimate in others. MOH is most common among middle-aged adults in their 40 s to 50 s. MOH is more predominant in females except in one study.

ICHD-3 beta Criteria A and B for the diagnosis of MOH can be easily applied in future prevalence studies but questionnaires must incorporate a clearer definition for “pre-existing headache disorder.” A modified version that is consistent with Criterion A for chronic migraine and chronic tension-type headache would be easier to implement.

Criterion C (exclusion of other diagnoses) is not feasible in large epidemiologic studies.

Footnotes

Funding

MW received a research grant from IMK Almene Fond and the patient association Hovedpine og Migræneforeningen.

Conflict of interest

MW received travel grants from Pfizer and Lundbeck Fund to present results of related studies. RJ has given lectures for Pfizer, Berlin-Chemie, Allergan, Merck, and Autonomic Technologies; is a member of the advisory boards of Autonomic Technologies, Medotech, and Neurocore; and is co-director of Lifting The Burden. The other authors have nothing to declare.

References

Diener

Limmroth

. Medication-overuse headache: A worldwide problem. Lancet Neurol 2004; 3: 475–483.

Katsarava

Jensen

. Medication-overuse headache: Where are we now? Curr Opin Neurol 2007; 20: 326–330.

Evers

Marziniak

. Clinical features, pathophysiology, and treatment of medication overuse headache. Lancet Neurol 2010; 9: 391–401.

Diener

Katsarava

Limmroth

. Headache attributed to a substance or its withdrawal. Handb Clin Neurol 2010; 97: 589–599.

Ad Hoc Committee on Classification of Headache. Classification of Headache. JAMA 1962; 179: 717–718.

Wilkinson

Introduction. In: Diener

Wilkingson

(eds). Drug-induced headache, Berlin: Springer-Verlag, 1988, pp. 1–2.

Headache Classification Committee of the International Headache Society. Classification and diagnostic criteria for headache disorders, cranial neuralgias and facial pain. Cephalalgia 1988; 8(Suppl 7): 1–96.

Silberstein

Lipton

Solomon

. Classification of daily and near-daily headaches: Proposed revisions to the IHS criteria. Headache 1994; 34: 1–7.

Headache Classification Subcommittee of the International Headache Society. The International Classification of Headache Disorders. Cephalalgia 2004; 24(Suppl 1): 1–160.

10.

Silberstein

Olesen

Bousser

. The International Classification of Headache Disorders, 2nd edition (ICHD-II)—revision of criteria for 8.2 Medication-overuse headache. Cephalalgia 2005; 25: 460–465.

11.

Olesen

Bousser

Diener

. New appendix criteria for a broader concept of chronic migraine. Cephalalgia 2006; 26: 742–746.

12.

Olesen

Steiner

Bousser

M-G

. Proposals for new standardized general diagnostic criteria for the secondary headaches. Cephalalgia 2009; 29: 1331–1336.

13.

Headache Classification Committee of the International Headache Society. The International Classification of Headache Disorders, 3rd edition (beta version). Cephalalgia 2013; 33: 629–808.

14.

Rasmussen

Jensen

Olesen

A population-based analysis of the International Headache Society Classification. In: Olesen

(eds). Headache classification and epidemiology, New York: Raven Press, 1994, pp. 69–76.

15.

Sun-Edelstein

Bigal

Rapoport

. Chronic migraine and medication overuse headache: Clarifying the current International Headache Society classification criteria. Cephalalgia 2008; 29: 445–452.

16.

Ferrari

Coccia

Sternieri

. Past, present, and future prospects of medication overuse headache classification. Headache 2008; 48: 1096–1102.

17.

Olesen

. Analgesic headache. BMJ 1995; 310: 479–480.

18.

Srikiatkhachorn

Maneesri

Govitrapong

. Derangement of serotonin system in migrainous patients with analgesic abuse headache: Clues from platelets. Headache 1998; 38: 43–49.

19.

Isler

. Treatment of analgesic-induced headache [in German]. Ther Umsch 1986; 43: 221–225.

20.

Edmeads

. Analgesic-induced headaches: An unrecognized epidemic. Headache 1990; 30: 614–615.

