Job Stress among Female Residents during Postgraduate Training at a University Hospital in Karachi , Pakistan

Background: Postgraduate medical training is a stressful experience for both male and female doctors but previous studies have shown that stress might be more prevalent amongst female residents. Since women make up 70-80% of medical graduates in Pakistan we conducted this study to determine the prevalence of stress among female residents during postgraduate training and identify the most common stressors. Methodology: All female residents enrolled in postgraduate training programmes at the Aga Khan University meeting the inclusion criteria were recruited and informed written consent was obtained. Job Stress Inventory (JSI) was used to identify symptoms and sources of job stress. A cut-off value of >25 was used to indicate job stress. Results: A total of 154 female residents met the inclusion criteria. The response rate of the survey was 77%. Sixty one residents scored 26 or more on the JSI, indicating the presence of job stress. The three most common causes cited for job stress were, lack of control, environmental factors and workplace politics. Conclusion: Our study indicates a high prevalence of stress among female residents, independent of the marital status, number of children, the chosen specialty or the level of training. We suggest that programme directors and supervising faculty acknowledge the need for change in the training programmes refraining from favoritism and the need for politics towards constructive feedback based on merit and hard work and allow the residents to have more control over their working environment.


Introduction
Postgraduate medical training is a stressful experience for both male and female doctors (Nesbitt, 1988), but women tend to report higher levels of stress reaction during residency than men (Butterfield, 1988;Hughes et al., 1991) These stressors have been reported to lead to depression (Butterfield, 1988;Hsu et al., 1987), burn-out (Dyer, 1994), anger and irritability (Scneider & Phillips, 1993) anxiety (Hughes et al., 1992), substance abuse (Hughes et al., 1991) and drop out.Most of this data is from western countries and as stress maybe influenced by unique cultural factors, the level and sources of stress may be different in different regions.
Pakistan is perceived to be a conservative society and in order to pursue their medical careers, women doctors have to overcome several prejudices and pressures.Since women make up 70-80% of medical graduates in Pakistan (Rebecca, 2013;Zofeen, 2010) we conducted this study to explore workplace stress in female residents and identify the most common sources of stress based on the Job Stress Inventory (JSI).

Methods
This cross-sectional study was conducted at the Aga Khan University during 2012-2013.Ethics committee approval was obtained prior to starting the survey and Associate Dean Postgraduate Medical Education (PGME) was informed.All female residents enrolled in the post graduate training programmes were recruited.Informed written consent was obtained from all participants.Exclusion criteria were trainees who were enrolled in fellowship programmes, those on leave or placed in external elective rotations outside the main campus.
The JSI was used to identify symptoms and sources of job stress.A hard copy of the JSI was sent to all the participants along with a letter explaining the purpose of this study.We aimed for a response rate of 75%.All survey forms returned by the residents were collected by the principal investigator and the data was coded to protect the identity of the participants.Data was entered on to SPSS by a research officer and double checked.Cross-tabulation and frequency were computed to analyze symptoms and sources of job stress by using statistical software (SPSS Inc.,Chicago,.A cut-off value of 26 or above was used to indicate job stress.Cronbach's alpha was computed to assess the reliability of the questionnaire.

Results
One hundred and fifty four women residents met the inclusion criteria.Completed questionnaires were returned by one hundred and nineteen subjects (77.27%).

Characteristics of the residents
As shown in Table 1, residents were divided into three groups based on specialty; Medicine and allied specialties, Surgery and allied specialties and the RAP group which included Radiology, Anaesthesia and Pathology.The majority of (n=61, 51%) residents belonged to Medicine and allied specialties were junior residents (in the first 2 years of training) while seventy two (60.5%) were senior residents, i.e. in the third, fourth and final year of training.Forty-nine (41%) were married and 56% of the married residents had not yet started a family.Thirty two percent had one child, 8% had two children and 4% had more than two.

Job Stress
The questions used to identify the symptoms of job stress had a Cronbach's alpha of 0.92.Three questions had Cronbach's alpha less than 0.50 and removing them did not improve the reliability to more than 0.92.The mean JSI score of the whole group was 28.13 (2.00-70.00;SD 14.07).With the recommended cutoff value of 26 on the JSI tool, the estimated prevalence of job stress was 51%.No statistically significant difference was observed in the mean JSI score amongst residents in the three specialty groups, senior or junior, married or single and number of offsprings the married residents had (p value 0.64, 0.87, 0.46, 0.62 respectively).This is shown in Table 2.

