gms | German Medical Science

Psychological stress, especially in the field of health professions, has been investigated systematically since 1981. In the model developed by the psychologists Maslach and Jackson, burnout syndrome is determined using the Maslach Burnout Inventory (MBI) and broken down into the sub-categories of emotional exhaustion, depersonalisation and personal accomplishment [1].

In recent decades, an increased risk of developing burnout syndrome was found not only amongst doctors and in particular of dentists [2], [3], [4] but even amongst students of human and dental medicine [5], [6].

The effects of psychosocial stress on students of human and dental medicine have already been investigated several times. These revealed high incidences of burnout syndrome, stress, anxiety and depersonalisation. High anxiety scores, a high incidence of burnout and depersonalization are described in the literature. Some studies indicate that the causes for the underlying stress for students of the above mentioned subjects are especially due to the lack of free time, high examination pressure, the financial realities of studying and the interaction with patients [5], [6], [7], [8], [9], [10], [11], [12].

A publication from 2008 simultaneously studied students of dental medicine at the universities of Manchester, Belfast, Helsinki, Amsterdam and Cork using the MBI. In total 39% of the students displayed emotional exhaustion, 22% depersonalization and 41% had limited personal accomplishment [8].

A study by Pöhlmann et al. [6], which studied a total of 161 students of dental medicine in the 4th and 5th year of study at the universities of Bern, Freiburg and Dresden came to the conclusion that especially depersonalisation was prominent, with 28% of students surveyed showing increased values. This study used the Maslach Burnout Inventory (MBI) as a measuring instrument [1]. Depersonalisation is characterized by impaired and distorted perception of oneself, of others and one's environment and it manifests itself as an affective-symptomatic lack of empathy. Loss of motivation and a sense of isolation can occur [13], [14], [15], [16]. Hunter et al. [17] showed increased depersonalisation amongst Anglo-American students in epidemiological comparative studies, compared with the general population. Depersonalisation can manifest itself as a disease in response to negative and stressful everyday situations [17], [18]. Depersonalisation can adversely affect contact with patients in the form of cynical and negative attitudes. To protect oneself against negative experiences when in direct contact with a patient, the affected practitioner objectifies their counterpart and does not perceive them as a person but as a true object [1], [19]. Because the intensive contact with patients and patient-related aspects are considered causes for an increased perception of stress [8], [12], this could be a possible cause of depersonalisation, anxiety and burnout.

Under guidance, dental students from the 7^th to 9^th semester carry out dental treatment in treatment courses for 20 semester hours. Because patient contact in human medicine clearly differs, our study used students studying human medicine from an appropriate semester as a comparison group to the dental students.

Stress management and coping skills of dental students have already been investigated in several studies [20], [21] but they have not been assessed in connection with the parameters of anxiety, depression, burnout and depersonalisation, which is why the focus of this study will take this into consideration.

In principle, the question arises whether the high values ??for anxiety, depression, burnout and depersonalisation found in the literature are at all detectable at the University of Erlangen-Nuremberg.

Based on the above findings, we propose the following hypotheses:

• A. Dental students tend toward higher values ??with respect to scale values for burnout, anxiety and depression compared to students of human medicine.
• B. The pathological values increase as the students progress to later semesters and more contact with patients takes place.
• C. Functional coping strategies are associated with low scale values. In contrast, students with exceptionally high values ??frequently have dysfunctional coping strategies.

The survey took place in April 2008 at the beginning of the summer semester at the Friedrich-Alexander University (FAU) of Erlangen/Nuremberg. Students of dentistry and human medicine in the 4^th and 5^th academic year were invited to participate in the study. The anonymous survey was carried out using questionnaires once during a lecture and was voluntary, so data is not available for the entire cohort of each semester.

1. Demographic Data

Data on age, year of study, degree course and gender was collected.

2. Burnout Syndrome

The Maslach Burnout Inventory (MBI) [1], [19] is composed of three subscales with a total of 22 items and is considered a standard instrument for measuring burnout syndrome. The subscales emotional exhaustion (EE, nine items), depersonalisation (DP, 5 items) and personal accomplishment (PA, 8 items) are given on a 7-point Likert scale (0=never to 6=every day) and are added up to a total value. The German version of the MBI was used [22].

