|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Department of Public Health (Drs. Huovinen and Kaprio), University of Helsinki, Helsinki, Finland; Department of Medicine (Dr. Laitinen), University of Helsinki Central Hospital, Helsinki, Finland; and Department of Public Health (Dr. Koskenvuo), University of Turku, Turku, Finland.
| Abstract |
|---|
|
|
|---|
Design: Prospective cohort study.
Participants: A population of 11,540 Finnish adult men and women, initially 18 to 45 years of age, who returned a health questionnaire in 1975, 1981, and 1990 as part of the Finnish Twin Cohort study.
Methods: Age-standardized prevalences and cumulative incidences among individuals were calculated for asthma, hay fever, and chronic bronchitis. The incidence of asthma among subjects with and without hay fever or chronic bronchitis was analyzed in the entire cohort as well as in twin pairs discordant for incident asthma.
Results: The prevalence of asthma increased slightly from 1975 (2.0% in men and 2.2% in women) to 1990 (2.9% in men and 3.1% in women). The prevalence of hay fever showed a larger increase in men and women (from 6.8% and 9.8% to 11.8% and 15.3%, respectively). Compared with figures for 1976 to 1981, no significant increase in asthma incidence occurred from 1982 to 1990, whereas the incidence of hay fever was lower during the latter period among men (incidence rate ratio, 0.7; 95% confidence interval, 0.6 to 0.9) as was the incidence of chronic bronchitis among women (incidence rate ratio, 0.7; 95% confidence interval, 0.6 to 0.9). Hay fever and chronic bronchitis were usually diagnosed before asthma. Both diseases increased the risk of asthma significantly on the basis of analyses of all individuals and of discordant twin pairs.
Conclusions: The pattern of increase in asthma and hay fever prevalence with time was similar, and hay fever was a strong predictor of asthma. These diseases showed no significant increase in incidence.
Key Words: asthma chronic bronchitis cohort study hay fever incidence prevalence
| Introduction |
|---|
|
|
|---|
Prevalence refers to the cases of a disease that exist at a specified time, and incidence to the number of new cases that develop during some specified time interval. Thus, prevalence is a function of incidence and duration of the disease. Longitudinal epidemiologic studies show that the natural history of asthma includes phases of prolonged remission and even of recovery,10 and that asthma reduces survival in adults.11 ,12 Notable differences between prevalence and incidence studies of asthma are therefore to be expected.
Incidence studies on asthma are rare compared with prevalence studies. Yunginger et al13 noted an increase in asthma incidence among children from 1964 to 1983, whereas rates among adults remained constant. The incidence of asthma among young Israeli adults was also stable during 1987 to 1989.14 However, a Finnish study found a rising trend of medication-requiring asthma incidence among young adults for the period 1986 to 1993, although there was only a slight increase among women aged 50 to 64 years, and even a decrease in men of the same age.15
Hay fever prevalences have also risen in recent decades,6 ,7 but no such trend in hay fever incidence was found by Ross and Fleming16 during the last 12 years. Because diagnostic variation is a common problem when studying chronic bronchitis, trends in its prevalence and incidence are not easily clarified.17 ,18
This prospective cohort study examined the incidence of asthma, hay fever, and chronic bronchitis among adults during a 15-year follow-up period, as well as the prevalence of asthma and hay fever. The co-occurrence of hay fever and chronic bronchitis among asthmatic subjects and their influence on asthma incidence were also studied.
| Materials and Methods |
|---|
|
|
|---|
The incidence and prevalence analyses were based on different sets of subjects selected from the Finnish Twin Cohort. The study population for incidence analysis consisted of all twins who answered all three questionnaires, whereas prevalence analyses were based on all subjects responding to the questionnaire of the year for which prevalence was calculated, ie, 1975, 1981, or 1990. The incidence analyses respondents were born from 1930 to 1957 (N = 11,540), and 45% were men. Both twins of 3,753 pairs responded in addition to 4,034 twins whose twin sibling did not respond. Prevalences, by contrast, were calculated among subjects aged 30 to 59 years at the time of each questionnaire, except for the 1990 study, when the range was 33 to 59 years. The numbers of subjects were 14,468 in 1975, 15,317 in 1981, and 12,219 in 1990. The data for calculating 1975 prevalences included some singletons (N = 2,085, 14.4% of all subjects), whereas 1981 and 1990 prevalences were based on twins only. All prevalences and incidences were calculated from individuals.
