Research Article Factors related to staff stress in HIV/AIDS related palliative care Prabha S. Chandra, K. R. Jairam, Anila Jacob Department of Psychiatry, NIMHANS, Bangalore, India Correspondence: Dr. Prabha Chandra National Institute of Mental Health and Neuro Sciences, Bangalore - 560029, India. E-mail: prabhachandra@rediffmail.com Abstract Aims: Staff stress in HIV related palliative care has been identified as an important problem worldwide. This study aimed at estimating prevalence of staff stress and its correlates in a sample of palliative caregivers in HIV/AIDS in India. Materials and Methods: Fifty-two participants (29 female and 23 male) completed the Maslach Burnout Inventory (MBI), AIDS Contact Scale (ACS) and AIDS Stress Scale (ASS) and a semi-structured questionnaire. Results: The majority (92%) had average to high scores on at least one domain of MBI. High scores on the factors Emotional Exhaustion, Depersonalisation, and Personal Accomplishment were seen in 10%, 17% and 58% of the sample respectively. ASS score, severity of stress in dealing with persons living with AIDS and having considered leaving HIV related work were predictors of high Emotional Exhaustion scores. ASS score, ACS score and severity of stress with death of a person with AIDS were predictors of high Depersonalisation scores. Female gender was a predictor of a high Personal Accomplishment score. Conclusions: The study emphasises the need to address issues related to staff burnout in HIV palliative care. Key Words: HIV, palliative care, staff stress Introduction Palliative care in India involves mainly those affected by cancer and more recently HIV/AIDS. In the current Indian setting where HIV/AIDS is associated with extreme social stigma, patients will continue to reach health care facilities late for aggressive treatment of the illness. Even though there is a gradual decrease in new HIV infections in India, the epidemic is still spreading. The number of new infections in 2001 has been estimated to be 0.11 million. 1 With an expected rise in HIV prevalence to epidemic proportions in India, and the high cost of anti retroviral therapy, the number of people infected with HIV requiring palliative care will rise exponentially and draw on health care resources both in terms of infrastructure and personnel. 48 Hence, it is of utmost importance to utilize the resources available in palliative care by ensuring efficient functioning of the available palliative caregivers, alongside training of new hands. As in any care giving profession, in palliative care too, the psychological well being of the caregivers plays a major role in improving quality of care. This may be important as exhaustion syndrome is a potential risk for palliative care workers and families because of their special contact with suffering. 2 Burnout has been reported to be associated with sub optimal patient care practices. 3 It is therefore important to identify prevalence of stress and burnout among carers and identify factors that contribute to this stress in the palliative care settings and design suitable intervention programmes for those under stress.
Staff stress in HIV/AIDS related palliative care Most of the early literature on stress experienced by staff caring for dying focused on nurses experiences. 4 As research data on staff stress accumulated, other theoretical frameworks or models came to be discussed. The studies done vary in terms of the sample size and character and the tools used. McKusick & Horstman 5 studied the psychological experience of professionals working in AIDS. They found that respondents experienced more depression, anxiety, overwork and fear of death since starting to work with AIDS. While the number of years of working with AIDS did not correlate with psychological distress, the percentage of total work time spent in AIDS unit was related to depression, and they suggested that burnout is a function of the amount of concentrated exposure rather than of longitudinal contact with the disease. Ross & Seeger 6 found that 34% of health professionals reported stress and 43% reported overwork in their work with AIDS. The major stressors were youth of the people with AIDS, neurological aspects of the illness and death. Stress had the highest correlation for burnout, followed by needing more information about emotional needs of people with AIDS. Maj 7 highlighted the lack of data on the prevalence of the burnout syndrome in AIDS caregivers. One of the major limitations of the evidence