Portal de Periódicos da CAPES

Reflux Symptoms in Professional Opera Choristers

2007; Elsevier BV; Volume: 132; Issue: 3 Linguagem: Inglês

10.1053/j.gastro.2007.01.047

1528-0012

Giovanni Cammarota, Giovanna Masala, Rossella Cianci, Domenico Palli, Pasquale Capaccio, Antonio Schindler, Lucio Cuoco, Jacopo Galli, Enzo Ierardi, O Cannizzaro, M. Caselli, Maria Pina Dore, Benedetta Bendinelli, Antonio Gasbarrini,

Dysphagia Assessment and Management

Background & Aims: A specific, occupation-related susceptibility of professional singers to experience gastroesophageal reflux was hypothesized. We investigated the prevalence of gastroesophageal reflux symptoms in a series of professional opera choristers in comparison with a general population sample. Methods: A total of 351 professional opera choristers from well-known choirs in different Italian regions were identified and a sample of 578 subjects residing in the same areas with a similar distribution in age and sex was selected. Reflux symptoms in the year preceding the survey together with selected individual characteristics and lifestyle habits were investigated in both study groups through a structured questionnaire. Prevalence rate ratios, adjusted for sex, age, body mass index, smoking status, alcohol consumption, and other confounding factors, were computed. Results: Opera choristers reported a statistically significant higher prevalence of heartburn, regurgitation, cough, and hoarse voice than the population sample, with adjusted prevalent rate ratios of 1.60 (95% confidence interval [CI], 1.32–1.94), 1.81 (95% CI, 1.42–2.30), 1.40 (95% CI, 1.18–1.67), and 2.45 (95% CI, 1.97–3.04), respectively. Regurgitation appeared to be associated consistently with the cumulative lifetime duration of singing activity (P = .04) and with the weekly duration of singing activity (P = .005) when different multivariate models were applied. Conclusions: Opera choristers reported a higher prevalence of reflux symptoms than the population sample. Future studies will be needed to clarify whether gastroesophageal reflux in professional opera choristers is stress-induced and therefore may be considered as a work-related disease. Background & Aims: A specific, occupation-related susceptibility of professional singers to experience gastroesophageal reflux was hypothesized. We investigated the prevalence of gastroesophageal reflux symptoms in a series of professional opera choristers in comparison with a general population sample. Methods: A total of 351 professional opera choristers from well-known choirs in different Italian regions were identified and a sample of 578 subjects residing in the same areas with a similar distribution in age and sex was selected. Reflux symptoms in the year preceding the survey together with selected individual characteristics and lifestyle habits were investigated in both study groups through a structured questionnaire. Prevalence rate ratios, adjusted for sex, age, body mass index, smoking status, alcohol consumption, and other confounding factors, were computed. Results: Opera choristers reported a statistically significant higher prevalence of heartburn, regurgitation, cough, and hoarse voice than the population sample, with adjusted prevalent rate ratios of 1.60 (95% confidence interval [CI], 1.32–1.94), 1.81 (95% CI, 1.42–2.30), 1.40 (95% CI, 1.18–1.67), and 2.45 (95% CI, 1.97–3.04), respectively. Regurgitation appeared to be associated consistently with the cumulative lifetime duration of singing activity (P = .04) and with the weekly duration of singing activity (P = .005) when different multivariate models were applied. Conclusions: Opera choristers reported a higher prevalence of reflux symptoms than the population sample. Future studies will be needed to clarify whether gastroesophageal reflux in professional opera choristers is stress-induced and therefore may be considered as a work-related disease. Gastroesophageal reflux disease (GERD), affecting 10%–30% of the population in Western countries, is one of the most common conditions reported to primary care physicians and gastroenterologists. It is a chronic disease that results from the abnormal exposure of the esophageal mucosa to refluxed gastric contents (including acid and pepsin).1Holtmann G. Reflux disease: the disorder of the third millennium.Eur J Gastroenterol Hepatol. 