Factors associated with a negative outcome of medication-overuse headache: A 3-year follow-up (the ‘CARE’ protocol)

Abstract

Aim

To evaluate factors associated with a negative outcome in a 3-year follow-up of subjects diagnosed with medication-overuse headache (MOH) (revised-ICHD-II criteria).

Methods

All consecutive patients entering the center’s inpatient detoxification program were analyzed in a prospective, non-randomized fashion. All participants were assessed by a neurologist using an ad hoc patient record form. Personality was assessed using the Minnesota Multiphasic Personality Inventory (MMPI)-2, Chi-square test, one-way analysis of variance (ANOVA), and odds ratios (OR) were calculated as appropriate.

Results

One-hundred and fifty patients completed the follow-up (79.3% females, age 46.40 ± 11.31 years): 13 never stopped their drug overuse (A), 38 stopped their overuse, but relapsed at least once (B), and 99 stopped and never relapsed (C). The Group A patients differed from those in B + C as they were more frequently single (OR 0.134; p = 0.007) and unemployed (OR 3.273; p = 0.04), took a higher number of drug doses (p < 0.001), and less frequently drank coffee (OR 3.273; p = 0.044). Personality profile: subjects in A scored higher than those in C on the following scales: Hypochondriasis (p = 0.007), Depression (p = 0.003), Paranoia (p = 0.025), Fears (p = 0.003), Obsessiveness (p = 0.026), Bizarre Mentation (p = 0.046), Social Discomfort (p = 0.004), Negative Treatment Indicators (p = 0.040), Repression (p = 0.007), Overcontrolled Hostility (p = 0.040), Addiction Admission (p = 0.021), Social Responsibility (p = 0.039), and Marital Distress (p = 0.028).

Conclusion

Disease outcome in MOH patients is influenced negatively by overuse severity and by specific psychological and socio-economic variables. Other possible modifier factors were voluptuary habits.

Keywords

Medication-overuse headache prognosis follow-up personality

Introduction

Medication-overuse headache (MOH) has become one of the major challenges in the field of headache treatment. In Europe, up to 30% of patients attending tertiary care centers present MOH, and in the USA the figure is even higher, exceeding 50%. The prevalence of MOH in the general population is about 1–2% (1).

Debate on MOH, which is fierce and ongoing, mainly revolves around diagnosis, pathophysiological mechanisms, therapeutic strategies, factors related to the chronification process, and the risk of relapse after withdrawal of offending drug(s).

The International Classification of Headache Disorders, 2nd Edition (ICHD-II) (2) includes diagnostic criteria for MOH. This edition also introduced the sub-category of triptan-overuse headache. However, after the publication of the ICHD-II, a lively debate on the classification of MOH soon led to some major developments: elimination of headache characteristics from the diagnostic criteria, introduction of a new sub-type (MOH attributed to combination of acute medications) (3), and the proposal of criteria for a broader concept of chronic migraine and MOH (4). The debate on the nosographic framing of MOH continues and additional evidence is needed in order to define diagnostic criteria that can be readily used by researchers as well as clinicians (5).

The pathophysiological mechanisms of MOH are at best speculative, and several hypotheses have been advanced, ranging from involvement of central sensitization (6), genetic factors (7), and endocrine/neurotransmitter function (8,9) to altered orbitofrontal cortex metabolism (10). A role, albeit still far from clear, may be played by psychological factors, including psychiatric disorders (11,12), cognitive components (13), personality, and behaviors of dependence (14 –16).

More is known about the role of symptomatic drugs in the chronification process. In principle, all acute headache drugs may be associated with chronic headache, although it is not certain whether and how the headache characteristics of MOH are related to overused drug(s) (17 –19).

There is also debate over the optimal treatment for MOH, with options ranging from structured inpatient care programs to simple advice to withdraw the overused drug(s) (20,21). Some authors have suggested that the choice of treatment could be determined by stratification of MOH patients into simple and complex forms, on the basis of clinical and anamnestic features (21,22).

Prognosis after withdrawal therapy is another critical issue. The relapse rate 1 year after withdrawal ranges from 14% to 41% (23 –29), regardless of whether inpatient, outpatient, or advice-alone protocols were adopted (28). A relapse rate of 45% was found at 4-year follow-up (29), and of 39% in a 5-year follow-up study (30). Another 4-year follow-up study found a relapse rate of 34%, and reported a range of 21–60% in other follow-ups of 4 years or more (31). Crucially important, for clinical management as well as research purposes, is the analysis of factors predicting relapse of MOH, even though existing studies are difficult to compare because of differences in endpoints and other important aspects (32). In a previous 1-year follow-up study (26), we found intake of more than 30 doses/month, smoking, alcohol consumption, non-confirmation of MOH diagnosis 2 months after detoxification, and return to overused drug(s) to be predictors of relapse. Another 1-year follow-up study (33) identified other risk factors: the presence of migraine as the primary diagnosis, the occurrence of combined migraine and tension-type headache, and the use of analgesics as opposed to ergots or triptans. Another study (24) in patients with low medical needs, found the presence of migraine on more than eight headache days/month, a higher frequency of migraine after drug withdrawal, and a greater number of previous preventive treatments to be possible predictors of relapse.

The main aim of the present study was to evaluate the factors associated with a negative outcome in a 3-year follow-up of a population of subjects diagnosed with MOH.

Methods

Sample recruitment and treatment

This study was conducted at the headache center (a tertiary referral center) of the C. Mondino National Institute of Neurology Foundation in Pavia, Italy. All consecutive patients with MOH beginning (for the first time) the center’s inpatient detoxification program between May 2004 and October 2006 were analyzed in a prospective, non-randomized fashion. They were enrolled as outpatients and gave their verbal informed consent to undergo the protocol, which consisted of an inpatient detoxification treatment and three follow-up appointments over the following year. During the 2 months preceding the detoxification treatment, the patients kept a detailed headache diary (34), which was reviewed on their admission to hospital for detoxification. All the patients underwent the standard CARE withdrawal protocol, described in a previous study (35), and continued to keep a headache diary throughout the hospital stay and the 3-year follow-up. The diagnosis of MOH was made according to the revised ICHD-II criteria (4).

All the participants were assessed by the same neurologist using an ad hoc patient record form. A comprehensive personal and family history of each patient was taken, paying particular attention to the patient’s headache history and profile of drug use. Each subject was examined by a psychologist at the beginning of the study, by means of clinical interviews and the Structured Clinical Interview for DSM-IV Disorders (SCID-I) (36) to make diagnoses according to Axis I of the DSM-IV (37). Personality profile was assessed using the Minnesota Multiphasic Personality Inventory (MMPI-2) (38), which is the most widely used objective personality test. Patients are presented with defined and structured items to which they are required to respond as accurately and truthfully as possible, choosing from a limited number of possibilities. The questionnaire includes 567 statements and subjects have to answer ‘true’ or ‘false’ according to what is predominantly true or false for them. The statements pertain to heterogeneous topics ranging from somatic symptoms to sexuality, from family to religious issues, from culture to personal relationships, and so on. The test is designed for individuals aged 18 years and older.

