Performance of the Mexican nursing labor market: a repeated cross-sectional study, 2005–2019

According to the classic perspective of markets, supply and demand of labor are balanced when the maximum number of individuals is engaged in the existing market spaces–whether public or private. However, supply and demand forces are dynamic, rendering it difficult to hold an equilibrium in markets. Unemployment, a classic indicator of disequilibrium, reflects an over-abundant labor force, or an excessive supply of labor, relative to demand. In professional markets, underemployment provides a key indicator as well. While the latter denotes low working hours and incomes in general labor markets, in professional markets, it is understood as the volume of professionals who find no work in their corresponding markets and end up working in areas unrelated to their professional expertise.

As mentioned above, precarious employment is a growing phenomenon among health professionals. This indicates a greater level of disequilibrium in the market since it deprives salaried workers of labor conditions adequate for fulfilling their tasks commensurate with their educational level [13]. Although the nature of precarious employment differs from that of unemployment and underemployment, it is important to understand how it contributes to market imbalance.

The proposed performance assessment is relevant in that it offers a comprehensive perspective of the labor market. Performance should not be understood merely as the capacity to balance the supply and demand of health workers. It should also be seen as the ability to provide workers with conditions that allow them to function as a key element for achieving the objectives of health systems and institutions. From the supply standpoint, these conditions involve the training of students at all levels; on the demand side, they concern the assurance of protected labor conditions, in accordance with international conventions and national regulations.

The deterioration of professional labor markets is subject to exogenous and endogenous structural determinants that interact with one another despite their different natures [14]. The first relates to a transformation in global economic relations as powers compete for capital reproduction spaces across the globe. This phenomenon has permeated all economic sectors, with recent emphasis on service production. It is in this particular context that health services and their associated endogenous determinants evolve. In low- and middle-income countries, the incorporation of capital has managed to insinuate itself into health systems of all types. To a varying degree, health system reforms facilitate the incorporation of private capital to finance the production of highly profitable services. This has modified labor relationships in the production of both private and public services as efforts have shifted their focus to the need for maximizing capital reproduction, minimizing costs and reducing the budgets of the public sector [15]. The effects of these exogenous determinants can be assessed by measuring the aforementioned indicators of unemployment, underemployment, and precarious employment, with the latter understood as a summary measure of specific job characteristics such as salary, benefits, excessive work and contractual instability.

Nursing professionals in Mexico are trained at three educational levels: (a) auxiliary, (b) technical and (c) university. The first two were initiated over a century ago in hospitals associated with religious orders and were later assumed by the government. Public and private hospitals have continued training nurses to ensure an ad hoc production of personnel that meets their specific needs. Social Security institutions provide exclusively technical training including technical specialization programs. Aspiring nurses began attending university five decades ago, with the number of graduates growing exponentially over the past 20 years. Those who earn a graduate degree can pursue postgraduate (master and doctorate) credits and clinical specialization [16]. Demand for nursing graduates is segmented by educational level and the type of institution requiring personnel. Public institutions follow various recruitment regimes. Some, including the Ministry of Health, have opened their doors to university graduates as their production has grown, and have modified their structure of job and salary codes somewhat to distinguish among nursing educational levels [17]. Others, such as the Social Security institutions, prefer technical personnel who scale through the labor code structure as they acquire higher training [18]. Although the latter have begun to engage a larger number of university professionals in recent years, they have yet to adjust their structure of job and salary codes to differentiate technical from university-level personnel. Private hospitals are also incorporating more university professionals; however, they lack a common structure of job and salary codes and tend to offer precarious employment conditions [19].

The performance of professional labor markets is often assessed on the basis of individual indicators, primarily unemployment; underemployment is evaluated to a lesser degree. Both reflect the extent to which specific markets have the capacity to incorporate personnel with the corresponding level of training [20]. A third indicator, precarious employment, reflects not the incapacity of markets to integrate individuals in areas corresponding to their professions, but their incapacity to engage them in such areas under protected and stable labor conditions. Furthermore, inadequate labor conditions have also been associated to reduced quality of care and patient safety [21, 22]. The set of indicators used for assessing the performance of professional markets therefore needs to be integrated and tracked over time. At present, analyses of the health labor market use indicators such as salary, work schedule, geographic distribution and productivity [23]. Indicators of labor waste are not considered even though this factor prevents thousands of workers from participating in the markets for which they were trained. Precariousness is also ignored given its recent incorporation into analyses of this type [24].

We performed a repeated cross-sectional analysis of secondary data on the Mexican nursing labor market collected between 2005 and 2019. We drew the information from the National Survey of Occupation and Employment (ENOE by its initials in Spanish), which is a public, cross-sectional, probabilistic and retrospective population-based survey representative at the national and state levels (Mexico has 32 states), including rural/urban strata. It is the main data source used for generating official labor indicators and their individual and household socioeconomic and demographic correlates. This survey is administered quarterly to individuals 15 years or older residing in more than 100,000 households [25].

This study did not involve human participants and the data for the analysis were requested and obtained from the surveys public repository hosted at the INEGI (see in https://​www.​inegi.​org.​mx/​programas/​enoe/​15ymas/​#Microdatos). Therefore, the approval for this secondary analysis was not obtained from any Ethics, Research or Biosecurity Committee.

