Using the Berlin Questionnaire To Identify Patients at Risk for the Sleep Apnea Syndrome

The obstructive sleep apneahypopnea syndrome is a potentially disabling condition characterized by excessive daytime sleepiness, disruptive snoring, repeated episodes of upper airway obstruction during sleep, and nocturnal hypoxemia. Epidemiologic surveys indicate associations among snoring, sleep apnea, and cardiovascular disease (1). A 1993 population-based study (2) of workers in Wisconsin found that 2% of women and 4% of men had symptoms of sleepiness with associated levels of sleep apnea believed to indicate at least a moderate degree of illness. Prevalence estimates from other countries and other U.S. studies are similar (3-5). Recognition of sleep apnea by community physicians is, however, low. In the Wisconsin study (6), only 7% of women and 12% of men who had moderate to severe illness reported receiving a diagnosis of sleep apnea from a medical encounter. Two studies observed that specialist intervention with diagnostic equipment (7) or intensive physician education on taking a sleep history (8) improved recognition of sleep apnea in primary care practices. However, both approaches required substantial professional and technical resources. Asking patients to report their symptoms is a simple alternate approach that has been shown to be helpful in sleep referral clinics and community surveys (1). The Berlin Questionnaire asks about risk factors for sleep apnea, namely snoring behavior, waketime sleepiness or fatigue, and the presence of obesity or hypertension. We evaluated the usefulness of this instrument in identifying patients with sleep apnea in primary care settings. Methods The Berlin Questionnaire The Berlin Questionnaire was an outcome of the Conference on Sleep in Primary Care, which involved 120 U.S. and German pulmonary and primary care physicians and was held in April 1996 in Berlin, Germany. Questions were selected from the literature to elicit factors or behaviors that, across studies, consistently predicted the presence of sleep-disordered breathing (1, 9-15). By consensus, the instrument focused on a limited set of known risk factors for sleep apnea. One introductory question and four follow-up questions concern snoring; three questions address daytime sleepiness, with a subquestion about sleepiness behind the wheel (that is, while driving a motor vehicle). One question concerns history of high blood pressure. Patients are also asked to provide information on age, weight, height, sex, neck circumference, and ethnicity. Obesity was quantified by calculating body mass index from self-reported weight and height. The responses to these questions have utility in non-primary care settings (1). The conference also proposed a plan for risk grouping to simplify recognition of sleep apnea; this strategy was shown to be useful in sleep clinic and community surveys (11, 13, 15). Predetermination of high risk and lower risk for sleep apnea was based on responses in three symptom categories. In category 1, high risk was defined as persistent symptoms (>3 to 4 times/wk) in two or more questions about their snoring. In category 2, high risk was defined as persistent (>3 to 4 times/wk) waketime sleepiness, drowsy driving, or both. In category 3, high risk was defined as a history of high blood pressure or a body mass index more than 30 kg/m2. To be considered at high risk for sleep apnea, a patient had to qualify for at least two symptom categories. Those who denied having persistent symptoms or who qualified for only one symptom category were placed in the lower risk group. Survey Distribution One thousand questionnaires in batches of 200 per study site were provided to individual physicians at five sites in the Cleveland, Ohio, area. The sites were chosen on the basis of geographic and socioeconomic diversity (further information is available from the authors on request). Three physicians were solo practitioners and 2 were members of a practice group; all practices were part of a hospital-owned network that at the time of study included 92 primary care physicians who cared for adults. All 5 participating physicians were Board-certified in internal medicine, and 2 had more advanced training (rheumatology or pulmonary medicine). By design, all participating physicians had practiced primary care medicine for more than 4 years and had stable practice patterns, each handling a panel of 2500 to 3000 patients. According to network records, no physician had referred more than 2 patients for sleep studies in the previous year. Office staff handed out questionnaires to consecutive patients who visited the study physician for any reason. Each site was instructed to return the questionnaires to the sleep center. Completed questionnaires were included in our analysis if they met the following criteria: They had to be dated, the date had to fall within 3 weeks of distribution, and they had to be returned to the sleep center within 1 month. The study was approved by