21.

Adelman

. Analgesic induced headache. N C Med J 1991; 52: 345–346.

22.

Kudrow

Possible mechanisms and treatment of analgesic-induced chronic headache. In: Diener

Wilkinson

(eds). Drug-induced headache, Berlin: Springer-Verlag, 1988.

23.

Diener

Dichgans

Scholz

. Analgesic-induced chronic headache: Long-term results of withdrawal therapy. J Neurol 1989; 236: 9–14.

24.

Rapoport

. Analgesic rebound headache. Headache 1988; 28: 662–665.

25.

Dichgans

Diener

Gerber

. Chronic analgesic-induced headache [in German]. Dtsch Med Wochenschr 1984; 109: 369–373.

26.

Fumal

Laureys

Di Clemente

. Orbitofrontal cortex involvement in chronic analgesic-overuse headache evolving from episodic migraine. Brain 2006; 129: 543–550.

27.

Pini

Cicero

Sandrini

. Long-term follow-up of patients treated for chronic headache with analgesic overuse. Cephalalgia 2001; 21: 878–883.

28.

Young

Silberstein

. Long-term follow-up of patients treated for chronic headache with analgesic overuse. Cephalalgia 2001; 21: 873–873.

29.

Williams

Stark

. Intravenous lignocaine (lidocaine) infusion for the treatment of chronic daily headache with substantial medication overuse. Cephalalgia 2003; 23: 963–971.

30.

Linton-Dahlöf

Linde

Dahlöf

. Withdrawal therapy improves chronic daily headache associated with long-term misuse of headache medication: A retrospective study. Cephalalgia 2000; 20: 658–662.

31.

Antonaci

. Drug abuse headache: Recognition and management. Cephalalgia 1998; 18(Suppl 22): 47–52.

32.

Diener

. A personal view of the classification and definition of drug dependence headache. Cephalalgia 1993; 13 (Suppl 12): 68–71.

33.

Bekkelund

Salvesen

. Drug-associated headache is unrecognized in patients treated at a neurological centre. Acta Neurol Scand 2002; 105: 120–123.

34.

Silverman

. Behavioural changes in rats suggesting drug-induced headache [proceedings]. Br J Pharmacol 1976; 58: 439P–440P.

35.

Allain

Weintraub

. Drug induced headache. Ration Drug Ther 1980; 14: 1–6.

36.

Diener

Wilkinson

. Drug-induced headache, Berlin: Springer-Verlag, 1988.

37.

Silberstein

. Drug-induced headache. Neurol Clin 1998; 16: 107–123.

38.

Spierings

. Prednisone as initial treatment of drug-induced daily headache. Cephalalgia 2000; 20: 748–748.

39.

Tribl

Schnider

Wöber

. Are there predictive factors for long-term outcome after withdrawal in drug-induced chronic daily headache? Cephalalgia 2001; 21: 691–696.

40.

Mathew

Kurman

Perez

. Drug induced refractory headache—clinical features and management. Headache 1990; 30: 634–638.

41.

Moore

Noble

. Drug treatment of migraine: Part I. Acute therapy and drug-rebound headache. Am Fam Physician 1997; 56: 2039–2048, 2051–2054.

42.

Askmark

Lundberg

Olsson

. Drug-related headache. Headache 1989; 29: 441–444.

43.

Diener

Tfelt-Hansen

Headache associated with chronic use of substances. In: Olesen

Tfelt-Hansen

Welch

KMA

(eds). The headaches, New York: Raven Press Ltd, 1993, pp. 721–727.

44.

Diener

Dahlof

Headache associated with chronic use of substances. In: Olesen

Tfelt-Hansen

Welch

KMA

(eds). The headaches, Philadelphia: Lippincott Williams & Wilkins, 1999, pp. 871–878.

45.

Panconesi

Del Bianco

Curradi

. Headache caused by analgesic and/or ergotamine abuse [in Italian]. Clin Ter 1991; 137: 169–183.

46.

Parajuá Pozo

Ballesteros Martínez

. Headache caused by analgesic abuse [in Spanish]. Aten Primaria 1994; 14: 547–549.

