Epidemiology of chronic bronchitis in Prague

Bull. Org. mond. SantJ } 1970,42,711-722Bull. Wid Hith Org. Epidemiology of Chronic Bronchitis in Prague F. BOUDIK,1 J. R. GOLDSMITH,2 V. TEICHMAN1 & P.-C. KAUFMANN a A number of epidemiological studies in several countries have implicated cigarette smoking, occupation, age, sex and air pollution as well as other environmental factors in the epidemiology of chronic respiratory disease. This paper reports the results of a respiratory-symptom and pulmonary-function survey among a large population of men aged 50-65 years in Prague. In general, the findings of other studies have been confirmed, showing that these relationships occur in an even wider range of geographic and cultural backgrounds. In addition, the study uncovered an apparent relationship between the occurrence of respiratory symptoms and a family history of chronic bronchitis and tuber- culosis, suggesting that future surveys should include studies offamilial aggregations of chronic respiratory disease. Similar studies have been carried out in other parts of Czecho- slovakia with different levels of air pollution for comparison with the results of this study. A number of epidemiological surveys on pre- valence of chronic bronchitis have been carried out in different countries during the last 15 years- especially in the USA and the United Kingdom (Brinkman & Coates, 1962; College of General Practitioners, 1961; Deane, Goldsmith&Tuma, 1965; Ferris & Anderson, 1962; Fletcher et al., 1959; Gold- smith et al., 1962; Higgins et al., 1956). Evidence obtained suggested that several factors play an important role in the etiology of chronic bronchitis, for example, cigarette smoking (Deane, Gold- smith & Tuma, 1965; Ferris & Anderson, 1962; Fletcher et al., 1959; Goldsmith et al., 1962; Olsen & Gilsen, 1960; Stanek et al., 1966), envi- ronmental and social factors, as well as occupa- tion, age and sex (College of General Practitioners, 1961; Colley & Holland, 1967; Enterline, 1967; Holland & Reid, 1965; Kourilsky et al., 1966). Hereditary factors may also contribute to the development of chronic bronchitis (Oswald, Harold & Martin, 1953). The present paper presents results obtained from a large population in Prague, Czechoslovakia, con- 1 The 2nd Medical Clinic, Faculty of Medicine, Prague, Czechoslovakia. 'The Environmental Epidemiology Unit, California State Department of Public Health, Berkeley, Calif., USA, Work supported by a grant from the World Health Organi- zation, Geneva, Switzerland. ' Statistician, Health Statistical Methodology, World Health Organization, Geneva, Switzerland. firming the known relationship between the above- mentioned factors and chronic bronchitis. Some of our data suggest predisposing factors. Similar studies in areas with serious air pollution and less polluted rural areas in Czechoslovakia are being completed and the results will be reported later. METHODS Population selection, methods and criteria All males between the ages of 50 years and 65 years (born 1898-1913) were selected from the election lists in Prague 2. Before starting the survey in November 1964, the lists were adjusted to bring them up to date. The total number of males in the 50-65-year age-group was 11 842. From this figure, 441 subjects should have been excluded according to findings by the end of the survey for the following reasons: 116 males had moved, 36 males on the lists were older or younger than the required age, 205 had died, and 84 subjects with active pulmonary tuberculosis were not asked to attend the examination. The study lasted from November 1964 to the end of July 1965 and, after the second invitation, the total number of males available for evaluation was 8292. At the end of the survey the age range of the subjects was 52-67 years. In relation to the adjusted number of eligibles, the coverage in the survey was 72.8%. The short form of the British questionnaire (Medical Research Council Committee on the 2510 -711 712 F. BOUDIK AND OTHERS Aetiology of Chronic Bronchitis, 1960) of respiratory symptoms was used with minor modifications. The questionnaire together with an invitation to visit the clinic was sent by mail. The subject brought the completed questionnaire with him to the