To investigate the co-occurrence of hay fever and chronic bronchitis with asthma, those with missing data on asthma, hay fever, or chronic bronchitis in any questionnaire (n = 862) were excluded. Subjects reporting asthma in 1981 or 1990 were then defined as "new asthma cases," ie, asthma diagnosed from 1976 to 1990 (n = 219).
Statistical Analysis
The prevalences of asthma and hay fever were calculated in 1975,
1981, and 1990 after excluding those with missing data on the
respective disease in each questionnaire. The European standard
population22
was used to calculate age-standardized
prevalences by 5-year age groups separately for men and women.
Confidence intervals (CIs) were computed using an appropriate software
(Confidence Interval Analysis; M. J. Gardner and British Medical
Journal; London, UK) program.23
Separate incidences were computed among those reporting no asthma, hay fever, or chronic bronchitis in the 1975 questionnaire for the whole follow-up time (1976 to 1990) and for the first 6 years of follow-up (1976 to 1981). Cumulative incidences were calculated separately for men and women in age groups based on birth decade (1930 to 1939, 1940 to 1949, 1950 to 1957). Age-trend analyses were based on these three birth-decade groups and performed with the Rate Analysis program24 derived from the formulas given by Miettinen.25 Similarly incidence rate ratios (IRRs) by sex and their CIs were calculated by the Rate Analysis program.24 ,25 Basic statistics were derived by a software package (SAS; SAS Institute; Cary, NC).
To compare incidences of asthma, hay fever, and chronic bronchitis during the first 6 and last 9 years of the follow-up time, incidences were calculated separately for two periods (1976 to 1981 and 1982 to 1990). The initial ages of subjects (ie, in 1975 and 1981) were used, and only those aged 25 to 44 years at the beginning of each follow-up were included in these analyses to make the age distributions of the two periods as comparable as possible. Person-years were calculated by multiplying the number of subjects at risk by the number of years (6 or 9) of follow-up. The European standard population22 was used to calculate age-standardized incidences per 10,000 person-years by 5-year age groups during 1976 to 1981 and 1982 to 1990, and CIs were computed by the Confidence Interval Analysis program.23 IRRs and their CIs were computed by stratifying 5-year age groups using the Rate Analysis program.24 ,25
To study the effects of hay fever and chronic bronchitis on the risk of developing asthma, the cumulative incidence of asthma during 1976 to 1990 was calculated among subjects reporting hay fever or chronic bronchitis in 1975, and among those not reporting it. IRRs were computed by stratifying birth decade using the Rate Analysis program.24 ,25
Twin pairs discordant for incident asthma in the 1981 or 1990 questionnaire, or in both (N = 161), were identified to examine whether the risk of asthma in relation to a history of hay fever and chronic bronchitis (in 1975) differed from the risk in their age-matched twins (either dizygotic or monozygotic) without these diseases. These co-twins represent persons who have generally shared the same childhood environment as the diseased (asthma) subject, and have part or all of their genes in common by descent. The estimate of relative risk was calculated from the ratio of number of pairs in which the diseased twin had had the condition (hay fever or chronic bronchitis) and the healthy twin had not, to the number of pairs in which the opposite had occurred; CIs were also computed.23
| Results |
|---|
|
|
|---|
The age-standardized prevalence of asthma in 1975 was 2.0% (95% CI, 1.6 to 2.5%) among men and 2.2% (95% CI, 1.8 to 2.6%) among women, and there were no significant prevalence rate differences between the singletons and twins. The 1981 prevalence was quite similar, being 2.1% (95% CI, 1.7 to 2.5%) and 2.3% (95% CI, 1.9 to 2.7%). In 1990, there was some increase in asthma prevalence both among men 2.9% (95% CI, 2.4 to 3.4%) and women 3.1% (95% CI, 2.6 to 3.5%). The prevalence of hay fever in 1975 was 6.8% (95% CI, 6.1 to 7.5%) for men and 9.4% (95% CI, 8.7 to 10.2%) for women. In 1981 it was 7.9% (95% CI, 7.3 to 8.6%) and 10.6% (95% CI, 9.8 to 11.4%), and in 1990, 11.8% (95% CI, 10.8 to 12.8%) and 15.3% (95% CI, 14.3 to 16.4%), respectively. Age-specific prevalences for asthma and hay fever are shown in Figure 1 by sex and time of assessment.
