Incentives to quit smoking in primary care

Abstract

Spirometry with pictorial feedback on lung age, not just raw data, improves quit rates

In the accompanying randomised controlled trial, Parkes and colleagues assess the effect of telling patients over 35 years of age their estimated spirometric lung age as an incentive to quit smoking.¹ Support for conducting the trial comes from a recent Polish observational study on the potential association between smoking cessation and participants’ spirometry results, as communicated using Fletcher and Peto’s diagram (a pictorial representation of how smoking ages the lungs).² The Polish study showed higher smoking cessation rates at one year in smokers with airway obstruction than in those with normal spirometric parameters. However, the study had no control group without spirometry testing or without feedback on such testing. The authors called for a large randomised controlled trial comparing the effect of providing spirometry results versus no spirometry results on smoking cessation.

In Parkes and colleagues’ trial, participants in the intervention group received comprehensive information about their spirometry results including individualised interpretation, estimated lung age, and Fletcher and Peto’s diagram. People in the control group received written results as raw data on forced expiratory volume in one second, with no further explanation. Participants in both groups were advised to quit smoking and were offered an optional referral to an intensive support service. Smokers randomised to the intervention group were about twice as likely to be not smoking at 12 months’ follow-up than those in the control group. A subgroup analysis found no evidence of a dose-response relation between “lung age deficit” (lung age minus chronological age) and the outcome, as quitters and non-quitters had similar lung age deficits. This study did not look at the potential health benefit of screening for chronic obstructive pulmonary disease because all participants underwent spirometry testing.

Another recent randomised controlled trial investigated a closely related research question in smokers aged 18-24 years.³ It focused on intermediate psychological outcomes, based on health behaviour theories such as the “health belief model.” This model states that people are likely to follow a particular health action if they think they are susceptible to a condition that they consider serious, and if they believe that the benefits of the action outweigh the costs.⁴ The intervention group received a smoking cessation booklet plus feedback about their spirometric lung age and respiratory symptoms, and the control group received only the smoking cessation booklet. Perceived risks, worries, and desire to quit smoking were assessed using 7 point Likert-type scales at study entry and after delivery of the intervention. No significant differences were found between groups at either time point. They also assessed the perceived relevance of lung age and feedback on respiratory symptoms in the intervention group using the “10 item personal involvement inventory.” A significant inverse correlation was seen between lung age deficit and perceived relevance of lung age feedback, perhaps as a defensive reaction against potentially worrying information..

A systematic review explored the effect of biomedical risk assessment as an aid for smoking cessation.⁵ It included trials in which a measurement—such as exhaled carbon monoxide, spirometry, or genetic testing—was used to increase motivation to quit. For trials to be eligible, the control group had to receive all parts of the intervention except for the biomedical feedback. Only one trial of spirometry was eligible for this review.⁶ It found no significant difference in smoking cessation at 12 months’ follow-up between participants receiving spirometry feedback and repeated counselling and those receiving counselling but not spirometry testing (odds ratio for 7 day abstinence at 12 months in the intervention group compared with the control group 1.21, 95% confidence interval 0.60 to 2.42). An ongoing updated search found another eligible paper that had similar results.⁷ Parkes and colleagues investigated a slightly different research question (comprehensive, illustrated, and individualised oral feedback versus short, raw, and written feedback) than these two trials where participants did not undergo spirometry if they were allocated to the control condition.

On the basis of the evidence so far, general practitioners have to decide whether to wait for a trial comparing the potential benefit for smoking cessation of spirometry testing using lung age feedback versus no spirometry testing or whether to adopt the strategy suggested by Parkes and colleagues. In making this decision they should be aware of the limitations of the trial—for example, the lack of information about the comparability of the study sample with the entire recruitment population, the longer duration of contact between participants and caregivers in the intervention group than in the control group, and outcome data that are limited to point-prevalence abstinence.⁸ Despite these limitations, however, providing feedback on lung age with graphic displays seems to be the best option so far for communicating the results of spirometry. This strategy might also be an opportunity for general practitioners to tailor smoking cessation messages to the individual, as recommended in the recent National Institute for Health and Clinical Excellence (NICE) guidance on smoking cessation.⁹