Key Points
Question
Do teleneurology services in rural primary care, compared with standard care, increase the proportion of patients with neurological symptoms or signs that are managed solely in general practice?
Findings
In a stepped-wedge cluster randomized clinical trial, teleneurology services did not lead to an increased proportion of patients solely managed by general practice. They led to fewer referrals to ambulatory neurologists but to an increase in referrals to other specialists and hospitalization.
Meaning
Referral rates to ambulatory neurologists may be reduced by teleneurology, but no reduction of overall referrals, including to other specialists and hospitals, was demonstrated in this study; long-term effects remain to be evaluated.
Abstract
Importance
Incorporating teleneurology into primary care aimed to improve access to neurological care in a rural region in Germany. A telemedicine network connecting neurologists, general practitioners (GPs), and patients was established.
Objective
To determine whether teleneurology services in rural primary care increased the proportion of patients with neurological symptoms managed solely in general practice.
Design, Setting, and Participants
The NeTKoH study (Neurological Teleconsultation With General Practitioners to Strengthen Specialist Care in Western Pomerania, Germany) was a stepped-wedge cluster randomized clinical trial conducted from January 1, 2021, through July 31, 2025, to compare teleconsultations vs standard care. Patients were continuously approached and eligible if they were aged 18 years or older and presented at one of 41 participating GP practices in northeast Germany with symptoms for which a neurological consultation was deemed necessary. The duration of patient follow-up was 3 months. Data analysis was conducted from January to July 2025.
Intervention
Practices were equipped with a telemedicine system to enable a video conference between the GP and patient on 1 side and a neurologist on the other side to obtain immediate neurological assessment and guide further care decisions.
Main Outcomes and Measures
The primary outcome was the proportion of patients who continued to be managed solely within general practice.
Results
Between October 15, 2021, and October 25, 2024, 986 patients were enrolled (intervention: 517 patients; control: 469 patients); 3 of these patients were excluded, leaving 983 patients for analysis (605 [61.5%] female; median [range] age, 55 [18-90] years). Teleconsultations, compared with standard care, resulted in fewer patients managed solely within general practice (38.3% vs 50.7%; adjusted odds ratio, 0.58; 95% CI, 0.38-0.88). Referrals to neurologists decreased with teleconsultations compared with standard care (36.4% vs 41.4%), while referrals to other specialists (11.9% vs 4.7%) and hospitals (12.8% vs 2.6%) increased. No significant differences were observed in quality of life or health status. Adherence to recommendations was lower in the intervention phase. At follow-up, more than half of the patients in either group were still managed solely in general practice.
Conclusions and Relevance
This stepped-wedge cluster randomized clinical trial suggests that neurological teleconsultations in general practice may have helped triage patients (eg, to other specialists or hospitals) and therefore contributed to improved medical care, particularly in structurally underserved areas; however, an immediate reduction in the use of secondary care was not found. The inability to blind GPs to the randomization may have introduced selection effects, which should be considered in future studies.
Trial Registration
German Clinical Trials Register Identifier: DRKS00024492
This stepped-wedge cluster randomized clinical trial evaluates whether teleneurology services in rural primary care in northeast Germany increased the proportion of patients with neurological symptoms or signs managed solely in general practice.
Introduction
It is estimated that 42% to 50% of the global population has a neurological disorder.1 Due to an aging population, a further increase is expected.2 Access to specialized neurological care is limited particularly in rural areas, mainly due to a workforce shortage.1 The growing demand for neurological care poses significant challenges for general practitioners (GPs) and neurologists working in outpatient settings. Incorporating teleneurology into primary care has the potential to improve access to ambulatory care, reduce waiting times, avoid unnecessary referrals, and thus improve quality of care. Previous studies have shown that teleneurology is feasible3,4,5 and that its use is expanding.6,7 Most teleneurology services focus on specific health problems8 and are not integrated in primary care.9 The majority of studies were retrospective, observational, or surveys.8 Comprehensive teleneurology services in general practice in Germany have not been established and evaluated with a randomized clinical trial (RCT). The NeTKoH project (Neurological Teleconsultation With General Practitioners to Strengthen Specialist Care in Western Pomerania, Germany) established such a neurological telemedicine service.