47.

Lantéri-Minet

. Headaches caused by analgesic abuse [in French]. Pathol Biol (Paris) 2000; 48: 707–714.

48.

Parajua Pozo

Serra Mestres

López Pousa

. Headache caused by ergotamine addiction [in Spanish]. Rev Clin Esp 1991; 189: 347–347.

49.

Peters

Horton

. Headache; with special reference to the excessive use of ergotamine tartrate and dihydroergotamine. J Lab Clin Med 1950; 36: 972–973.

50.

Feoktistov

Filatova

Vei˘n

. Medication abuse-related headache [in Russian]. Zh Nevrol Psikhiatr Im S S Korsakova 1999; 99: 58–61.

51.

Hering

Glover

Pattichis

. 5HT in migraine patients with medication-induced headache. Cephalalgia 1993; 13: 410–412.

52.

Zed

Loewen

Robinson

. Medication-induced headache: Overview and systematic review of therapeutic approaches. Ann Pharmacother 1999; 33: 61–72.

53.

Mathew

. Medication misuse headache. Cephalalgia 1998; 18(Suppl 21): 34–36.

54.

Klapper

. Denial of hospitalization by insurers for inpatient treatment of medication rebound headaches. Headache 1994; 34: 601–602.

55.

Evans

. Warning: The Excedrin migraine warning label is inadequate to warn consumers of the risk of medication rebound headache. Headache 1999; 39: 679–680.

56.

Silberstein

Welch

. Painkiller headache. Neurology 2002; 59: 972–974.

57.

Page

. ‘Rebound headache’ from ergotamine withdrawal. JAMA 1981; 246: 719–719.

58.

Couch

. Rebound-withdrawal headache (medication overuse headache). Curr Treat Options Neurol 2006; 8: 11–19.

59.

Andersson

. Ergotamine headache. Headache 1975; 15: 118–121.

60.

Rowsell

Neylan

Wilkinson

. Ergotamine induced headaches in migrainous patients. Headache 1973; 13: 65–67.

61.

Shea

. Drug- and caffeine-induced headaches. CMAJ 1997; 157: 510–510.

62.

Göbel

Stolze

Heinze

. Easy therapeutical management of sumatriptan-induced daily headache. Neurology 1996; 47: 297–298.

63.

Shorofsky

Lamm

. Caffeine-withdrawal headache and fasting. N Y State J Med 1977; 77: 217–218.

64.

Katsarava

Fritsche

Muessig

. Clinical features of withdrawal headache following overuse of triptans and other headache drugs. Neurology 2001; 57: 1694–1698.

65.

Liberati

Altman

Tetzlaff

. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: Explanation and elaboration. PLoS Med 2009; 6: e1000100–e1000100.

66.

Booth

. How much searching is enough? Comprehensive versus optimal retrieval for technology assessments. Intl J of Technology Assess Health Care 2010; 26: 431–435.

67.

Stovner

Hagen

Jensen

. The global burden of headache: A documentation of headache prevalence and disability worldwide. Cephalalgia 2007; 27: 193–210.

68.

Scher

Midgette

Lipton

. Risk factors for headache chronification. Headache 2008; 48: 16–25.

69.

Stovner

Andree

. Prevalence of headache in Europe: A review for the Eurolight project. J Headache Pain 2010; 11: 289–299.

70.

Stark

Ravishankar

Siow

. Chronic migraine and chronic daily headache in the Asia-Pacific region: A systematic review. Cephalalgia 2013; 33: 266–283.

71.

Castillo

Muñoz

Guitera

. Kaplan Award 1998. Epidemiology of chronic daily headache in the general population. Headache 1999; 39: 190–196.

72.

Wang

Fuh

. Chronic daily headache in Chinese elderly: Prevalence, risk factors, and biannual follow-up. Neurology 2000; 54: 314–319.

73.

Fuh

Chen

. Chronic daily headache in Taipei, Taiwan: Prevalence, follow-up and outcome predictors. Cephalalgia 2001; 21: 980–986.

74.

