Insights into radiographic investigations for headaches in general practice

Insights into radiographic investigations for headaches in general practice Abstract Background/Objective Headaches present commonly in general practice, and doctors face uncertainties and pressures in dealing with them. This study identifies key characteristics of headaches investigated through head imaging, in the hope of strengthening existing guidelines. Method A 7-year retrospective analysis of head imaging scans ordered for investigating headaches among patients aged ≥18 in an Australian general practice setting. Results A total of 109 of the 517 (21.1%) patients required head imaging, although 14 (2.7%) of these had repeat scans. Three-quarters were females, and most of the headaches were recurrent (56.9%), severe (62.4%) or had associated red flags (78.0%). Computed tomography (CT) and magnetic resonance imaging (MRI) were the only scans adopted (4:1 and 1:1 for first and repeat scans, respectively). Twelve (11.0%) scans had findings likely to explain the headaches after the initial scans, with no difference in findings between diagnoses from CTs and MRIs (P = 0.41). Repeat scans offered no additional benefits, and roughly one in three patients were referred to third-party carers (mostly neurologists). Females had more recurrent headaches (OR = 2.63; CI = 1.09–6.35; P = 0.03). Patients with psycho-morbidities were less likely to have scan findings that explained their headaches (OR = 0.22; CI = 0.06–0.88; P = 0.03), and, though not quite significant, were also more likely to undergo imaging (OR = 1.47; CI 0.96–2.27; P = 0.08). Conclusion Significant intracranial findings were uncommon following imaging for headaches, and MRIs offer no advantages over CTs. Repeating head scans within 5 years offers no clear benefits. Psycho-morbidities should be considered when deciding the imaging needs, given the lesser chance of findings. Larger studies will help validate these findings. Brain, general practice, head, headaches, imaging, investigation, migraines, primary care, scan Introduction Headaches are common reasons for consulting GPs (1–3) and can account for up to 1.5% of all general practice consultations (1). They are listed among the top 10 disabling illnesses by the WHO (2) and make up the most common new neurological complaints to both GPs and neurologists (3–5). The social, economic and personal costs of headaches are believed to be considerable (6,7). In Australia, findings from the 2005 Bettering the Evaluation and Care of Health (BEACH) study, covering data from 1998 to 2004, showed that headaches were seen at an annual rate of 1.9% per annum, with about 1.9 million yearly cases across the country (8). They are usually not the only symptom for patients presenting with it, and present as sole symptoms only in 30% of cases (8). Despite the foregoing statistics, dangerous causes of headaches are very rare (1,3,7), and most cases in general practice are treated by physical examination, drug prescriptions (1) and reassurances (4) only. However, where deemed relevant (like with the presence of ‘red flags’), GPs do refer patients for investigations (blood tests and imaging of head/brain) or specialist opinions (4). Such referrals are still issued in some cases where there are no ‘red flags’, or in cases where the headaches are not medically explained, perhaps due to high anxiety from the patients, or uncertainties and fear of medico-legal issues by the GPs (7,9–12). Red flags include sudden, severe headaches or those different to previous episodes, exacerbated by coughing, sneezing, straining, eye movements and postural changes, as well as those associated with blurred vision, stiff neck, rash, altered consciousness, confusion and focal neurological signs (5). Given the common nature of headache presentations, the uncertainty on the part of the GPs in diagnosis and management, the pressure from anxious patients and the potential medico-legal implications of missing important findings, GPs need to have some robust, evidence-based and up-to-date guidelines that will assist them in deciding when to investigate, who to investigate and when to refer to third-party practitioners (specialists and other allied health staff). In fact, a number of studies have claimed that managing headaches efficiently in general practice can be improved through diagnostic and treatment approaches (7,13,14). Evidence-based improvements in the management of headaches in primary care can help enhance both direct (avoiding unnecessary imaging tests) and indirect (by saving time and work-related implications) costs to patients and healthcare funders. Also, the management of headaches in general practice is one area where effective communication with patients can improve outcomes (15). Therefore, a good support network to GPs will help boost their confidence, and provide more pragmatic frameworks for delivering such communications, and this study aims to achieve this. In addition, our work will add to the existing database of studies on headaches in primary care, an area that has been surprisingly understudied, and provide recommendations for both policy changes and future research into the management of headaches in general practice. Unfortunately, researches into headaches in general practice have been poor (1), and a number of publications in the United Kingdom have argued that existing guidelines may be fraught with uncertainties (16,17). These existing guidelines, both in Australia and globally, can be strengthened with further research. To achieve the forgoing objectives, this study explored the imagings done to investigate headaches and identified the characteristics of those headaches in terms of their severity, recurrence and red flags. The types of scans utilized, the proportion of scans that showed significant findings and the ensuing follow-up and referral patterns (to either specialists or allied health experts) after the scans were also analysed. The potential impact of psychological co-morbidities on the imaging requests ordered as part of the investigation of headaches was also explored. Methods Setting This is a 7-year retrospective study that explored the patterns of all brain and head (hereinafter used interchangeably) imaging scans ordered as part of investigations for patients presenting with headaches in an Australian general practice setting between the 1 January 2010 and 1 April 2017. The study focused on a group of three geographically distinct, but administratively and functionally inter-connected, group of general practice surgeries named ‘Kennedy Drive Medical Centre (KDMC)’, ‘PKG Medical Centre’ and the ‘Bilambil Heights Medical Centre (BHMC)’. All three are located in the ‘Tweed Shire Area’ of Northern New South Wales (NSW), Australia, also called the ‘Tweed Area’ or simply ‘The Tweed’ (18). ‘The Tweed’ covers about 1303 km2, and sits at the border of the states of NSW and Queensland, about 102 km to the South of Brisbane and 830 km North of Sydney (19). Alone, the Tweed’s population is approximately 90000 (18). However, the three surgeries are among a few others serving an estimated patient population of about 150000 (20,21), giving that patients that live in the Tweed, as well as the southern suburbs of the City of Gold Coast, all patronize them. The records and files of all patients, as well as the management system of all the three surgeries are inter-linked, and, combined, they have a registered patient population of about 31000, served by a total of about 11 doctors at each particular time (seven doctors in KDMC, three in PKG and two in BHMC). Demographics Even though the median age of the population in the Tweed Area was about 54 years according to the 2011 census, the larger ‘Gold Coast–Tweed Area’, which our study area focused on, has a median age of 38 years, a number similar to the Australian average of 37 years (19). The older median age in the Tweed is because the area is a hub of retirees, hosting about 14 separate retirement villages in the area (19). About 87% of the adult population in the Gold Coast–Tweed Area is employed (either full or part-time), with a male to female ratio of 51:49, and the child to family ratio of 2:1 (19). More than 80% of the residents are Caucasians, whereas 87% has English as their dominant first language (19). Inclusion and exclusion criteria The inclusion criteria included all adults aged ≥18 years of age at the time of their presentation, who saw a doctor within the defined period in any of the three identified practices, with headache as a major complaint, and requiring head imaging as part of their management. Headaches of all kinds were included, provided it was the main reason for the consultation. Patients who presented with headaches in this period, but had no imaging done, were excluded from the study. Also excluded were those who had head scans for other reasons like sinusitis, dizziness and suspected strokes, but did not present primarily with headaches. To be included, the requests for the head scans must have been ordered on the same day of the consultation for the headaches. Patient selection The ‘search tool’ in the Medical Director® software was used to identify all the patients seen in the defined period who met the study criteria. The search filter was broadened to include all forms of headaches and migraines with all their identified variations, including cervicogenic, sinus, chronic, cluster, tension, migraine (common, classical, cluster, migrainous aura) and posthead injury. All patients identified were then manually reviewed to ensure that they met all the inclusion–exclusion criteria. All potential forms of imaging were allowed, including computed tomography (CT), magnetic resonance imaging (MRI), magnetic resonance angiography, ultrasound and X-rays. Basic patient demographics, duration of headache, findings in the imaging results and immediate treatment offered were also obtained. Pre-defined terms were stated ahead of the selection, so as to avoid bias, and these were applied strictly. They are contained in Box 1. Box 1: Definition of terms Dates of the scans: Given that this is a retrospective study, the ages of the patients at the time of our data collection and analysis (May–June 2017) were expectedly different to the ages at the times they had the imaging scans. For this reason, the dates of their respective scans were collected and used to work out the patients’ ages at the times they did the scans. These latter ages were used in most of our analysis unless otherwise stated. For the sake of the analysis, the age categorization was into the four groups of ‘18–24’, ‘25–44’, ‘45–64’ and ‘≥65’, following the same pattern established by the popular Australian Bettering the Evaluation and Care of Health study (8). Severity of the headaches were recorded as documented in the patients’ notes. Where no such documentation is found, we recorded it as ‘unknown’. For the purpose of our work, headaches are classified as ‘recurrent’if it is documented as such, or if there were two or more documented primary presentations of headaches in the period surveyed. Findings from the scan: The findings were grouped into ‘none’, ‘intracranial’ and ‘extracranial’ findings. Given that some of the intracranial findings were likely to be incidental, they were further divided into those likely to explain the headaches, and those unlikely to do so. Identified red flags that informed the scans: A number of danger signs that would alert the GPs to the need to carry out imaging scans have been identified and listed earlier. Most GPs document these as the basis to justify doing scans for patients presenting with headaches. Where possible from the records, data on these were collected. Psychological co-morbidities of the patients: Extra care was taken to identify all cases of psychological co-morbidities existing in the patients that were seen for headaches. These include depression, anxiety, post-traumatic stress disorder, bipolar diseases and stress. To be included as a co-morbidity for a patient, there must be evidence of either a pharmacological (medications) or psychological (correspondences with psychologists and psychiatrists, including treatments offered through Mental Health Care Plans or otherwise) treatment for the disorder. This criterion was included to avoid spurious or transient forms of psychological disorders like in cases used as excuses for work absence certificates, minor grief situations, and transient work stress. Follow-up/referrals following the scans: All follow ups directly related to the imaging scans were noted. The kinds of third-party referrals to either specialist doctors or allied health staff patients were recorded, and where possible, those who initiated the referrals were identified (patients themselves or their GPs). Pre-defined terms were stated ahead of the selection, so as to avoid bias, and these were applied strictly and are defined. Data collection Data collection was done between May and June of 2017. For each patient that met the study criteria, data were collected on gender, date of birth, type(s) of scan done, date(s) of the scans, severity of the headaches, recurrence of the headaches, finding(s) from the scan(s), identified red flags that necessitated the scans, psychological co-morbidities of the patients and follow-up/referrals following the scans. In cases where a patient had more than one scan in the 7-year period, relevant data similar to that previously described were also collected (up to a maximum of two consultations), provided that the reason for the subsequent scans was for the investigation of a headache complaint. Data analysis Analysis was with the IBM SPSS Version 24. To identify significant associations, binary logistics regression (BLR) was used. The probability (P value) was set at <0.05. One part of the regression analysis focused solely on the patients who had head imaging, and for this, three outcome (dependent) variables were tested (‘gender’, ‘age’ and ‘psychological co-morbidities’) against two predictor (independent) variables (‘recurrence of headaches’ and ‘findings on imaging likely to explain headaches or not’). Results We found that 109 of the 517 (21.1%) adult patients who presented with headaches over the 7-year period required imaging as part of the investigations for their headaches. Scan 1 The reports for the first set of scans are summarized in Table 1. The age range (in years) at the time the scans were done, ranged from 18 to 88, with an average of 41.2 years (SD 15.47). The majority were females (81% or 74.3%). Only CT and MRI scans were done, with CTs used in 91 of the 109 patients (83.5%). Table 1. Demographics and characteristics of patients investigated initially with head imaging for headaches over a 7-year period in an Australian general practice setting (n = 109) S. No  Patient variable  Number (%)  Number by gender (%)  1  Gender  Female  81 (74.3)  Male  28 (25.7)  2  Age range (as at time of second scans)  18–24  20 (18.3)  Female = 14 (70.0)  Male = 6 (30.0)  25–44  46 (42.2)  Female = 33 (71.7)  Male = 13 (28.3)  44–64  34 (31.2)  Female = 27 (79.4)  Male = 7 (20.6)  ≥65  9 (8.3)  Female = 7 (77.8)  Male = 2 (22.2)  