Since at least the late 1970s, “flexible production” has commonly been considered as a positive and necessary innovation to ensure sustainable economic growth.^1,2 The need to be “flexible” has been proposed for workplace technical systems, schedules and salaries, and “flexibility” has even been recognised as a positive feature of a worker’s personality. Increasing labour flexibility means reducing the constraints on the movement of workers into and out of jobs previously constrained by labour laws, union agreements, training systems or labour markets that protect workers’ income and job security.³

Within this context, one of the best-known outcomes of labour market flexibility has been the growth of “atypical” forms of employment and the decline of the “standard” full-time, permanent jobs. Thus, the standard full-time permanent job with benefits is now often replaced with different forms of non-standard work arrangements such as contingent, part-time contract, unregulated underground work or home-based work, many of which are characterised by variable work schedules, reduced job security, lower wages, hazards at the workplace and stressful psychosocial working conditions.⁴

There are a number of reasons why public health researchers should be concerned about the growth of non-standard employment relationships.⁵ Workers in flexible jobs share many labour market characteristics (eg, lower credentials, low income, women, immigrant and non-white) with the unemployed, while themselves experiencing bouts of unemployment, a factor strongly associated with adverse health outcomes.^6,7 In addition, evidence suggests that these new types of work arrangements can be as dangerous as traditional unemployment for workers’ health.^8,9

In the last decade, the main focus in the study of flexible work has been job insecurity,¹⁰ a concept defined as “the discrepancy between the level of job security a person experiences and the level she might prefer”.¹¹ Studies of self-reported job insecurity after plant closings present consistent evidence that job insecurity can have significant adverse effects on self-reported physical and mental health.¹² For example, results show that relative to workers who remained in secure employment, self-reported morbidity was higher among workers reporting insecurity in their jobs. Workers exposed to chronic job insecurity had the highest self-reported morbidity, indicating that job insecurity might act as a chronic stressor. Despite these findings, knowledge on the health effects of flexible work is still scarce and many questions remain unanswered. In fact, more research on subjective job insecurity and other perceptions of the psychosocial work environment only provide a partial picture of these new employment relationships and is insufficient to explain how new work arrangements affect the health of the flexible workforce. For example, self-perceived job insecurity may not be able to capture the effect of workplace structural determinants (eg, lack of unionisation, benefits or domination) on workers’ health.¹³

In that sense, the analysis of temporary employment and fixed-term contracts has proved to be informative.¹⁴ Temporary workers are often exposed to strenuous and tiring positions, intense noise and repetitive movements, have less freedom to choose when to take personal leave¹⁵ and are seldom represented in health and safety committees.¹⁶ There is also evidence that non-permanent workers have less information about their work environment,¹⁷ enjoy less job autonomy and control over schedules than workers on permanent contracts, are likely to be occupied in less skilled jobs ¹⁸ and experience worse health outcomes compared with permanent workers.^19–21 However, temporary work only refers to contracts of expected limited duration, whereas flexible, contingent or non-standard contracts do not necessarily provide an inferior status to permanent jobs. For example, in the US about one-third of non-standard jobs such as professional consultants or self-employed people have above average incomes.²² Employees who work in a temporary capacity “voluntarily” are likely to be more interested in characteristics of the job that enhance their quality of life, whereas employees who “involuntarily” work in a temporary position are likely to be more dissatisfied than their permanent coworkers.²³

The limitations of the approaches presented above highlight the need to develop conceptual and measurement alternatives based on the social structure of work organisation.³ Probably the most important alternative can be found in the concept of “precarious employment”, a sociological construct,^24–29 generally defined as the lack of regulations that support the standard employment relationship, making workers more vulnerable.³ Precarious workers are likely to work under different power relationships than those in standard jobs, with limited rights at work.³⁰ In fact, the popular meaning of control under precarious employment relationships can go beyond the notion of “decision authority” and create new types of uncertainty in expectations regarding issues such as future work, income, benefits or schedules. Precarious work is thus located on a continuum, with the standard of social security provided by a standard (full-time, year-round, unlimited duration and with benefits) employment contract at one end and a high degree of precariousness at the other. Historically, precarious employment was once common but declined in the developed countries with increased government regulation and political influence of labour. Currently, precarious employment is becoming more common in developed countries and is widespread in developing economies. A particular form of precarious employment occurs in the informal economy, which poses significant health risks because working conditions are unregulated. The millions of workers who today, mainly in poor regions and countries, are under situations of slavery (at least 12 million people worldwide)³¹ or the problem of child labour (about 250 million)³² are extreme cases of precarious employment. Thus, precarious workers may experience adverse health effects through material and social deprivation, may be more exposed to hazardous work environments and the effects of precarious employment may extend to family members and dependents.

Current conceptual and empirical scholarship on precarious employment and health is still limited. Precarious employment can be defined as a multidimensional construct defined according to four dimensions: continuity (ie, temporality), vulnerability (ie, powerlessness), protection (ie, limited benefits) and income (ie, low level of earnings).^33,34 Nevertheless, the specific dimensions are not yet well defined and detailed studies have not been conducted. Although some problems have already received attention, many significant issues such as the relative lack of information on gender have also received insufficient attention,³⁵ partly because they are related to emerging changes in labour markets, partly because progress requires additional effort developing conceptual clarity and valid measures and partly because precarious workers are hard-to-reach populations. Although available psychosocial models (ie, demand/control and effort/reward factors) have been proposed as two major means to explain the relationship between work environment and health, these models may not be able to capture other more distal social and organisational determinants of health.³⁶ Sociological and public health studies of precarious employment suggest that it may be necessary to take into account other objective and structural social factors related to inequalities in power and class relationships.^37,38

Indeed, the study of precarious employment and health is still in its infancy. To develop a new research agenda on this topic, a series of fundamental challenges need to be addressed at various key levels: gathering of quality data within improved information systems, clarification of precarious employment dimensions, development of causal theories and pathways, and creation of instruments capable of measuring the mechanisms through which precarious employment may damage workers’ health. Finally, future research should be able to capture multiple situations of precariousness associated with flexible employment in different social contexts and for different types of workers.^2,9,38