Prencipe

Casini

Ferretti

. Prevalence of headache in an elderly population: attack frequency, disability, and use of medication. J Neurol Neurosurg Psychiatry 2001; 70: 377–381.

75.

Scher

Stewart

Lipton

. Is analgesic use a risk factor for chronic daily headache? Neurology 2001; 56: A312–A312.

76.

Scher

Lipton

Stewart

. Patterns of medication use by chronic and episodic headache sufferers in the general population: Results from the frequent headache epidemiology study. Cephalalgia 2010; 30: 321–328.

77.

Colás

Muñoz

Temprano

. Chronic daily headache with analgesic overuse: Epidemiology and impact on quality of life. Neurology 2004; 62: 1338–1342.

78.

Zwart

Dyb

Hagen

. Analgesic overuse among subjects with headache, neck, and low-back pain. Neurology 2004; 62: 1540–1544.

79.

Wiendels

Knuistingh Neven

Rosendaal

. Chronic frequent headache in the general population: Prevalence and associated factors. Cephalalgia 2006; 26: 1434–1442.

80.

Dyb

Holmen

Zwart

. Analgesic overuse among adolescents with headache: The Head-HUNT-Youth Study. Neurology 2006; 66: 198–201.

81.

Wang

Fuh

. Chronic daily headache in adolescents. Prevalence, impact, and medication overuse. Neurology 2006; 66: 193–197.

82.

Aaseth

Grande

Kværner

. Prevalence of secondary chronic headaches in a population-based sample of 30–44-year-old persons. The Akershus study of chronic headache. Cephalalgia 2008; 28: 705–713.

83.

Aaseth

Grande

Lundkvist

. What is chronic headache in the general population? The Akershus study of chronic headache. Acta Neurol Scand 2009; 120(Suppl 189): 30–32.

84.

Rueda-Sánchez

Díaz-Martínez

. Prevalence and associated factors for episodic and chronic daily headache in the Colombian population. Cephalalgia 2008; 28: 216–225.

85.

Straube

Pfaffenrath

Ladwig

. Prevalence of chronic migraine and medication overuse headache in Germany—the German DMKG headache study. Cephalalgia 2010; 30: 207–213.

86.

Katsarava

Dzagnidze

Kukava

. Primary headache disorders in the Republic of Georgia: Prevalence and risk factors. Neurology 2009; 73: 1796–1803.

87.

Da Silva

Jr Costa

Gomes

. Chronic headache and comorbibities: A two-phase, population-based, cross-sectional study. Headache 2010; 50: 1306–1312.

88.

Jonsson

Hedenrud

Linde

. Epidemiology of medication overuse headache in the general Swedish population. Cephalalgia 2011; 31: 1015–1022.

89.

Jonsson

Linde

Hensing

. Sociodemographic differences in medication use, health-care contacts and sickness absence among individuals with medication overuse headache. J Headache Pain 2012; 13: 281–290.

90.

Lipton

Manack

Ricci

. Prevalence and burden of chronic migraine in adolescents: Results of the chronic daily headache in adolescents study (C-dAS). Headache 2011; 51: 693–706.

91.

Linde

Stovner

Zwart

. Time trends in the prevalence of headache disorders. The Nord-Trondelag Health Studies (HUNT 2 and HUNT 3). Cephalalgia 2011; 31: 585–596.

92.

Chu

Kim

B-K

. Prevalence and impact of chronic daily headache in Korea: Results from Korean headache survey. Cephalalgia 2011; 31 (Suppl 1): 138–139.

93.

Shahbeigi

Fereshtehnejad

Mohammadi

. Epidemiology of headaches in Tehran urban area: A population-based cross-sectional study in district 8, year 2010. Neurol Sci 2013; 34: 1157–1166.

94.

Ertas

Baykan

Orhan

. One-year prevalence and the impact of migraine and tension-type headache in Turkey: A nationwide home-based study in adults. J Headache Pain 2012; 13: 147–157.

95.

Liu

Zhao

. The prevalence and burden of primary headaches in China: A population-based door-to-door survey. Headache 2012; 52: 582–591.

96.

Ayzenberg

Katsarava

Sborowski

. The prevalence of primary headache disorders in Russia: A countrywide survey. Cephalalgia 2012; 32: 373–381.

