Purpose Results of a study of stress and negative affect levels in postgraduate year 1 (PGY1) and postgraduate year 2 (PGY2) pharmacy residents are presented.
Methods A cross-sectional approach was used. Pharmacy residency program directors received e-mailed invitation letters requesting that they ask their residents to participate in an online survey in 2011. The main study outcomes included evaluation of resident scores on the 10-item Perceived Stress Scale (PSS10) and the Multiple Affect Adjective Checklist–Revised (MAACL-R) anxiety, depression, hostility, and dysphoria subscales.
Results Of the 524 pharmacy residents included in the study, 75.4% were female, 41.2% were under 26 years of age, and 41% reported working more than 60 hours per week. There were no significant differences between PGY1 and PGY2 residents in stress levels, as assessed with the PSS10 (mean ± S.D. score, 19.05 ± 5.96 versus 19.09 ± 5.77). MAACL-R scores for hostility were, on average, higher among PGY2 residents (mean ± S.D., 50.83 ± 10.02) than among PGY1 residents (48.62 ± 8.96), while there were no significant differences in anxiety, depression, and dysphoria levels. Relative to residents who worked 60 or fewer hours per week, those who worked more than 60 hours had higher perceived stress levels as well as higher depression, hostility, and dysphoria scores.
Conclusion Pharmacy residents exhibited high levels of perceived stress, especially those who worked more than 60 hours per week. Perceived stress was highly correlated to negative affect levels.
The numbers of pharmacy residency programs and resident applicants continue to increase. In 2015, the ASHP Resident Matching Program had the highest numbers of available positions and applicants matched since its inception in 1979.1 Pharmacy residency training is intensive and requires residents to work long hours, potentially resulting in stress for residents.
Much attention has been paid to the impact of stress on medical students and medical residents. One study revealed that one third of medical residents in general surgery met criteria for clinical psychological distress.2 Rosen and colleagues3 conducted a study of 47 interns at the University of Pennsylvania School of Medicine and found that stress during first-year medical residency training was associated with chronic sleep deprivation and mood disturbances. Stress and workload were found to affect medical residents’ sexual health and quality of life in another study.4 The impact of stress on medical residents resulted in changes in medical residency regulations and led to an 80-hour workweek restriction intended to decrease emotional exhaustion among residents and improve their quality of life.5,6
The experience of stress as it pertains to pharmacy residents is not known. Relevant studies, which are limited to the pharmacy student population in academic settings, have provided some information about the levels and sources of stress experienced by students.7,8 A better understanding of this topic would enable pharmacy educators and residency directors to assist residents in coping with stress that may negatively impact their performance and quality of life. The study described here was conducted in order to understand the stress and emotional experience of pharmacy residents.
A cross-sectional study was conducted to examine stress and negative affect levels among postgraduate year 1 (PGY1) and postgraduate year 2 (PGY2) pharmacy residents. Invitation e-mails were sent to program directors requesting that they ask pharmacy residents to participate in an online password-protected, institutional review board–approved questionnaire survey during the 2011–12 residency year. Data collection was conducted in February and March 2012, when residents had completed more than 50% of the training year. Data on the ASHP Resident Matching Program indicate that 2,321 pharmacy residency positions were filled during the 2011– 12 training year9; based on that figure, a sample size of at least 330 subjects was deemed sufficient to carry out our online questionnaire study with a confidence level of 95% and a margin of error of 5%. G*Power analyses (G*Power 3.2.1, Heinrich Heine University Düsseldorf, Düsseldorf, Germany) were also conducted to determine the minimum sample size needed to address the research questions. With 95% power, a significance level of 0.05, and a moderate effect size (Cohen’s f = 0.25, the minimum number of participants needed was determined to be 210.
The 10-item Perceived Stress Scale (PSS10) is a free, simple, and validated survey that can be administered easily by residency program directors to assess stress levels among pharmacy residents.
In a sample of pharmacy residents surveyed using the PSS10, those who worked more than 60 hours per week had increased levels of perceived stress and elevated depression, hostility, and dysphoria scores.
By measuring trainee stress levels, residency programs can help prevent negative emotional outcomes that can be detrimental to residents and hinder work performance.
The 10-item Perceived Stress Scale (PSS10), a validated psychological instrument, was used to assess perceived stress among the study subjects.10 The PSS10 contains 10 questions, with answers ranked using a 5-point scale ranging from 0 to 4. The PSS10 assesses stressful experiences as well as responses to stress during the previous month. Total scores range from 0 to 40, with higher scores indicating higher perceived stress levels, and mean scores are used to compare survey samples.
