An Assessment of Work Environment Factors Correlating with Burnout Status of Physicians Employed with the Public Hospitals Authority in The Bahamas

Wallace – Bain¹, M. Gonzales¹, R. Roberts – Carter², S. Zonicle – Newton³, Carnille Farquharson⁴, MAC Frankson⁵, S. Pinder – Butler⁶

^1,2,3 Public Hospitals Authority, Nassau, Bahamas
⁴Family Medicine Postgraduate Programme, University of the West Indies, School of Clinical Medicine and Research, Nassau, Bahamas
⁵ Statistics Department, University of the West Indies, School of Clinical Medicine and Research, Nassau, Bahamas
⁶Family Medicine Postgraduate Programme, University of the West Indies, School of Clinical Medicine and Research, Nassau, Bahamas

Corresponding Author

Dr. Merrilyn Wallace-Bain
Public Hospitals Authority, Nassau, Bahamas
P.O. Box EE16595, Nassau, Bahamas
Email: [email protected]

DOI: doi.org/10.48107/CMJ.2021.07.002

Copyright: This is an open-access article under the terms of the Creative Commons Attribution License which permits use, distribution, and reproduction in any medium, provided the original work is properly cited.

©2021 The Authors. Caribbean Medical Journal published by Trinidad & Tobago Medical Association.

ABSTRACT

Objective
To identify the prevalence and factors associated with the Burnout Syndrome in physicians working in the Public Hospitals Authority (PHA), Nassau, Bahamas.

Methods
A cross-sectional study was done in 2014 utilising a self-administered survey comprising of demographics, general health, work environment, and Maslach Burnout Inventory items to assess burnout among physicians working in nine (9) departments of PHA. The IBM SPSS (v. 20.0 or more current) Statistical Analyses software was used for data analysis.

Results
The study participants were 153 physicians. Their mean age was 35.84 (± 7.09) years old; median 34.00 (IQR: 31.00, 40.00) years old, 99 (64.7%) were females. No associations were found among these, other socio-demographic variables measured and burnout status. Physicians in the department of Internal Medicine represented 34 (22.2%), Family Medicine 31 (20.3%), Emergency Medicine 30 (19.6%), Paediatrics 21 (13.7%), and physicians in other departments 34 (22.3%). Of the physicians employed under the PHA, 55.7% collectively exhibited burnout. Poor balance of family life, 15 work environment potential stressors and 4 potential stress relievers were found to each be weakly or very weakly related to burnout status. Postgraduate programme year, irregular sleep pattern and lack of appreciation were moderately strong positively related. Logistic regression analysis showed the key predictors of burnout status to be lack of appreciation (OR=1.69, p=.002) and number of years worked post-internship (OR=.94, p=.039).

Conclusion: Physician’s sense of appreciation and number of years post internship were clear predictors of burnout.

INTRODUCTION

The phenomenon known as burnout has become increasingly significant in the medical community today.  This syndrome was coined by psychologist Herbert Freudenberger who identified this among individuals in service professions brought on by repeated unrelieved stress.¹

A physician’s risk for burnout may be dependent on work environmental factors inclusive of the length of time as a physician, sleep deprivation, and the number of hours worked. Specialties such as emergency medicine may be expected to carry the highest prevalence for burnout syndrome among physicians.²  However, studies have shown that this phenomenon may be equal or higher in other subspecialties, as they experience their unique challenges and stressors within their work environment.³  Regardless of the specialty or other work environment factors, any medical environment without proper recourse could lead to this phenomenon which could manifest as emotional exhaustion, depersonalization, and a decreased sense of personal achievement. Ultimately, this could place the physician’s general physical and mental health at risk and may potentially lead to suboptimal patient care.⁴

The adverse outcome of patient care secondary to physician burnout has resulted in several studies worldwide to identify the various stressors and their impact on the general health and work ethics of the physician.  However, there have been few previous publications assessing factors related to burnout among Bahamian physicians. Therefore, this research project aimed to identify factors associated with Burnout Syndrome considering the prevalence of that syndrome in physicians working in the Public Hospitals Authority (PHA), Nassau, Bahamas.