2001; 13: S5-S11PubMed Google Scholar, 2Locke 3rd, G.R. Talley N.J. Fett S.L. Zinsmeister A.R. Melton III, L.J. Prevalence and clinical spectrum of gastroesophageal reflux: a population-based study in Olmsted County, Minnesota.Gastroenterology. 1997; 112: 1488-1496Abstract Full Text PDF Scopus (1933) Google Scholar Heartburn is the predominant symptom of GERD, impairing patient quality of life as determined by the frequency and severity of heartburn, irrespective of whether or not esophagitis also exists.3Stanghellini V. Armstrong D. Monnikes H. Bardhan K.D. Systematic review: do we need a new gastro-oesophageal reflux disease questionnaire?.Aliment Pharmacol Ther. 2004; 19: 463-479Crossref PubMed Scopus (100) Google Scholar, 4Revicki D.A. Wood M. Maton P.N. Sorensen S. The impact of gastroesophageal reflux disease on health-related quality of life.Am J Med. 1998; 104: 252-258Abstract Full Text Full Text PDF PubMed Scopus (414) Google Scholar Thus, it has been widely accepted that GERD is likely to be present when heartburn and/or regurgitation occur 2 or more days a week.5Dent J. Brun J. Fendrick A.M. Fennerty M.B. Janssens J. Kahrilas P.J. Lauritsen K. Reynolds J.C. Shaw M. Talley N.J. An evidence-based appraisal of reflux disease management—the Genval Workshop Report.Gut. 1999; 44: S1-S16Crossref PubMed Google Scholar, 6Dent J. Jones R. Kahrilas P. Talley N.J. Management of gastro-oesophageal disease in general practice.BMJ. 2001; 322: 344-347Crossref PubMed Scopus (100) Google Scholar, 7Dent J. Armstrong D. Delaney B. Moayyedi P. Talley N.J. Vakil N. Symptom evaluation in reflux disease: workshop background, processes, terminology, recommendations and discussion outputs.Gut. 2004; 53: 1-24PubMed Google Scholar However, a range of other symptoms can be associated with the disease, including chest pain and dysphagia. In addition, in the past decade, a variety of laryngopharyngeal (extra-esophageal) signs and symptoms, including chronic cough and hoarseness, often have been related to GERD. GERD, therefore, places a major burden on health care services worldwide and impacts society by affecting time off from work and work productivity.8Henke C.J. Levin T.R. Henning J.M. Potter L.P. Work loss costs due to peptic ulcer disease and gastroesophageal reflux disease in a health maintenance organization.Am J Gastroentrol. 2000; 95: 788-792Crossref PubMed Google Scholar, 9Vakil N. Ryden-Bergsten T. Bergenheim K. Systematic review: patient-centered and points in economic evaluations of gastroesophageal reflux disease.Aliment Pharmacol Ther. 2002; 16: 1469-1480Crossref PubMed Scopus (16) Google Scholar, 10Wahlqvist P. Carlsson J. Stalhammar N. Wiklund I. Validity of a work productivity and activity impairment questionnaire for patients with symptoms of gastro-esophageal reflux disease (WPAI-GERD)—results from a cross sectional study.Value Health. 2002; 5: 106-113Abstract Full Text PDF PubMed Scopus (133) Google Scholar In normal subjects, as in patients with pathologic reflux, nearly all episodes of reflux are caused by 1 of the following 4 pathologic mechanisms: transient complete relaxation of the lower esophageal sphincter, transient increase in intra-abdominal pressure that overcomes the resistance of the antireflux barrier (stress reflux), spontaneous reflux through a permanently hypotonic sphincter, and other mechanisms related to anatomic distortion of the gastroesophageal junction as with hiatus hernia.11Kahrilas P.J. Supraesophageal complications of reflux disease and hiatal hernia.Am J Med. 2001; 111: 51S-55SAbstract Full Text Full Text PDF PubMed Google Scholar, 12van Herwaarden M.A. Samsom M. Smout A.J. Excess gastroesophageal reflux in patients with hiatus hernia is caused by mechanisms other than transient LES relaxations.Gastroenterology. 