Evaluating headache referral trends and practices across different settings in neurology clinics: insights from an international cross-sectional multicenter study

2.1 Study design

This follow-up study of a former one designed as a multinational and multicenter cross-sectional study was conducted five weeks from 01 April to 16 May 2022. During this period, 69 neurologists included patients on a designated day each week. The patient admission days were picked using the “Research Randomizer Program”. Neurologists evaluated all patients in detail, and all investigators received structured training on the International Classification of Headache Disorders-3 (ICHD-3) criteria [13] to increase study reliability.

All volunteers aged 18 years and older with headaches as the primary cause for admission were included in the study. It excluded patients younger than 18, without headaches, and did not agree to participate in the study. Online informed consent forms were obtained from volunteers who participated in the study. A structured questionnaire (Head-MENAA Study Questionnaire—Supplementary material 1) was administered to participants, including demographic information and headache characteristics questions [3]. Headache severity was evaluated by a numbered visual scale (NVS). Neurologists coded headache subtypes according to ICHD-3 criteria [13] in the same questionnaire.

Patients referred to NC were evaluated in this substudy of the Head-MENAA study. Patients were grouped into three different settings as referred from: (a) emergency services (ESs), (b) other specialty clinics (OSCs), and (c) those assessed in private offices (POs). The study population consisted of participating regions from the Middle East (Egypt, Iran), North Africa (Ivory Coast, Chad, Senegal, Sudan, Ethiopia, Morocco), and Asia (Türkiye, Turkish Republic of Northern Cyprus, Azerbaijan, Tatarstan, Mongolia) (Fig. 1). Therefore, the name of the study was determined as “Head-MENAA” (Middle East, North Africa, Asia) using the initials of these regions. The study coordinator obtained Ethics Committee approval from the Clinical Research Ethics Committee, University of the Health Sciences, Van Training, and Research Hospital, Van, Türkiye (Decision no: 2022/05–01, Date: 02 March 2022).

Fig. 1.The distribution of referred patients according to the three different clinical settings in neurology. (NOCs: Neurology Outpatient Clinics, *Only Tatarstan region joined the study from Russia, #The countries are listed in alphabetical order.)

2.2 Statistical analysis

Normality control of continuous variables was established with the Shapiro-Wilk test. Parametric tests were used for the variables that fit the normal distribution, and non-parametric tests were used for the ones that did not. One-way Analysis of Variance (ANOVA) and Kruskal Wallis tests were used to carry out the comparative analysis of more than two groups. In addition, the Tukey test was used as the post-hoc test. The chi-square test was applied to analyze categorical data. The statistical significance level was taken as p < 0.05. The analysis of the data was carried out through the TIBCO Statistica 13.5.0.17 program (TIBCO Software Inc., Palo Alto, CA, USA).

4.1 Consultation patterns of neurologic care for headaches

Many headache patients do not seek medical attention for their symptoms [5]. For example, headaches in Africa are primarily self-treated, mainly due to limited primary care and neurological consultation opportunities [6]. In addition to patients being reluctant to seek medical attention for their headaches, some patients lack access to healthcare providers due to socioeconomic, time, and transportation reasons [6, 18]. On the other hand, in many countries, healthcare services cannot meet the needs of many patients with headaches, even when they seek medical attention, due to the lack of neurologists and diagnostic support [11, 12, 19, 20, 21, 22]. Therefore, improving knowledge and developing effective referral systems may be essential in headache management.

In many regions, headaches are managed in primary care [7, 19]. In some regions, direct referrals can be made to neurology care or private headache clinics [23, 24]; in countries such as Spain and Norway, headache clinics are designed on a mixed referral model based on collaboration with primary care [25, 26].

Headaches are among the most frequent problems in primary care and the foremost reason for neurological consultations [25]. A comprehensive interview and physical examination of the patient can furnish the information needed to establish a correct headache diagnosis. It can also prevent unnecessary neuroimaging tests by allowing the exclusion of structural secondary causes [26]. However, these patients with primary headaches are often triaged at low priority and must be adequately treated in primary care [7]. Another area for improvement is patients’ difficulties receiving correct diagnosis and adequate treatment in primary care settings such as the ESs. In one study, only 18% of patients received analgesic treatment upon discharge from the ESs, leading to repeated emergency visits [27]. The unmet need for satisfactory acute headache treatment is the leading reason for ESs visits for primary headache patients [28]. The applicability of the ICHD criteria [13] in an ES is hampered by crowded ESs, nonlinear acceptance of severe cases, time constraints for obtaining a detailed history, and difficulty obtaining a history from the patient during the painful period, besides the lack of knowledge and training on the headache diagnosis [27, 29]. In a way that supports this, a study conducted in 2007 demonstrated that the application of ICHD criteria [13] in the ESs allowed correct diagnosis in only two-thirds of patients [30]. Another major problem is the excessive and inappropriate use of neuro-imaging methods in primary headaches due to several reasons, such as malpractice risk and patient anxiety [24]. Adopting suitable approaches in primary care and establishing a structured referral chain can support the efficient use of resources, the provision of correct diagnosis and treatment, and the determination of appropriate priorities among patients to be referred.

