This was a retrospective cohort study conducted at a single secondary referral center in Mongolia. This study was approved by the Research Ethical Committee of the MNUMS on 21 December 2018 (No. 2018/3-18). We retrospectively analyzed prospectively collected data from patients diagnosed with BPPV. Informed consent was obtained from all participants.

Patients were recruited between January 2019 and January 2021 at ENT Hospital. The diagnosis of BPPV was made in accordance with the 2017 American Academy of Otolaryngology–Head and Neck Surgery Foundation (AAO–HNSF) BPPV guidelines. The inclusion criteria were age ≥18 years, agreement to participate in the study, and completion of follow-up evaluations.

BPPV was diagnosed in 421 patients. Among these, 144 patients who did not exhibit positional vertigo or nystagmus at the time of evaluation (resolved BPPV) and 115 patients who did not complete follow-up examinations were excluded. Consequently, 162 patients were included in this study (Figure 1). Clinical questionnaires and the Dizziness Handicap Inventory (DHI) were completed, and videonystagmography was recorded and analyzed for all patients.

No patients or members of the public were involved in the design, conduct, reporting, or dissemination plans of this research.

The questionnaire consisted of three parts. The first part collected general patient information, including name, age, sex, and address. The second part gathered details about the history of the current illness, such as the onset of symptoms, most recent episode, frequency and duration of episodes, and potential risk factors. The third part included four questions addressing the major signs and symptoms of BPPV, as outlined in the 2017 AAO-HNSF BPPV guidelines (10). These questions assessed whether vertigo (dizziness) was triggered by changes in head position (e.g., lying down, turning over in a supine position, or bending down), whether the duration of dizziness was 5–60 s, whether nausea was experienced, and whether blurry vision occurred.

The severity and handicap associated with dizziness were evaluated before and after treatment using the DHI, developed by Jacobson and Newman in 1990 (11).

Spontaneous nystagmus and positional nystagmus were recorded and analyzed before and after treatment using a videonystagmography system (Easy-eyes®; SLMED, Seoul, Republic of Korea). The video recording sequence included the following positions: sitting, lying down with the head centered, head rolled to the right side, head rolled to the left side, sitting, Dix–Hallpike maneuver on the right side, Dix–Hallpike maneuver on the left side, sitting, head hanging, sitting, and head bent (12, 13). Detailed results of the nystagmus analysis are presented in Supplementary Table S1.

Patients with posterior canal (PC) BPPV were treated using the Epley maneuver followed by three consecutive Semont maneuvers (14, 15). Patients with horizontal canal (HC) BPPV underwent the barbecue roll maneuver followed by three consecutive Gufoni maneuvers (16, 17). Patients displaying anterior canal (AC) BPPV were treated with three consecutive Yacovino maneuvers (18). Patients were instructed to repeat the maneuvers at home three times daily for 7 days. Treatment outcomes were evaluated at 7, 14, and 28 days. If nystagmus persisted during a follow-up visit, treatment was continued, and the patient was re-evaluated at the subsequent follow-up.

To minimize potential measurement bias, all patients were evaluated using standardized diagnostic criteria, evaluation tools (VNG, DHI), and treatment protocols. However, as this study was conducted at a single secondary referral center, selection bias may exist, and the findings may not be fully generalizable to primary care settings in Mongolia.

The study size was determined by the number of eligible patients who presented to the clinic during the two-year study period and consented to participate. No a priori sample size calculation was performed as the primary objective was descriptive, aiming to assess the current status of medical services for BPPV.

Costs were estimated based on 2024 pricing to reflect current economic burdens, although clinical data were collected from 2019 to 2021.

Patients with incomplete data were handled on an analysis-by-analysis basis; cases with missing values were excluded only from the specific analysis requiring that variable. Continuous variables are presented as mean ± SD and categorical variables as n (%). Between-group comparisons were performed using the chi-square test (or Fisher’s exact test when appropriate) for categorical variables and Student’s t-test for continuous variables; Welch’s t-test was used when variance assumptions were not met. Paired t-tests were used to compare pre- and post-treatment DHI scores.

For diagnostic delay, the interval from symptom onset to the BPPV diagnosis at the first specialist dizziness-clinic visit was calculated. For patients with pre-referral MRI, additional timing intervals were calculated using recorded dates: symptom onset to MRI acquisition, and MRI acquisition to the first specialist visit.

To identify factors associated with MRI usage, multivariate logistic regression analysis was performed. Clinical variables were entered into the model, and a stepwise approach was used to determine significant predictors. Results are presented as odds ratios (ORs) with 95% confidence intervals (CIs). A two-sided p < 0.05 was considered statistically significant. Statistical analyses were conducted using SPSS Statistics for Windows version 23.0 (IBM Corp., Armonk, NY, USA) and R version 4.5.2 (R Foundation for Statistical Computing, Vienna, Austria).

We prepared this manuscript following the STROBE (Strengthening the Reporting of Observational studies in Epidemiology) statement. The completed STROBE checklist is available as a Supplementary material.

The mean age of the patients was 50.0 ± 11.7 years (range, 22–82 years). Thirty-two patients (19.8%) were men, and 130 patients (80.2%) were women. The cause of BPPV was idiopathic in 115 patients (70.9%); the remaining cases were attributed to trauma (40 patients, 24.7%), prolonged bed rest (four patients, 2.5%), and otologic disease (three patients, 1.9%).

