Towards a Brief Definition of Burnout Syndrome by Subtypes

Development of the "Burnout Clinical Subtypes Questionnaire" (BCSQ-12)

Jesús Montero-Marín; Petros Skapinakis; Ricardo Araya; Margarita Gili; Javier García-Campayo

Health and Quality of Life Outcomes. 2011;9(74)

Discussion

The BCSQ-12 has been proposed as a definition of burnout that could cover common ground between the typological and standard approaches.^[1,2,4,7,8] Its factor and criterial validity had not been tested until now. By using a multi-occupational sample of university employees, EFA and CFA were performed on different sub-samples, a ROC curve analysis was carried out with the MBI-GS as a standard criterion and a contrast of hypotheses was made for both models with respect to sex and occupation.

The prevalence values obtained for the study sample according to the classical dimensions were high, although within the expected range. The structure of the BCSQ-12 behaved consistently throughout the factor analyses. All the items loaded perfectly on the factors following the original design, and they were all well explained. Internal consistency was very good in all cases and all items contibuted to its increase. The restrictions imposed by the model were well fitted to all the data, from both an absolute and incremental perspective. The discriminatory capacity of the classifier and the accuracy associated with the proposed cut-off points were good. The sensitivity and specificity shown by the dimensions of the BCSQ-12 when predicting those of the MBI-GS do not show the values that we normally expect to obtain from an ideal classifier, however, they are seen to be moderately high and all significant, far from those of random behaviour. Although the likelihood of being a 'non-case' among unexposed subjects offered an excellent score that of being a 'case' among exposed subjects offered a more limited score, which made the misclassification increase in this sense. Nevertheless, the likelihood of being a 'case' among exposed subjects was much greater than those who were not exposed, the likelihood of attaining the status of 'exposed' was greater among the 'cases' and the likelihood of attaining the status of 'unexposed' was greater among 'non-cases'. No significant differences were found with regard to sex, but there were differences depending on occupation. 'TRS' showed higher levels of 'exhaustion' than 'ASP'. 'TRS' and 'TRA' presented higher levels of 'overload' and ASP showed higher levels of 'lack of development'. 'TRA' showed lower levels of 'neglect' than 'ASP'.

As limitations to the study, we should mention that the scores for variables considered were self-reported and therefore may have been weakened by the effects of socially desirable responses. The utilization of a sample obtained from a sole organization may have limited the external validity of the obtained results. Still, this is a broad and multi-occupational sample made up of workers with very diverse jobs, which reinforces the possibility of generalization. Certainly, the RRs obtained with regard to occupation were different and could have introduced a possible selection bias that may have affected the representative nature of the sample. However, we would also mention that this does not produce an important reduction in the statistical power for comparing the groups. We found that teaching and research staff were significantly less participative than administration and service personnel and trainees. Nonetheless, all the response rate values obtained from these groups, although low, fell within the range that could be expected when using this data collection procedure.^[10,11] Our opinion is that this pattern of response could be due to differences in the type of burnout mostly present in each occupational category, which follows the line put forward by Montero-Marín et al.^[4] and is in agreement with the results obtained in this study concerning the differences between groups. The fact that teaching and research staff show a greater tendency to suffer from overload may influence their being less participative, owing to the little time they have and their strong focus on accomplishing their own goals. Administration and service personnel, showing a greater tendency to experience lack of development, would appear to be more participative perhaps as this allows them a momentary break from the monotony of their work. The trainees, showing outstandingly low levels of neglect, appear to be a participative group, most likely owing to the nature of their jobs and to their scarce exposure in time to the rigidity of the organizational structure of the institution, which would leave them feeling less worn out. Consequently, the different response rates obtained depending on occupational categories could be explained in relation to the differences between the burnout types encountered. This point gains in importance if we are to obtain representative samples for the calculation of prevalence values for burnout syndrome depending on the different occupational strata.^[5] Therefore, this will have to be taken into account when recruiting participants in future research projects. Finally, the criterion was established from a psychometric level, given the lack of consensus in the contemporary scene from a clinical perspective. As strengths of the study, we would underscore the quality of the data, which was controlled by eliminating the possible errors from the transcription process by means of purpose-designed software. Likewise, the obtention of convergent results between exploratory and confirmatory analyses, carried out on different sub-samples, increases the confidence of our results.

