Journal of Environmental and Public Health / 2012 / Article / Tab 4

Review Article

A Synthesis of the Evidence for Managing Stress at Work: A Review of the Reviews Reporting on Anxiety, Depression, and Absenteeism

Table 4

Effectiveness of SMIs by level and outcome of intervention (results based only on meta-analyses).

InterventionIndividualOrganisationalMixed/unspecified
OutcomeIndividualOrganisationalIndividualOrganisationalIndividualOrganisational

Parks and Steelman, 2008Outcome: absenteeism
Wellness programme effect size (mean difference between those with and without a wellness programme, weighted for sample size)
d = −0.3, 95%  CI: −0.48, −0.22***

Richardson and Rothstein, 2008
GHQ, STAI,
SCL-90, various anxiety measures; TQ4, anxiety measure
by Bending 1956
CBT for overall psychological outcomes combined d = 1.154**
Anxiety d = 2.390***
For psychological outcomes combined d = 0.507***
Anxiety d = 0.611***
Absenteeism d = 0.213Organisational (support groups and participatory action groups) for psychological outcomes combined d = 0.134
Mental Health d = 0.167
Organisational support groups and participatory action groups for absenteeism d = −0.159Stress d = 7.27***
Anxiety d = 0.678***
Mental Health d = 0.441***

Bond et al., 2006More control lead to reduced absenteeism 4 studies d = −.11 95% CI: −.15 to −.08
Small and significant
More support leads to less absenteeism 1 study d = −16 95% CI: −.24 to −.09
Small and significant
Communication leads to less absenteeism d = −.23 Small-to-medium effect and significant

Marine et al., 2006
Measures of state and trait anxiety STAI, GHQ, Beck
Person-directed interventions versus control:
State anxiety: WMD = −9.42, CI: −16.92 to −1.93
Trait anxiety WMD = −6.91; 95% CI: −12.80, −1.01
Findings sustained in medium term:
State anxiety: WMD = −0.831, CI: −11.49 to −5.13
Trait anxiety: WMD = −4.09, CI: −7.6 to −0.58
GHQ: person directed interventions did not reduce symptoms: WMD = −11.87, CI: −27.24 to 3.49
GHQ symptoms reduced following combination of knowledge skills training, programme planning
WMD: −2.9, CI: −5.16 to −0.64.
Anxiety not measured
Other single studies using SCL and GHQ did not change results

Van der Klink et al., 2001
Not listed
CBT on anxiety d = 0.70***
Relaxation on anxiety d = 0.25*
Summation on anxiety d = 0.54***
CBT on depression d = 0.23
Relaxation on depression d = 0.11
Summation on depression d = 0.33**
CBT on absenteeism d = −0.18
Relaxation on absenteeism d = −0.09
Summation on absenteeism d = −0.12
Depression d = 0Absenteeism d = 0Anxiety d = 0.50***
Depression d = 0.59***

Saunders et al., 1996
Measures of state or trait anxiety, STAI. Others not listed
Performance anxiety r = 0.509**
State anxiety r = 0.373**

Penalba, McGuire, Leite, 2009 (electronic searches on 12/5/08
Hand searches 1973–1990)
One primary prevention study Backman (1997): mental imaging training versus control

SCL-90 depression subscale
GHQ
Depression outcome: MD (fixed effect) −2.14, CI: −4 to −0.28 at end point (in favour of intervention)
18 months later: MD = −0.97, CI: −2.43 to 0.49
GHQ outcome: MD = 2.74, CI: 0.78–4.7 (in favour of control)
18 month followup: MD = 1.3, CI: −0.61 to 3.21
One secondary prevention study intervention versus control: Norvell (1993)
Depression: MD= −7.32, CI: −11.79 to −2.85 in favour of intervention
Anxiety: MD = −3.1, CI: −6.94 to 0.6 in favour of intervention
All prevention versus control post hoc selected outcomes:
Depression: (Backman, 1977; Norvell, 1993)
Depression outcome: MD = −0.8, CI: −1.36 to −0.24, in favour of intervention

Martin 2009
Published 1997–2000
Standardised measures, anxiety, depression and composite measure
CES-D
BSI
Depression: SMD = Small but significant effect:
Depression: SMD = 0.28, CI: 0.12–0.44
Anxiety: SMD = 0.29, CI: 0.06–0.51
Single trial of stress management programme: depression: SMD = 0.69Depression: SMD:0.31, CI: 0.1–0.51
Anxiety
SMD: 0.25, CI: 0.03–0.53
Improvement maintained at followup (only 9 of 17 studies report followup) 5 studies had different methods:
Depression in smoking cessation trial: worsening SMD = −0.08 (on CESDscale), no difference in BSI outcomes of depression: SMD = 0.03, anxiety SMD = 0.05

Conn 2009
(electronic searches 1969–2007)
Mood (self report-measure not reported)
Work attendance
Unclear which studies that were reported contributed to the effect sizes, and whether they used individual or organisational interventions
Mood
2 group post test: mean of effect size (MES): 0.13, CI: −0.05 to 0.31 (NS)
2 group pre- and posttest:MES = 0.21, CI: 0.07 to 0.36**
Treatment pre and post: MES + 0.31, CI: 0.22 to 0.4***
Work attendance:
2 group post test: MES = 0.19, CI: 0.11 to 0.27***
2 group pre post test: MES = 0.05, CI: −0.19 to 0.29
Treatment pre post test: MES 0.02, CI: −0.08 to 0.13 (NS)
Workplace interventions had better results, as did intervention in paid time, studies with onsite fitness facilities

Van Wyk (2010)
State and trait anxiety index
Career identity training in one study does not improve anxiety in nurses: mean difference: −0.06, CI: −0.44 to 0.32
Von Baeyer’s 3 session stress management training showed marginal benefit.
Standardised mean difference: −1.45, CI: −2.67 to 0.22
Weir (1997) assessed effect of management intervention to improve process consultation between nurse managers and staff on mean hours absence of staff in a community hospital No difference: mean difference = 20.35, CI: −10.65 to 51.35

Noordik (2010)
Specific outcomes not listed
Absenteeism not given as separate outcome from work function
Effects on Anxiety
2 studies in meta-analysis, SMD = −0.54, CI: −1.26 to 0.16
Group exposure CBT and medication versus only medication: SMD = 0.87, CI: 0.34 to 1.39
Exposure in vivo and medication versus only medication: 1, CI: 0.52 to 1.49
Large effect sized (>0.8) judged to indicate significant result without formal statistical tests

KEY: d = effect size, SMD = standardised mean difference, WMD = weighted mean difference, CI = confidence interval. ***P = 0.001, **P = 0.01, *P = 0.05.
When intervention types are not specified the intervention summed in the respective cell are multiple and too many to list. Bold denotes a statistically significant outcome.

We are committed to sharing findings related to COVID-19 as quickly and safely as possible. Any author submitting a COVID-19 paper should notify us at help@hindawi.com to ensure their research is fast-tracked and made available on a preprint server as soon as possible. We will be providing unlimited waivers of publication charges for accepted articles related to COVID-19.