The first 370 items are divided into 10 clinical scales (Hs: Hypochondriasis; D: Depression; Hy: Hysteria; Pd: Psychopathic Deviate; Mf: Masculinity–Femininity; Pa: Paranoia; Pt: Psychasthenia; Sc: Schizophrenia; Ma: Hypomania; Si: Social Introversion) and three validity scales (L: Lie; K: Correction; F: Frequency). Content scales contain clusters of items concerning the same psychological dimension and behavioral area (ANX: Anxiety; FRS: Fears; OBS: Obsessiveness; DEP: Depression; HEA: Health Concerns; BIZ: Bizarre Mentation; ANG: Anger; CYN: Cynicism; ASP: Antisocial Practices; TPA: Type A; LSE: Low Self-Esteem; SOD: Social Discomfort; FAM: Family Problems; WRK: Work Interference; TRT: Negative Treatment Indicators). In this study we also used the following supplementary scales to investigate broad personality characteristics: A: Anxiety; R: Repression; Es: Ego Strength; Do: Dominance; Re: Social Responsibility; Generalized Emotional Distress – Mt: College Maladjustment; PSPK: Post-Traumatic Stress Disorder; MDS: Marital Distress Scale; Behavioral Dyscontrol – Ho: Hostility; O-H: Overcontrolled Hostility; MAC-R: MacAndrew Alcoholism Scale-Revised; AAS: Addiction Admission Scale; APS: Addiction Potential Scale; and Gender Role – GM: Masculine; GF: Feminine.

During hospitalization, all the patients received educational and rehabilitation counseling, including instructions and treatment prescriptions for the post-discharge period (for details see (26)). On discharge, we asked patients to continue keeping the headache diary and to attend scheduled follow-up visits at 2, 6, 12, 18, 24, and 36 months post discharge. At each of these visits, the pattern of the headache reported by the patient was checked against the information reported in the diary. Two telephone contacts were also scheduled: the first a month after discharge and the second after 9 months.

Definition of outcomes

We considered the occurrence (or not) of at least one episode of relapse into drug overuse during the 3-year follow-up period.

On the basis of our findings, the population was subdivided into three groups:

Group A: patients who never stopped (over the observation period of 3 years) overusing drugs after the detoxification treatment;

Group B: patients who stopped their drug overuse following detoxification, but then relapsed at least once over the 3-year observation period;

Group C: patients who stopped their drug overuse following detoxification and never relapsed over the 3-year observation period.

We then compared the groups looking for possible predictors of:

the presence of at least one episode of relapse into drug overuse during the overall 3-year follow-up (analysis of this outcome included all the patients, namely Groups A + B + C);

the presence of ongoing drug overuse at the end of the third year of follow-up (in this case, patients who never stopped overusing drugs, namely Group A, were excluded from the analysis).

The following variables were analyzed as possible predictors of the above outcomes: gender, age, socio-demographic characteristics, alcohol consumption, coffee consumption, smoking, positive family history of drug abuse or positive family history of headache, presence of psychiatric comorbidity, past medical history, type of primary headache, type of overused drug(s), and pattern of overuse (days and doses per month, duration of overuse in months, duration of chronic headache in months). We also considered ‘confirmation of MOH at 2 months’, which corresponded to the presence of headache on <15 days/month in a patient no longer overusing the previously overused drug(s).

Personality characteristics, as assessed using MMPI-2, were also taken into account as possible predictors. Patients overusing more than one class of drug were labeled ‘polyabusers’. Patients not attending the scheduled follow-up appointments were considered drop-outs.

Statistics

Categorical variables were analyzed using the Chi-square test. For quantitative variables, statistical differences were analyzed using the one-way analysis of variance (ANOVA), after controlling for normal distribution; multiple comparisons were adjusted with Bonferroni’s correction. Crude odds ratios (OR) with their 95% confidence intervals (CI) were calculated for dichotomous outcomes in a univariate model. Adjusted OR with their 95% CI were calculated by means of binary logistic regression in a multivariate model. An alpha of 0.05 was used for all statistical tests.

Results

Of a total of 243 patients, 150 completed the follow-up at 3 years (79.3% females, age 46.40 ± 11.31). A, B, and C groups were all observed for 3 years. Of these, 13 patients never stopped their drug overuse (Group A), 38 patients stopped their drug overuse following detoxification, but then relapsed at least once during the 3 years (Group B), and 99 patients stopped their drug overuse as a consequence of the detoxification program and never relapsed (Group C) (Figure 1).

Figure 1.

General clinical pattern of the 3-year follow-up.

Differences between groups

The patients who never stopped their drug overuse (Group A) differed from the other two groups (B + C) in a number of respects: living arrangement (they were less likely to live with a partner and children: OR 0.134, 95% CI 0.031–0.580, p = 0.007), employment (they were more likely to be unemployed: OR 3.273, 95% CI 1.03–10.39, p = 0.04), coffee consumption (they were more likely not to drink coffee: OR 3.273 95% CI 1.03–10.39, p = 0.044), and monthly intake of symptomatic drug doses (they took a greater number of drug doses per month p < 0.001).

As regards personality profile assessed using the MMPI-2, the patients in Group A recorded higher scores on the Lie scale (p = 0.004) compared with all the other subjects (Groups B and C), and on the following scales compared with the patients who stopped their drug overuse and never relapsed (Group C): Frequency (p = 0.020), Hypochondriasis (p = 0.007), Depression (p = 0.003), Paranoia (p = 0.025), Fears (p = 0.003), Obsessiveness (p = 0.026), Bizarre Mentation (p = 0.046), Social Discomfort (p = 0.004), Negative Treatment Indicators (p = 0.040), Repression (p = 0.007), Overcontrolled Hostility (p = 0.040), Addiction Admission Scale (p = 0.021), Social Responsibility (p = 0.039), Marital Distress Scale (p = 0.028).

Comparison of patients who experienced at least one relapse after improving (Group B) with those who never relapsed (Group C) revealed a significant difference in confirmation of MOH at 2 months: specifically, Group B was more likely to include subjects in whom MOH could not be confirmed because of the persistence of a chronic headache pattern (OR 3.727, 95% CI 1.613–8.613, p = 0.002). Groups B and C also showed differences in employment status and moderate alcohol consumption (<0.5l): Group B was more likely to include unemployed patients (OR 4.05, 95% CI 1.790–9.161, p = 0.001) and moderate drinkers (OR 2.4, 95% CI 1.105–5.211, p = 0.027). On the MMPI scales, the Group B patients recorded higher scores only on College Maladjustment.

