The Maslach Burnout Inventory is not a measure of burnout

For decades, the Maslach Burnout Inventory (MBI) has been regarded as the “gold standard” for the assessment of burnout [1, 2]. The instrument has been employed in thousands of studies and remains, by far, the most widely used measure of the entity. The MBI is purported to assess the three defining components of burnout, namely, exhaustion, cynicism, and inefficacy [1]. The definition of burnout attached to the MBI has been so influential that the World Health Organization adopted it in 2019 [3]. In the present paper, we demonstrate that, despite its popularity and referential role in burnout research, the MBI fails to measure the construct it is supposed to measure. On a deeper level, the problems affecting the MBI suggest that the idea of burnout may be misconceived.

Burnout has been described as a syndrome of exhaustion, cynicism, and inefficacy [1, 3]. By definition, a syndrome refers to a set of co-occurring signs and symptoms reflective of a given condition [4]. At first glance, the MBI looks consistent with this characterization. The MBI manual underscores that exhaustion, cynicism, and inefficacy constitute the “essential elements of burnout” (p. 1) and the “core aspects of the burnout experience” (p. 73) [1]. The MBI developers further emphasize the syndromal quality of burnout by warning against a reduction of burnout to only one of its facets (e.g., exhaustion). Expectedly, the questionnaire comprises items that reference each of the components of the syndrome. The use of terms such as “essential elements,” “core aspects,” and “components” dovetail with the notion that exhaustion, cynicism, and inefficacy are the constituent parts of a larger whole—burnout. However, the apparent correspondence between the conceptualization and measurement of burnout dissolves as one discovers the prescriptions of the MBI developers regarding the use of their instrument. The MBI manual indeed stipulates that “   each respondent’s three MBI scale scores should be calculated and interpreted separately” and stresses that “   responses to MBI items should not be combined to form a single ‘burnout’ score” (p. 4) [1]. The manual further indicates: “It is NOT appropriate to add the three scale scores to create a total burnout score” (p. 23). Such warning messages are repeated no fewer than nine times across the various sections of the MBI manual [1]. Making their prescriptions even clearer, the MBI developers write that “[t]he three dimensions of exhaustion, cynicism and inefficacy     are not so highly correlated as to constitute a single, one-dimensional phenomenon” (p. 72). All in all, the definition of burnout thus appears to be self-contradictory. As a syndrome, burnout is supposed to emerge from the combination of its three components (exhaustion, cynicism, and inefficacy), but such a combination is considered ill-advised and expressly discouraged. Investigators are eventually left with three individual constructs, none of which, according to the MBI developers themselves, can be equated with burnout [1].

Ironically, the creators of the MBI are partially correct when suggesting that exhaustion, cynicism, and inefficacy should be approached separately. The evidence available indeed indicates that the three entities cannot easily be subsumed under a general or higher-order “burnout” factor [5]. In fact, the MBI has repeatedly shown problematic psychometric properties [5–7], including inadequate factorial validity. Though correct when pointing out that exhaustion, cynicism, and inefficacy do not form a cohesive syndrome, the MBI developers overlook key conclusions deriving from such a state of affairs. First, the MBI assesses exhaustion, cynicism, and inefficacy, not burnout. Second, if exhaustion, cynicism, and inefficacy constitute disparate entities that cannot be combined, then burnout has nothing to emerge from and there is no syndrome in sight. The MBI summons burnout without effectively operationalizing it.

The problems affecting the MBI may be unsurprising given the genesis of the burnout construct. Burnout emerged in the mid-1970s as a largely predefined construct [8]. Burnout’s definition was not predicated on robust empirical investigations or sound theorizing, nor did it stem from a systematic review of the literature on stress and health [9–11]. The MBI was developed based on prenotions and anecdotal evidence, and its creation involved a host of arbitrary choices [10, 11]. Interestingly, many investigators viewed burnout as a parody of a construct to which the MBI, as a standardized quantitative measure, offered a scientific veneer [9]. Judging from the current state of the art [8], the skepticism that surrounded the introduction of the burnout construct was not unfounded.

Despite its allure for researchers, the MBI does not assess the construct it is claimed to assess. Given the hegemony of the MBI in burnout research, addressing this matter is important. It is of note that other measures deemed to assess burnout, such as the Oldenburg Burnout Inventory (OLBI) [12], suffer from similar inconsistencies and fail to measure what they are purported to measure. The OLBI assesses exhaustion and disengagement, two separate constructs reflected in two distinct scores. Once more, burnout is nowhere to be found. To conclude, we urge burnout researchers to pay closer attention to the practices that they implement and promote. In addition to being scientifically problematic, these practices involve a wasteful use of our limited research resources and weaken our ability to support our workforce. Occupational health is a central concern that requires solid constructs and dependable indicators. Robust alternatives to the MBI and the OLBI are available [5, 14]; these alternative measures may help us address job-related distress more effectively.