Portfolio Management Process

Department of the Premier and Cabinet
Social impact investment – statement of requirements
Note: this statement formed part of the documentation for the South
Australian Government’s Expression of Interest, which closed in
February 2015. This is an archive copy.
BACKGROUND …………………………………………………………………………………………………………3  
GENERAL INFORMATION …………………………………………………………………………………………4  
Social impact bond structure …………………………………………………………………………………4  
Pay by results contracts ……………………………………………………………………………………….4  
Competitive dialogue ……………………………………………………………………………………………4  
Requesting further information ………………………………………………………………………………5  
INFORMATION TO INFORM PROPOSALS IN THE GOVERNMENT’S FOCUS AREAS …….6  
1.   Preventing and reducing hospital admissions/transfers for residential aged care
facility residents…………………………………………………………………………………………………6  
1.1 Nature of the problem ……………………………………………………………………………………..6  
1.2 Size of the problem …………………………………………………………………………………………6  
1.3 Government costs ………………………………………………………………………………………….7  
1.4 Target cohort …………………………………………………………………………………………………8  
1.5 Measurement framework …………………………………………………………………………………8  
1.6 Program requirements …………………………………………………………………………………….8  
2.   Helping long stay older patients in public hospitals find more appropriate
accommodation ……………………………………………………………………………………………….13  
2.1 Nature of the problem ……………………………………………………………………………………13  
2.2 Government cost ………………………………………………………………………………………….13  
2.3 Target cohort ……………………………………………………………………………………………….14  
2.4 Outcomes sought …………………………………………………………………………………………14  
2.5 Measurement framework ……………………………………………………………………………….14  
2.6 Program requirements …………………………………………………………………………………..14  
3.   Preventing and reducing hospital admissions for people with Borderline
Personality Disorder …………………………………………………………………………………………15  
3.1 Nature of the problem ……………………………………………………………………………………15  
3.2 Size of the problem ……………………………………………………………………………………….16  
3.3 Cost of the problem ………………………………………………………………………………………16  
3.4 Target cohort ……………………………………………………………………………………………….17  
3.5 Outcomes sought …………………………………………………………………………………………17  
3.6 Potential savings to Government …………………………………………………………………….17  
3.7 Measurement framework ……………………………………………………………………………….17  
3.8 Program requirements …………………………………………………………………………………..17  
1
4.   Homelessness ………………………………………………………………………………………………….19  
4.1 Nature of the problem ……………………………………………………………………………………19  
4.2 Size of the problem ……………………………………………………………………………………….19  
4.3 Use of homelessness services ……………………………………………………………………….20  
4.4 Government costs ………………………………………………………………………………………..21  
4.5 A current homelessness social impact bond example ………………………………………..25  
4.6 Measurement framework ……………………………………………………………………………….26  
4.7 Program requirements …………………………………………………………………………………..26  
5.   Out-of-Home Care …………………………………………………………………………………………….27  
5.1 Objective ……………………………………………………………………………………………………..27  
5.2 Background …………………………………………………………………………………………………27  
5.3 Size of child protection in South Australia ………………………………………………………..27  
5.4 Cost of the problem ………………………………………………………………………………………29  
5.5 Indicative benefit …………………………………………………………………………………………..30  
6.   Recidivism ……………………………………………………………………………………………………….32  
6.1 The issue …………………………………………………………………………………………………….32  
6.2 Size of the problem ……………………………………………………………………………………….32  
6.3 Government costs ………………………………………………………………………………………..34  
6.4 Target cohort ……………………………………………………………………………………………….34  
6.5 Potential savings ………………………………………………………………………………………….35  
6.6 Program requirements …………………………………………………………………………………..36  
6.7 SA Government data …………………………………………………………………………………….36  
2
BACKGROUND
The information contained in this document is provided to help inform proposals in the
following focus areas:
•
preventing and reducing hospital presentations and admissions for older people
from residential aged care facilities (RACFs)
•
improving outcomes for people experiencing homelessness
•
helping long stay older patients in public hospitals find more appropriate
accommodation
•
preventing and reducing hospital admissions for people with Borderline
Personality Disorder
•
reducing the number of children and young people entering the out-of-home care
system
•
reducing recidivism.
The key driver for a social impact bond or pay by results contract trial in South Australia is to
address social problems that cause significant costs to the community but which could be
potentially avoided or alleviated through earlier interventions.
While this document contains information relating to the above focus areas, proposals in
other areas will be considered if it can be demonstrated that they meet the following criteria:
1.
New or innovative approaches to intervention exist, and there is evidence to
suggest that they will be effective in addressing the problem, based on
experience either locally or in other jurisdictions.
