D3.124 - Effect of psycho-behavioural comorbidities on asthma symptom control scores: implications on asthma remission assessment

Instruments such as the Asthma Control Test (ACT) are used to assess asthma symptom control and are incorporated into asthma remission definitions. ACT captures patient-reported symptom burden, which may reflect psycho-behavioural comorbidities in addition to asthma biology. We examined the relative contributions of biological and psycho-behavioural factors to ACT. 

Patients were assessed at baseline for ACT, lung function (FEV₁%), blood eosinophils, fractional exhaled nitric oxide (FeNO), dysfunctional breathing (Nijmegen Questionnaire), depression (Patient Health Questionnaire-9, PHQ9), and anxiety (General Anxiety Disorder-7 questionnaire, GAD7), and followed up for 1 year. A multiple linear regression model was fitted with baseline ACT as the dependent variable, grouping predictors into demographics (age, sex), biology (FEV₁%, log₁₀ FeNO, log₁₀ blood eosinophils) and psycho-behavioural comorbidities (Nijmegen, PHQ-9, GAD-7). An exploratory structural equation model (SEM) specified a single latent psycho-behavioural factor measured by Nijmegen, PHQ-9 and GAD-7, with baseline ACT as the observed outcome. A separate linear regression model examined associations between change in ACT (ΔACT) and change in FEV₁% (ΔFEV₁%), adjusting for age and sex.

Baseline data of 158 patients are shown in Table 1. The multivariable model explained 23% of the variance in baseline ACT. Nijmegen was independently associated with ACT (β −0.13, 95% CI −0.22 to −0.05; p = 0.002). Dominance analysis showed that the psycho-behavioural domain contributed 0.18 R² units (77% of total variance), compared with 0.04 (19%) for demographics and 0.01 (4%) for biological markers. In the SEM, Nijmegen, PHQ-9 and GAD-7 loaded significantly on the latent factor supporting a common psycho-behavioural construct. Psycho-behavioural burden was associated with ACT (β −0.23, 95% CI −0.37 to −0.09; p = 0.001), independent of age and sex. For ΔACT, ΔFEV₁%, age and sex explained 23% of the variance. ΔFEV₁% (β 0.18, 95% CI 0.10 to 0.26; p < 0.001) and age (β −0.09, 95% CI −0.16 to −0.01; p = 0.033) were associated with ΔACT. 

ACT predominantly reflects symptom burden associated with psycho-behavioural comorbidities while biological markers contribute less to cross-sectional variance. Changes in ACT over time is reflective of change in lung function. These findings support interpreting ACT with psycho-behavioural comorbidities in mind when assessing asthma remission.