Predictors of relapse

In a univariate analysis, the following factors were found to be associated with a higher risk of recurrence of at least one episode of drug overuse during the 3-year follow-up: being unemployed (OR 4.050, 95% CI 1.790–9.161), drinking in moderation (<0.5l) (OR 2.4, 95% CI 1.105–5.211), having a positive family history of headache (OR 3.733, 95% CI 1.053–13.234), taking a higher number of drug doses per month (OR 1.018, 95% CI 1.003–1.033), recording a lower score on the MMPI Ego Strengthscale (OR 0.918, 95% CI 0.852–0.989), and higher scores on the following MMPI scales: Hypochondriasis (OR 1.145 95% CI 1.028–1.276), Depression (OR 1.145, 95% CI 1.048–1.251), Paranoia (OR 1.120, 95% CI 1.004–1.249), Psychasthenia (OR 1.076, 95% CI 1.014–1.142), Schizophrenia (OR 1.056, 95% CI 1.001–1.113), Social Introversion (OR 1.056, 95% CI 1.002–1.135), Anxiety (OR 1.28, 95% CI 1.023–1.243), Fears (OR 1.181, 95% CI 1.043–1.339), Health Concerns (OR 1.146, 95% CI 1.028–1.278), Work Interference (OR 1.093, 95% CI 1.003–1.190), College Maladjustment (OR 1.101, 95% CI 1.029–1.177) and Post-Traumatic Stress Disorder (OR 1.078, 95% CI 1.012–1.149). On the contrary, confirmation of MOH at 2 months emerged as a protective factor (OR 0.268, 95% CI 0.116–0.620) (Tables 1A and 1B).

Table 1A.

Crude OR of relapse for categorical variables during the 3-year follow-up.

		Not relapsed frequency (%)	Relapsed frequency (%)	Crude OR (95% CI)
Gender	Male	21 (70)	9 (30)	1.153 (0.474–2.806)
Gender	Female	78 (72.9)	29 (27.1)	Ref
Education	≤8 years	37 (68.5)	17 (31.5)	Ref
Education	>8 years	62 (74.7)	21 (25.3)	0.737 (0.345–1.573)
Marital status	Married	73 (71.6)	29 (28.4)	1.148 (0.480–2.744)
Marital status	Not married	26 (74.3)	9 (25.7)	Ref
Living arrangement	Partner	27 (81.8)	6 (18.2)	0.756 (0.199–2.865)
	Partner and children	55 (67.1)	27 (32.9)	1.699 (0.557–5.006)
	Other	17 (77.3)	5 (22.7)	Ref
Employed	Yes	81 (80.2)	20 (19.8)	Ref
Employed	No	18 (50)	18 (50)	4.050 (1.790–9.161)
Alcohol consumer	Yes	27 (60)	18 (40)	2.4 (1.105–5.211)
Alcohol consumer	No	72 (78.3)	20 (21.7)	Ref
Coffee drinker	Yes	82 (72.6)	31 (27.4)	0.918 (0.347–2.428)
Coffee drinker	No	17 (70.8)	7 (29.2)	Ref
Smoker	Yes	22 (64.7)	12 (35.3)	1.615 (0.703–3.712)
	No	77 (74.8)	26 (25.2)	Ref
	>20/day	4 (40)	6 (60)	4.5 (0.939–21.557)
	<20/day	18 (75)	6 (25)	Ref
Insomnia	Yes	56 (71.8)	22 (28.2)	1.056 (0.495–2.250)
Insomnia	No	43 (72.9)	16 (27.1)	Ref
Snoring	Yes	25 (65.8)	13 (34.2)	1.477 (0.657–3.321)
Snoring	No	71 (74.0)	25 (26)	Ref
Family history of drug abuse	Yes	10 (83.3)	2 (16.7)	0.495 (0.103–2.369)
Family history of drug abuse	No	89 (71.2)	36 (28.8)	Ref
Positive family history of headache	Yes	75 (68.2)	35 (31.8)	3.733 (1.053–13.234)
Positive family history of headache	No	24 (88.9)	3 (11.1)	Ref
Psychiatric disorders	Yes	23 (69.7)	10 (30.3)	1.180 (0.500–2.788)
Psychiatric disorders	No	76 (73.1)	28 (26.9)	Ref
Benzodiazepine treatment	Yes	10 (58.8)	7 (41.2)	2.010 (0.704–5.736)
Benzodiazepine treatment	No	89 (74.2)	31 (25.8)	Ref
SCID (total psychiatric disorders)	3	32 (74.4)	11 (25.6)	2.403 (0.265–21.721)
	2	33 (70.2)	14 (29.8)	2.966 (0.333–26.381)
	1	15 (75)	5 (25)	2.330 (0.228–23.853)
	0	7 (87.5)	1 (12.5)	Ref
	No	77 (74.0)	27 (26.0)	Ref
Primary headache diagnosis	Migraine with/without aura	74 (75.5)	24 (24.5)	0.579 (0.351–1.675)
Primary headache diagnosis	Migraine and tension-type headache	25 (64.1)	14 (35.9)	Ref
Type of medication overused	Ergotamines	4 (80)	1 (25)	0.361 (0.034–3.788)
	Triptans	32 (82.1)	7 (17.9)	0.316 (0.097–1.028)
	Opioids	3 (75.0)	1 (25.0)	0.481 (9.043–5.401)
	Analgesics	22 (68.8)	10 (31.3)	0.657 (0.212–2.036)
	Analgesics with barbiturates and/or caffeine	25 (71.4)	10 (28.6)	0.578 (0.188–1.775)
	Polyabusers	13 (59.1)	9 (40.9)	Ref
MOH confirmed at 60-day follow-up	Yes	82 (78.8)	22 (21.2)	0.268 (0.116–0.620)
MOH confirmed at 60-day follow-up	No	16 (50)	16 (50)	Ref

Significant OR are in bold.

MOH: medication-overuse headache; Ref: reference category; SCID: Structured Clinical Interview for DSM-IV Disorders.

Table 1B.

Crude OR of relapse during the 3-year follow-up for continuous variables.