The aim was to evaluate the effect of neurological teleconsultations in general practice on the management of patients presenting with neurological symptoms or signs that might justify referral to a neurologist. The main research question was whether neurological teleconsultations, compared with standard care, increase the proportion of patients with neurological symptoms or signs that are managed solely in general practice? Additional questions were as follows: What were the dominant symptoms or signs among patients enrolled? To what extent did patients adhere to care management recommendations, and were they seen by a neurologist or another specialist or hospitalized? Were there group differences in health status or quality of life between baseline and follow-up?
Methods
Study Design and Participants
The study was a stepped-wedge cluster RCT conducted from January 1, 2021, through July 31, 2025 (trial protocol appears in Supplement 1). Each GP was considered a cluster, and GPs in joint practice were randomized together. GPs in northeast Germany were eligible if they provided care to patients insured by a statutory health insurance. The trial period was divided into 11 steps, each lasting 12 weeks. Steps including public holidays or vacation periods were extended by up to 2 weeks. In the first step, all GPs were in the control phase, providing standard care. Prior to each step, 3 to 4 randomly selected GPs transitioned to the intervention, with neurological teleconsultations. By the final step, all GPs had transitioned to the intervention (Figure 1). Due to the nature of the intervention, blinding was impossible. Patients were continuously identified by GPs and were eligible if they (1) were aged 18 years or older, (2) presented with symptoms or signs for which a neurology consultation was deemed appropriate, and (3) were able to give informed consent. Exclusion criteria were symptoms indicative of an acute stroke or other conditions requiring emergency care. Patients could only participate once.
Figure 1. Study Flowchart.
Three patients were excluded from the neurological teleconsultation group (2 due to technical issues with the teleconsultation and 1 due to data protection reasons).
Initially, participation was limited to patients insured with 1 particular statutory health insurance. When the targeted sample size for the first steps was not reached, eligibility was extended for the following steps to allow all patients to enroll in the study, provided they were insured by any statutory health insurance.
The study was approved by the Ethics Committee of the University Medicine Greifswald, Germany. The first patient was enrolled on October 15, 2021, and the last patient on October 25, 2024. Written informed consent was obtained from all participants at the GP practice. Three months after enrollment, a trained study nurse conducted a follow-up telephone interview. This article follows the Consolidated Standards of Reporting Trials (CONSORT) reporting guidelines and the CONSORT Extension reporting guidelines for stepped wedge cluster RCTs.10 A more detailed description of the study methods has been published previously.11 Data analysis was conducted from January to July 2025.
Intervention
GPs transitioning into the intervention phase were equipped with a certified telemedicine system provided by MEYTEC. The device was a face-to-face video conferencing tool that linked the GP and patient to one of 7 participating teleneurologists from University Medicine Greifswald, some of whom had subspecialty training in migraine and headache, peripheral neurology, epilepsy, and stroke. The primary purpose was to obtain an immediate neurological assessment and guidance for further management. Neurologists recommended whether the patient could be managed in general practice, required referral to an ambulatory specialist, or required an immediate referral to the nearest appropriate hospital. The teleneurologist documented the recommendation and transmitted it electronically to the GP. Teleconsultation was generally available Mondays to Fridays from 8 am to 5 pm.