2.
The impact of the intervention program(s) on the focus area can be measured
accurately and with confidence against current or recent historical performance
and cost or a matched control group.
3.
The expected outcomes of the new approach can be demonstrated to:
a. Improve wellbeing for individuals and the broader community.
b. Reduce the cost per intervention on the public sector within a five to seven
year timeframe and deliver financial savings to the State Government.
Proposals may be for new programs or for the expansion of existing programs.
3
GENERAL INFORMATION
Social impact bond structure
Government’s preferred structure is to contract with a single party which could either be the
service delivery agency or a special purpose entity that subcontracts to one or more service
delivery agencies to deliver services to meet agreed outcomes. The contract will specify
payments to be made dependent on the outcomes achieved. Legal agreements with
investors will be managed by the contracting entity.
The successful respondent/s will have ultimate responsibility for performance against the
agreed outcomes. The Government will not have any contractual arrangements with
investors or with sub-contracted service providers as part of the social impact bond trial.
Returns to investors will vary depending on program performance (measured against a predetermined baseline). The Government’s expectation is that investor principal will be at risk
for below baseline performance, while performance at baseline will result in the return of
investor principal only. Returns on investment will be paid for performance above baseline
with higher returns achievable for exceptional performance.
Pay by results contracts
Proposals may be submitted for pay by results contracts with Government without the
involvement of external investors. Pay by results proposals may be for existing contracts
with Government or new contract proposals.
Proposals for pay by results contracts should address all relevant criteria identified in this
statement of requirements and Part C Response Schedule, other than those criteria that
relate to relationships with external investors (bond structure, investor engagement).
At a minimum, pay by results proposals should include information on the target population
and cohort, program location, entry and referral arrangements, outcome measures,
baselines and data availability, how outcome measures are tied to payments, and program
timeframes.
Competitive dialogue
This request for expressions of interest is the first stage of a possible two-stage procurement
process. If the expression of interest stage does not result in proposals that are assessed
as viable, the Government would not proceed to the second stage. If viable proposals are
received, preferred proponent(s) will be invited to engage with Government on a confidential
basis to further develop the details on the technical, financial and legal aspects of their
proposals, including:
•
the target cohort (identify the sub group of the target population to receive
intervention under the social impact investment trial)
•
location(s) of program delivery
•
program referral and entry arrangements (note that Government intends to
participate in program referral arrangements)
•
outcome measurement(s)
•
data to support outcome measurements and program referral
•
bond duration*
•
timeframes for key program milestones
•
bond structure and contractual relationships*
4
•
risk allocation between parties
•
size of bond issue*
•
investor returns and payment structure*
•
investor engagement*
•
baselines, comparison groups
•
intellectual property
•
dispute resolution provisions and termination clauses
•
program evaluation
•
independent audit arrangements
•
program delivery personnel (programs to be delivered by personnel with
appropriate qualifications and approvals)
•
cash flows for program delivery (drawdown requirements).
(* does not apply to a pay by results trial)
Proponents should consider these issues when developing proposals and address them in
response to the expression of interest.
Requesting further information
Should respondents require further data or information to inform their proposals they may
request this information by contacting the nominated contact person until the last queries
date (see Part A Expression of Interest).
The only person authorised by the Principal to communicate with respondents is the Contact
Person. Therefore, respondents cannot rely on communications with any other person. Any
communication with the Contact Person should be in writing and addressed to the Contact
Person.
Requests for data and information will be attended to in accordance with the Bid Rules.
5
INFORMATION TO INFORM PROPOSALS IN THE GOVERNMENT’S
FOCUS AREAS
1. Preventing and reducing hospital admissions/transfers for residential
aged care facility residents
1.1 Nature of the problem
The South Australian Government is seeking to ensure the appropriateness of admission to
hospital for residents of residential aged care facilities (RACFs).
RACF residents may be inappropriately transferred/admitted to hospital for a range of
reasons; including:
•
due to poor understanding of available palliative clinical support options
•
RACF reduced capacity to manage and care for patients at the end of life
•
family preference.
Appropriate care of RACF residents, especially those at end of life, is an expectation of
individuals, families and communities. Contemporary practice supports the minimisation of
acute hospital clinical interventions, as these interventions are ineffective and/or of marginal
value for residents of RACFs.1
The South Australian Government supports the universal application of end-of-life planning
and palliative care provided within RACFs, to optimise end-of-life care. This is to support the
uptake of advance care plans, ensuring individual needs are known and can be supported.
