Asthma risk prediction score accounts for early treatment response
medwireNews: Researchers have developed a measure for predicting future asthma risk based on patient characteristics and current level of asthma control.
“Patients’ individual risk can now be estimated in an easy way, as proposed but not specified, by asthma management guidelines,” say Persijn Honkoop (Leiden University Medical Centre, the Netherlands) and fellow team members in The Journal of Allergy and Clinical Immunology: In Practice.
The risk prediction score they have developed includes six “easy to obtain” predictors, namely sex, smoking status, Asthma Control Questionnaire (ACQ) score, and exacerbations in the previous year at baseline, as well as ACQ score at 3 months and the number of exacerbations in the first 3 months as an indication of early treatment response.
Based on regression coefficients, the score classifies patients as being at low, intermediate, or high risk of having uncontrolled asthma (ACQ score ≥1.5) at 12 months or at least one severe exacerbation (admission to emergency department or at least 3 days of prednisolone) during the final 6 months.
The model was tested in 304 primary care patients with asthma aged an average of 40.2 years. Of these, 27.3% experienced an event at 12 months, with asthma remaining uncontrolled in 52 individuals, severe exacerbations occurring in 19, and both endpoints seen in 12 patients.
In the first model, only baseline patient characteristics – sex, current smoking, ACQ score, exacerbations in the previous year – were included and this yielded an area under the receiver operating characteristic curve (AUC) of 0.78.
But the researchers note that when early treatment response – ACQ assessment at 3 months and occurrence of exacerbations in the first 3 months – was also considered the AUC increased to 0.84.
“Furthermore by adding the ACQ as a predictor in our model, it is possible for the clinician to estimate the current level of control and the level of guideline-defined future risk, all within the same risk model,” they add.
The odds ratios for future risk associated with these individual elements were 2.05 for women versus men (p=0.03), 1.49 for current smokers versus nonsmokers at baseline (p=0.35), 1.74 per 0.5 point increase in ACQ score at baseline (p<0.001), 2.45 for at least one exacerbation versus none in the year preceding baseline (p<0.01), 1.93 per 0.5 point increase in ACQ score at 3 months (p<0.001), and 6.40 for at least one exacerbation versus none in the first 3 months (p=0.02).
The participants were given a total score from 0 to 16 based on the factors present, and from this the team could classify the patient’s level of future risk, as low with a score of 0–4, intermediate with a score of 5–8, or high with a score of 9–16.
The absolute risks for each category were 11.7%, 47.0%, and 72.7%, respectively, and the corresponding proportion of study participants in each group were 64.2%, 23.6%, and 12.2%.
The findings were replicated when the risk prediction score was validated in an independent cohort of 195 patients, yielding an AUC of 0.77.
The researchers acknowledge that “the ACQ is as yet not routinely collected in primary care,” but they assume that “the result would not differ greatly from other asthma symptom scores, since outcomes are correlated.”
They advise that “[t]he developed risk model should be used alongside other important components of a structured asthma review; e.g. inhaler technique, adherence and patient education.”
And conclude that “with the inclusion of early treatment response in the model, a review of effectiveness of treatment is included, thereby minimizing exposure to ineffective treatment.”
By Lucy Piper
medwireNews is an independent medical news service provided by Springer Healthcare. © 2018 Springer Healthcare part of the Springer Nature group