The referral decision is influenced by many factors, including diagnostic uncertainty, treatment decisions, patient anxiety and the referrer’s suspicion of underlying secondary causes needing neurologic care [8, 9]. Sources of referral to neurology include general practitioners, other specialists (especially neurosurgeons), other healthcare personnel, hospital staff and family members, in addition to their own applications [25, 31, 32, 33, 34]. The rates of referred patients may vary according to healthcare systems. For example, as mentioned above, in regions where the referral system is mandatory, the rate of patients referred to neurology by primary care is 55.3–71.7% [25, 33], while in areas where the referral system is not mandatory, this rate drops to 23–30% [31, 34]. In our study, patients from the ESs and OSCs were referred to neurology by general practitioners (GPs) and other specialists, reflecting their real-life experiences.

In a study conducted in the USA, the main complaint of 2.2–3.2 out of every 100 people who applied to ESs was headache [35, 36]. In a study conducted in Zagreb, one in five patients who applied to the ESs had a headache, three-quarters of which were primary headaches [37]. The rate of neurology consultations for headaches in patients who applied to the ESs differs between centers. In a study conducted in Türkiye, 0.83% of patients who applied to the ESs consulted neurology, but the headache was not among the reasons [38]. In contrast, in another study conducted in Türkiye, headache was the second most common reason for consultation (13.5% of consulted patients) [39]. These different patterns reflected that the referral system is not structured and its reasoning is not well-established.

In a study conducted in the UK, which uses the GP system more effectively, 2.1 out of 100 patients diagnosed with headaches in general practice were referred to neurology [21]. The rate of headache among patients referred to NOCs was 30.04% in our study, which was similar to studies in Cameroon (31.9%) [40] and Ecuador (33.2%) [34]. In contrast, in a study conducted in New Zealand, 69% of patients referred to neurology were headache patients [7]. Among the patients evaluated for headaches in our study, those referred from the ESs constituted 0.02% of all patients admitted for any reason, those referred from OSCs constituted 0.03%, and those evaluated in POs constituted 0.1%.

4.2 Demographic characteristics of the patients admitted to different settings

A study in the United Kingdom reported that young women had the highest headache consultation rates [21]. Indeed, in our research, there was a similar female predominance in areas other than the ESs. Our study also found that the proportion of male patients admitted to the ESs was significantly higher than in other clinical settings (Table 1). While 49.4% of patients referred from the ESs were male, this rate was 35.5% in a study conducted in Zagreb [37]. This may be due to men being less likely to seek routine medical care, longer waiting times to access a neurologist, socioeconomic reasons, and the fact that ESs are more accessible for pain management. According to one study, men are 60% less likely to seek medical attention than women [41]. Male patients are thought to delay seeking advice for headaches and use more over-the-counter analgesics to manage this condition. Indeed, a higher incidence of MOH was observed in men in one series [42]. In addition, a study conducted in Ireland showed that longer waiting times for appointments at the NCs and economic reasons were tribulations in headache management [43]. The literature still needs to learn the effects of many factors that may affect differences in gender preferences, such as sociocultural, economic, psychological and race.

There was no difference in mean ages between headache patients consulted from ESs and OSCs and those evaluated in POs, as seen in Table 1. Similar rates were also obtained in other studies. For example, the mean age of patients consulted from a private headache clinic was 42 years [25]. In another study evaluating patients referred from primary care to neurology, it was 48.5 [29]. Headache severity was higher in patients referred from ESs (7.48 ± 1.86), as expected, and also for those evaluated in POs (7.69 ± 1.3) compared to those admitting to NOCs (6.98 ± 1.74) and consulted from OSCs (6.32 ± 1.75). The headache frequency and lifetime pain duration of the patients evaluated in our study can be listed from the lowest to highest as ESs < OSCs < POs (Table 1). Similar to our study, the pain intensity of patients assessed in the ESs in Italy was also high (8.8 ± 1.6) [26]. It seems that individuals whose headaches are not severe enough and can tolerate headache symptoms generally do not admit to the ESs [5]. Since the neurologists working in POs in our study typically consist of physicians specializing in headaches, this group’s pain intensity, frequency, and lifetime pain duration were higher, probably because resistant headache cases are referred to these specialists.