A social impact investment reducing and preventing hospital admissions for RACF residents
may present an opportunity to both improve health outcomes for RACF residents and reduce
preventable hospital admissions.
1.2 Size of the problem
10% of acute hospitalisations are RACF transferred patients. As the elderly population in
South Australia continues to increase, the acute demand for hospital services from RACF
residents is forecast to increase.
In 2012-13 there were 4,710 separations of RACF residents aged 65 and over from
metropolitan public hospitals that had been admitted through emergency departments,
accounting for 25,293 bed days at a cost of $35.2 million.
Transfers to emergency departments and hospitals from RACFs are a common first
response when a resident’s condition deteriorates. It is noted that prior General Practice
assessment is only undertaken in 25% of these cases.
The causes for the transfer of RACF residents vary across individual RACFs. Each entity
has its own individual operational capacity, processes and staffing models. In some
instances, ineffective staffing models result in skill gaps and an inability of the workforce to
support hospital avoidance strategies.
Table 1.1 and Table 1.2 below show metropolitan Adelaide RACF resident hospitalisations
through emergency department, by Service Related Group cost (top 5) and associated
1
Ian Scott (2010) Operating Hospitals Differently, AHHA-AIHPS 2010 Congress Adelaide.
6
separations.2 Table 1.5 and Table 1.6 show all Service Related Group conditions and costs
for that group.
There are opportunities for innovative approaches in the clinical management and/or
preventative service delivery of:
•
Respiratory Infections/Inflammation
•
Other Orthopaedics – Surgical
•
Hip & Knee Replacement
•
Dementia, Delirium & Non-Traumatic Stupor/Coma
•
Non-Acute Rehabilitation
•
Kidney & Urinary Tract Infections
•
Major Psychiatric Disorder
Table 1.1: RACF Patients (age 65+) Top 5 Conditions, Emergency Admissions, Metropolitan,
2011-12
Service Related Group – Condition
Separations
Days
Total Cost
Respiratory Infections/Inflammation
493
2,989
$4,235,833
Other Orthopaedics – Surgical
146
1,394
$3,016,443
Dementia, Delirium & Non-Traumatic Stupor/Coma
170
2,035
$2,747,911
Major Psychiatric Disorder
58
1,775
$1,901,443
Hip & Knee Replacement
76
682
$1,658,739
TOTAL
943
8,875
$13,560,368
2011-12 Metropolitan Hospital Cost and Morbidity Data
Table 1.2: RACF Patients (age 65+) Top 5 Conditions, Emergency Admissions, Metropolitan,
2012-13
Service Related Group – Condition
Separations
Days
Total Cost
Respiratory Infections/Inflammation
538
2,861
$3,640,709
Other Orthopaedics – Surgical
155
1,319
$2,675,451
Dementia, Delirium & Non-Traumatic Stupor/Coma
231
2,227
$2,559,144
Kidney & Urinary Tract Infections
247
1,226
$1,631,899
Hip & Knee Replacement
77
655
$1,504,443
1,248
8,288
$12,011,646
TOTAL
2012-13 Metropolitan Hospital Cost and Morbidity Data
1.3 Government costs
The total cost of clinical care for RACF residents aged 65 and over that had been admitted
to metropolitan hospitals through emergency departments for 2011-12 was $39,358,193 and
for 2012-13 was $35,238,826. The breakdown by local health network (LHN) is provided
below:
2
Separation numbers are provided as a proxy for hospital admissions.
7
Table 1.3: RACF Patients LHN Metropolitan 2011-12
Local Health Network
Separations
Days
Total Cost
2,275
12,718
$19,440,977
Central Adelaide
Northern Adelaide
468
3,418
$4,438,294
Southern Adelaide
1,450
10,005
$15,478,920
TOTAL
4,193
26,141
$39,358,193
2011-12 Metropolitan Hospital Cost and Morbidity Data
Table 1.4: RACF Patients LHN Metropolitan 2012-13
Local Health Network
Separations
Days
Total Cost
Central Adelaide
2,422
12,323
$17,131,900
Northern Adelaide
1,076
5,961
$6,332,064
Southern Adelaide
1,212
7,009
$11,774,861
TOTAL
4,710
25,293
$35,238,826
2012-13 Metropolitan Hospital Cost and Morbidity Data
1.4 Target cohort
The South Australian Government is seeking appropriate clinical management and/or
preventative services for metropolitan RACF residents, who:
•
are 65 years or older;
•
are likely to be admitted to an acute public hospital service through an
emergency department; and
•
have a usual type of accommodation as a residential aged care facility.