available then was that it had been obtained exclusively in developed countries. In her review of literature on impact of working with people with HIV/AIDS, Barbour 8 postulated that stress in staff members may differ depending on their training, previous experience, personality, sexual identity and reason for working in the field. Kleiber et al 9 attempted to identify and characterize burnout in AIDS caregivers by comparing health workers in AIDS and non-aids fields. They found no effect of occupational group (medical vs psychosocial) on burnout, however those in AIDS health care were less burnt out than workers in cancer care or geriatrics. In contrast to Kleiber s results, Catalan 10 found that a majority of AIDS workers had some degree of stress detected by high scores on at least one MBI subscale. In their sample, AIDS workers were found to be five times more likely than oncology workers to score high on MBI with doctors scoring higher on depersonalization than nurses. More recently Hayter 11 studied burnout in HIV care nurses and found that 66% had moderate or high burnout on EE or PA of the MBI. Sherman 12 carried out a qualitative study among nurses working with AIDS patients and found that most nurses coped with physical and emotional risks and stress of AIDS caregiving by problem focused and emotion-focused coping. Understanding the evolution of staff stress is important in preventing and treating the same. One of the popular models used to study staff stress is the person environment model, which suggests that job satisfaction and occupational stress are the result of a dynamic interaction between the person holding a particular job and the environment in which he or she is employed. Most researchers have focused on identifying the correlates based on coping and support alongside demographic and work related variables. Attempting to identify predictors of retention among HIV health care professionals, Brown et al 13 studied health professionals in hemophilia treatment centers. The burnout as measured by MBI was found to be low, but over a four-year period, more than one third left their jobs. Less perceived stress with colleagues and being married were significant predictors of job retention. The various other factors described as related to stress and burnout include: age, 14,15,4,16,17 communication, role ambiguity, team problems, 4,18,19 poor adjustment related to work, social relations and leisure, 10,20 and staff support. 18,4 The palliative care field in India is fraught with problems of developing nations- limited access to palliative care, illiteracy, poverty, poor status of women, stigma against the disease and lack of an organized social security system. 21 To our knowledge there has been no study to date on staff stress in palliative care in India. Even though there have been numerous studies from the West, the distinct features of palliative care in India makes it difficult to draw conclusions in our context based on these studies. The current study was designed to identify the prevalence of stress among staff working in HIV related palliative care and to identify its correlates. Materials and Methods Semi structured questionnaires were given to 52 staff members in four palliative care centres in Bangalore, 49
Chandra PS, et al. and also to staff offering palliative care for the HIV infected in hospitals. The questionnaire (available from the author on request) comprised of items designed to assess (1) the clinical background, (2) their sociodemographic information and (3) issues related to work. The work related section of the questionnaire had items that assessed the following: Degree of satisfaction with patients, families, colleagues, institutional administration, society, including media and government Degree of satisfaction with opportunities for case supervision/ guidance, case discussion with peers and opportunities to improve knowledge and skills Degree of stress perceived dealing with various aspects of HIV/AIDS including stage of disease, age of patients, psychosocial and ethical issues Strategies used to cope with stress. The following structured tools were also used: AIDS Contact Scale 22 (ACS) The ACS is a 15 item self administered questionnaire concerning the number of AIDS patients seen, time spent with AIDS patients, the number of contacts with family and friends of patients and the frequency of a variety of specific physical contacts (touching, handling bedclothes, handling blood and other fluids and handling