Sample Size and Power Calculation
The sample size calculation was based on the primary outcome, considering the stepped-wedge cluster design with fixed time and random cluster effects.12 The proportion of patients with neurological symptoms or signs who could be cared for in general practice was estimated to be 20% in standard care and projected to increase to 30% with teleconsultations. With a type I error of .05, a coefficient of variation of 0.25, and a power of 80%, the required sample size was 1089 patients (intervention: 534 patients; control: 555 patients). The calculation was performed using the R package swCRTdesign (R Foundation).13
Randomization
Randomization was performed at the cluster level after recruitment of the participating GPs, using R (version ≥4.1.0). Following design adjustments in 2022 that reduced the number of clusters to 33 and increased the number of steps to 11, a rerandomization was performed in which clusters that had previously been assigned to the first sequence group remained in this group.
Outcomes and Data Collection
The primary outcome was the proportion of patients with neurological symptoms or signs who received continued care at their GP’s practice, ie, the proportion of patients who did not require a referral to a neurologist, other specialist, or hospital. Data were collected using paper questionnaires completed by the GP in the control phase and were entered online into the database by the teleneurologist in the intervention phase.
Secondary outcomes included care management recommendation to ambulatory specialists or hospitalization and the health status and quality of life of the patients at baseline and after 3 months. Quality of life and self-rated health status were assessed using the EQ-5D-5L and the visual analog scale (EQ-VAS) with a self-completed paper questionnaire in general practice at baseline and a telephone interview at follow-up.
Additional analyses explored the recorded patient diagnoses (International Statistical Classification of Diseases and Related Health Problems, Tenth Revision codes) as the medical reasons necessitating a neurological consultation (eTable 1 in Supplement 2), and adherence with care management recommendations as captured in the follow-up interviews. The duration of patient follow-up was 3 months.
Statistical Analysis
Descriptive statistics were computed for GP and patient characteristics. Categorical variables were presented as proportions; continuous variables were summarized using means with SDs or medians with ranges or IQRs. Under an assumption of missing completely at random, observations with missing data were excluded from individual analyses.
For differences in care coordination at baseline, logistic mixed models with random intercepts adjusting for age and sex of the patient as well as time were computed.
Odds ratios and their 95% CIs were estimated. Differences in the distribution of categories for further care management between control and intervention were assessed for statistical significance using the Cramér–von Mises test for 2 independent samples.
Patients lost to follow-up were excluded for further analysis. Multiple visits to different health care professionals at follow-up were considered.
Group differences in health status and quality of life between baseline and follow-up were calculated using generalized linear mixed regression models, adjusting for patients’ age and sex as well as time and reported as the risk difference. The statistical analysis was performed using R version 4.5.1 and RStudio version 2024.12.1.
Results
Of 1056 screened patients, 986 (93.4%) were enrolled by a total of 41 GPs. Eight GPs dropped out, mainly due to time and resource constraints (because GPs were administering COVID-19 vaccines) or an insufficient number of eligible patients early in the study (eTable 2 in Supplement 2). They were replaced by newly recruited GPs (Figure 1). The 33 GPs who participated until the end had a median (range) age of 47 (34-65) years and a median (IQR) work experience of 17 (13-25) years, and 21 (63.6%) were male. The majority (23 GPs [69.7%]) had no previous experience with telemedicine (eTable 3 in Supplement 2).
Of 986 patients (intervention: 517 patients; control: 469 patients), 3 in the intervention had to be excluded from further analysis: 2 due to technical issues with the telemedicine system and 1 due to data protection reasons.
Among the 983 remaining patients, the median (range) age was 55 (18-90) years, most patients were female (605 [61.5%]), and the most common neurological symptoms or signs were neuropathies, headaches, migraine, and dizziness (Table). The mean (SD) EQ-VAS scores at baseline in the intervention and control phases were 65 (20.0) and 64 (18.5) of 100, respectively. Overall, baseline characteristics were balanced between the 2 groups.