1.5 Measurement framework
•
an overall reduction in the numbers of hospital separations for RACF residents
1.6 Program requirements
Proposals for interventions that target the problem should clearly identify:
•
the particular cohort that will be targeted
•
the nature of the intervention and how this will demonstrate it will meet the
outcomes and potential savings to Government.
8
Table 1.5: RACF Patients (age 65+) All Conditions, Emergency Admissions, Metropolitan,
2011-12
Service Related Group – Condition
Abdominal Pain
Ami W/O Invasive Cardiac Inves Proc
Anal & Stomal Procs
Back & Neck Procedures
Bronchitis & Asthma
Cellulitis
Chest Pain
Cholecystectomy
Chronic Obstructive Airways Disease
Colonoscopy
Conisation, Vagina, Cervix & Vulva Procedures
Coronary Bypass
Craniotomy
Dementia, Delirium & Non-Traumatic Stupor/Coma
Dental & Oral Disease Excluding Extractions
Dermatology
Diabetes
Digestive Malignancy
Disorders Of Liver, Biliary Tract & Pancreas
Drug & Alcohol
ERCP
Extensive Burns, Medical
Extensive Burns, Surgical
Gastroscopy
GI Obstruction
Haematological Surgery
Head Injuries
Headache
Heart Failure & Shock
Hip & Knee Replacement
Injuries – Non-Surgical
Injuries To Limbs – Medical
Invasive Cardiac Inves Proc
Kidney & Urinary Tract Infections
Lymphoma & Non-Acute Leukaemia
Major Psychiatric Disorder
Major S & L Bowel Procs Incl Rectal Resection
Microvascular Tissue Transfer Or Skin Grafts
Non-Procedural ENT
Non-Procedural Gynaecology
Non-Procedural Neurosurgery
Non-Procedural Ophthalmology
Non-Surgical Back & Neck Problems
Oesophagitis, Gastroent & Misc Digestive System Disorders
Or Procedures For Injuries
Other Cardiology
Other Cardiothoracic Surgery
9
Separations
32
68
4
2
4
75
86
4
150
15
1
1
3
170
8
17
32
13
44
2
13
1
2
58
44
5
55
3
143
76
145
142
10
227
10
58
16
9
36
5
24
19
45
53
4
128
3
Days
83
319
41
72
24
450
132
52
805
116
2
11
83
2,035
10
96
154
105
272
8
103
3
6
362
191
101
265
4
881
682
456
570
55
1,100
121
1,775
313
114
101
13
209
71
277
200
42
470
36
Total Cost
$124,565
$445,952
$63,224
$156,755
$30,005
$557,685
$222,725
$93,945
$1,096,998
$164,038
$4,661
$37,074
$156,981
$2,747,911
$15,366
$136,170
$224,215
$140,704
$414,536
$8,623
$143,680
$6,802
$15,674
$559,952
$281,938
$171,613
$406,908
$8,750
$1,127,482
$1,658,739
$623,505
$807,970
$92,645
$1,530,657
$191,065
$1,901,443
$579,926
$183,733
$148,303
$19,298
$329,036
$99,506
$389,329
$295,405
$66,048
$771,994
$52,330
Service Related Group – Condition
Other Endocrinology & Metabolic Dis
Other Eye Procedures
Other Gastroenterology
Other General Medicine
Other General Surgery
Other Haematology
Other Interventional Cardiology
Other Medical Oncology
Other Neurology
Other Non-Procedural Urology
Other Orthopaedics – Non-Surgical
Other Orthopaedics – Surgical
Other Plastic & Reconstructive Surgery
Other Procedural ENT
Other Psychiatry
Other Renal Medicine
Other Respiratory Medicine
Other Upper GIT Surgery
Other Urological Procedures
Other Vascular Surgery Procedures
Otitis Media & URTI
Percutaneous Coronary Angioplasty W Ami
Percutaneous Coronary Angioplasty W/O Ami
Peripheral Vascular Disease Incl Skin Ulcers
Poisoning/Toxic Effects Of Drugs & Other Substances
Post-Operative Infections & Sequelae Of Treatment
Red Blood Cell Disorders
Renal Dialysis
Renal Failure
Respiratory Infections/Inflammation
Respiratory Neoplasms
Rheumatology
Seizures
Septicaemia & Other Infectious Diseases
Stroke
Syncope & Collapse
TIA
Tracheostomy
TURP
Ungroupable
Unstable Angina
Urinary Stones And Obstruction
Viral Illness
TOTAL
2011-12 Metropolitan Hospital Cost and Morbidity Data
10
Separations
82
2
189
36
14
21
13
18
84
56
20
146
1
2
30
47
142
9
9
38
10
2
2
48
15
19
59
1
58
493
16
40
42
64
105
113
43
5
2
6
16
6
4
4,193
Days
387
9
693
141
171
141
65
92
617
247
139
1,394
3
4
234
202
744
133
68
522
59
9
2
311
53
101
211
1
397