the institutional review board of University Hospitals of Cleveland. Sleep Studies Portable monitoring of respiratory disturbances during sleep was offered to both high-risk and lower-risk patients. The intent was to study approximately 20% of respondents, equally distributed in both risk groups. From an alphabetically ordered list, the first 75 patients in the high-risk group and the first 65 patients in the lower-risk group were contacted by telephone and asked to participate. Patients who agreed to sleep studies were visited at home, instructed on the use of the monitor, and monitored overnight; the monitor was retrieved the next day. Patients gave written consent for portable monitoring and for results to be sent to their primary care physician. Monitoring was performed with a six-variable, four-channel Eden Tec recorder (Nellcor Puritan Bennett, Minneapolis, Minnesota). Variables measured included nasal and oral airflow by thermistor, chest wall movement by impedance electrodes, and oxygen saturation (Sao 2) and pulse rate by pulse oximeter. A respiratory disturbance event was defined as a decrease in nasal or oral airflow, alone or with chest wall movement of approximately 50% that lasted for 10 seconds or more. A decrease in Sao 2 of 4% or more was considered significant oxygen desaturation. The recorder was taken to the patient's home, where he or she was instructed on how to use the recording device and to turn it on at bedtime and to turn it off upon arising (13). Measurements from a full-disclosure printout were manually scored for a respiratory disturbance index (RDI) (measured as the number of respiratory events per hour in bed) and the oxygen desaturation index (number of decreases in Sao 2 of 4% per hour in bed). Acceptable records were those in which the patients spent at least 6 hours in bed and good to excellent recording of Sao 2 and respiration (either impedance or thermistor records or both) was achieved (13). A single researcher who had no knowledge of the questionnaire results performed the scoring. Statistical Analysis The quantitative distribution of returned questionnaires, individual patient variables, responses to individual questions about sleep-related symptoms, and results of home sleep monitoring are expressed by descriptive statistics (frequencies, mean SD, and range). Missing data and data that are not applicable are expressed in the percentage of the returned questionnaires and in total number of patients for each variable. Answers to questions on sex and study site were evaluated by using the chi-square test and were expressed by the significance level. The Pearson correlation test and level of significance were used to compare questionnaire responses and risk groupings. We used a logistic regression model that examined the relative effects of age, sex, and the three symptom categories and risk group. The predictive accuracy (16) of risk grouping and of each category was assessed for RDIs of 5 or less, more than 5, more than 15, and more than 30; these arbitrary cut-off values are similar to those used in previous studies (2, 6) and those proposed as diagnostic criteria (17). Computations were performed by using SPSS 7.5 for Windows (SPSS, Inc., Chicago, Illinois). Results Of 1008 questionnaires (one physician had distributed an additional 8 questionnaires), 744 (74%) were entered for analysis. The variability in return rate resulted from time constraints and unavailability of staff rather than patient refusal. The return rate did not correlate with the socioeconomic profile of the practice site; solo practices had greater response rates. One male respondent and one female respondent reported that they had received a diagnosis of or treatment for sleep apnea; their results were included in the analysis. Characteristics of the respondents are shown in Table 1. Because responses to the questions on neck circumference and ethnicity were often not provided, these results were not included in the analyses. Table 1. Self-Reported Characteristics of the 744 Survey Respondents Prevalence of Symptoms Of the 744 respondents, 388 (52.2%) reported that they snored, 223 (30.0%) denied snoring, 118 (15.9%) did not know whether they snored, and 15 (2%) did not respond to this question. Ninety-four of all respondents (24.6%) reported that their snoring was louder than normal speech and 289 (75.4%) did not snore louder than normal speech. Two hundred three (47.9%) respondents reported snoring at least three to four times per week, and 221 (52.1%) said that they did not snore more than one to two times per week. Two hundred seventy-four (54.9%) respondents reported that their snoring bothered other people, whereas 225 (45.1%) denied that it did. In 66 (11.1%) respondents, breathing pauses during sleep were observed by others at least 1 to 2 times per month; in 31 (5.2%) respondents, breathing pauses were observed more than 3 to 4 times per week. Two hundred forty-three (33.8%) respondents stated that they did not feel rested after a