According to social exchange theory, the establishment of reciprocal relations is essential for the health and well-being of individuals. Perception of the lack of reciprocity in a work environment plays a fundamental role in the development of burnout syndrome and increases the risk of individuals suffering from emotional disorders.^[37–39] This is due to the imbalance between effort and gratification being an important source of stress.^[40] The manifestation of burnout through different clinical subtypes corresponds to coping with feelings of frustration produced through differing levels of commitment.^[3–8]

Individuals suffering from "frenetic" burnout experience the feeling of 'overload' when they try to maximize their rewards by taking on a volume and pace of work that become excessive.^[3–8] This feeling constitutes a classic aetiological factor of burnout,^[41–43] which was observed to be associated with 'exhaustion' in our study. According to Karasek's model, high demands and low autonomy in the workplace increase exhaustion levels and thus the likelihood of developing the syndrome, particularly in workers with poor time management skills and a low level of resources.^[44–46] The "frenetic" subtype offers a profile of active coping that could benefit from interventions directed at reducing activation, for the purpose of removing accumulated tension and preventing exhaustion; improvement in time management to make room for the total satisfaction of personal needs; and development of self-assertion in order to place limits on the acceptance of responsibilities.

The "underchallenged" subtype balances rewards by carrying out tasks in a superficial manner, leading to feelings of meaninglessness and lack of personal development in the workplace.^[3–8] This has an influence on the negative assessment of work conditions,^[47] constitutes a risk factor for burnout^[48,49] and has been associated with boredom, indifference and a mechanical performance.^[8] It has been associated with 'cynicism' in our study. From a non-linear perspective, Karasek's model explains the origin of feeling of frustration as the absence of challenges resulting from monotony owing to low demands in the workplace.^[50] The "underchallenged" subtype, situated between active and passive coping modes although closer to the latter, may benefit from interventions that encourage interest, satisfaction and personal development through training of conscious attention towards tasks and through the establishment of challenging and significant targets.

The "worn-out" subtype optimizes rewards by reducing efforts through 'neglect' of responsibilities and chooses this as a consequence of the defencelessness learned in the individual's experience with the organization.^[3–8] This 'neglect' is the opposite of commitment^[7,51] and is seen in our study to be associated with the perception of 'lack of efficacy' in the carrying out of tasks. According to Karasek's model, experiences of lack of control play an important part in the health of workers and reduce their productivity,^[44,52] leading to a breaking of an individual's commitment through the erosion they cause in expectations of self-efficacy, given the modulating role these play in the maintenance of behaviours.^[53,54] The "worn-out" subtype presents a profile of passive coping that could benefit from interventions directed at treatment for despair and increased confidence through the regaining of control and the perception of self-efficacy.

A definition of the syndrome that is able to discriminate the type of experienced burnout by means of the identification of clinical profiles according to a three-dimensional definition, such as that presented in the BCSQ-12, offers understanding into the type of dysfunctional attitudes associated with each case, favouring the development of more specific interventions within a conceptual framework according to the classical perspective. From our point of view, this is due to the fact that the model provided by the BCSQ-12 extends the standard definition of burnout, allowing greater differentiation to be made using clinical subtypes; but at the cost of becoming a little distanced from the core of the syndrome as it has been considered using the classical model. Extra validity will be given to the proposed model through the clinical benefits that this new definition may produce by means of the design of new and more specific interventions for the syndrome.