2,989
118
174
154
518
872
291
174
144
21
70
43
15
17
26,141
Total Cost
$539,365
$20,867
$996,238
$185,792
$266,162
$186,587
$212,241
$121,776
$801,114
$344,630
$211,594
$3,016,443
$7,357
$14,500
$244,438
$280,353
$1,184,110
$284,541
$119,422
$910,379
$71,747
$20,706
$11,941
$433,130
$60,552
$140,054
$278,801
$1,215
$548,857
$4,235,833
$172,565
$188,937
$263,476
$777,170
$1,457,580
$398,274
$203,300
$534,930
$36,249
$140,090
$77,603
$28,341
$23,070
$39,358,193
Table 1.6: RACF Patients (age 65+) All Conditions, Emergency Admissions, Metropolitan,
2012-13
Service Related Group – Condition
Respiratory Infections/Inflammation
Other Orthopaedics – Surgical
Dementia, Delirium & Non-Traumatic Stupor/Coma
Kidney & Urinary Tract Infections
Hip & Knee Replacement
Other Respiratory Medicine
Major Psychiatric Disorder
Heart Failure & Shock
Chronic Obstructive Airways Disease
Stroke
Other Gastroenterology
Injuries To Limbs – Medical
Other Neurology
Septicaemia & Other Infectious Diseases
Other Vascular Surgery Procedures
Other Endocrinology & Metabolic Dis
Injuries – Non-Surgical
Other Cardiology
Renal Failure
Cellulits
AMI W/O Invasive Cardiac Inves Proc
Gastroscopy
Syncope & Collapse
Tracheostomy
Other General Surgery
Peripheral Vascular Disease Incl Skin Ulcers
Oesophagitis, Gastroent & Misc Digestive System
Disorders
Head Injuries
Other Medical Oncology
Digestive Malignancy
Other Interventional Cardiology
Other Renal Medicine
Non-Procedural Neurosurgery
Other Cardiothoracic Surgery
Red Blood Cell Disorders
ERCP
Craniotomy
Major S & L Bowel Procs Incl Rectal Resection
Disorders Of Liver, Biliary Tract & Pancreas
Rheumatology
Other Urological Procedures
Seizures
Non-Surgical Back & Neck Problems
Other Psychiatry
GI Obstruction
Colonoscopy
11
Separations
538
155
231
247
77
191
49
174
191
112
205
191
85
65
28
108
188
126
68
79
77
49
139
4
18
50
59
Days
2,861
1,319
2,227
1,226
655
977
1,240
989
887
698
655
638
663
464
349
443
422
362
398
421
372
283
352
96
235
280
282
Total Cost
$3,640,709
$2,675,451
$2,559,144
$1,631,899
$1,504,443
$1,387,898
$1,234,542
$1,124,524
$1,103,792
$904,162
$857,194
$814,664
$811,342
$605,616
$598,651
$578,671
$569,379
$546,896
$535,392
$518,159
$489,719
$466,754
$430,574
$425,973
$400,792
$386,404
$338,832
66
17
18
22
60
25
5
80
20
7
10
37
59
18
41
45
26
41
12
261
141
242
104
221
208
109
199
141
138
147
192
209
130
178
175
239
142
148
$327,894
$320,421
$309,118
$308,049
$294,243
$291,506
$264,953
$257,093
$256,433
$256,405
$248,518
$243,349
$238,428
$231,718
$219,865
$201,794
$201,267
$193,758
$187,940
Service Related Group – Condition
Other General Medicine
Or Procedures For Injuries
Diabetes
Post-Operative Infections & Sequelae Of Treatment
Other Orthopaedics – Non-Surgical
Non-Procedural Ent
Respiratory Neoplasms
Other Non-Procedural Urology
Ungroupable
Poisoning/Toxic Effects Of Drugs & Other Substances
Lymphoma & Non-Acute Leukaemia
Chest Pain
Abdominal Pain
Non-Procedural Ophthalmology
Back & Neck Procedures
Haematological Surgery
Cholecystectomy
Other Upper GIT Surgery
Microvascular Tissue Transfer Or Skin Grafts
Invasive Cardiac Inves Proc
Otitis Media & Urti
TIA
Percutaneous Coronary Angioplasty W AMI
Other Eye Procedures
Bronchitis & Asthma
Unstable Angina
Other Haematology
Appendicectomy
Wrist & Hand Procedures Incl Carpal Tunnel
Urinary Stones And Obstruction
Other Procedural ENT
Coronary Bypass
Dermatology
Skin, Subcutaneous Tissue & Breast Procedures
TURP
Headache
Non-Procedural Gynaecology
Anal & Stomal Procs
Percutaneous Coronary Angioplasty W/O Ami
Lens & Glaucoma Procedures
Viral Illness
Dental & Oral Disease Excluding Extractions
Drug & Alcohol
Extensive Burns, Medical
Other Plastic & Reconstructive Surgery
TOTAL
2012-13 Metropolitan Hospital Cost and Morbidity Data
12
Separations
47
7
32
25
27
51
10
52
5
27
14
72
34
17
1
7
6
5
