. 2016 Aug 31;45(1):28–47. doi: 10.1007/s10488-016-0760-3

Table 2.

Evidence from experimental or quasi-experimental studies on the Clubhouse model

Citation		Outcome	Comparison group	Sample	Significance
Randomized Controlled Trials
Beard et al. (1963)**	Proportion Re-hospitalized: Clubhouse participants had a lower proportion re-hospitalized at every time interval during the 2 year follow-up study. By 9 months, 46 % of controls and 28 % of Clubhouse members had been re-hospitalized.		Persons referred to other community services	352 (274 Clubhouse, 78 comparison)	p < 0.01
Beard et al. (1978)**	Proportion Re-hospitalized: A smaller proportion of Clubhouse members were re-hospitalized at 6, 12, and 24 months. The subgroup of Clubhouse members receiving 2 years of reaching out services had a lower proportion re-hospitalized at 5 years.		Persons referred to other community services	333 (252 Clubhouse, 81 comparison); all had been hospitalized in the prior 4 months	p < 0.01 (6 months) p < 0.05 (12 months) p < 0.02 (24 months) p < 0.02 (5 years, subgroup)
	Days Hospitalized: Over 9 years of follow-up, Clubhouse members spent less time in the hospital (39 vs. 50 months).		Persons referred to other community services		p < 0.05
	Proportion Re-hospitalized: A smaller proportion of Clubhouse members were re-hospitalized at 6, 12, 18, and 24 months.		Persons referred to other community services	74 (40 Clubhouse, 34 comparison); all had been hospitalized in the prior 4 months	p < 0.05 (6, 12, 24 months) p < 0.02 (18 months)
Gold et al. (2016)	Global Quality of Life: Clubhouse participants reported greater global quality of life improvement, particularly with the social and financial aspects of their lives, as well as greater self-esteem. Clubhouse participants who worked a competitive job reported greater service satisfaction compared to other Clubhouse participants.		PACT	167 (83 Clubhouse, 84 PACT)	P < .05 Effect Size = 0.00 Self Esteem: (est. = 1.02, SE = 0.41, t = 2.50, p = 0.01) (M = 24.5, SD = 4.7, n = 38 vs. M = 21.3, SD = 5.9, n = 36);
Johnsen et al. (2004)*^a	Employment Days: Clubhouse members in transitional employment positions had a greater number of days worked compared to persons in jobs set aside for mentally ill persons.		Persons receiving Assertive Community Treatment (PACT)	175	p < 0.01, N = 17
Johnsen et al. (2004)*^a	Employment—hourly wage: Clubhouse members in transitional employment positions had greater wage than persons employed in set-aside jobs.		Persons receiving Assertive Community Treatment (PACT)	175	p < 0.01, N = 17
Macias et al. (2001)*^a	Employment rate: A similar percentage of both groups became employed (66 % in PACT and 70 % in Clubhouse).		PACT	166 (80 Clubhouse, 86 PACT)	p = 0.581
Macias et al. (2001)*^a	Employment tenure: There were no significant differences in work duration between Clubhouse and PACT.		PACT	166 (80 Clubhouse, 86 PACT)	Data not reported
Macias et al. (2006)*^a	Employment rate: There was no difference in employment rates between PACT (64 %) and Clubhouse (47 %) or in days to first job.		PACT	174 (58 Clubhouse, 63 ACT)	p = 0.06 (employment rate) p = 0.492 (days to first job)
	Employment hours: Clubhouse members worked more total hours (median 494 vs. 234).				p = 0.040
	Employment wage: Clubhouse members earned more (median $3456 vs. $1252).				p = 0.023
	Employment duration: Clubhouse members worked longer (median 199 days vs. 98 days).				p = 0.048
Schonebaum et al. (2006)*^a	Employment Placement: No significant differences between groups (60 % Clubhouse vs. 74 % PACT.		PACT	170 (86 Clubhouse, 84 PACT)	p = 0.052
	Employment Duration: Clubhouse members worked more weeks per job than PACT (mean of 21.8) vs. 13.1 weeks).				CI = 9.8-16.4; x ² = 6.37, df = 1, p < 0.01
	Employment Wage: Clubhouse members earned more (mean of $7.38/h vs. $6.30/h).				Clubhouse CI = $6.74-$8.02, Pact CI = $6.03-$6.58;, x ² = 7.72, df = 1, p < 0.01
