IJE Advance Access originally published online on May 13, 2005
International Journal of Epidemiology 2005 34(5):1012-1018; doi:10.1093/ije/dyi091
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Article |
Trends in primary care consultations for asthma in Switzerland, 1989–2002
1 Waldheimstrasse 51, 3012 Berne, Switzerland
2 Department of Social and Preventive Medicine, University of Berne, Finkenhubelweg 11, 3012 Bern, Switzerland
3 Swiss Federal Office of Public Health, 3003 Bern, Switzerland
4 Department of Social Medicine, University of Bristol, Canynge Hall, Whiteladies Road, Bristol, BS8 2PR, UK
* Corresponding author. Department of Social and Preventive Medicine, University of Berne, Finkenhubelweg 11, 3012 Bern, Switzerland. E-mail: low{at}ispm.unibe.ch
 |
Abstract |
|---|
 Top Abstract MethodsResultsDiscussionReferences |
|---|
Background There is widespread debate about trends in the occurrence of asthma in industrialized countries. This study was conducted to investigate time trends in consultations for asthma in primary care in Switzerland.
Methods Prospective observational study from 1989 to 2002 within the Swiss Sentinel Surveillance Network; a primary care surveillance system. We used time series analysis and non-parametric smoothing methods to investigate long-term and short-term trends in rates of asthma episodes per 1000 consultations. From 1994 to 2002 we compared rates of first episodes with all subsequent consultations for asthma.
Results Overall consultation rates for asthma per 1000 primary care consultations increased from 1989 to 1994 then stabilized and have declined since 2000. Long-term trends showed a small decline in first consultations for asthma from an average of 0.78 (95% credibility intervals (CI) 0.74–0.81) in 1999 to 0.62 (95% CI 0.55–0.69) per 1000 consultations in 2002. Subsequent consultations for asthma have been declining since at least 1994, from an average of 1.5 (95% CI 1.40–1.61) per 1000 consultations in 1994 to 0.93 (95% CI 0.82–1.04) in 2002. In addition, the ratio of subsequent to first episodes of asthma fell in all age groups.
Conclusions In Switzerland, primary care consultations for asthma, subsequent to the initial diagnosis, have been declining since 1994. This is more likely to be owing to an increase in the use of home medication than to a shift in care to hospital settings. The incidence of diagnosed asthma might also be decreasing.
Keywords Asthma, primary health care, family practice, incidence, prevalence
Accepted 5 April 2005
There is widespread debate about trends in the occurrence of asthma in industrialized countries. Although the methodological quality of many studies has been criticized,1 an increase in the prevalence of childhood asthma up to the mid-1990s has been consistently reported.2–5 Reports of stabilizing or declining prevalence in both high-income and middle-income countries,6 and a fall in the incidence of diagnosed asthma in primary care in England,7 now raise the question as to whether the 'epidemic' of asthma has ended.6 However, trends in the prevalence of risk factors for asthma reported from different countries are less clear, and there is no consensus about the reasons for such changes.8,9 A long-standing concern is that, in the absence of an accepted objectively determined definition of asthma, these trends might have been influenced by changes in diagnostic labelling practices among doctors10 or in symptom awareness by patients.11
In Switzerland primary health care is delivered principally by general practitioners, general internists, and paediatricians. The Swiss Sentinel Surveillance Network (SSSN) has been collecting information about primary care consultations with physicians from these groups since 1986. We have previously used the SSSN to study long-term epidemiological characteristics and seasonal patterns of asthma in different age groups.12,13 This study was conducted to investigate time trends in all consultations for episodic asthma, in first consultations, and the relationship between first and subsequent consultations.
|
Methods |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
The SSSN has been described in detail elsewhere.13,14 Briefly, from 1989 to 2002 the SSSN has consistently included
3% of all physicians reporting a similar proportion of all annual consultations in Switzerland (Table 1). Of the participants 70% are general practitioners, 15–20% general physicians, and 10–15% paediatricians. For this study, we included data from sentinel physicians who reported for 
75% of the time >90% of participants) to ensure consistency.
|
The definition of asthma in the SSSN, which did not change during the study period, includes exertional, allergic, and infectious causes: bronchial obstruction or hyper-responsiveness diagnosed by the presence of wheezing, dyspnoea or cough on or after physical exertion, or on contact with pollen, dust, or animal dander; and cough at night without an acute respiratory infection, or for >2 weeks after an acute respiratory infection. Age and gender were available for each reported asthma episode. From 1994 onwards, the first recorded episode of asthma was reported separately. In this paper we refer to first recorded consultations as first asthma episodes and to all remaining consultations as subsequent episodes.