	Not relapsed mean ± SD (N = 99)	Relapsed mean ± SD (N = 38)	Crude OR (95% CI)
Age	(99) 45.39 ± 10.79	(38) 49.03 ± 11.03	1.031 (0.996–1.068)
BMI	(99) 23.10 ± 3.51	(38) 23.92 ± 3.50	1.067 (0.961–1.185)
Age at onset	(99) 14.41 ± 6.98	(37) 15.24 ± 7.36	1.017 (0.964–1.072)
Drug doses per month	(99) 33.75 ± 21.40	(38) 44.84 ± 29.86	1.018 (1.003–1.033)
Days with intake/days with headache per month	(99) 21.78 ± 6.79	(38) 24.13 ± 7.15	1.052 (0.995–1.112)
Duration of abuse (months)	(99) 53.39 ± 62.82	(38) 57.76 ± 59.77	1.001 (0.995–1.007)
Duration of chronic headache (months)	(99) 64.65 ± 75.59	(38) 60.63 ± 61.04	0.999 (0.994–1.005)
MMPI-2 (L)	(61) 6.44 ± 2.64	(19) 7.21 ± 2.86	1.114 (0.917–1.353)
MMPI-2 (F)	(61) 6.38 ± 3.62	(19) 8.26 ± 5.66	1.104 (0.982–1.242)
MMPI-2 (K)	(61) 14.20 ± 4.81	(19) 12.84 ± 4.75	0.941 (0.842–1.052)
MMPI-2 (HS)	(61) 13.69 ± 5.16	(19) 17.53 ± 6.49	1.145 (1.028–1.276)
MMPI-2 (D)	(61) 25.80 ± 6.87	(19) 31.79 ± 6.47	1.145 (1.048–1.251)
MMPI-2 (HY)	(61) 27.69 ± 5.98	(19) 30.42 ± 5.70	1.081 (0.989–1.182)
MMPI-2 (PD)	(61) 17.15 ± 5.90	(19) 18.68 ± 6.89	1.040 (0.959–1.129)
MMPI-2 (MF)	(61) 31.11 ± 5.46	(19) 31.95 ± 4.02	1.034 (0.931–1.147)
MMPI-2 (PA)	(61) 9.66 ± 4.11	(19) 12.32 ± 6.01	1.120 (1.004–1.249)
MMPI-2 (PT)	(61) 15.05 ± 8.81	(19) 21.21 ± 9.55	1.076 (1.014–1.142)
MMPI-2 (SC)	(61) 14.20 ± 8.42	(19) 19.37 ± 12.01	1.056 (1.001–1.113)
MMPI-2 (MA)	(61) 15.11 ± 4.35	(19) 15.16 ± 4.84	1.002 (0.892–1.126)
MMPI-2 (SI)	(61) 29.98 ± 8.72	(19) 34.95 ± 9.66	1.056 (1.002–1.135)
MMPI-2 (ANX)	(61) 7.79 ± 5.25	(19) 11.53 ± 6.25	1.128 (1.023–1.243)
MMPI-2 (FRS)	(61) 6.62 ± 3.88	(19) 9.74 ± 4.91	1.181 (1.043–1.339)
MMPI-2 (OBS)	(61) 5.18 ± 3.38	(19) 6.58 ± 3.78	1.119 (0.967–1.295)
MMPI-2 (DEP)	(61) 9.10 ± 6.52	(19) 12.26 ± 7.48	1.067 (0.992–1.148)
MMPI-2 (HEA)	(61) 13-34 ± 5.09	(19) 16.95 ± 5.45	1.146 (1.028–1.278)
MMPI-2 (BIZ)	(61) 2.64 ± 2.57	(19) 4.11 ± 3.72	1.168 (0.989–1.380)
MMPI-2 (ANG)	(61) 5.69 ± 2.85	(19) 6.32 ± 3.20	1.077 (0.901–1.286)
MMPI-2 (CYN)	(61) 12.07 ± 5.35	(19) 13.68 ± 5.15	1.062 (0.959–1.177)
MMPI-2 (ASP)	(61) 6.89 ± 3.43	(19) 7.68 ± 2.77	1.079 (0.919–1.266)
MMPI-2 (TPA)	(61) 8.84 ± 3.18	(19) 10.58 ± 3.67	1.182 (0.998–1.401)
MMPI-2 (LSE)	(61) 6.30 ± 4.16	(19) 7.16 ± 3.89	1.052 (0.929–1.192)
MMPI-2 (SOD)	(61) 7.21 ± 4.31	(19) 9.42 ± 4.94	1.114 (0.992–1.252)
MMPI-2 (FAM)	(61) 5.33 ± 3.80	(19) 6.00 ± 5.13	1.040 (0.919–1.176)
MMPI-2 (WRK)	(61) 9.57 ± 5.82	(19) 13.00 ± 7.06	1.093 (1.003–1.190)
MMPI-2 (TRT)	(61) 7.41 ± 4.72	(19) 9.21 ± 4.28	1.086 (0.972–1.212)
MMPI-2 (A)	(61) 14.66 ± 8.25	(19) 18.37 ± 9.75	1.051 (0.989–1.117)
MMPI-2 (R)	(61) 18.03 ± 4.15	(19) 18.95 ± 3.72	1.058 (0.931–1.202)
MMPI-2 (ES)	(61) 31.74 ± 6.98	(19) 27.32 ± 7.25	0.918 (0.852–0.989)
MMPI-2 (MAC-R)	(61) 19.48 ± 3.31	(19) 19.37 ± 3.71	0.991 (0.850–1.155)
MMPI-2 (FB)	(61) 3.67 ± 3.30	(19) 5.53 ± 4.99	1.125 (0.989–1.279)
MMPI-2 (OH)	(61) 14.74 ± 3.02	(19) 14.11 ± 3.94	0.941 (0.802–1.104)
MMPI-2 (APS)	(61) 20.00 ± 3.47	(19) 20.26 ± 4.28	1.020 (0.885–1.175)
MMPI-2 (AAS)	(61) 1.57 ± 1.19	(19) 2.26 ± 2.38	1.306 (0.947–1.803)
MMPI-2 (DO)	(61) 15.11 ± 3.37	(19) 13.79 ± 2.48	0.880 (0.747–1.035)
MMPI-2 (RE)	(61) 20.80 ± 3.71	(19) 19.79 ± 3.54	0.930 (0.812–1.066)
MMPI-2 (MT)	(61) 15.23 ± 7.82	(19) 21.84 ± 9.23	1.101 (1.029–1.177)
MMPI-2 (GM)	(61) 27.80 ± 7.14	(19) 24.37 ± 7.48	0.937 (0.871–1.008)
MMPI-2 (GF)	(61) 34.25 ± 5.39	(19) 34.16 ± 4.59	0.997 (0.903–1.101)
MMPI-2 (PSPK)	(61) 11.85 ± 7.64	(19) 17.05 ± 9.24	1.078 (1.012–1.149)
MMPI-2 (MDS)	(61) 3.49 ± 2.56	(19) 4.26 ± 3.07	1.111 (0.919–1.342)

Significant OR are in bold.

BMI: body mass index; Hs: Hypochondriasis; D: Depression; Hy: Hysteria; Pd: Psychopathic Deviate; Mf: Masculinity–Femininity; Pa: Paranoia; Pt: Psychasthenia; Sc: Schizophrenia; Ma: Hypomania; Si: Social Introversion. Three validity scales – L: Lie; K: Correction; F: Frequency. Content scales have clusters of items relating to the same psychological dimension and behavioral area ANX: Anxiety; FRS: Fears; OBS: Obsessiveness; DEP: Depression; HEA: Health Concerns; BIZ: Bizarre Mentation; ANG: Anger; CYN: Cynicism; ASP: Antisocial Practices; TPA: Type A; LSE: Low Self-Esteem; SOD: Social Discomfort; FAM: Family Problems; WRK: Work Interference; TRT: Negative Treatment Indicators. A: Anxiety; R: Repression; Es: Ego Strength; Do: Dominance; Re: Social Responsibility; Generalized Emotional Distress Mt: College Maladjustment; PSPK: Post-Traumatic Stress Disorder; MDS: Marital Distress Scale; Behavioral Dyscontrol Ho: Hostility; O-H: Overcontroled Hostility; MAC-R: MacAndrew Alcoholism Scale-Revised; AAS: Addiction Admission Scale; APS: Addiction Potential Scale; Gender Role – GM: Masculine; GF: Feminine.

