Behaviour change: The key to implementing evidence on COPD prevention, diagnosis and management

Translating clinical evidence into prevention behaviours and healthcare practice is complex and difficult. In healthcare systems, this translation (or ‘implementation’) involves someone doing something differently: the CEO endorses a local policy; a healthcare team clarifies members' responsibilities; a clinician discusses prevention strategies with a patient; the patient quits smoking. In other words, implementation requires behaviour change. We know that changing behaviour is difficult, but not impossible. Empirical research and theoretical development in behavioural science show that some behaviour change techniques1 are likely more effective than others, and that these effects depend on contextual, organizational and personal factors. Implementation science (the study of methods to support uptake of evidence-based interventions into practice and policy to improve health2) draws on bodies of knowledge from behavioural science, organizational disciplines and systems thinking to design interventions to support change. Nonetheless, ‘implementation interventions’, even if plausible, require evaluation to identify whether they are effective and cost-effective. Randomized evaluations of implementation interventions show that they can be effective. For example, reviews published by the Cochrane group, Effective Practice and Organisation of Care (EPOC; https://epoc.cochrane.org/), show positive effects on clinical behaviour change resulting from implementation strategies such as audit-and-feedback, clinical pathways or reminder systems. However, the heterogeneity around effects sizes is large and unexplained. There is still much to learn about how implementation strategies can be optimized to support changes in clinical practice and health behaviour. As we mark the World COPD Day on 17 November 2021, embracing the theme of ‘Healthy Lungs: Never More Important’, this is an opportune time to reflect on improving behaviour change and implementation strategies for the care of people living with chronic obstructive pulmonary disease (COPD). These principles can be broadly applied to prevention, biomarkers, diagnosis and management of COPD. Cigarette smoking is the predominant cause of COPD in many countries. A systematic review of smoking cessation interventions for people with COPD3 found that pharmacological therapy combined with behavioural counselling was more effective than each strategy separately. Yet provision of smoking cessation support by healthcare providers is low, with barriers including role identity, skills, time and guideline familiarity. Designing behavioural support interventions involves selecting behaviour change techniques to target specific barriers such as these. COPD also develops in never smokers exposed to risk factors including indoor and outdoor air pollution, occupational exposures, chronic asthma, tuberculosis and low maximally attained lung growth.4 Public health interventions can reduce exposure to biomass smoke from traditional cooking and heating methods in low- and middle-income countries. However, solutions such as improved cooking stoves and ventilation require significant educational and economic support and community engagement to successfully change both the built environment and culturally embedded behaviours. Biomarkers are biological or clinical markers that are useful indicators of disease activity, prognosis or treatment response. Many putative biomarkers have been investigated in COPD.4 For example, in the ECLIPSE study, markers of mortality at 8 years of follow-up included frequent exacerbations and hospitalizations, airflow obstruction and markers of systemic inflammation (IL-6, neutrophils and surfactant protein D).5 To be relevant for clinical use, biomarkers should add significant value to routine clinical decision-making, to facilitate precision health. Challenges for biomarker discovery and validation in COPD include appropriate study design, sample size and patient selection (including controls); consistency and magnitude of results across populations and COPD disease severities; and cost-effectiveness of and accessibility to testing. Translation from bench to bedside, especially for molecular biomarkers for COPD, first requires clinicians to understand the biomarker discovery process, which is increasingly more complex with ‘-omics’ technologies.6 Then, clinicians would need to adopt the use of biomarkers in day-to-day clinical practice, once they are convinced that evidence shows testing would improve patient outcomes. Clinicians are typically time-poor, so it is challenging for them to adopt additional practices, especially when they have to invest time and cognitive effort during the ‘learning curve’ phase of adoption. Indeed, it has been estimated to take an average of 17 years for research evidence to be translated into clinical practice.7 The mission of implementation science is to build evidence about the most effective strategies for accelerating the uptake of evidence-based practices. Symptoms of COPD include dyspnoea, chronic cough and/or sputum production and recurrent bronchitis in addition to a relevant history, including history of noxious exposure, long-standing asthma and genetic, socioeconomic and other factors. A firm diagnosis cannot be made without spirometric confirmation of poorly reversible airflow obstruction. Yet there are significant barriers to performance of spirometry in primary care where, ideally, the diagnosis should be made. A significant proportion of patients do not undergo confirmatory testing until they are admitted to hospital with an exacerbation.8, 9 Previous studies have identified only 50% of patients admitted to hospital with a diagnosis of COPD have had confirmatory spirometry,9 and that a false-positive diagnosis of COPD is present in as many as 60% of people who are labelled as having the disease (‘doctor diagnosis’ without spirometric confirmation).10 Factors influencing the performance of spirometry in general practice include lack of access to a well-maintained spirometer, lack of expertise and low confidence in interpreting spirometry results, among others. These three identified barriers to performing spirometry in primary care correspond to three of the ‘domains’ in the Theoretical Domains Framework of clinical behaviour change, which proposes 14 kinds of barriers.11 Using this framework to explore barriers to practice change has two important advantages over less structured approaches. First, it results in identification of a wider range of potential barriers that could be addressed. Second, the domains can be mapped on to evidence-based behaviour change techniques for supporting practice change, in this case, higher and more consistent uptake of spirometry. Management begins with confirmation of diagnosis through spirometry at a minimum, and other tests including more complex lung function tests and assessments of comorbidities (especially cardiac disease), as appropriate. Guidelines for management include the international GOLD guidelines and locally developed guidelines such as, in Australia, COPD-X.12 Non-pharmacological management includes prevention of progression through smoking cessation, as well as increased physical activity and pulmonary rehabilitation. Pharmacological treatment requires assessment of the presence of co-existent asthma, where inhaled corticosteroids will likely be indicated. In its absence, treatment with long-acting bronchodilators may improve symptoms and quality of life, and prevent exacerbations. Inhaled corticosteroids may be used in addition if exacerbations are a feature. A multidisciplinary team approach is appropriate for COPD management, which includes dyspnoea management strategies, management of comorbidities, assessment of need for oxygen therapy and discussion regarding goals of care. Management thus involves a complex constellation of behaviours, performed by multiple people: providers, patients and families. Some behaviours will be barriers to performing others (e.g., providers will be hesitant to suggest increased physical activity if patients are likely to resist), while some will interact to have a positive effect. For example, self-management approaches suggested by providers may have more potent effects if family members enthusiastically support the patient to adopt new self-management behaviours. Taking a behavioural perspective may therefore be useful when considering clinical management strategies. Despite helpful guidelines for preventing, diagnosing and managing COPD, there is room for improvement in clinical practice and patient behaviour. This is consistent with many other clinical fields in which there is strong evidence that guidelines, while necessary, are seldom sufficient on their own to change practice.13 Further support in the form of behavioural implementation strategies may be required to optimize care and to support patients and families in their attempts to prevent exacerbations and effectively self-manage COPD. Christine F. McDonald and Ian A. Yang are members of the Lung Foundation Australia COPD-X Guidelines Committee.