The Simple 10-Item Predicting Asthma Risk in Children Tool to Predict Childhood Asthma-An External Validation

External validation of the Predicting Asthma Risk in Children tool in a clinical cohort

INTRODUCTION

The Predicting Asthma Risk in Children (PARC) tool uses questionnaire-based respiratory symptoms collected from preschool children to predict asthma risk 5 years later. The tool was developed and validated in population cohorts but not validated using a clinical cohort. We aimed to externally validate the PARC tool in a pediatric pulmonology clinic setting.

METHODS

The Swiss Paediatric Airway Cohort (SPAC) is a prospective cohort of children seen in pediatric pulmonology clinics across Switzerland. We included children aged 1-6 years with cough or wheeze at baseline who completed the 2-year follow-up questionnaire. The outcome was defined as current wheeze plus use of asthma medication. We assessed performance using: sensitivity, specificity, negative predictive value (NPV) and positive predictive value (PPV), area under the curve (AUC), scaled Brier's score, and Nagelkerke's R² scores. We compared performance in SPAC to that in the original population, the Leicester Respiratory Cohort (LRC).

RESULTS

Among 346 children included, 125 (36%) reported the outcome after 2 years. At a PARC score of 4: sensitivity was higher (95% vs. 79%), specificity lower (14% vs. 57%), and NPV and PPV comparable (0.84 vs. 0.87 and 0.37 vs. 0.42) in SPAC versus LRC. AUC (0.71 vs. 0.78), R² (0.18 vs. 0.28) and Brier's scores (0.13 vs. 0.22) were lower in SPAC.

CONCLUSIONS

The PARC tool shows some clinical utility, particularly for ruling out the development of asthma in young children, but performance limitations highlight the need for new prediction tools to be developed specifically for the clinical setting.

The external validation of the asthma prediction tool in the French ELFE cohort

BACKGROUND

Existing predictive scores for early identification of children at high risk of developing asthma include invasive procedures, and hence have limited utility in a primary care setting. The Leicestershire respiratory cohort (LRC) has developed a noninvasive asthma prediction tool (APT) for children with promising results. We aimed to perform its external validation in the French general population Étude Longitudinale Française depuis l'Enfance (ELFE) cohort.

METHODS

Predictive scores were determined at Age 1 and the primary outcome of asthma was defined as parental reporting of "asthma ever or "wheezing in the past 12 months" at Age 5. Logistic regression was used to calculate the odds ratio (OR) and performance measures, and discriminative performance was reported using the receiver operating curve and area under curve (AUC). Calibration was assessed using Hosmer-Lemeshow goodness-of-fit test and visualized with a calibration plot. Overall performance was determined using Brier scores.

RESULTS

Of the 10,689 children analyzed: 84.9% were at low, 13.1% medium, and 2% at high risk of developing asthma at Age 5. Children in the medium-risk category were three times more likely to develop asthma (OR = 3.3, 95% confidence interval [CI] = 2.97-3.78) whereas 13 times more likely in the high-risk category (OR = 13.8, 95% CI = 10.2-18.8). The tool's AUC was comparable: LRC 0.74 versus ELFE 0.68; as were the Brier scores LRC 0.16 versus ELFE 0.14. The tool's performance was robust to changes in inclusion criteria and outcome definitions.

CONCLUSIONS AND RELEVANCE

Results of the present study and previous validation studies performed in high-risk populations provide a comprehensive measure of the effectiveness of the APT, providing encouragement for its application by general practitioners.

Development of a Symptom-Based Tool for Screening of Children at High Risk of Preschool Asthma

IMPORTANCE

Despite advances in asthma therapeutics, the burden remains highest in preschool children; therefore, it is critical to identify primary care tools that distinguish preschool children at high risk for burdensome disease for further evaluation. Current asthma prediction tools, such as the modified Asthma Predictive Index (mAPI), require invasive tests, limiting their applicability in primary care and low-resource settings.

OBJECTIVE

To develop and evaluate the use of a symptom-based screening tool to detect children at high risk of asthma, persistent wheeze symptoms, and health care burden.

DESIGN, SETTING, AND PARTICIPANTS

The cohort for this diagnostic study included participants from the CHILD Study (n = 2511) from January 1, 2008, to December 31, 2012, the Raine Study from January 1, 1989, to December 31, 2012 (n = 2185), and the Canadian Asthma Primary Prevention Study (CAPPS) from January 1, 1989, to December 31, 1995 (n = 349), with active follow-up to date. Data analysis was performed from November 1, 2019, to May 31, 2022.