equipment contaminated by such fluids) and social contact (general conversation with patients, discussion of physical and emotional problems related to the disease). Each item is scored from 1 to 4, the total score being the mean of the item scores, with higher scores indicating a higher degree of contact with persons with AIDS. The scale has been seen to have internal consistency (α = 0.83). AIDS Stress Scale 23 (ASS) The ASS is an 8 item scale that was developed to assess the challenges posed to health care workers as a result of working with AIDS and persons with AIDS. Using a 4-point Liekert type scale, respondents indicate their degree of comfort with AIDS patients and with friends and family of persons with AIDS as well as the degree of risk they perceive as a result of their jobs. There are 5 yes-no questions assessing stress related to working with patients with AIDS, whether they feel their knowledge is sufficient to deal with physical and emotional needs of AIDS patients and 50 with the family and friends of AIDS patients. Each item is scored from 1 to 4 and the score for the scale is obtained by the mean of the item scores. The mean score is an indicator of AIDS stress, with higher scores indicating more stress. The reliability of the scale has been demonstrated with a Cronbach s α = 0.668. Maslach Burnout Inventory 24 (MBI) The MBI is a 22 item scale that is recognized as a valid measure of burnout. The inventory is comprised of 3 subscales that assess 3 aspects of burnout, namely Emotional Exhaustion (EE), Depersonalization (DP) and Personal Accomplishment (PA). Higher scores on EE and DP indicate higher levels of burnout and higher scores on PA indicate lower levels of burnout. In addition to the raw scores, cutoffs have been suggested to classify the score as falling into low, average and high levels of burnout. Scores less than 19 on EE, 6 on DP and more than 39 on PA suggest low level of burnout. Scores of 19-26 on EE, 6-9 on DP and 34-39 on PA suggest average level of burnout. High level of burnout is indicated by a score more than 26 on EE, 9 on DP and less than 34 on PA. Data Analysis Data was entered into SPSS 7.5 for analysis purpose. An initial descriptive analysis was carried out to determine the sample characteristics. t tests, ANOVA and Pearson s bivariate correlation were carried out to determine correlates of burnout. The analysis was carried out using the raw scores on the MBI scales. For assessing prevalence of burnout, the cutoff suggested 24 was used to categorize the scores as indicating low, average or high level of burnout. Results The sample comprised of 52 subjects, 29 females and 23 males. The mean age of the sample was 34.42 years (SD 9.64). Majority of the sample were doctors (46%), followed by counsellors (17%), nurses (14%), social workers (10%) and others (13%). 54% of the subjects had post graduate education. Thirty three subjects were married (64%) and 16 subjects (31%) had children. The majority (48%) of the subjects had been working in the field of palliative care for a period of 1-5 years. 64% of the participants were working full time in palliative care. Most of the participants started
Staff stress in HIV/AIDS related palliative care working in the field either as part of work assigned to them (49%) or as an area of professional interest (35%). Frequency of contact with HIV cases per week was less than 5 cases in 41%, up to 10 cases in 31% and more than 10 cases in 29%. The majority (55%) worked more than 5 days a week with HIV patients. ASS (AIDS Stress Scale) Scores The ASS scores ranged from 1.25 to 4.13 with a mean of 2.42 (0.73). Those who had children scored significantly lower (1.94 ± 0.53) than those without (2.61 ± 0.72); P= 0.003 indicating lower degree of stress among those with children. Prevalence of burnout Over all, across the domains, 92% of the respondents were seen to have average to high levels of burnout in at least one of the 3 domains. 66% of the 44 participants who responded to all the items had high levels of burnout in at least one domain. Emotional Exhaustion: The mean score in this domain was 13.60 (SD 10.71). 62% of the respondents (n=50) reported low levels, 28% average and 10% high levels of burnout in this domain. Depersonalization: The mean score of the respondents in this domain was 5.22 (SD 4.85). Of the total 46 participants who responded to all the items in the domain, 61% had low level, 22% had average level and 17% had high level of burnout in this domain. Personal Accomplishment: The mean score in this domain was 31.12 (SD 8.28). 