The Multiple Affect Adjective Checklist–Revised (MAACL-R),11 which can be used only by license (Educational and Industrial Testing Service, San Diego, CA), was used to assess negative affect levels among the study subjects. The MAACL-R is a widely used psychometric tool for assessing anxiety, depression, hostility, and dysphoria. The MAACL-R is available in 2 forms. The “In General,” or Trait, form of the MAACL-R captures the personality of study subjects but was not used in our study. The MAACL-R “Today,” or State, form captures how subjects currently feel at a particular point in time; it was used in this study to assess situational emotional states. Raw scores for anxiety, depression, hostility, and dysphoria were calculated from responses to 132 items on the State form. Using the MAACL-R protocol, we converted raw scores to standard t scores. The MAACL-R standard t score was useful in evaluating the relationship of participants’ scores to scores in larger populations or other groups by putting them on a common scale. The MAACL-R survey also includes certain items to help identify “uncooperativeness, random checking, or intentional manipulation” behaviors; this mechanism helps increase the validity of the survey.
The collected data were analyzed using IBM SPSS, version 19.0 (IBM Corporation, Armonk, NY). Descriptive analyses were performed to describe the demographic variables, perceived stress scores, and MAACL-R affect level subscale scores. Preliminary analyses were conducted to examine the data for potential covariates. Specifically, cross-tabulations with Pearson chi-square analyses were conducted to assess the relationships between the categorical demographic variables. Analysis of variance (ANOVA) and multivariate analysis of variance (MANOVA) were also conducted to examine the effect of the demographic variables on the perceived stress scores and MAACL-R affect level subscale scores. Preliminary analyses also included Pearson product–moment correlations to examine the relationship between the continuous variables. An independent-sample t test was conducted to examine the effect of residency year on perceived stress scores. A 1-sample t test was used to compare the perceived stress scores with norms, as measured in an L. Harris Poll,10 and with scores in previously published literature. An ANOVA and a MANOVA were conducted to examine the effect of residency year on MAACL-R anxiety, depression, hostility, and dysphoria subscale scores. A Pearson product–moment correlation was conducted to examine the relationships between perceived stresses and affect levels and between 2 residency years. A multiple linear regression was performed to predict the main variables (i.e., perceived stress scores and negative affect levels) from the demographic variables (i.e., sex, marital status, age, and hours worked per week).
Among the 2,321 pharmacy residents enrolled in U.S. residency programs in 2011, a total of 643 PGY1 and PGY2 residents participated in the study (a response rate of 27.7%). There were 95 participants who did not complete the survey questionnaire; as a result, they were excluded from the study. There also were 12 participants who completed the survey but whose responses did not meet the validity criteria of the MAACL-R protocol due to uncooperativeness, random checking, or intentional manipulation. There were 524 participants included in the study. As shown in Table 1, majorities of participants were PGY1 (versus PGY2) pharmacy residents, female, and single. Relationships between the demographic variables and pharmacy residency year are shown in Table 2. Residency year was significantly related to resident age. Compared with PGY1 pharmacy residents, a greater proportion of PGY2 pharmacy residents worked over 60 hours per week.
Perceived stress scores are shown in Table 3. Participants’ stress scores ranged from 3 to 37, with a mean ± S.D. perceived stress score of 19.06 ± 5.90. Residency year did not have a significant effect on participants’ PSS10 scores. Because number of hours worked per week (40–60 hours versus more than 60 hours per week) and age (20–25 years versus 26 years or older) were related to pharmacy residency year, these variables were included in a multiple linear regression to identify factors predictive of PSS10 scores. As shown in Table 4, number of hours worked per week was a significant (p = 0.001) predictor of perceived stress, with participants who worked more than 60 hours per week being significantly more likely to have higher stress levels. Age and pharmacy residency year were not significant predictors of PSS10 scores.
As shown in Table 5, pharmacy residency year had a significant (p = 0.017) effect on MAACL-R hostility scores. PGY2 pharmacy residents had significantly higher MAACL-R hostility scores than PGY1 pharmacy residents. As shown in Table 6, number of hours worked per week was a significant predictor of dysphoria scores (β = 0.119, p = 0.007), indicating that those who worked more than 60 hours per week were significantly more likely than those who worked 40–60 hours per week to have higher MAACL-R dysphoria scores.
The relationship between PSS10 and MAACL-R scores was examined for the entire study population and for PGY1 and PGY2 pharmacy residents separately. As shown in Table 7, for the full sample, PSS10 scores were positively correlated with anxiety, depression, hostility, and dysphoria scores (β values ranged from 0.437 to 0.611, p < 0.001 for all variables), indicating that participants with higher PSS10 scores tended to have higher anxiety, depression, hostility, and dysphoria scores. Relative to PGY1 residents with lower stress scores, PGY1 pharmacy residents with higher stress scores tended to have higher anxiety, depression, hostility, and dysphoria scores (r values ranged from 0.413 to 0.601, p < 0.001 for all variables); this was also true for PGY2 pharmacy residents (r values ranged from 0.513 to 0.646, p < 0.001 for all variables).