Table. Patient Characteristics and Categorized Diagnoses.
| Patient characteristic | Total (n = 983) | Phase | |
|---|---|---|---|
| Intervention (n = 514) | Control (n = 469) | ||
| Age, median (range), y | 55 (18-90) | 56 (18-89) | 55 (18-90) |
| Sex, No. (%) | |||
| Female | 605 (61.5) | 309 (60.1) | 296 (63.1) |
| Male | 378 (38.5) | 205 (39.9) | 173 (36.9) |
| EQ-5D-5L VAS score at baseline, mean (SD) | 64.9 (19.2) | 65.5 (20.0) | 64.6 (18.5) |
| Disease categories diagnosed by GPs, No. (%)a | (n = 980) | (n = 512) | (n = 468) |
| Neuropathy | 188 (19.2) | 104 (20.3) | 84 (17.9) |
| Headache | 175 (17.9) | 80 (15.6) | 95 (20.3) |
| Migraine | 168 (17.1) | 78 (15.2) | 90 (19.2) |
| Dizziness | 162 (16.5) | 102 (19.9) | 60 (12.8) |
| Movement disorder | 85 (8.7) | 44 (8.6) | 41 (8.8) |
| Other | 45 (4.6) | 20 (3.9) | 25 (5.3) |
| Cerebrovascular disorder | 38 (3.9) | 24 (4.7) | 14 (3.0) |
| Mental disorder | 28 (2.9) | 14 (2.7) | 14 (3.0) |
| Musculoskeletal disorder | 25 (2.6) | 12 (2.3) | 13 (2.8) |
| Impaired consciousness | 19 (1.9) | 11 (2.1) | 8 (1.7) |
| Epilepsy | 18 (1.8) | 9 (1.8) | 9 (1.9) |
| Demyelinating diseases of the central nervous system | 10 (1.0) | 3 (0.6) | 7 (1.5) |
| Vision disorder | 10 (1.0) | 6 (1.2) | 4 (0.9) |
| Dementia | 5 (0.5) | 2 (0.4) | 3 (0.6) |
| Insomnia | 4 (0.4) | 3 (0.6) | 1 (0.2) |
Abbreviations: GPs, general practitioners; VAS, visual analog scale.
Corresponding International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10) codes in eTable 1 in Supplement 2.
In total, 135 of 983 patients (13.7%) were lost to follow-up after 3 months (intervention: 66 patients [12.8%]; control: 69 patients [14.7%]). There was no sex difference in loss to follow-up, but the proportion of patients with headache and movement disorders at baseline was higher in participants lost to follow-up.
Patients with clear symptoms indicative of an acute stroke or with other conditions requiring emergency care were to be excluded. Nevertheless, 6 patients in the intervention phase and 2 patients in the control phase were referred to hospital care after symptoms of acute stroke were observed during the teleconsultation or the regular consultation, respectively.
Care Coordination at Baseline
Initially, 197 patients (38.3%) in the intervention phase and 238 patients (50.7%) in the control phase were triaged to further management solely in general practice (eFigure 1 in Supplement 2). Here, triage refers to the care management recommendation documented at baseline by the GP in the control phase and by the teleneurologist in the intervention phase, where GPs stated that they would implement the recommendation completely in 466 cases (90.7%) and partially in another 46 (8.9%).
The regression analysis showed an odds ratio of 0.58 (95% CI, 0.38-0.88) for patients being triaged to further management in general practice in the intervention phase compared with the control phase. The distribution of triage recommendation to an ambulatory specialist or to a hospital differed between intervention and control (eFigure 1 in Supplement 2): a slightly lower proportion of patients in the intervention phase were triaged for neurology referral (36.4%) compared with the control phase (41.4%). With the intervention, as compared with the control, significantly more patients were triaged to other ambulatory specialists (61 [11.9%] vs 22 [4.7%]) or to a hospital (66 [12.8%] vs 12 [2.6%]). Adjusted multivariate logistic regressions showed the differences in the latter 2 referral categories to be statistically significant when compared with remaining in general practice for management (eTable 4 in Supplement 2).
