Obstructive sleep apnea syndrome (OSA) is a multifactorial clinical condition whose pathophysiology is produced by the conjunction of anatomical, muscular, neurological factors and other factors not yet identified, causing a collapse of the upper airway as a result of an imbalance between forces that tend to close it and those that keep it open, resulting in multiple episodes of complete (apnea) or partial (hypopnea) obstruction of the upper airway. This obstruction leads to poor sleep quality and intermittent hypoxemia with vascular repercussions (1). OSA is a risk factor related to cardiovascular diseases (1–3), hypertension, and stroke (4, 5), so the frequent association with other comorbidities makes management more complex. The most frequent nocturnal symptoms are snoring, observed apneas, gasping or choking episodes, abnormal movements, diaphoresis, or frequent awakenings. In contrast, the usual daytime symptoms are excessive daytime sleepiness, nonrestorative sleep, tiredness, headache, irritability, or depression (3). The most important risk factors are age, male sex, and body mass index (BMI), and variables influencing their onset or aggravation are alcohol, tobacco, sedatives, hypnotics, barbiturates, and supine position (5).

OSA is a highly prevalent condition (1, 3, 6–8). Benjafield et al. (6), in their study indicate OSA prevalence in the general population is between 4% and 30%. The International Consensus Document on obstructive sleep apnea, indicates that in men over 40 years of age, the prevalence of OSA ranges between 48% and 73% (7). In Spain, between 3% and 6% of the population suffers from OSA, with a very severe condition between 24% and 26% of this (3). It affects 0.7%–3% of children aged between 4 and 5 years (8). In adult women, the prevalence of OSA is 2%–4%, with the ratio between men and women being 2-3/1, with a tendency to equalize after menopause. The prevalence of OSA increases with age, tripling in the elderly compared to the middle-aged population (8, 9).

In Spain and other countries, only 5%–9% of patients with OSA have been diagnosed and treated. The lack of accessibility to diagnosis is considered the main cause of this situation (10). Polysomnography (PSG) has been considered the gold standard for diagnosing OSA. However, PSG requires an infrastructure and human resources that not all centers of the health system can cover because it requires the nocturnal hospitalization of the subjects. One of the alternative methods to PSG to try to avoid these drawbacks is home respiratory polygraphy (HRP), a procedure that consists of nocturnal monitoring of oxygen saturation, oronasal airflow, and breathing movements; it can be performed at home by the patients themselves, with the convenient prior training of the patient. This tool has high accuracy, with a sensitivity of 93.3%–96.6%, a specificity of 82.9%–100%, and a positive predictive value of 96.5% for diagnosing OSA. In patients with a strong or moderate suspicion of OSA, HRP represents a procedure that can replace PSG. However, given its lower specificity, doubtful results, although few, should be confirmed by PSG (11, 12).

Currently, no curative treatment is available for OSA, but the mechanical problem (upper airway closure) can be solved with the use of continuous positive airway pressure (CPAP) devices (1, 3), which has been associated with decreased risk of falls (13), strokes, cardiovascular, morbidity and mortality (5), and with an improvement in health-related quality of life (14).

Given the high hidden morbidity of OSA, and the lack of accessibility to the diagnostic procedures usually used, OSA screening tools, such as the Berlin questionnaire (BQ) (12), could be useful tools for healthcare professionals in primary care settings (PC). To date, no validated version of the BQ is available in Spain, so more studies are needed to provide higher consistency to the results obtained.

The main aim of this study was to validate the Spanish translation of the BQ to detect moderate or severe OSA in a population aged 40 years or older residing in Spain. The specific objectives were: (a) to perform the transcultural adaptation of the BQ from English into Spanish, (b) to determine the criterion validity, and (c) to verify its reproducibility in terms and its internal consistency.

The cross-cultural adaptation was successful because 8 out of the 10 items of the BQ were equivalent, and only 2 items showed minor modifications without affecting the meaning of the question. Consequently, this questionnaire was used to analyze its validity and reliability.

Initial eligibility of 283 was evaluated, of which 255 were invited to participate, signing the informed consent. No patient was excluded or withdrew from the study (completion rate: 100%).

The patients had a mean age of 54.76 years (SD: 6.57; range: 40–60; 95% CI: 53.53–55.99), without statistical differences between men and women, men being 54.12% (95% CI: 47.96–60.27). Regarding the BMI, the mean was 31.17 kg/m² (SD: 6.57; range: 19.53–60.18; 95% CI: 30.36–31.98), and the mean neck circumference was 38.33 cm (SD: 5.78; range: 24–53; 95% CI: 37.62–39.05), with significant differences by sex in neck circumference (p < 0.001) and BMI (p = 0.002).

