Introduction

Nurses providing around–the–clock monitoring and bedside care to sick babies in Intensive Care Unit (ICU) face a challenging situation [1]. Stress has been found to be related to nurses’ activities in intensive care units, and also is related to characteristics of the ICU environment [2]. A bibliographic review of factors leading to stress in pediatric and neonatal ICU nurses [3] found stress was related to lack of support from colleagues [4], strained work relationships [5, 6], high shift rotation, double shifts, insufficient information and professional inability (including inability to cater to patients’ needs) [6–9], insufficient/malfunctioning equipment, chaotic work environment, large number of emergencies, need for constant attention [5, 6, 8, 9], relationship with the patients’ relatives and death of the newborn [5, 6]. In earlier studies, stress in nurses has been found to be associated with burnout [8, 10], stress related physical symptoms, dissatisfaction with work [2] and psychological symptoms [11, 12]. Earlier studies in India conducted on ward, intensive care, operating room and emergency staff nurses [13, 14] found them experiencing moderate to high stress levels.

Researchers have been studying positive and negative effects of caring on the caregiver. Positive effect is conceptualized as compassion satisfaction and negative effect as compassion fatigue. Compassion fatigue consists of two constructs: burnout and secondary traumatic stress. Both share negative effect where burnout is about feeling worn out and secondary traumatic stress is about feeling afraid [15]. Professional quality of life has a complex relationship with characteristics of work environment (which in itself has been shown to be causing stress as summarized above), individual’s personal environment (other life demands, individual’s assessment of stress agents) and individual’s exposure to primary and secondary trauma in work setting (Fig. 1).

Fig. 1
figure 1

Theoretical conceptualization of professional quality of life constructs in relation to work, client and personal environment. (© Beth HudnallStamm, 2009. www.ProQOL.org)

Full size image

In a country like India, nurse to patient ratio may range from 1:3 to 1:25 [13] depending on hospital setup. In western countries guidelines mandate a NICU nurse to patient ratio of 1:1, allowing 1:3 for immediate care and/or septic isolation. Earlier study done in NICU environment has studied link between NICU admissions and maternal stress levels [16]. However there are no studies pertaining to stress and professional quality of life in NICU nurses in Indian setting. Considering the different local context, lack of earlier studies and potential for adverse impact on neonatal outcomes this study surveyed the nurses working in the NICU and evaluated their perceived stress levels and professional quality of life.

Material and Methods

Nurses from nine hospitals spread out over six main cities of Gujarat participated from six private and three government NICU setups, representative of the best and largest NICU setups in the state. The best private NICU setups in Gujarat were identified by one of the investigators based on previous visits to many of the NICU setups in Gujarat, and with knowledge of patient flow in NICUs and on the level of care provided. All the NICUs provide level 3 care which includes advanced neonatal ventilation and are headed by pediatricians who have been further trained in Neonatology as a subspecialty. Government institutions were selected on the basis of geography in those areas where good private NICU setups were not available. Emphasis was to represent the major areas of the Gujarat state, which is why the top 5 populous cities of Gujarat were selected, the 6th city being chosen by default as the study was conceived there. All nurses included in the study were full time qualified registered nurses with at least 1 y of work experience in NICU work settings. The nurses were diploma holders with General Nursing and Midwifery qualifications with no special qualifications in Neonatology or Pediatrics. Institutional ethics committee approval was obtained from the institution where the study was conceived prior to the conduct of the study. In the participating institutions, permission of NICU in-charge was obtained. Informed consent of all nurses participating was obtained. Of 141 eligible nurses, 133 could be contacted and 129 took either English or Gujarati version of an anonymous self-administered written survey, with facilitation by the investigator. Gujarati version was developed using translation and back-translation method. Face validity and cultural adaptability of translation was judged based on consensus amongst the experts.

Survey included a demographic questionnaire, Perceived Stress Scale 14 (PSS14) [17] and Professional Quality of Life scale Version 5 (ProQOL5) [15].

The demographic questionnaire identified potential stressors in work and personal environment such as age, marital status, number of children, work shift hours, interpersonal relations at work, satisfaction with salary perks and job security. The relationship with peers and colleagues was classified as excellent (no need for improvement), average (some scope for improvement) and poor (significant interpersonal problems).

PSS14 is a valid and reliable measure of the degree to which situations in one’s life are appraised as stressful [18]. As professional quality of life may get affected by nurses’ work and personal environment, the authors used a measure of appraised stress for life in general. Items are rated on a five-point Likert scale from 0 = ‘never’ to 4 = ‘very often’. Scale score ranges from 0 to 56, with higher score indicating higher stress level. The stress score was arbitrarily classified as Mild (less than 1st quartile), Moderate (between 1st and 3rd quartile) and High (more than 3rd quartile) similar to the scheme used by Amr et al. [19].

ProQOL5 is a valid and reliable measure of professional quality of life with 3 subscales viz. compassion satisfaction, burnout and secondary traumatic stress [15]. Compassion satisfaction measured satisfaction in ability to be an effective caregiver. Burnout measured feelings of hopelessness and difficulty in dealing with work. Secondary traumatic stress measured exposure to frightening experiences at work. Thirty statements are rated on a five-point Likert scale from 1=“never” to 5=“very often” based on hours spent at work. Raw scores were converted to t-scores (mean score 50, SD 10) and categorized as high (cut-off 57 or more), moderate or low (cut-off 43 or less) according to ProQOL5 scoring manual. Score as a continuous variable was used for computing correlations. High values on burnout and secondary traumatic stress scale and low values on compassion satisfaction scale are of concern. High value on the compassion satisfaction scale is desirable.