Headaches and migraine were the most common symptoms in both phases and the most common symptoms among patients who continued management with general practice. Neuropathies, migraine, and movement disorders were the most common symptoms of patients referred to an ambulatory neurologist by the teleneurologist (eTable 5 in Supplement 2). The most common symptom for which the teleneurologist recommended referral to other specialists was dizziness. Furthermore, more patients in the intervention group with cerebrovascular disorder, dizziness, neuropathy, epilepsy or impaired consciousness, migraine, and headache were recommended for hospitalization (eTable 5 in Supplement 2).
All patients in the intervention had an initial neurologist contact in the teleconsultation, which was conducted on the same day for 507 of 514 patients (98.6%).
Follow-Up
Patients lost to follow-up were excluded from the following analysis. More than 70% of patients triaged to further care in general practice (intervention: 125 of 174 patients [71.8%]; control: 145 of 197 patients [73.6%]) reported no neurologist, specialist, or hospital visits (Figure 2). Among all patients, 57.8% (259 of 448) in the intervention phase and 55.8% (223 of 400) in the control phase received care exclusively in general practice.
Figure 2. Flowchart Showing Care Management Recommendations.
Care management recommendation (triage) to general practice, neurology referral, referral to other specialists, or hospital by the general practitioner (GP) in the control phase and by the teleneurologist in intervention phase for participants not lost to follow-up, and their actual management as self-reported in the 3-month follow-up. The numbers listed in the grid for 3-month follow-up indicate the recommendation (shown by the line’s color under the value), and the column indicates what the patients actually did. For example, 6 patients in the control group who were referred to the hospital had an appointment with a neurologist instead. The numbers above the thin black line indicate those who were initially recommended multidisciplinary care. Some patients had consulted several health care professionals within 3 months.
Of the patients triaged to ambulatory neurologists, 33.3% (55 of 165) in the intervention phase and 47.1% (80 of 170) in the control phase had been seen by a neurologist. Among all patients with follow-up, the proportion of patients who were seen by an ambulatory neurologist was lower in the intervention phase than in the control phase (23.0% [103 of 448] vs 28.0% [112 of 400], respectively).
Of the patients triaged to other ambulatory specialists, 35.3% (18 of 51) in the intervention phase and 36.8% (7 of 19) in the control phase self-reported a consultation with another specialist by the time of their follow-up interview. Overall, 18.3% (82 of 448) in the intervention phase and 21.8% (87 of 400) in the control phase consulted another specialist.
Of the 56 patients advised to visit a hospital after the teleconsultation, only 32 reported a hospitalization, 20 an appointment with an ambulatory neurologist, and 6 with another specialist, while 17 reported none of these (multiple visits to different health care professionals were possible) (Figure 2). The proportion of self-reported hospitalizations nearly doubled in the intervention phase compared with the control phase (55 of 448 [12.3%] vs 26 of 400 [6.5%]).
Quality of Life and Health Status
The mean across the 5 dimensions of the EQ-5D-5L did not reveal any relevant differences in the quality of life between the intervention and control phases (eFigure 2 in Supplement 2). The mean (SD) EQ-5D-5L index for the intervention was 0.8 (0.2) at both baseline and follow-up, which was not significantly higher than in the control phase (mean [SD], 0.7 [0.2]).
The health status (EQ-VAS) showed no significant difference between the phases. The mean (SD) increase in the EQ-VAS score in the intervention phase from 65.5 (20.0) to 69.6 (20.1) was similar to that in the control phase, which increased from 64.6 (18.5) to 68.8 (19.8) (eFigure 2 in Supplement 2).
Regressions showed no statistically significant difference in change in quality of life (risk difference, −0.01; 95% CI, −0.04 to 0.03) or health status (risk difference, 0.14; 95% CI, −2.81 to 3.09).
Discussion
The expected effect, that the provision of neurological teleconsultations in general practice would increase the proportion of patients who do not require a referral, was not confirmed. While referrals to ambulatory neurologists decreased, referrals to other ambulatory specialists or the hospital increased, which may be indicative of more targeted patient referrals. Thus, teleneurological expertise may speed up access to diagnostic workup by improving the accuracy and timeliness of referrals, thereby improving management and treatment for patients with neurological symptoms and signs.