Sources of job stress
The overall reliability of the scores to identify sources of job stress on the JSI was 0.94.Table 3 shows the frequency of responses grouped into two categories, i.e. (never + occasionally) and (somewhat often + frequently + almost always).The three most common causes cited for job stress were lack of control ("I feel trapped in a situation without any real options", "I am unable to influence decisions that affect me", and "there are a lot of requirements that get in the way of my doing certain tasks"), environmental factors ("I have to deal with many little hassles") and workplace politics ("I sense that popularity and politics are more important than performance").The organization that I work for pressures employees to do things that are unethical or unsafe 97(82.9)20(17.1%)

Discussion
While women entering medical profession faced an uphill struggle in the nineteenth century (Gunning, 1996), the twenty first century is seeing a feminization of medicine (Brian, 2008).The percentage of female medical graduates in the United States in 2011 was 48.3% (Barzansky & Etzel, 2011).
According to Royal College of Physicians, at least 50% of new UK medical graduates have been female for a decade, rising to over 60% between 2007and 2009(Elston, 2009)).In Pakistan 70-80% of medical students are females (Rebecca, 2013;Zofeen, 2010).Despite the increased enrollment of women in medical schools, the medical workforce in Pakistan is severely lopsided., 2013].Results of this study show a high prevalence of stress amongst female postgraduate female residents working in a private medical university in Pakistan and this might be a contributing factor to lessen female specialists in the country.However this level of stress is comparable to the physicians' stress reported in previous studies, i.e. 48% doctors including trainees, general practitioners and consultants graded job stress from high to very high in a study conducted at the same institution in 2002, with higher levels of stress reported by females (Khuwaja et al., 2004).Firth-Cozen reported that 50% of junior house officers exhibited emotional distress (Firth-Cozens, 1987) and Caplan reported that about half of senior medical staff suffered from high levels of stress and anxiety (Caplan, 1994).
Women are underrepresented in most surgical fields globally (Allen, 1989;Maheux et al., 1989;Wakeford et al., 1989) due to the female physician's awareness of the competing demands made on her time by career, marriage and children (McCue, 1985).This was also seen in our survey where, Medicine was the career choice of 51% residents, followed by the RAP group (34%), with only 15% of women doctors taking up surgery.The surgical specialty is perceived to be more stressful but in our survey though the surgical residents had the highest mean JSI score, this was not statistically different from medicine and the RAP group.However the poor response rate (56.25%) from residents from the surgical and allied specialties might have skewed the results.
Junior house officers have been identified by North American authors to be most vulnerable to stress (Small, 1981;Ziegher et al., 1984;Ford, 1983;Mazie, 1985).In our survey the junior residents (year 1 and 2) had a higher mean JSI score as compared to the senior residents (year 3, 4 & 5) but this did not reach statistical significance.
We commenced the study with the assumption that married women residents with multiple personal and professional demands will be more at risk of experiencing job stress, but the results showed that there was no statistical significance between the mean JSI scores of married and single residents.Several studies have similarly found that marriage and strong social contacts reduce the perception of stress (Landau et al., 1986;Kelner, & Rosenthal, 1986;Bickel, 2000).Residency is the most common period for female physicians to have their first child (Philips, 2000).In our survey 56% of the married residents had not started a family.Residents with two children were the most stressed but this was not statistically significant.
Job-related stress, career dissatisfaction and psychological distress are greatest when high levels of demands are placed on the physician while providing little individual control over work and little autonomous decision-making power (Amick, 1998;Mumford, 1983).This study has also highlighted the fact that lack of control, having to deal with small hassles, and politics and favoritism are the most prevalent causative factors of stress among the female residents.
In an editorial Mumford stressed that educators in medical disciplines should make changes in the training system after identifying "who" specifically is vulnerable and "what" specifically is harmful to medical training (Mumford, 1983).
We have identified that women residents in a major teaching hospital are at risk of stress and lack of control over environment, politics and favoritism at the workplace and those are harmful aspects of their training.We plan to share the findings of this survey with our postgraduate medical education committee.We suggest that programme directors and supervising faculty should acknowledge the need for change in the training programme, refraining from favoritism and the need for politics and more towards constructive feedback based on merit and hard work instead of popularity.Strategies focused towards reduction and prevention of stress could include formal education of the residents about stresses inherent in the training programmes, workshops directed at managing stress and gripe sessions of residents and interested faculty to openly discuss their concerns and feelings.
There are some limitations to this study.This was a single center study at a private university hospital that attracts high-achievers; career oriented, motivated female doctors and may not reflect the issues faced by female residents working in other institutions or public sector hospitals.There are limitations to the tool that we used to measure stress.The JSI identifies presence of job stress and sources of stress related to the workplace, but does not provide measures of work stress unique to women.We feel that there is a further need to explore sources of stress unique to women physicians working in our conservative society utilizing an open-ended qualitative methodology.

Conclusion
This study showed that a high percentage of female residents suffer from stress (51%).These results were not dependent on year of residency programme, marital status or chosen specialty.