Among the 162 patients, 96 (59.3%) had PC BPPV, 44 (27.1%) had HC BPPV, 17 (10.5%) had AC BPPV, and five (3.1%) had multiple canal BPPV. HC BPPV was classified as geotropic in 21 (12.9%) patients and apogeotropic in 23 (14.2%) patients. The right ear was affected in 96 (59.3%) patients, the left ear in 60 (37.0%) patients, and both ears in two (1.2%) patients; the affected side was not determined in four (2.4%) patients (Table 1). Pearson’s chi-squared test revealed a higher incidence on the right side of the PC (p < 0.001) (Table 2).

As expected given the inclusion/diagnostic criteria based on the AAO-HNSF guideline framework, all patients responded “yes” to three key symptom questions: provocation by head position change, duration of 5–60 s, and nausea. In total, 142 patients (87.7%) also reported blurry vision.

The mean diagnostic delay (interval from symptom onset to BPPV diagnosis at the first specialist dizziness-clinic visit) was 16.7 ± 17.1 days. Of the 162 patients, 102 (62.5%) were initially evaluated at other hospitals before referral. BPPV was correctly diagnosed in only six patients (3.7%) before referral. The remaining patients were either undiagnosed (n = 75) or misdiagnosed (n = 21) with conditions such as transient ischemic attack or stroke (Table 1).

MRI was performed in 88 patients (54.3%) before their specialist visit (No MRI, n = 74). The interval from symptom onset to the diagnosis of BPPV was significantly longer in the MRI group (mean ± SD, 24.4 ± 19.5 days) than in the no-MRI group (7.5 ± 5.9 days). The mean difference was 16.8 days (95% CI, 12.5 to 21.1 days; p < 0.001, Welch’s t-test) (Table 3). The identified MRI centers are summarized in Supplementary Table S2.

To further characterize the timing of delays among patients who underwent MRI, we calculated the interval from symptom onset to MRI acquisition and from MRI acquisition to the first specialist visit. Among patients with MRI (n = 88), the mean (± SD) time from symptom onset to MRI was 11.3 ± 10.9 days, and the mean time from MRI to specialist visit was 13.1 ± 13.6 days (Table 4).

In addition, we reviewed the available imaging reports and medical records to describe MRI acquisition pathways (patient-initiated vs. clinician-associated) and facility type (private vs. public/unknown) in this cohort. These findings motivated an additional review of MRI acquisition pathways and facility type, as described below.

Among the 88 patients who underwent pre-referral brain MRI, 68% (60/88) were documented as patient-initiated imaging (performed at the patient’s request or own decision according to available records). Among these patient-initiated cases (n = 60), 70% (42/60) underwent MRI in private hospitals, whereas the remaining 30% (18/60) were performed in public facilities or the facility name was not recorded. The referring specialty was not consistently documented; however, when identifiable, neurology was the most frequently recorded specialty (33 documented instances), followed by emergency physicians, general practitioners, and otolaryngologists.

Overall treatment resolution rates after 7, 14, and 28 days were 75.9% (95% CI 68.6–82.3), 93.8% (95% CI 88.9–97.0), and 99.4% (95% CI 96.6–100.0), respectively (Figure 2). When compared among clinical types, no significant differences in treatment success rates were observed (Table 5).

A significant improvement in DHI scores was observed following treatment (p < 0.001, paired t-test). The mean DHI score decreased from 39.93 ± 19.3 before treatment to 4.12 ± 4.6 after treatment (mean difference 35.8, 95% CI 32.9 to 38.7; p < 0.001). Before treatment, 52.46% of patients reported a moderate or severe handicap; after treatment, 99.4% reported no handicap (Table 6).

Patients who underwent MRI scans experienced a significantly longer delay from symptom onset to diagnosis (24.4 ± 19.5 days) compared to those who did not (7.5 ± 5.9 days; p < 0.001). However, in the multivariate logistic regression analysis, no specific clinical symptoms (e.g., hearing loss, severe nausea, or type of vertigo) were found to be significant predictors of MRI usage (all p > 0.05). Interestingly, the variable ‘no change in hearing’ was retained in the final stepwise model with an Odds Ratio of 2.18 (95% CI 0.77–7.14, p = 0.163), suggesting a paradoxical trend where patients without auditory symptoms were potentially more likely to undergo neuroimaging, although this did not reach statistical significance.

Clinical symptoms mimicking central pathology, such as hearing loss or continuous vomiting, were observed in less than 3% of patients. While all patients with these specific red flags underwent MRI (100%), they were excluded from the multivariate model due to their rarity causing complete separation. In the final stepwise logistic regression model excluding these sparse variables, no remaining clinical symptoms (e.g., visual blurring, tinnitus) significantly predicted MRI usage. This suggests that MRI utilization may be influenced by factors beyond the recorded clinical symptoms, rather than by specific “red flag” symptoms alone.

Based on 2024 healthcare cost metrics in Ulaanbaatar, we estimated the financial burden of potentially low-yield imaging. The average cost of a brain MRI in a mid-tier private hospital was estimated at 475,000 MNT (~$140 USD), whereas the specialist consultation fee was approximately 50,000 MNT (~$15 USD), yielding a cost ratio of 9.5.

For a median income earner in Mongolia, one MRI scan represents 21.3% of their monthly income. For minimum wage workers, the financial toxicity reaches catastrophic levels, with a single scan consuming 72.0% of their monthly earnings. In our cohort, the 88 potentially low-yield MRIs resulted in a direct waste of approximately 41.8 million MNT (~$12,300 USD), excluding indirect productivity losses. The aggregated wasted cost of 41.8 million MNT observed in this single-center cohort alone could have covered the cost of establishing basic vestibular screening kits (e.g., Frenzel goggles) for dozens of primary care clinics.