2000; 119: 1439-1446Abstract Full Text Full Text PDF PubMed Scopus (259) Google Scholar Moreover, a balance of aggressive and defensive factors can underlie the occurrence of reflux. We hypothesized a specific susceptibility of opera choristers to stress reflux, specifically related to this occupation. In fact, opera choristers perform singing tasks that require rapid changes of subglottal pressure with consistent use of the diaphragm. They regularly activate the diaphragm when there is a need for a rapid decrease in subglottal pressure, which causes an abrupt and prolonged increase in intra-abdominal pressure, deep inspiration, and straining. These mechanisms, repeated several times per day and over many years of professional activity, could, in theory, increase the occurrence of reflux symptoms by disabling the diaphragmatic sphincter. If our hypothesis is true, opera choristers experience what may be considered as a work-related disease. This multicenter study was designed to evaluate the prevalence of gastroesophageal reflux symptoms (GERS) and GERD, in the year preceding the study, in a series of professional opera choristers as compared with a sample of the general population (controls). The directors of several choirs in different towns in Italy were contacted to verify their availability to participate in the study. Ten choirs from 9 different Italian towns gave their consent to participate in the study. These well-known choirs included the choir of the Rome Opera Theatre; the choir of the "Teatro Comunale" in Florence, the choir of the "La Fenice" Theatre in Venice, the choir of the "S. Carlo" Theatre in Naples, the choir and the Conservatory of music "Francesco Cilea" in Reggio Calabria, the choir of "La Scala" Theatre in Milan, the choir of "La Polifonica di S. Cecilia" in Sassari, the choir of the "Giordano" Theatre in Foggia, and the choir of the "Vittore Veneziani" Choral Academy and the "S. Gregorio" choir in Ferrara. A study questionnaire was delivered by hand to each chorister who was present during a 1-hour break of a daily song test on a specific date for each choir. A local researcher briefly introduced the aim of the study before distributing the questionnaire. The communication was aimed to explain the purpose of the study, to emphasize the importance of the answers of each subject, to ensure confidentiality of the information provided (the questionnaire was structured in an anonymous format), and to thank the subjects for their participation. The subjects were given 1 hour to complete the questionnaires, which then were collected immediately. Of the 398 opera choristers who were present during the questionnaire distribution, 47 (11.8%) refused to participate in the study. Of the remaining 351 choristers (157 men, 194 women) who completed the study questionnaire, 64 were recruited in Rome, 62 in Florence, 42 in Venice, 27 in Naples, 34 in Reggio Calabria, 27 in Milan, 25 in Sassari, 22 in Foggia, and 50 in Ferrara. Furthermore, 151 opera choristers of the 10 enrolled choirs were not present (as a result of various reasons) at the song tests on the specific day of the survey and therefore were not contacted further. Our control group consisted of a general population sample enrolled in the same study areas as for the opera choristers. These subjects were recruited at the hospital entrance among the relatives and visitors of patients admitted to the local hospitals in the study period (July 2004 to October 2005) and were frequency matched with opera choristers for age and sex. Among the 631 potentially eligible subjects contacted, 53 (8.38%) refused to participate in the study (principally because of a lack of time). Thus, a total of 578 (246 men, 332 women) subjects were recruited and formed our control group. They received the same survey questionnaire from the same researchers after a brief explanation of the study. As described previously, the questionnaire had to be completed in 1 hour. All enrolled subjects gave their consent to participate in the study. In the first phase of the study, 2 gastroenterologists (G. C., G. G.) developed a self-administered questionnaire. The questionnaire was not designed as a diagnostic tool for physicians, but rather with the aim to screen for GERS. The questionnaire included a word picture to describe the symptoms to facilitate the interpretation of the questions. It was assumed that the symptoms of GERD would be similar in each recruitment