An increased degree of burnout is present if the total value of the scales emotional exhaustion and depersonalisation are high and a low value for personal accomplishment [1], [19]. The third edition of the Maslach Burnout Inventory contains cut-off values specifically for the medical field. For MBI-EE, this is the sum value of 18, for MBI-DP the sum value if 10 and for the personal accomplishment the value of 33. Values above these cut-offs indicate a high degree of burnout [19]. The internal consistency in the study by Enzmann et al. (1989) for the German version was α=0.82 (emotional exhaustion), α=0.67 (depersonalisation) and α=0.75 (personal accomplishment) [22].

3. Depersonalisation

The Cambridge Depersonalisation Scale (CDS) was translated into German and first used in 2004 by Michal et al. [13], after the first English version of Sierra and Berrios was published in 2000 [23].

The short-scale CDS-9 used in this study is a screening version, which covers the core symptoms of depersonalisation through nine questions from the CDS [13], [24].

The items to be determined on the short scale relate to the last six months and capture frequency (0=never, to 4=constantly) and duration (1=a few seconds longer, to 6=longer than a week) of the individual symptoms. The item scores are summed up to a total value of the nine questions, resulting in achievable values between 0-90.

At a sensitivity of 90.7% and specificity of 87.5%, the study by Michal et al. set the cut-off value at a total score of 19. The short-scale CDS-9 showed adequate internal consistency and temporal stability (α=0.92, test-retest reliability 10-14 days: r_tt=0.86) in this study [13].

4. Anxiety and Depression

The Hospital Anxiety and Depression Scale (HADS) was developed by Zigmond and Snaith as a self-assessment tool in 1983 to detect the level of anxiety and depression [25]. The German version (HADS-D) by Herrmann et al. was used [26]. The HADS-D provides an anxiety scale (HADS-D/A) and a depression scale (HADS-D/D), each with seven items. Each item has four answer grades, resulting in values ??0-3. By adding these up, a sum value for the two scales is obtained. Values ??from 0 to 7 are seen as normal, 8-10 as borderline and sum values ??between 11-21 are considered suspicious.

The internal consistency of the German version (Cronbach's alpha) is at α=00.80 (anxiety scale) and α=0.81 (depression scale). The test-retest reliability of the anxiety scale (up to 14 days: r_tt=0.81) and the depression scale (up to 14 days: r_tt=0.89) is sufficient [26].

5. Coping Strategies

Brief COPE is a short version of the COPE questionnaire and was published in 1997 by Carver [27]. The German version of Brief COPE was translated in 2002 by Knoll in her dissertation [28] and used in 2005 by Rieckmann and Schwarzer in their study [29].

The questionnaire consists of 14 scales with two items each which can be answered using a four-point Likert scale (1=not at all, to 4=very).

To improve the practicalities of using Brief COPE, the items were reduced to three scales in this study. Following this reduction, the three new scales were checked to see if they fit together correctly in terms of logic and content. To achieve this, a factor analysis with varimax rotation was carried out. A three-factor solution shows optimal content structure for answering the questions. The items were then combined to form new scales without weighting.

The first factor therefore consists of items 2, 5, 7, 10, 14, 15, 21, 23, 25 and is known as active-functional coping. This classification covers the areas of active coping, for example items 7 and 23 (“I've acted to improve the situation” and “I've been trying to get advice or assistance from other people”). The second scale consists of items 12, 17, 18, 20, 22, 24, 27, 28 and represents the cognitive-functional coping factor. Items 17 and 28 are representative of this strategy (“I tried to find something good in what happened to me”; “I saw the funny side of it all”). The first and second scale is therefore assigned to the functional coping strategies. Items 3, 4, 8, 11, 13, 26 form the third scale, dysfunctional coping, and reflect dysfunctional coping strategies (for example item 4: “I have taken alcohol or other drugs to feel better” or item 13: “I have criticised and blamed myself”).

For the statistical analysis SPSS 15.0 was used.