exa- mination, where the doctors and research technicians from the 2nd Medical Clinic checked the answers to the principal questions on cough, phlegm, dyspnoea, wheezing, smoking habits and occupation. From the non-respondents, a representative sample of 100 subjects was selected and examined for the prevalence of chronic bronchitis (using the ques- tionnaire only). No important differences were found in responses to major questions. For most of the subjects who completed the ques- tionnaire,radiographs 10cm x 10 cm(postero-anterior and right lateral projection in inspiratory position) were taken. On about half the subjects the following pulmonary-function tests were performed: forced expiratory spirogram (FES, from which the forced expiratory volume in 1 second, FEV1.o, and the forced ventilatory capacity, FVC, were evaluated) and peak expiratory flow rate (PEFR). Availability of equipment rather than the examiner's judgement determined which subjects were examined (except those who refused). We are not aware of any bias being introduced by not examining the entire group. We have compared the prevalence of symptoms in those with and without X-ray and with and without pulmonary-function tests and find no significant differences. For the FES examination a plastic, water-filled spirometer, and for PEFR determination Wright's peak-flow meter (Wright & McKerrow, 1959), were used. If the patient's co-operation was good, only 2 measurements for FES and PEFR were performed. Additional tests were made for subjects who had a difference between test values of more than 10%. The maximum values of FES and PEFR, adjusted to ambient temperature and pressure saturated with vapour (ATPS), were coded. Techni- cally imperfect records were excluded from the coding. Repeat measurements of PEFR and FEV1.0 were done in a subsequent follow-up study starting a year later. The first 100 values of both examinations of this follow-up study have been compared with those obtained during the prevalence survey. The correlation coefficients are 0.91 for FEV1.o and 0.90 for PEFR. For the diagnosis of chronic bronchitis the symptom group " persistent cough and persistent phlegm, lasting more than 2 years" was used (WHO Expert Committee on Cor Pulmonale, 1961). The term " persistent cough " was used when the following replies were obtained: Yes to morning cough in winter and/or Yes to day cough in winter for 3 months in the year. The term " persistent phlegm " was used in the similar sense. For the evaluation of the radiographs, an enlarge- ment to 35 cm x 35 cm by means of a " Helio- contrastor de Oude" was used. From the radiographs several pathological changes were evaluated, the most important of them being: tuberculous intraparenchymal changes (divided into 3 groups according to the extent), post-inflammatory, non-tuberculous intraparenchymal, and pleural changes (also divided into 3 groups). The data were processed on the WHO IBM 360/40 computer. For the training of doctors and research technicians participating on this survey the British instructions for evaluating the questionnaire were used.' The " unknown answers " were coded and evaluated separately for particular symptoms. In combination with other symptoms and dyspnoea they were combined with "'No' answers Degree of air pollution The area of Prague has sulfur dioxide pollution averaging about 0.2 mg/i3 and 0.5 mg/i3 of pollu- tant particles measured by the membrane filter method (V. Spurny, Prague). Demographic information Prague 2, with a total population of 104 000, is a district in the centre of Prague. The population density is about 25 000 per km2. Manual workers (designated socio-economic category I) were pre- dominant (about 60 %), and the remainder were white-collar workers (socio-economic category II). These categories conform to an official classification used in Czechoslovakia. In socio-economic cate- gory I, the following occupation groups are included: factory workers, workers in building industries, traffic and post office workers, employees in different services and trades, and unskilled workers. Socio- economic category II includes technical employees, administrative employees (both with college educa- tion), and males with higher school qualification (e.g., doctors, scientists, teachers, artists). 