9
10
11
25
2
3
11
17
8
2
6
7
1
1
9
3
1
3
7
2
2
3
3
6
3
2
1
4,710
Days
142
75
128
93
121
115
114
108
67
88
103
107
90
61
34
41
46
62
47
44
63
55
30
32
33
42
33
21
16
23
11
16
26
11
13
5
8
10
2
4
8
9
6
4
8
25,293
Total Cost
$172,509
$170,730
$164,826
$160,023
$155,302
$148,973
$141,943
$136,461
$126,853
$122,695
$121,537
$119,634
$104,592
$102,431
$99,651
$95,276
$94,575
$88,733
$82,701
$73,841
$72,523
$68,964
$65,056
$64,213
$48,436
$48,191
$43,804
$38,618
$36,210
$34,836
$32,772
$30,569
$21,806
$20,466
$17,772
$17,107
$15,050
$13,155
$13,137
$12,999
$11,479
$10,659
$10,220
$7,271
$5,900
$35,238,826
2. Helping long stay older patients in public hospitals find more
appropriate accommodation
2.1 Nature of the problem
There are older people in public hospitals who have completed their acute and post-acute
care and have been assessed for Commonwealth aged care. These people remain in
hospital during the process of securing an appropriate community and/or residential aged
care place. This means that hospital resources are maintaining these patients rather than
providing acute complex care to other cases.
Maintenance care is defined as care in which the primary clinical purpose or treatment goal
is support for a patient with:
•
impairment
•
activity limitation, or
•
participation restriction.
A long stay older patient is defined as a patient who:
•
is over 65 years of age
•
is receiving maintenance care
•
has a length of stay greater than 35 days
•
does not require acute care, and
•
is medically ready for discharge.
Public hospitals do not provide the most appropriate accommodation for these patients.
Their needs are best met in residential accommodation. It also means public hospital beds
will be available for acute hospital care.
Maintenance care patients represent 1% of total public hospital separations, however they
contribute to 6% of total bed days across South Australia. The South Australian Government
is interested in ensuring these patients can find and move to the most appropriate
accommodation in a timely manner.
2.2 Government cost
In 2012-13, there were 1,853 separations from metropolitan public hospitals in South
Australia for maintenance care of patients aged 65 years and older. The average cost of
each separation was $12,052. The total cost of these patients in public hospitals was
$22,332,539 (Table 2.1 below).3
Table 2.1: Metropolitan Adelaide Maintenance Care Patients 2012-13
Local Health Network
Separations
Total Days
Total Cost
Central Adelaide
873
8,708
$8,388,228
Northern Adelaide
177
1,678
$1,687,749
Southern Adelaide
803
9,935
$12,256,562
1,853
20,321
$22,332,539
TOTAL
12-13 costs for patient with an episode of care = maintenance care and 65+ years by LHN – Metropolitan
Hospitals
3
Separation numbers are provided as a proxy for hospital admissions.
13
2.3 Target cohort
The South Australian Government is interested in interventions that will target patients in
metropolitan public hospitals that:
•
are 65 years or over (50 years or over for Aboriginal people);
•
have been assessed by an Aged Care Assessment Team as being eligible for
permanent aged care services (residential care or packaged care); and are
unable to return to the community without that care in place; and
•
no longer require inpatient or post-acute care (including rehabilitation) and are
declared medically ready for discharge.
2.4 Outcomes sought
•
long stay older patients who are ready for discharge are placed in the most
appropriate accommodation in a prompt manner.
This will contribute to:
•
improved quality of life, health and wellbeing for individuals
•
improved optimal end-of-life care for individuals and their families
•
appropriate use and allocation of public health resources across South Australia
•
improved capacity to support people to die at home if that is their wish.
