Re(de)fining success: tenancy issues, provider supports, and tenancy outcomes in an Australian Permanent Supportive Housing programme

Figures & data

Figure 1. A framework to characterize the success of PSH tenancy outcomes.

Table 1. Sample composition.

	Mean/%
Outcome variables
Tenancy duration, in days	1,274
Tenancy ended, %
Yes	34.3
No	65.7
Tenancy outcome, %
Unsuccessful	25.2
Successful	74.8
Explanatory variables
Year of entry, %
2012	36.0
2013/2015	19.9
2016/2018	22.8
2019/2022	21.3
Age at tenancy entry, %
Up to 24 years	19.9
25–49 years	61.2
50+ years	18.9
Gender, %
Woman	51.8
Man	48.2
Ethno-cultural group, %
Non-English-speaking background	14.9
Aboriginal or Torres Strait Islander background	13.7
Neither	71.5
Pre-tenancy living arrangements
Homeless	53.5
Not homeless	46.5
Arrears issues during tenancy, %	17.7
Unit-condition issues during tenancy, %	13.7
Behaviour issues during tenancy, %	27.1
Sustaining tenancy plan, %	36.2
n (Tenancies)	417

Notes: Administrative data collated by Brisbane Common Ground, July 2012–November 2022.

Figure 2. (a) Kaplan Meier survival curve, survival rates by age. Notes: Administrative data collated by Brisbane Common Ground, July 2012 – November 2022. (b) Kaplan Meier survival curve, survival rates by gender. Notes: Administrative data collated by Brisbane Common Ground, July 2012 – November 2022. (c) Kaplan Meier survival curve, survival rates by ethno-cultural group. Notes: Administrative data collated by Brisbane Common Ground, July 2012 – November 2022. (d) Kaplan Meier survival curve, survival rates by previous homelessness. Notes: Administrative data collated by Brisbane Common Ground, July 2012 – November 2022.

Table 2. Results from event-history models of tenancy exits, hazard ratios.

	Model 1	Model 2
Age at tenancy entry (reference: Up to 24 years)
25–49 years	0.56*	0.57*
50+ years	0.38*	0.41*
Woman (reference: man)	1.04	1.04
Ethno-cultural group (reference: neither)
Non-English-speaking background	0.76	0.74
Aboriginal or Torres Strait Islander background	0.74	0.77
Formerly homeless	1.14	1.41*
Arrears issues		1.91*
Unit-condition issues		0.83
Behaviour issues		1.88*
Sustaining tenancy plan		0.36*
n (Tenancies)	417	417
Pseudo R^2	0.01	0.02
Chi^2	40.54	71.86
p (Chi^2)	<0.01	<0.01

Notes: Administrative data collated by Brisbane Common Ground, July 2012 – November 2022. Semi-parametric Cox regression models. The models also control for entry year. Statistical significance: *p < 0.05.

Figure 3. (a) Predicted probabilities from marginal effects, by gender. Notes: Administrative data collated by Brisbane Common Ground, July 2012 – November 2022. Based on Model 1 in Table 3. (b) Predicted probabilities from marginal effects, by previous homelessness. Notes: Administrative data collated by Brisbane Common Ground, July 2012 – November 2022. Based on Model 1 in Table 3.

Table 3. Results from logistic regression models of successful tenancy outcomes, odds ratios.

	Model 1	Model 2
Age at tenancy entry (reference: Up to 24 years)
25–49 years	0.67	0.55
50+ years	1.26	0.68
Woman (reference: man)	1.86*	1.49
Ethno-cultural group (reference: neither)
Non-English-speaking background	1.79	1.46
Aboriginal or Torres Strait Islander background	1.63	1.93
Formerly homeless	0.12*	0.13*
Arrears issues		0.11*
Unit-condition issues		0.85
Behaviour issues		0.21*
Sustaining tenancy plan		0.91
n (Tenancies)	417	417
Pseudo R^2	0.20	0.37
Chi^2	92.05	172.15
p (Chi^2)	<0.01	<0.01

Notes: Administrative data collated by Brisbane Common Ground, July 2012–November 2022. Logistic regression models. The models also control for entry year. Statistical significance: *p < 0.05.

Figure 4. (a) Kaplan Meier survival curve, survival rates by experience of arrears issues. Notes: Administrative data collated by Brisbane Common Ground, July 2012 – November 2022. (b) Kaplan Meier survival curve, survival rates by experience of unit-condition issues. Notes: Administrative data collated by Brisbane Common Ground, July 2012 – November 2022. (c) Kaplan Meier survival curve, survival rates by experience of behaviour issues. Notes: Administrative data collated by Brisbane Common Ground, July 2012 – November 2022. (d) Kaplan Meier survival curve, survival rates by receipt of a Sustaining Tenancy Plan. Notes: Administrative data collated by Brisbane Common Ground, July 2012 – November 2022.

Figure 5. (a) Predicted probabilities from marginal effects, by experience of arrears issues. Notes: Administrative data collated by Brisbane Common Ground, July 2012 – November 2022. Based on Model 2 in Table 3. (b) Predicted probabilities from marginal effects, by experience of behaviour issues. Notes: Administrative data collated by Brisbane Common Ground, July 2012 – November 2022. Based on Model 2 in Table 3.