1 Instructions for the use of the questionary on respiratory symptoms. A pamphlet prepared for and approved by the Medical Research Council's Committee on the Aetiology of Chronic Bronchitis. EPIDEMIOLOGY OF CHRONIC BRONCHITIS IN PRAGUE TABLE I DISTRIBUTION OF THE SURVEY POPULATION BY THE PLACE OF BIRTH Birthplace No %(1 December 1930 Census) No. Prague 2 648 31.9 Other towns in Czechoslovakia with populations: 0-24 999 4 660 56.2 25 000-49 999 140 1.7 50 000-99 999 21 1.3 >100 000 186 2.2 Born abroad 403 4.9 Unknown 234 2.8 Total 8 292 100.0 The distribution of the studied population is shown according to the place of birth in Table 1, and by socio-economic category, occupation and smoking historv in Table 2. RESULTS The prevalence of respiratory symptoms is shown in Table 3, wvhere the over-all prevalence of chronic bronchitis (defined as persistent cough and persistent phlegm lasting more than 2 years) in the age-group 52-67 years is shown to be 31.8 %. Apart from cough grade 1, phlegm grade 1 and persistent cough and phlegm lasting less than 2 years, the other symptoms show a moderate gradient by age. The gradient is highest for persistent cough and phlegm and dysp- noea grade 5. This is in contrast to the findings of some other authors (College of General Practitioners, 1961). The figure for dyspnoea grade 5 without other symptoms includes all persons with this complaint (shortness of breath on washing and dressing) regardless of the cause of dyspnoea, e.g., including those with heart disease. Of the subjects answering positively to this question, 9.1 % indicated that a diagnosis of myocardial infarction has been reported to them. The prevalence of myocardial infarction was much higher in those who also had persistent cough and phlegm than among those who did not (Table 4). Respiratory symptoms were found to occur more frequently among cigarette smokers, as is seen in Fig. 1, where data are given on morning, day and persistent phlegm by smoking habits. The effects of age and smoking on the prevalence of symptoms are demonstrated in Fig. 2 and 3. Fig. 2 shows the relation between smoking and persistent phlegm in the different age-groups. Within each smoking category the increase in prevalence of cough with TABLE 2 PERCENTAGE DISTRIBUTION BY SMOKING HABITS WITHIN SOCIO-ECONOMIC CATEGORY AND OCCUPATIONAL GROUPa Socio-economic category I Socio-economic category 11 Smoking Totalhabits Group Group Group Group Group Group Group Group Group Total 0 1 - 2 3 4 5 6 7 8 Cigarette 45.0 50.8 58.6 51.3 47.8 52.3 44.2 42.4 36.7 46.0 Pipe/cigar 2.9 3.8 5.8 2.8 2.9 4.8 2.4 2.5 4.4 3.2 Ex-smokers 28.8 22.5 17.3 23.7 23.5 20.8 25.3 28.4 26.6 24.9 Never smoked 23.3 22.9 18.3 22.2 25.8 22.2 28.1 26.8 32.3 25.9 Total no. of per- sonspergroup 378 866 191 675 2195 442 787 1625 1000 8159 a Group 0 - unknown Group 1 - factory workers Group 2 - building Industry employees Group 3 - traffic and post office employees Group 4 - craftsmen and tradesmen Group 5 - unskilled workers Group 6- technical employees with college education Group 7 - administrative employees with college education Group 8 - university educated persons (scientists, doctors, artists). 713 F. BOUDIK AND OTHERS TABLE 3 PREVALENCE OF RESPIRATORY SYMPTOMS IN PRAGUE Age-group (no. per age-group in parentheses) Symptom a, 52-54 55-59 60-64 65-67 Total (1815) (3064) (2271) (1142) (8292) Morning cough in winter 52.0 55.4 55.7 55.1 54.7 Day cough in winter 32.9 37.6 37.5 36.9 36.4 Cough in summer 29.7 32.4 32.8 33.0 32.0 Persistent cough 37.8 42.9 42.9 54.2 41.8 Morning phlegm in winter 48.7 53.2 53.9 54.8 52.6 Day phlegm in winter 31.0 35.8 38.6 39.0 36.0 Phlegm In summer 28.4 31.8 32.1 32.4 31.2 Persistent phlegm 35.3 40.4 39.8 40.6 39.2 One exacerbation 2.9 3.4 3.3 3.8 3.3 Two or more exacerbations 22.3 22.9 22.0 24.0 22.7 Dyspnoea grade 5 3.1 6.9 8.6 10.3 7.0 Wheezy on lying down 24.6 28.1 28.1 28.6 27.4 History of pneumonia 27.7 30.2 30.5 31.8 30.0 History of lung tuberculosis 7.1 8.0 7.6 9.4 7.9 Cough grade 1 10.8 10.3 11.5 10.7 10.8 Phlegm grade 1 10.2 10.7 9.2 9.1 10.0 Cough grade 2 26.5 32.1 30.8 31.1 30.4 Phlegm grade 2 24.5 28.9 30.1 31.4 28.6 Persistent cough and phlegm lasting more than 2 years 