By the time of the 3-month follow-up, more than half of the patients in either phase continued to be managed solely in general practice, and a larger proportion of patients in the intervention phase reported a hospitalization during the follow-up period. Between baseline and follow-up, quality of life and health status showed similar developments across both phases.
The high enrollment rate of eligible patients suggests that GPs and patients welcomed the opportunity to get access to a neurological teleconsultation. With few exceptions, a teleconsultation was arranged for all patients on the same day, providing timely care to patients at their local general practice.
A larger proportion of patients in the intervention phase presented to a hospital during the follow-up period compared with the control phase. However, 43% of patients in the intervention phase and 36% in the control phase did not follow the medical recommendation of presentation to a hospital. In sum, a large proportion of patients did not receive the subsequent care that had been recommended to them. Possible explanations are that (1) patients did not follow the recommendation, (2) GPs did not refer the patients, or (3) patients did not receive an appointment with an ambulatory specialist. Inherent limitations of teleconsultations, safety concerns, less knowledge about the patients, and limited experience in ambulatory care might have led teleneurologists to more frequently refer to the emergency department or other ambulatory specialists. The low rate of ambulatory specialty consultations by the follow-up potentially highlights the long waiting times for specialist appointments.
The acuity and severity of neurological symptoms and signs were not recorded across phases, but differences seem likely. Given that blinding was impossible, GPs knew whether a teleconsultation would or would not take place.14 Furthermore, an identification and recruitment bias cannot be ruled out, where patients with unclear or more severe symptoms may initially have been presented via teleconsultation in the intervention phase. This may have been the case for the 6 patients who were not directly referred to a hospital by their GP with suspected stroke, which was deemed to be nonacute and therefore suitable for a teleconsultation.
In our study, the most common symptoms and signs were headache, migraine, and dizziness. Similar findings have been reported in other studies,5,15 most likely due to high diagnostic uncertainty associated with treatment of headaches in general practice.16 The safety and efficacy of telemedicine for nonacute headaches were previously demonstrated in a Norwegian study.9 However, that intervention was not conducted in general practice; therefore, a direct comparison with our study is impossible.
Headaches are very common, and many individuals choose over-the-counter medications as their first management option rather than seeking medical care.17,18 It is plausible that GPs in this study used the opportunity to obtain expert advice regarding their patients’ needs that would likely not have been sought under routine care conditions. This also applies to dizziness, a frequent reason for consultation, although diagnostic uncertainty appears to be even greater here due to the diverse and interdisciplinary causes and the difficulty of differential diagnosis.19 GPs may also have used teleneurology to reduce diagnostic uncertainty for the symptom of dizziness. Patient and clinician satisfaction with telemedicine services for dizziness was previously demonstrated.20
The population studied here is comparable to those of other rural and small-town general practices in Germany. However, our results may not apply to urban areas with better access to specialist care. As GPs who were interested in telemedicine and research agreed to participate in the study, this group may have been overrepresented. The majority had not used telemedicine before.
Strengths and Limitations
To our knowledge, this is the first study evaluating neurological teleconsultation in general practice in Germany, providing a robust framework for evaluation of teleneurology in routine care, allowing transition to the intervention phase for all practices, and addressing logistic constraints.
The study has several limitations, including underrecruitment (−10%), potential but nondifferential recall bias, a short follow-up period limiting pathway assessment, and forms of documenting primary outcomes that differed between phases. Lack of blinding may have favored recruitment of more severe cases in the intervention phase, where a teleconsultation may have provided the fastest access to a neurologist for patients in a rural area with extended waiting times for specialist appointments. Other approaches less prone to selection bias might be more suited to evaluate telemedicine services.21
Conclusions
This stepped-wedge cluster RCT demonstrates that teleneurology in rural primary care settings can alter referral patterns, potentially improving health care efficiency and effectiveness. Teleneurology represents a promising approach to enhancing neurological care in underserved areas, but further research is needed to assess its long-term impact and to develop strategies for improving patient compliance with care recommendations.