METHODS

Measurements
This was a cross-sectional study utilising a questionnaire to assess general health and a second questionnaire to measure the prevalence of burnout among physicians employed under the Public Hospitals Authority (PHA) at the Princess Margaret Hospital (PMH), Community Clinics, and Family Medicine Clinic in Nassau, Bahamas.  The study period was October through November 2014.  The targeted medical specialties included physicians in Family Medicine, Internal Medicine, Emergency Medicine, and other departments. The sample size was 153 physicians. Inclusion criteria were willing physicians working in PMH, including the Family Medicine Department, and other physicians under PHA.  Medical students, visiting/non-salary working physicians, physicians unwilling to participate in the survey were excluded.  Approval for the study was granted by the Joint Ethics Committee of the Public Hospitals Authority and the University of the West Indies.  Confidentiality was upheld throughout. The researchers gave oversight to the self-administered survey and collected completed ones. The general health questionnaire assessed socio-demographics, work environment, potential stressors, stress reduction factors, job satisfaction, and general health.  Subsections on potential stressors, stress reduction factors, and job satisfaction were adapted with modifications from a cross-sectional survey, assessing burnout among public doctors in Hong Kong.⁵

The Maslach Burnout Inventory – Human Services Survey (MBI – HSS), a validated instrument measuring burnout syndrome was procured and utilised.  It uses a 22 item questionnaire with a Likert Scale ranging from 0 to 6.⁶ It is used to assess three components of burnout syndrome: emotional exhaustion (EE), depersonalization (DP), and personal accomplishment (PA).  This study’s researchers adhered to the scoring key provided and interpreted burnout as high emotional exhaustion and/or depersonalization.

The IBM SPSS (v. 20.0 or more current) Statistical Analysis application software was used to perform both descriptive and inferential quantitative data analyses.  Descriptive statistics included determining counts, percentages, modes, and medians along with respective interquartile ranges presented in tables and figures.  Inferential statistics included the Mann-Whitney U Test to assess median differences and the Chi-Squared tests of both independence and homogeneity.  Associations were assessed using Cramer’s V, Phi, and Spearman’s correlation coefficients. Type 1 error (alpha) was predetermined at ≤5% (α≤0.05) to assess the statistical significance of inferential study findings.

RESULTS
In this study of 153 participating physicians, 99 (64.7%) were females, while 54 (35.3 %) were males. With respect to race, 140 (91.5%) were black.  The largest nationality represented were Bahamians at 129 (84.3%).  In the participating group of physicians, more than half of the individuals were single 81 (54.9%), while less than half were found to be married 68 (44.4%).  Separated and divorced individuals accounted for the remaining percentages at one (0.7%) and three (2%) respectively.  See Table 1.

Table 1: Characteristics of participants in survey

The 34 (22.2%) physicians from the Department of Internal Medicine represented the largest participating group in the survey.  This was followed by physicians in Family Medicine at 31(20.3%), Emergency Medicine at 30 (19.6%), and Paediatrics at 21 (13.7%).  Physicians from other departments also made up the sample studied.  Concerning current posts, Consultants represented 19 (12.4%), Senior Registrars 10 (6.5%), Registrars 12 (7.8%), Senior House Officers 95 (62.1%), and Interns 17 (11.1%). As it relates to a residency program, 105 (69.1%) physicians were not currently enrolled in a programme while 47 (30.9%) were.  With respect to the participants who were enrolled in a residency program, seven (14.6 %) were in year one of a programme while 18 (37.5%), 10 (20.8%), nine (18.8%), one (2.1%), and three (6.3%) represent individuals enrolled in year two to greater than five years respectively. See Table 2.

Table 2: Job related characteristics of participating physicians studied

The median duration practicing as a doctor post-internship was six (IQR: 3, 10) years. Within a week, the median number of weekend days worked by these participants was three (IQR: 2.75, 3) days.  While 13.3% reported working none, 15 (10.0%) worked on Saturdays only, two (1.3%) Sundays only, while the modal percent 113 (75.3%), worked on both weekend days. Three persons (2.0%) refrained from reporting on this issue.  The mean number of days worked per week was 5.85 (± 0.89).  The median number of hours worked per week was 40-50 (IQR: 40-50, >60-80) hours. More specifically, six (4.1%) physicians interviewed worked <40 hrs., 82 (55.8%) 40-50 hrs., 17 (11.6%) 51-60 hrs., 29 (19.7%) >60-80 hrs., and 13 (8.8%) >80 hrs. Six (3.9%) persons did not share this information.