Our study shows how the BCSQ-12 went further than the standard MBI-GS in characterizing work-related discomfort experienced with regard to occupation. Taking into account the series of inconsistencies presented by the classic standard,^[55,56] the BCSQ-12 may provide a more solid definition of the syndrome at a structural level. The therapeutic interventions derived from the standard model has not produced very promising results to date,^[57] perhaps because not enough attention has been given to the matter of the type of dissatisfaction and burnout experienced. Generally speaking, the evidence shows that levels of satisfaction in the workplace have a decisive influece on the health of workers.^[58] Future research will need to clarify whether this new perspective will be able to produce more effective interventions for burnout and for the improvement of workers' health status.

1 2 3 4 5 6

Next Section

Cite this: Towards a Brief Definition of Burnout Syndrome by Subtypes - Medscape - Sep 20, 2011.

Table 1. Characteristics of the study participants
variables	total sample n_T = 826	sub-sample 1 n₁ = 413	sub-sample 2 n₂ = 413	p
Age				0.242
Md (SD)	40.26 (9.52)	40.64 (9.59)	39.87 (9,46)
Sex				0.362
male	366 (44.31)	176 (42.62)	190 (46.00)
Stable Relationship				0.999
yes	647 (78.33)	324 (78.45)	323 (78.21)
Education				0.667
secondary	119 (14.41)	64 (15.50)	55 (13.32)
university	423 (51.21)	208 (50.36)	215 (52.06)
doctorate	284 (34.38)	141 (34.14)	143 (34.62)
Occupation				0.988
TRS	372 (45.04)	185 (44.79)	187 (45.28)
ASP	351 (42.49)	176 (42.62)	175 (42.37)
TRA	103 (12.47)	52 (12.59)	51 (12.35)
Length of service				0.210
< 4 years	184 (22.28)	85 (20.58)	99 (23.97)
4–16 years	353 (42.74)	172 (41.65)	181 (43.83)
> 16 years	289 (34.99)	156 (37.77)	133 (32.20)
Contract duration				0.775
permanent	503 (60.90)	254 (61.50)	249 (60.29)
Contract type				0.718
full-time	750 (90.80)	377 (91.28)	373 (90.31)
Sick leave				0.201
yes	256 (30.99)	119 (28.81)	137 (33.17)

The figures represent frequencies, percentages (in brackets) and the p value associated with an χ² contrast between sub-sample 1 and sub-sample 2 except for the age variable where the figures represent means, standard deviations and the p value associated with a t contrast.

Table 2. Exploratory Factor Analysis – weightings and communalities
Items	Factor
Items	1	2	3	h²
3. When things at work don't turn out as well as they should, I stop trying	0.72	0.13	0.07	0.54
6. I give up in response to difficulties in my work	0.85	0.15	0.14	0.76
9. I give up in the face of any difficulties in my work tasks	0.73	0.17	0.14	0.58
12. When the effort I invest in work is not enough, I give in	0.82	0.12	0.09	0.70
2. I would like to be doing another job that is more challenging for my abilities	0.02	0.85	0.05	0.73
5. I feel that my work is an obstacle to the development of my abilities	0.29	0.68	0.22	0.62
8. I would like to be doing another job where I can better develop my talents	0.12	0.92	0.04	0.86
11. My work doesn't offer me opportunities to develop my abilities	0.22	0.72	0.02	0.58
1. I think the dedication I invest in my work is more than what I should for my health	0.07	0.13	0.80	0.67
4. I neglect my personal life when I pursue important achievements in my work	0.09	0.02	0.82	0.67
7. I risk my health when I pursue good results in my work	0.06	0.01	0.77	0.60
10. I overlook my own needs to fulfil work demands	0.20	0.11	0.68	0.52

Extraction method: Maximum Likelihood with Varimax orthogonal rotation on sub-sample 1. h² = communalities. Bold = belonging factor