3  Recurrence of headache  Recurrent  62 (56.9)  Female = 51 (82.3)  Male = 11 (17.7)  Not recurrent  47 (43.1)  Female = 30 (63.8)  Male = 17 (36.2)  4  Severity of headaches  Severe  68 (62.4)  Female = 46 (67.6)  Male = 22 (32.4)  Not recorded  41 (37.6)  Female = 35 (85.4)  Male = 6 (14.6)  5  Type of scan  CT head/brain  91 (83.5)  Female = 67 (73.6)  Male = 24 (26.4)  MRI head/brain  18 (16.5)  Female = 14 (77.8)  Male = 4 (22.2)  6  Findings from scan  Intracranial  17 (15.6)  Female = 15 (88.2)  Male = 2 (11.8)  Extracranial  14 (12.8)  Female = 7 (50.0)  Male = 7 (50.0)  No pathological finding  78 (71.6)  Female = 59 (75.6)  Male = 19 (24.4)  7  Findings from scan likely to explain headache  Yes  12 (11.0)  Female = 11 (91.7)  Male = 1 (8.3)  No  97 (89.0)  Female = 70 (72.2)  Male = 27 (27.8)  8  Red flags present or not  Yes  85 (78.0)  Female = 65 (76.5)  Male = 20 (23.5)  No  24 (22.0)  Female = 16 (66.7)  Male = 8 (33.3)  9  Follow-up/referral  Of the 28 patients referred to specialists and other allied health experts, 2 (both females, and one each to neurologist and neurosurgeon) requested for the referrals themselves, whereas the other referrals were initiated by the GPs  GP  81 (74.3)  Female = 57 (70.4)  Male = 24 (29.6)  Neurologist  18 (16.5)  Female = 15 (83.3)  Male = 3 (16.7)  Neurosurgeon  3 (2.8)  Female = 3 (100.0)  Male = 0 (0.0)  ENT  3 (2.8)  Female = 2 (66.7)  Male = 1 (33.3)  Optometrist  2 (1.8)  Female = 2 (100.0)  Male = 0 (0.0)  Physiotherapist  2 (1.8)  Female = 2 (100.0)  Male = 0 (0.0)  10  Psychological co-morbidities (including depression, anxiety, schizophrenia and bipolar disorder)  Present  48 (44.0)  Female = 40 (83.3)  Male = 8 (16.7)  Absent  61 (56.0)  Female = 41 (67.2)  Male = 20 (32.8)  S. No  Patient variable  Number (%)  Number by gender (%)  1  Gender  Female  81 (74.3)  Male  28 (25.7)  2  Age range (as at time of second scans)  18–24  20 (18.3)  Female = 14 (70.0)  Male = 6 (30.0)  25–44  46 (42.2)  Female = 33 (71.7)  Male = 13 (28.3)  44–64  34 (31.2)  Female = 27 (79.4)  Male = 7 (20.6)  ≥65  9 (8.3)  Female = 7 (77.8)  Male = 2 (22.2)  3  Recurrence of headache  Recurrent  62 (56.9)  Female = 51 (82.3)  Male = 11 (17.7)  Not recurrent  47 (43.1)  Female = 30 (63.8)  Male = 17 (36.2)  4  Severity of headaches  Severe  68 (62.4)  Female = 46 (67.6)  Male = 22 (32.4)  Not recorded  41 (37.6)  Female = 35 (85.4)  Male = 6 (14.6)  5  Type of scan  CT head/brain  91 (83.5)  Female = 67 (73.6)  Male = 24 (26.4)  MRI head/brain  18 (16.5)  Female = 14 (77.8)  Male = 4 (22.2)  6  Findings from scan  Intracranial  17 (15.6)  Female = 15 (88.2)  Male = 2 (11.8)  Extracranial  14 (12.8)  Female = 7 (50.0)  Male = 7 (50.0)  No pathological finding  78 (71.6)  Female = 59 (75.6)  Male = 19 (24.4)  7  Findings from scan likely to explain headache  Yes  12 (11.0)  Female = 11 (91.7)  Male = 1 (8.3)  No  97 (89.0)  Female = 70 (72.2)  Male = 27 (27.8)  8  Red flags present or not  Yes  85 (78.0)  Female = 65 (76.5)  Male = 20 (23.5)  No  24 (22.0)  Female = 16 (66.7)  Male = 8 (33.3)  9  Follow-up/referral  Of the 28 patients referred to specialists and other allied health experts, 2 (both females, and one each to neurologist and neurosurgeon) requested for the referrals themselves, whereas the other referrals were initiated by the GPs  GP  81 (74.3)  Female = 57 (70.4)  Male = 24 (29.6)  Neurologist  18 (16.5)  Female = 15 (83.3)  Male = 3 (16.7)  Neurosurgeon  3 (2.8)  Female = 3 (100.0)  Male = 0 (0.0)  ENT  3 (2.8)  Female = 2 (66.7)  Male = 1 (33.3)  Optometrist  2 (1.8)  Female = 2 (100.0)  Male = 0 (0.0)  Physiotherapist  2 (1.8)  Female = 2 (100.0)  Male = 0 (0.0)  10  Psychological co-morbidities (including depression, anxiety, schizophrenia and bipolar disorder)  Present  48 (44.0)  Female = 40 (83.3)  Male = 8 (16.7)  Absent  61 (56.0)  Female = 41 (67.2)  Male = 20 (32.8)  View Large Table 1. Demographics and characteristics of patients investigated initially with head imaging for headaches over a 7-year period in an Australian general practice setting (n = 109) S. No  Patient variable  Number (%)  Number by gender (%)  1  Gender  Female  81 (74.3)  Male  28 (25.7)  2  Age range (as at time of second scans)  18–24  20 (18.3)  Female = 14 (70.0)  Male = 6 (30.0)  25–44  46 (42.2)  Female = 33 (71.7)  Male = 13 (28.3)  44–64  34 (31.2)  Female = 27 (79.4)  Male = 7 (20.6)  ≥65  9 (8.3)  Female = 7 (77.8)  Male = 2 (22.2)  3  Recurrence of headache  Recurrent  62 (56.9)  Female = 51 (82.3)  Male = 11 (17.7)  Not recurrent  47 (43.1)  Female = 30 (63.8)  Male = 17 (36.2)  4  Severity of headaches  Severe  68 (62.4)  Female = 46 (67.6)  Male = 22 (32.4)  Not recorded  41 (37.6)  Female = 35 (85.4)  Male = 6 (14.6)  5  Type of scan  CT head/brain  91 (83.5)  Female = 67 (73.6)  Male = 24 (26.4)  MRI head/brain  18 (16.5)  Female = 14 (77.8)  Male = 4 (22.2)  6  Findings from scan  Intracranial  17 (15.6)  Female = 15 (88.2)  Male = 2 (11.8)  Extracranial  14 (12.8)  Female = 7 (50.0)  Male = 7 (50.0)  No pathological finding  78 (71.6)  Female = 59 (75.6)  Male = 19 (24.4)  7  Findings from scan likely to explain headache  Yes  12 (11.0)  Female = 11 (91.7)  Male = 1 (8.3)  No  97 (89.0)  Female = 70 (72.2)  Male = 27 (27.8)  8  Red flags present or not  Yes  85 (78.0)  Female = 65 (76.5)  Male = 20 (23.5)  No  24 (22.0)  Female = 16 (66.7)  Male = 8 (33.3)  9  Follow-up/referral  Of the 28 patients referred to specialists and other allied health experts, 2 (both females, and one each to neurologist and neurosurgeon) requested for the referrals themselves, whereas the other referrals were initiated by the GPs  GP  81 (74.3)  Female = 57 (70.4)  Male = 24 (29.6)  Neurologist  18 (16.5)  Female = 15 (83.3)  Male = 3 (16.7)  Neurosurgeon  3 (2.8)  Female = 3 (100.0)  Male = 0 (0.0)  ENT  3 (2.8)  Female = 2 (66.7)  Male = 1 (33.3)  Optometrist  2 (1.8)  Female = 2 (100.0)  Male = 0 (0.0)  Physiotherapist  2 (1.8)  Female = 2 (100.0)  Male = 0 (0.0)  10  Psychological co-morbidities (including depression, anxiety, schizophrenia and bipolar disorder)  Present  48 (44.0)  Female = 40 (83.3)  Male = 8 (16.7)  Absent  61 (56.0)  Female = 41 (67.2)  Male = 20 (32.8)  S. No  Patient variable  Number (%)  Number by gender (%)  1  Gender  Female  81 (74.3)  Male  28 (25.7)  2  Age range (as at time of second scans)  18–24  20 (18.3)  Female = 14 (70.0)  Male = 6 (30.0)  25–44  46 (42.2)  Female = 33 (71.7)  Male = 13 (28.3)  44–64  34 (31.2)  Female = 27 (79.4)  Male = 7 (20.6)  ≥65  9 (8.3)  Female = 7 (77.8)  Male = 2 (22.2)  3  Recurrence of headache  Recurrent  62 (56.9)  Female = 51 (82.3)  Male = 11 (17.7)  Not recurrent  47 (43.1)  Female = 30 (63.8)  Male = 17 (36.2)  4  Severity of headaches  Severe  68 (62.4)  Female = 46 (67.6)  Male = 22 (32.4)  Not recorded  41 (37.6)  Female = 35 (85.4)  Male = 6 (14.6)  5  Type of scan  CT head/brain  91 (83.5)  Female = 67 (73.6)  Male = 24 (26.4)  MRI head/brain  18 (16.5)  Female = 14 (77.8)  Male = 4 (22.2)  6  Findings from scan  Intracranial  17 (15.6)  Female = 15 (88.2)  Male = 2 (11.8)  Extracranial  14 (12.8)  Female = 7 (50.0)  Male = 7 (50.0)  No pathological finding  78 (71.6)  Female = 59 (75.6)  Male = 19 (24.4)  7  Findings from scan likely to explain headache  Yes  12 (11.0)  Female = 11 (91.7)  Male = 1 (8.3)  No  97 (89.0)  Female = 70 (72.2)  Male = 27 (27.8)  8  Red flags present or not  Yes  85 (78.0)  Female = 65 (76.5)  Male = 20 (23.5)  No  24 (22.0)  Female = 16 (66.7)  Male = 8 (33.3)  9  Follow-up/referral  Of the 28 patients referred to specialists and other allied health experts, 2 (both females, and one each to neurologist and neurosurgeon) requested for the referrals themselves, whereas the other referrals were initiated by the GPs  GP  81 (74.3)  Female = 57 (70.4)  Male = 24 (29.6)  Neurologist  18 (16.5)  Female = 15 (83.3)  Male = 3 (16.7)  Neurosurgeon  3 (2.8)  Female = 3 (100.0)  Male = 0 (0.0)  ENT  3 (2.8)  Female = 2 (66.7)  Male = 1 (33.3)  Optometrist  2 (1.8)  Female = 2 (100.0)  Male = 0 (0.0)  Physiotherapist  2 (1.8)  Female = 2 (100.0)  Male = 0 (0.0)  10  Psychological co-morbidities (including depression, anxiety, schizophrenia and bipolar disorder)  Present  48 (44.0)  Female = 40 (83.3)  Male = 8 (16.7)  Absent  61 (56.0)  Female = 41 (67.2)  Male = 20 (32.8)  View Large Table 2 shows that 17 (15.6%) of the 31 (28.4%) patients scanned had intracranial findings, but as shown in Table 3, only 12 (11.0% of all patients) had findings that were likely to have explained their headaches. Of the 12 intracranial findings, 9 came from CTs and 3 from MRIs, with no significant differences in using either (OR = 0.55; CI = 0.13–2.27; P = 0.41). Table 2. All findings (intracranial and extracranial) on head scans done for patients presenting with headaches over a 7-year period in an Australian general practice setting   Group of findings  Number reported in scan 1  Number reported in scan 2  1  Atrophy (including age and non-age related, cerebellar, cortical, cerebral, early)  4  1  2  Prominent cerebrospinal spaces  2  1  Arachnoid cyst  1  1  3  Developmental venous anomaly  1  –  4  Traumatic attenuation  1  –  5  Combined atrophy + ischemia  1  –  6  Hyperintensities/demyelination  2  –  7  Meningioma  1  1  8  Sinus mucosal thickenings and sinus bone dysplasia  15  11  9  Ischemia (acute or chronic)  2  –  10  Combined cyst prominent cerebrospinal spaces  1  1    Total  31  7    Group of findings  Number reported in scan 1  Number reported in scan 2  1  Atrophy (including age and non-age related, cerebellar, cortical, cerebral, early)  4  1  2  Prominent cerebrospinal spaces  2  1  Arachnoid cyst  1  1  3  Developmental venous anomaly  1  –  4  Traumatic attenuation  1  –  5  Combined atrophy + ischemia  1  –  6  Hyperintensities/demyelination  2  –  7  Meningioma  1  1  8  Sinus mucosal thickenings and sinus bone dysplasia  15  11  9  Ischemia (acute or chronic)  2  –  10  Combined cyst prominent cerebrospinal spaces  1  1    Total  31  7  View Large Table 2. All findings (intracranial and extracranial) on head scans done for patients presenting with headaches over a 7-year period in an Australian general practice setting   Group of findings  Number reported in scan 1  Number reported in scan 2  1  Atrophy (including age and non-age related, cerebellar, cortical, cerebral, early)  4  1  2  Prominent cerebrospinal spaces  2  1  Arachnoid cyst  1  1  3  Developmental venous anomaly  1  –  4  Traumatic attenuation  1  –  5  Combined atrophy + ischemia  1  –  6  Hyperintensities/demyelination  2  –  7  Meningioma  1  1  8  Sinus mucosal thickenings and sinus bone dysplasia  15  11  9  Ischemia (acute or chronic)  2  –  10  Combined cyst prominent cerebrospinal spaces  1  1    Total  31  7    Group of findings  Number reported in scan 1  Number reported in scan 2  1  Atrophy (including age and non-age related, cerebellar, cortical, cerebral, early)  4  1  2  Prominent cerebrospinal spaces  2  1  Arachnoid cyst  1  1  3  Developmental venous anomaly  1  –  4  Traumatic attenuation  1  –  5  Combined atrophy + ischemia  1  –  6  Hyperintensities/demyelination  2  –  7  Meningioma  1  1  8  Sinus mucosal thickenings and sinus bone dysplasia  15  11  9  Ischemia (acute or chronic)  2  –  10  Combined cyst prominent cerebrospinal spaces  1  1    Total  31  7  View Large Table 3. Specific intracranial findings on head scans that are likely to explain headaches among patients presenting in an Australian general practice setting First scans  Second scans  Findings  Type of scan  Findings  Type of scan  Arachnoid cyst  CT  Intraventricular arachnoid cyst + calcification  CT  Significant distention of right ventricle + calcification + cyst  CT  Prominent ventricular system; atrophy; hydrocephalus  CT  Meningioma  MRI  Prominent cerebrospinal spaces (distention) + frontal atrophy (not with age)  CT  Hypodense R cerebral hemisphere (can be traumatic, infectious, malignant or vascular, etc.)  CT  Prominent cerebrospinal space  MRI  Attenuation of left parietal lobe posttrauma  CT  Microangiopathic white matter ischemia and vasculitis + generalized atrophy  MRI  Total = 3  CT = 1; MRI = 2  Developmental venous anomaly  CT  Subcortical intensities: likely vasculitis or ischemia  MRI  Meningioma  CT  Hyperintense white matter: likely demyelination or vasculitis  MRI  Ischemia (acute/subacute)  CT  Total = 12  CT = 9; MRI = 3  First scans  Second scans  Findings  Type of scan  Findings  Type of scan  Arachnoid cyst  CT  Intraventricular arachnoid cyst + calcification  CT  Significant distention of right ventricle + calcification + cyst  CT  Prominent ventricular system; atrophy; hydrocephalus  CT  Meningioma  MRI  Prominent cerebrospinal spaces (distention) + frontal atrophy (not with age)  CT  Hypodense R cerebral hemisphere (can be traumatic, infectious, malignant or vascular, etc.)  CT  Prominent cerebrospinal space  MRI  Attenuation of left parietal lobe posttrauma  CT  Microangiopathic white matter ischemia and vasculitis + generalized atrophy  MRI  Total = 3  CT = 1; MRI = 2  Developmental venous anomaly  CT  Subcortical intensities: likely vasculitis or ischemia  MRI  Meningioma  CT  Hyperintense white matter: likely demyelination or vasculitis  MRI  Ischemia (acute/subacute)  CT  Total = 12  CT = 9; MRI = 3  View Large Table 3. Specific intracranial findings on head scans that are likely to explain headaches among patients presenting in an Australian general practice setting First scans  Second scans  Findings  Type of scan  Findings  Type of scan  Arachnoid cyst  CT  Intraventricular arachnoid cyst + calcification  CT  Significant distention of right ventricle + calcification + cyst  CT  Prominent ventricular system; atrophy; hydrocephalus  CT  Meningioma  MRI  Prominent cerebrospinal spaces (distention) + frontal atrophy (not with age)  CT  Hypodense R cerebral hemisphere (can be traumatic, infectious, malignant or vascular, etc.)  CT  Prominent cerebrospinal space  MRI  Attenuation of left parietal lobe posttrauma  CT  Microangiopathic white matter ischemia and vasculitis + generalized atrophy  MRI  Total = 3  CT = 1; MRI = 2  Developmental venous anomaly  CT  Subcortical intensities: likely vasculitis or ischemia  MRI  Meningioma  CT  Hyperintense white matter: likely demyelination or vasculitis  MRI  Ischemia (acute/subacute)  CT  Total = 12  CT = 9; MRI = 3  First scans  Second scans  Findings  Type of scan  Findings  Type of scan  Arachnoid cyst  CT  Intraventricular arachnoid cyst + calcification  CT  Significant distention of right ventricle + calcification + cyst  CT  Prominent ventricular system; atrophy; hydrocephalus  CT  Meningioma  MRI  Prominent cerebrospinal spaces (distention) + frontal atrophy (not with age)  CT  Hypodense R cerebral hemisphere (can be traumatic, infectious, malignant or vascular, etc.)  