Original Article
A diagnostic model for the detection of sensitization to wheat allergens was developed and validated in bakery workers

https://doi.org/10.1016/j.jclinepi.2009.10.008Get rights and content

Abstract

Objectives

To develop and validate a prediction model to detect sensitization to wheat allergens in bakery workers.

Study Design and Setting

The prediction model was developed in 867 Dutch bakery workers (development set, prevalence of sensitization 13%) and included questionnaire items (candidate predictors). First, principal component analysis was used to reduce the number of candidate predictors. Then, multivariable logistic regression analysis was used to develop the model. Internal validation and extent of optimism was assessed with bootstrapping. External validation was studied in 390 independent Dutch bakery workers (validation set, prevalence of sensitization 20%).

Results

The prediction model contained the predictors nasoconjunctival symptoms, asthma symptoms, shortness of breath and wheeze, work-related upper and lower respiratory symptoms, and traditional bakery. The model showed good discrimination with an area under the receiver operating characteristic (ROC) curve area of 0.76 (and 0.75 after internal validation). Application of the model in the validation set gave a reasonable discrimination (ROC area = 0.69) and good calibration after a small adjustment of the model intercept.

Conclusion

A simple model with questionnaire items only can be used to stratify bakers according to their risk of sensitization to wheat allergens. Its use may increase the cost-effectiveness of (subsequent) medical surveillance.

Introduction

What is new?

A simple diagnostic questionnaire model for detection of sensitization to wheat allergens was developed and validated. This model will enable stratification of risk of sensitization to wheat allergens among bakers.

In the past few decades, occupational allergic diseases (OADs) including occupational asthma (OA) have been shown to be a major respiratory problem. Many OA cases are related to exposure to high–molecular weight (HMW) allergens, including wheat flour and fungal α-amylase allergens in bakery and flour producing industries [1], [2], [3]. Although highly exposed individuals are more likely to have serious complaints and disability [4], workers with intermittent or occasional exposures may also be affected [5]. Recent studies suggest that there is no reason to assume that the burden of OAD is decreasing. So, prevention should focus on finding new cases, detecting subclinical illness, and early intervention in already asthmatic workers [1], [3].

A genuine diagnosis of OAD, such as OA, can only be made in a particular person at a particular time in a clinical setting. Therefore, identification of OAD in bakers exposed to flour allergens should not focus on clinically established allergic disease but on highly associated preliminary symptoms and signs. Sensitization to allergens is an outcome strongly associated with OA and the most appropriate characteristic that can easily be investigated. Sensitized subjects, when continuously exposed to such allergens, are at high risk of developing occupational allergy outcomes [6]. A logical approach is therefore to first identify sensitized workers. After detecting workers with a high risk of wheat sensitization, additional clinical investigations may be restricted to this group, leaving a considerable number of workers with a low risk in which no or less far reaching medical investigations are needed [7].

For that reason, a diagnostic questionnaire model was developed to estimate the individual probability of the presence of sensitization to wheat allergens. This model enables objective and standardized quantification of the individual probability of wheat sensitization in a cost-effective manner by avoiding as much as possible the use of (invasive) advanced and costly reference tests. With the development of a diagnostic model for sensitization to wheat allergens, the authors aimed to easily detect individuals with a high probability of having OAD [8], [9].

The accuracy of prediction rules is generally lower compared with the accuracy in the data on which the prediction rule is developed [10], [11]. Therefore, the authors evaluated whether the developed diagnostic model performed accurately in new workers, and finally whether the performance of the diagnostic model could be improved (updated) before reliable implementation will be accepted in daily care use by occupational physicians in medical surveillance programs. The so obtained final diagnostic model was then simplified into an easy-to-use scoring rule to enhance its use in practice.

Section snippets

Populations

Data from a study performed in the framework of a National Occupational Respiratory Allergy Surveillance Program among bakery workers in The Netherlands were used to develop the diagnostic model (development set) [12]. Between June 2005 and June 2006, 341 and 28 randomly selected traditional and industrial bakeries, respectively were approached. From these companies, 64.5% (238 of 369) participated in the study, comprising 1,249 workers. Twelve bakeries no longer existed, 116 companies refused

Results

Table 1 shows the distribution of the workers' characteristics in the development and validation sets. Most workers in the development set were traditional bakers (74%) compared with 39% in the validation set. The prevalence of wheat sensitization was 13% and 20% in the development and validation sets, respectively. Table 2 shows the univariable association between the candidate predictors and sensitization in the development set. Nasoconjunctival symptoms, asthma symptoms, shortness of breath

Discussion

A simple diagnostic model for the detection of sensitization to wheat allergens was developed and validated in bakery workers. The model included six items: nasoconjunctival symptoms in the last 12 months, asthma problems in the last 12 months, shortness of breath and wheeze, work-related upper and lower respiratory symptoms, and type of bakery (traditional yes or no). Exposure to wheat allergens was presented by a proxy variable, that is, type of bakery. The authors intentionally did not

Acknowledgments

This study was supported by Productschap GZP, The Hague, and Aladdin program for occupational health research, The Netherlands Ministry Social Affairs and Employment.

References (30)

  • D. Gautrin et al.

    Long-term outcome of a prospective cohort of apprentices exposed to high-molecular-weight agents

    Am J Respir Crit Care Med

    (2008)
  • E. Suarthana et al.

    Predicting occupational diseases

    Occup Environ Med

    (2009)
  • E. Meijer et al.

    A strategy for health surveillance in laboratory animal workers exposed to high molecular weight allergens

    Occup Environ Med

    (2004)
  • E. Suarthana et al.

    Diagnostic model for sensitization in workers exposed to occupational high molecular weight allergens

    Am J Ind Med

    (2005)
  • E.W. Steyerberg et al.

    Validation and updating of predictive logistic regression models: a study on sample size and shrinkage

    Stat Med

    (2004)
  • Cited by (0)

    View full text