Table 3. Exactness of BCSQ-12 according to MBI-GS
	criterion: 'exhaustion' (cut-off point 'overload'≥3.38)		criterion: 'cynicism' (cut-off point 'L.development'≥3.63)		criterion: 'inefficacy' (cut-off point 'neglect'≥2.63)
	index	95% IC	index	95% IC	index	95% IC
Sensitivity *	75.89	70.07–81.72	70.71	65.21–76.22	71.19	64.23–78.14
Specificity *	62.35	58.40–66.30	70.57	66.66–74.48	67.03	63.33–70.72
PPV^a *	42.82	37.83–47.81	55.15	49.87–60.44	37.06	31.78–42.34
NPV^b *	87.44	84.19–90.69	82.48	78.93–86.06	89.51	86.68–92.33
OMR^c *	33.98	30.69–37.27	29.38	26.22–32.55	32.08	28,84–35.33
PLR^d	2.02	1.78–2.29	2.40	2.07–2.79	2.16	1.87–2.49
NLR^e	0.39	0.30–0.49	0.42	0.34–0.50	0.43	0.34–0.55
OR^f	5.25^g	3.62–7.60	6.77^h	4.79–9.57	5.21ⁱ	3.57–7.60

*values given as percentages. a = Positive predictive value. b = Negative predictive value. c = Overal misclassification rate. d = Positive likelihood ratio. e = Negative likelihood ratio. f = Adjusted Odds Ratio by means of multivariate logistic regression models controlling age, sex, stable relationship, education, occupation, length of service, contract duration and contract type. g = Wald p < 0.001; H-L p = 0.451. h = Wald p < 0.001; H-L p = 0.093. i = Wald p < 0.001; H-L p = 0.216. Values obtained from the total sample (n_T).

Table 4. Descriptive statistics for BCSQ-12 items
items	Mn	SD	Q₁	Mdn	Q₃	min	max	asym^a	kurt^b	Item-rest
1. I think the dedication I invest in my work is more than what I should for my health	3.83	1.66	3.00	4.00	5.00	1.00	7.00	0.09	-0.80	0.75
4. I neglect my personal life when I pursue important achievements in my work	3.10	1.71	2.00	3.00	4.00	1.00	7.00	0.57	-0.56	0.75
7. I risk my health when I pursue good results in my work	3.43	1.69	2.00	3.00	5.00	1.00	7.00	0.33	-0.73	0.74
10. I overlook my own needs to fulfil work demands	3.53	1.63	2.00	3.00	5.00	1.00	7.00	0.21	-0.73	0.69
2. I would like to be doing another job that is more challenging for my abilities	3.42	1.86	2.00	3.00	5.00	1.00	7.00	0.31	-0.90	0.77
5. I feel that my work is an obstacle to the development of my abilities	3.08	1.64	2.00	3.00	4.00	1.00	7.00	0.61	-0.29	0.72
8. I would like to be doing another job where I can better develop my talents	3.68	1.86	4.00	4.00	5.00	1.00	7.00	0.14	-1.01	0.82
11. My work doesn't offer me opportunities to develop my abilities	3.53	1.86	2.00	3.00	5.00	1.00	7.00	0.30	-0.96	0.73
3. When things at work don't turn out as well as they should, I stop trying	2.46	1.26	1.00	2.00	3.00	1.00	7.00	0.92	1.08	0.61
6. I give up in response to difficulties in my work	2.36	1.24	1.00	2.00	3.00	1.00	7.00	0.88	0.90	0.74
9. I give up in the face of any difficulties in my work tasks	2.12	1.09	1.00	2.00	3.00	1.00	7.00	1.05	1.84	0.68
12. When the effort I invest in work is not enough, I give in	2.48	1.20	1.00	3.00	3.00	1.00	7.00	0.69	0.64	0.74

Mn = mean. SD = standard deviation. Mdn = median. Q₁ = percentile-25. Q₃ = percentile-75. min = minimum score. max = maximum score. asym = asymmetry. kurt = kurtosis. Item-rest = correlation coefficient item-rest (r between each item and the remaining items belonging to the same factor). a = typical asymmetry error = 0.08 for all items. b = typical kurtosis error = 0.17 for all items. Values obtained from the total sample (n_T = 826).

