Table 1.

Characteristics of studies excluded from meta-analysis (included in descriptive synthesis)

Study ID/origin	Design	Intervention/comparison	Outcomes	Results
- (Glover, 1995) [14].	-RTC.	- Three case managers received training from Charles Rapp, the	-Consumer functioning by the Utah Case Management Consumer Assessment Record.	-Study group showed significant improvement in level of functioning compared to the control group P < .01.
-USA	-Participants should havemet the definition of severe persistent mental illness to be included in the study, participants with primary diagnosis of substance misuse or organic mental disorders were excluded from the study.	developer of the strengths model as well as 20 hours of direct supervision for using the strengths model, weekly supervision meetings were held to enforce adherence to the principles of the strengths model, in addition, caseloads were 20 clients. Moreover, Case managers provided services to clients in the community. All of the former adheres to the fidelity of the strengths model scale.	-Hospital admissions were measured 42 months after implementation of the model by reviewing Valley Mental Health Records retrospectively for evidence of hospitalization 18 months before the implementation of the case management and 42 months after initiation of case management.	-At 42 months, 33% of the experimental group and 13% of the control had been hospitalised; differences between groups were not statistically significant.
	- 136 participants were randomly assigned to study and control group; 67 participants in the study group and 69 in the control.	-comparison was traditional case management services.
(Macias, 1997) [15]	- Pre-post quasi experimental study.	-Strengths case management services were community based. Program assessment records focused on achieving clients personal goals which commit to the principles of the strengths model apparent in The fidelity of strengths modelscale.	-Quality of life and functioning variables by the Self Report inventory (Macias and Jackson, 1990) [28].	-Regarding therapists’ assessment of consumer symptomatology, strengths case management group showed significant reduction in symptoms. The MANCOVA produced significant time effect p < .01, focused t test showed attribution of time effect to case management on the symptoms subscale P < .001.
-USA	−97 participants were included in this study; 48 in the study group and 49 in the control group	- Comparison was treatment in usual.	-Depression, anxiety, and somatisation by the Brief Psychological Well-being (Macias and Kinney, 1990) [29].
			-Consumer functioning by Utah Case Management Consumer Assessment Record (CCAR)	-therapists’ CCAR assessment correlates significantly with self-report measures of psychiatric symptomatology p < .05 which correlated significantly with family members’ assessment of consumer symptomatology p < .01.
- (Stanard, 1999) [16].	-Quasi experimental study.	-Case managers for the study group received 40 hours of training on the strengths model. Manipulation check was conducted to determine the effect of training on case managers, findings showed that case managers operated according to the principles of the strengths model.	- Quality of life was measured by the Quality of Life Inventory by (Frisch, 1992) [30].	-Repeated measures ANOVA on QOL showed significant interaction effect p < .05 which indicated that the experimental group showed satisfaction with QOL compared to the control group.
-USA	−44 participants took part in this study; 29 in the study group and 15 in the control group.	-strengths case management versus generalist case management	-Residential Living and Vocational Educational Status by a tool developed during strengths case management training.	-There was no significant effect regarding hospital days p > .05.
			-Hospitalisation Rate and Number of Hospital Days which was reported by case managers for 3 months before the study and during the study.	-Results showed no statistically significant differences between the study and control group regarding psychiatric symptoms p > .05.
			-Symptoms by using Hopkins Symptoms Checklist-90	-regarding hospitalisation rate before and after treatment for the whole sample, chi square test showed no statistically significant difference between both groups p > .05.
				-chi square was performed before and after treatment for residential living outcomes and it showed significance p < .001 suggesting that there was differences in residential treatment favouring the study group.