CT  Prominent cerebrospinal space  MRI  Attenuation of left parietal lobe posttrauma  CT  Microangiopathic white matter ischemia and vasculitis + generalized atrophy  MRI  Total = 3  CT = 1; MRI = 2  Developmental venous anomaly  CT  Subcortical intensities: likely vasculitis or ischemia  MRI  Meningioma  CT  Hyperintense white matter: likely demyelination or vasculitis  MRI  Ischemia (acute/subacute)  CT  Total = 12  CT = 9; MRI = 3  View Large A total of 28 (25.7%) patients needed referrals to third-party carers other than their GPs following their initial scans, with nearly two-thirds of these being to neurologists, whereas 48 patients (44%) had known psychological co-morbidities (Table 1). Scan 2 This is summarized in Table 4, and show that only 14 of the 109 (12.8%) patients who had the initial scans required a repeat scan within the 7-year period surveyed. The longest time between the initial and the repeat scans was 5 years. Table 4. Demographics and characteristics of patients investigated for the second time with head imaging for headaches over a 7-year period in an Australian general practice setting (n = 14) S. No  Variable    Number (%)  Number by gender (%)  1  Age range (as at time of second scans)  18–24  2 (14.3)  Female = 2 (100.0)  Male = 0 (0.0)  25–44  6 (42.9)  Female = 4 (66.7)  Male = 2 (33.3)  44–64  5 (35.7)  Female = 5 (100.0)  Male = 0 (0.0)  ≥65  1 (7.1)  Female = 1 (100.0)  Male = 0 (0.0)  2  Severity of headaches  Severe  9 (64.3)  Female = 7 (77.8)  Male = 2 (22.2)  Not recorded  5 (35.7)  Female = 5 (100.0)  Male = 0 (0.0)  3  Type of scan  CT head/brain  7 (50.0)  Female = 6 (85.7)  Male = 1 (14.3)  MRI head/brain  7 (50.0)  Female = 6 (85.7)  Male = 1 (14.3)  4  Findings from scan  Intracranial  4 (28.6)  Female = 4 (100.0)  Male = 0 (0.0)  Extracranial  3 (21.4)  Female = 2 (66.7)  Male = 1 (33.3)  No pathological finding  7 (50.0)  Female = 6 (85.7)  Male = 1 (14.3)  5  Findings from scan likely to explain headache  Yes  3 (21.4)  Female = 3 (100.0)  Male = 0 (0.0)  No  11 (78.6)  Female = 9 (81.8)  Male = 2 (18.2)  6  Red flags present or not  Yes  10 (71.4)  Female = 9 (90.0)  Male = 1 (10.0)  No  4 (28.6)  Female = 3 (75.0)  Male = 1 (25.0)  7  Follow-up/referral  GP  9 (64.3)  Female = 8 (88.9)  Male = 1 (11.1)  Neurologist  5 (35.7)  Female = 4 (80.0)  Male = 1 (20.0)  S. No  Variable    Number (%)  Number by gender (%)  1  Age range (as at time of second scans)  18–24  2 (14.3)  Female = 2 (100.0)  Male = 0 (0.0)  25–44  6 (42.9)  Female = 4 (66.7)  Male = 2 (33.3)  44–64  5 (35.7)  Female = 5 (100.0)  Male = 0 (0.0)  ≥65  1 (7.1)  Female = 1 (100.0)  Male = 0 (0.0)  2  Severity of headaches  Severe  9 (64.3)  Female = 7 (77.8)  Male = 2 (22.2)  Not recorded  5 (35.7)  Female = 5 (100.0)  Male = 0 (0.0)  3  Type of scan  CT head/brain  7 (50.0)  Female = 6 (85.7)  Male = 1 (14.3)  MRI head/brain  7 (50.0)  Female = 6 (85.7)  Male = 1 (14.3)  4  Findings from scan  Intracranial  4 (28.6)  Female = 4 (100.0)  Male = 0 (0.0)  Extracranial  3 (21.4)  Female = 2 (66.7)  Male = 1 (33.3)  No pathological finding  7 (50.0)  Female = 6 (85.7)  Male = 1 (14.3)  5  Findings from scan likely to explain headache  Yes  3 (21.4)  Female = 3 (100.0)  Male = 0 (0.0)  No  11 (78.6)  Female = 9 (81.8)  Male = 2 (18.2)  6  Red flags present or not  Yes  10 (71.4)  Female = 9 (90.0)  Male = 1 (10.0)  No  4 (28.6)  Female = 3 (75.0)  Male = 1 (25.0)  7  Follow-up/referral  GP  9 (64.3)  Female = 8 (88.9)  Male = 1 (11.1)  Neurologist  5 (35.7)  Female = 4 (80.0)  Male = 1 (20.0)  All 14 patients scanned the second time had recurrent headaches, out of which 12 (85.7%) were females, whereas 2 (14.3%) were males. All referrals for follow-ups were to neurologists and initiated by the GPs themselves. No referral was initiated by the patients themselves, and no referrals were made to neurosurgeons or other allied health therapists. View Large Table 4. Demographics and characteristics of patients investigated for the second time with head imaging for headaches over a 7-year period in an Australian general practice setting (n = 14) S. No  Variable    Number (%)  Number by gender (%)  1  Age range (as at time of second scans)  18–24  2 (14.3)  Female = 2 (100.0)  Male = 0 (0.0)  25–44  6 (42.9)  Female = 4 (66.7)  Male = 2 (33.3)  44–64  5 (35.7)  Female = 5 (100.0)  Male = 0 (0.0)  ≥65  1 (7.1)  Female = 1 (100.0)  Male = 0 (0.0)  2  Severity of headaches  Severe  9 (64.3)  Female = 7 (77.8)  Male = 2 (22.2)  Not recorded  5 (35.7)  Female = 5 (100.0)  Male = 0 (0.0)  3  Type of scan  CT head/brain  7 (50.0)  Female = 6 (85.7)  Male = 1 (14.3)  MRI head/brain  7 (50.0)  Female = 6 (85.7)  Male = 1 (14.3)  4  Findings from scan  Intracranial  4 (28.6)  Female = 4 (100.0)  Male = 0 (0.0)  Extracranial  3 (21.4)  Female = 2 (66.7)  Male = 1 (33.3)  No pathological finding  7 (50.0)  Female = 6 (85.7)  Male = 1 (14.3)  5  Findings from scan likely to explain headache  Yes  3 (21.4)  Female = 3 (100.0)  Male = 0 (0.0)  No  11 (78.6)  Female = 9 (81.8)  Male = 2 (18.2)  6  Red flags present or not  Yes  10 (71.4)  Female = 9 (90.0)  Male = 1 (10.0)  No  4 (28.6)  Female = 3 (75.0)  Male = 1 (25.0)  7  Follow-up/referral  GP  9 (64.3)  Female = 8 (88.9)  Male = 1 (11.1)  Neurologist  5 (35.7)  Female = 4 (80.0)  Male = 1 (20.0)  S. No  Variable    Number (%)  Number by gender (%)  1  Age range (as at time of second scans)  18–24  2 (14.3)  Female = 2 (100.0)  Male = 0 (0.0)  25–44  6 (42.9)  Female = 4 (66.7)  Male = 2 (33.3)  44–64  5 (35.7)  Female = 5 (100.0)  Male = 0 (0.0)  ≥65  1 (7.1)  Female = 1 (100.0)  Male = 0 (0.0)  2  Severity of headaches  Severe  9 (64.3)  Female = 7 (77.8)  Male = 2 (22.2)  Not recorded  5 (35.7)  Female = 5 (100.0)  Male = 0 (0.0)  3  Type of scan  CT head/brain  7 (50.0)  Female = 6 (85.7)  Male = 1 (14.3)  MRI head/brain  7 (50.0)  Female = 6 (85.7)  Male = 1 (14.3)  4  Findings from scan  Intracranial  4 (28.6)  Female = 4 (100.0)  Male = 0 (0.0)  Extracranial  3 (21.4)  Female = 2 (66.7)  Male = 1 (33.3)  No pathological finding  7 (50.0)  Female = 6 (85.7)  Male = 1 (14.3)  5  Findings from scan likely to explain headache  Yes  3 (21.4)  Female = 3 (100.0)  Male = 0 (0.0)  No  11 (78.6)  Female = 9 (81.8)  Male = 2 (18.2)  6  Red flags present or not  Yes  10 (71.4)  Female = 9 (90.0)  Male = 1 (10.0)  No  4 (28.6)  Female = 3 (75.0)  Male = 1 (25.0)  7  Follow-up/referral  GP  9 (64.3)  Female = 8 (88.9)  Male = 1 (11.1)  Neurologist  5 (35.7)  Female = 4 (80.0)  Male = 1 (20.0)  All 14 patients scanned the second time had recurrent headaches, out of which 12 (85.7%) were females, whereas 2 (14.3%) were males. All referrals for follow-ups were to neurologists and initiated by the GPs themselves. No referral was initiated by the patients themselves, and no referrals were made to neurosurgeons or other allied health therapists. View Large CTs and MRIs were, again, the only scans done, and were used in equal proportions of seven each. There were no findings in seven of the cases, whereas intracranial lesions were found in four, out of which three were likely to explain the headaches (one from CT and two from MRIs). Associations The results of the BLR analysis are shown in Table 5. The only significant findings were that females were more likely to report recurrent headaches (OR = 2.63; CI = 1.09–6.35; P = 0.03), whereas patients with identified psychological co-morbidities were less likely to have scan findings that explained their headaches (OR = 0.22; CI = 0.06–0.88; P = 0.03). It was also found that patients with psycho-morbidities were more likely to be investigated with head imaging compared with those without such co-morbidities (OR = 1.47; CI = 0.96–2.27; P = 0.08). However, this latter finding slightly missed statistical significance. Table 5 Binary logistics regression of associations between patient characteristics and independent variables among patients investigated with head scans for headaches in an Australian general practice setting over a 7-year period (all at first scans) S. No  Dependent variable (only patients who had neuroimaging, unless otherwise stated)  Independent variables  Odds ratio (OR)  95% CI of OR  Significance (P value)  Lower  Upper  1  Gender (female versus male) (n = 109)  Headache recurrence (yes versus no)  2.63  1.09  6.35  0.03**  Findings on CT/MRI head likely to explain headache (yes versus no)  4.24  0.52  34.47  0.18  2  Psychological co-morbidity (present versus not present) (n = 109)  Headache recurrence (yes versus no)  0.57  0.26  1.23  0.15  Findings on CT/MRI head likely to explain headache (yes versus no)  0.22  0.06  0.88  0.03**  3  Age group 1 (‘≥18–≤24’ versus others) (n = 20)  Headache recurrence (recurrent versus not recurrent)  0.91  0.34  2.42  0.85  Findings on CT/MRI head likely to explain headache (yes versus no)  0.88  0.18  4.36  0.87  4  Age group 2 (‘≥25–≤44’ versus others) (n = 46)  Headache recurrence (recurrent versus not recurrent)  1.37  0.53  2.45  0.74  Findings on CT/MRI head likely to explain headache (yes versus no)  0.66  0.19  2.32  0.51  5  Age group 3 (‘≥45–≤64’ versus others) (n = 34)  Headache recurrence (recurrent versus not recurrent)  1.12  0.49  2.55  0.78  Findings on CT/MRI head likely to explain headache (yes versus no)  1.68  0.49  5.71  0.41  6  Age group 4 (‘≥65’ versus others) (n = 9)  Headache recurrence (recurrent versus not recurrent)  0.58  0.15  2.29  0.44  Findings on CT/MRI head likely to explain headache (yes versus no)  1.01  0.12  0.87  0.99  7  Psychological co-morbidity among all patients with headaches (present versus not present) (n = 517)  Neuroimaging done (yes versus no)  1.47  0.96  2.27  0.08*  S. No  Dependent variable (only patients who had neuroimaging, unless otherwise stated)  Independent variables  Odds ratio (OR)  95% CI of OR  Significance (P value)  Lower  Upper  1  Gender (female versus male) (n = 109)  Headache recurrence (yes versus no)  2.63  1.09  6.35  0.03**  Findings on CT/MRI head likely to explain headache (yes versus no)  4.24  0.52  34.47  0.18  2  Psychological co-morbidity (present versus not present) (n = 109)  Headache recurrence (yes versus no)  0.57  0.26  1.23  0.15  Findings on CT/MRI head likely to explain headache (yes versus no)  0.22  0.06  0.88  0.03**  3  Age group 1 (‘≥18–≤24’ versus others) (n = 20)  Headache recurrence (recurrent versus not recurrent)  0.91  0.34  2.42  0.85  Findings on CT/MRI head likely to explain headache (yes versus no)  0.88  0.18  4.36  0.87  4  Age group 2 (‘≥25–≤44’ versus others) (n = 46)  Headache recurrence (recurrent versus not recurrent)  1.37  0.53  2.45  0.74  Findings on CT/MRI head likely to explain headache (yes versus no)  0.66  0.19  2.32  0.51  5  Age group 3 (‘≥45–≤64’ versus others) (n = 34)  Headache recurrence (recurrent versus not recurrent)  1.12  0.49  2.55  0.78  Findings on CT/MRI head likely to explain headache (yes versus no)  1.68  0.49  5.71  0.41  6  Age group 4 (‘≥65’ versus others) (n = 9)  Headache recurrence (recurrent versus not recurrent)  0.58  0.15  2.29  0.44  Findings on CT/MRI head likely to explain headache (yes versus no)  1.01  0.12  0.87  0.99  7  Psychological co-morbidity among all patients with headaches (present versus not present) (n = 517)  Neuroimaging done (yes versus no)  1.47  0.96  2.27  0.08*  **Statistically significant. *Narrowly missed statistical significance. View Large Table 5 Binary logistics regression of associations between patient characteristics and independent variables among patients investigated with head scans for headaches in an Australian general practice setting over a 7-year period (all at first scans) S. No  Dependent variable (only patients who had neuroimaging, unless otherwise stated)  Independent variables  Odds ratio (OR)  95% CI of OR  Significance (P value)  Lower  Upper  1  Gender (female versus male) (n = 109)  Headache recurrence (yes versus no)  2.63  1.09  6.35  0.03**  Findings on CT/MRI head likely to explain headache (yes versus no)  4.24  0.52  34.47  0.18  2  Psychological co-morbidity (present versus not present) (n = 109)  Headache recurrence (yes versus no)  0.57  0.26  1.23  0.15  Findings on CT/MRI head likely to explain headache (yes versus no)  0.22  0.06  0.88  0.03**  3  Age group 1 (‘≥18–≤24’ versus others) (n = 20)  Headache recurrence (recurrent versus not recurrent)  0.91  0.34  2.42  0.85  Findings on CT/MRI head likely to explain headache (yes versus no)  0.88  0.18  4.36  0.87  4  Age group 2 (‘≥25–≤44’ versus others) (n = 46)  Headache recurrence (recurrent versus not recurrent)  1.37  0.53  2.45  0.74  Findings on CT/MRI head likely to explain headache (yes versus no)  0.66  0.19  2.32  0.51  5  Age group 3 (‘≥45–≤64’ versus others) (n = 34)  Headache recurrence (recurrent versus not recurrent)  1.12  0.49  2.55  0.78  Findings on CT/MRI head likely to explain headache (yes versus no)  1.68  0.49  5.71  0.41  6  Age group 4 (‘≥65’ versus others) (n = 9)  Headache recurrence (recurrent versus not recurrent)  0.58  0.15  2.29  0.44  Findings on CT/MRI head likely to explain headache (yes versus no)  1.01  0.12  0.87  0.99  7  Psychological co-morbidity among all patients with headaches (present versus not present) (n = 517)  Neuroimaging done (yes versus no)  1.47  0.96  2.27  0.08*  S. No  Dependent variable (only patients who had neuroimaging, unless otherwise stated)  Independent variables  Odds ratio (OR)  95% CI of OR  Significance (P value)  Lower  Upper  1  Gender (female versus male) (n = 109)  Headache recurrence (yes versus no)  2.63  1.09  6.35  0.03**  Findings on CT/MRI head likely to explain headache (yes versus no)  4.24  0.52  34.47  0.18  2  Psychological co-morbidity (present versus not present) (n = 109)  Headache recurrence (yes versus no)  0.57  0.26  1.23  0.15  Findings on CT/MRI head likely to explain headache (yes versus no)  0.22  0.06  0.88  0.03**  3  Age group 1 (‘≥18–≤24’ versus others) (n = 20)  Headache recurrence (recurrent versus not recurrent)  0.91  0.34  2.42  0.85  Findings on CT/MRI head likely to explain headache (yes versus no)  0.88  0.18  4.36  0.87  4  Age group 2 (‘≥25–≤44’ versus others) (n = 46)  Headache recurrence (recurrent versus not recurrent)  1.37  0.53  2.45  0.74  Findings on CT/MRI head likely to explain headache (yes versus no)  0.66  0.19  2.32  0.51  5  Age group 3 (‘≥45–≤64’ versus others) (n = 34)  Headache recurrence (recurrent versus not recurrent)  1.12  0.49  2.55  0.78  Findings on CT/MRI head likely to explain headache (yes versus no)  1.68  0.49  5.71  0.41  6  Age group 4 (‘≥65’ versus others) (n = 9)  Headache recurrence (recurrent versus not recurrent)  0.58  0.15  2.29  0.44  Findings on CT/MRI head likely to explain headache (yes versus no)  1.01  0.12  0.87  0.99  7  Psychological co-morbidity among all patients with headaches (present versus not present) (n = 517)  Neuroimaging done (yes versus no)  1.47  0.96  2.27  0.08*  **Statistically significant. *Narrowly missed statistical significance. View Large Discussion The finding of a predominance of females (74.3% in first scans; 85.7% in second scans) among the patients who required imaging for headaches is grossly consistent with the report of the Australian Bettering the Evaluation and Care of Health (BEACH) Study (8), which found that 63.4% of all headache presentations were among females. Also, the observation that the age group of 25–44 required head scans the most (42.2% in first scans; 42.9% in second scans), whereas the age group ≥65 the least (8.3% and 7.1%, respectively), also corresponds to the BEACH report, which concluded that these age groups present with the most