43 participants responded to all the items in this domain. Of them, 58% had a high level of burnout, 28% and 14% were seen to have average and low level of burnout respectively. Correlates of burnout Emotional Exhaustion (EE): There was no association of the scores in this domain with any of the sociodemographic variables considered. There was significant positive correlation between the ASS score and the EE scores (r = 0.418; P = 0.004). Those who reported using some strategy to deal with their stress were seen to score significantly lower than those who did not. Those who perceived higher severity of stress in dealing with ethical issues in persons living with AIDS, stigma attached to the work, and death of a young person with AIDS scored higher on EE as well. Also, higher scores on EE were associated with reporting having considered leaving AIDS related palliative care work. Depersonalization (DP): Those who did not have children were seen to score significantly higher (6.19 ± 4.94) in this domain compared to those with children (3 ± 3.96), (t (44)=-2.129; P= 0.039). There was also a significant positive correlation of DP scores with ASS scores (r = 0.541; P = 0.015) and ACS scores (r = 0.381; P = 0.038). Those who reported more severity of stress in dealing with death of a person with AIDS and stigma attached to the work scored significantly higher on DP as well. Those who had higher scores on DP also reported having considered leaving palliative care work more often. Personal Accomplishment: There was a significant gender difference in the scores on this domain with females scoring higher (33.92 ± 8.11) than males (27.22 ± 7); P = 0.006 indicating a greater amount of burnout in this domain among males. A significant association was found between PA score and ASS score (r = -0.286, P = 0.037). Those who reported that they took short breaks from work to deal with stress were seen to have a significantly higher score (suggesting lower levels of burnout) than those who did not. A comparison of subjects in part time and full time work with HIV palliative care did not reveal any significant differences between the two groups (Table 1). When professions were compared on burnout scores, doctors showed less scores on all burnout domains compared to nurses and social workers (Table 2). To identify the predictors of the domains of MBI, regression analysis was carried out with the variables found to be significantly associated for each domain. ASS scores, as well as degree of stress perceived with dealing with person living with AIDS, not using some strategy to deal with stress and having considered leaving HIV related work were found to be the predictors for EE scores. For Depersonalization scores, scores on the ASS and ACS and severity of stress related to death of a person with AIDS were significant predictors. For PA, however only gender was a significant predictor; being female predictive of scoring high on PA. (Table 3). Discussion The findings of the current study that 92% of respond- 51
Chandra PS, et al. Table 1: Duration of work and burnout Your work with HIV persons n Mean Std. Deviation Std. Error Mean Emotional exhaustion score Part time 19 12.21 9.92 2.28 Full time 31 14.45 11.24 2.02 Depersonalisation score Part time 17 4.59 5.14 1.25 Full time 29 5.59 4.73 0.88 Personal accomplishment Part time 14 27.79 9.11 2.43 Full time 29 32.72 7.50 1.39 Table 2: Profession and burnout n Mean Std. Deviation Emotional exhaustion score Doctor 20 13.15 9.65 Nurse 6 16.33 16.50 Social worker, counsellor 9 17.11 10.46 Others 4 9.75 6.65 Total 39 14.21 10.69 Depersonalisation score Doctor 19 4.68 4.19 Nurse 6 7.67 7.17 Social worker, counsellor 9 5.11 4.08 Others 4 3.00 2.94 Total 38 5.08 4.63 Personal accomplishment Doctor 19 29.74 8.90 Nurse 6 37.67 7.53 Social worker, counsellor (3-6) 9 30.00 7.07 Others 4 31.25 5.56 Total 38 31.21 8.23 Table 3: Regression analysis to determine predictor variables for MBI domains Domain Variable B t p Emotional exhaustion ASS Score 3.752 2.399 0.023 Considered leaving HIV related work 3.556 3.043 0.005 Strategy to deal with stress 9.185 3.296 0.003 Perceived stress dealing with PLWA 3.537 2.754 0.010 Depersonalization ASS Score 2.393 2.681 0.015 ACS Score 2.975 2.350 0.030 Stress dealing with death of person with AIDS 1.785 2.864 0.010 Personal accomplishment Gender 7.059 2.725 0.010 PLWA- People Living with AIDS ents had average to high levels of burnout and that 66% had high burnout in at least one domain are of some concern. This indicates that high levels of burnout are present in palliative caregivers of HIV/AIDS. Though we did not study a control group, the very fact that a majority of the sample had burnout is significant. We did not find any association of burnout with age, occupational role, or duration of work in the field of palliative care as reported in western studies. 