In the last decades, social and economical changes in Western Countries caused deep changes in the labour market. Particularly, unemployment and flexibility have grown, leading to the new concept of job insecurity.1 In particular, job insecurity is defined as “the discrepancy between the level of job security a person experiences and the level she might prefer”.2

Within the European Union (EU), the number of employed people is still lower than it used to be before the 2008 crisis by 2.2 million workers.3 Specifically, talking about the Italian situation, the occupational rate shows an overall fall of 2.3% in 2008, whereas the occupational rate among the youth population (15-34 years old) is still 16.5% below the EU average rate.3 Data describe a growth of short-term workers within the total employed population in the EU, increasing from 17.5% to 19.6%, between 2008 and 2015.3 Moreover, about 50% of short-term contracts last less than one year and only 20% of them last at least 12 months.3 This trend is more common among the young workers population. Among workers aged 20 to 24, the rate of short-term employees has risen from 19.4% to 45.6% among the men population and from 26.1% to 49.5% among women. We can also see an increase for men and women aged 25 to 29 and 30 to 34 years old.3

The role played by occupation as a health determinant is well known, but the need of a better comprehension of the job insecurity has also emerged. Many studies assessed the spatial and temporal work's flexibility impact on health. However, the evidence on this possible association is not consistent. A meta-analysis considering 27 studies4 has demonstrated the existence of an association between short-term jobs and poor mental health. This association seems to be stronger when the insecurity grows, and it changes according to gender and social background. There are also other studies which not only underline the associations between short-term contracts and poor psychological outcomes, but do also prove a causal connection between them.5–8 At the same time the literature present a variety of results9–11 that could be related to the variety of outcomes assessed in these papers. Other studies show how lower income,12 reduced access to benefits and poorer future perspectives13–15 or gender16 can affect the association. The results are not easily generalized, because each country has its own welfare system, providing different plans of assistance to its citizens. In light of this, it seems to be clear that job insecurity needs to be better assessed as a social determinant of health. Our research aimed to explore the job insecurity as a possible risk factor for poor mental health, alcohol and drugs abuse and smoking habit in the Southern Europe context. This pilot study was used to identify the key points and possibly relevant aspects that can help this assessment.

The team performed a cross-sectional pilot study in the city of Torino, Italy. STROBE guideline for observational studies was used for reporting.17 The participants (18-40 years old) were enrolled by using a convenience sample, recruited in public recreational areas, like concerts club and bars, between April and June 2016. The participants were informed about the study and signed the informed consent. Two researchers performed direct personal interviews. Each interview lasted around 20minutes. Different Italian translated evaluation scales were used to evaluate the presence of mental health problems, drug and alcohol abuse and tobacco addiction.

The final sample size for the pilot study consisted of 504 subjects. In Table 1 the main demographic, occupational and social features of the sample are resumed stratified by gender and perception of job security. Data show how the workers who perceive their job situation as insecure are more often unmarried (69.2% vs. 50.9%; p <0.001). Speaking about family structure and economical independence, it is shown that insecure workers more often live with their parents than the secure workers do (27.2% vs. 18.2%; p <0.001) or do still rely economically on them, even when living on their own (15.8% vs. 3.8%; p <0.001). On the other hand, a larger proportion of secure workers lives on its own and is not only more frequently economically independent (61.2% vs. 51.9%; p <0.001), but it also appears to share more often part of its income with parents and/or relatives (16.8% vs. 5.1%; p <0.001). These differences may be partially explained with the difference in mean age. The mean age in the secure group is 30,05±5,23 years while in the insecure group is 28,01±4,56 years (p <0.001). While no statistically significant differences are shown between secure and insecure workers on the basis of their educational qualification, relevant differences appear to exist among the different kind of job contract in the sample. Analysing the association between familiar and individual annual income groups and the perception of job security or insecurity, Table 1 shows that both groups feel their situation as insecure. By assessing the association between the perception of personal income and job security, it emerged that workers who describe themselves secure are more likely to consider their income adequate (72.4% vs. 51.57%; p <0.001) or excellent (7,6% vs. 4.4%; p <0.001), while, on the other side, perceived job insecurity may lead to consider the income as unsatisfactory (5% vs. 1.4%; p <0.001) or poor (39% vs. 18.6%; p <0.001).