and the least headaches in Australia, respectively (8). The consistency of our findings on gender and age with the reports of the BEACH study implies that our data follow established patterns and are therefore arguably representative of the wider Australian population. An important point to note is that only 11% of the patients undergoing the first scans had findings that were likely to have explained their headaches. This finding is consistent with reports in the published literature, which had variously reported that dangerous findings following imaging for head scans are low (1,7). It is not surprising that no other scanning techniques apart from CTs and MRIs were used for the imaging investigations, with the former being used in more over 80% of cases. This is similar to the report of the BEACH study, which found CT scans as the highest imaging technique adopted for headache investigations (8). This finding is also consistent with the recommendations of available guidelines in Australia, which advocates CTs primarily (and MRIs secondarily) as imaging tools for non-acute headaches warranting imaging investigations (22). Cost and availability may also be factors in the higher use of CTs, given that they are more widely available and cheaper compared with MRIs (23). We also found that the rate of MRIs increased if repeat scans for headaches were needed (50% each for CT and MRI, respectively). Again, this is in line with the available guidelines (22), which seem to be based on the claim that MRIs are more sensitive than CTs in investigating intracranial pathologies (23), meaning that clinicians are more likely to adopt it if the initial query was unsatisfactorily resolved with the original scans. However, it is interesting to note that we found no statistical differences between the use of MRIs and CTs with respect to finding significant intracranial pathologies that might explain headaches (P = 0.41). This finding appears to be supported by the conclusions of a paper that reviewed all the relevant studies from 1966 to 1998, which looked at the sensitivity, specificity and predictive values of different neuroimaging techniques in investigating non-acute headaches (24). The paper found that, although MRIs may be superior to CTs in identifying white matter lesions and developmental anomalies (which were largely clinically insignificant in diagnosing causes of non-acute headaches), they offered no advantages in identifying clinically significant pathologies relevant to these headaches. Another interesting finding was that repeat scans, be it with CT or MRI, revealed no new pathologies different from the first ones. The biggest time lapse between initial and the repeat scans was 5 years, leading us to infer that repeating head imaging for recurrent headaches within a 5-year period offer no additional benefits. This conclusion may be taken with caution though, given the small samples involved, but larger studies will be required to disprove it. Should this observation be replicated in larger studies, the implications for policy and practice are huge, given that it may then be more clinically beneficial to refer patients with recurrent headaches to third-party practitioners, rather than to go on to repeat scans if they re-present within a period of 5 years from an earlier scan. A key component of the discussion from the preceding paragraph is the fact that the use of MRIs on repeat scans after initial CT scans identified no new pathologies. This appears to re-confirm our conclusion in the earlier paragraph that MRIs have no advantage over CTs in investigating headaches in general practice, a finding backed by at least one previous publication (24) as already stated. Given the cost benefits of CTs over MRIs, these observations, if replicated in larger studies, will have significant cost-saving benefits (23). Yet another significant result from this work was that patients with identified psycho-morbidities were 78% less likely to have findings that will explain their headaches following head scans, relative to those without these co-morbidities (P = 0.03; Table 5). It is not clear if this is a result of the fact that practitioners tend to order more scans for patients with mental health issues, possibly out of higher level of ‘pressures’ known to be associated with these patients (4,11,25). On the other hand, it raises the question of whether patients with these psycho-morbidities are less likely to have structural brain pathologies than those without it. Whatever is the case, this finding is even made more significant given the observation that, among all the 517 patients seen for headaches in the 7-year period, those with psycho-morbidities were about 1.5 times more likely to get head imaging tests done when compared with those without such co-morbidities. Even though this latter observation narrowly missed statistical significance (P−0.08), combined to the earlier one, it appears to make a case that psycho-morbidities need to be considered in all cases of headaches before head imaging is ordered, given that, among these group of patients, scans are unlikely to reveal useful findings, yet doctors order them more. Whether these findings are peculiar to the population studied, or reflects a wider national, or global phenomenon is hard to ascertain within the limitations of this study. Larger studies to explore these is recommended, given the huge practice implications this portends. Limitations One limitation of our study is that our work was based on the documentations of doctors over a 7-year period. Omissions are possible, particularly with respect to the characteristics of the headaches (severity, recurrence and red flags) and the documentation of psychological problems. However, given that many years were involved, spanning across multiple visits from patients across more than one GP in most cases, we believe that most data required to make reliable observations were captured for most of the patients analysed. Our approach of relying on proven pharmacological and psychological treatments for patients with mental health issues also helped limit potential errors from poor documentation on mental health issues, as this ensured that only patients with real psychological problems were identified as such. Another potential limitation is on the possibility of human error in data collection, particularly with respect to the radiological reports and clinical notes entered for the patients. We tried to limit these by having all entries checked twice. The fact that the demographics of our patients, and the baseline proportions reported were not much different from the reports of the BEACH study also reassures that our data were reasonably accurate. Finally, as already acknowledged in the main paper, involving a larger number of surgeries, through a national and/or international study, may be necessary to help validate or disprove the generalizability of the findings of this work. Conclusions We conclude that just over one patient out of five (21.1%) presenting with headaches in general required neuroimaging as part of their investigations, with the majority being in the age range of 25–44 years (42.2%). Females were nearly three times more likely to report recurrent headaches (P = 0.03). Only CT and MRI imaging scans were used, with the former preferred at a ratio of about 4:1. However, there was no difference in using either of them with respect to significant intracranial findings (P = 0.41). Just about 15% of the scans revealed intracranial findings, with only 11% likely to explain the headaches. About one of every four patients who had scans were referred to other non-GP providers, with neurologists accounting for over 60% of these referrals. Also, 44% of the patients requiring head scans had associated psychological co-morbidities, and these subgroups were 78% less likely to have lesions that will explain their headaches (P = 0.03). Also, even though statistical significance was narrowly missed, headache presentations among patients with psycho-morbidities were more likely to be investigated with neuroimaging when compared with those without such co-morbidities (P = 0.08). Finally, nearly 13% of the patients required repeat scans, and these were all within a 5-year period. However, these additional scans provided no additional findings to the initial scans, be it CT or MRI. Recommendations Firstly, we recommend that patients presenting with recurrent headaches after initial head scans may not benefit from a repeat scan within a 5-year period and that clinicians should consider referring to third-party providers if there are concerns or uncertainty with the diagnosis. Secondly, given that there are no statistically significant differences in the findings whether CTs or MRIs were used, we recommend that the cheaper and more widely available technique, which is CT, be adopted where head imaging for headaches is needed in general practice. Thirdly, we advise that need GPs pay more attention to the psycho-morbidities of patients presenting with headaches. A full evaluation may be important before ordering imaging and involving psychiatrists and psychologists may be worthwhile options. Finally, we advocate that larger studies of national and international dimensions are worth undertaking so as to fully explore all the above-mentioned findings, given the huge implications for policy and practice, as well as the potential cost–saving benefits involved. Acknowledgements We wish to acknowledge the immense support from Mrs. Maureen Firehock, the Managing Director and CEO of all three practices used in this work (Kennedy Drive, PKG and Bilambil Heights Medical Centers, Tweed Heads, Australia). We are also appreciative of the efforts of all the staff in these surgeries who assisted with the data collection and other logistics needed to make this study a success. Declaration Funding: No external source of funding was required for this study. Ethical approval: Ethical approval was obtained from the Human Research Ethics Committee (HREC) of the Griffith University (GU Ref No: 2017/354) prior to the commencement of our study. Conflicts of interest: None. References 1. Frese T, Druckrey H, Sandholzer H. Headache in general practice: frequency, management, and results of encounter. Int Sch Res Notices  2014; 2014: 169428. Google Scholar PubMed  2. Rasmussen BK, Jensen R, Schroll Met al.   Epidemiology of headache in a general population—a prevalence study. J Clin Epidemiol  1991; 44: 1147– 57. Google Scholar CrossRef Search ADS PubMed  3. Latinovic R, Gulliford M, Ridsdale L. Headache and migraine in primary care: consultation, prescription, and referral rates in a large population. J Neurol Neurosurg Psychiatry  2006; 77: 385– 7. Google Scholar CrossRef Search ADS PubMed  4. O’Flynn N, Ridsdale L. Headache in primary care: how important is diagnosis to management? Br J Gen Pract  2002; 52: 569– 73. Google Scholar PubMed  5. Beran RG. Management of chronic headache. Aust Fam Physician  2014; 43: 106– 10. Google Scholar PubMed  6. Wonderling D, Vickers AJ, Grieve Ret al.   Cost effectiveness analysis of a randomised trial of acupuncture for chronic headache in primary care. BMJ  2004; 328: 747. Google Scholar CrossRef Search ADS PubMed  7. Kernick D, Stapley S, Hamilton W. GPs’ classification of headache: is primary headache underdiagnosed? Br J Gen Pract  2008; 58: 102– 4. Google Scholar CrossRef Search ADS PubMed  8. Britt H, Valenti L, Miller Get al.   Presentations of diarrhoea in Australian general practice. Aust Fam Physician  2005; 34: 218– 9. Google Scholar PubMed  9. Stanley IM, Peters S, Salmon P. A primary care perspective on prevailing assumptions about persistent medically unexplained physical symptoms. Int J Psychiatry Med  2002; 32: 125– 40. Google Scholar CrossRef Search ADS PubMed  10. Page LA, Howard LM, Husain Ket al.   Psychiatric morbidity and cognitive representations of illness in chronic daily headache. J Psychosom Res  2004; 57: 549– 55. Google Scholar CrossRef Search ADS PubMed  11. Couchman GR, Forjuoh SN, Rajab MHet al.   Nonclinical factors associated with primary care physicians’ ordering patterns of magnetic resonance imaging/computed tomography for headache. Acad Radiol  2004; 11: 735– 40. Google Scholar CrossRef Search ADS PubMed  12. Kernick DP, Ahmed F, Bahra Aet al.   Imaging patients with suspected brain tumour: guidance for primary care. Br J Gen Pract  2008; 58: 880– 5. Google Scholar CrossRef Search ADS PubMed  13. De Klippel N, Jansen JP, Carlos JS. Survey to evaluate diagnosis and management of headache in primary care: headache management pattern programme. Curr Med Res Opin  2008; 24: 3413– 22. Google Scholar CrossRef Search ADS PubMed  14. Kowacs PA, Twardowschy CA, Piovesan EJet al.   General practice physician knowledge about headache: evaluation of the municipal continual medical education program. Arq Neuropsiquiatr  2009; 67: 595– 9. Google Scholar CrossRef Search ADS PubMed  15. Stewart MA. Effective physician-patient communication and health outcomes: a review. CMAJ  1995; 152: 1423– 33. Google Scholar PubMed  16. Sandrini G, Friberg L, Jänig Wet al.   Neurophysiological tests and neuroimaging procedures in non-acute headache: guidelines and recommendations. Eur J Neurol  2004; 11: 217– 24. Google Scholar CrossRef Search ADS PubMed  17. American Academy of Neurology. Report of the quality standards sub-committee of the American Academy of Neurology. The utility of neuro imaging in the evaluation of headache in patients with normal neurological examinations . 2008. 18. Tweed Shire Council. The Tweed 2017. http://www.tweed.nsw.gov.au/TheTweed (accessed on 17 July 2017). 19. Australian Bureau of Statistics. 2011 Census QuickStats: Tweed Heads 2017. http://www.censusdata.abs.gov.au/census_services/getproduct/census/2011/quickstat/SSC12332 (accessed on 26 September 2017). 20. Citipedia. Gold Coast, Australia—list of cities, towns, villages 2017. http://www.citipedia.info/province/general/Australia_Gold+Coast (accessed on 17 July 2017). 21. City of Gold Coast. Gold Coast population fact sheets 2017. http://www.goldcoast.qld.gov.au/social-planning-and-development-factsheets-2006-2010-5152.html (accessed on 17 July 2017). 22. Government of Western Australia. Diagnostic imaging pathways—headache (adult) 2017. http://www.imagingpathways.health.wa.gov.au/index.php/imaging-pathways/neurological/headache#pathway (accessed on 20 July 2017). 23. De Luca GC, Bartleson JD. When and how to investigate the patient with headache. Semin Neurol  2010; 30: 131– 44. Google Scholar CrossRef Search ADS PubMed  24. Frishberg BM, Rosenberg JH, Matchar DBet al.   Evidence-Based Guidelines in the Primary Care Setting: Neuroimaging in Patients with Nonacute Headache . St Paul, MN: US Headache Consortium, 2000. 25. Morgan M, Jenkins L, Ridsdale L. Patient pressure for referral for headache: a qualitative study of GPs’ referral behaviour. Br J Gen Pract  2007; 57: 29– 35. Google Scholar PubMed  © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Family Practice Oxford University Press

Insights into radiographic investigations for headaches in general practice

Family Practice , Volume Advance Article – Jan 24, 2018

Loading next page...