During the month prior to completing the questionnaire the hours of overtime work was zero (0, 21.5) hours.  The median number of hours participants spent in after-work hours meetings or similar activities was three (0, 6) hours.

Approximately half of the physicians had high scores for emotional exhaustion, a quarter had high scores for depersonalization, and a quarter had low scores for personal achievement.  See Table 3.

Table 3:  MBI burnout subscales prevalence among physicians

The prevalence of burnout using the definition of high emotional exhaustion and/or high depersonalization was 55.7% in this study. See Table 4,

There was a very strong relationship between physicians’ departmental placement and burnout status (Cramer’s V=.365, p=0.011).  By specialty department burnout prevalence was: three (75.0%) for radiology, 15 (71.4%) for Paediatrics, 23 (69.7%) for Internal Medicine, six (66.7%) for Anaesthesiology, 19 (65.5%) for Emergency Medicine, two (40.0%) for other departments, 12 (38.7%) for Family Medicine, four (36.4%) for Psychiatry, and one (11.1%) for Surgery. This study showed that the departments with the highest prevalence of burnout were Paediatrics, Internal Medicine, Emergency Medicine, and Anaesthesiology. The Internal Medicine and Emergency Medicine Departments represented approximately 50% of all the physicians who had burnout. See Table 4.

Table 4: Prevalence of burnout among physicians

Regardless of physician burnout status, small percentages (≤6 %) of physicians perceived illicit drugs as a stress reduction factor. However, 38.5% of the physicians experiencing burnout perceived alcohol as a stress reduction factor, while 25.7% without burnout did so. Irrespective of burnout status, 80% or more of the physicians perceived spousal support, exercise, and co-worker support to be likely stress reduction interventions.  See Figure 1.

Figure 1: Percentages concerning perceived stress reduction covariates by physicians’ burnout relief status

There was an overall decrease in burnout prevalence as the physicians’ “number of children” increased from no children to three children. The two physicians with four children had burnout and thus 100% burnout prevalence. The physicians who were “very dissatisfied”, “dissatisfied”, and “neutral” about their job had a higher prevalence of burnout. The physicians who were satisfied and very satisfied about their job had a lower prevalence of burnout. Among the physicians sampled, greater than 50% denied having any medical conditions while 10.5% admitted to having hypertension, 5.9% had asthma 2% were diabetics, and 6.5% admitted to having other medical conditions.

There were 70.4% of the physicians surveyed who had reported taking no sick days in the past month prior to the survey, while 13.8% had taken at least one sick day, 5.9% had taken two sick days, 0.7% had taken three sick days and 9.2% of the physicians had taken greater than three sick days.  The median number of days participants reported being sick was zero (IQR: 0, 1). There were 96.5% of physicians in this study who denied being hospitalized, while 3.5% admitted to hospitalization. The median number of days hospitalized was zero (IQR: 0, 0). Concerning medication, 62.1% did not take any medication, 19% admitted to taking one tablet, 9.3% took two tablets and 5.2% had taken three tablets, while 1.3% and 0.7% of physicians in the study were found to have taken four and five tablets per day, respectively. The median number of medications taken was zero (IQR: 0, 1).

There were 142 (93.4%) who admitted to having never smoked, four (2.6%) did so a few times per year or less, two (1.3%) about once per month, two (1.3%) once weekly and two (1.3%) smoked a few times per week or more frequently.  With regards to exercise, 11 (7.2%) of the participating physicians reported never doing so, 14 (9.2%) did so a few times per year, 23 (15%) admitted to exercising a few times per month, 11 (7.2%) participated in exercise at least once per week, 62 (40.5%) confessed to a few times per week, and nine (5.9%) admitted to everyday.

In the sampled group of physicians, seven (4.7%) never consumed a healthy diet, four (2.7%) said they consumed a healthy diet a few times a year, six (4.0%) confessed to having this once a month or less, 13 (8.7%) claimed to do so once per week, 72 (48%) reported having this a few times per week, and 34 (22.7%) reported having a healthy diet daily.