4.3 Headache types and subtypes of the patients admitted to different settings

In our study, the primary headache rates in patients evaluated in the ESs, OSCs and POs were 89.2%, 90.3%, and 93.5%, respectively. In studies, the rate of headache among patients presenting to the ESs is around 2.2–3.2% [22, 35, 37]. In another study evaluating ESs patients in Ireland, the rate of primary headaches was 31% [22]. In Ecuador, 72% of headache patients consulting neurology were diagnosed with primary headache [34]. In our study, the rate of migraine headaches among primary headaches in patients assessed in the ESs, OSCs, and POs was 69.9%, 62.4%, and 72.5%, respectively, while the rate of TTH was 24.1%, 35.5% and 23.1% (Table 2). Approximately 50% of migraine patients and 16% of TTH patients consult a general practitioner for headache complaints [44]. In a study conducted in the USA, 63.5% of headaches in ESs were caused by migraine and vascular headaches, while 3.4% were caused by TTH [35]. Although TTH constitutes the most significant percentage of total headaches in population-based studies, migraine is observed at a higher rate in primary care [45]. The reason for the higher rate of migraine may be due to increased pain intensity, longer attack duration, and increased aura frequency [27]. In our study, patients with frequent episodic TTH were referred to NOCs more frequently from OSCs than POs.

In contrast, patients referred with probable TTH to NOCs were more often from ESs and OSCs than POs. When all these findings are evaluated together, it can be concluded that primary headaches (especially TTH) are not sufficiently recognized and treated in the settings participating in the study (especially by general practitioners and other specialists). At the same time, this situation may increase the burden on the health system and neurologists. A study conducted in low and middle-income countries showed that TTH needs to be better recognized, supporting this conclusion [46]. Another review suggested that there is an unmet need for effective diagnosis, treatment and management of migraine in East Asia [4]. Short-lasting unilateral neuralgiform headache attacks and probable TACs were more common in patients referred from ESs. Hemicrania continua was more common in patients referred from OSCs than in POs. The reason for this difference, which may prompt emergency visits, is that TACs are characterized by severe pain that awakens from sleep and has pulsatile/sharp quality.

Our study demonstrated that the secondary headache rates of patients evaluated in the ESs, OSCs, and POs were very similar, 36.1%, 33.3% and 30.6%, respectively. Although secondary headaches were more common in the ESs than in OSCs, this was not statistically significant. On the contrary, in a study evaluating patients in the ESs, 15% of patients with headaches had serious secondary causes [22]. Secondary headaches were observed in 21% of patients consulted with headaches to neurology in Ecuador [34]. Remarkably, we found that two diverse causes, “Headache attributed to CVT” and “Headache attributed to somatization disorder,” were observed more frequently in patients referred from the ESs than those referred from POs (Table 2). The most common neurological causes of referral within somatic symptoms are conversion disorders, non-epileptic attacks, and chronic benign headaches [47]. Somatic symptoms are widespread in patients with chronic migraine and chronic daily headaches, especially if these patients have severe headaches, anxiety or depression [48]. The perception of pain during a headache attack and the presence of stressful events and/or psychiatric comorbidities such as anxiety or depression may explain the decision to seek medical help in ESs [49, 50]. A study of adult patients in ESs in the UK found a surprisingly low rate of somatization and somatoform disorders of 3.8% [51]. Although it is unknown how high this rate is, the true ESs prevalence lies somewhere between this figure and the rates seen in primary care and outpatient clinics. Another exciting study result is the higher prevalence of CVT in patients referred from the ESs, possibly because headaches in CVT are acute and more severe [52]. No significant difference was found between the groups for other headache subtypes in which acute and severe headache was observed. This may be because they were mortal, clinical urgency caused neglect to question the headache symptom, and these patients are consulted OSCs, such as neurosurgery. For instance, as seen in Table 2, any patients with headaches attributed to trauma or injury to the head and/or neck were not consulted by neurologists by the ESs. Another striking factor among the secondary headache causes in the study was the intensity of MOH (Table 2). The risk of MOH increases as the number of unsuccessful treatments and doctors consulted increases. Another issue is the recent adoption of the approach of taking medications as early as possible during a migraine attack. This approach increases the likelihood that the patient will frankly take more medicines than necessary and may pave the way for the development of MOH [53]. Chagas et al. [42] reported in their study that 55% of headache patients referred by primary and secondary health care used medications prescribed by a doctor, while 34% self-medicated. Another study found that 13.3% of patients with TTH had MOH and that combined analgesics were preferred due to their popularity, availability without a prescription, and being cheap and readily available [54]. Therefore, in regions where the demand for neurology consultation is high, overuse of simple analgesics in patients who cannot access appropriate diagnosis and prophylactic treatment may increase the risk of MOH [7]. The most effective way to prevent MOH is to specify patients at risk and educate them about acute medication use [55].