28.8 32.4 32.3 33.8 31.8 Persistent cough and phlegm lasting less than 2 years 2.4 3.4 2.7 2.7 2.9 Persistent cough and phlegm lasting more than 2 years and exacerbation 15.0 16.5 16.0 17.9 16.2 Persistent cough and phlegm and dyspnoea grade 5 1.8 4.5 5.2 6.5 4.4 a Cough grade 1- morning cough or day/night cough on most days for as much as 3 months in the winter. Cough grade 2-morning cough and day/night cough on most days for as much as 3 months in the winter. Phlegm grade 1-morning phlegm or day/night phlegm on most days for as much as 3 months In the winter. Phlegm grade2-morning phlegm and day/night phlegm on most days foras much as3 months in the winter. TABLE 4 PREVALENCE OF MYOCARDIAL INFARCTION (Ml) BY AGE AND RESPIRATORY SYMPTOMS Total Age-group (years) Respiratory symptoms 52-54 55-59 60-64 65-67 Respiratory symptoms No. of o wth obser- Ml x | ober %with obserf % with ob%serAwith o ser A withvations obe- Ml obe- Ml -er M I obe- MI vations vations vations vations Persistent cough + persistent phlegm 2 507 8.5 498 3.8 938 7.5 702 11.5 369 12.2 lasting more than 2 years 6.42 No persistent cough or 5 559 6.8 1 276 3.1 2 028 6.3 1 511 9.3 744 9.3 persistent phlegm 0.63 Persistent cough + persistent phlegm 226 8.0 41 2.4 98 5.1 58 13.8 29 13.8 lasting less than 2 years 714 EPIDEMIOLOGY OF CHRONIC BRONCHITIS IN PRAGUE FIG. I FREQUENCY OF RESPIRATORY SYMPTOMS IN SMOKERS AND NON-SMOKERS 76 CIGARETTE SMOKERS EX-SMOKERS PIPE/CIGAR SMOKERS NEVER SMOKED MORNING PHLEGM DAY PERSISTENT PHLEGM PHLEGM FIG. 2 FREQUENCY OF PERSISTENT COUGH IN CIGARETTE SMOKERS AND NON-SMOKERS Inn AGE-GROI (YEARS) 90 S2-54- 80 n55-S9.Q.- -60-6 70~~ 6*5 -67- 1-14 15-24 CIGARETTES FER DAY FIG. 3 FREQUENCY OF PERSISTENT COUGH AND PERSISTENT PHLEGM AND DYSPNOEA GRADE 5 IN NON-SMOKERS AND CIGARETTE SMOKERS 14 AJE.WU (YEARS) 12 -0,52-54 10-- --_ 8- 65-G7 CL - 1 * - CIGARETMS PER DAY IN THE PAST age is small, but is apparently more marked among those persons who smoke moderately. A similar pattern is shown in Fig. 3 where the prevalence of persistent cough and phlegm with dyspnoea grade 5 is compared with the previous consumption of cigarettes. Here, the increase in prevalence with age is almost as great as the increase relating to present cigarette consumption. In Table 5, the values obtained for the 2 simple ventilatory tests, FEVL.0 and PEFR, are given in relation to cigarette-smoking history. For both tests the values decrease, in general, as the quan- tity of cigarettes smoked increases although the differences are relatively slight. Similar observations have been made by other workers (Goldsmith et al., 1962; Holland & Reid, 1965). In view of the possible association of symptom prevalence with migration from small towns to urban areas, an attempt was made to evaluate the data from this point of view. For the main respira- tory symptoms, no important differences were found between the two chief groups of subjects, those born in Prague and those born in rural areas (Table 6). The association between socio-economic category and symptom prevalence is demonstrated in Fig. 4. For persistent cough and persistent phlegm a pre- valence of 37.5% was found in socio-economic category I, and a prevalence of 27% was found in socio-economic category II. The influence ofchanges of occupation upon symptom prevalence is also shown in this figure. Only those changes in which the new occupation lasted more than 4 years are considered. Marital status was also associated with the prevalence of chronic bronchitis. The prevalence of persistent cough and persistent phlegm was higher in the divorced or widowed males than in the married ones (Table 7 and Fig. 5, X2 = 26.06). When this symptom was combined with dyspnoea grade 5 no difference was found (4.57% and 4.33 %, Table 7). The relationship between chronic bronchitis and body-weight, well known in clinical practice, is documented in Fig. 6, showing body-weight distribu- tion, and in Fig. 7, showing the prevalence of persistent cough and persistent phlegm in relation to body-weight. There was a significantly higher prevalence a botht extremes of body-weight (50 kg- 59 kg and 101 kg and over) than for the 70 kg- 74 kg category. The statistically significant difference remains when evaluating persistent