Statistical analysis
We calculated 4 weekly rates of all reported asthma episodes per 100 consultations from 1989 to 2002. From 1994 to 2002 we examined the first recorded consultation and all other subsequent consultations separately. We then used standard time series analysis methods to explore trends. A time series of rates consists of a trend, a seasonal component, and random error. We estimated the seasonal component15–17 and then examined long-term and short-term trends on deseasonalized data, using 'loess', a robust non-parametric smoothing method implemented in the statistical package S-plus (6.1, Insightful, US). The algorithm used by loess is as follows: a window is placed about time t and the data points that lie inside it are weighted so that nearby points get more weight.15 Then, a robust weighted regression is used to predict the value of yt. The parameter f controls the window size and is the proportion of data (total 9 years) that are included in estimation. By increasing f we obtain smoother functions over a longer time period. We used the default S-plus value for f of two-thirds of the data to define the longer term (6 years). For shorter trends, we chose f of about one-third of the data (3 years). We estimated 95% credibility intervals (CI) for these data using a generalized additive model with loess smoother to calculate pointwise twice-standard-error curves around estimated trends.15,18 To examine changes in consultation patterns over time we calculated a yearly ratio of subsequent to first consultations for asthma from 1994 to 2002 in age groups 0–4, 5–16, 17–45, 46–65 and >65 years and examined these over time using chi-square tests for trend.
|
Results |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
Of 13 050 014 consultations recorded in the SSSN from 1989 to 2002, 24 871 were for acute asthma. Total yearly asthma episodes increased from 1.47 (95% CI 1.39–1.56) per 1000 consultations in 1989 to 2.23 (95% CI 2.14–2.33) in 1994. They fluctuated around this level until 2000 and fell to 1.65 in 2002 (95% CI 1.57–1.73) (Table 1 and Figure 1). First ever episodes accounted for an increasing proportion of all asthma episodes over time, from 28% (95% CI 26–30, 578/2073) in 1994 to 41% (95% CI 39–44, 671/1632) in 2002 (P for trend < 0.001). Of 18 440 asthma consultations from 1994 onwards 26% were in children <5 years, 29% in 5- to 16-year olds, 22% in 17- to 45-year olds, and 23% in adults >45 years. The <16 years age group accounted for 61% of first asthma episodes compared with 16% in the >45 years age group.
|
Figure 2 shows loess long-term and short-term trends with upper and lower 95% CIs for first and subsequent asthma episodes from 1994. Long-term trends show that consultations for first asthma episodes increased to an estimated average of 0.78 (95% CI 0.74–0.81) per 1000 consultations in 1999 and then fell to 0.62 (95% CI 0.55–0.69) in 2002, although the short-term trend suggests another upturn in 2001. Consultations for subsequent asthma episodes have fallen by 38.0% since 1994, from an estimated average of 1.50 (95% CI 1.40–1.61) per 1000 consultations to 0.93 (95% CI 0.82–1.04) in 2002. Short-term trends indicate a plateau from 1997 to 2000 before a steep decline. Figure 3 shows that yearly ratios of subsequent to first episodes declined over time in all age groups (P for trend in all age groups <0.001), demonstrating that subsequent episodes represented a decreasing proportion of all episodes. These ratios were highest in the two oldest age groups in which the decrease was also greatest. In 5- to 16-year olds the ratio increased from 1998 to 2001 but the overall trend was downwards.
|
|
|
Discussion |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
Time trend analysis from the SSSN shows that primary care consultation rates for first episodes of asthma in Switzerland began to fall in 2000. Rates of subsequent episodes have been falling since at least 1994. The yearly ratio of subsequent to first episodes has declined since 1994 in all age groups.
Methodological issues
The main advantage of the SSSN is that it is a national system with centralized collation, analysis, and dissemination of data. Sentinel practices are stratified by geographic area, sociodemographic characteristics and physician speciality19 and coverage exceeds that of other similar systems.20,21 Participation is voluntary, which may lead to some selection but improves the consistency of reporting. Another strength of this study was the use of a robust non-parametric method to estimate the non-linear trends in our data. This was more appropriate than using parametric methods such as quadratic polynomials7 for exploring non-linear trends because they do not specify a particular form for the trend. They also allow for shorter or longer trend detection by changing the smoothing parameters and both contribute to our understanding of the data.
There are some limitations in using primary care surveillance data, which need to be taken into consideration in interpreting our results. First, we calculated rates per 1000 consultations rather than population-based rates because practices do not serve defined populations. The use of different denominators in primary care epidemiology has been discussed23,24 and practice consultations are acknowledged to be an appropriate substitute when population denominators are not available.24 We believe that the time trends observed here are a reflection of the national situation because the numbers of physicians and consultations in the SSSN as a proportion of the respective totals in Switzerland remained stable over time, as did the number of consultations per physician. Second, consultation rates for asthma are sensitive to the number of paediatricians in the system. The proportion of paediatricians was reasonably stable and the age distribution of the population consulting for asthma remained constant over time. Third, although the definition of asthma was in accordance with the international paediatric asthma consensus group statement22 and remained unchanged throughout the study period it relied on clinical diagnosis, which might have differed between physicians and over time. However, the problem of subjectivity applies to all similar studies, including those using self-reported symptoms because there are no objective diagnostic tests for asthma. Fourth, consultations were recorded anonymously so we could not relate subsequent consultations to first episodes of asthma on an individual basis. We therefore focused on trends in the ratio of subsequent to first consultations rather than absolute figures.
The end of the asthma epidemic?
Our results are consistent with findings from a primary care surveillance system in England, in which new diagnoses of asthma fell from 50 per 100 000 population in 1993 to 30 per 100 000 in 1998.7 This was not explained by changes in patterns of health care use or diagnostic preference of doctors. First recorded consultations approximate the incidence of asthma diagnosis although absolute rates may be overestimated because of differences in the interpretation of a first episode. Taken together, these two studies provide some evidence that the incidence of asthma diagnosis in western Europe might be decreasing.
Studies that have investigated both reported symptoms in the past 12 months and lifetime diagnosis of asthma have found a decreasing or stable prevalence of reported symptoms in children25 and adults11 but an increase in reported prevalence of diagnosed asthma. This apparent discrepancy is interpreted as being attributable to a change in diagnostic labelling, with milder symptoms being labelled as asthma,11,25,26 but could also indicate a true decline in underlying prevalence.25 A recent fall in newly diagnosed cases of asthma, as reported here, might also reflect this trend, and is not incompatible with rising lifetime prevalence, particularly, for chronic conditions with low mortality such as asthma.
We also observed a fall in the proportion of asthma patients who had hay fever (data not shown), so a fall in atopic asthma might explain this trend and is consistent with other reports from Switzerland.27,28 The prevalence of asthma and hay fever fell in children of Swiss farmers whose respiratory health and sensitization to common allergens were measured regularly from 1983 to 1999.27 The Swiss Study on Childhood Allergy and Respiratory Symptoms, which assessed symptom prevalence using the International Study on Asthma and Allergy in Children core questions and specific IgE levels in serum, observed a levelling in trends between 1992 and 1998.28 Decreases in the prevalence of eczema and hay fever have also been reported from the British Isles,25 although a survey in Australia has reported an increase in the prevalence of eczema and hay fever, despite a fall in asthma prevalence.29 We did not distinguish between different asthma phenotypes (transient wheezing, non-atopic wheezing, and atopy-associated asthma)9 so we cannot comment on the relative contribution of these to the observed trends.
Changes in asthma management or health service utilization?
Decreasing morbidity and mortality from asthma in New Zealand in the 1990s30 and declining mortality in the United States of America31 have been attributed to improved management, with a shift in focus towards an increase in home medication with inhaled steroids and beta-agonists and a reduction in unscheduled primary care visits. The European Community Respiratory Health Survey found a large increase in the reported use of medications for asthma over the past decade.11 Our findings of a fall in subsequent asthma consultations are compatible with the hypothesis that management of asthma has changed, although formal self-management plans with regular physician review are not widely implemented in Switzerland, even amongst well-motivated patients.32 In our study, subsequent consultation rates increased in 5- to 16-year olds from 1998 to 2001, consistent with the finding that the highest rates of emergency visits to general practitioners for asthma are in children.32
A fall in consultation rates in primary care could also reflect changes in patterns of health care utilization. In Switzerland, patients have direct access to primary care, specialist physicians, and emergency departments.33 A rise in health insurance premiums in Switzerland since the introduction of a new scheme34 might have reduced use of primary care services, and the use of emergency departments in Swiss hospitals for chronic conditions including asthma has increased. We do not believe that this explains our observations. First, increasing attendances at emergency departments are thought to be owing to demographic changes in adults.33 Second, hospital referral rates for asthmatic patients in the SSSN from 1994 onwards were consistently low, ranging between 2.4 and 3.7%. Although we cannot exclude an increase in emergency room attendances without referral by a doctor, overall hospital admissions for asthma in children in Switzerland have also decreased by 20% from 1995 to 2002. Future studies on the incidence and prevalence of clinically diagnosed asthma should cover primary, secondary, and tertiary health care settings to facilitate the interpretation of secular trends. Studies should also begin to examine explanations for observed changes. Differentiation of the different phenotypes of asthma would provide insights into the relative contributions of allergic and other triggers of bronchoconstrictive disorders.
In conclusion, primary care consultations for asthma, subsequent to the initial diagnosis, have been declining in Switzerland since 1994. This is more likely to be owing to an increase in the use of home medication than to a shift in care to hospital settings. The incidence of diagnosed asthma might also be decreasing.
KEY MESSAGES
|
|
Acknowledgments |
|---|
We particularly thank the primary health care physicians participating in the SSSN for their incessant efforts to collect these data and Urban Wirz, president of the SSSN, for his support and encouragement. We thank Douglas Fleming, Birmingham Research Unit, UK, and Claudia Kuehni, Institute of Social and Preventive Medicine, Bern, Switzerland for constructive comments on previous versions of this manuscript.
|
References |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
1 Magnus P, Jaakkola JJ. Secular trend in the occurrence of asthma among children and young adults: critical appraisal of repeated cross sectional surveys. BMJ 1997;314:1795–99.
2 Burr ML, Butland BK, King S, Vaughan-Williams E. Changes in asthma prevalence: two surveys 15 years apart. Arch Dis Child 1989;64:1452–56.
3 Mitchell EA, Asher MI. Prevalence, severity and medical management of asthma in European school children in 1985 and 1991. J Paediatr Child Health 1994;30:398–402.[Web of Science][Medline]
4 Peat JK, van den Berg RH, Green WF, Mellis CM, Leeder SR, Woodcock AJ. Changing prevalence of asthma in Australian children. BMJ 1994;308:1591–96.
5 Omran M, Russell G. Continuing increase in respiratory symptoms and atopy in Aberdeen schoolchildren. BMJ 1996;312:34.
6 Shafazand S, Colice G. Asthma. The epidemic has ended, or has it? Chest 2004;125:1969–70.
7 Fleming DM, Sunderland R, Cross KW, Ross AM. Declining incidence of episodes of asthma: a study of trends in new episodes presenting to general practitioners in the period 1989–98. Thorax 2000;55:657–61.
8 The International Study of Asthma and Allergies in Childhood (ISAAC) Steering Committee. Worldwide variation in prevalence of symptoms of asthma, allergic rhinoconjunctivitis, and atopic eczema: ISAAC. Lancet 1998;351:1225–32.[CrossRef][Web of Science][Medline]
9 Kuehni CE, Davis A, Brooke AM, Silverman M. Are all wheezing disorders in very young (preschool) children increasing in prevalence? Lancet 2001;357:1821–25.[CrossRef][Web of Science][Medline]
10 Hill R, Williams J, Tattersfield A, Britton J. Change in use of asthma as a diagnostic label for wheezing illness in schoolchildren. BMJ 1989;299:898.
11 Chinn S, Jarvis D, Burney P et al. Increase in diagnosed asthma but not in symptoms in the European Community Respiratory Health Survey. Thorax 2004;59:646–51.
12 Bollag U, Cloetta J, Oberreich J, Paget WJ. Asthma trends in Switzerland: reports by primary care physicians in the Swiss Sentinel Surveillance Network between 1988 and 1996. Eurosurveillance 1999;4:21–24.
13 Bollag U, Paget WJ, Oberreich J, Cloetta J. Asthma in the community. Observations by the Swiss Sentinel Surveillance Network over a ten year period (1988–1997). Eur J Gen Pract 2000;6:122–29.
14 Matter HC, Zwahlen M, Cloetta J, Helbling P, Zimmermann HP. Re: 'Medical practice-based influenza surveillance: viral prevalence and assessment of morbidity'. Am J Epidemiol 1995;142:1347–49.
15 Venables WN. Modern Applied Statistics with S-PLUS. 3rd edn. New York: Springer, 1999.
16 Cleveland RB, Cleveland WS, McRae JE, Terpenning I. STL: a seasonal-trend decomposition procedure based on loess (with discussion). J Offic Statistics 1990;6:3–73.
17 Shumway RH, Stoffer DS. Time Series Analysis and its Applications. New York: Springer, 2000.
18 Hastie TJ, Tibshirani RJ. Generalized Additive Models. London: Chapman and Hall, 1996.
19 Schuler M, Nef R. Räumliche Typologien des Schweizerischen Zentrum-Peripherie Musters. Nationales Forschungs-Programm 'Regionalprobleme in der Schweiz', Arbeitsbericht 35. Bern: Swiss National Science Foundation, 1983.
20 Foundation of the Netherlands Institute of Primary Health Care (NIVEL). Continuous Morbidity Registration Sentinel Stations in the Netherlands. Utrecht: NIVEL, 1997.
21 Nutting PA, Baier M, Werner JJ, Cutter G, Reed FM, Orzano AJ. Practice patterns of family physicians in practice-based research networks: a report from ASPN. Ambulatory Sentinel Practice Network. J Am Board Fam Pract 1999;12:278–84.[Abstract]
22 International consensus report on diagnosis and management of asthma. Eur Respir J 1992;5:601–41.[Web of Science][Medline]
23 Schlaud M, Brenner MH, Hoopmann M, Schwartz FW. Approaches to the denominator in practice-based epidemiology: a critical overview. J Epidemiol Community Health 1998;52:13S–19S.
24 Fleming DM. The denominator for audit in general practice. Fam Pract 1985;2:76–81.
25 Anderson HR, Ruggles R, Strachan DP. Trends in prevalence of symptoms of asthma, hay fever, and eczema in 12–14 year olds in the British Isles, 1995–2002: questionnaire survey. BMJ 2004;328:1052–53.
26 Weiland SK, Pearce N. Asthma prevalence in adults: good news? Thorax 2004;59:637–38.
27 Gassner-Bachmann M, Wüthrich B. Bauernkinder leiden selten an Heuschnupfen und Asthma. Deutsch Med Wochenschr 2000;125:924–31.
28 Braun-Fahrländer C, Gassner M, Grize L et al. No further increase in asthma, hayfever and atopic sensitisation adolescents living in Switzerland. Eur Respir J 2004;23:407–13.
29 Robertson CF, Roberts MF, Kappers JH. Asthma prevalence in Melbourne schoolchildren: have we reached the peak? Med J Aust 2004;180:273–76.[Web of Science][Medline]
30 Garrett J, Kolbe J, Richard G, Whitlock T, Rea H. Major reduction in asthma morbidity and continued reduction in asthma mortality in New Zealand: what lessons have been learned? Thorax 1995; 50:303–11.
31 Sly RM. Decreases in asthma mortality in the United States. Ann Allergy Asthma Immunol 2000;85:121–27.[Web of Science][Medline]
32 Kuehni CE, Frey U. Age-related differences in perceived asthma control in childhood: guidelines and reality. Eur Respir J 2002;20:880–89.
33 Santos-Eggimann B. Increasing use of the emergency department in a Swiss hospital: observational study based on measures of the severity of cases. BMJ 2002;324:1186–87.
34 Federal Health Insurance Law. Federal Office of Social Insurance, Berne, 18 March 2004.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  U. Bollag, L. Grize, and C. Braun-Fahrlander Is the ebb of asthma due to the decline of allergic asthma? A prospective consultation-based study by the Swiss Sentinel Surveillance Network, 1999-2005 Fam. Pract., April 1, 2009; 26(2): 96 - 101. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Pearce, N. Ait-Khaled, R. Beasley, J. Mallol, U. Keil, E. Mitchell, C. Robertson, and and the ISAAC Phase Three Study Group Worldwide trends in the prevalence of asthma symptoms: phase III of the International Study of Asthma and Allergies in Childhood (ISAAC) Thorax, September 1, 2007; 62(9): 758 - 766. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Pearce and J. Douwes Commentary: Asthma time trends--mission accomplished? Int. J. Epidemiol., October 1, 2005; 34(5): 1018 - 1019. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||