2.5 Measurement framework
Specific targets that include:
•
a reduction in the number of maintenance bed days
•
a decrease in the number of separations for patients (65 years+) with
maintenance care and subsequently placed in appropriate accommodation.
2.6 Program requirements
Proposals for interventions that target the problem should clearly identify:
•
the particular cohort that will be targeted
•
the nature of the intervention and how this will demonstrate it will meet the
outcomes and potential savings to Government.
14
3. Preventing and reducing hospital admissions for people with
Borderline Personality Disorder
3.1 Nature of the problem
Borderline Personality Disorder is a highly complex condition with significant negative
consequences characterised by instability across several areas of an individual’s life.
Borderline Personality Disorder is often hard to diagnose, varies amongst consumers in
presentation, complexity and severity and can affect individuals across all social groups. The
potential consequences of the condition include loss of productivity, social and occupational
breakdown, carer stress, family breakdown and risk to self and others. Treatment requires a
sustained, flexible and multidisciplinary approach in order to maximise potential for recovery.
The diagnosis of Borderline Personality Disorder often co-exists with depression and
anxiety, eating disorders, psychosis, PTSD, substance misuse, bipolar disorder and a range
of medical conditions. Assessment and treatment requires a structured approach that may
involve multiple agencies. Whilst recognising the challenges, evidence based and effective
interventions are available and present opportunities to enhance service provision through
the implementation of a stepped approach building on the services that already exist. The
interpersonal interactional style that is associated with Borderline Personality Disorder is
such that treatment in a community based low stigma setting can optimise the potential for
recovery and reduce the risk of harm.
Consumers and carers often report that a diagnosis of Borderline Personality Disorder can
result in discrimination and exclusion. People with a diagnosis of Borderline Personality
Disorder have reported feeling “actively excluded from services, not feeling respected, and
their concerns dismissed as not important or their mental illness considered not being severe
enough”.
Borderline Personality Disorder has a marked effect on health services and consumers and
carers who live with the impact of frequent suicidal thoughts as well as self-harming
behaviours. These can be distressing to both consumers with Borderline Personality
Disorder and carers. The threat of suicide generates considerable anxiety for carers
particularly and sometimes these are child carers of adults with Borderline Personality
Disorder.
Emergency Departments are often the first port of call for people with Borderline Personality
Disorder who have engaged in deliberate self-harm so it is important that attending staff are
appropriately trained to deal with these presentations. Consumers have reported
experiencing negative stigma in these environments, that waiting times can be long, that
their injuries are sometimes overlooked, disregarded or trivialised despite the seriousness or
that they are blamed for their distress. This results in consumers feeling dismissed and their
carers and family experiencing high levels of anxiety relating to the survival and wellbeing of
their loved one.
Across South Australia there is a great variability of services provided to people with
Borderline Personality Disorder. Whilst there are some “pockets of excellence” in
metropolitan Adelaide, particularly in the north, north east, outer south and west, there are
also areas with very few or no specific services for consumers with a diagnosis with
Borderline Personality Disorder.
Evidence-based psychological therapies – informed by an individualised management plan
and thorough assessment – can significantly improve symptoms and functions, stability and
growth. Establishing consistency across these service domains and across healthcare
settings through education, service development and linkage stands to deliver benefits
beyond the health care system.
15
3.2 Size of the problem
There were 1,264 separations from public hospitals in South Australia in 2013-14 with a
diagnosis of Borderline Personality Disorder recorded in the primary, secondary or other
field. These patients had 8,326 bed days and the average length of stay was 6.6 days.
As illustrated below, there was an increase of 19.2% in the number of hospital bed days for
people diagnosed with Borderline Personality Disorder between 2009-10 and 2013-14.
Table 3.1: Borderline Personality Disorder separations, length of stay and average length of
stay 2009-10 to 2013-14
Separation FY
Separations
Length of stay (days)
Average LOS (days)
2009/10
1,259
6,985
5.5
2010/11
1,119
7,356
6.6
2011/12
1,384
7,710
5.6
2012/13
1,243
7,853
6.3
2013/14
1,264
8,326
6.6
5
1,341
1.0
0.4%
19.2%
18.7%
Variance (13/14 to 09/10)
% Change
In addition, there were 690 separations in 2012-13 involving self-harm, involving 7,001
hospital bed days (average length of stay 10.1 days).
People with a diagnosis of Borderline Personality Disorder will often present to an
emergency department with physical complaints which, on discharge, will be coded as the
reason for admission (and not the Borderline Personality Disorder). Accordingly the figures
above will not reflect the entire number of people in South Australia with a diagnosis of
Borderline Personality Disorder who have accessed public hospitals.