Table 2 describes how the large majority of those who have a permanent contract feel their situation as secure, while this proportion is reduced by about its half when referring to temporary jobs and other kinds of contract (p <0.001).

Table 3 shows the main differences regarding mental problems’ and substances consumption's prevalence between the male and the female population. A statistically significant higher proportion of anxiety disturbance is recorded in the female population (30.8% show mild anxiety, 10.2% moderate anxiety and 4.1% severe generalized anxiety disorder, versus 23.8%, 4.7% and 1.7% respectively for the male population; p <0.001). On the other side, the male population appears to be statistically significant more related towards hazardous alcohol consumption (17.9% vs. 7.9%; p <0.001). The same statistically significant results are shown for drugs abuse (p=0.043).

The perceived job security among the workers interviewed in this study seems to be also related to mental disturbances and substances use and/or abuse: secure workers showed less depression symptoms than insecure workers (p <0.001) as reported in Table 3. The same pattern is noticeable for the presence of anxiety-related symptoms (p <0.001). For what it concerns alcohol consumption and tobacco addiction, we can still find a preponderance among insecure workers rather than secure (p <0.001).

The association between unemployment and mental disturbances in the young population has been widely investigated, especially during the last decade, since the worldwide financial crisis started to display its influence on society. The increase of mental illnesses’ incidence among the unemployed young population appears to be a common trend in all European countries.26 Other studies conducted in Southern Europe countries have not only confirmed this trend, but have also pointed out a major association between unemployment and mood disorders, particularly with depression and anxiety.27–29 Moreover, similar results were found in other non-European countries, suggesting a possible important role of the crisis in the influence of labour market changes and on related health issues across the world.30,31 According to our preliminary results, the perception of job insecurity appears to be related not only to mental disturbances, but also to hazardous behaviours. These findings are confirmed in literature.32,33 In particular, one study found no significant differences between unemployed and subjects with an insecure work status according to the association with mental illnesses and substances abuse.34

Furthermore, this pilot study shows that an adequate or excellent income is a protective factor of depressive/anxious symptoms and alcohol intake, differing from one other study.35

In conclusion, the analysis of the existence of associations within labour market and mental health issues may lead to controversial conclusions. Because of this, it is important to study and consider all the main aspects that could intervene and influence specific outcomes. It is necessary to perform other specific longitudinal studies and to develop a strong advocacy of labour's policy making, coherent with the special needs of this specific portion of young workers.

Moreover, it is important to underline both points of strength and weakness of the study. First of all, few similar studies have been conducted to analyse the young worker's situation after the 2008 financial crisis. All the included studies collected data from different countries, thus making them difficult to compare among each other, since each country has its own welfare system. Moreover, a set of different validated scales was used to strengthen the assessment of the outcomes. On the other hand, the way the sample was selected, and the pilot study design may have led to selection biases and it cannot be considered as fully representative of the population of Italian young workers; at the same time, cross-sectional studies cannot give strong evidence of causal connections among the different variables. Another potential limitation of this and other studies that touch sensitive topics is the risk of underreporting of some important data about mental health problems or substances abuse. Since these topics have been assessed with questionnaires, the interviewed subjects may have omitted some information about their real condition.

This pilot study conducted in a single Italian city provided important information about how to perform more specific multicentre studies with larger sample sizes. These studies could explore this issue in southern Europe, in order to point out potential differences with other economical systems, based on stronger evidence.

The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committees on human experimentation and with the Helsinki Declaration of 1964 and its later amendments.

Carlos Álvarez-Dardet.

The corresponding author on behalf of the other authors guarantee the accuracy, transparency and honesty of the data and information contained in the study, that no relevant information has been omitted and that all discrepancies between authors have been adequately resolved and described.

All the authors substantially contributed in the conception and design of the work, data collection, analysis and/or interpretation of the data. They gave their contribution in writing the article or gave an important intellectual contribution. They all approved the final version of the paper. Furthermore, each of them agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

None.

None.