center. Therefore, the judgment of all local researchers involved in the study was sought to confirm that the questionnaire was comprehensible to the study subjects.13Talley N.J. Wiklund I. Patient reported outcomes in gastroesophageal reflux disease: an overview of available measures.Qual Life Res. 2005; 14: 21-33Crossref PubMed Scopus (28) Google Scholar The questionnaire contained general questions about individual characteristics and lifestyle habits and specific questions investigating GERS presence (heartburn, regurgitation, chest pain, dysphagia, hoarseness, and cough in detail) according to the modified questionnaire by Locke et al2Locke 3rd, G.R. Talley N.J. Fett S.L. Zinsmeister A.R. Melton III, L.J. Prevalence and clinical spectrum of gastroesophageal reflux: a population-based study in Olmsted County, Minnesota.Gastroenterology. 1997; 112: 1488-1496Abstract Full Text PDF Scopus (1933) Google Scholar, 14Locke G.R. Talley N.J. Weaver A.L. Zinsmeister A.R. A new questionnaire for gastroesophageal reflux disease.Mayo Clin Proc. 1994; 69: 539-547Abstract Full Text Full Text PDF PubMed Scopus (399) Google Scholar and taking into account the experts' recommendations.7Dent J. Armstrong D. Delaney B. Moayyedi P. Talley N.J. Vakil N. Symptom evaluation in reflux disease: workshop background, processes, terminology, recommendations and discussion outputs.Gut. 2004; 53: 1-24PubMed Google Scholar The following definitions therefore were used to identify 6 principal GERS in the questionnaire: (1) heartburn: an epigastric or retrosternal burning pain or discomfort feeling that rises into the chest; (2) regurgitation: a bitter- or sour-tasting fluid influx into the throat or mouth; (3) chest pain: any pain or discomfort felt inside the chest but not including heartburn or any pain that is primarily in the abdomen; (4) dysphagia (trouble swallowing): a feeling that food sticks in the throat or chest; (5) hoarseness: rough and harsh voice; and (6) cough: persistent coughing episodes as often as 4–6 times a day. The frequency of these symptoms was measured on the following scale: 1, never (in the past year); 2, less than once a month; 3, approximately once a month; 4, approximately once a week; 5, several times a week; and 6, daily. A subject was deemed to be suffering from GERS when he/she reported having the symptoms during the year preceding the survey. Only those subjects experiencing heartburn and/or regurgitation more than once a week were considered to have GERD.5Dent J. Brun J. Fendrick A.M. Fennerty M.B. Janssens J. Kahrilas P.J. Lauritsen K. Reynolds J.C. Shaw M. Talley N.J. An evidence-based appraisal of reflux disease management—the Genval Workshop Report.Gut. 1999; 44: S1-S16Crossref PubMed Google Scholar, 6Dent J. Jones R. Kahrilas P. Talley N.J. Management of gastro-oesophageal disease in general practice.BMJ. 2001; 322: 344-347Crossref PubMed Scopus (100) Google Scholar, 7Dent J. Armstrong D. Delaney B. Moayyedi P. Talley N.J. Vakil N. Symptom evaluation in reflux disease: workshop background, processes, terminology, recommendations and discussion outputs.Gut. 2004; 53: 1-24PubMed Google Scholar Subjects who already were known to have GERD (because of a previous diagnosis and after antireflux treatment) were considered as having both heartburn and regurgitation several times a week. Subjects experiencing GERS and who were under specific treatment were requested to report characteristics of the symptoms before therapy. The study subjects also were asked to classify the intensity of their heartburn and regurgitation as follows: mild, when easily tolerated; moderate, when the discomfort interfered with normal activities, and severe/very severe, when the subject was incapacitated and unable to perform normal activities. Other questions including additional upper-gut symptoms (such as epigastric pain, vomiting, and nausea), physician visits for GERD, and medication use also were collected. Sociodemographic and individual characteristics including age and sex, weight and height (self-reported), lifestyle habits such as tobacco and alcohol consumption, physical activity in leisure time (yes/no), and eating habits (whether or not dinner was consumed