Longitudinal comparison of depression, coping, and turnover among NHS and private sector staff caring for people with dementia

Relatives caring for people with dementia show high levels of psychological distress and depression.¹ However, the psychological health of staff in private and NHS care facilities for people with dementia has not been evaluated. Staff turnover is often high in these facilities, and mental health could be a contributory factor.

Active coping strategies reduce depression and psychological distress in family caregivers,² and the same could be true for professional carers. We compared the prevalence of psychological distress among professional staff in private sector and NHS facilities and assessed the relation with coping strategies and rates of staff turnover.

Participants, methods, and results

We measured emotional wellbeing (28 item general health questionnaire) and the use of positive coping strategies (active coping, planning, seeking social support, positive reinterpretation, and acceptance-COPE³) in the care staff of private sector residential or nursing homes and NHS continuing care facilities. All nine private facilities with over 30 residents within two catchment areas and all four NHS facilities were asked to participate. Questionnaire data were discounted from one facility in each category because of the poor return rate (<25%), although the staff turnover was calculated in all 12. We compared results using the χ² test and the Mann-Whitney U test. All data were analysed with SPSS (version 9).

A total of 225 staff completed all of the assessments (161/176 (91%) from private facilities and 64/89 (72%) from NHS facilities). There were 48 nurses (27 from NHS, 21 from private homes) and 177 care assistants or senior care assistants (37 from NHS, 140 from private homes).

Forty five (20%) respondents scored ⩾5 on the general health questionnaire (the cut off point for caseness), 35 (22%) from private facilities and 10 (16%) from the NHS (χ² =1.7, df =1, P=0.30). The table gives further details. Staff scoring below 5 on the general health questionnaire were more likely to use positive coping strategies (active coping, Z=2.1, P=0.03). Nurses used active coping (Z=2.6, P=0.008) and planning (Z=2.5, P=0.011) more than the care assistants, although the differences were not significant for private staff (table).

Seventy five (38%) staff in private facilities and 19 (16%) in NHS homes had left in the 12 months before the assessment (χ²=16.4, df=1, P<0.0001). There was no relation between staff turnover and the mean general health questionnaire score (Spearman's r=0.21, P=0.52).

Comment

We found psychological distress in about 20% of professionals caring for people with dementia in private and NHS facilities. This level is low compared with reported frequencies of 50% in other healthcare workers⁴ and relatives caring for people with dementia.¹ Levels of stress in NHS homes were lower than in private facilities (16% v 22%), although the difference was not significant. The study did not have sufficient statistical power to detect a significant difference of this magnitude. Our study confirms previous reports that positive coping strategies protect against psychological distress and indicates that nurses are more likely to use positive coping strategies than care assistants, particularly in NHS settings. This emphasises the potentially important role of nursing staff in developing such skills in care homes.

The assumption that stress is an important factor in the high turnover of professional carers seems to be unfounded. Although staff turnover was high, particularly in the private sector, it was unrelated to the level of psychological distress within individual facilities. Possible explanations for the high turnover include the poor wages and poor career structure. The lower staff turnover rates in the NHS facilities were perhaps because staff had a better sense of “community” as they were part of a larger organisation. Promotion of better links between the NHS and private sector could increase stability and allow efficient delivery of training programmes and the development of mutually supportive staff groups.

Table.

Prevalence of stress among staff in private and NHS residential homes and stress coping mechanisms

	Private homes				NHS homes				Overall comparison of private and NHS facilities
	Care assistant* (n=140)	Nurse (n=21)	P value †		Care assistant (n=37)	Nurse (n=27)	P value†		Private (n=161)	NHS (n=64)	P value‡
No (%, 95% CI) with general health questionnaire score:
⩾5	31 (22%, 15% to 29%)	4 (19%, 3% to 38%)	1.0		7 (19%, 7% to 31%)	3 (11%, 0% to 23%)	0.5		35 (22%, 19% to 25%)	10 (16%, 7% to 25%)	0.3 (χ2=1.7, df=1)
<5	109 (78%, 71% to 85%)	17 (81%, 65% to 97%)			30 (81%, 69% to 93%)	24 (89%, 77% to 100%)			126 (78%, 75% to 81%)	54 (84%, 75% to 93%)
Staff turnover§	—	—	—		—	—	—		75 (38%, 31% to 45%)	19 (16%, 9% to 23%)	<0.0001 (χ2=16.4, df=1)
Median (range) score for coping mechanisms:
Active coping	11 (4-16)	21 (4-16)	0.2		12 (6-16)	13 (7-16)	0.03		11 (4-16)	12 (6-16)	0.008
Planning	11 (2-16)	10 (4-16)	0.2		11 (6-16)	13 (7-16)	0.01		11 (2-16)	12 (6-16)	0.011
Seeking support	12 (3-16)	12 (4-16)	0.8		11 (6-16)	12 (4-15)	0.8		12 (3-16)	11 (4-16)	0.5
Reinterpretation	11 (2-16)	11 (4-16)	0.02		11 (6-16)	13 (8-16)	0.1		11 (2-16)	12 (6-16)	0.03
Acceptance	11 (4-16)	11 (3-16)	0.7		10 (5-16)	12 (9-16)	<0.001		11 (3-16)	12 (5-16)	0.14

^*Care assistants and senior care assistants. ^†Fisher's exact test for general health questionnaire analyses and Mann-Whitney U test for coping mechanism scores. 

^‡χ²test for general health questionnaire and turnover; Mann-Whitney U test for coping mechanism scores. 

^§Staff turnover rate was calculated by using all staff (198 for private homes and 117 for NHS homes) not just those who completed questionnaires. 

Because some of the demographic variables could overlap with those pertaining to coping and psychological distress we carried out logistic regression analyses (with NHS v private facilities as the dependent variable and age, sex, general health questionnaire score, active coping, planning, seeking support, reinterpretation, and acceptance as the independent variables) for nursing staff (seeking support Wald 6.3, P=0.001 and planning Wald 3.3 P=0.07 entered into the equation but not general health questionnaire) and care assistants separately (none of the variables entered into the equation).