In a multivariate analysis, the factors found to predict a higher risk of recurrence of at least one episode of drug overuse during the 3-year follow-up were: monthly intake of drug doses (OR 1.028; 95% CI 1.005–1.051) and MMPI-2 Depression scale score (OR 1.155, 95% CI 1.045–1.278).

According to the results of a univariate analysis, the factors associated with a higher risk of the presence of drug overuse at the end of the third year of the follow-up were: being unemployed (OR 2.253, 95% CI 1.029–6.183), drinking in moderation (<0.5l) (OR 2.469, 95% CI 1.033–5.902), taking a higher number of drug doses per month (OR 1.028, 95% CI 1.011–1.045), having more days with headache per month (OR 1.105, 95% CI 1.030–1.186), and recording higher scores on the following MMPI-2 scales: Depression (OR 1.117, 95% CI 1.021–1.224), Psychasthenia (OR 1.072, 95% CI 1.006–1.142), Anxiety (OR 1.141, 95% CI 1.025–1.271) and College Maladjustment (OR 1.100, 95% CI 1.023–1.183). Once again, confirmation of MOH at two months was a protective factor (OR 0.400, 95% CI 0.160–1.002) (Tables 2A and 2B).

Table 2A.

Crude OR of the presence of overuse at the end of the third year of follow-up for categorical variables.

		Non-overusers frequency (%)	Overusers frequency (%)	Crude OR (95% CI)
Gender	Male	24 (80)	16 (20)	1.087 (0.393–3.010)
Gender	Female	87 (81.3)	20 (18.7)	Ref
Education	≤8 years	44 (81.5)	10 (18.5)	Ref
Education	>8 years	67 (80.7)	16 (19.3)	1.051 (0.437–2.525)
Marital status	Married	82 (80.4)	20 (19.6)	1.179 (0.431–3.223)
Marital status	Not married	29 (82.9)	6 (17.1)	Ref
Living arrangement	Partner	31 (93.9)	2 (6.1)	1.025 (0.334–3.147)
	Partner and children	63 (76.8)	19 (23.2)	0.219 (0.038–1.254)
	Other	17 (77.3)	5 (22.7)	Ref
Employed	Yes	86 (85.1)	15 (14.9)	Ref
Employed	No	25 (69.4)	11 (30.6)	2.253 (1.029–6.183)
Alcohol consumer	Yes	32 (71.1)	13 (28.9)	2.469 (1.033–5.902)
Alcohol consumer	No	79 (85.9)	13 (14.1)	Ref
Coffee drinker	Yes	93 (82.3)	20 (17.7)	0.645 (0.227–1.830)
Coffee drinker	No	18 (75)	6 (25)	Ref
Smoker	Yes	25 (73.5)	9 (26.5)	1.821 (0.724–4.581)
	No	86 (83.5)	17 (16.5)	Ref
	<20/day	18 (75)	6 (25)	Ref
	>20/day	7 (70)	3 (30)	1.286 (0.250–6.615)
Insomnia	Yes	62 (79.5)	16 (20.5)	1.264 (0.527–3.031)
Insomnia	No	49 (83.1)	10 (16.9)	Ref
Snoring	Yes	29 (76.3)	9 (23.7)	1.442 (0.579–3.594)
Snoring	No	79 (82.3)	17 (17.7)	Ref
Family history of alcohol abuse	Yes	5 (100)	—	—
Family history of alcohol abuse	No	106 (80.3)	26 (19.7)	Ref
Family history of drug abuse	Yes	12 (100)	—	—
Family history of drug abuse	No	99 (79.2)	26 (20.8)	Ref
Family history of headache	Yes	86 (78.2)	24 (21.8)	3.488 (0.771–15.783)
Family history of headache	No	25 (92.6)	2 (7.4)	Ref
Psychiatric diseases	Yes	26 (78.8)	7 (21.2)	1.204 (0.456–3.182)
Psychiatric diseases	No	85 (81.7)	19 (18.3)	Ref
Benzodiazepine treatment	Yes	13 (76.5)	4 (23.5)	1.371 (0.408–4.607)
Benzodiazepine treatment	No	98 (81.7)	22 (18.3)	Ref
SCID (total psychiatric disorders)	3	7 (87.5)	1 (12.5)	1.360 (0.144–12.838)
	2	38 (80.9)	9 (19.1)	1.656 (0.180–15.194)
	1	17 (85)	3 (15)	1.234 (0.109–13.980)
	0	7 (87.5)	1 (12.5)	Ref
Primary headache diagnosis	Migraine with/without aura	81 (82.7)	17 (17.3)	Ref
Primary headache diagnosis	Migraine and tension-type headache	30 (76.9)	9 (23.1)	1.429 (0.575–3.551)
Type of medication abused	Ergotamines	4 (80)	1 (20)	0.667 (0.061–7.230)
	Triptans	34 (87.2)	5 (12.8)	0.392 (0.104–1.479)
	Opioids	4 (100)	—	—
	Analgesics	26 (81.3)	6 (18.8)	0.615 (0.169–2.239)
	Analgesics with barbiturates and/or caffeine	27 (77.1)	8 (22.9)	0.790 (0.232–2.692)
	Polyabusers	16 (72.7)	6 (27.3)	Ref
MOH confirmed at 60-day follow-up	Yes	88 (84.6)	16 (15.4)	0.400 (0.160–1.002)
MOH confirmed at 60-day follow-up	No	22 (68.8)	10 (31.3)	Ref

Significant OR are in bold.

MOH: medication-overuse headache; Ref: reference category; SCID: Structured Clinical Interview for DSM-IV Disorders.

Table 2B.

Crude OR of presence of overuse at the end of the third year of follow-up for continuous variables.