late in the evening) also were collected. Opera choristers were requested to record, in detail, the number of years of singing activity, the number of hours per week of singing exercises, and their singing specialty (soprano, mezzo, and alto for women; tenor, baritone, and bass for men). Finally, the general population subjects were requested to record their occupation. The distribution of a series of individual characteristics among opera choristers and the general population sample was compared. Continuous variables were expressed as the mean and SD and comparisons between groups were performed with the t test. Comparisons of categoric variables were performed with the χ2 test. The recruitment centers were divided into 3 geographic areas within the national territory (north, center, south), according to the area of residence, which may potentially reflect varying cultural characteristics, lifestyle habits, and terminologies. The comparison of the frequency distribution of each symptom, according to the classification adopted in the questionnaire, between the 2 groups was performed with the χ2 test. A P value of less than .05 (2-tailed) was considered significant. The prevalence of each symptom was calculated based on the presence of the symptom at least once in the year before the survey. The crude prevalence rate ratios (PRRs) of each symptom according to the presence or absence in the year before the survey (yes/no) among opera choristers and the control population were computed. To consider possible differences between the 2 study groups in lifestyle habits and anthropometric characteristics potentially related to GERD, we also computed the PRR adjusted for sex, age (years), area of residence (northern, central, and southern Italy), smoking habits (never smokers, former smokers, and current smokers), drinking habits (yes/no), body mass index (BMI) (<25; 25–29; ≥30), physical activity in leisure time (yes/no), habit of consuming dinner late in the evening (yes/no); and 95% confidence intervals (CIs) were computed.15Spiegelman D. Hertzmark E. Easy SAS calculation for risk or prevalence ratios and differences.Am J Epidemiol. 2005; 162: 199-200Crossref PubMed Scopus (1477) Google Scholar The same multivariate model was applied to evaluate the PRR of symptoms according to the duration of singing activity as well as the number of hours/week of exercise. Because of the relevance as a potential confounder of BMI,16Aro P. Ronkainen J. Talley N.J. Storskrubb T. Bolling-Sternevald E. Agreus L. Body mass index and chronic unexplained gastrointestinal symptoms: an adult endoscopic population based study.Gut. 2005; 54: 1377-1383Crossref PubMed Scopus (96) Google Scholar, 17Jacobson B.C. Somers S.C. Fuchs C.S. Kelly C.P. Camargo Jr, C.A. Body-mass index and symptoms of gastroesophageal reflux in women.N Engl J Med. 2006; 354: 2340-2348Crossref PubMed Scopus (490) Google Scholar all the analyses also were performed based on models in which BMI was considered as a continuous variable. Analyses were performed using SAS statistical software (SAS/STAT version 9.1; SAS Institute Inc, Cary, NC). Among the 351 professional opera choristers who consented to participate in the survey, the mean age was 40.8 years (SD 10.3 years). Among the 157 men, 76 reported being tenors, 38 were baritones, and 36 were bass. Among the 194 women, 102 reported being sopranos, 46 were mezzos, and 39 were altos. For 14 subjects the information was not available. The mean duration of singing activity was 18.1 years (SD 9.9 years), with a mean amount of 11.2 hours per week (SD 8.3 hours/week) of singing exercise. Overall, in our control group formed by 578 subjects from the general population (246 men), the mean age was 36.2 years (SD 11.8 years). Approximately 71% (409 of 578) reported being currently employed; none reported being a professional singer. The most frequent occupation reported was that of clerk (26.4%). Overall, the prevalence of overweight and obesity was significantly higher among opera choristers in comparison with the general population sample. In addition, the prevalence of never-smokers and the proportion of subjects who reported being physically active in their