Introduction

Mental health professionals are faced with a variety of stressors that occur in the work environment. In mental health settings, most employees encounter situations in their careers that are uncomfortable or distressing. In general, each individual responds in a slightly different manner to each presented situation; and each individual may report a different level of discomfort to similar situations. A variety of factors might influence a professional's response to stressful situations including past work experiences, professional training, and each individual's personality style. The potential for staff turnover and burnout caused by significant discomfort when dealing with coworkers and patients is high (₁).

In the mental health setting, appropriate interactions between staff members and their patients, coworkers, and supervisors are important for improved quality of services. However, developing the skills to interact appropriately in the work setting can be challenging. In fact, many educational programs for health care professionals do not provide extensive training and preparation regarding interpersonal interactions in the work environment. Additionally, there is little research available to assess which types of interactions might be perceived as most difficult for persons in mental health professions.

Few inventories have been developed to assess difficult interpersonal situations in a number of areas. One group of researchers developed the Gay and Lesbian Oppressive Situations Inventory- Frequency and Effect (GALOSI-F and GALOSI-E) (₂). This inventory was designed to measure negative situations often encountered by gay and lesbian singles and couples. In the area of medical and patient care, the Threatening Medical Situations Inventory (TMSI) was developed to measure cognitive confrontation and avoidance within the area of medical threat (₃). Although not specifically designed to measure difficult interpersonal situations, the TMSI was designed to address differences in coping styles in response to medical situations.

Although scales have been developed to assess perceptions regarding difficult interpersonal situations, few human service researchers have attempted to assess staff perceptions of difficult employment situations. Dunn, Umlauf, and Mermis were perhaps the first researchers to address staff perceptions of difficult work situations. The authors developed the Rehabilitation Situations Inventory (RSI), which was designed to measure staff perceptions of difficult behavioral situations in a rehabilitation setting (₄). The measure had good internal reliability and later subscale development revealed a six-factor structure (₅). Research with the RSI has been conducted with medical students (₆), students in health professions (₇), and rural versus urban college students preparing for human service professions (₈). Findings generally revealed no significant differences between disciplines and scores on the RSI. However, years of employment in disability care produced significant differences between groups on the RSI.

To date, there has been no specific measure developed specifically for mental health professionals to identify which staff-coworker and staff-patient interactions are most difficult. Further, there is no specific outcome-based evaluation that adequately measures a program's ability to modify perceptions of professionals in dealing with difficult interpersonal encounters in psychiatric settings. Outcome-based evaluation is important for many purposes including analysis of a program's effectiveness and cost benefit (₉). The purpose of this study is to present development of a scale designed to measure staff perceptions regarding difficult psychiatric situations, the Psychiatric Situations Scale. The authors sought to develop a reliable measure that serves as a preliminary step in developing an outcome measure for the evaluation of staff training.

Method

Procedure

The first author met with members of several disciplines including psychology (n = 4), social work (n = 4), nursing (n = 4), and psychiatry (n = 1) in order to identify uncomfortable, awkward, or stressful situations that individuals might encounter in the mental health field. Participants were recruited from inpatient and outpatient settings of a private hospital that served persons with mental health needs. A demographic sheet was developed, and 69 statements identified by participants as uncomfortable or distressing were combined to create the inventory.

The Psychiatric Situations Scale then was completed by 115 staff members from 3 major disciplines- medical, mental health, and administrative. Scales were distributed to those who provided inpatient and/or outpatient services in the Midwestern and Eastern regions of the United States. Respondents in all disciplines had significant exposure to persons with mental health needs. Demographic characteristics of the sample according to discipline can be found in Table 1.

Figure 1

Table 1: Demographic Characteristics of the Sample by Discipline

The respondents were asked to rate each statement on a Likert scale anchored by “1 = situation was least disturbing or not applicable to a particular position of employment” to “5 = situation was most disturbing.” Items generally pertained to staff- patient interaction (e.g., patient threatens to kill staff), staff- staff interaction (e.g., coworker frequently calls in sick for his/her shift), or staff-supervisor interaction (e.g., psychologist/psychiatrist overrides healthcare workers' recommendations).

Data Analyses

Results

Reduction of the original 69 items was accomplished by examining the interquartile ranges of each item. Items were eliminated if the item had an interquartile range of less than two between the 25th and 75th percentile. Items with interquartile ranges greater than or equal to two were chosen because they were considered to have sufficient variation to allow for discrimination between groups of respondents. Of the 69 original items, 42 items met this criterion. Remaining analyses were conducted only on the 42 items. The 42-item version of the Psychiatric Situations Scale is provided as an Appendix.

Internal reliability was calculated by two different methods. First internal reliability was calculated using Cronbach's coefficient alpha (α = .93). Then, the split-half correlation coefficient between first and last half of the items of the scale were calculated (r = .86, p < .01). Both methods of calculation revealed good internal reliability of the instrument. Item-total correlations were calculated, as well as Cronbach's coefficient alpha if individual items were deleted. Item-total statistics for the 42 items are reported in Table 2.