For comparison of the two degree courses (Hypothesis A) regarding the clinical scales of the individual questionnaires, a two-factorial MANOVA with gender and degree course as the main factors was conducted. Gender was also included in the analysis because the gender distribution was different in both degree courses.

To compare the stress of dental students in the 7th, 8th and 9th semester (Hypothesis B), one way ANOVA was carried out for the clinical scales.

To show the relation between the individual coping strategies (Hypothesis C) and the results of the clinical scales, a three-factorial MANOVA was carried out. The three factors of the Brief COPE were used in dichotomised form as orthogonal key factors. The correlations were not calculated because they do not take interaction into account. The Maslach Burnout Inventory (MBI), the Hospital Anxiety and Depression Scale (HADS) and the Cambridge Depersonalization Scale (CDS-9) were dependent variables in all multivariate methods.

It total, 182 students participated in the written questionnaire, 109 medical and 73 dental students. The demographic variables are shown in Table 1 [Tab. 1]. The variable “age” showed no statistical relevance and is therefore not listed.

As can be seen in Table 2 [Tab. 2], the values of the used scales were more often outside of the defined standard values for students of dental medicine than for students of human medicine.

The two-factorial MANOVA with group factors for gender (male, female) and degree course (human and dental medicine) showed both a significant effect of gender (F=2.558, df(Hypothesis)=9, df(Error)=147, p=0.008) and the degree course (F=4.653, df(Hypothesis)=9, df(Error)=147, p>0.001). Female students and dental students showed significantly pathological values in comparison with male students and students of human medicine. An interaction between gender and degree course could not be detected. The mean values of the individual scales (see Table 3 [Tab. 3]) showed significant differences between the degree courses on the Cambridge Depersonalization Scale (CDS-9) (F=9.430, df(Hypothesis)=1, df(Error)=155, p=0.003) and on both subscales of the HADS (anxiety: F=21.330, df(Hypothesis)=1, df (Error)=155, p<0.001 and depression: F=4.926, df(Hypothesis)=1, df(Error)=155, p=0.028), with dental students showing higher values.

For all scales, except for “personal accomplishment” of the Maslach Burnout Inventory (MBI), the one-factor ANOVA showed significant differences between students in semester 7 (n=34), 8 (n=35) and 9 (n=38). The students of the seventh semester had the most noticeable values, students of the ninth semester the least noticeable (see Figure 1 [Fig. 1]). The ANOVA results are as follows: MBI emotional exhaustion (F=8.929, df(Hypothesis)=2, df(Error)=93, p<0.001), MBI depersonalisation (F=4.164, df(Hypothesis)=2 , df(Error)=93, p=0.018), Cambridge Depersonalisation Scale (CDS-9) (F=5.091, p=0.008), HADS-anxiety (F=6.046, df(Hypothesis)=2, df(Error)=93, p=0.003) and HADS-depression (F=3.862, df(Hypothesis)=2, df(Error)=93, p=0.024).

In a three-factorial MANOVA, where the three factors of the Brief COPE (both in high and low expression) represent the group variables, it became clear that only the third factor (dysfunctional coping) has a significant effect on each of the clinical scales (F=5.279 to F=17.159, df(Hypothesis)=1, df(Error)=89, p=0.024 to p<0.001). Thus it follows from the significance that the studied students showed a statistical relationship between high values on HADS-anxiety, HADS-depression, the Cambridge Depersonalisation Scale (CDS-9), MBI-emotional exhaustion, MBI depersonalisation, MBI-personal accomplishment and having a dysfunctional coping strategy (see Figure 2 [Fig. 2]).

It was also shown that there is an interaction between the second factor (cognitive-functional coping) and the third factor (dysfunctional coping) (F=2.623, df(Hypothesis)=6, df(Error)=89, P=0.022) but that at the level of the individual scales this cannot be classified as significant. The active-functional and cognitive-functional coping strategies show, in relation to the studied scales, no proven correlation. The internal consistency (Cronbach’s alpha) of the three factors are 0.798 for the first factor, 0.706 for the second factor and 0.661 for the third factor - dysfunctional coping.