	Non-overusers mean ± SD	Overusers mean ± SD	Crude OR (95% CI)
Age	(111) 46.10 ± 11.13	(26) 47.69 ± 10.18	1.013 (0.975–1.054)
BMI	(111) 23.43 ± 3.54	(26) 22.91 ± 3.41	0.957 (0.841–1.088)
Age at onset	(111) 14.75 ± 7.09	(26) 14.15 ± 7.11	0.988 (0.928–1.051)
Drug doses per month	(111) 33.23 ± 20.93	(26) 52.19 ± 31.93	1.028 (1.011–1.045)
Days with intake/days with headache per month	(111) 21.60 ± 6.86	(26) 25.96 ± 6.28	1.105 (1.030–1.186)
Duration of abuse (months)	(111) 53.56 ± 60.67	(26) 59.08 ± 67.51	1.001 (0.995–1.008)
Duration of chronic headache (months)	(111) 63.74 ± 72.46	(26) 62.65 ± 69.48	1.000 (0.994–1.006)
MMPI-2 (L)	(65) 6.60 ± 2.73	(15) 6.73 ± 2.60	1.019 (0.826–1.256)
MMPI-2 (F)	(65) 6.60 ± 3.82	(15) 7.80 ± 5.76	1.065 (0.940–1.207)
MMPI-2 (K)	(65) 14.12 ± 4.77	(15) 12.80 ± 4.96	0.942 (0.834–1.064)
MMPI-2 (HS)	(65) 14.14 ± 5.33	(15) 12.80 ± 4.96	1.085 (0.975–1.207)
MMPI-2 (D)	(65) 26.26 ± 6.99	(15) 31.40 ± 6.86	1.117 (1.021–1.224)
MMPI-2 (HY)	(65) 27.92 ± 5.94	(15) 30.13 ± 6.08	1.064 (0.968–1.170)
MMPI-2 (PD)	(65) 17.23 ± 6.07	(15) 18.73 ± 6.49	1.039 (0.952–1.134)
MMPI-2 (MF)	(65) 31.28 ± 5.36	(15) 31.47 ± 4.27	1.007 (0.902–1.125)
MMPI-2 (PA)	(65) 9.91 ± 4.36	(15) 11.93 ± 5.98	1.086 (0.972–1.214)
MMPI-2 (PT)	(65)15.40 ± 8.97	(15) 21.33 ± 9.55	1.072 (1.006–1.142)
MMPI-2 (SC)	(65) 14.69 ± 8.86	(15) 18.60 ± 12.02	1.041 (0.984–1.101)
MMPI-2 (MA)	(65) 15.18 ± 4.43	(15) 14.87 ± 4.63	0.984 (0.865–1.119)
MMPI-2 (SI)	(65) 30.62 ± 8.87	(15) 33.53 ± 10.22	1.037 (0.973–1.105)
MMPI-2 (ANX)	(65) 7.91 ± 5.37	(15) 12 ± 6.05	1.141 (1.025–1.271)
MMPI-2 (FRS)	(65) 7.02 ± 4.25	(15) 8.87 ± 4.45	1.100 (0.969–1.249)
MMPI-2 (OBS)	(65) 5.20 ± 3.42	(15) 6.87 ± 3.68	1.142 (0.974–1.339)
MMPI-2 (DEP)	(65) 9.32 ± 6.70	(15) 12.13 ± 7.23	1.058 (0.979–1.145)
MMPI-2 (HEA)	(65) 13.72 ± 5.21	(15) 16.27 ± 5.76	1.096 (0.982–1.224)
MMPI-2 (BIZ)	(65) 2.77 ± 2.69	(15) 3.93 ± 3.75	1.127 (0.948–1.341)
MMPI-2 (ANG)	(65) 5.71 ± 2.91	(15) 6.40 ± 3.07	1.085 (0.894–1.317)
MMPI-2 (CYN)	(65) 12.14 ± 5.26	(15) 13.80 ± 5.56	1.064 (0.951–1.191)
MMPI-2 (ASP)	(65) 6.91 ± 3.35	(15) 7.80 ± 2.96	1.088 (0.914–1.296)
MMPI-2 (TPA)	(65) 8.88 ± 3.18	(15) 10.87 ± 3.74	1.088 (0.914–1.296)
MMPI-2 (LSE)	(65) 6.45 ± 4.18	(15) 6.73 ± 3.81	1.017 (0.887–1.166)
MMPI-2 (SOD)	(65) 7.40 ± 4.32	(15) 9.20 ± 5.28	1.091 (0.964–1.235)
MMPI-2 (FAM)	(65) 5.28 ± 3.74	(15) 6.40 ± 5.59	1.066 (0.934–1.217)
MMPI-2 (WRK)	(65) 9.85 ± 6.00	(15) 12.73 ± 7.04	1.076 (0.983–1.178)
MMPI-2 (TRT)	(65) 7.58 ± 4.71	(15) 8.93 ± 4.40	1.063 (0.944–1.197)
MMPI-2 (A)	(65) 14.83 ± 8.46	(15) 18.60 ± 9.43	1.052 (0.985–1.123)
MMPI-2 (R)	(65) 18.06 ± 4.11	(15) 19.07 ± 3.77	1.063 (0.925–1.222)
MMPI-2 (ES)	(65) 31.25 ± 7.27	(15) 28.27 ± 6.85	0.946 (0.876–1.021)
MMPI-2 (MAC-R)	(65)19.54 ± 3.37	(15) 19.07 ± 3.53	0.959 (0.809–1.136)
MMPI-2 (FB)	(65) 3.92 ± 3.76	(15) 4.93 ± 4.10	1.066 (0.930–1.223)
MMPI-2 (OH)	(65) 14.69 ± 3.01	(15) 14.13 ± 4.19	0.948 (0.797–1.128)
MMPI-2 (APS)	(65) 19.89 ± 3.44	(15) 20.80 ± 4.51	1.071 (0.917–1.250)
MMPI-2 (AAS)	(65) 1.58 ± 1.24	(15) 2.40 ± 2.50	1.357 (0.963–1.911)
MMPI-2 (DO)	(65) 14.95 ± 3.39	(15) 14.13 ± 2.95	0.925 (0.778–1.053)
MMPI-2 (RE)	(65) 20.82 ± 3.66	(15) 19.47 ± 3.64	0.911 (0.788–1.053)
MMPI-2 (MT)	(65) 15.55 ± 7.95	(15) 22.20 ± 9.47	1.100 (1.023–1.183)
MMPI-2 (GM)	(65) 27.28 ± 7.48	(15) 25.73 ± 6.66	0.972 (0.900–1.049)
MMPI-2 (GF)	(65) 34.45 ± 5.29	(15) 33.27 ± 4.73	0.961 (0.869–1.062)
MMPI-2 (PSPK)	(65) 12.31 ± 8.07	(15) 16.47 ± 8.68	1.061 (0.992–1.134)
MMPI-2 (MDS)	(65) 3.52 ± 2.56	(15) 4.33 ± 3.20	1.115 (0.909–1.368)

Significant OR are in bold.

Finally, in a multivariate analysis, the following factors emerged as predictors of the presence of drug overuse at the end of the third year of the follow-up: monthly intake of drug doses (OR 1.040; 95% CI 1.014–1.067) and College Maladjustment score (OR 1.100, 95% CI 1.017–1.189).

Discussion

The identification of risk factors predicting relapse of medication overuse and headache in patients with MOH is critically important because of its clinical and pathophysiological implications. In the present study, we evaluated risk factors for two different types of outcome: one more ‘dynamic’ (recurrence of episodes of drug abuse during the entire 3-year follow-up), and the other more precise (presence of principal drug abuse at the end of the third year of follow-up).

In multivariate analyses, factors related to disease severity (higher intake of drug doses per month) and to psychopathological elements (higher Depression scores) were associated with the recurrence of at least one episode of drug overuse during the overall study period, whereas factors related to disease severity (intake of a higher number of drug doses per month) and socio-economic status (higher College Maladjustment scores) were associated with the presence of drug overuse at the end of the third year.