There were 48 (31.4%) physicians who reported to have never consumed alcohol, 24 (15.7%) drank a few times or less per year, 20 (13.1%) did so about once per month, 23 (15.0%) stated having alcohol a few times per month, 20 (13.1%) said once per week, 17 (11.1%) claimed to do so a few times per week, and one (0.7%) drank every day.

With respect to weight and height, both were normally distributed in frequency, the means being 175.2 (± 43.3) lbs., and 66.8 (± 3.9) inches.  The calculated BMI distribution was slightly skewed to the right with the mean being 27.5 (± 5.8) kg/m², median BMI was 26.8 (IQR: 23.8, 30.8) kg/m².

With regards to factors that may prevent burnout, spousal support and alcohol consumption had a stronger association with burnout than exercise and illicit drug use. Support from co-workers had no association with burnout. Most work environmental factors and potential stressors were associated with burnout. See Table 5.

Multiple logistic regression done on the data, without regard for the physicians’ specific specialty department, revealed that the two statistically significant variables serving as significant predictors for burnout were physicians sensing of a lack of appreciation (OR=1.687; 95% CI: 1.219, 2.336; B=.523, p=.002), and the number of years post-internship (OR=.940; 95% CI: 887, .997; B=-.061, p=.039). The constant in this model was not statistically significant (OR=.481; B=-.732, p=.176), but serves to complete the associated predictive equation.  Note that there were other potential stressors reviewed that were not statistically significant.  See Table 5.

Table 5: Factors found to be significantly associated with burnout

DISCUSSION
Physician’s burnout prevalence was 55.7% and manifested mainly as emotional exhaustion. Similarly, in the Bahamas in a study on 66 health care personnel using the abbreviated Maslach Inventory Human Services Survey for Medical Personnel, emotional exhaustion was the only dimension reported being experienced among physicians enrolled in a residency program and medical students.⁷  Marcelino et al., in a Portugal study on 153 physicians using a modified Maslach Burnout Inventory found similar results.⁸ Berger et al; found that 45.8% of all physicians had at least one symptom of burnout, with emotional exhaustion being the most common at 37.9 %.⁹ In comparison, physicians in a study done in Jamaica using the Maslach Burnout Inventory found 70% of Emergency Physicians reporting moderate to high levels of emotional exhaustion.¹⁰  With respect to the work environment, we found that the specialties with the highest rate of burnout were Anaesthesiology (66.7%), Paediatrics (71.4%), Internal Medicine (69.7%), and Emergency Medicine (65.5%), which was comparable to a study by Berger et al; who found a high rate of 65% in emergency physicians and 55% in general Internal Medicine (2).⁹  Likewise, Hutchinson et al in Jamaica found a high percentage of burnout manifested as emotional exhaustion among Emergency Medicine workers.¹⁰

Burnout among participating physicians was associated with certain socio-demographic and stress reduction factors. Spousal support, co-worker support, and exercise were considered as stress reduction factors by most physicians. In general, there was a decrease in burnout prevalence as the physician’s “number of children” increased to a maximum of three children. This was supported by IsHak et al who indicated that parenting could act as a protective factor against burnout because it results in less detachment and depersonalization.¹¹ The data also indicated that job satisfaction may also be a protective factor against burnout.

Alcohol consumption was the only general health factor that had an association with burnout which may be explained by the small sample size.

The majority of work environment and potential stressor factors were found to be associated with a high prevalence of burnout. The finding of the senior house officer post in the early years after internship being associated with a higher level of burnout was also highlighted in a study done on public doctors in Hong Kong by Siu et al, Mckenzie et al in a study done in The Bahamas, and a literature review by IsHak et al.^5,7,9 This may be as a result of early-career physicians being subjected to working more hours with less flexibility than their senior counterparts.⁷  Subsequently, this could have introduced bias, as young, careered physicians may have been more willing to participate, as a form of protest to working conditions.

Lack of appreciation and the number of years post-internship were the two independent risk factors associated with high burnout prevalence in the multivariate analysis.

One of the limitations in this study was the lack of equal distribution in the sample size across specialties.

CONCLUSION
This study highlighted the prevalence of burnout among physicians employed under the Public Hospitals Authority, showing that lack of appreciation and number of years post-internship were the two factors that were mostly associated with burnout.

