Methods

Study design

This analysis is based on a subgroup selected from the most recent laboratory study of the TAHS. The details of the methodology and some results from this study have been reported elsewhere.16 17 18 19 20 In brief, TAHS commenced in 1968 by recruiting 8583 Tasmanian children born in 1961, who were surveyed for respiratory problems and underwent clinical examination and lung function measurements. Subsequent follow-up surveys were completed at the ages of 13 (in 1974), 20 (in 1981) and 31 (in 1992). The current follow-up started in 2004 when the probands were 44 years of age. We have traced 85.2% (n = 7312) of the original 1968 cohort to an address and achieved a response of 78.4% (n = 5729) to a postal survey.20 A subgroup of these respondents were selected based on their participation in the previous follow-up studies, samples of which were enriched for asthma, and invited to participate in a more detailed laboratory study. Of 2373 invited, 1389 (58.5%) took part in a full laboratory visit, 354 (15%) completed a telephone questionnaire only and 630 (26.5%) withdrew.

Questionnaires and clinical tests

During the laboratory phase of the study, participants completed an interview-administered questionnaire which captured information about their demographics and, where relevant, details of their asthma (past and present), smoking history, medication use and healthcare service utilisation.

Lung function testing

Lung function was measured with an Easyone ultrasonic spirometer (ndd, Medizintechnik, AG, Switzerland). Participants were asked not to smoke for 4–6 h prior to testing. Forced expiratory volume in 1 s (FEV₁) was recorded as the best of three blows that met American Thoracic Society criteria.21 Predicted values for FEV₁ and forced vital capacity (FVC) were calculated from age, height and gender using equations by Gore et al.22 Spirometry was repeated 10 min after 200 μg (2 puffs) of salbutamol administered via a spacer.

Definitions

Current asthma

Asthma-ever was defined as a positive response by participants in the current laboratory follow-up to the question “Have you ever had asthma?” Participants were defined as having current asthma if they had experienced symptoms within the last 12 months.

Asthma severity and minimal adequate medication

The definition of asthma severity was adopted from the NAC classification of asthma for patients with untreated newly diagnosed asthma and based on self-reported morning, daytime and night-time symptoms and number of flare-ups (table 1). Similarly, the minimum level of adequate preventer medication was defined according to the NAC guidelines and the medication regimens used here were essentially the minimum level of treatment recommended to be prescribed at initial assessment by a physician at a particular level of current severity.

Onset of current asthma

Participants with current asthma were defined as having “early-onset current asthma” or “late-onset current asthma” based on asthma data from the 1968, 1974 and 2004 surveys. If the participants with current asthma had reported either ever asthma in 1968 or current asthma in 1974 or both, they were classified as having early-onset current asthma. If the participants had current asthma in 2004 but had no ever asthma recorded in 1968 or 1974, they were classified as having late-onset current asthma.

Statistical analysis

All statistical analyses were performed using Stata version 9 (StataCorp, College Station, Texas, USA). The distributions of personal characteristics such as gender, smoking status, family history and level of education across the asthma severity groups were examined using χ² tests. We also performed multiple logistic regression analyses to examine the associations between these personal characteristics and adequacy of preventer medication use by participants with current asthma. In addition, we performed multiple linear regressions to assess the relationships between lung function measurements and adequate preventer medication use while accounting for the asthma severity. We further stratified these associations by the adequacy groups to investigate any variation in the association between adequacy of medication and lung function across the severity groups. All multiple linear regression models were adjusted for gender, smoking status, age, weight and height. A p value of <0.05 was regarded as statistically significant.

Results

Personal characteristics of the participants by asthma severity

In total, 702 participants reported asthma ever and, of these, 351 (50%) had current asthma. Of those with current asthma, 103 (29.3%) had intermittent asthma, 98 (27.9%) had mild persistent asthma, 92 (26.2%) had moderate persistent asthma and 58 (16.5%) had severe persistent asthma. The distribution of relevant personal characteristics by severity groups is shown in table 2. The distributions of the proportions of women, the proportions of those with allergic conditions and those with parents and/or siblings with asthma across the severity groups were found to be statistically significant. On the other hand, distributions of level of education and smoking across the severity groups were not statistically significant, even though it was notable that over 40% of the patients with severe persistent asthma were current smokers.

Asthma medication use by current asthma participants

Approximately 85% of participants with current asthma had used some form of asthma medication in the past 12 months. A breakdown of the types of medications used is presented in table 2. Overall, the majority of participants with current asthma had used reliever medication (78%, n = 274). Only 16% (n = 55) had used ICS alone and 26% (n = 93) had used a combination of ICS and LABA medication. In addition, four individuals were taking LABA alone, including one with severe asthma (table 2).