The stratification of the patients by overuse behavior (patients who never stopped their drug overuse, those who stopped their overuse but relapsed at least once, and those who stopped overusing and never relapsed) provided some important findings. The patients who never stopped their drug overuse showed more psychosocial problems (being out of work, not living with partner or children) and more pathological personality characteristics (i.e. as measured on the Depression, Hypochondriasis, Paranoia, Fears, Social Discomfort, Obsessiveness scales) compared with the other two groups. These patients did not experience remission of headache after the detoxification treatment, but it is interesting to note that they took a higher number of doses (>45) of acute drug(s) per month than the other patients did. Therefore, it is possible that they failed to stop overusing simply because they had a more severe drug dependence. The fact that these subjects recorded significantly higher scores on the Addiction Admission Scale supports this hypothesis. In previous studies in which we considered MMPI-2 scale scores (14,39) – and our findings are at variance with the recent literature on this topic (15,16) – we did not find any evidence supporting the presence of personality characteristics related to addiction in MOH patients. Our new data, on the other hand, seem to point to the existence of a small subgroup of patients with chronic headache and symptomatic medication overuse in whom addiction-related personality and behavior may strongly influence (foster) the development and maintenance of the drug overuse and chronic headache. In agreement with Lake (22), we therefore suggest that patients with complicated MOH exist and need therapy that is tailored, using an approach previously outlined (26), to their specific clinical needs.

The higher intake of acute drugs observed in our group of relapsers may theoretically reflect a less adequate control of the attacks, with the need to take more drugs to get a sustained pain-free response, as suggested by some authors (40). In this frame, the observed tendency toward a better outcome in triptan users might reflect a more effective control of attacks with these specific antimigraine drugs. However, it is also true that triptans are associated with a quite high rate of recurrence (41), therefore further studies specifically designed to address this issue are needed.

Noteworthy, any single psychiatric disorder (DSM-IV criteria) did not predict the prognosis of MOH. On the contrary, personality dysfunctional characteristics as Depression did. The association of MOH with psychiatric comorbidity (mainly depression and anxiety disorders) has been frequently reported (42) and it has been suggested to play a role in the chronification process (12,42). However, in the present, prospective study, we did not find any negative predictive value for psychiatric comorbidity, not even when considering multiple comorbid psychiatric disorders. Therefore, it is possible that psychiatric comorbidity plays a role in the transformation of an episodic primary headache into MOH, while it might have a limited role in the relapse back to overuse once the patient is adequately diagnosed, treated, and followed-up. Further studies are needed to test this hypothesis.

In our search for likely specific predictors of relapse we decided to consider the entire study period, as well as the end of the third year, because we felt that the whole period would probably provide more accurate information about the true pattern of MOH.

Our findings showed that not being in regular employment, drinking in moderation, taking a higher number of symptomatic drug doses, and on a higher number of days per month, having a higher number of headache days per month, and having a positive family history of headache increased the risk of recurrence of an episode of drug overuse over the 3 years. Recurrence of overuse in association with an episodic pattern of headache (MOH confirmation) at 2 months post detoxification was instead found to be a protective factor. It is worth noting that the triptans, compared with the other classes of acute drugs, showed a trend toward a protective effect. In addition, according to 4-year (or more) follow-up studies, we know that remission at 1 year is a predictor of good prognosis in the general population (43) and that in the same population the use of non-steroidal anti-inflammatory drugs and/or triptans as opposed to ergots and/or opioids is associated with a better outcome.

In the present study, past medical history, BMI, smoking, and single psychiatric disorders did not show any significant association with the outcome measures.

As regards the psychological factors, numerous MMPI-2 scales were found to have significant negative prognostic value: Hypochondriasis, Depression, Paranoia, Psychasthenia, Schizophrenia, Social Introversion, Anxiety, Fears, Health Concerns, Work Interference, Ego Strength, College Maladjustment and Post-Traumatic Stress Disorder.

In a 4-year follow-up study (31) that evaluated psychiatric comorbidity as a possible outcome predictor, anxiety, and/or depression scores were found to have negative prognostic value (other non-significant outcome measures were sex, age, education level, employment status, numbers of years with headache, type of intervention, primary headache diagnosis, more than two types of intervention). Another long-term follow-up study (44) found psychiatric disorders to be negative predictors only in MOH and not in other forms of chronic headache. In our previous study (1-year follow-up) (34), we found only a trend toward significant prediction of MOH relapse. The present study instead supports the negative prognostic value of multiple psychiatric conditions, mainly in patients with several concomitant personality-related problems.

Our study confirmed the role of the number of acute drug doses as a significant predictor of prognosis (34), whereas, unlike other long-term follow-up studies (45,46), we found no predictive value for gender, type of overused drug(s), headache diagnosis at onset, age at headache onset, duration of medication overuse, or primary headache type (47,48). However, comparisons between studies are difficult because of differences in the endpoints used as well as in the samples analyzed, all of which precludes definitive conclusions. General consensus on this topic is needed in order to open up the way for more comparable follow-up studies in this important research field (49). We suggest that psychosocial variables should be included as outcome measures in future studies on risk of relapse in MOH. In our study, a negative prognosis seemed to be related not only to the characteristics of the headache (e.g. family recurrence) and overuse (e.g. number of doses, type of drug(s) overused), but also to the patient’s general circumstances (e.g. being unemployed or living alone) and psychological status (presence of risk factors for psychiatric and personality disorders).

In conclusion, the present study provides additional information on the factors that are associated with the outcome of MOH and highlights the importance of overall ‘CARE’ of the patient, in which clinical intervention is designed to address both medical and psychological problems. Furthermore, the present findings suggest that MOH is a complex disorder resulting from the mutual interplay of genetic, social, psychological, biological, and behavioral determinants. Its management should be approached accordingly.

Clinical implications

Not only drug abuse, but also psycho-social factors predict negative prognosis for MOH.

Personality profile (Hypochondriasis, Depression, Paranoia, Fears, Obsessiveness, Bizarre Mentation, Repression, Overcontrolled Hostility, Addiction Admission, Social Responsibility, and Marital Distress) predicted the worst outcome (never stopping of abuse).

Footnotes

Acknowledgements

This study has been realized in collaboration with the University Consortium for Adaptive Disorders and Head pain (UCADH).

Funding

This research received the funding of the Ministry of Health, Italy (RC 2008-2010).

Conflict of interest

None declared.

References

Katsarava

Jensen

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

Headache Classification Subcommittee of the International Headache SocietyThe International Classification of Headache Disorders: 2nd edition. Cephalalgia 2004; 24: 1–160.

Silberstein

Olesen

Bousser

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

Olesen

Bousser

Diener

. Headache Classification CommitteeNew appendix criteria open for a broader concept of chronic migraine. Cephalalgia 2006; 26: 742–746.

Evers

Marziniak

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

Ayzenberg

Obermann

Nyhuis

. Central sensitization of the trigeminal and somatic nociceptive systems in medication overuse headache mainly involves cerebral supraspinal structures. Cephalalgia 2006; 26: 1106–1114.

Cevoli

Mochi

Scapoli

. Family history for chronic headache and drug overuse as risk factor for headache chronification. Headache 2008; 48: 412–418.