cough and phlegm in combination with dyspnoea grade 5. No such association was found between symptom 715 F. BOUDIK AND OTHERS TABLE 5 FORCED EXPIRATORY VOLUME IN 1 SECOND (FEVi.o) AND PEAK EXPIRATORY FLOW RATE (PEFR) BY AGE AND CIGARETTES SMOKED a Age-group (years) Number of cigarettes snmoked per day No. 52-54 Mean Standarc ,value deviatior 55-59 60-64 No. Mean Standard N Mean Standard1i tvalue deviation, value deviation 65-67 No. Mean |Standardvalue deviation FEVi.o (in ml) None 292 1-14 181 15-24 229 25-34 80 ;-35 50 Pipe/cigar smokers 29 PEFR (in litres/min) None 1-14 15-24 25-34 >35 Pipe/cigar smokers 391 233 296 110 56 30 " Pulmonary function values are adjusted to ATPS. TABLE 6 PERCENTAGE PREVALENCE OF SYMPTOMS BY PLACE OF BIRTH Birthplace (1 December 1930 census) Prague Other towns in Czechoslovakia with population: 0-24 999 25 000-49 999 50 000-99 999 100 000 Born abroad Unknown Persistent Persistent cough phlegm 42.6 41.5 31.9 42.9 40.0 40.7 45.8 38.8 39.3 30.7 42.9 41.3 40.5 43.9 FIG. 4 FREQUENCY OF PERSISTENT COUGH AND PERSISTENT PHLEGM BY SOCIO-ECONOMIC CATEGORY SOCIO-ECONOMIC CATEGORY 7 . SOCIO-ECONOMIC CATEGORY 11 _ 2 7 . 0 CATEGORY I CHANGED TO CATEGORY 11 31.8 CATEGORY Jl CHANGED TO CATEGORY CHANGE OF. OCCUPATIONAL GROUP 23.1 WITHIN CATEGORY 11 CHANGE OF OCCUPATIONAL GROUP 4. WIlTHIN CATEGORY P1020 3 40 PERCENTAGE 2 642 2 629 2 544 2 465 2 408 2 683 555 573 601 536 586 468 482 287 423 179 95 63 2 510 2 436 2 288 2 256 2 202 2 300 517 555 521 646 578 539 268 225 275 147 85 36 2 320 2 162 2 058 2 073 1 966 2 156 537 599 590 560 578 562 106 124 143 70 37 16 2 153 1 974 1 985 1 853 1 757 1 956 543 677 621 525 609 627 483 475 459 445 455 471 87 97 96 110 109 91 610 363 549 240 126 71 474 439 430 424 408 419 86 94 97 102 108 87 377 319 392 184 93 49 448 413 404 393 384 419 87 100 101 100 106 110 162 193 203 87 51 27 431 96 393 104 389 113 379 105 331 106 375 114 716 EPIDEMIOLOGY OF CHRONIC BRONCHITIS IN PRAGUE 717 TABLE 7. FREQUENCY OF PERSISTENT COUGH AND PERSISTENT PHLEGM WITHOUT AND WITH DYSPNOEA GRADE 5 BY MARITAL STATUS AND AGE I Age-group (years) 52-54 No. 55-59 60-64 % No. No. 65-67 No. Persistent cough and phlegm 71 35.21 133 33.83 142 33.80 69 30.43 1 417 30.20 2 365 34.46 1 644 33.21 801 32.71 134 36.57 223 39.46 219 42.01 124 55.65 193 23.83 343 32.07 266 32.33 148 34.46 Persistent cough and phlegm and dyspnoea grade 5 71 1.41 133 3.76 142 7.04 69 5.80 1 417 1.98 2365 4.44 1 644 4.93 801 6.99 134 0.75 223 4.48 219 6.39 124 5.65 193 1.55 343 4.96 266 4.51 148 4.73 FIG. 5 FREQUENCY OF PERSISTENT COUGH AND PERSISTENT PHLEGM BY MARITAL STATUS 60 - - SO MARRIED * WIDOWED FIG. 6 DISTRIBUTION OF SUBJECTS BY BODY-WEIGHT 14C 'OC looc OCCZ 6X 40C 20C AGE-GROUP (YEARS) 549 75. 6 85- 97B5- M- 59 64 69 74 79 84 89 99 BOD Y-WE IGHT t FIG. 7 FREQUENCY OF PERSISTENT COUGH AND PERSISTENT PHLEGM BY BODY-WEIGHT 60 4 _X ns|1 r | l l 1 r n 1El r n -I H r --| r r l X r| _ _ _ . . | _ | _ _ - I L L _ l L L s50 50- 60- 5 70- 75- 0 85- 90 IOOf'59 64 69 74 79. 84 89 99 BOOY-WEtGHT (kg) Marital status Single Married Widowed or divorced Unknown Single Married Widowed or divorced Unknown so 40 w -a ui 30 w cc ta Q. 20 10 O' F. BOUDIK AND OTHERS FIG. 8 DISTRIBUTION OF SUBJECTS BY HEIGHT prevalence and height as shown in the Fig. 8 and 9. We do not and shall not assume that the relation- ship of cough and sputum to myocardial infarction, social class, marital status, body-weight and height are all independently acting with respect to other variables, especially with the cigarette-smoking practices. We are aware that in other studies, for example, divorced persons tend to be heavier smokers than married persons and that smokers FIG. 9 FREQUENCY OF PERSISTENT COUGH AND PERSISTENT PHLEGM BY HEIGHT 50 " -_-__ _ t | _ _ _ _ l * _ _ s _ _ _ l - - - -| _ _ _ _L _ L L I _ _ _ _ r _ r r | _ _ _ _l _ __ _ I L _ _ L L____ _ F _ _ l I _ _ _ _ _ _ _ _ . . _ __ _ _ _ _ S154 155- 1iS0- 165- 170- 175- 1a0- 1185- I59 164 174 179 I8 189 HEIGHT (e") 19047 tend to have a higher prevalence of respiratory and coronary disease. Quite possibly the relationship of marital status to cough and sputum may be dependent on cigarette-smoking practices. Appro- priate methods for standardization of the smoking practices among these groups are now being applied and the results will be reported when the work is completed. The prevalence of persistent cough and persistent phlegm lasting more than 2 years in the group of TABLE 8 RESPIRATORY SYMPTOM PREVALENCE BY AGE AND FAMILY HISTORY OF CHRONIC BRONCHITIS No chronic bronchitis In family history Chronic bronchitis in family history Age-group x2 a(years) With Without % with With Without % with symptom symptom symptom symptom symptom symptom Persistent cough and phlegm lasting more than 2 years 52-54 373 968 27.82 75 113 39.89 11.61 55-59 675 1 537 30.52 168 191 46.80 44.51 60-64 488 1 137 30.03 130 158 45.14 25.53 65-67 244 536 31.28 76 86 46.91 14.61 Total or mean 1780 [ 4178 [ 29.88 449 548 45.04 90.13 Persistent cough and phlegm and dyspnoea grade 5 a 3 degrees of freedom for the total and 1 degree of freedom for each age-group. of T 718 40 w ..a 30I.-. W ci n 20CL 10 = EPIDEMIOLOGY OF CHRONIC BRONCHITIS IN PRAGUE non-respondents was found to be 34.7%, slightly higher than in respondents. In our previous smaller epidemiological survey, as well as in our clinical practice, we had sometimes found familial occurrence of chronic bronchitis. Similar findings have been reported by other authors (Hole & Wasserman, 1965; Hurst, 1959; Larson & Barman, 1965; Oswald, Harold & Martin, 1953). In processing the data of the Prague survey we tried to ascertain if such a trend could be confirmed from the survey data. These results are given in Table 8 and show a significantly higher prevalence of per- sistent cough and persistent phlegm with or without dyspnoea grade 5 in subjects whose families (parents and/or siblings) had chronic bronchitis. The data on the occurrence of chronic bronchitis in respondents' families are based on their own answers only. Those answers are being checked in a subsequent follow-up study on 500 bronchitics. These data appear likely to underestimate the true impact of family history. For instance, in several cases the respondent's father was killed in the First World War at a comparatively young age, and no evidence of family history of the disease was given. For the detection of chronic bronchitis in the family history a Czech term was used denoting the com- bination of chronic cough and/or phlegm with dyspnoea of pulmonary origin. Such data are biased by many factors of which it is hard to ascertain the importance. Of course, the data cannot be regarded as precise evidence of the familial occurrence of chronic bronchitis or of the influence of genetic factors. Neither clinical nor customary epidemiological methods provide the means of getting more accurate information on this problem. One can agree with McKusick, Goodman & Danks (1963) that " age-dependence of the disorder imposes difficulties. By the time the proband comes to attention the parents are likely to be dead, the sibs are widely scattered, and off- spring are not yet old enough to show evidence of disease ". The study of twins, as shown by the Karolinska group in Sweden, provides a better way of studying this question, and their results do appear to indicate a genetic factor in the disease (Cederlof et al., 1966). Should the validity of the family history data be confirmed, the effect of similarity in other factors could account for such an association, for example, between brothers with similar smoking habits (D. D. Reid, personal communication). The figures of preva- lence for persistent cough and persistent phlegm (with and without dyspnoea grade 5) in subjects who have never smoked are given in Table 9 and show that the prevalence is significantly higher in subjects with positive family history for chronic bronchitis. A possible role of pulmonary tuberculosis in the development of chronic obstructive pulmonary disease is discussed by Gaensler & Lindgren (1959). These authors believe that the chronic bronchitis of patients with pulmonary tuberculosis could be an important etiological factor in the development of obstructive emphysema. Some of our clinical findings seem to support the possibility of such an association. Therefore, data on the occurrence of healed pulmonary tuberculosis in the respondents' history and in their family history were evaluated according to the frequency of respiratory symptoms. The results of this analysis are given in Table 10. A significantly higher prevalence (in the statistical sense) of persistent cough and persistent phlegm (with and without dyspnoea grade 5) was found in persons having pulmonary tuberculosis in their personal or family history. As mentioned above, all persons with active pulmonary tuberculosis were excluded from this survey. TABLE 9 RESPIRATORY SYMPTOM PREVALENCE BY FAMILY HISTORY OF CHRONIC BRONCHITIS: NON-SMOKERS a No chronic bronchitis in family history Chronic bronchitis in family history Symptom With Without % with With Without % with X2b symptom symptom symptom symptom symptom symptom Persistent cough and phlegm 386 1 761 17.97 97 199 32.76 35.88 Persistent cough and phlegm anddyspnoeagrade5 40 ] 2107 1.85 19 277 6.42 23.16 a Age range: 52-67 years. b One degree of freedom. 719 F. BOUDIK AND OTHERS TABLE 10 RESPIRATORY SYMPTOM PREVALENCE BY PERSONAL AND FAMILY HISTORY OF PULMONARY TUBERCULOSIS No tuberculosis in personal history Tuberculosis in personal history Symptom With Without % with With Without % with x symptom symptom. symptom symptom symptom symptom Persistent cough and phlegm 2 340 4 881 32.41 272 347 43.94 42.23 Persistent cough and phlegm and dyspnoea grade 5 272 6 949 3.77 61 558 9.85 51.95 No tuberculosis in family history Tuberculosis in family history With Without % with With Without % with symptom symptom symptom symptom symptom symptom Persistent cough and phlegm 2 110 4 469 32.07 564 841 40.14 33.85 Persistent cough and phlegm and dyspnoea grade 5 272 6307 4.13 75 1 330 5.34 4.04 a Age range: 52-67 years. TABLE 11 PERCENTAGE PREVALENCE OF RESPIRATORY SYMPTOMS IN DIFFERENT TOWNS Symptom 1 Prague a London b Los Angeles c San FranciscoC Morning cough in winter 54.1 51.8 31.4 24.5 Day cough in winter 35.8 38.7 24.3 18.4 Morning cough in summer 20.8 25.7 22.4 31.4- Day cough in summer 10.7 20.0 14.3 Persistent cough 41.0 44.5 32.9 18.4 Morning phlegm in winter 51.6 49.6 37.1 24.5 Day phlegm in winter 34.1 41.6 30.0 20.4 Morning phlegm in summer 12.6 30.0 18.4 30.6 Day phlegm In summer 11.9 28.6 16.3 Persistent phlegm 38.5 44.5 40.0 24.5 Cough grade 1 10.5 17.5 14.3 4.1 Phlegm grade 1 10.5 19.0 12.9 10.2 Cough grade 2 30.0 27.0 18.6 14.3 Phlegm grade 2 27.2 25.5 27.1 14.3 a Males aged 52-59 years, general population, N=4 879. b Males aged 50-59 years, mail-van drivers and vehicle-maintenance men in central London, N=137 (Holland & Reid, 1965). c Males aged 50-59 years, outside plant telephone workers, N=70 for Los Angeles and N=49 or San Francisco (Deane, Goldsmith & Tuma, 1965). 720 EPIDEMIOLOGY OF CHRONIC BRONCHITIS IN PRAGUE All these data are based on the respondents' information only and are biased in the same way as the data on the presence of chronic bronchitis in family history. The data in the personal history are likely to be less biased by lack of information since there has been a high level of tuberculosis care in Czechoslovakia in the last 30-40 years. It was possible to validate some of the findings given above using the radiographic findings. Sub- jects with post-primary tuberculosis of minimal degree (defined as one large focus or several small foci no more extensive than one sub-segment) were evaluated according to the frequency of persistent cough and persistent phlegm. The prevalence of this symptom combination was found to be significantly higher in subjects with these radio- graphic findings than in subjects without them (X2 = 12.66, significant at the 1% level). When combined with dyspnoea grade 5, no statistical significance was found, and when combined with dyspnoea grade 3, the statistical significance was at the 10% level. The degree of significance was lower because subjects with a positive personal his- tory of pulmonary tuberculosis sometimes refused X-ray examination in this survey (they are frequently examined by X-ray in the health centres even if they suffer from tuberculosis of minimal degree). There is no universal definition of chronic bron- chitis and therefore a number of criteria have been used in different surveys, even in those using the standard questionnaire, and it is difficult to make comparisons. That is why it is more advantageous to compare the prevalence of responses