4 Our findings also indicate that reason for working in the field was not significantly associated with MBI scores. We also did not find any significant difference in the MBI scores between different occupational groups though nurses scored the maximum on all burnout measures. This suggests the possibility that palliative care may mitigate the stress on any particular group of professionals due to the team approach to care rather than individual specialist approach followed generally in medicine. However, nurses continue to be at risk for high levels of stress. Our finding that higher scores on ASS were predic- 52
Staff stress in HIV/AIDS related palliative care tive of emotional exhaustion and depersonalization underline the possibility that AIDS related work is inherently stressful. We found that severity of stress in dealing with persons living with AIDS (that would indicate a continued interaction with the patient) was predictive of higher EE, and that severity of stress in dealing with death of a person with AIDS (which would indicate a loss of relationship for the caregiver) was predictive of DP scores. Whether it is the inherent stress due to AIDS care, as measured by ASS, that translates as higher degree of burnout on the MBI requires further exploration. Also, the intensity and frequency of contact with AIDS patients denoted by ACS score was seen to be predictive of depersonalization. This definitely raises the question as to whether there are unique features in AIDS care that induce stress in a caregiver. This also allows scope for designing interventions aimed at relieving stress in these situations. Our finding that subjects who used some strategy to deal with stress scored lower on EE domain indicate that interventions aimed at promoting stress reduction method may decrease staff stress in AIDS palliative care. It is of interest to note that MBI scores in our study are different from that reported in previous studies. Our subjects scored less on EE compared to other samples. 11,13 However it is of interest to note that the scores on PA in our sample was also lower than in these studies indicating poor personal accomplishment. It seems that compared to Western studies, though the degree of emotional exhaustion was found to be less, feelings of self worth in work as measured by PA scale in MBI were less. Low PA maybe due to inadequate resources and inability to meet the physical and financial needs of people living with HIV AIDS which might result in lower self worth. We found that higher scores on EE and DP were associated with reporting having considered leaving the work. This underscores the importance and usefulness of using the MBI in this work setting to identify those staff who are likely to leave AIDS related palliative care. We also found that those who had children reported lower levels of depersonalization. Similar results have been reported by Beck Friis, 25 who found that being married and having children was associated with better job satisfaction. Our finding that being female was associated with higher personal accomplishment scores has not been reported before. Some variables such as age, duration of work, occupational group, frequency of contact with patients and reason for starting to work in the field of HIV/AIDS, which were found relevant in western literature did not seem to have any association with the MBI scores in the current study. The difference in sample characteristics between this study and others may have contributed to the variation in results. The limitations of the current study include a small sample size and lack of a control group. Also we did not attempt to identify possible personal variables such as personality traits, ways of coping, available social supports and current and past stressful life events that could contribute to stress. It would be useful for future studies to address burnout issues in people in full time HIV palliative care and also compare the profile of burnout between cancer and HIV related palliative care. The strengths of this study include the use of a