S.TRUEMAN PhD THESIS 2016

102

with the smallest n = 10 (Reed & Fitzgerald, 2005) and the largest n = 163 (Clark et al., 2005). Slaven and Kisely (2002) had n = 25 but, of this number, only n = 13 (55%) participants were nurses. Three of the studies related to ‘mental health’ while Slaven and Kisely (2002) specifically researched the topic of alcohol misuse, which narrowed its relevance to this review. Slaven and Kisely (2002), Reed and Fitzgerald (2005) and Jelinek et al. (2011) were qualitative studies reporting their thematic analysis of participant interviews; it was only Clark et al. (2005) that utilised a validated questionnaire tool for data collection, enabling the reporting of statistical analysis (Analysis of Variance [ANOVA], un-paired t- tests, SDs, means, p -values and Spearman’s rho correlation coefficients) across three domains of enquiry (therapeutic commitment, role competency and role support). Jelinek et al. (2011) is very limited as it only has three nurses out of n = 16 who are in rural and regional settings and is therefore included on the basis of completeness. Interestingly, while of marginal benefit to this review, the findings mirror many of the other paper’s findings. Slaven and Kisely (2002) is also of limited value due to its small nursing participant numbers and focused subject matter. Clark et al. (2005) through the use of a questionnaire tool, reports extensive and in-depth statistical responses and relationships combined with the greatest number of participants making it the most useful and reliable source of evidence concerning the topic of interest. Generally, rural and remote nurses preferred not to treat or care for mental health clients and felt unsure of their skills irrespective of their years of nursing experience. Nurses often prioritised the physical needs of clients ahead of mental healthcare (Gillette, Bucknell & Meegan, 1996; Fleming & Szmukler, 1992). Rural and remote healthcare practice is largely generalist in nature (Bushy, 2002; Francis & Mills, 2011; Henwood,