Our study results suggest that PGY1 and PGY2 pharmacy residents experience similar stress levels and that residency training affects their levels of perceived stress. The mean ± S.D. PSS10 score for the full sample of pharmacy residents was 19.06 ± 5.90. Those values were significantly higher than the values reported among 18- to 29-year-old participants in a 1983 survey of a probability sample of healthy U.S. adults (mean ± S.D. PSS10 score, 14.2 ± 6.2)10; that age group had the highest average score of all age groups evaluated in the 1983 survey, and 93.3% of pharmacy residents in our sample were under 30 years of age. Waldman and colleagues12 conducted a study assessing stress levels among medical residents in cardiology in Argentina. The cardiology residents (n = 106) were found to have a mean ± S.D. PSS10 score of 20.3 ± 7.4, significantly higher than values in a sample of age- and sex-matched non-medical professionals (16.7 ± 6.7). The researchers also found that the cardiology residents worked a mean ± S.D. of 64.5 ± 30.5 hours per week and that 80.2% suffered symptoms of burnout such as emotional exhaustion and de-personalization. This was a pertinent finding, as mean stress scores in our sample of pharmacy residents were similar to those of the cardiology residents assessed by Waldman et al.12
In our study, pharmacy residents who worked more than 60 hours per week were likely to have more stress than those working fewer hours. More than one half of the surveyed PGY2 pharmacy residents (51.5%) worked more than 60 hours per week, as compared with more than one third of PGY1 pharmacy residents (37.4%); this factor may have contributed to the slightly higher mean PSS10 score of PGY2 pharmacy residents. Pharmacy residents who reported working more than 60 hours per week were significantly more likely than those working fewer hours to have higher MAACL-R subscale scores for depression, hostility, and dysphoria. When the pressures of being overworked exceed a resident’s ability to cope, his or her psychological well-being is in danger and there is the potential for physical exhaustion and feelings of burnout, distress, and depression.13 For residents who feel overworked, learning may be hindered. ASHP has regulations in place that require duty-hour limits for pharmacy residencies. Per ASHP requirements, duty hours must be limited to 80 hours per week inclusive of all inhouse call activities and “moonlighting.”14 Our study findings not only confirm the importance of residents not exceeding duty-hour limits but highlight the fact that residents can experience stress even within those limits.
The difference in pharmacy residency training years did not have a significant effect on MAACL-R anxiety, depression, and dysphoria scores. However, PGY2 residents exhibited higher hostility levels than PGY1 residents. The ramifications of hostility on PGY2 pharmacy residents’ health are beyond the scope of this study. However, studies have found that “high-hostile” and “low-hostile” subjects differed in heart rate and systolic blood pressure.15 Other research indicated that healthy men with feelings of hostility and exhibiting depressive symptoms, even in the mild-to-moderate range, were at heightened risk for cardiac events.16
Pharmacy residents who had higher PSS10 scores tended to have higher anxiety, depression, hostility, and dysphoria scores. This is a notable finding since this correlation was not found in the literature.
There were several limitations to the study. A questionnaire survey depends on the subjectivity of each individual in self-evaluating his or her stress levels. The PSS10 and MAACL-R instruments both are designed to capture data within a “moment” or short time frame. The overall survey response rate was low due to reliance on pharmacy directors forwarding the survey to residents and residents taking the time to complete the survey in its entirety. We believed that the best time to conduct the survey was when residents were entering into the second half of the residency year, a time when their stress might be expected to be at a peak; thus, the window for conducting the survey was narrow. Since the study participants remained anonymous, it was unclear if there was adequate representation from different geographic areas and settings. No data on normal stress levels in the general population have been published since 1988 (the data from the aforementioned 1983 study of Cohen and Williamson9); thus, more up-to-date, “real-time” comparisons of residents and nonresidents were not possible.
Pharmacy residents are subject to professional, situational, and personal stresses during their residency. The PSS10 questionnaire, which is simple to administer and available free of charge, can be used as a tool to screen for stress in pharmacy residents. If there is a desire to collect more detailed information about negative affect levels, the more complicated MAACL-R questionnaire may be used. By evaluating the stress levels of residents, residency program directors and preceptors can facilitate stress management for their trainees in order to improve the learning experience and prevent negative emotional outcomes that may hinder work performance. Residency programs should continue to monitor and prevent residents from not only exceeding duty-hour maximums but consider decreasing work hours to less than 60 per week to prevent increased levels of stress that could be detrimental.
Pharmacy residents exhibited high levels of perceived stress, especially those who worked more than 60 hours per week. Perceived stress was highly correlated to negative affect levels.
The authors have declared no potential conficts of interest.
- Copyright © 2017 by the American Society of Health-System Pharmacists, Inc. All rights reserved.