 
/lp/ou_press/insights-into-radiographic-investigations-for-headaches-in-general-nMi2NGyftR
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
ISSN
0263-2136
eISSN
1460-2229
D.O.I.
10.1093/fampra/cmx136
Publisher site
See Article on Publisher Site

Abstract

Abstract Background/Objective Headaches present commonly in general practice, and doctors face uncertainties and pressures in dealing with them. This study identifies key characteristics of headaches investigated through head imaging, in the hope of strengthening existing guidelines. Method A 7-year retrospective analysis of head imaging scans ordered for investigating headaches among patients aged ≥18 in an Australian general practice setting. Results A total of 109 of the 517 (21.1%) patients required head imaging, although 14 (2.7%) of these had repeat scans. Three-quarters were females, and most of the headaches were recurrent (56.9%), severe (62.4%) or had associated red flags (78.0%). Computed tomography (CT) and magnetic resonance imaging (MRI) were the only scans adopted (4:1 and 1:1 for first and repeat scans, respectively). Twelve (11.0%) scans had findings likely to explain the headaches after the initial scans, with no difference in findings between diagnoses from CTs and MRIs (P = 0.41). Repeat scans offered no additional benefits, and roughly one in three patients were referred to third-party carers (mostly neurologists). Females had more recurrent headaches (OR = 2.63; CI = 1.09–6.35; P = 0.03). Patients with psycho-morbidities were less likely to have scan findings that explained their headaches (OR = 0.22; CI = 0.06–0.88; P = 0.03), and, though not quite significant, were also more likely to undergo imaging (OR = 1.47; CI 0.96–2.27; P = 0.08). Conclusion Significant intracranial findings were uncommon following imaging for headaches, and MRIs offer no advantages over CTs. Repeating head scans within 5 years offers no clear benefits. Psycho-morbidities should be considered when deciding the imaging needs, given the lesser chance of findings. Larger studies will help validate these findings. Brain, general practice, head, headaches, imaging, investigation, migraines, primary care, scan Introduction Headaches are common reasons for consulting GPs (1–3) and can account for up to 1.5% of all general practice consultations (1). They are listed among the top 10 disabling illnesses by the WHO (2) and make up the most common new neurological complaints to both GPs and neurologists (3–5). The social, economic and personal costs of headaches are believed to be considerable (6,7). In Australia, findings from the 2005 Bettering the Evaluation and Care of Health (BEACH) study, covering data from 1998 to 2004, showed that headaches were seen at an annual rate of 1.9% per annum, with about 1.9 million yearly cases across the country (8). They are usually not the only symptom for patients presenting with it, and present as sole symptoms only in 30% of cases (8). Despite the foregoing statistics, dangerous causes of headaches are very rare (1,3,7), and most cases in general practice are treated by physical examination, drug prescriptions (1) and reassurances (4) only. However, where deemed relevant (like with the presence of ‘red flags’), GPs do refer patients for investigations (blood tests and imaging of head/brain) or specialist opinions (4). Such referrals are still issued in some cases where there are no ‘red flags’, or in cases where the headaches are not medically explained, perhaps due to high anxiety from the patients, or uncertainties and fear of medico-legal issues by the GPs (7,9–12). Red flags include sudden, severe headaches or those different to previous episodes, exacerbated by coughing, sneezing, straining, eye movements and postural changes, as well as those associated with blurred vision, stiff neck, rash, altered consciousness, confusion and focal neurological signs (5). Given the common nature of headache presentations, the uncertainty on the part of the GPs in diagnosis and management, the pressure from anxious patients and the potential medico-legal implications of missing important findings, GPs need to have some robust, evidence-based and up-to-date guidelines that will assist them in deciding when to investigate, who to investigate and when to refer to third-party practitioners (specialists and other allied health staff). In fact, a number of studies have claimed that managing headaches efficiently in general practice can be improved through diagnostic and treatment approaches (7,13,14). Evidence-based improvements in the management of headaches in primary care can help enhance both direct (avoiding unnecessary imaging tests) and indirect (by saving time and work-related implications) costs to patients and healthcare funders. Also, the management of headaches in general practice is one area where effective communication with patients can improve outcomes (15). Therefore, a good support network to GPs will help boost their confidence, and provide more pragmatic frameworks for delivering such communications, and this study aims to achieve this. In addition, our work will add to the existing database of studies on headaches in primary care, an area that has been surprisingly understudied, and provide recommendations for both policy changes and future research into the management of headaches in general practice. Unfortunately, researches into headaches in general practice have been poor (1), and a number of publications in the United Kingdom have argued that existing guidelines may be fraught with uncertainties (16,17). These existing guidelines, both in Australia and globally, can be strengthened with further research. To achieve the forgoing objectives, this study explored the imagings done to investigate headaches and identified the characteristics of those headaches in terms of their severity, recurrence and red flags. The types of scans utilized, the proportion of scans that showed significant findings and the ensuing follow-up and referral patterns (to either specialists or allied health experts) after the scans were also analysed. The potential impact of psychological co-morbidities on the imaging requests ordered as part of the investigation of headaches was also explored. Methods Setting This is a 7-year retrospective study that explored the patterns of all brain and head (hereinafter used interchangeably) imaging scans ordered as part of investigations for patients presenting with headaches in an Australian general practice setting between the 1 January 2010 and 1 April 2017. The study focused on a group of three geographically distinct, but administratively and functionally inter-connected, group of general practice surgeries named ‘Kennedy Drive Medical Centre (KDMC)’, ‘PKG Medical Centre’ and the ‘Bilambil Heights Medical Centre (BHMC)’. All three are located in the ‘Tweed Shire Area’ of Northern New South Wales (NSW), Australia, also called the ‘Tweed Area’ or simply ‘The Tweed’ (18). ‘The Tweed’ covers about 1303 km2, and sits at the border of the states of NSW and Queensland, about 102 km to the South of Brisbane and 830 km North of Sydney (19). Alone, the Tweed’s population is approximately 90000 (18). However, the three surgeries are among a few others serving an estimated patient population of about 150000 (20,21), giving that patients that live in the Tweed, as well as the southern suburbs of the City of Gold Coast, all patronize them. The records and files of all patients, as well as the management system of all the three surgeries are inter-linked, and, combined, they have a registered patient population of about 31000, served by a total of about 11 doctors at each particular time (seven doctors in KDMC, three in PKG and two in BHMC). Demographics Even though the median age of the population in the Tweed Area was about 54 years according to the 2011 census, the larger ‘Gold Coast–Tweed Area’, which our study area focused on, has a median age of 38 years, a number similar to the Australian average of 37 years (19). The older median age in the Tweed is because the area is a hub of retirees, hosting about 14 separate retirement villages in the area (19). About 87% of the adult population in the Gold Coast–Tweed Area is employed (either full or part-time), with a male to female ratio of 51:49, and the child to family ratio of 2:1 (19). More than 80% of the residents are Caucasians, whereas 87% has English as their dominant first language (19). Inclusion and exclusion criteria The inclusion criteria included all adults aged ≥18 years of age at the time of their presentation, who saw a doctor within the defined period in any of the three identified practices, with headache as a major complaint, and requiring head imaging as part of their management. Headaches of all kinds were included, provided it was the main reason for the consultation. Patients who presented with headaches in this period, but had no imaging done, were excluded from the study. Also excluded were those who had head scans for other reasons like sinusitis, dizziness and suspected strokes, but did not present primarily with headaches. To be included, the requests for the head scans must have been ordered on the same day of the consultation for the headaches. Patient selection The ‘search tool’ in the Medical Director® software was used to identify all the patients seen in the defined period who met the study criteria. The search filter was broadened to include all forms of headaches and migraines with all their identified variations, including cervicogenic, sinus, chronic, cluster, tension, migraine (common, classical, cluster, migrainous aura) and posthead injury. All patients identified were then manually reviewed to ensure that they met all the inclusion–exclusion criteria. All potential forms of imaging were allowed, including computed tomography (CT), magnetic resonance imaging (MRI), magnetic resonance angiography, ultrasound and X-rays. Basic patient demographics, duration of headache, findings in the imaging results and immediate treatment offered were also obtained. Pre-defined terms were stated ahead of the selection, so as to avoid bias, and these were applied strictly. They are contained in Box 1. Box 1: Definition of terms Dates of the scans: Given that this is a retrospective study, the ages of the patients at the time of our data collection and analysis (May–June 2017) were expectedly different to the ages at the times they had the imaging scans. For this reason, the dates of their respective scans were collected and used to work out the patients’ ages at the times they did the scans. These latter ages were used in most of our analysis unless otherwise stated. For the sake of the analysis, the age categorization was into the four groups of ‘18–24’, ‘25–44’, ‘45–64’ and ‘≥65’, following the same pattern established by the popular Australian Bettering the Evaluation and Care of Health study (8). Severity of the headaches were recorded as documented in the patients’ notes. Where no such documentation is found, we recorded it as ‘unknown’. For the purpose of our work, headaches are classified as ‘recurrent’if it is documented as such, or if there were two or more documented primary presentations of headaches in the period surveyed. Findings from the scan: The findings were grouped into ‘none’, ‘intracranial’ and ‘extracranial’ findings. Given that some of the intracranial findings were likely to be incidental, they were further divided into those likely to explain the headaches, and those unlikely to do so. Identified red flags that informed the scans: A number of danger signs that would alert the GPs to the need to carry out imaging scans have been identified and listed earlier. Most GPs document these as the basis to justify doing scans for patients presenting with headaches. Where possible from the records, data on these were collected. Psychological co-morbidities of the patients: Extra care was taken to identify all cases of psychological co-morbidities existing in the patients that were seen for headaches. These include depression, anxiety, post-traumatic stress disorder, bipolar diseases and stress. To be included as a co-morbidity for a patient, there must be evidence of either a pharmacological (medications) or psychological (correspondences with psychologists and psychiatrists, including treatments offered through Mental Health Care Plans or otherwise) treatment for the disorder. This criterion was included to avoid spurious or transient forms of psychological disorders like in cases used as excuses for work absence certificates, minor grief situations, and transient work stress. Follow-up/referrals following the scans: All follow ups directly related to the imaging scans were noted. The kinds of third-party referrals to either specialist doctors or allied health staff patients were recorded, and where possible, those who initiated the referrals were identified (patients themselves or their GPs). Pre-defined terms were stated ahead of the selection, so as to avoid bias, and these were applied strictly and are defined. Data collection Data collection was done between May and June of 2017. For each patient that met the study criteria, data were collected on gender, date of birth, type(s) of scan done, date(s) of the scans, severity of the headaches, recurrence of the headaches, finding(s) from the scan(s), identified red flags that necessitated the scans, psychological co-morbidities of the patients and follow-up/referrals following the scans. In cases where a patient had more than one scan in the 7-year period, relevant data similar to that previously described were also collected (up to a maximum of two consultations), provided that the reason for the subsequent scans was for the investigation of a headache complaint. Data analysis Analysis was with the IBM SPSS Version 24. To identify significant associations, binary logistics regression (BLR) was used. The probability (P value) was set at <0.05. One part of the regression analysis focused solely on the patients who had head imaging, and for this, three outcome (dependent) variables were tested (‘gender’, ‘age’ and ‘psychological co-morbidities’) against two predictor (independent) variables (‘recurrence of headaches’ and ‘findings on imaging likely to explain headaches or not’). Results We found that 109 of the 517 (21.1%) adult patients who presented with headaches over the 7-year period required imaging as part of the investigations for their headaches. Scan 1 The reports for the first set of scans are summarized in Table 1. The age range (in years) at the time the scans were done, ranged from 18 to 88, with an average of 41.2 years (SD 15.47). The majority were females (81% or 74.3%). Only CT and MRI scans were done, with CTs used in 91 of the 109 patients (83.5%). Table 1. Demographics and characteristics of patients investigated initially with head imaging for headaches over a 7-year period in an Australian general practice setting (n = 109) S. No  Patient variable  Number (%)  Number by gender (%)  1  Gender  Female  81 (74.3)  Male  28 (25.7)  2  Age range (as at time of second scans)  18–24  20 (18.3)  Female = 14 (70.0)  Male = 6 (30.0)  25–44  46 (42.2)  Female = 33 (71.7)  Male = 13 (28.3)  44–64  