	Employment Positions Worked: No significant differences between groups (2.2 Clubhouse vs. 2.1 PACT).				Estimated at p = 0.676 (full data not provided)
Schonebaum and Boyd (2012)*^a	Employment Duration: Greater Clubhouse Work-Ordered Day participation prior to employment was associated with greater employment duration. Work-Ordered Day hours prior to and during competitive employment had a small significant correlation with each other.		PACT	43	t (36) = 3.38, p < .01 r(41) = .30, p < .05
Matched Designs
Henry et al. (1999)*		Hospitalizations: High Clubhouse attendees experienced a nominally greater decline in number of hospitalizations comparing the first to third years of enrollment.	Persons matched on gender and case management enrollment date)	862 (509 Clubhouse, 353 comparison)	p = 0.080
Henry et al. (1999)*		Emergency Mental Health Encounters: In the first year, high attending Clubhouse members had more emergency encounters than low attending members and the matched group. During the second and third years, high attendees had a significantly greater decline in emergency encounters compared to low attendees and the matched group.		862 (509 Clubhouse, 353 comparison)	p = 0.012 (year 1, high vs. low attendees) p < 0.001 (year 1, high attendees vs. matched) p < 0.001 (all comparisons, years 2 & 3, high attendees vs. low attendees and high attendees vs. matched)
Mowbray et al. (2009)**		Quality of Life: Clubhouse members reported a higher quality of life controlling for demographics, symptomatology, and disability.	Geographically matched centers	31 Clubhouses and 31 Consumer Drop In Centers (>1800 consumers)	Coefficient^b = .15 p = 0.048
Mowbray et al. (2009)**		Recovery Orientation: A greater proportion of Clubhouse members self-reported to be in recovery from mental illness (71 % vs. 52 %).	Geographically matched centers		Coefficient^b = .43 p = 0.004 OR = 1.54
Tsang et al. (2010)**		Employment Rate: A greater proportion of Clubhouse members were employed during the 6-month follow-up (24 % vs. 2 %).	Age and sex-matched individuals from a regional outpatient clinic	92 (46 pairs)	p < 0.01
Tsang et al. (2010)**		Quality of Life: At baseline, Clubhouse members had lower physical health-related QOL. At three and six month follow-ups, Clubhouse members showed improvements in physical, psychological, social relationships and environmental QOL domains.		92 (46 pairs)	p < 0.01 (baseline) Cohen’s d effect size = <0.01, p < 0.001 for all follow-ups (Author reports these results are not significant after Bonferroni correction)
Warner et al. (1999)*		Social Support: A greater proportion of Clubhouse members reported having close friends (92 % vs. 62 %) and someone to rely on when they needed help (100 % vs. 63 %).	Group of patients matched on diagnosis, age, sex, psychiatry history, and previous service use.	76 (38 pairs)	p = 0.002 (close friend) p < 0.001 (someone to rely on)
		Quality of Life: Clubhouse members reported better QOL for finances, legal/safety, and global well-being.			(legal/safety t = 2.18, df = 69, p < 0.01) (finances t = 2.18, df = 69, p < .05), global well-being t = 2.4, df = 74, p < .05)
		Hospitalization: During the first 6 months, a higher proportion of Clubhouse members were hospitalized (13 % vs. 3 %).			p = 0.108
		Employment Rate: A higher proportion of Clubhouse members were employed (45 % vs. 34 %).			p = 0.327
		Employment hours: Clubhouse members worked less hours.			p = 0.003

Evidence on Effects of Clubhouse

* Clubhouse Accredited by Clubhouse International

** Clubhouse Adheres to Standards and/or had Fidelity Check

^aParticipants from a single study, (the EIDP), were used in separate analyses for all five publications: (Macias et al. 2001) n = 166, (Macias et al. 2006) n = 174, Schonebaum et al. (2006) n = 170, (Schonebaum and Boyd 2012; Schonebaum et al. 2006) n = 43, (Johnsen et al. 2004) n = 175, (Gold et al. 2016), n = 167