Table 5. Descriptive statistics, Cronbach's α and contrasts with regard to sex and occupation for the BCSQ-12 and MBI-GS dimensions
	(n)	BCSQ-12			MBI-GS
	(n)	Overload	L. Development	Neglect	Exhaustion	Cynicism	Efficacy
Total	826
Mn		3.47	3.43	2.35	2.24	2.01	4.47
SD		1.42	1.57	1.00	1.42	1.57	0.97
Mdn		3.25	3.25	2.25	2.00	1.50	4.58
Q₁		2.50	2.25	1.50	1.20	0.75	3.83
Q₃		4.50	4.50	3.00	3.20	3.00	5.17
min		1.00	1.00	1.00	0.00	0.00	0.00
max		7.00	7.00	6.25	6.00	6.00	6.00
asym^a		0.34	0.28	0.48	0.71	0.78	-0.72
kurt^b		-0.50	-0.62	0.06	-0.14	-0.23	0.71
α		0.87	0.89	0.85	0.91	0.92	0.82
Male	366
Mdn		3.25	3.50	2.25	1.80	1.75	4.50
Q₁		2.50	2.25	1.50	1.00	1.00	3.83
Q₃		4.50	4.62	3.00	3.00	3.00	5.17
α		0.86	0.88	0.86	0.91	0.91	0.81
Female	460
Mdn		3.25	3.25	2.50	2.00	1.50	4.67
Q₁		2.50	2.25	1.50	1.00	1.00	3.83
Q₃		4.50	4.25	3.00	3.20	2.94	5.17
α		0.88	0.89	0.84	0.92	0.92	0.83
p ^c		0.502	0.082	0.480	0.194	0.108	0.124
TRS	372
Mdn		3.75	3.00	2.25	2.00	1.50	4.50
Q₁		3.00	1.75	1.50	1.40	0.75	3.83
Q₃		5.00	4.00	3.00	3.60	3.00	5.00
α		0.87	0.86	0.84	0.92	0.92	0.82
ASP	351
Mdn		3.00	4.00	2.50	1.80	1.75	4.67
Q₁		2.25	3.00	1.50	1.00	1.00	4.00
Q₃		3.50	5.00	3.00	2.80	3.00	5.17
α		0.85	0.90	0.86	0.90	0.91	0.82
TRA	103
Mdn		3.50	3.00	2.00	2.00	1.50	4.50
Q₁		2.50	1.75	1.25	1.00	0.75	3.67
Q₃		5.25	4.00	2.75	3.40	2.75	5.50
α		0.87	0.91	0.86	0.93	0.94	0.85
p ^d		< 0.001	< 0.001	0.016	0.006	0.305	0.155
TRS vs ASP		< 0.001	< 0.001	0.322	0.001	0.123	0.056
TRS vs TRA		0.456	0.622	0.023	0.466	0.786	0.344
ASP vs TRA		< 0.001	< 0.001	0.004	0.202	0.501	0.863

Mn = mean. SD = standard deviation. Mdn = median. Q₁ = percentile-25. Q₃ = percentile-75. min = minimum score. max = maximum score. asym = asymmetry. kurt = kurtosis. a = typical asymmetry error = 0.08. b = typical kurtosis error = 0.17. c = Mann-Whitney contrast. d = Kruskal-Wallis contrast.

Comments

Commenting is limited to medical professionals. To comment please Log-in.

Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.

processing....

Towards a Brief Definition of Burnout Syndrome by Subtypes

Development of the "Burnout Clinical Subtypes Questionnaire" (BCSQ-12)

Discussion

Tables

Tables

Tables

Tables

Tables

References

Authors and Disclosures

Authors and Disclosures

Comments

Email This

Feedback