|
|
|
|
|
|
| Discussion |
|---|
|
|
|---|
60 years at the end of the follow-up, we
avoided most diagnostic problems common among the elderly. Subjects
reporting the study diseases in the first questionnaire in 1975 were
excluded from the incidence analyses. It is possible that some of the
diseases diagnosed in childhood were not reported in 1975 because they
were in remission or forgotten. As Grol et al26
have
recently concluded, disappearance of symptoms does not mean the disease
is cured. Thus, defining an incident case of asthma is much more
difficult than assessing prevalent cases. Nonetheless, the low rates of
diagnosed asthma and hay fever in the 1960s and earlier in
Finland5
strongly suggest that the incidences of asthma,
hay fever, and chronic bronchitis in this study are largely based on
cases diagnosed for the first time in adulthood. We used a questionnaire-based classification of physician-diagnosed asthma, hay fever, and chronic bronchitis. The advantage of using these diagnoses lay in the consistency of follow-up information provided by the identical questions in each questionnaire. Finns tend to know quite well what diseases they have when their knowledge is compared with medical records.27 The validity of questionnaire-based diagnoses of asthma is adequate according to previous studies.20 ,28 ,29 ,30 Moreover, questionnaire-based diagnoses of asthma are both sensitive and specific as predictors of asthma deaths.12 In the present study, about 40% of those who reported asthma in 1981 for the first time no longer reported it in 1990, consistent with previous study findings.20 Hay fever was reported somewhat more consistently than asthma in the present study, and a Norwegian study shows good agreement between self-reported diagnosis of hay fever and the physician's clinical examination.31 Thus, questionnaire-based information for hay fever, too, can be considered adequate. Of those reporting chronic bronchitis in 1981, half no longer reported it 9 years later, probably because definition problems are common in all epidemiologic studies of chronic bronchitis, and the occurrence of symptoms of chronic bronchitis is known to vary over time. However, cumulative incidences were based on diagnoses in one or both questionnaires to discover the total probability of having these diseases diagnosed during the follow-up period.
Asthma prevalences in 1975 were consistent with those of previous studies in Finland and Sweden.29 ,32 ,33 The present study detected only a slight rise in the prevalence of asthma during 1975 to 1990, whereas a Finnish study on conscripts showed as high as a sixfold increase in recent decades.5 The age distribution of our study group differed from that of Haahtela et al,5 because our data included no young adults < 30 years old in prevalence calculations. Thus, the greater rise in asthma prevalence of other studies is probably because of an increase among children and young adults. Moreover, our prevalence figures are based on subjects aged 30 to 59 years old at the time of each survey, and thus on different subjects from our incidence analysis.
Cumulative incidence provides an estimate of the probability, or risk, that an individual will develop a disease during a specified period of time. The probability of developing asthma during the 15-year period was approximately 2% for both men and women, but risk was dependent on age, even though all subjects were adults. Thus, on average, 1.6 new asthma diagnoses were made annually per 1,000 adults, which is consistent with the annual age- and sex-adjusted incidence for definite asthma (1.38/1,000) in the study of Yunginger et al13 and with the standardized incidence (2.10/1,000/yr) found by McWhorter et al.34 The probability of being diagnosed with hay fever during the same period was higher, with men having a risk of almost 10% and women, 13.6%, although age had no effect. This means that annually there is an average of six new hay fever diagnoses among 1,000 adult men and nine among 1,000 adult women.
The increase in asthma incidence between 1976 to 1981 and 1982 to 1990 was more obvious among men than women, although neither was statistically significant; at the same time the incidence of chronic bronchitis decreased. It is possible that these trends are partly because of changes in diagnostic criteria, especially among men, with a preference for an asthma diagnosis. Atopy is found to be a predisposing factor for chronic bronchitis,35 but allergic subjects with symptoms such as cough and phlegm are more likely to be considered as asthmatics than chronic bronchitis sufferers. One reason is that in Finland the costs of asthma medication have been better subsidized by the Social Insurance Institution of Finland throughout the whole study period, whereas medication costs for chronic bronchitis have been included in this higher subsidy category only since 1986.
The association between smoking and chronic bronchitis is well-known.36 The proportion of smoking men has been decreasing in Finland, and Finnish women are also currently smoking less than before, although a rising trend appeared among women during the 1980s.37 Thus, the decrease in incidence of chronic bronchitis cannot be explained wholly by smoking habits. In the present study, the women were more likely to have chronic bronchitis, whereas a Swedish study on the prevalence of chronic bronchitis showed a significantly higher prevalence among men than women.38 Our proportion of subjects reporting chronic bronchitis in 1981 but no longer in 1990 was higher among men than women; thus men diagnosed soon after the 1981 questionnaire might have forgotten to report it in the next questionnaire, especially if they had quit smoking during the follow-up interval.