Lung function levels and severity of asthma

Lung function for participants with current asthma was significantly lower than for those who did not have current asthma. The mean FEV₁ pre-BD for the participants with current asthma was 2.99 l compared with 3.26 l in participants with non-current asthma (mean difference −0.27 l (95% CI −0.36 to −0.17), p<0.001). Similarly, post-BD FEV₁/FVC in the participants with current asthma was 75.2% compared with 79.4% for those with non-current asthma (mean difference −4.20% units (95% CI −5.52% to −2.87%), p<0.001). Lung function measurements stratified by current asthma severity are summarised in table 3. Lung function was significantly associated with increasing severity (table 3). In participants with severe persistent asthma, pre-BD FEV₁ was 0.3 l lower than in participants with mild persistent asthma (95% CI −0.52 to −0.07, p = 0.011).

Adequacy of preventer medication and its determinants

Of the 351 participants with current asthma, only 26% were taking minimally adequate preventer medication. We found that participants with current asthma who had family members with a history of asthma were more likely to use adequate preventer medication (table 4). Moreover, adequate preventer medication users were more likely to have visited general practitioner/casualty/hospital admission in the last year. Univariate logistic regression analysis showed that participants with early-onset asthma were more likely to use adequate preventer medication; however, this association attenuated after adjusting for “hospital admission for asthma ever” since participants with early-onset asthma were more likely to have been admitted to hospital for asthma (table 4). The associations between other determinants including gender, smoking, social class, and other allergic diseases with adequate preventer medication use were not found to be statistically significant.

Adequacy of preventer medication use by severity

For the group of participants with current asthma, approximately 71% (n = 248) had persistent disease activity. While 45% (n = 113) of the participants with persistent asthma reported having used any kind of ICS, only 33% (n = 82) used them on a regular basis. Of those taking regular ICS, not all were using adequate doses or combination treatment in relation to their severity. Approximately 28% of the participants with moderate persistent asthma and the 48% of participants with severe persistent asthma were using ICS regularly, whereas only 17% of the participants with moderate and 26% of the participants with severe persistent asthma were using defined minimal adequate preventer medication. Of participants with mild persistent asthma, 29% were found to be taking adequate preventer medication.

When compared with the group of participants with mild persistent asthma, those with moderate persistent asthma had reduced odds of 57% for using adequate preventer medication (adjusted OR 0.43 (95% CI 0.20 to 0.91), p = 0.027). The odds of adequate preventer medication use by participants with severe persistent asthma were also found to be lower (a reduction in 48%) compared with the group of participants with mild persistent asthma, but this finding did not reach statistical significance (adjusted OR 0.52 (95% CI 0.22 to 1.20), p = 0.123, table 4).

Fixed airflow obstruction in adequate and inadequate preventer medication groups

Lung function data were available for approximately 75% of the participants with current asthma. We analysed the association between lung function and severity by adequacy of preventer medication use. In the adequate medication group, those with severe asthma had lower FEV₁ levels compared with the mild group. In addition, we found a statistically significant and progressive reduction in FEV₁/FVC (from mild to severe) for the inadequately treated group but not for the adequately treated group (table 5).

Discussion

Our aim was to evaluate the adequacy of preventer medication use by Australians with current asthma and its relationship to severity and lung function. To our knowledge, this is the first study to define adequacy of preventer medication using established national asthma guidelines and to assess how well asthma is being managed against such criteria. We found that the majority of patients with current asthma had persistent asthma, yet few were taking even minimally adequate preventer medication. Importantly, for those who were not taking adequate preventer therapy, their lung function measurements declined significantly with increasing severity while those taking adequate therapy had no such decline. Other allergic conditions and family history were observed to be strongly associated with increasing asthma severity by group. We found that, if participants had adult-onset asthma, had no family history and no history of hospital admission, they were more likely to be taking inadequate medications.

We investigated the implications of our findings for asthma outcomes and found a statistically significant fall-off in the post-BD FEV₁/FVC ratio from mild to severe groups for the inadequate preventer users but not for the adequate preventer users. Although the numbers in these analyses were relatively small, we hypothesise that adequate use of preventer medication is protective against the progressive development of fixed airway obstruction associated with increasing asthma severity. This finding is supported by a recent randomised controlled trial of combination therapy versus formoterol alone for as-needed treatment, which found significantly better FEV₁ for those on ICS/LABA combination therapy.23

A few participants with current persistent asthma reported not taking any medication for asthma in the last year. Of the three participants with severe persistent asthma who were not taking any asthma medication, one reported an emergency visit in the previous year. Lung function measurements were not available for this participant but were available for the other two; one had FEV₁ post-BD <80% predicted. This patient with severe persistent asthma and impaired lung function had asthma onset in childhood. Similarly, of the nine participants with moderate persistent asthma who were not taking any medication for asthma in the last year, one had FEV₁ post-BD <80%. None of these patients reported emergency visits/hospital admission in the last year.