Trial Protocol
eTable 1. Disease Categories and Corresponding ICD Codes
eTable 2. Reasons for Attrition (8 GP Practices)
eTable 3. Characteristics and Professional Experience of General Practitioners
eTable 4. Odds Ratios for Referral Recommendations Compared to GP Follow-Up
eTable 5. Complaint Prevalence Among Patients by Group According to the Disease Categories of eTable 1, Stratified by Referral
eFigure 1. Care Management Recommendation (Triage) (Number of Participants Managed in GP Practice, Referrals to a Neurologist, Other Specialists, or Hospital) According to Suggested Management of the GPs (Control Phase) and Teleneurologists (Intervention Phase), at Baseline
eFigure 2. Change in Quality of Life Measured With the EQ-5D-5L Questionnaire and Change in Health Status Measured With the EQ VAS, for Patients Without Loss to Follow-Up
Data Sharing Statement
References
- 1.World Health Organization . Global Status Report on Neurology. World Health Organization; 2025. [Google Scholar]
- 2.Feigin VL, Vos T, Nichols E, et al. The global burden of neurological disorders: translating evidence into policy. Lancet Neurol. 2020;19(3):255-265. doi: 10.1016/S1474-4422(19)30411-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Chua R, Craig J, Wootton R, Patterson V. Randomised controlled trial of telemedicine for new neurological outpatient referrals. J Neurol Neurosurg Psychiatry. 2001;71(1):63-66. doi: 10.1136/jnnp.71.1.63 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Janssen F, Awadallah M, Alhalabi A, et al. Telemedicine in general neurology: use of audiovisual consultation for on call back-up service in an acute care hospital. J Neurol. 2018;265(4):880-884. doi: 10.1007/s00415-018-8756-4 [DOI] [PubMed] [Google Scholar]
- 5.Mantese CE, Aquino ERDS, Figueira MD, Rodrigues L, Basso J, Raupp Da Rosa P. Telemedicine as support for primary care referrals to neurologists: decision-making between different specialists when guiding the case over the phone. Arq Neuropsiquiatr. 2021;79(4):299-304. doi: 10.1590/0004-282x-anp-2020-0137 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Hatcher-Martin JM, Adams JL, Anderson ER, et al. Telemedicine in neurology: Telemedicine Work Group of the American Academy of Neurology update. Neurology. 2020;94(1):30-38. doi: 10.1212/WNL.0000000000008708 [DOI] [PubMed] [Google Scholar]
- 7.Larner AJ. Teleneurology: an overview of current status. Pract Neurol. 2011;11(5):283-288. doi: 10.1136/practneurol-2011-000090 [DOI] [PubMed] [Google Scholar]
- 8.Sahu JK, Coan AC, Chan J, et al. Applications and impact of telemedicine for persons with epilepsy: a scoping review. Seizure. 2026;136:107-116. doi: 10.1016/j.seizure.2026.01.016 [DOI] [PubMed] [Google Scholar]
- 9.Müller KI, Alstadhaug KB, Bekkelund SI. A randomized trial of telemedicine efficacy and safety for nonacute headaches. Neurology. 2017;89(2):153-162. doi: 10.1212/WNL.0000000000004085 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Hemming K, Taljaard M, McKenzie JE, et al. Reporting of stepped wedge cluster randomised trials: extension of the CONSORT 2010 statement with explanation and elaboration. BMJ. 2018;363:k1614. doi: 10.1136/bmj.k1614 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Wainwright K, Mayer I, Oliveira Gonçalves AS, et al. ; NeTKoH Consortium . Effect evaluation of a tele-neurologic intervention in primary care in a rural area in Germany—the NeTKoH study protocol of a stepped-wedge cluster randomized trial. BMC Health Serv Res. 2023;23(1):756. doi: 10.1186/s12913-023-09724-w [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Hussey MA, Hughes JP. Design and analysis of stepped wedge cluster randomized trials. Contemp Clin Trials. 2007;28(2):182-191. doi: 10.1016/j.cct.2006.05.007 [DOI] [PubMed] [Google Scholar]