The prevalence of allergic rhinitis has increased among young Swedish men,6 and there has been an increase in hay fever prevalence in all age groups in England and Wales,7 without the same trend in hay fever incidence.16 The present study showed no change in the incidence of hay fever during the two follow-up periods among women, but among men there was a significant decrease. However, the prevalence of hay fever rose significantly during follow-up. For incidence calculations we studied hay fever cases diagnosed among subjects aged 25 to 44 years at the beginning of the two follow-up periods.
Atopy is known to be a predisposing factor for asthma, especially in childhood and among young adults39 ; chronic bronchitis is usually associated with older asthma patients.40 We were able to specify the temporal relationship of atopy and asthma, as well as that of asthma and chronic bronchitis, in the present study. We found that diagnoses of hay fever and chronic bronchitis were almost always made before or at the same time as the diagnosis of asthma, but rarely afterwards.
The incidence of asthma was significantly higher among our subjects with hay fever or chronic bronchitis. These diseases cannot be considered purely as predisposing factors for asthma, because they may also be early symptoms or manifestations of asthma, as seen in the temporal relationship of the diagnoses. Thus, the increase in asthma risk caused by these diseases is probably overestimated, but still significant. The data on our discordant-twin pairs indicate that hay fever and chronic bronchitis increase the risk of developing asthma, even when compared with the risk in siblings.
In conclusion, there was a slight increase in the prevalence of asthma during 1975 to 1990 and also some evidence of a rise in asthma incidence. The prevalence of hay fever has also increased during recent decades, even though no rise in hay fever incidence has occurred. That hay fever was a strong predictor of new asthma cases and that hay fever was nearly always found before asthma supports the role of hay fever as a strong predisposing factor for asthma.
| Acknowledgements |
|---|
| Footnotes |
|---|
Correspondence to: Jaakko Kaprio MD, PhD, Department of Public Health, PO Box 41, University of Helsinki, Helsinki FIN 00014, Finland; e-mail: Jaakko.Kaprio@Helsinki.Fi
Abbreviations: CI = confidence interval; IRR = incidence rate ratio
Received for publication December 2, 1997. Accepted for publication November 25, 1998.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
Y.-L. Lee, T.-R. Hsiue, C.-H. Lee, H.-J. J. Su, and Y. Leon Guo Home exposures, parental atopy, and occurrence of asthma symptoms in adulthood in southern taiwan. Chest, February 1, 2006; 129(2): 300 - 308. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Brogger, T. Eagan, G.E. Eide, P. Bakke, and A. Gulsvik Bias in retrospective studies of trends in asthma incidence Eur. Respir. J., February 1, 2004; 23(2): 281 - 286. [Abstract] [Full Text] [PDF] |
||||
![]() |
T.M.L. Eagan, P.S. Bakke, G.E. Eide, and A. Gulsvik Incidence of asthma and respiratory symptoms by sex, age and smoking in a community study Eur. Respir. J., April 1, 2002; 19(4): 599 - 605. [Abstract] [Full Text] [PDF] |
||||
![]() |
M Albin, L Rylander, Z Mikoczy, L Lillienberg, A Dahlman Hoglund, J Brisman, K Toren, B Meding, K Kronholm Diab, and J Nielsen Incidence of asthma in female Swedish hairdressers Occup. Environ. Med., February 1, 2002; 59(2): 119 - 123. [Abstract] [Full Text] [PDF] |
||||
![]() |
X. BASAGANA, J. SUNYER, J.-P. ZOCK, M. KOGEVINAS, I. URRUTIA, J. A. MALDONADO, E. ALMAR, F. PAYO, and J. M. ANTO Incidence of Asthma and Its Determinants among Adults in Spain Am. J. Respir. Crit. Care Med., October 1, 2001; 164(7): 1133 - 1137. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. Huovinen, J. Kaprio, and D. Strachan Twins and asthma BMJ, March 3, 2001; 322(7285): 556 - 556. [Full Text] |
||||
![]() |
M. Lindstrom, J. Kotaniemi, E. Jonsson, and B. Lundback Smoking, Respiratory Symptoms, and Diseases* : A Comparative Study Between Northern Sweden and Northern Finland: Report From the FinEsS Study Chest, March 1, 2001; 119(3): 852 - 861. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Hakala, B. Stenius-Aarniala, and A. Sovijarvi Effects of Weight Loss on Peak Flow Variability, Airways Obstruction, and Lung Volumes in Obese Patients With Asthma Chest, November 1, 2000; 118(5): 1315 - 1321. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |