The Natural History of Asthma in a Primary Care Cohort

The article reporting the natural history of asthma in a primary care cohort argues that the use of BHR testing has limited value in predicting subsequent respiratory tract disease for patients who have asthma diagnosed by a physician. Our research on the natural history of asthma in a community cohort shows very different results to those shown in this study. We have shown that a positive BHR test is an important risk factor for adult asthma [1], persistence of wheeze [2] and also has an impact on reduced growth of FEV1 [3].

With such conflicting results we should ask what are the differences between studies. Firstly, the rate of asymptomatic BHR of 39% is much higher than we observe in Australia. In a recent study of 8 – 11 year old school children in Belmont we found the prevalence of asymptomatic BHR was 8.3% [4]. Other European studies also report asymptomatic BHR at a rate lower than 39%. Secondly, the classification of outcome appears quite variable and could be made at any time over a period of 10 years by chest physicians, internists, pediatricians or family physicians. Finally, there were only 16 subjects who developed asthma and therefore the study has limited power to draw conclusions about AHR as a risk factor for developing asthma. These differences may help to explain the different results.

Interestingly, the combination of symptoms and BHR was predictive of an asthma diagnosis. This classification which we have called ‘current asthma’ identifies a group with clinically important asthma [5].

We believe that BHR testing does provide additional information beyond that obtained by questionnaire. Although BHR testing is not a gold standard test for asthma and we are not recommending use in routine clinical practice we have previously demonstrated that it does have important prognostic significance in a community-based cohort.

1. Toelle BG, Xuan W, Peat JK, Marks GB. Childhood factors that predict asthma in young adulthood. European Respiratory Journal 2004; 23:66-70.

2. Xuan W, Marks GB, Toelle BG, Belousova E, Peat JK, Berry G, Woolcock AJ. Risk factors for onset and remission of atopy, wheeze, and airway hyperresponsiveness. Thorax 2002; 57:104-109.

3. Xuan W, Peat JK, Toelle BG, Marks GB, Berry G, Woolcock AJ. Lung function growth and its relation to airway hyperresponsiveness and recent wheeze. American Journal of Respiratory and Critical Care Medicine 2000; 161:1820-1824.

4. Toelle BG, Ng K, Belousova E, Salome CM, Peat JK, Marks GB. Prevalence of asthma and allergy in schoolchildren in Belmont, Australia: three cross sectional surveys over 20 years. British Medical Journal 2004; 328:386-387.

5. Toelle BG, Peat JK, Salome CM, Mellis CM, Woolcock AJ. Toward a definition of asthma for epidemiology. American Review of Respiratory Disease 1992; 146:633-637.

Competing interests:   None declared

We are very encouraged with the comments we received on our article “ The Natural History of Asthma in a Primary Care Cohort”{1}. It offers an opportunity to reflect more in the implications of our findings for patients with asthma in family practice. We would like to start to comment on the methodology of our study. First, the diagnostic criteria require clarification. As stated in the paper the diagnosis of asthma in 1989 was based upon a screening of symptoms (questionnaire), spirometry and BHR testing according to international guidelines{2,3}. The outcomes of the 1989 screening were related to GP recorded respiratory tract morbidity (asthma, allergic rhinitis, upper and lower respiratory tract infections and allergic rhinitis) in the Nijmegen Continuous Morbidity Registration (CMR) practices{4,5} between 1990 to 2000. Thomas questions the quality of the GPs diagnosis of asthma. The GPs in our study based the diagnosis asthma upon criteria reflecting the knowledge of that time: respiratory symptoms like (episodic) dyspnoea, wheezing and or coughing (with phlegm) and reversibility on B2sympaticomimetica, according to national agreements. Evidence-based guidelines were introduced in Dutch family practice ahead of most other countries {6-11}. All practice visits in the period 1990-2000 were considered for the diagnosis asthma to be made.

The prevalence of asthma in the study population of 20 in 1,000 is in the high-normal range of the Dutch adolescent primary care population{12,13}which may also indicate that the CMR is representative of the family practice setting. Dr. Fardy refers to the important question of the number of diagnoses the GPs could enter. Some databases suffer from bureaucratic restrictions of limited numbers of diagnoses that practitioners can enter. This severely limits the validity of such databases for research. The CMR, however, is a research directed database and the GPs can enter as many diagnostic codes as needed to describe the clinical problem of the patient. Even more important for clinical research is the fact that the code can be corrected on the basis of subsequently collected information, and only the correct diagnosis will feature in the database. This is particularly important for conditions, like asthma, that vary over time and often can only be diagnosed on the basis of repeated observations.

Dr. Thomas refers to the interesting idea of NO measurements to identify airways inflammation. However, if he would refute our findings because of the diagnostic criteria researchers and GPs are not in line with current concepts, we strongly disagree. We have high expectations from GPs, but we fail to see how research data collected around 1989 could be compatible to techniques of 2004 like NO measurements. At that time, bronchial hyper responsiveness (BHR) was considered a hallmark of asthma, and PC20-methacholine was the state of the art of measuring it. Our study casts doubts on this since there were no indications of BHR without symptoms being a risk factor for asthma (operationalised as a GP diagnosed asthma). From this we concluded that airways inflammation in the absence of respiratory symptoms is no helpful risk indicator. Thomas presents an alternative hypothesis: the test to identify airways inflammation might have been wrong. This makes perfect sense, but it requires meticulous testing of his NO test – comparable to the way we tested the impact of PC20-methacholine –before we can shape clinical practice accordingly. At the moment we are conducting a follow-up study in which we again invite the participants of the 1989 study to perform a BHR test, test their reversibility on B2sympaticomimetica and take a symptom questionnaire. This way we can get a more objective diagnosis of asthma and compare this to the diagnosis in 1989 and the GP recorded respiratory morbidity. It will also be a way to validate the GPs diagnosis of asthma, Jackson emphasises correctly. At this time we advice GPs to base their diagnosis primarily on symptoms (and be alert of non-presented symptoms) and to be careful with testing like BHR.

1. WinklerPrins V, van den Nieuwenhof L, van den Hoogen H, Bor H, van Weel C. The natural history of asthma in a primary care cohort. Ann Fam Med. 2004;2:110-115.

2. Standards for the diagnosis and care of patients with chronic obstructive pulmonary disease (COPD) and asthma. This official statement of the American Thoracic Society was adopted by the ATS Board of Directors, November 1986. Am Rev Respir Dis. 1987; 136:225-244

3. National Heart, Lung and Blood Institute. Department of Health and Human services. National Institutes of Health. Global Initiative for Asthma, Globa. Strategy for Asthma Management and Prevention NHLBI/WHO workshop report. Bethesda, Md: National Institutes of Health; 1995; 95- 3659.

4. Van Weel C, van den Bosch WJ, van den Hoogen HJ, Smits AJ. Development of respiratory illness in childhood—a longitudinal study in general practice. J R Coll Gen Pract. 1987;37:404-408.

5. Van Weel, Smith, Beasley JW. Family practice research networks. Experiences from 3 countries. J Fam Pract. 2000; 49:938-943.

6. Geijer, R. M. M. et al. "NHG-Standaard COPD en Astma bij volwassenen: diagnostiek."[NHG guideline on COPD and Asthma in adults:diagnosis] Huisarts.Wet. 44.3 (2001): 107-17.

7. Geijer, R. M. M. et al. "NHG-Standaard COPD en astma bij volwassenen: diagnostiek." [NHG guideline on COPD and Asthma in adults: diagnosis] Huisarts.Wet. 40.9 (1997): 416-29.

8. Geijer, R. M. M. et al. "NHG-Standaard astma bij volwassenen: Behandeling[NHG guideline on asthma in adults: treatment]." Huisarts.Wet. 44 (2001): 153-64.

9. Geijer, R. M. M. et al. "NHG-Standaard astma bij volwassenen: behandeling[NHG guideline on asthma in adults: treatment]." Huisarts.Wet. 40.9 (1997): 443-54.

10. van der Waart, M. A. C. et al. "NHG-standaard CARA bij volwassenen: behandeling [NHG guideline on CNSLD in adults: treatment]." Huisarts.Wet. 35.11 (1992): 437-43.

11. Dirksen, W. J. et al. "NHG standaard Astma bij kinderen[NHG guideline on asthma in children]." Huisarts Wet 41 (1998): 130-43.

12. Lamberts H. In Het Huis van de Huisarts. Verslag van het Transitieproject. 2nd ed. Lelystad: Meditekst; 1994.

13. van de Lisdonk EH, van den Bosch WHJM, Huygen FJH, Lagro-Janssen ALM, eds. 3rd ed. Ziekten in de Huisartspraktijk. Maarssen: Bunge/Elsevier; 1999.