34 (31.2)  Female = 27 (79.4)  Male = 7 (20.6)  ≥65  9 (8.3)  Female = 7 (77.8)  Male = 2 (22.2)  3  Recurrence of headache  Recurrent  62 (56.9)  Female = 51 (82.3)  Male = 11 (17.7)  Not recurrent  47 (43.1)  Female = 30 (63.8)  Male = 17 (36.2)  4  Severity of headaches  Severe  68 (62.4)  Female = 46 (67.6)  Male = 22 (32.4)  Not recorded  41 (37.6)  Female = 35 (85.4)  Male = 6 (14.6)  5  Type of scan  CT head/brain  91 (83.5)  Female = 67 (73.6)  Male = 24 (26.4)  MRI head/brain  18 (16.5)  Female = 14 (77.8)  Male = 4 (22.2)  6  Findings from scan  Intracranial  17 (15.6)  Female = 15 (88.2)  Male = 2 (11.8)  Extracranial  14 (12.8)  Female = 7 (50.0)  Male = 7 (50.0)  No pathological finding  78 (71.6)  Female = 59 (75.6)  Male = 19 (24.4)  7  Findings from scan likely to explain headache  Yes  12 (11.0)  Female = 11 (91.7)  Male = 1 (8.3)  No  97 (89.0)  Female = 70 (72.2)  Male = 27 (27.8)  8  Red flags present or not  Yes  85 (78.0)  Female = 65 (76.5)  Male = 20 (23.5)  No  24 (22.0)  Female = 16 (66.7)  Male = 8 (33.3)  9  Follow-up/referral  Of the 28 patients referred to specialists and other allied health experts, 2 (both females, and one each to neurologist and neurosurgeon) requested for the referrals themselves, whereas the other referrals were initiated by the GPs  GP  81 (74.3)  Female = 57 (70.4)  Male = 24 (29.6)  Neurologist  18 (16.5)  Female = 15 (83.3)  Male = 3 (16.7)  Neurosurgeon  3 (2.8)  Female = 3 (100.0)  Male = 0 (0.0)  ENT  3 (2.8)  Female = 2 (66.7)  Male = 1 (33.3)  Optometrist  2 (1.8)  Female = 2 (100.0)  Male = 0 (0.0)  Physiotherapist  2 (1.8)  Female = 2 (100.0)  Male = 0 (0.0)  10  Psychological co-morbidities (including depression, anxiety, schizophrenia and bipolar disorder)  Present  48 (44.0)  Female = 40 (83.3)  Male = 8 (16.7)  Absent  61 (56.0)  Female = 41 (67.2)  Male = 20 (32.8)  S. No  Patient variable  Number (%)  Number by gender (%)  1  Gender  Female  81 (74.3)  Male  28 (25.7)  2  Age range (as at time of second scans)  18–24  20 (18.3)  Female = 14 (70.0)  Male = 6 (30.0)  25–44  46 (42.2)  Female = 33 (71.7)  Male = 13 (28.3)  44–64  34 (31.2)  Female = 27 (79.4)  Male = 7 (20.6)  ≥65  9 (8.3)  Female = 7 (77.8)  Male = 2 (22.2)  3  Recurrence of headache  Recurrent  62 (56.9)  Female = 51 (82.3)  Male = 11 (17.7)  Not recurrent  47 (43.1)  Female = 30 (63.8)  Male = 17 (36.2)  4  Severity of headaches  Severe  68 (62.4)  Female = 46 (67.6)  Male = 22 (32.4)  Not recorded  41 (37.6)  Female = 35 (85.4)  Male = 6 (14.6)  5  Type of scan  CT head/brain  91 (83.5)  Female = 67 (73.6)  Male = 24 (26.4)  MRI head/brain  18 (16.5)  Female = 14 (77.8)  Male = 4 (22.2)  6  Findings from scan  Intracranial  17 (15.6)  Female = 15 (88.2)  Male = 2 (11.8)  Extracranial  14 (12.8)  Female = 7 (50.0)  Male = 7 (50.0)  No pathological finding  78 (71.6)  Female = 59 (75.6)  Male = 19 (24.4)  7  Findings from scan likely to explain headache  Yes  12 (11.0)  Female = 11 (91.7)  Male = 1 (8.3)  No  97 (89.0)  Female = 70 (72.2)  Male = 27 (27.8)  8  Red flags present or not  Yes  85 (78.0)  Female = 65 (76.5)  Male = 20 (23.5)  No  24 (22.0)  Female = 16 (66.7)  Male = 8 (33.3)  9  Follow-up/referral  Of the 28 patients referred to specialists and other allied health experts, 2 (both females, and one each to neurologist and neurosurgeon) requested for the referrals themselves, whereas the other referrals were initiated by the GPs  GP  81 (74.3)  Female = 57 (70.4)  Male = 24 (29.6)  Neurologist  18 (16.5)  Female = 15 (83.3)  Male = 3 (16.7)  Neurosurgeon  3 (2.8)  Female = 3 (100.0)  Male = 0 (0.0)  ENT  3 (2.8)  Female = 2 (66.7)  Male = 1 (33.3)  Optometrist  2 (1.8)  Female = 2 (100.0)  Male = 0 (0.0)  Physiotherapist  2 (1.8)  Female = 2 (100.0)  Male = 0 (0.0)  10  Psychological co-morbidities (including depression, anxiety, schizophrenia and bipolar disorder)  Present  48 (44.0)  Female = 40 (83.3)  Male = 8 (16.7)  Absent  61 (56.0)  Female = 41 (67.2)  Male = 20 (32.8)  View Large Table 1. Demographics and characteristics of patients investigated initially with head imaging for headaches over a 7-year period in an Australian general practice setting (n = 109) S. No  Patient variable  Number (%)  Number by gender (%)  1  Gender  Female  81 (74.3)  Male  28 (25.7)  2  Age range (as at time of second scans)  18–24  20 (18.3)  Female = 14 (70.0)  Male = 6 (30.0)  25–44  46 (42.2)  Female = 33 (71.7)  Male = 13 (28.3)  44–64  34 (31.2)  Female = 27 (79.4)  Male = 7 (20.6)  ≥65  9 (8.3)  Female = 7 (77.8)  Male = 2 (22.2)  3  Recurrence of headache  Recurrent  62 (56.9)  Female = 51 (82.3)  Male = 11 (17.7)  Not recurrent  47 (43.1)  Female = 30 (63.8)  Male = 17 (36.2)  4  Severity of headaches  Severe  68 (62.4)  Female = 46 (67.6)  Male = 22 (32.4)  Not recorded  41 (37.6)  Female = 35 (85.4)  Male = 6 (14.6)  5  Type of scan  CT head/brain  91 (83.5)  Female = 67 (73.6)  Male = 24 (26.4)  MRI head/brain  18 (16.5)  Female = 14 (77.8)  Male = 4 (22.2)  6  Findings from scan  Intracranial  17 (15.6)  Female = 15 (88.2)  Male = 2 (11.8)  Extracranial  14 (12.8)  Female = 7 (50.0)  Male = 7 (50.0)  No pathological finding  78 (71.6)  Female = 59 (75.6)  Male = 19 (24.4)  7  Findings from scan likely to explain headache  Yes  12 (11.0)  Female = 11 (91.7)  Male = 1 (8.3)  No  97 (89.0)  Female = 70 (72.2)  Male = 27 (27.8)  8  Red flags present or not  Yes  85 (78.0)  Female = 65 (76.5)  Male = 20 (23.5)  No  24 (22.0)  Female = 16 (66.7)  Male = 8 (33.3)  9  Follow-up/referral  Of the 28 patients referred to specialists and other allied health experts, 2 (both females, and one each to neurologist and neurosurgeon) requested for the referrals themselves, whereas the other referrals were initiated by the GPs  GP  81 (74.3)  Female = 57 (70.4)  Male = 24 (29.6)  Neurologist  18 (16.5)  Female = 15 (83.3)  Male = 3 (16.7)  Neurosurgeon  3 (2.8)  Female = 3 (100.0)  Male = 0 (0.0)  ENT  3 (2.8)  Female = 2 (66.7)  Male = 1 (33.3)  Optometrist  2 (1.8)  Female = 2 (100.0)  Male = 0 (0.0)  Physiotherapist  2 (1.8)  Female = 2 (100.0)  Male = 0 (0.0)  10  Psychological co-morbidities (including depression, anxiety, schizophrenia and bipolar disorder)  Present  48 (44.0)  Female = 40 (83.3)  Male = 8 (16.7)  Absent  61 (56.0)  Female = 41 (67.2)  Male = 20 (32.8)  S. No  Patient variable  Number (%)  Number by gender (%)  1  Gender  Female  81 (74.3)  Male  28 (25.7)  2  Age range (as at time of second scans)  18–24  20 (18.3)  Female = 14 (70.0)  Male = 6 (30.0)  25–44  46 (42.2)  Female = 33 (71.7)  Male = 13 (28.3)  44–64  34 (31.2)  Female = 27 (79.4)  Male = 7 (20.6)  ≥65  9 (8.3)  Female = 7 (77.8)  Male = 2 (22.2)  3  Recurrence of headache  Recurrent  62 (56.9)  Female = 51 (82.3)  Male = 11 (17.7)  Not recurrent  47 (43.1)  Female = 30 (63.8)  Male = 17 (36.2)  4  Severity of headaches  Severe  68 (62.4)  Female = 46 (67.6)  Male = 22 (32.4)  Not recorded  41 (37.6)  Female = 35 (85.4)  Male = 6 (14.6)  5  Type of scan  CT head/brain  91 (83.5)  Female = 67 (73.6)  Male = 24 (26.4)  MRI head/brain  18 (16.5)  Female = 14 (77.8)  Male = 4 (22.2)  6  Findings from scan  Intracranial  17 (15.6)  Female = 15 (88.2)  Male = 2 (11.8)  Extracranial  14 (12.8)  Female = 7 (50.0)  Male = 7 (50.0)  No pathological finding  78 (71.6)  Female = 59 (75.6)  Male = 19 (24.4)  7  Findings from scan likely to explain headache  Yes  12 (11.0)  Female = 11 (91.7)  Male = 1 (8.3)  No  97 (89.0)  Female = 70 (72.2)  Male = 27 (27.8)  8  Red flags present or not  Yes  85 (78.0)  Female = 65 (76.5)  Male = 20 (23.5)  No  24 (22.0)  Female = 16 (66.7)  Male = 8 (33.3)  9  Follow-up/referral  Of the 28 patients referred to specialists and other allied health experts, 2 (both females, and one each to neurologist and neurosurgeon) requested for the referrals themselves, whereas the other referrals were initiated by the GPs  GP  81 (74.3)  Female = 57 (70.4)  Male = 24 (29.6)  Neurologist  18 (16.5)  Female = 15 (83.3)  Male = 3 (16.7)  Neurosurgeon  3 (2.8)  Female = 3 (100.0)  Male = 0 (0.0)  ENT  3 (2.8)  Female = 2 (66.7)  Male = 1 (33.3)  Optometrist  2 (1.8)  Female = 2 (100.0)  Male = 0 (0.0)  Physiotherapist  2 (1.8)  Female = 2 (100.0)  Male = 0 (0.0)  10  Psychological co-morbidities (including depression, anxiety, schizophrenia and bipolar disorder)  Present  48 (44.0)  Female = 40 (83.3)  Male = 8 (16.7)  Absent  61 (56.0)  Female = 41 (67.2)  Male = 20 (32.8)  View Large Table 2 shows that 17 (15.6%) of the 31 (28.4%) patients scanned had intracranial findings, but as shown in Table 3, only 12 (11.0% of all patients) had findings that were likely to have explained their headaches. Of the 12 intracranial findings, 9 came from CTs and 3 from MRIs, with no significant differences in using either (OR = 0.55; CI = 0.13–2.27; P = 0.41). Table 2. All findings (intracranial and extracranial) on head scans done for patients presenting with headaches over a 7-year period in an Australian general practice setting   Group of findings  Number reported in scan 1  Number reported in scan 2  1  Atrophy (including age and non-age related, cerebellar, cortical, cerebral, early)  4  1  2  Prominent cerebrospinal spaces  2  1  Arachnoid cyst  1  1  3  Developmental venous anomaly  1  –  4  Traumatic attenuation  1  –  5  Combined atrophy + ischemia  1  –  6  Hyperintensities/demyelination  2  –  7  Meningioma  1  1  8  Sinus mucosal thickenings and sinus bone dysplasia  15  11  9  Ischemia (acute or chronic)  2  –  10  Combined cyst prominent cerebrospinal spaces  1  1    Total  31  7    Group of findings  Number reported in scan 1  Number reported in scan 2  1  Atrophy (including age and non-age related, cerebellar, cortical, cerebral, early)  4  1  2  Prominent cerebrospinal spaces  2  1  Arachnoid cyst  1  1  3  Developmental venous anomaly  1  –  4  Traumatic attenuation  1  –  5  Combined atrophy + ischemia  1  –  6  Hyperintensities/demyelination  2  –  7  Meningioma  1  1  8  Sinus mucosal thickenings and sinus bone dysplasia  15  11  9  Ischemia (acute or chronic)  2  –  10  Combined cyst prominent cerebrospinal spaces  1  1    Total  31  7  View Large Table 2. All findings (intracranial and extracranial) on head scans done for patients presenting with headaches over a 7-year period in an Australian general practice setting   Group of findings  Number reported in scan 1  Number reported in scan 2  1  Atrophy (including age and non-age related, cerebellar, cortical, cerebral, early)  4  1  2  Prominent cerebrospinal spaces  2  1  Arachnoid cyst  1  1  3  Developmental venous anomaly  1  –  4  Traumatic attenuation  1  –  5  Combined atrophy + ischemia  1  –  6  Hyperintensities/demyelination  2  –  7  Meningioma  1  1  8  Sinus mucosal thickenings and sinus bone dysplasia  15  11  9  Ischemia (acute or chronic)  2  –  10  Combined cyst prominent cerebrospinal spaces  1  1    Total  31  7    Group of findings  Number reported in scan 1  Number reported in scan 2  1  Atrophy (including age and non-age related, cerebellar, cortical, cerebral, early)  4  1  2  Prominent cerebrospinal spaces  2  1  Arachnoid cyst  1  1  3  Developmental venous anomaly  1  –  4  Traumatic attenuation  1  –  5  Combined atrophy + ischemia  1  –  6  Hyperintensities/demyelination  2  –  7  Meningioma  1  1  8  Sinus mucosal thickenings and sinus bone dysplasia  15  11  9  Ischemia (acute or chronic)  2  –  10  Combined cyst prominent cerebrospinal spaces  1  1    Total  31  7  View Large Table 3. Specific intracranial findings on head scans that are likely to explain headaches among patients presenting in an Australian general practice setting First scans  Second scans  Findings  Type of scan  Findings  Type of scan  Arachnoid cyst  CT  Intraventricular arachnoid cyst + calcification  CT  Significant distention of right ventricle + calcification + cyst  CT  Prominent ventricular system; atrophy; hydrocephalus  CT  Meningioma  MRI  Prominent cerebrospinal spaces (distention) + frontal atrophy (not with age)  CT  Hypodense R cerebral hemisphere (can be traumatic, infectious, malignant or vascular, etc.)  CT  Prominent cerebrospinal space  MRI  Attenuation of left parietal lobe posttrauma  CT  Microangiopathic white matter ischemia and vasculitis + generalized atrophy  MRI  Total = 3  CT = 1; MRI = 2  Developmental venous anomaly  CT  Subcortical intensities: likely vasculitis or ischemia  MRI  Meningioma  CT  Hyperintense white matter: likely demyelination or vasculitis  MRI  Ischemia (acute/subacute)  CT  Total = 12  CT = 9; MRI = 3  First scans  Second scans  Findings  Type of scan  Findings  Type of scan  Arachnoid cyst  CT  Intraventricular arachnoid cyst + calcification  CT  Significant distention of right ventricle + calcification + cyst  CT  Prominent ventricular system; atrophy; hydrocephalus  CT  Meningioma  MRI  Prominent cerebrospinal spaces (distention) + frontal atrophy (not with age)  CT  Hypodense R cerebral hemisphere (can be traumatic, infectious, malignant or vascular, etc.)  CT  Prominent cerebrospinal space  MRI  Attenuation of left parietal lobe posttrauma  CT  Microangiopathic white matter ischemia and vasculitis + generalized atrophy  MRI  Total = 3  CT = 1; MRI = 2  Developmental venous anomaly  CT  Subcortical intensities: likely vasculitis or ischemia  MRI  Meningioma  CT  Hyperintense white matter: likely demyelination or vasculitis  MRI  Ischemia (acute/subacute)  CT  Total = 12  CT = 9; MRI = 3  View Large Table 3. Specific intracranial findings on head scans that are likely to explain headaches among patients presenting in an Australian general practice setting First scans  Second scans  Findings  Type of scan  Findings  Type of scan  Arachnoid cyst  CT  Intraventricular arachnoid cyst + calcification  CT  Significant distention of right ventricle + calcification + cyst  CT  Prominent ventricular system; atrophy; hydrocephalus  CT  Meningioma  MRI  Prominent cerebrospinal spaces (distention) + frontal atrophy (not with age)  CT  Hypodense R cerebral hemisphere (can be traumatic, infectious, malignant or vascular, etc.)  CT  Prominent cerebrospinal space  MRI  Attenuation of left parietal lobe posttrauma  CT  Microangiopathic white matter ischemia and vasculitis + generalized atrophy  MRI  Total = 3  CT = 1; MRI = 2  Developmental venous anomaly  CT  Subcortical intensities: likely vasculitis or ischemia  MRI  Meningioma  CT  Hyperintense white matter: likely demyelination or vasculitis  MRI  Ischemia (acute/subacute)  CT  Total = 12  CT = 9; MRI = 3  First scans  Second scans  Findings  Type of scan  Findings  Type of scan  Arachnoid cyst  CT  Intraventricular arachnoid cyst + calcification  CT  Significant distention of right ventricle + calcification + cyst  CT  Prominent ventricular system; atrophy; hydrocephalus  CT  Meningioma  MRI  Prominent cerebrospinal spaces (distention) + frontal atrophy (not with age)  CT  Hypodense R cerebral hemisphere (can be traumatic, infectious, malignant or vascular, etc.)  