Health behaviours of education professionals: any room for improvement? A study of French employees

Background

Education professionals have a role to play in health education, both as knowledge providers and role-models. To appraise their health awareness, we compared their health behaviours to those of non-education employees.

Methods

Using inclusion data of the CONSTANCES French cohort (2012-2019) and adjusting for important confounders (sex, age, education,...), we alternately compared three indicators of risky conducts (at-risk drinking, current smoking, past year-cannabis use) and two indicators of unhealthy lifestyle (low physical activity, overweight/obesity) between education professionals (n = 14730) and a random sample of non-education employees (n = 34244). Among education professionals, we distinguished between teachers (n = 12820), school principals (n = 372), principal educational advisers (n = 189), school health professionals (n = 128), and school manual/service staff (n = 1221).

Results

Teachers were less likely than non-education employees to be at-risk drinker, to smoke, to have used cannabis in the past year and to be overweight/obese. Other non-teaching education professionals were also rather less involved in risky conducts than non-education employees. Nonetheless, school principals and principal educational advisers reported more often low physical activity and school principals and manual/service staff were more prone to overweight/obesity than non-education employees.

Conclusions

In this large nationwide sample of French employees, education professionals were rather less involved in risky conducts than other non-education employees with a similar demographic and socioeconomic profile. Yet, non-teaching education professionals showed punctually unhealthy lifestyle indicators, suggesting a window of opportunity to improve both their own health and, indirectly through role-model, that of the youth with whom they interact daily.

Key messages

Education professionals, especially teachers, appear more health-conscious than average. There may still be room for improvement toward a healthier lifestyle. In addition, the average observed may be quite far from the recommended public health target, so that any action to enhance education professional’s health behaviours will have societal benefits.

Use of artificial intelligence to estimate population health indicators in France

Background

The use of artificial intelligence is increasing to estimate and predict health outcomes from large data sets. The main objectives were to develop two algorithms using machine learning techniques to identify new cases of diabetes (case study I) and to classify type 1 and type 2 (case study II) in France.

Methods

We selected the training data set from a cohort study linked with French national Health database (i.e., SNDS). Two final datasets were used to achieve each objective. A supervised machine learning method including eight following steps was developed: the selection of the data set, case definition, coding and standardization of variables, split data into training and test data sets, variable selection, training, validation and selection of the model. We planned to apply the trained models on the SNDS to estimate the incidence of diabetes and the prevalence of type 1/2 diabetes.

Results

For the case study I, 23/3468 and for case study II, 14/3481 SNDS variables were selected based on an optimal balance between variance explained and using the ReliefExp algorithm. We trained four models using different classification algorithms on the training data set. The Linear Discriminant Analysis model performed best in both case studies. The models were assessed on the test datasets and achieved a specificity of 67% and a sensitivity of 62% in case study I, and a specificity of 97 % and sensitivity of 100% in case study II. The case study II model was applied to the SNDS and estimated the prevalence of type 1 diabetes in 2016 in France of 0.3% and for type 2, 4.4%. The case study model I was not applied to the SNDS.

Conclusions

The case study II model to estimate the prevalence of type 1/2 diabetes has good performance and will be used in routine surveillance. The case study I model to identify new cases of diabetes showed a poor performance due to missing necessary information on determinants of diabetes and will need to be improved for further research.