A total of 61.57% (95% CI: 55.57–67.57) of the patients were diagnosed with OSA (AHI >5). Moderate/severe OSA (AHI >15) was found in 38.43% of subjects (Table 1).

Table 2 shows the results of criterion validity parameters according to sex. The BQ validated in Spanish for the population of 40 years or older, with a cut-off point of 4.5, presented a sensitivity of 76.77, a specificity of 74.49%, a positive predictive value of 75.25% (95% CI: 66.34–84.16), a negative predictive value of 76.04% (95% CI: 66.98–85.10), a LR⁺ of 3.01, and LR⁻ of 0.312. The AUC was 0.786 (95% CI: 0.721–0.851) (Figure 1).

Regarding the reliability of the questionnaire, Cronbach’s alpha coefficient was 0.730 (95% CI: 0.668–0.784), and the kappa index was 0.739 (95% CI: 0.384–1.000).

The BQ is an easy and quick to complete tool. From its origin, it has already proven to be of clinical utility as a screening tool because it has a high predictive capacity to detect OSA in those patients with high clinical suspicion of presenting this condition. This questionnaire resulted from the Sleep Conference, which brought together 120 American and German family practice physicians and sleep researchers in Berlin, Germany, in 1966 (12). Questions were selected from the literature to obtain factors or behaviors that, in all studies, systematically predicted the presence of sleep-disordered breathing. By consensus, the tool focused on a limited set of known risk factors for OSA. Its validation was conducted on 744 adults, of which only 100 underwent sleep studies, representing a methodological limitation. It presented a sensitivity of 86%, a specificity of 77%, and a positive predictive value of 89%. Questions about symptoms demonstrated good internal consistency (Cronbach’s alpha from 0.86 to 0.92) (12).

In the Spanish language, the validation of this questionnaire in the Colombian population detected in sleep centers was published in 2013 (19). This validation obtained a sensitivity of 87%, a specificity of 70%, a PPV of 98%, a NPV of 21%, and an AUC of 0.785, presenting an acceptable internal consistency with a Cronbach’s alpha coefficient of 0.725, and a high reproducibility, with a Cohen’s kappa index of 0.815. The authors found a clinically and statistically significant correlation between the adaptation performed and the PSG.

No validated version of the BQ for the Spanish adult population was available to date. Therefore, for cultural reasons or idiosyncrasies of each Spanish-speaking country (and probably also clinical or epidemiological reasons), it is necessary to ensure that it has the same qualities regarding its psychometric properties when applied to a population recruited in PC.

In the present study, we have successfully translated the BQ according to established guidelines. This study aligned with previous studies that examined sensitivity, specificity, PPV, and NPV. Table 3 shows that the sensitivity results are similar to the studies conducted in Greece (20), which are slightly higher than in Denmark (21). However, the specificity of our study is higher than that obtained in the rest of the published studies with which it has been compared.

There are other questionnaires, such as the STOP-Bang (22–27) for the detection of OSA, that also show consistent results, as we have been able to confirm by administering it to the same sample of our population (23). This questionnaire showed a sensitivity not very different to BQ (range: 84.85%–93.8%), but a lower specificity (range: 55.10% to 63.4%) (23, 24, 26, 27), although presents the advantage of its greater usability, being even easier to fill in and interpret than the BQ.

Our study demonstrates that BQ has a highly diagnostic capacity to detect clinically relevant OSA (AHI ≥15). BQ is simple and easy to understand and fill in, so it can be applied in a PC setting. The use of BQ could potentially identify many patients with OSA, especially when it is moderate or severe, leading to earlier correct diagnosis and faster treatment, decreasing the risk of complications and unwanted effects. With the translation, back-translation, and consensus of experts, a final version of the BQ applicable to Spain was obtained, which is understandable and usable, and retains the meaning of the questionnaire in its source language. This final version can be used for any Spanish patient over 40 years old, regardless of gender, educational level, and/or socioeconomic status. Extending the study with a PSG would be indicated only in the cases of a possible mild OSA. The results of this validation were similar to those of the source version, obtaining a qualified, affordable, and easy-to-use prediction tool in the PC setting with high diagnostic efficacy and acceptable reliability. This tool allows identifying patients with a higher risk of having OSA to schedule treatment more quickly and thus reduce the associated comorbidities.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

The studies involving human participants were reviewed and approved by Comite de etica e investigacion clinica del hospital Reina Sofia, Córdoba, Spain. The patients/participants provided their written informed consent to participate in this study.

RM-G, JS-M, LAP-dT, MV-A, and EN-M contributed to the conception and design of the study. JS-M, EN-M, and LAP-dT reviewed the database and performed the statistical analysis. EN-M wrote the first draft of the manuscript. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.