leisure time as well as those who reported consuming dinner late was significantly higher among choristers. In comparison with the control sample, the opera choristers reported a less frequent consumption of alcoholic beverages (Table 1).Table 1Distribution of Selected Individual Characteristics in 351 Opera Choristers and 578 Subjects From the General PopulationOpera choristersGeneral population sampleN(%)N%Sex Men157(44.73)246(42.56) Women194(55.27)332(57.44)P = .518BMI Underweight/normal (<25)194(55.27)408(70.71) Overweight (≥25 and <30)133(37.89)139(24.09) Obese (≥30)24(6.84)30(5.20)P < .0001Area of residence North118(33.62)187(32.35) Center129(36.75)236(40.83) South104(29.63)155(26.82)P = .438Smoking status Never smoker218(62.11)308(53.38) Ex-smoker78(22.22)107(18.54) Current smoker55(15.67)162(28.08)P < .0001Alcohol consumption No89(25.36)112(19.38) Yes262(74.64)466(80.62)P = .032Physical activity in leisure time No143(40.86)310(53.91) Yes207(59.14)265(46.09)P = .0001Consume dinner late in the evening No106(30.2)305(52.95) Yes245(69.8)271(47.05)P < .0001NOTE. Because of missing data not all numbers add up to the total. P values are from the χ2 test (n − 1 degree of freedom). Open table in a new tab NOTE. Because of missing data not all numbers add up to the total. P values are from the χ2 test (n − 1 degree of freedom). The height was similar in the 2 study groups although the average weight was significantly higher among choristers (Table 2).Table 2Mean Values (and SD) of Age and Selected Anthropometric Measurement of 351 Opera Singers and 578 Subjects From the General PopulationOpera choristersGeneral population samplet test P valuesMeanSDMeanSDAge40.7610.2936.211.85.004Height, m1.69.081.68.09.260Weight, kg71.316.2367.0113.14<.0001 Open table in a new tab When we considered the average weight by sex this difference still was evident in men (average weight, 82.0 kg [SD 14.9 kg] in opera choristers and 76.0 kg [SD 11.1 kg] in men from the general population sample; P < .0001), but was not apparent in women (average weight, 62.6 kg [SD 11.4 kg] in opera choristers and 60.4 kg [SD 10.2 kg] in women from the general population sample; P = .087). The distributions of GERS and other symptoms in opera choristers and in the general population sample according to the frequency scores are shown in Table 3, Table 4. Heartburn, regurgitation, cough, and hoarseness were significantly different between the 2 groups, both when subjects were classified according to the more detailed classification (Table 3) and when the "ever/never in the last year" classification was adopted (Table 4).Table 3Distribution of the Self-Reported Frequency of a Series of GERS in 351 Opera Choristers and 578 Subjects From the General PopulationOpera choristersGeneral population sampleN(%)N(%)Heartburn Never203(57.83)434(75.09) 1/wk12(3.42)5(.87) 1/day5(1.42)6(1.04)P < .0001Regurgitation Never229(65.24)477(82.53) 1/wk5(1.42)3(.52) 1/day2(.57)5(.87)P < .0001Chest pain Never275(79.02)482(83.39) 1/wk5(1.44)5(.87) 1/day0(.00)1(.17)P = .271Dysphagia Never279(79.49)469(81.14) 1/wk13(3.70)8(1.38) 1/day5(1.42)7(1.21)P = .062Epigastric pain Never199(56.70)330(57.09) 1/wk8(2.28)15(2.60) 1/day0(.00)3(.52)P = .263Nausea Never256(72.93)395(68.34) 1/wk8(2.28)12(2.08) 1/day0(.00)5(.87)P = .408Vomiting Never306(87.18)490(84.78) 1/wk1(.28)1(.17) 1/day0(.00)0(.00)P = .388Cough Never198(56.41)378(65.40) 1/wk11(3.13)14(2.42) 1/day5(1.42)9(1.56)P = .006Hoarseness Never198(56.41)463(80.10) 1/wk8(2.28)3(.52) 1/day2(.57)5(.87)P < .0001NOTE. Because of missing data not all numbers add up to the total. P values are from the χ2 test (n − 1 degree of freedom). Open table in a new tab Table 4GERS in 351 Opera Choristers and 578 Subjects From the General PopulationOpera choristersGeneral population sampleN(%)N(%)Heartburn Never in the previous year203(57.83)434(75.09) At least once in the previous year148(42.17)144(24.91)P < .0001Regurgitation Never in the previous year229(65.24)477(82.53) At least once in the previous year122(34.76)101(17.47)P < .0001Chest pain Never in the previous year275(79.02)482(83.39) At least once in the previous year73(20.98)96(16.61)P = .096Dysphagia Never in the previous