Table 2

Findings and appraisals of studies using stress related scale

Russler (1991), USABerg et al (1994), Hallberg and Norberg (1993), SwedenMelchior et al (1996), Netherlands
Symbols •, Seems no problem; T, treatment group; (E), environmental management; P, placebo group; (P), personnel support; C, control group.
Design features
Type of studyRCTPCSPCS
Population57 student nurses2 wards (31 nurses)161 nurses
T19, P19, C19T16, C15T60, C101
Focus of interventionCognitive education + role playing (P)Nursing method change + supervision (E)Nursing method change + support training (E)
Duration1 semester (16h)1 year constantly1 year constantly
Reliability and validity of scalePresent R data reported3 scales, present R&V data reportedPrevious R&V mentioned
Authors’ conclusionEffective when analysed as overall subjectsReduce negative effectsPotentially effective (decreased turnover and stable stress)
Validity of results
RandomisationStrategy not clearN/AN/A
Sample attritionNot clear8, not included in analysisNot reported
BlindingNot givenNot givenNot given
Similarity at start
Confounding factorsEffects of placebo on stress not consideredNot reportedImitation of intervention by C
Authenticity of differences
Main outcome (scale)Not reportedLarger decrease in T than CStable in both T and C
Statistical testANOVAMann-WhitneyANCOVA
Not significant on all subscalesNot significant on most subscalesNot consistent
Way of comparisonOverall analysisTotal score not analysedTotal score not analysed
Within group analysis
Clinical significanceNot reportedNot clearNot clear
Conclusion of appraisalsEvaluation impossiblePotentially effectiveEvaluation impossible