We focused our evaluation of adequacy of preventer medication use on participants with persistent asthma only because, according to NAC guidelines, participants with intermittent asthma do not need to be on ICS.3 The proportion of adequate preventer medication use by participants with persistent asthma was approximately 24% (95% CI 18% to 29%). This is lower than the findings from Italy by De Marco et al12 who reported that 52% of patients with persistent asthma used adequate preventer medication. However, the mean age in the Italian study was 34 years compared with 44 years in our study population. It is perhaps surprising that younger patients may be adhering better to preventer medication than we have observed in our study. Patients’ understanding of medication and adherence to asthma management are critically important, and hence it poses a major challenge for healthcare professionals to motivate patients to take preventer medications even when they are asymptomatic.24

Participants with mild persistent asthma were more likely to use adequate medication, and this is likely to be causal (that is, taking proper therapy leads to less severe disease and better outcomes). Even though we do not have information on disease severity for the participants at the beginning of preventer medication use, there is evidence to support this suggestion that regular use of adequate preventer medication for adequate periods of time results in patients shifting from “severe” to “less severe” disease activity.25 26 It is critical that patients using medication are followed up regularly over a considerable period and monitored iteratively using several outcome indices to assess control and management of their disease.

Our definition of adequacy of preventer medication is approximately equivalent to what a patient with asthma should be prescribed when being assessed by a GP for the first time. Many would need a subsequent increase of treatment to gain good control and perhaps future back-titration of ICS. Regardless, we thought it is reasonable to use the basal expected treatment to define “minimal adequacy” for the purposes of this study, recognising that this may well underestimate true needs.

Contrary to what might be expected, we observed a higher proportion of our adequate user group to have made use of healthcare services in the last year. The most likely explanation for this association is reverse causation (ie, those who made use of healthcare services were more likely to receive correct treatment advice and to adhere to treatment after an episode sufficient to attract medical intervention). We also found that the adequate preventer medication users were more likely to have asthma since childhood. This association was attenuated when we adjusted for “hospital admission for asthma ever”, which suggests participants with early-onset asthma were more likely to use healthcare services. Hence, it is possible that the adequate preventer medication users in our study population had a more long standing and serious asthma condition and therefore they were more likely to make use of healthcare services.

The availability of spirometric data were a major strength of our study. However, we did not include lung function as part of the severity classification, partly because it was not available for everyone but mainly because we wanted to assess spirometric measurements, especially degree of fixed airflow obstruction, as outcomes. Adequate preventer medication use was found to protect against the effects of progressive severity on development of fixed airflow obstruction. This should be an encouragement to both patients and practitioners to persist with good asthma management and especially appropriate medication titrated against disease severity criteria.

One limitation of our study is the lack of information on when or how frequently participants consult their doctors about their asthma. However, since those on preventer therapy would need regular prescriptions, we assume that there would have been frequent opportunities to intervene, either by the GP or pharmacist. A recent publication has emphasised how effective intervention from healthcare professionals can be, including improving quality of life, when an inadequate level of therapy relative to disease severity is noticed.27 Bereznicki et al show strong evidence that patients are prepared to change therapeutic practice under good professional guidance.27 This contrasts with published evidence that physicians have found it challenging to adhere to NAC guidelines, suggesting perceived limitations to such a regimented “one size fits all” approach. Some of these limitations include heterogeneity of asthma and symptom presentations, variations in response to treatment, lack of clearcut asthma diagnosis and cost of asthma management.28 29 Furthermore, it has been demonstrated that emergency department physicians rarely use the NAC guidelines to assess acute asthma status and tend to underestimate the level of severity.30 Overall, the general picture seems to be one of many lost opportunities to intervene effectively in asthma care and to use core aspects of guidelines to assess asthma well and to initiate adequate levels of therapy accordingly.

Conclusion

Overall, the use of appropriate asthma therapy by our study population was found to be grossly inadequate. The groups at greatest risk were those with adult-onset persistent asthma, moderate or severe persistent asthma and those without a family history of asthma. Efforts for improvement should therefore be especially focused on these groups. Further research is needed on methods of improving patients’ understanding of and adherence with effective medication use. Our current models of asthma care delivery need to be questioned and re-assessed. Patients with persistent symptomatic asthma, in particular, need more effective encouragement to use ICS and LABA combination therapy on a regular basis, and there are strategies that work. Adams et al31 reported that more positive interactions in terms of willingness of doctors to spend more time with and explain asthma management to patients appears to be associated with increased preventer use, independent of the level of asthma symptoms. Pharmacist interventions can also be successful.27 32 Little specific attention has been given to the specific characteristics and needs of middle-aged patients with asthma, and yet they are a group especially vital to families, business and society in general. Our findings indicate that there is a large gap for the majority of patients with moderate to severe asthma between their current level of asthma management and optimal management, and that this is likely to have long-term consequences for their health.