We assessed the influence of the labor sector on the likelihood of having non-precarious work by estimating two groups of repeated cross-sectional, multiple logistic regression models [30‐32]. The first group referred to the sample of underemployed nurses and other workers in the health sector, while the second pertained to the sample of health-sector employees. We constructed three models for each sample, differentiated by means of adjusted covariates: the first model was adjusted only by quarter (Trim) and survey year (Survey) as well as by the relationship between the labor sector (Sector) and survey year; the second was further adjusted by the individual characteristics mentioned above; and the third adjusted model (the most comprehensive) included the contextual features. The three models were specified as follows:where p represented the occurrence probability of the event of interest, and Z a vector that included the sociodemographic, labor and residence characteristics mentioned above for labor sector k, measured for each nursing professional j surveyed. β′s and \({\varepsilon }_{j}\) represented a vector of parameters to be estimated and the error term, respectively. Post-estimate and goodness-of-fit tests showed that the regression models exhibiting a good fit for the data had been correctly specified. Lastly, we adjusted the prevalence rates and corresponding 95% CIs of non-precarious health-sector employees by survey period and socioeconomic region of residence.

During the period analyzed, the unemployment rate among our sample of nursing professionals varied by 5%, the percentage of those employed in the health sector diminished by 10.5% (dropping from 68.6% in 2005–2006 to 61.4% in 2019), and the percentage of those underemployed grew by 26.2% (rising from 26.7% in 2005–2006 to 33.7% in 2019) (Fig. 1).

During the same period, 25.2% of the study population self-reported precarious underemployment and 4.9% non-precarious underemployment (Table 1). 28.7% of the sample self-reported non-precarious and 36.6% precarious employment conditions in the health sector. More men than women reported being precariously (29.7% vs. 24.5%) and not precariously (6.7% vs. 4.6%) underemployed. Conversely, a smaller percentage of men reported being employed in the public and private health sectors, whether precariously (25.0% vs. 29.5%) or not (31.9% vs. 37.3%). The unemployment rate decreased with the age of the nursing professionals; this was in contrast to what we observed with non-precarious underemployment, where the percentage of underemployed nurses increased with age.

In addition, we observed a U-inverse relationship between the percentage of nurses employed in the health sector under non-precarious conditions and age: 15.1% (95% CI: 12.8%-17.4%) among individuals 24 years old or younger, 41.9% (95% CI: 40.5%-43.3%) in those 25 to 54 years old, and 30.7% (95% CI: 26.5%-39.6%) in those over 54 years old. Surprisingly, the percentage of unemployment was similar for nurses who had received a university education (4.4%, 95% CI: 3.8%-5.1%) and those who reported having a technical background (4.7%, 95% CI: 3.9%-5.5%). This was not the case regarding the percentage of people in the other job categories. In particular, the percentage of nurses employed in the health sector under non-precarious conditions was 23% higher (32.7% vs. 40.3%) for those with university-level education. Precarious underemployment was more prevalent in the private (57.9%) than in the public (5.0%) sector. However, the percentage of nursing professionals employed under precarious and non-precarious conditions was higher among those employed in the public (36.2%, 95% CI: 34.6%-37.7% and 55.1%, 95% CI: 53.5%-56.6%) as opposed to the private (21.2%, 95% CI: 19.7%-22.8% and 13.8%, 95% CI: 12.5%-15.1%) sector, respectively. The prevalence rates of both non-precarious underemployment and non-precarious employment in the health sector increased according to the degree of urbanity of the place of residence and the level of wealth of the region in which the nursing professionals resided (Table 1).

We also observed that the percentage of employees in the private sector working under highly precarious conditions had grown notably over the last fifteen years, from 19.5% in 2005–2006 to 31.2% in 2019 (Fig. 2a). On the other hand, the percentage of employees working in non- or low-precarious conditions in the private sector decreased by 22.5% (from 60.8% in 2005–2006 to 47.1% in 2019, P < 0.001). The situation was different in the public sphere (Fig. 2b). Although the percentage of nurses employed in highly precarious conditions practically doubled (from 3.8% in 2005–2006 to 7.3% in 2019), the percentage of those employed in non- or low-precarious conditions fell by 11.5% (from 89.9% in 2005–2006 to 79.6% in 2019).

The aOR of having a non-precarious job for nursing professionals who were underemployed or employed in the health sector, were higher among public-sector workers (aOR 7.11, 95% CI: 4.91–10.31). These odds ratios dropped considerably when the estimates were limited to those who reported being employed (aOR 2.17, 95% CI: 1.36–3.46) (Table 2). Among both female and male working nurses, the percentage of those employed in non-precarious conditions was notably higher in the public than in the private sector and increased according to the levels of wealth of their regions of residence, although this difference has been narrowing in recent years (Fig. 3a). Correspondingly, the percentage of people employed under non-precarious conditions in the health sector grew according to the level of regional wealth (Fig. 3b). These models showed that although this percentage had been declining in the public sector, it tended to equal that estimated for the private sector, especially at the end of the period analyzed.