standardized tool (MBI) with established reliability and validity. We included different occupational sub groups to facilitate comparison with most palliative care centers in India, which also use interdisciplinary teams to provide care. The importance of the current study lies in the fact that this is the first attempt to identify and explore staff stress in palliative care set up related to HIV in India. The findings of our study suggest that staff stress in palliative care is an issue that needs to be addressed by professionals in the field. It is important that research on staff stress and burnout in the Indian context be fuelled to provide data on this potentially vexing problem that may slow the growth of palliative care in India. We propose that further studies need to be carried out with larger samples and across settings (HIV/AIDS and cancer). Studies would have to look at all the variables including socio-demographic, personal and environmental to yield a comprehensive picture of the issue of staff stress. More studies are required from developing countries to facilitate a thorough understanding of staff stress in palliative care. P Chandra, Additional Professor K R Jairam, Research Officer A Jacob, Senior Resident NIMHANS, Bangalore, India 53
Chandra PS, et al. References 1. NACO. http://naco.nic.in/vsnaco/indianscene/esthiv.htm. In; 2002. 2. Astudillo W, Mendinueta C. Exhaustion syndrome in palliative care. Support Care Cancer 1996;4:408-15. 3. Shanafelt TD, Bradley KA, Wipf JE, Back AL. Burnout and self-reported patient care in an internal medicine residency program. Ann Intern Med 2002;136:358-67. 4. Vachon ML. Staff stress in hospice/palliative care: A review. Palliat Med 1995;9:91-122. 5. McKusick L, Horstman W, editors. The impact of AIDS on physician. Los Angeles: University of California Press; 1986. 6. Ross M, Seeger V. Determinants of reported burnout in health care professionals associated with the care of patients with AIDS. AIDS 1988;2:395-7. 7. Maj M. Psychological problems of families and health workers dealing with people infected with human immunodeficiency virus 1. Acta Psychiatr Scand 1991;83:161-8. 8. Barbour RS. The impact of working with people with HIV/ AIDS: A review of the literature. Soc Sci Med 1994;39:221-32. 9. Kleiber D, Enzmann D, Gusy B, editors. Stress and Burnout in AIDS Health Care: Are There Special Characteristics? London: Harwood Academic Publishers; 1995. 10. Catalan J, Burgess A, Pergami A, Hulme N, Gazzard B, Phillips R. The psychological impact on staff of caring for people with serious diseases: The case of HIV infection and oncology. J Psychosom Res 1996;40:425-35. 11. Hayter M. Burnout and AIDS care-related factors in HIV community clinical nurse specialists in the North of England. J Adv Nurs 1999;29:984-93. 12. Sherman DW. Experiences of AIDS-dedicated nurses in alleviating the stress of AIDS caregiving. J Adv Nurs 2000;31:1501-8. 13. Brown LK, Schultz JR, Forsberg AD, King G, Kocik SM, Butler RB. Predictors of retention among HIV/hemophilia health care professionals. Gen Hosp Psychiatry 2002;24:48-54. 14. Krikorian DA, Moser DH. Satisfactions and stresses experienced by professional nurses in hospice programs. Am J Hosp Care 1985;2:25-33. 15. Masterson-Allen S, Mor V, Laliberte L, Monteiro L. Staff burnout in a hospice setting. Hosp J 1985;1:1-15. 16. Raphael B. Psychological distress among volunteer AIDS cousellors. Med J Aust 1990;152:275. 17. Bennett L, Michie P, Kippax S. Quantitative analysis of burnout and its associated factors in AIDS nursing. AIDS Care 1991;3:181-92. 18. Cooper C, Mitchell S. Nursing the critically ill and dying. Hum Rel 1990;43:297-311. 19. Bene B, MJ F. Death anxiety and job stress in hospice and medical-surgical nurses. Hosp J 1991;7:25-41. 20. Klonoff E, Ewers D. Care of AIDS patients as a source of stress to nursing staff. AIDS Educ Prev 1990;2:338-48. 21. Rajagopal MR. Involvement of the community in palliative care delivery. Indian J Palliat Care 2001;7:1-3. 22. Pleck J, editor. AIDS-Contact Scale. 2 nd Ed. California: Sage Publications; 1998. 23. Pleck J, editor. AIDS-Stress Scale. 2 nd Ed. California: Sage Publications; 1998. 24. Maslach C, Jackson SE, Leiter MP. Maslach Burnout Inventory Manual. 3 rd Ed. Palo Alto, California: Consulting Psychologists Press, Inc; 1996. 25. Beck-Friis B, Strang P, Sjoden P. Caring for severely ill cancer patients: A comparison of working conditions in hospital-based home care and in hospital. Support Care Cancer 1993;1:145-51. 54