EXPOSURES

The CHILDhood Asthma Risk Tool (CHART) identified factors associated with asthma in patients at 3 years of age (timing and number of wheeze or cough episodes, use of asthma medications, and emergency department visits or hospitalizations for asthma or wheeze) to identify children with asthma or persistent symptoms at 5 years of age.

MAIN OUTCOMES AND MEASURES

Within the CHILD Study cohort, CHART was evaluated against specialist clinician diagnosis and the mAPI. External validation was performed in both a general population cohort (Raine Study [Australia]) and a high-risk cohort (CAPPS [Canada]). Predictive accuracy was measured by sensitivity, specificity, area under the receiver operating characteristic curve (AUROC), and positive and negative predicted values.

RESULTS

Among 2511 children (mean [SD] age at 3-year clinic visit, 3.08 [0.17] years; 1324 [52.7%] male; 1608 of 2476 [64.9%] White) with sufficient questionnaire data to apply CHART at 3 years of age, 2354 (93.7%) had available outcome data at 5 years of age. CHART applied in the CHILD Study at 3 years of age outperformed physician assessments and the mAPI in predicting persistent wheeze (AUROC, 0.94; 95% CI, 0.90-0.97), asthma diagnosis (AUROC, 0.73; 95% CI, 0.69-0.77), and health care use (emergency department visits or hospitalization for wheeze or asthma) (AUROC, 0.70; 95% CI, 0.61-0.78). CHART had a similar predictive performance for persistent wheeze in the Raine Study (N = 2185) in children at 5 years of age (AUROC, 0.82; 95% CI, 0.79-0.86) and CAPPS (N = 349) at 7 years of age (AUROC, 0.87; 95% CI, 0.80-0.94).

CONCLUSIONS AND RELEVANCE

In this diagnostic study, CHART was able to identify children at high risk of asthma at as early as 3 years of age. CHART could be easily incorporated as a routine screening tool in primary care to identify children who need monitoring, timely symptom control, and introduction of preventive therapies.

Prediction models for childhood asthma: A systematic review

BACKGROUND

The inability to objectively diagnose childhood asthma before age five often results in both under-treatment and over-treatment of asthma in preschool children. Prediction tools for estimating a child's risk of developing asthma by school-age could assist physicians in early asthma care for preschool children. This review aimed to systematically identify and critically appraise studies which either developed novel or updated existing prediction models for predicting school-age asthma.

METHODS

Three databases (MEDLINE, Embase and Web of Science Core Collection) were searched up to July 2019 to identify studies utilizing information from children ≤5 years of age to predict asthma in school-age children (6-13 years). Validation studies were evaluated as a secondary objective.

RESULTS

Twenty-four studies describing the development of 26 predictive models published between 2000 and 2019 were identified. Models were either regression-based (n = 21) or utilized machine learning approaches (n = 5). Nine studies conducted validations of six regression-based models. Fifteen (out of 21) models required additional clinical tests. Overall model performance, assessed by area under the receiver operating curve (AUC), ranged between 0.66 and 0.87. Models demonstrated moderate ability to either rule in or rule out asthma development, but not both. Where external validation was performed, models demonstrated modest generalizability (AUC range: 0.62-0.83).

CONCLUSION

Existing prediction models demonstrated moderate predictive performance, often with modest generalizability when independently validated. Limitations of traditional methods have shown to impair predictive accuracy and resolution. Exploration of novel methods such as machine learning approaches may address these limitations for future school-age asthma prediction.

The Journal of Allergy and Clinical Immunology: In Practice 2019 Highlights

This article provides highlights of the clinically impactful original studies and reviews published in The Journal of Allergy and Clinical Immunology: In Practice in 2019 on the subjects of anaphylaxis, asthma, dermatitis, drug allergy, food allergy, immunodeficiency, immunotherapy, rhinitis/sinusitis, and urticaria/angioedema/mast cell disorders. Within each topic, practical aspects of diagnosis and management are emphasized. Treatments discussed include lifestyle modifications, allergen avoidance therapy, positive and negative effects of pharmacologic therapy, and various forms of immunologic and desensitization management. We designed this review to help readers consolidate and use this extensive and practical knowledge for the benefit of their patients.