97.

Hagen

Linde

Steiner

. Risk factors for medicine over-use headache: An 11-year follow-up study. The Nord-Trøndelag Health Studies. Pain 2012; 153: 56–61.

98.

Hagen

Zwart

Vatten

. Head-HUNT: Validity and reliability of a headache questionnaire in a large population-based study in Norway. Cephalalgia 2000; 20: 244–251.

99.

Katsarava

Kukava

Mirvelashvili

. A pilot methodological validation study for a population-based survey of the prevalences of migraine, tension-type headache and chronic daily headache in the country of Georgia. J Headache Pain 2007; 8: 77–82.

100.

Hagen

Zwart

Aamodt

. The validity of questionnaire-based diagnoses: The third Nord-Trøndelag Health Study 2006–2008. J Headache Pain 2010; 11: 67–73.

101.

Ayzenberg

Katsarava

Mathalikov

. The burden of headache in Russia: Validation of the diagnostic questionnaire in a population-based sample. Eur J Neurol 2011; 18: 454–459.

102.

Limmroth

Katsarava

Fritsche

. Features of medication overuse headache following overuse of different acute headache drugs. Neurology 2002; 59: 1011–1014.

103.

Maizels

Burchette

. Rapid and sensitive paradigm for screening patients with headache in primary care settings. Headache 2003; 43: 441–450.

104.

Haag

. Medication overuse headache—comments on the current International Headache Society classification criteria. Cephalalgia 2010; 30: 1409–1411.

105.

Abrignani

Ferrante

Castellini

. Description and validation of an Italian ICHD-II-based questionnaire for use in epidemiological research. Headache 2012; 52: 1262–1282.

106.

Rao

Kulkarni

Gururaj

. The burden of headache disorders in India: Methodology and questionnaire validation for a community-based survey in Karnataka State. J Headache Pain 2012; 13: 543–550.

107.

Russell

Lundqvist

. Prevention and management of medication overuse headache. Curr Opin Neurol 2012; 25: 290–295.

108.

Meskunas

Tepper

Rapoport

. Medications associated with probable medication overuse headache reported in a tertiary care headache center over a 15-year period. Headache 2006; 46: 766–772.

109.

Bigal

Tepper

Sheftell

. Chronic daily headache: Correlation between the 2004 and 1998 International Headache Society Diagnostic Criteria. Headache 2004; 44: 684–691.

110.

Bigal

Rapoport

Tepper

. The classification of chronic daily headache in adolescents—a comparison between the second edition of the international classification of headache disorders and alternative diagnostic criteria. Headache 2005; 45: 582–589.

111.

Zeeberg

Olesen

Jensen

. Medication overuse headache and chronic migraine in a specialized headache centre: Field-testing proposed new appendix criteria. Cephalalgia 2009; 29: 214–220.

112.

Wang

Fuh

. Medication overuse headache in Taiwan. Cephalalgia 2008; 28: 1240–1242.

113.

Midgette

Scher

. The epidemiology of chronic daily headache. Curr Pain Headache Rep 2009; 13: 59–63.

114.

Cao

Zhao

. The burden of headache in China: Validation of diagnostic questionnaire for a population-based survey. J Headache Pain 2011; 12: 141–146.

115.

Vincent

. Controversy over the classification of medication overuse headache. Curr Pain Headache Rep 2012; 16: 80–85.

116.

Sampson

McGowan

Cogo

. An evidence-based practice guideline for the peer review of electronic search strategies. J Clin Epidemiol 2009; 62: 944–952.

Definitions of medication-overuse headache in population-based studies and their implications on prevalence estimates: A systematic review

Abstract

Background

Methods

Results

Conclusions

Keywords

Introduction

Methods

Results

Diagnostic criteria

Prevalence in adults and the elderly

Prevalence in young people

Reports within countries

Discussion

Case ascertainment in population-based studies

On comparisons over time

On global prevalence and comparing prevalence between countries

A new working definition

Methodological considerations and limitations

Conclusions

Clinical implications

Footnotes

Funding

Conflict of interest

References