to individual questions obtained by means of the standard ques- tionnaire. To illustrate the symptom prevalence among the male population in towns from different countries, some data obtained by other authors are given in Table 11. ACKNOWLEDGEMENTS We are grateful to Professor Herles, M.D., Director of the 2nd Medical Clinic in Prague for his advice and encouragement during this survey, and to Miss Margaret Deane, of the California State Department of Public Health, for assistance in statistical planning and analys,is of the data. RESUME EPIDtMIOLOGIE DE LA BRONCHITE CHRONIQUE A PRAGUE De novembre 1964 a juillet 1965, une enquete sur la prevalence de la bronchite chronique et sur l'importance etiologique de divers facteurs a et menee parmi la popu- lation masculine ag6e de 52 a 67 ans d'un district de Prague (Tch6coslovaquie). Pour ces investigations por- tant sur un total de 8292 personnes, on a utilise un questionnaire relatif aux sympt6mes respiratoires sub- jectifs, pratiqu6 un examen radiographique des poumons chez la plupart des sujets et soumis la moiti6 environ d'entre eux a des epreuves de la fonction ventilatoire. On a enregistre une prevalence globale de la bronchite chronique (definie comme atoux et expectoration chro- niques, persistant depuis plus de deux ans #) de 31,8 %. La fr6quence des symptomes respiratoires etait plus elevee chez les fumeurs de cigarettes, pour lesquels les resultats des tests fonctionnels etaient egalement moins bons, meme en l'absence de tout symptome. La bron- chite chronique etait plus repandue chez les travailleurs manuels, chez les hommes veufs ou divorces et chez ceux dont le poids corporel s'ecartait, en plus ou en moins. de la normale. Le syndrome <(toux et expectoration chroniques, axec ou sans dyspn6ee a ete observe avec une frequence nettement plus elevee chez les sujets qui comptaient. parmi leurs parents proches, une personne atteinte de bronchite chronique, ce qui plaide en faveur de l'influence possible d'un facteur hereditaire. De meme, la prevalence de la bronchite chronique etait plus forte parmi les sujets dont les ant6cedents personnels ou familiaux compor- taient la mention de tuberculose pulmonaire et chez ceux qui, a l'examen radiographique, presentaient des lesions tuberculeuses cicatricielles peu etendues. 721 722 F. BOUDIK AND OTHERS REFERENCES Brinkman, G. L. & Coates, E. O., Jr (1962) Amer. Rev. resp. Dis., 86, 47-54 Cederlof, R., Friberg, L., Jonsson, E. & Kaij, L. (1966) Arch. environm. HIth, 13, 726-736 College of General Practitioners (1961) Brit. med. J., 2, 973-979 Colley, J. R. T. & Holland, W. W. (1967) Arch. environm. Hlth, 14,157-160 Deane, M., Goldsmith, J. R. & Tuma, D. (1965) Arch. environm. Hith, 10, 323-331 Enterline, P. E. (1967) Arch. environm. Hlth, 14, 189-198 Ferris, B., Jr & Anderson, D. 0. (1962) Amer. Rev. resp. Dis., 86, 165-177 Fletcher, C. M., Elmes, F. C., Fairbarn, A. S. & Wood, C. M. (1959) Brit. med. J., 2, 257-266 Gaensler, E. A. & Lindgren, I. (1959) Amer. Rev. resp. Dis., 80,185-193 Goldsmith, J. R., Hechter, H. H., Perkins, N. M. & Borhani, N. 0. (1962) Amer. Rev. resp. Dis., 86, 867-874 Higgins, I. T. T., Oldham, P. D., Cochrane, A. L. & Gilson, J. C. (1956) Brit. med. J., 2, 904-909 Hole, B. V. & Wasserman, K. (1965) Ann. intern. Med., 63, 1009-1017 Holland, W. W. & Reid, D. D. (1965) Lancet, 1, 445-448 Hurst, A. (1959) Amer. Rev. resp. Dis., 80, 179-180 Kourilsky, R., Brille, D., Hutte, J., Carton, J. & Hinglais, J. C. (1966) Enquete sur l'etiologie et la prophylaxie de la bronchite chronique et l'emphysime pulmonaire, Paris. Larson, R. K. & Barman, M. L. (1965) Ann. intern. Med., 63, 1001-1008 McKusick, V. A., Goodman, R. M., & Danks, D. M. (1963) Arch. environm. HiIth, 6, 25-36 Medical Research Council Committee on the Aetiology of Chronic Bronchitis (1960) Brit. med. J., 2, 1665 Olsen, H. C. & Gilson, J. C. (1960) Brit. med. J., 1, 450-456 Oswald, N. C., Harold, J. T. & Martin, W. J. (1953) Lancet, 2, 639-643 Stanek, V., Fodor, J., Hejl, Z., Charvdt, P., Santr-iek, M., Widimsk9, J. & Zajic, F. (1966) eas. lIk. des., 105, 689-695 WHO Expert Committee on Chronic Cor Pulmonale (1961) Wld Hlth Org. techn. Rep. Ser., No. 213 Wright, B. M. & McKerrow, C. B. (1959) Brit. med. J., 2, 1041-1047