The study shows that symptoms of burnout, anxiety, depression and depersonalisation are serious issues both for students of dentistry and human medicine, even if baseline values were not collected before university entry. The high values for the Maslach Burnout Inventory of this study do not match all values from previous studies by other authors [5], [6]. In the paper by Guthrie et al. students of human medicine were studied at the University of Manchester. The proportion of students of human medicine in the clinical section who showed elevated values was 9.6%-12.9% for emotional exhaustion, 14.0%-16.1% for reduced personal accomplishment and 3.8%-7.1% for the depersonalisation scale [5]. These values are significantly lower than those found by us: 22.7%, 37.9% and 25% for the three MBI subscales.

Amongst dental students, the differences between our results and similar investigations are less pronounced. In a study by Pöhlmann et al. at the universities of Dresden, Freiburg and Bern amongst dental students, 10% had non-normal values regarding emotional exhaustion, 17% regarding reduced personal accomplishment and 28% regarding depersonalisation [6]. In the present study, the corresponding percentages of non-normal values were 37.7%, 34.3% and 29.8%. A possible reason for these values being higher than those in the literature regarding reduced personal accomplishment could be the special teaching situation at the University of Erlangen-Nuremberg.

The degree of anxiety and depression, as measured by HADS, can be compared with the publication by Newbury-Birch et al. [11] at the University of Newcastle, which discussed stress amongst students of human and dental medicine. In the fifth year of study, dental students frequently showed elevated values on the subscales anxiety and depression (67%) in comparison to students of human medicine (26%) [11]. This result is consistent with our results. The difference between students of dental and human medicine is most clearly visible on the HADS anxiety scale, with about four times as many dental students (27.5%) displaying anxiety scores above the cut-off value of 11 compared to students of human medicine (6.5%). In line with our hypotheses, the sum values of the investigated scales amongst dental students are significantly higher than amongst students of human medicine. The biggest difference exists in the area of depersonalisation. This result may be partly explained by the intensive patient contact for students of dentistry during early phases of study. The assumption that due to the stresses caused by the extensive treatment courses, the values of the clinical scales for students of later semesters could be higher than those of students of earlier semesters could not be confirmed. The scale values for students of later semesters were significantly lower compared to students of the earlier semesters (see Figure 2 [Fig. 2]). The highest levels of burnout symptoms were measured amongst dental students beginning with the treatment of patients in the seventh semester. The period immediately before or after the start of dental treatment courses would appear to offer itself for counteracting psychosocial stress. Regarding this aspect, as early as 1984 a controlled study by Tisdell et al. showed the positive influence of a corresponding seminar on the stress levels. This seminar taught dental students relaxation techniques, mechanisms for coping with stress and how to organise their treatment appointments [30]. A study by Harvard Medical School in 2002 also was able to demonstrate the positive effect of a similar seminar for students of various disciplines [31]. Psychosomatic and psychological support accompanying clinical courses and seminars could therefore lead to an improvement of the psychosocial situation of students.

The dysfunctional coping strategies correlated positively, as shown in Figure 1, with the values of the scales HADS-anxiety, HADS-depression, CDS-9 and the Maslach Burnout Inventory. Students who showed non-normal values on the clinical scales simultaneously showed high levels of dysfunctional coping. A study by the University of Freiburg shows how relevant this issue is as it shows a relatively high number of students of human medicine with problematic patterns of alcohol consumption, something which is regarded as a dysfunctional coping strategy [32]. Accordingly, it would be desirable to target coping mechanisms to prevent the formation of dysfunctional strategies.

It would also be helpful for students who have already reached critical levels of anxiety, depression, depersonalisation and burnout symptoms to reduce these through appropriate strategies.

In future investigations it would be interesting to conduct a prospective longitudinal study of a student cohort, beginning with a survey at the beginning of the degree course. This could clarify whether the origin of the high levels of anxiety, depression, depersonalisation and burnout syndrome are directly related to the progression of studies.

Our study points to a tendency for dental students to be suffering more frequently from anxiety, depression, burnout and depersonalisation than students of human medicine. The fact this changes the further into the course that students advance and the situation appears to improve but also that targeted teaching of effective strategies could prevent students from developing negative coping strategies. Psychology and psychosomatic medicine in dentistry could therefore play a much greater role.

The authors declare that they have no competing interests.