CT  Prominent cerebrospinal space  MRI  Attenuation of left parietal lobe posttrauma  CT  Microangiopathic white matter ischemia and vasculitis + generalized atrophy  MRI  Total = 3  CT = 1; MRI = 2  Developmental venous anomaly  CT  Subcortical intensities: likely vasculitis or ischemia  MRI  Meningioma  CT  Hyperintense white matter: likely demyelination or vasculitis  MRI  Ischemia (acute/subacute)  CT  Total = 12  CT = 9; MRI = 3  View Large A total of 28 (25.7%) patients needed referrals to third-party carers other than their GPs following their initial scans, with nearly two-thirds of these being to neurologists, whereas 48 patients (44%) had known psychological co-morbidities (Table 1). Scan 2 This is summarized in Table 4, and show that only 14 of the 109 (12.8%) patients who had the initial scans required a repeat scan within the 7-year period surveyed. The longest time between the initial and the repeat scans was 5 years. Table 4. Demographics and characteristics of patients investigated for the second time with head imaging for headaches over a 7-year period in an Australian general practice setting (n = 14) S. No  Variable    Number (%)  Number by gender (%)  1  Age range (as at time of second scans)  18–24  2 (14.3)  Female = 2 (100.0)  Male = 0 (0.0)  25–44  6 (42.9)  Female = 4 (66.7)  Male = 2 (33.3)  44–64  5 (35.7)  Female = 5 (100.0)  Male = 0 (0.0)  ≥65  1 (7.1)  Female = 1 (100.0)  Male = 0 (0.0)  2  Severity of headaches  Severe  9 (64.3)  Female = 7 (77.8)  Male = 2 (22.2)  Not recorded  5 (35.7)  Female = 5 (100.0)  Male = 0 (0.0)  3  Type of scan  CT head/brain  7 (50.0)  Female = 6 (85.7)  Male = 1 (14.3)  MRI head/brain  7 (50.0)  Female = 6 (85.7)  Male = 1 (14.3)  4  Findings from scan  Intracranial  4 (28.6)  Female = 4 (100.0)  Male = 0 (0.0)  Extracranial  3 (21.4)  Female = 2 (66.7)  Male = 1 (33.3)  No pathological finding  7 (50.0)  Female = 6 (85.7)  Male = 1 (14.3)  5  Findings from scan likely to explain headache  Yes  3 (21.4)  Female = 3 (100.0)  Male = 0 (0.0)  No  11 (78.6)  Female = 9 (81.8)  Male = 2 (18.2)  6  Red flags present or not  Yes  10 (71.4)  Female = 9 (90.0)  Male = 1 (10.0)  No  4 (28.6)  Female = 3 (75.0)  Male = 1 (25.0)  7  Follow-up/referral  GP  9 (64.3)  Female = 8 (88.9)  Male = 1 (11.1)  Neurologist  5 (35.7)  Female = 4 (80.0)  Male = 1 (20.0)  S. No  Variable    Number (%)  Number by gender (%)  1  Age range (as at time of second scans)  18–24  2 (14.3)  Female = 2 (100.0)  Male = 0 (0.0)  25–44  6 (42.9)  Female = 4 (66.7)  Male = 2 (33.3)  44–64  5 (35.7)  Female = 5 (100.0)  Male = 0 (0.0)  ≥65  1 (7.1)  Female = 1 (100.0)  Male = 0 (0.0)  2  Severity of headaches  Severe  9 (64.3)  Female = 7 (77.8)  Male = 2 (22.2)  Not recorded  5 (35.7)  Female = 5 (100.0)  Male = 0 (0.0)  3  Type of scan  CT head/brain  7 (50.0)  Female = 6 (85.7)  Male = 1 (14.3)  MRI head/brain  7 (50.0)  Female = 6 (85.7)  Male = 1 (14.3)  4  Findings from scan  Intracranial  4 (28.6)  Female = 4 (100.0)  Male = 0 (0.0)  Extracranial  3 (21.4)  Female = 2 (66.7)  Male = 1 (33.3)  No pathological finding  7 (50.0)  Female = 6 (85.7)  Male = 1 (14.3)  5  Findings from scan likely to explain headache  Yes  3 (21.4)  Female = 3 (100.0)  Male = 0 (0.0)  No  11 (78.6)  Female = 9 (81.8)  Male = 2 (18.2)  6  Red flags present or not  Yes  10 (71.4)  Female = 9 (90.0)  Male = 1 (10.0)  No  4 (28.6)  Female = 3 (75.0)  Male = 1 (25.0)  7  Follow-up/referral  GP  9 (64.3)  Female = 8 (88.9)  Male = 1 (11.1)  Neurologist  5 (35.7)  Female = 4 (80.0)  Male = 1 (20.0)  All 14 patients scanned the second time had recurrent headaches, out of which 12 (85.7%) were females, whereas 2 (14.3%) were males. All referrals for follow-ups were to neurologists and initiated by the GPs themselves. No referral was initiated by the patients themselves, and no referrals were made to neurosurgeons or other allied health therapists. View Large Table 4. Demographics and characteristics of patients investigated for the second time with head imaging for headaches over a 7-year period in an Australian general practice setting (n = 14) S. No  Variable    Number (%)  Number by gender (%)  1  Age range (as at time of second scans)  18–24  2 (14.3)  Female = 2 (100.0)  Male = 0 (0.0)  25–44  6 (42.9)  Female = 4 (66.7)  Male = 2 (33.3)  44–64  5 (35.7)  Female = 5 (100.0)  Male = 0 (0.0)  ≥65  1 (7.1)  Female = 1 (100.0)  Male = 0 (0.0)  2  Severity of headaches  Severe  9 (64.3)  Female = 7 (77.8)  Male = 2 (22.2)  Not recorded  5 (35.7)  Female = 5 (100.0)  Male = 0 (0.0)  3  Type of scan  CT head/brain  7 (50.0)  Female = 6 (85.7)  Male = 1 (14.3)  MRI head/brain  7 (50.0)  Female = 6 (85.7)  Male = 1 (14.3)  4  Findings from scan  Intracranial  4 (28.6)  Female = 4 (100.0)  Male = 0 (0.0)  Extracranial  3 (21.4)  Female = 2 (66.7)  Male = 1 (33.3)  No pathological finding  7 (50.0)  Female = 6 (85.7)  Male = 1 (14.3)  5  Findings from scan likely to explain headache  Yes  3 (21.4)  Female = 3 (100.0)  Male = 0 (0.0)  No  11 (78.6)  Female = 9 (81.8)  Male = 2 (18.2)  6  Red flags present or not  Yes  10 (71.4)  Female = 9 (90.0)  Male = 1 (10.0)  No  4 (28.6)  Female = 3 (75.0)  Male = 1 (25.0)  7  Follow-up/referral  GP  9 (64.3)  Female = 8 (88.9)  Male = 1 (11.1)  Neurologist  5 (35.7)  Female = 4 (80.0)  Male = 1 (20.0)  S. No  Variable    Number (%)  Number by gender (%)  1  Age range (as at time of second scans)  18–24  2 (14.3)  Female = 2 (100.0)  Male = 0 (0.0)  25–44  6 (42.9)  Female = 4 (66.7)  Male = 2 (33.3)  44–64  5 (35.7)  Female = 5 (100.0)  Male = 0 (0.0)  ≥65  1 (7.1)  Female = 1 (100.0)  Male = 0 (0.0)  2  Severity of headaches  Severe  9 (64.3)  Female = 7 (77.8)  Male = 2 (22.2)  Not recorded  5 (35.7)  Female = 5 (100.0)  Male = 0 (0.0)  3  Type of scan  CT head/brain  7 (50.0)  Female = 6 (85.7)  Male = 1 (14.3)  MRI head/brain  7 (50.0)  Female = 6 (85.7)  Male = 1 (14.3)  4  Findings from scan  Intracranial  4 (28.6)  Female = 4 (100.0)  Male = 0 (0.0)  Extracranial  3 (21.4)  Female = 2 (66.7)  Male = 1 (33.3)  No pathological finding  7 (50.0)  Female = 6 (85.7)  Male = 1 (14.3)  5  Findings from scan likely to explain headache  Yes  3 (21.4)  Female = 3 (100.0)  Male = 0 (0.0)  No  11 (78.6)  Female = 9 (81.8)  Male = 2 (18.2)  6  Red flags present or not  Yes  10 (71.4)  Female = 9 (90.0)  Male = 1 (10.0)  No  4 (28.6)  Female = 3 (75.0)  Male = 1 (25.0)  7  Follow-up/referral  GP  9 (64.3)  Female = 8 (88.9)  Male = 1 (11.1)  Neurologist  5 (35.7)  Female = 4 (80.0)  Male = 1 (20.0)  All 14 patients scanned the second time had recurrent headaches, out of which 12 (85.7%) were females, whereas 2 (14.3%) were males. All referrals for follow-ups were to neurologists and initiated by the GPs themselves. No referral was initiated by the patients themselves, and no referrals were made to neurosurgeons or other allied health therapists. View Large CTs and MRIs were, again, the only scans done, and were used in equal proportions of seven each. There were no findings in seven of the cases, whereas intracranial lesions were found in four, out of which three were likely to explain the headaches (one from CT and two from MRIs). Associations The results of the BLR analysis are shown in Table 5. The only significant findings were that females were more likely to report recurrent headaches (OR = 2.63; CI = 1.09–6.35; P = 0.03), whereas patients with identified psychological co-morbidities were less likely to have scan findings that explained their headaches (OR = 0.22; CI = 0.06–0.88; P = 0.03). It was also found that patients with psycho-morbidities were more likely to be investigated with head imaging compared with those without such co-morbidities (OR = 1.47; CI = 0.96–2.27; P = 0.08). However, this latter finding slightly missed statistical significance. Table 5 Binary logistics regression of associations between patient characteristics and independent variables among patients investigated with head scans for headaches in an Australian general practice setting over a 7-year period (all at first scans) S. No  Dependent variable (only patients who had neuroimaging, unless otherwise stated)  Independent variables  Odds ratio (OR)  95% CI of OR  Significance (P value)  Lower  Upper  1  Gender (female versus male) (n = 109)  Headache recurrence (yes versus no)  2.63  1.09  6.35  0.03**  Findings on CT/MRI head likely to explain headache (yes versus no)  4.24  0.52  34.47  0.18  2  Psychological co-morbidity (present versus not present) (n = 109)  Headache recurrence (yes versus no)  0.57  0.26  1.23  0.15  Findings on CT/MRI head likely to explain headache (yes versus no)  0.22  0.06  0.88  0.03**  3  Age group 1 (‘≥18–≤24’ versus others) (n = 20)  Headache recurrence (recurrent versus not recurrent)  0.91  0.34  2.42  0.85  Findings on CT/MRI head likely to explain headache (yes versus no)  0.88  0.18  4.36  0.87  4  Age group 2 (‘≥25–≤44’ versus others) (n = 46)  Headache recurrence (recurrent versus not recurrent)  1.37  0.53  2.45  0.74  Findings on CT/MRI head likely to explain headache (yes versus no)  0.66  0.19  2.32  0.51  5  Age group 3 (‘≥45–≤64’ versus others) (n = 34)  Headache recurrence (recurrent versus not recurrent)  1.12  0.49  2.55  0.78  Findings on CT/MRI head likely to explain headache (yes versus no)  1.68  0.49  5.71  0.41  6  Age group 4 (‘≥65’ versus others) (n = 9)  Headache recurrence (recurrent versus not recurrent)  0.58  0.15  2.29  0.44  Findings on CT/MRI head likely to explain headache (yes versus no)  1.01  0.12  0.87  0.99  7  Psychological co-morbidity among all patients with headaches (present versus not present) (n = 517)  Neuroimaging done (yes versus no)  1.47  0.96  2.27  0.08*  S. No  Dependent variable (only patients who had neuroimaging, unless otherwise stated)  Independent variables  Odds ratio (OR)  95% CI of OR  Significance (P value)  Lower  Upper  1  Gender (female versus male) (n = 109)  Headache recurrence (yes versus no)  2.63  1.09  6.35  0.03**  Findings on CT/MRI head likely to explain headache (yes versus no)  4.24  0.52  34.47  0.18  2  Psychological co-morbidity (present versus not present) (n = 109)  Headache recurrence (yes versus no)  0.57  0.26  1.23  0.15  Findings on CT/MRI head likely to explain headache (yes versus no)  0.22  0.06  0.88  0.03**  3  Age group 1 (‘≥18–≤24’ versus others) (n = 20)  Headache recurrence (recurrent versus not recurrent)  0.91  0.34  2.42  0.85  Findings on CT/MRI head likely to explain headache (yes versus no)  0.88  0.18  4.36  0.87  4  Age group 2 (‘≥25–≤44’ versus others) (n = 46)  Headache recurrence (recurrent versus not recurrent)  1.37  0.53  2.45  0.74  Findings on CT/MRI head likely to explain headache (yes versus no)  0.66  0.19  2.32  0.51  5  Age group 3 (‘≥45–≤64’ versus others) (n = 34)  Headache recurrence (recurrent versus not recurrent)  1.12  0.49  2.55  0.78  Findings on CT/MRI head likely to explain headache (yes versus no)  1.68  0.49  5.71  0.41  6  Age group 4 (‘≥65’ versus others) (n = 9)  Headache recurrence (recurrent versus not recurrent)  0.58  0.15  2.29  0.44  Findings on CT/MRI head likely to explain headache (yes versus no)  1.01  0.12  0.87  0.99  7  Psychological co-morbidity among all patients with headaches (present versus not present) (n = 517)  Neuroimaging done (yes versus no)  1.47  0.96  2.27  0.08*  **Statistically significant. *Narrowly missed statistical significance. View Large Table 5 Binary logistics regression of associations between patient characteristics and independent variables among patients investigated with head scans for headaches in an Australian general practice setting over a 7-year period (all at first scans) S. No  Dependent variable (only patients who had neuroimaging, unless otherwise stated)  Independent variables  Odds ratio (OR)  95% CI of OR  Significance (P value)  Lower  Upper  1  Gender (female versus male) (n = 109)  Headache recurrence (yes versus no)  2.63  1.09  6.35  0.03**  Findings on CT/MRI head likely to explain headache (yes versus no)  4.24  0.52  34.47  0.18  2  Psychological co-morbidity (present versus not present) (n = 109)  Headache recurrence (yes versus no)  0.57  0.26  1.23  0.15  Findings on CT/MRI head likely to explain headache (yes versus no)  0.22  0.06  0.88  0.03**  3  Age group 1 (‘≥18–≤24’ versus others) (n = 20)  Headache recurrence (recurrent versus not recurrent)  0.91  0.34  2.42  0.85  Findings on CT/MRI head likely to explain headache (yes versus no)  0.88  0.18  4.36  0.87  4  Age group 2 (‘≥25–≤44’ versus others) (n = 46)  Headache recurrence (recurrent versus not recurrent)  1.37  0.53  2.45  0.74  Findings on CT/MRI head likely to explain headache (yes versus no)  0.66  0.19  2.32  0.51  5  Age group 3 (‘≥45–≤64’ versus others) (n = 34)  Headache recurrence (recurrent versus not recurrent)  1.12  0.49  2.55  0.78  Findings on CT/MRI head likely to explain headache (yes versus no)  1.68  0.49  5.71  0.41  6  Age group 4 (‘≥65’ versus others) (n = 9)  Headache recurrence (recurrent versus not recurrent)  0.58  0.15  2.29  0.44  Findings on CT/MRI head likely to explain headache (yes versus no)  1.01  0.12  0.87  0.99  7  Psychological co-morbidity among all patients with headaches (present versus not present) (n = 517)  Neuroimaging done (yes versus no)  1.47  0.96  2.27  0.08*  S. No  Dependent variable (only patients who had neuroimaging, unless otherwise stated)  Independent variables  Odds ratio (OR)  95% CI of OR  Significance (P value)  Lower  Upper  1  Gender (female versus male) (n = 109)  Headache recurrence (yes versus no)  2.63  1.09  6.35  0.03**  Findings on CT/MRI head likely to explain headache (yes versus no)  4.24  0.52  34.47  0.18  2  Psychological co-morbidity (present versus not present) (n = 109)  Headache recurrence (yes versus no)  0.57  0.26  1.23  0.15  Findings on CT/MRI head likely to explain headache (yes versus no)  0.22  0.06  0.88  0.03**  3  Age group 1 (‘≥18–≤24’ versus others) (n = 20)  Headache recurrence (recurrent versus not recurrent)  0.91  0.34  2.42  0.85  Findings on CT/MRI head likely to explain headache (yes versus no)  0.88  0.18  4.36  0.87  4  Age group 2 (‘≥25–≤44’ versus others) (n = 46)  Headache recurrence (recurrent versus not recurrent)  1.37  0.53  2.45  0.74  Findings on CT/MRI head likely to explain headache (yes versus no)  0.66  0.19  2.32  0.51  5  Age group 3 (‘≥45–≤64’ versus others) (n = 34)  Headache recurrence (recurrent versus not recurrent)  1.12  0.49  2.55  0.78  Findings on CT/MRI head likely to explain headache (yes versus no)  1.68  0.49  5.71  0.41  6  Age group 4 (‘≥65’ versus others) (n = 9)  Headache recurrence (recurrent versus not recurrent)  0.58  0.15  2.29  0.44  Findings on CT/MRI head likely to explain headache (yes versus no)  1.01  0.12  0.87  0.99  7  Psychological co-morbidity among all patients with headaches (present versus not present) (n = 517)  Neuroimaging done (yes versus no)  1.47  0.96  2.27  0.08*  **Statistically significant. *Narrowly missed statistical significance. View Large Discussion The finding of a predominance of females (74.3% in first scans; 85.7% in second scans) among the patients who required imaging for headaches is grossly consistent with the report of the Australian Bettering the Evaluation and Care of Health (BEACH) Study (8), which found that 63.4% of all headache presentations were among females. Also, the observation that the age group of 25–44 required head scans the most (42.2% in first scans; 42.9% in second scans), whereas the age group ≥65 the least (8.3% and 7.1%, respectively), also corresponds to