RECOMMENDATIONS
It is recommended that further studies be done in this area to help in identifying causes of burnout, versus merely associations identified as in this study.  As lack of appreciation and number of years post-internship were factors identified to be most associated with burnout, policies should be implemented to address these.  Furthermore, this study should be expanded to include all physicians across the archipelago of the Bahamas, with a repeat study post the implementation of mitigating factors to determine the effectiveness of the new policies.

Ethical Approval Statement: Ethical approval to conduct this study was granted by the Joint Ethics Committee of the Public Hospitals Authority and the University of the West Indies.  Researchers obtained informed consent from each patient before their participation.  There were no consequences for refusal to participate in the study.

Conflict of Interest Statement: Throughout the development and post completion of this manuscript the authors of this paper certifies that there were no financial or non-financial conflict of interest.

Informed Consent Statement: Informed consent was obtained from each participant.

Funding Statement: Funding for this research was solely provided by the researchers.  Expenditures included the purchase of the Maslach Burnout Inventory and photocopies of questionnaires.

Author Contributions: MAC Frankson conducted data analysis, participated in study design, data analysis and interpretation, critically revised manuscript, and approved the final version. S. Pinder–Butler participated in research question selection, revision of the manuscript, and approved the final version. M. Wallace–Bain, M. Gonzales, R. Roberts–Carter and S. Zonicle-Newton participated in research question selection, study design, data analysis and interpretation, revision of the manuscript, and approved the final version. C. Farqhuarson participated in the review of the manuscript.

REFERENCES

1. Korczak D, Kister C, Huber B. What is Burnout Syndrome? [Internet]. http://www.ncbi.nlm.nih.gov/pubmedhealth/PMH0050545. 2010 [cited 8 September 2014]. Available from: PubMed Health. What is burnout syndrome? http://www.ncbi.nlm.nih.gov/pubmedhealth/PMH0050545.
2. Popa F, Raed A, Purcarea V, Lala A, Bobimac G, Davilla c. Occupational Burnout Levels in Emergency Medicine – a national study and analysis. Journal of Medicine and Life. 2010;3(3):207-215.
3. Arigoni F, Bovier P, Sappino A. Trend of burnout among Swiss doctors. Swiss Medical Weekly. 2010;140.
4. Nason G, Liddy S, Murphy T, Doherty E. A cross-sectional observation of burnout in a sample of Irish junior doctors. Irish Journal of Medical Science. 2013;182(4):595-599.
5. Siu C, Yuen S, Cheung A, Burnout Among Public Doctors. Hong Kong Medical Journal. 2012;18(3):186-192.
6. Maslach C, Jackson S. Maslach Burnout Inventory – Human Services Survey [Internet]. http://www.mindgarden.com/products/mbi.htm. 2010 [cited 8 September 2014]. Available from: Maslach C, Jackson S. Maslach Burnout Inventory – Human Services Survey. http://www.mindgarden.com/products/mbi.htm
7. Mckenzie B, Walker T, Chin C. (2018, August). Prevalence of Burnout in Bahamian Health [Letter to the Editor]. Journal of Occupational and Environmental Medicine, 434-435. Retrieved from https://journals.lww.com/joem/Fulltext/2018/08000/Prevalence_of__Burnout_in_ Bahamian_Health_Care.27.aaspx
8. Marcelino G, Cerveira J, Carvalho I, Costa J, Lopes M, Calado N et al. Burnout levels among Portuguese family doctors: a nationwide survey. British Medical Journal 2012;2(3):e001050
9. Berger E. Physician Burnout- Emergency Physicians see Triple Risk of Career Affliction. Annals of Emergency Medicine. 2013;61(3):A17-A19.
10. Hutchinson TA, Hasse S, French S, McFarlane TA. Stress, Burnout and Coping among Emergency Physicians at a Major Hospital in Kingston, Jamaica. West Indian Medical Journal. 2014;63(3):262-266
11. IsHak W, Lederer S, Mandili C, Nikravesh R, Seligman L, Vasa M et al. Burnout During Residency Training: A Literature Review. Journal of Graduate Medical Education. 2009;1(2):236-242

Appendix 1: Questionnaire

SECTION A:  Demographics

1. What is your Gender?

□₀    Male        □₁  Female

2.What is your Race?