Figure 2

Table 2: Item-Total Statistics for the 42-Item Scale

A series of one-way analyses of variance (ANOVA's) were conducted; and Bonferroni adjustments were calculated for multiple comparisons (p =.05/2 = .025). Due to unequal sample sizes between disciplines, preliminary analysis was necessary to address potential violation of homogeneity of variances. The Levene's test was used for preliminary analysis; and results were not significant between variances of Administrative (M = 98.1; SD = 33.3), Medical (M = 126.0; SD = 32.2), and Mental Health (M = 121.3; SD = 24.9) groups on the Psychiatric Situations Scale [F (2, 112) = 1.474, p = .233]. The first ANOVA was then conducted to determine if differences existed between groups on the mean combined total scores of the 42 items of the Psychiatric Situations Scale. A one-way ANOVA was conducted, and results were significant [F (2, 112) = 5.728; p = .004]. Post hoc t-tests were then conducted using the Bonferroni statistic for multiple comparisons. Testing revealed that the Administrative group's total scores on the Psychiatric situations Scale were significantly lower when compared to both the Medical (p = .007) and Mental Health (p = .007) groups. Mean scores between the Medical and Mental Health groups were not significant (p = .921).

The Levene's test was used to confirm the assumption of homogeneity of variances between male (n = 29) and female (n = 86) participants' scores on the Psychiatric Situations Scale. Results were not significant between variances of groups on the Psychiatric Situations Scale [F (1, 113) = .681, p = .441]. One-way ANOVA was conducted in order to determine if there were significant differences between mean scores on the Psychiatric Situations Scale based upon gender. Analysis revealed no significant differences between mean scores based upon gender [F (1, 113) = .004; p = .953].

A series of a priori two-tailed t-tests were conducted to address the hypothesis that those professionals with over 10 years of mental health experience (n = 40; M = 114.3; SD = 29.5) would report significantly lower scores on the Psychiatric Situations Scale (i.e., suggesting less discomfort) than those with less than 5 years experience (n = 37; M = 112.8; SD = 27.8), and those with 5-10 years of experience (n = 37; M = 113.8; SD = 26.3). The comparison between those with less than 5 years experience and those with 10 or more years of experience was not significant [t (75) = -.248; p = .81]. The comparison between those professionals with less than 5 years of experience and individuals with 5-10 years of experience was not significant [t (72) = -.163; p = .87]. Finally, the comparison between those with 5-10 years experience and those with over 10 years experience also was not significant [t (75) = -.091, p = .93].

Table 3 lists items perceived to be the top three most difficult situations by medical, mental health, and administrative disciplines. For medical and mental health professions, the top two items pertained to patient threats of harm to the professionals, other staff, or their families. For the administrative sample, only one of the top two items pertained to threats by patients to harm the employee or their family.

Figure 3

Table 3: Ratings of the Top Three Most Difficult Situations for Each Discipline

Discussion

This paper presents preliminary findings regarding development of an instrument designed to assess the perception of difficult psychiatric situations by mental health professionals. The Psychiatric Situations Scale was developed and tested among members of three mental health disciplines (i.e., medical staff and professionals, mental health professionals, and administrative staff who deal directly with persons who have mental health needs).

One important finding regarding analysis of the instrument was that mean scores on the Psychiatric Situations Scale were significantly lower for the Administrative group when compared to the Medical and Mental Health groups. This finding suggests that of the 42-items comprising the instrument, the Administrative group found the items less distressing/uncomfortable than the other groups. One explanation for this finding is that Administrative professionals were less likely to encounter the psychiatric situations described in the Psychiatric Situations Scale.

There were several limitations of the study. One limitation of the study was that the sample was one of convenience. Additional reliability and validity studies will be necessary in order to determine usefulness of the tool for medical, mental health, and administrative professionals. The Psychiatric Situations Scale could be potentially applicable as an evaluation tool in both inpatient and outpatient psychiatric settings. Also, a professional's perception of a variety of psychiatric situations does not necessarily translate into lack of skill in dealing with a difficult situation. As pointed out in previous research, a professional's skill level or competence at performing job duties improves at a faster rate than perceived comfort with difficult interpersonal situations encountered in the job setting (₄,₁₂,₁₃).

Another limitation of the study was with the scaling of the original 42-item instrument (i.e., scaling of 1 = situation as least disturbing or not applicable to particular position of employment). This might have presented a measurement problem by artificially inflating the variance of the scale. Therefore, the scale is presented in the Appendix with a “not applicable (N/A)” scaling option. Future research is necessary to assess the utility of the new scale with the “not applicable (N/A)” scaling option.

Due to the small sample size, exploratory factor analysis was not conducted to determine what type of factor structure exists. Content of the Psychiatric Situations Scale suggests items pertain to one of three factors including staff-patient interaction, staff-staff interaction, and staff-supervisor interaction. However, exploratory analysis with a large sample is necessary to determine the exact type of factor structure. If different subscales are found, the subscales might provide specific areas of focus for staff training; and the Psychiatric Situations Scale might prove useful in order to further assist with evaluating staff training programs. Furthermore, subscale development could provide important information regarding perceptions of different mental health disciplines.

Future research is necessary to assess whether mental health professionals who go through training to increase competence and reduce discomfort in difficult psychiatric situations will have lower scores on the Psychiatric Situations Scale. Preliminary analysis suggests that the Psychiatric Situations Scale may prove to be useful in the measurement of challenging psychiatric situations, in the face of the mental health professional's comfort.

In conclusion, this study presents the preliminary construction and analysis of reliability for a scale measuring discomfort with situations often encountered in mental health care. The study also provides an important first step in the construction of a scale to assess staff's level of comfort in difficult interpersonal situations. The ultimate objective for the development of this particular scale is to eventually produce an outcome measure for staff training, as well as to identify which situations should be emphasized in staff training.

Appendix

Figure 4

Figure 5

Figure 6

Correspondence to

Glen A. Palmer, Ph.D. Lanning Center for Behavioral Services Mary Lanning Memorial Hospital 715 North St. Joseph Avenue Hastings, NE 68901 E-mail: gpalmer@mlmh.org