year279(79.49)469(81.14) At least once in the previous year72(20.51)109(18.86)P = .537Epigastric pain Never in the previous year199(56.70)330(57.09) At least once in the previous year152(43.30)248(42.91)P = .905Nausea Never in the previous year256(72.93)395(68.34) At least once in the previous year95(27.07)183(31.66)P = .138Vomiting Never in the previous year306(87.18)490(84.78) At least once in the last year45(12.82)88(15.22)P = .310Cough Never in the previous year198(56.41)378(65.40) At least once in the previous year153(43.59)200(34.60)P = .006Hoarse voice Never in the previous year198(56.41)463(80.10) At least once in the previous year153(43.59)115(19.90)P < .0001NOTE. Frequency distributions (never/ever in the previous year) and percentages are shown. Because of missing data not all numbers add up to the total. P values are from the χ2 test (n − 1 degree of freedom). Open table in a new tab NOTE. Because of missing data not all numbers add up to the total. P values are from the χ2 test (n − 1 degree of freedom). NOTE. Frequency distributions (never/ever in the previous year) and percentages are shown. Because of missing data not all numbers add up to the total. P values are from the χ2 test (n − 1 degree of freedom). Among the 351 opera choristers, 148 (42.17%) reported heartburn in the previous year, whereas among the 578 subjects from the general population sample, 144 (24.91%) reported heartburn in the previous year (P < .0001). The crude and adjusted PRRs were 1.69 (95% CI, 1.40–2.04) and 1.60 (95% CI, 1.32–1.94), respectively (Table 5).Table 5Prevalence of GERS in Opera Choristers in Comparison With a Sample of the General PopulationCrude PRR95% CIAdjusted PRRaPRR adjusted for sex, age, area of residence (northern, central, and southern Italy), BMI class, physical activity, consuming dinner late, alcohol consumption, and smoking status.95% CIHeartburn1.691.40–2.041.601.32–1.94Regurgitation1.991.58–2.501.811.42–2.30Cough1.261.07–1.481.401.18–1.67Hoarseness2.191.79–2.782.451.97–3.04Chest pain1.26.96–1.661.23.92–1.64Dysphagia1.09.83–1.421.17.89–1.55Epigastric pain1.0187–1.181.12.96–1.30Nausea.85.69–1.05.95.76–1.17Vomiting.84.60–1.18.99.69–1.42NOTE. Prevalence of GERS occurring at least once in the previous year or never.a PRR adjusted for sex, age, area of residence (northern, central, and southern Italy), BMI class, physical activity, consuming dinner late, alcohol consumption, and smoking status. Open table in a new tab NOTE. Prevalence of GERS occurring at least once in the previous year or never. A total of 122 (34.76%) opera choristers and 101 (17.47%) subjects from the general population sample recorded regurgitation in the past year (P < .0001). The crude and adjusted PRRs were 1.99 (95% CI, 1.58–2.50) and 1.81 (95% CI, 1.42–2.30), respectively (Table 5). Heartburn was associated with regurgitation in 91 (25.9%) opera choristers and 63 (10.9%) controls at least once in the past year (crude PRR, 2.38; 95% CI, 1.78–3.19; adjusted PRR, 1.97; 95% CI, 1.45–2.67). Twenty-nine (8.26%) opera choristers and 21 (3.63%) subjects from the general population had these 2 associated symptoms at least once a month (crude PRR, 2.27; 95% CI, 1.32–3.92; adjusted PRR, 1.77; 95% CI, 1.01–3.11), whereas 8 (2.28%) opera choristers and 8 (1.38%) controls (P = .30) reported both heartburn and regurgitation at least once a week. In multivariate analyses, both heartburn and regurgitation prevalence in the choristers were associated significantly with the duration (in years) of the singing activity (P for trend was .02 and .04, respectively). Regurgitation also was associated significantly with the increasing number of hours spent in singing activity each week (P = .005). There was no significant differences between the prevalence of these symptoms and the singing specialties of the choristers (baritone, soprano, and so forth, data not shown). Finally, when heartburn and regurgitation were detailed for severity characteristics, opera choristers reported more severe symptoms (Table 6).Table 6Heartburn and Regurgitation Severity Scores in 351 Opera Choristers and 578 Subjects From the General PopulationOpera choristersGeneral population sampleN(%)N(%)Heartburn None203(57.83)434