Di Lorenzo

Sances

. Drug consumption in medication overuse headache is influenced by brain-derived neurotrophic factor Val66Met polimorphism. J Headache Pain 2009; 10: 349–355.

Srikiatkhachorn

Anthony

. Serotonin receptor adaptation in patients with analgesic-induced headache. Cephalalgia 1996; 16: 419–422.

10.

Fumal

Laureys

Di Clemente

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

11.

Cupini

De Murtas

Costa

. Obsessive-compulsive disorder and migraine with medication-overuse headache. Headache 2009; 49: 1005–1013.

12.

Radat

Creac'h

Swendsen

. Psychiatric comorbidity in the evolution from migraine to medication overuse headache. Cephalalgia 2005; 25: 519–522.

13.

Lauwerier

Paemeleire

Van Damme

. Medication use in patients with migraine and medication-overuse headache: The role of problem solving and attitudes about pain medication. Pain 2011; 152: 1334–1339.

14.

Galli

Pozzi

Frustaci

. Differences in the personality profile of medication overuse headache sufferers and drug addict patients: A controlled study using MMPI-2. Headache 2011; 51: 1212–1227.

15.

Radat

Creac'h

Guegan-Massardier

. Behavioral dependence in patients with medication overuse headache: A cross-sectional study in consulting patients using the DSM-IV criteria. Headache 2008; 48: 1026–1036.

16.

Fuh

Wang

Juang

. Does medication overuse headache represent a behavior of dependence? Pain 2005; 119: 49–55.

17.

Evers S, Gralow

Bauer

. Sumatriptan and ergotamine overuse and drug-induced headache: A clinicoepidemiologic study. Clin Neuropharmacol 1999; 22: 201–206.

18.

Limmroth

Katsarava

Fritsche

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

19.

Dodick

Silberstein

. How clinicians can detect, prevent and treat medication overuse headache. Cephalalgia 2008; 28: 1209–1217.

20.

Sances

Ghiotto

Loi

. A CARE pathway in medication-overuse headache: The experience of the Headache Centre in Pavia. J Headache Pain 2005; 6: 307–309.

21.

Rossi

Di Lorenzo

Faroni

. Advice alone vs. structured detoxification programmes for medication overuse headache: A prospective, randomized, open-label trial in transformed migraine patients with low medical needs. Cephalalgia 2006; 26: 1097–1105.

22.

Lake

III . Screening and behavioral management: Medication overuse headache – The complex case. Headache 2008; 48: 26–31.

23.

Tfelt-Hansen

Krabbe

. Ergotamine. Do patients benefit from withdrawal? Cephalalgia 1981; 1(1): 29–32.

24.

Rossi

Faroni

Nappi

. Medication overuse headache: Predictors and rates of relapse in migraine patients with low medical needs. Cephalalgia 2008; 28: 1196–1200.

25.

Boe

Salvesen

Mygland

. Chronic daily headache with medication overuse: Predictors of outcome 1 year after withdrawal therapy. Eur J Neurol 2009; 16: 705–712.

26.

Sances

Ghiotto

Galli

. Risk factors in medication overuse headache: A one-year follow-up study (“Care” II protocol). Cephalalgia 2010; 30: 329–336.

27.

Zidverc-Trajkovic

Pakmezovic

Jovanovic

. Medication overuse headache: Clinical features predicting treatment outcome at 1-year follow-up. Cephalalgia 2007; 27: 1219–1225.

28.

Evers

Marziniak

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

29.

Katsarava

Muessig

Dzagnidze

. Medication overuse headache: Rates and predictors for relapse in a 4-year prospective study. Cephalalgia 2005; 25: 12–15.

30.

Schnider

Aull

Baumgartner

. Long-term outcome of patients with headache and drug abuse after in-patient withdrawal: Five-year follow-up. Cephalalgia 1996; 16: 481–485.

31.

Hagen

Albretsen

Vilming

. A 4-year follow-up of patients with medication-overuse headache previously included in a randomized multicentre study. J Headache Pain 2011; 12: 315–322.

32.

Hagen

Jensen

Boe

Stovner

. Medication overuse headache: A critical review of end points in recent follow-up studies. J Headache Pain 2010; 11: 373–377.

33.

Katsarava

Limmroth

Finke

. Rates and predictors for relapse in medication overuse headache: A 1-year prospective study. Neurology 2003; 27:60: 1682–1683.

34.

Nappi

Jensen

Nappi

. Diaries and calendars for migraine. A review. Cephalalgia 2006; 26: 905–916.

35.

Ghiotto

Sances

Galli

. Medication-overuse headache and applicability of the ICHD-II diagnostic criteria: One-year follow-up study (CARE I protocol). Cephalalgia 2009; 29: 233–243.

36.

Mazzi F, Morosini P, De Girolamo G, et al. SCID-I CV: Italian Edition. Florence: O.S. Organizzazioni Speciali, 2000.

37.

Diagnostic and Statistical Manual of Mental Disorders, 4th edn. Washington: American Psychiatric Association, 1994.

38.

Butcher

Dahlstrom

Graham

. The Minnesota Multiphasic Personality Inventory-2 (MMPI-2): Manual for administration and scoring , Minneapolis, MN: University of Minnesota Press, 1989.

39.

Sances

Galli

Anastasi

. Medication-overuse headache and personality: A controlled study by means of the MMPI-2. Headache 2010; 50: 198–209.

40.

Tepper

. Breaking the cycle of medication overuse headache. Cleveland Clinic J Med 2010; 77: 236–242.

41.

Ferrari

Roon

Lipton

Goadsby

. Oral triptans (serotonin 5-HT(1B/1D) agonists) in acute migraine treatment: A meta-analysis of 53 trials. Lancet 2001; 358: 1668–1675.

42.

Guidetti V, Galli F and Sheftell F. Headache attributed to psychiatric disorders. In: Handbook of clinical neurology. Nappi, Moskovitz (EDS): Elsevier, 2011.

43.

Fontanillas

Colas

Munoz

. Long-term evolution of chronic daily headache with medication overuse in the general population. Headache 2010; 50: 981–988.

44.

Hagen

Linde

Steiner

. Risk factors for medication-overuse headache: An 11-year follow-up study. The Nord-Trondelag Health Studies. Pain 2011; 153(1): 56–61.

45.

Suhr

Evers

Bauer

. Drug-induced headache: long-term results of stationary versus ambulatory withdrawal therapy. Cephalalgia 1999; 19: 44–49.

46.

Katsarava

Muessig

Dzagnidze

. Medication overuse headache: rates and predictors for relapse in a 4-year prospective study. Cephalalgia 2004; 25: 12–15.

47.

Schnider

Aull

Baumgartner

. Long-term outcome of patients with headache and drug abuse after inpatient withdrawal: Five-year follow-up. Cephalalgia 1996; 16: 481–485.

48.

Pini

Ciero

AFG

Sandrini

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

49.

Hagen

Jensen

Boe

Stovner

. Medication overuse headache: A critical review of end points in recent follow-up studies. J Headache Pain 2010; 11: 373–377.