This study is the first to our knowledge that suggests the existence of associations between the different burnout subtypes (classified according to the degree of dedication to work) and the different sociodemographic and occupational characteristics that are congruent with the definition of each of the subtypes. The results of this work assist the clinical differentiation of subtypes by introducing sociodemographic and occupational variables into the differential burnout model as specific risk factors that are easy to identify. They also facilitate an understanding of the clinical phenomenology of the profiles, encouraging future working hypotheses of a causal nature to be considered among the variables and enabling more specific interventions to be developed for the syndrome.
The variables "number of hours worked per week" and "contract type" showed significance in the adjusted model for the "frenetic" burnout subtype. Those employees who invested more than forty hours per week in their jobs had a greater risk of presenting this type of burnout compared to those working fewer than thirty five hours. The number of hours worked per week was associated directly and linearly with the "frenetic" burnout sub-type in such a way that when the number of hours was increased, so was the risk of developing this burnout profile. This variable seems to be the key factor in the configuration of this profile and could contribute to the development of the syndrome by increasing worker exhaustion levels [15
]. Data regarding contract type show that workers in part-time employment present a higher risk of having this burnout subtype compared to full-time employees. This result may seem contradictory, but this is not the case when we consider that these subjects tend to have several jobs at the same time (e.g., adjunct lecturers and students on traineeships), which is associated with burnout syndrome in general [49
]. All of these results are consistent with what has been put forward in the qualitative works to which we previously referred [1
] and they enable the rapid identification of the burnout profile of concern to us. The significance of guilt feelings in the development and continuation of burnout syndrome [6
] has already been pointed out. Specifically, the "frenetic" subtype feels guilt when faced with the prospect of not achieving set goals, given the ambition and great need for achievement that characterise subjects with this profile [1
]. These individuals adopt active coping strategies and invest all of their possible efforts until they become exhausted and overloaded [17
]. Consequently, the treatment for this profile requires a holistic approach that takes into account the cause of their excessive ambition and their associated guilty feelings, in addition to a reduction of their involvement and lessening of their dedication to work in the interest of satisfying their personal needs.
On the other hand, the variables "occupation" and "gender" were statistically significant in the adjusted model for the "underchallenged" burnout subtype. In our study, the ASP group had a greater likelihood of developing this burnout profile when compared to the TRS group. Burnout can generally occur in all types of occupational groups [50
], but public assistance jobs, such as those performed by ASP group members, seem to have an increased risk [51
]. This risk is possibly due to the antecedent effect produced by the characteristics of this type of work [22
]. It is necessary to take the degree of worker satisfaction into consideration with respect to the characteristics of their jobs in order to address their discontent [52
], as dissatisfied workers present a greater risk of suffering from burnout [31
]. 33]. It is also important to pay attention to worker preferences with regard to the type of work they would like to perform [54
], given that a sustained organisational policy concerning these values improves satisfaction levels and reduces absenteeism in the long term [55
]. With regard to "gender", our study has found that males are at greater risk of suffering from "underchallenged" burnout than females, perhaps owing to the fact that the role of males has always been linked to social expectations of professional development [47
]. Generally, employees with the "underchallenged" profile have to cope with the disenchantment caused by feeling trapped in an occupational activity to which they are indifferent, which bores them and produces no gratification. These employees present a cynical attitude [17
] and are invaded by guilty feelings due to the ambivalence they feel for their work and by their desire for change. These people have lost their objectivity with respect to their natural right to experience needs for personal development and to try to pursue them [9
]. Basic components of treatments for this clinical profile should include restoring balance to this distorted view of their needs by approaching the associated guilty feelings, encouraging a renewal of interest and personal development at work by presenting job-related tasks in a significant light.
Lastly, "length of service", "level of education", "stable relationships" and "having children" were significant factors in the adjusted model for the "worn-out" burnout subtype. Employees with between four and sixteen years of service in the organisation and those with more than sixteen year of service were at greater risk of developing the "worn-out" profile in comparison with those with fewer than four years of service. "Length of service" in the organisation showed a direct linear association with the "worn-out" type, to the extent that the longer the service, the greater the likelihood of having this burnout profile. This variable has a certain ambivalence in its relationship with burnout syndrome in general, given that associations have been found that are both direct [35
], inverse [31
] and even absent [56
]. This contradiction may be due to the differential impacts of the various types of organisations on their employees [57
] and to the personal relations and forms of communication established in the workplace [36
], some of which offer protection from the development of the syndrome, while others induce it. Having a university degree, together with a stable relationship and the presence of children, was seen to be factors that protect from the "worn-out" burnout subtype, which is in line with results obtained in other studies for burnout syndrome in general [33
]. Our results suggest that the prolonged exposure to the environment provided by the organisation that was the object of our study turned out to be a significant risk factor for developing the helplessness characterising the "worn-out" profile. Employees with this profile adopt a passive coping strategy, becoming ineffective in performing work tasks and they feel guilty because they do not fulfil the responsibilities of their post [10
]. For this subtype, consideration is given to the suitability of treating not only the feelings of despair, passive coping and inefficacy that characterise it, but also of intervening in the actual contingency system of the organisation, directing its influence as much as possible towards developing commitment to tasks and encouraging the establishment of a social support network.
Through the analysis of the ROC curves, we have seen that the performance shown by the considered sociodemographic and occupational factors in predicting burnout types is superior to a random classifier. Nevertheless, they are far from being the ideal classifier, which means that it might be worth considering other variables that may be associated with the burnout subtypes, such as personality features or specific coping strategies. We should also not overlook the fact that as values for the considered variables were self-reported, they may have been influenced by socially-desirable responses. This phenomenon may have occurred more particularly in the subscales of involvement and neglect, as dedication to work is quite important in Western culture, dedication to work. Further, given that the minimum values for the former and the maximum values for the latter do not encompass the entire range of possible responses. On the other hand, the cross-sectional design of the study forces us to be cautious when drawing conclusions regarding the aetiology of burnout subtypes. However, confirmation of these types of associations does not come under the scope of this study. The main aim of this work was to ascertain in an exploratory fashion which sociodemographic and occupational variables could be associated with the different burnout subtypes in order to assist in the recognition and understanding of these clinical profiles. This goal does not require that the established associations must be of a causal nature. Nevertheless, the fact that these sociodemographic and occupational variables existed prior to the time of measurement (which implies the fulfilment of the premise of temporal precedence) and evidence of a dose-response relationship (statistically significant p values for linear trend analysis) would support that hypothesis. Therefore, our study makes advancement possible in the generation of new hypotheses that may be subsequently confirmed by means of a suitable research design [38
]. With regard to the representative nature of the sample, we believe that although the response rate obtained may seem low and the distribution by occupational levels may seem uneven, these values are comparable to those found in other studies using the same data collection procedures [40
]. We consider that one strength of this study lies in the fact that the work was carried out with a broad and multi-occupational sample of university employees in positions with very different characteristics, which reinforces the possibility of generalising our conclusions. Additionally, data quality was controlled by eliminating possible errors in the questionnaire transcription process through the use of purpose-designed software.