the BEACH report, which concluded that these age groups present with the most and the least headaches in Australia, respectively (8). The consistency of our findings on gender and age with the reports of the BEACH study implies that our data follow established patterns and are therefore arguably representative of the wider Australian population. An important point to note is that only 11% of the patients undergoing the first scans had findings that were likely to have explained their headaches. This finding is consistent with reports in the published literature, which had variously reported that dangerous findings following imaging for head scans are low (1,7). It is not surprising that no other scanning techniques apart from CTs and MRIs were used for the imaging investigations, with the former being used in more over 80% of cases. This is similar to the report of the BEACH study, which found CT scans as the highest imaging technique adopted for headache investigations (8). This finding is also consistent with the recommendations of available guidelines in Australia, which advocates CTs primarily (and MRIs secondarily) as imaging tools for non-acute headaches warranting imaging investigations (22). Cost and availability may also be factors in the higher use of CTs, given that they are more widely available and cheaper compared with MRIs (23). We also found that the rate of MRIs increased if repeat scans for headaches were needed (50% each for CT and MRI, respectively). Again, this is in line with the available guidelines (22), which seem to be based on the claim that MRIs are more sensitive than CTs in investigating intracranial pathologies (23), meaning that clinicians are more likely to adopt it if the initial query was unsatisfactorily resolved with the original scans. However, it is interesting to note that we found no statistical differences between the use of MRIs and CTs with respect to finding significant intracranial pathologies that might explain headaches (P = 0.41). This finding appears to be supported by the conclusions of a paper that reviewed all the relevant studies from 1966 to 1998, which looked at the sensitivity, specificity and predictive values of different neuroimaging techniques in investigating non-acute headaches (24). The paper found that, although MRIs may be superior to CTs in identifying white matter lesions and developmental anomalies (which were largely clinically insignificant in diagnosing causes of non-acute headaches), they offered no advantages in identifying clinically significant pathologies relevant to these headaches. Another interesting finding was that repeat scans, be it with CT or MRI, revealed no new pathologies different from the first ones. The biggest time lapse between initial and the repeat scans was 5 years, leading us to infer that repeating head imaging for recurrent headaches within a 5-year period offer no additional benefits. This conclusion may be taken with caution though, given the small samples involved, but larger studies will be required to disprove it. Should this observation be replicated in larger studies, the implications for policy and practice are huge, given that it may then be more clinically beneficial to refer patients with recurrent headaches to third-party practitioners, rather than to go on to repeat scans if they re-present within a period of 5 years from an earlier scan. A key component of the discussion from the preceding paragraph is the fact that the use of MRIs on repeat scans after initial CT scans identified no new pathologies. This appears to re-confirm our conclusion in the earlier paragraph that MRIs have no advantage over CTs in investigating headaches in general practice, a finding backed by at least one previous publication (24) as already stated. Given the cost benefits of CTs over MRIs, these observations, if replicated in larger studies, will have significant cost-saving benefits (23). Yet another significant result from this work was that patients with identified psycho-morbidities were 78% less likely to have findings that will explain their headaches following head scans, relative to those without these co-morbidities (P = 0.03; Table 5). It is not clear if this is a result of the fact that practitioners tend to order more scans for patients with mental health issues, possibly out of higher level of ‘pressures’ known to be associated with these patients (4,11,25). On the other hand, it raises the question of whether patients with these psycho-morbidities are less likely to have structural brain pathologies than those without it. Whatever is the case, this finding is even made more significant given the observation that, among all the 517 patients seen for headaches in the 7-year period, those with psycho-morbidities were about 1.5 times more likely to get head imaging tests done when compared with those without such co-morbidities. Even though this latter observation narrowly missed statistical significance (P−0.08), combined to the earlier one, it appears to make a case that psycho-morbidities need to be considered in all cases of headaches before head imaging is ordered, given that, among these group of patients, scans are unlikely to reveal useful findings, yet doctors order them more. Whether these findings are peculiar to the population studied, or reflects a wider national, or global phenomenon is hard to ascertain within the limitations of this study. Larger studies to explore these is recommended, given the huge practice implications this portends. Limitations One limitation of our study is that our work was based on the documentations of doctors over a 7-year period. Omissions are possible, particularly with respect to the characteristics of the headaches (severity, recurrence and red flags) and the documentation of psychological problems. However, given that many years were involved, spanning across multiple visits from patients across more than one GP in most cases, we believe that most data required to make reliable observations were captured for most of the patients analysed. Our approach of relying on proven pharmacological and psychological treatments for patients with mental health issues also helped limit potential errors from poor documentation on mental health issues, as this ensured that only patients with real psychological problems were identified as such. Another potential limitation is on the possibility of human error in data collection, particularly with respect to the radiological reports and clinical notes entered for the patients. We tried to limit these by having all entries checked twice. The fact that the demographics of our patients, and the baseline proportions reported were not much different from the reports of the BEACH study also reassures that our data were reasonably accurate. Finally, as already acknowledged in the main paper, involving a larger number of surgeries, through a national and/or international study, may be necessary to help validate or disprove the generalizability of the findings of this work. Conclusions We conclude that just over one patient out of five (21.1%) presenting with headaches in general required neuroimaging as part of their investigations, with the majority being in the age range of 25–44 years (42.2%). Females were nearly three times more likely to report recurrent headaches (P = 0.03). Only CT and MRI imaging scans were used, with the former preferred at a ratio of about 4:1. However, there was no difference in using either of them with respect to significant intracranial findings (P = 0.41). Just about 15% of the scans revealed intracranial findings, with only 11% likely to explain the headaches. About one of every four patients who had scans were referred to other non-GP providers, with neurologists accounting for over 60% of these referrals. Also, 44% of the patients requiring head scans had associated psychological co-morbidities, and these subgroups were 78% less likely to have lesions that will explain their headaches (P = 0.03). Also, even though statistical significance was narrowly missed, headache presentations among patients with psycho-morbidities were more likely to be investigated with neuroimaging when compared with those without such co-morbidities (P = 0.08). Finally, nearly 13% of the patients required repeat scans, and these were all within a 5-year period. However, these additional scans provided no additional findings to the initial scans, be it CT or MRI. Recommendations Firstly, we recommend that patients presenting with recurrent headaches after initial head scans may not benefit from a repeat scan within a 5-year period and that clinicians should consider referring to third-party providers if there are concerns or uncertainty with the diagnosis. Secondly, given that there are no statistically significant differences in the findings whether CTs or MRIs were used, we recommend that the cheaper and more widely available technique, which is CT, be adopted where head imaging for headaches is needed in general practice. Thirdly, we advise that need GPs pay more attention to the psycho-morbidities of patients presenting with headaches. A full evaluation may be important before ordering imaging and involving psychiatrists and psychologists may be worthwhile options. Finally, we advocate that larger studies of national and international dimensions are worth undertaking so as to fully explore all the above-mentioned findings, given the huge implications for policy and practice, as well as the potential cost–saving benefits involved. Acknowledgements We wish to acknowledge the immense support from Mrs. Maureen Firehock, the Managing Director and CEO of all three practices used in this work (Kennedy Drive, PKG and Bilambil Heights Medical Centers, Tweed Heads, Australia). We are also appreciative of the efforts of all the staff in these surgeries who assisted with the data collection and other logistics needed to make this study a success. Declaration Funding: No external source of funding was required for this study. Ethical approval: Ethical approval was obtained from the Human Research Ethics Committee (HREC) of the Griffith University (GU Ref No: 2017/354) prior to the commencement of our study. Conflicts of interest: None. References 1. Frese T, Druckrey H, Sandholzer H. Headache in general practice: frequency, management, and results of encounter. Int Sch Res Notices  2014; 2014: 169428. Google Scholar PubMed  2. Rasmussen BK, Jensen R, Schroll Met al.   Epidemiology of headache in a general population—a prevalence study. J Clin Epidemiol  1991; 44: 1147– 57. Google Scholar CrossRef Search ADS PubMed  3. Latinovic R, Gulliford M, Ridsdale L. Headache and migraine in primary care: consultation, prescription, and referral rates in a large population. J Neurol Neurosurg Psychiatry  2006; 77: 385– 7. Google Scholar CrossRef Search ADS PubMed  4. O’Flynn N, Ridsdale L. Headache in primary care: how important is diagnosis to management? Br J Gen Pract  2002; 52: 569– 73. Google Scholar PubMed  5. Beran RG. Management of chronic headache. Aust Fam Physician  2014; 43: 106– 10. Google Scholar PubMed  6. Wonderling D, Vickers AJ, Grieve Ret al.   Cost effectiveness analysis of a randomised trial of acupuncture for chronic headache in primary care. BMJ  2004; 328: 747. Google Scholar CrossRef Search ADS PubMed  7. Kernick D, Stapley S, Hamilton W. GPs’ classification of headache: is primary headache underdiagnosed? Br J Gen Pract  2008; 58: 102– 4. Google Scholar CrossRef Search ADS PubMed  8. Britt H, Valenti L, Miller Get al.   Presentations of diarrhoea in Australian general practice. Aust Fam Physician  2005; 34: 218– 9. Google Scholar PubMed  9. Stanley IM, Peters S, Salmon P. A primary care perspective on prevailing assumptions about persistent medically unexplained physical symptoms. Int J Psychiatry Med  2002; 32: 125– 40. Google Scholar CrossRef Search ADS PubMed  10. Page LA, Howard LM, Husain Ket al.   Psychiatric morbidity and cognitive representations of illness in chronic daily headache. J Psychosom Res  2004; 57: 549– 55. Google Scholar CrossRef Search ADS PubMed  11. Couchman GR, Forjuoh SN, Rajab MHet al.   Nonclinical factors associated with primary care physicians’ ordering patterns of magnetic resonance imaging/computed tomography for headache. Acad Radiol  2004; 11: 735– 40. Google Scholar CrossRef Search ADS PubMed  12. Kernick DP, Ahmed F, Bahra Aet al.   Imaging patients with suspected brain tumour: guidance for primary care. Br J Gen Pract  2008; 58: 880– 5. Google Scholar CrossRef Search ADS PubMed  13. De Klippel N, Jansen JP, Carlos JS. Survey to evaluate diagnosis and management of headache in primary care: headache management pattern programme. Curr Med Res Opin  2008; 24: 3413– 22. Google Scholar CrossRef Search ADS PubMed  14. Kowacs PA, Twardowschy CA, Piovesan EJet al.   General practice physician knowledge about headache: evaluation of the municipal continual medical education program. Arq Neuropsiquiatr  2009; 67: 595– 9. Google Scholar CrossRef Search ADS PubMed  15. Stewart MA. Effective physician-patient communication and health outcomes: a review. CMAJ  1995; 152: 1423– 33. Google Scholar PubMed  16. Sandrini G, Friberg L, Jänig Wet al.   Neurophysiological tests and neuroimaging procedures in non-acute headache: guidelines and recommendations. Eur J Neurol  2004; 11: 217– 24. Google Scholar CrossRef Search ADS PubMed  17. American Academy of Neurology. Report of the quality standards sub-committee of the American Academy of Neurology. The utility of neuro imaging in the evaluation of headache in patients with normal neurological examinations . 2008. 18. Tweed Shire Council. The Tweed 2017. http://www.tweed.nsw.gov.au/TheTweed (accessed on 17 July 2017). 19. Australian Bureau of Statistics. 2011 Census QuickStats: Tweed Heads 2017. http://www.censusdata.abs.gov.au/census_services/getproduct/census/2011/quickstat/SSC12332 (accessed on 26 September 2017). 20. Citipedia. Gold Coast, Australia—list of cities, towns, villages 2017. http://www.citipedia.info/province/general/Australia_Gold+Coast (accessed on 17 July 2017). 21. City of Gold Coast. Gold Coast population fact sheets 2017. http://www.goldcoast.qld.gov.au/social-planning-and-development-factsheets-2006-2010-5152.html (accessed on 17 July 2017). 22. Government of Western Australia. Diagnostic imaging pathways—headache (adult) 2017. http://www.imagingpathways.health.wa.gov.au/index.php/imaging-pathways/neurological/headache#pathway (accessed on 20 July 2017). 23. De Luca GC, Bartleson JD. When and how to investigate the patient with headache. Semin Neurol  2010; 30: 131– 44. Google Scholar CrossRef Search ADS PubMed  24. Frishberg BM, Rosenberg JH, Matchar DBet al.   Evidence-Based Guidelines in the Primary Care Setting: Neuroimaging in Patients with Nonacute Headache . St Paul, MN: US Headache Consortium, 2000. 25. Morgan M, Jenkins L, Ridsdale L. Patient pressure for referral for headache: a qualitative study of GPs’ referral behaviour. Br J Gen Pract  2007; 57: 29– 35. Google Scholar PubMed  © The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.

Journal

Family PracticeOxford University Press

Published: Jan 24, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off