- 13.Voldal EC, Hakhu NR, Xia F, Heagerty PJ, Hughes JP. SwCRTdesign: an Rpackage for stepped wedge trial design and analysis. Comput Methods Programs Biomed. 2020;196(105514):105514. doi: 10.1016/j.cmpb.2020.105514 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Lehmann BA, Lindert L, Ohlmeier S, Schlomann L, Pfaff H, Choi KE. “And then he got into the wrong group”: a qualitative study exploring the effects of randomization in recruitment to a randomized controlled trial. Int J Environ Res Public Health. 2020;17(6):1886. doi: 10.3390/ijerph17061886 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Bekkelund SI, Albretsen C. Evaluation of referrals from general practice to a neurological department. Fam Pract. 2002;19(3):297-299. doi: 10.1093/fampra/19.3.297 [DOI] [PubMed] [Google Scholar]
- 16.Bösner S, Hartel S, Diederich J, Baum E. Diagnosing headache in primary care: a qualitative study of GPs’ approaches. Br J Gen Pract. 2014;64(626):e532-e537. doi: 10.3399/bjgp14X681325 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Mehuys E, Paemeleire K, Van Hees T, et al. Self-medication of regular headache: a community pharmacy-based survey. Eur J Neurol. 2012;19(8):1093-1099. doi: 10.1111/j.1468-1331.2012.03681.x [DOI] [PubMed] [Google Scholar]
- 18.Porst M, Wengler A, Leddin J, et al. Migräne und Spannungskopfschmerz in Deutschland: Prävalenz und Erkrankungsschwere im Rahmen der Krankheitslast–Studie BURDEN 2020. September 9, 2020. Updated November 30, 2020. Accessed August 20, 2025. https://edoc.rki.de/handle/176904/6962.2
- 19.Stephan AJ, Kovacs E, Phillips A, Schelling J, Ulrich SM, Grill E. Barriers and facilitators for the management of vertigo: a qualitative study with primary care providers. Implement Sci. 2018;13(1):25. doi: 10.1186/s13012-018-0716-y [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Meinhardt G, Perez N, Sharrer C, et al. The role of telemedicine for evaluation and management of dizzy patients: a systematic review. Otol Neurotol. 2023;44(5):411-417. doi: 10.1097/MAO.0000000000003876 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Li MM, Rising KL, Goldberg EM. Transitioning to telehealth? a guide to evaluating outcomes. Health Policy Technol. 2022;11(3):100623. doi: 10.1016/j.hlpt.2022.100623 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Trial Protocol
eTable 1. Disease Categories and Corresponding ICD Codes
eTable 2. Reasons for Attrition (8 GP Practices)
eTable 3. Characteristics and Professional Experience of General Practitioners
eTable 4. Odds Ratios for Referral Recommendations Compared to GP Follow-Up
eTable 5. Complaint Prevalence Among Patients by Group According to the Disease Categories of eTable 1, Stratified by Referral
eFigure 1. Care Management Recommendation (Triage) (Number of Participants Managed in GP Practice, Referrals to a Neurologist, Other Specialists, or Hospital) According to Suggested Management of the GPs (Control Phase) and Teleneurologists (Intervention Phase), at Baseline
eFigure 2. Change in Quality of Life Measured With the EQ-5D-5L Questionnaire and Change in Health Status Measured With the EQ VAS, for Patients Without Loss to Follow-Up
Data Sharing Statement