Competing interests:   None declared

This is a welcome publication giving some insight into the natural history of asthma and other upper and lower respiratory tract disease as encountered in primary care.

The methodology is really interesting and I am envious of the systems that this Dutch group have in place for tracking morbidity in general practice. To be able to look at continuous morbidity registration in a reasonably stable population is a real epidemiological and research treasure.

In specific reference to this article on the natural history of asthma, it is interesting that this study reminds us that good clinical history taking is often more beneficial than various special tests. A question for all Primary Care physicians is: “ How can we encourage patient to tell us about their symptoms of asthma when they are not consulting us about their asthma ?”

One of the things I would like to know is how many fields of data entry the GPs in the CMR have to enter and whether they have the option of putting in multiple morbidities at the one consultation? If they are only able to enter one “reason for encounter”, there may be some un-diagnosis or under-diagnosis as the GPs put in the major reason for encounter and other issues which are dealt with do not get recorded.

Competing interests:   None declared

WinklerPrins presents data on the usefulness of symptoms and objective Bronchial Hyperresponsiveness (BHR) testing on subsequent respiratory consultations and diagnoses in a Dutch primary care cohort (1). The paper raises important points on the value of objective testing, but we feel a debate on the diagnosis of asthma is needed.

In this paper, a subsequent physician diagnosis of asthma is used to calculate the predictive odds-ratios of symptom scores and positive BHR testing. It is perhaps not surprising that symptoms perceived by patients predicted a subsequent asthma diagnosis, as symptoms are usually the patient factor causing a consultation, and the driver of the diagnostic process. It is perhaps more surprising that BHR does not predict a subsequent respiratory diagnoses or consultations, although as the authors comment, due to differing patient consultation behaviour, we cannot say for sure that increased respiratory morbidity did not occur. It may be that such patients have limited lifestyle to minimise symptoms, are poor perceivers of asthma symptoms or simply that one-off BHR testing is less predictive of asthma than previously thought.

There are however serious questions to be asked regarding the objectivity of the physician asthma diagnosis. There is now international agreement that asthma is fundamentally a chronic inflammatory disease of the airways characterised by BHR and consequent respiratory symptoms (2). There is however no simple ‘gold-standard’ diagnostic test, and as the symptoms of asthma are non-specific mis-diagnosis may occur (3). It is currently difficult to measure airways inflammation outside research settings, although potentially techniques such as measurement of sputum eosinophil counts, exhaled Nitric Oxide (eNO) and exhaled breath condensates can act as surrogate markers of inflammation. Although the diagnosis of asthma should rest on a minimum of symptoms, the demonstration or variable or reversible airflow obstruction and a response to treatment, there is evidence that in practice this does not always occur (4;5). It may be that due to the publicity surrounding under- diagnosis of asthma, GP’s have felt under pressure to make a diagnosis in patients presenting with respiratory symptoms, particularly in children. It would be interesting to have more information on the diagnostic criteria used in the participant practices in this study. Even when ‘standard’ diagnostic criteria are met however, recent studies have indicated that measuring inflammatory parameters may be superior to conventional approaches and may result in more accurate diagnosis and more appropriate therapy. The use of technologies such as eNO in the routine diagnosis and assessment of asthma needs further evaluation. This is an important study but should trigger a debate on the validity of physician-based asthma diagnoses, patient perception of asthma and its role in under diagnosis and the validity of BHR testing.

Reference List

1. WinklerPrins V, van den Nieuwenbof L, van den Hoogen H, van Weel C. The Natural History of Asthma in a Primary Care Cohort. Ann Fam Med 2004;2:110-5.

2. International consensus report on the diagnosis and treatment of asthma. Eur Respir J 1992;5:641.

3. Thomas M, Price D. The diagnosis and assessment of asthma in primary care. In Scadding G, O'Connor B, eds. Key Advances in the Clinical Management of Asthma , London: The Royal Society of Medicine Press Ltd, 2004.

4. Dennis S, Price JF, Vickers M, Frost CD, Levy ML, Barnes PJ. The management of newly identified asthma in primary care in England. Prim Care Resp J 2002;11:120-2.

5. Marklund B, Tunsater A, Bengtsson C. How often is the diagnosis bronchial asthma correct? Fam Pract 1999;16:112-6.

Competing interests:   None declared