Data was analyzed using Microsoft Excel and SPSS 14 (SPSS Inc., Chicago, IL, USA). Demographic information was depicted using descriptive statistics. The association between perceived stress and domains of professional quality of life was assessed through correlation coefficients. The overall and individual (un-confounded) association of socio-demographic variables with perceived stress was examined using multiple regression.

Results

Out of 129 nurses who participated in the study, most [96(74.4 %)] belonged to private hospitals. The mean (SD) age of the study population was 28.37(8.20) y. The nurse to patient ratio ranged from 1:4 to 1:8 in the private setups and 1:25 to 1:35 in government sector. Majority of nurses (77.2 %) were satisfied with salary benefits while about 16 % feared loss of their current job. The mean (SD) duration of duty hours for the study population was 8.12(0.76) and 43.6 % nurses were attending to more than 4 patients per shift. Most of them were single and reported average to excellent interpersonal relationships at work (Table 1).

Table 1 Demographic and work environment characteristics of nurses
Full size table

The mean (SD) perceived stress level was 22.19(7.18) [Range: 3 to 39]. Most of the nurses [61(47.3 %)] were identified as perceiving moderate stress while 38(29.5 %) perceived mild stress and 30(23.2 %) perceived high stress. High compassion satisfaction, high burnout, and high secondary traumatic stress on ProQOL5 was reported by 25(19.4 %), 30(23.3 %) and 30(23.3 %) nurses respectively (Table 2).

Table 2 Professional quality of life (ProQOL) scale summary
Full size table

PSS14 was negatively correlated with compassion satisfaction (r = -0.28) and positively correlated with burnout (r = 0.44), and secondary traumatic stress (r = 0.24). A multiple regression model with backward elimination method was constructed with perceived stress score as the dependent variable and age, marital status, number of children, satisfaction with salary and perks (yes/no), patients attended (≤4/ >4), relationship with doctors (excellent/average), and fear of losing job (yes/no) as independent variables. The predictive value of the socio-demographic and environmental variables in estimating perceived stress was very poor (r2 < 0.2) with no individual variable except age (p 0.035) and relationship with doctors (p 0.004) explaining more than 5 % variance in the perceived stress. (Data not shown).

Discussion

‘Stress’ should be viewed as a continuum in which an individual goes though feelings of eustress to increasing levels of distress. ‘Eustress’ represents positive aspects of stress; ‘Distress’ represents negative aspects of stress and may lead to negative physiological and psychological effects. Severe and prolonged distress may lead to burnout. Transition from eustress to distress depends upon an individual’s appraisal of stressful situation [20]. The results of the present study show that in this group of relatively young NICU nurses, 29.5 % perceived mild stress, 47.3 % perceived moderate stress and 23.2 % perceived high stress levels. Those perceiving mild to moderate stress on PSS14 might be either experiencing eustress or mild to moderate degrees of distress. Those perceiving high stress on PSS14 were experiencing prolonged distress and might be at higher risk of suffering burnout, secondary trauma or low compassion satisfaction [8, 10]. These findings are corroborated by weak to moderate positive correlation between perceived stress and burnout, secondary trauma and negative correlation between perceived stress and compassion satisfaction. In an earlier study among Indian emergency staff nurses with about 70 % nurses in the 26–30 y age group, 68.2 % perceived moderate stress and 19.1 % perceived high stress [14]. Another study from India assessing a more senior and experienced workforce representing ward, intensive care and operating room nurses reported 66 % perceiving moderate stress and 7.5 % perceiving high stress [13].

Contrary to findings in the studies from western settings, work and environmental factors like effect of duty hours [6, 7], workload [8, 9] and satisfaction with salary perks [11], job insecurity, and interpersonal relations in work place [5, 6, 10–12, 21] played no role on the stress experienced by NICU nurses in the Indian setting.

Perceived stress was positively correlated with burnout and secondary traumatic stress. Almost 25 % nurses screened positive on professional quality of life, burnout and secondary traumatic stress subscales, respectively. This is similar to levels reported in western literature. A survey of over 3000 nurses, nurse practitioners, respiratory care providers and physicians in 44 NICUs in USA, reported 26.9 % burnout in non-physicians [22]. A survey of over 2000 nursing staff covering 165 ICUs in France reported similar burnout (32 %) [23]. Those scoring high on burnout or secondary traumatic stress need to be screened for clinical depression and those scoring high on secondary traumatic stress need to be screened for post-traumatic stress disorder. Perceived stress was negatively correlated with compassion satisfaction, which is in agreement with previous studies [24, 25]. In NICUs, where nurses are the closest care givers to the patient, there is a high likelihood of experiencing death and dying and they are most often inadequately prepared to deal with the emotional needs of patients and their families [26].

The present study has certain limitations. PSS14 is a measure of life stress in general and variance in PSS14 score may be accounted for by other life areas as well. Future studies may be made more rigorous by using mixed methods (qualitative and quantitative) and taking into consideration nursing-job specific stress; stressful life events and other potential stressors in NICU nurses in Indian setting.

Conclusions

Most of the nurses [91(70.5 %)] were identified as perceiving moderate to high stress. NICU nurses in Gujarat experience perceived stress levels and burnout similar to those reported in previous Indian and Western literature, but variance is not explained by work and environmental factors reported in western literature. There is scope for improving professional quality of life in NICU nurses.