□₀  Black                           □₃  Asian (Indian)

□₁ White                           □₄ Asian (Phillipino)

□₂ Asian (Chinese) □₅ other   ___________

3.What is Nationality?

□₀ Bahamian                    □₄ North American                                     

□₁  Haitian                       □₅  Asian

□₂  Jamaican                    □₆  African

□₃  Other Caribbean        □₇ Other  ________________

4.What is your Marital Status?

□₀  Single                       □₃  Separated

□₁  Married                                □₄  Divorced

□₂  Common – Law          □₅ Widowed

5.How many Children do you have? _____________

6.How old are you? _____________

SECTION B: Work Environment

1.In which Department do you work?

□₀   Family Medicine                 □₂  Emergency Medicine

□₁  Internal Medicine               □₃  Other ______________________

2. What is your Current Post?

□₀ Intern                                 □₂ Registrar                   □Consultant

□₁ SHO                                  □₃ Senior Registrar

3(a). Are you in a residency program?

□₀ Yes                           □₁ No

3(b). If yes, what is your program year?

□₀ 1                □₁ 2                   □₂ 3                      □₃ 4                  □₄ 5

□₅ Other: ______________________

4. Which weekend days do you work in a week?

□₀  None                      □₁ Saturday                   □₂ Sunday                      □₃ Both

4. How many days do you work a week? ___________

5. How many hours do you work a week? ___________
6. How long have you been a doctor after internship? __________
7. In the last month how many of the following have you done:

a Number of on – call days   __________

b Number of overtime hours __________

c Number of after normal working hours meetings/activities __________                                                             

SECTION C: Potential Stressors

On a scale of 1-5 rate the following stress factors, with 1 being least likely to contribute to stress and 5 being the most likely to contribute to stress.

                                                1                      2                      3                      4                      5

Sleep deprivation                       □₀                     □₁                     □₂                     □₃                     □₄

Irregular sleep pattern              □₀                     □₁                     □₂                     □₃                     □₄

Lack of appreciation                 □₀                     □₁                     □₂                     □₃                     □₄

Excessive workload                   □₀                     □₁                     □₂                     □₃                     □₄

Poor balance of family              □₀                     □₁                     □₂                     □₃                     □₄

Life and work

Poor working relationship         □₀                     □₁                     □₂                     □₃                     □₄

With Colleagues

Lack of independence               □₀                     □₁                     □₂                     □₃                     □₄

Lack of job security                   □₀                     □₁                     □₂                     □₃                     □₄

Other:_______________________________________

       SECTION D: Stress Reduction Factors

 On a scale of 1-5 rate the following stress reduction factors with 1 being the least important stress reduction factor and 5 being most important stress reduction factor.

                                                1                      2                      3                      4                      5

Exercise                                    □₀                     □₁                     □₂                     □₃                     □₄

Alcohol consumption              □₀                     □₁                     □₂                     □₃                     □₄

Illicit Drug Use                         □₀                     □₁                     □₂                     □₃                     □₄

Spousal Support                      □₀                     □₁                     □₂                     □₃                     □₄

Support from co-workers        □₀                     □₁                     □₂                     □₃                     □₄

Other:_______________________________________________

SECTION E: Job Satisfaction

How satisfied are you with your job?

 SECTION F: General Health

 1. Which of the following medical problem(s) do you have?

□₀ None                                                □₅  Breast Cancer

□₁ Hypertension                                   □₆  Prostate Cancer

□₂ Cerebrovascular Disease                  □₇  Asthma

□₃ Coronary Heart Disease                  □₈  COPD

□₄ Diabetes Mellitus                             □₉ Psychiatric illnesses

□₁₀ Other  __________________

 2. In the last month:

a. How many sick days did you have? _______
b. How many days have you been hospitalized? _______
c. How many types of medications have you taken? _______                              

3. How often do you adapt to the following lifestyle habits?

0 – Never

1 – A few times a year or less              4 – Once a week

2 – Once a month or less                    5 – A few times a month

3 – A few times a month                     6 – Every day

a. ______ Smoking

b. ______ Alcohol Consumption

c. ______ Exercising

d. ______ Healthy Diet

4. What is your weight? _______ kg or ______ lb

 5. What is your height? _______ M or ______ ft.

SECTION G
Maslach Burnout Inventory Inserted Here