Systematic care for asthma in Australian general practice: a randomised controlled trial

Asthma has an impact on the quality of life of over 11% of the adult Australian population and is a cause of much absenteeism and many hospital admissions.1 As most people with asthma are managed in primary care,2 it is necessary to develop an evidence-based approach to the structural, organisational and personal aspects of asthma management in the primary care setting that will foster improved care. In the past, a reactive model of asthma management prevailed, with most patients considering the need to consult a general practitioner only when they had symptoms.

In 2001, the Australian Government introduced a policy initiative that recompensed GPs for completing the Asthma 3+ Visit Plan, encouraging them to adopt a more proactive approach to asthma management.3 This model was implemented without an adequate evidence base,4 and analysis indicated a poor uptake by practices before the commencement of this study in 2004.5

Our study aimed to test the efficacy of a process of systematic care for adult patients with asthma that focused on an integrated model of using up-to-date registers of patients with asthma and proactively recalling patients for review. There is some evidence that applying many of the elements of this model to children improves their care and health outcomes.6

Patients were identified through the practice’s prescription database. Eligible patients had to be 18 years or older, taking inhaled corticosteroids and assessed by the GP as having either moderate or severe asthma based on the guidelines of the National Asthma Council of Australia.7 Exclusion criteria included a diagnosis of chronic obstructive pulmonary disease only, inability to read and/or understand English, significant cognitive impairment and other severe illnesses likely to affect participation.

From the 5933 patients identified as potentially eligible, a letter was sent to 3367 inviting them to participate. In practices with fewer than 100 eligible patients, all patients were invited; for those with more than 100, a random selection of 100 eligible patients were approached. A total of 565 patients provided written informed consent and were enrolled between 8 June and 22 December 2004 (response rate, 16.8%) — 323 in SA and 242 in NSW. The mean number of patients per practice was 14 (range, 1–25).

The asthma management intervention comprised:

• assistance in setting up a register of patients with moderate or severe asthma;

• assistance in implementing a recall system, including the use of a postcard prompt;

• education of GPs and practice nurses about systematic asthma care (written information) and the use of spirometry (training workshops);

• education of staff about their role in systematic care and claiming the Asthma 3+ Visit Plan incentive payment (written information and one-on-one discussions); and

• provision of feedback by sending a summary of selected baseline data to GPs, with a reminder of what constituted best-practice care.

The central element was the posting of a yellow postcard to all participating patients in the intervention practices, outlining the benefits of asthma review and inviting them to attend their practice specifically for that purpose. This postcard was sent 2 months after patient recruitment. If there was no response to the postcard, one follow-up phone call was made by practice staff.

Control group practices continued to provide usual care for their asthma patients.

Box 1 shows the recruitment of practices and patients, and the retention of patients through the trial.

There were no differences between the 20 intervention and 20 control group practices in terms of size, socioeconomic index, or organisational structure. Only six control group and four intervention group practices had existing asthma registers, with one control and two intervention group practices previously running asthma clinics. There were 96 consenting GPs, of whom 74 (77%) were male; their mean age was 48.3 years, and there were no significant differences in the characteristics of control group and intervention group practice GPs.

Box 2 outlines the demographic and asthma characteristics of patients in the control (295 patients) and intervention (270 patients) groups, as well as a comparison with non-participants (eligible patients who were either not approached or who declined to participate). Overall, participants were older than non-participants, and were more likely to be female and to live in lower socioeconomic areas.

The intervention group had a slightly higher proportion of women compared with the control group, had lower socioeconomic status and were less likely to be employed.

Our trial of systematic asthma management resulted in a statistically significant difference in the rate of provision of a written asthma action plan by treating GPs, but not in other aspects of the process of care, quality of life, health outcomes or patient satisfaction. Written asthma action plans have been shown to improve the care of patients,12 and it is disappointing that there was still only a small proportion of patients receiving them.

Our results overall do not accord with a previous systematic review that found that a model of care incorporating self-monitoring, regular medical review and a written action plan improved many patient outcomes including the number of hospitalisations, unscheduled visits to doctors, days off from work and study, and bouts of nocturnal asthma.13

Despite our finding that the planned visits increased provision of asthma action plans, the benefits did not flow through to other measures at 12 months. Nevertheless, there were trends towards more comprehensive management and education, better patient understanding of asthma, improved quality of life and fewer asthma symptoms.

One explanation for the lack of sustainability could relate to the current structure of Australian general practice. A Finnish model of care achieved a reduction in the burden of asthma through a focus on GP and patient education, but was highly reliant on asthma nurses.14 Australian GPs do not currently have sufficient consultation time to provide a “complete package” of asthma care. Without a sustainable mechanism for funding nurses to help with educating patients about self-management and with case management, such models are unlikely to have a great impact.15 A method of funding practices that encourages broader infrastructure change, the development of a team culture and workforce innovation may be more effective.16,17

A second explanation could relate to the characteristics of the patients who attended. A discussion of the challenges for asthma management in a previous report includes focusing on difficult patients and recognising asthma as a long-term condition.18 While many patients in our study were willing to respond to a prompt for review, there was still resistance to visiting the GP. This may relate to denial about their condition, their perceptions of their asthma severity (only 37% self-rated this as moderate or severe), or poor perceptions of their level of control (64% rated control as good or very good). The relatively high levels of nocturnal asthma and exacerbations reported suggest that many did not have well controlled asthma. Mismatches between perception and objective assessment have previously been identified as barriers to attending doctors for review and to optimal self-management behaviour, compounded by a common view of asthma as an episodic condition.19,20 It is possible that if this model had targeted newly diagnosed, uncontrolled or non-compliant patients, it may have produced improvements in health outcomes. Cost of the visit was not seen as a deterrent, as the majority of practices bulk-billed.

Limitations of our study include our failure to reach the desired sample size; this may have reduced the power to detect differences. Also, the demographic characteristics of our patients did not fully represent those of the sampling frame, so our results may not be generalisable.

Because this was a pragmatic trial, eligibility was based on GP diagnosis of moderate or severe asthma. Even though patient report and spirometry testing cast some doubt on the diagnosis and/or severity of asthma in a few cases, we accepted the GP diagnosis. However, if some patients had only mild asthma, or they were already well controlled, there may have been little room for improvement from the intervention.

The failure to successfully implement the model of systematic asthma care we trialled in Australia calls into question whether a practice-based focus on proactive review for every patient with moderate or severe asthma is currently feasible. A previously described chronic illness model suggests that, as well as changes to the organisation of planned care similar to those carried out in our project, improving the quality of care in general practice requires broader workforce reform and changes to community awareness, self-management support and decision support.21

1 Recruitment of practices and patients, and patient retention throughout the trial

* Including partial case-note data for four patients in each arm who withdrew.