3.3 Cost of the problem
In 2009 the National Collaborating Centre for Mental Health in the United Kingdom published
“Borderline Personality Disorder: The NICE guideline on treatment and management” which
identified the costs of Borderline Personality Disorder to society in general and to the health
system in particular. The report’s key findings are relevant in the Australian context:
•
Besides functional impairment and emotional distress, Borderline Personality
Disorder is also associated with significant financial costs to the health care
system, social services and the wider society.
•
In comparison with people with other mental illness, those with personality
disorders have been reported to place a high economic cost on society, as
people with Borderline Personality Disorder frequently use intensive treatments,
such as emergency department visits and psychiatric hospital services leading to
higher related health care costs.
•
With the exception of self-help groups those with Borderline Personality Disorder
are more likely to use almost every type of psychosocial treatment and
compared with people with depression, will have used most classes of
medication.
•
Other areas warranting further consideration and which incur significant financial
and psychological costs to broader society include engagement with social
services, housing issues, unemployment and the criminal justice system
interaction. To date, little work relating to people with Borderline Personality
Disorder has been done in these areas.
16
3.4 Target cohort
The Government is interested in interventions that will target people in country and
metropolitan South Australia who:
•
have a diagnosis of Borderline Personality Disorder from a psychiatrist
•
attend a public hospital emergency department with a principal or secondary
diagnosis of Borderline Personality Disorder and meet agreed criteria for
complexity, severity and frequency of attendance, and
•
may have received services from community based mental health care teams.
3.5 Outcomes sought
People diagnosed with Borderline Personality Disorder receive the appropriate services in a
timely manner. This will contribute to:
•
reduced emergency department attendances
•
reduced acute crisis interventions
•
diminished distress levels among patients
•
joined up services providing care across the patients’ needs
•
better outcomes for carers and family members.
3.6 Potential savings to Government
It is increasingly understood that service optimisation actually leads to direct and indirect
savings to public health services as well as decreasing adverse outcomes to this group in
relation to hospital presentations, self-harm and suicide. Borderline Personality Disorder that
is not managed early can result in ongoing acute interventions occurring over a significant
number of years, hence reducing adverse outcomes early will prevent future ongoing public
health service use.
There are also longer-term benefits to broader society including reduced engagement with
social services, housing issues, unemployment and the criminal justice system interaction.
3.7 Measurement framework
The trial group would be measured relative to a control group benchmark or to a
proportionate comparable average of the previous year.
Specific targets that could be sought include:
•
a reduction in the number of emergency department attendances and hospital
bed days for people presenting with Borderline Personality Disorder.
3.8 Program requirements
Proposals for interventions that target this problem should clearly identify:
•
how they reach the proposed target cohort (including options to target people
with Borderline Personality Disorder who may not have been diagnosed or coded
correctly)
•
the geographical area that will be targeted
•
referral, intake and discharge mechanisms
17
•
the nature of the intervention and how this will demonstrate it will meet the
outcomes and potential savings to Government
•
partnership arrangements to facilitate appropriate service delivery and referral,
sector development, consistency of referral pathways and coordination of service
delivery and response.
18
4. Homelessness
4.1 Nature of the problem
The Government is seeking to improve the outcomes for people experiencing homelessness
in South Australia.
Governments across Australia currently fund specialist homelessness services to pursue a
number of outcomes relating to homelessness, including that:
•
fewer people will become homeless and fewer of these will sleep rough
•
fewer people will become homeless more than once
•
people at risk of or experiencing homelessness will maintain or improve
connections with their families and communities, and maintain or improve their
education, training or employment participation
•
people at risk of or experiencing homelessness will be supported by quality
services, with improved access to sustainable housing.4
Studies of homelessness have found that homeless people incur a much higher than
average cost for non-homeless services, most notably health care and justice services.5 6 7
Homelessness program evaluations have concluded that non-homeless service costs can be
reduced through the implementation of programs aimed at improving outcomes for homeless
people.8 9 10 Interventions targeting homelessness may present an opportunity both to
improve outcomes relating to homelessness and create savings to government through the
decreased use of non-homelessness services.
4.2 Size of the problem
On the night of the 2011 Census 5,982 people were recorded as homeless in South
Australia (56.5% male, 43.5% female). This represented a 6.7% increase from the 2006
Census, which was similar to the 5.4% increase seen in the total South Australian
population. Around 4% were recorded as sleeping rough or in improvised dwellings; the
most common accommodation arrangements on Census night were staying in supported
accommodation for the homeless and living in ‘severely’ crowded dwellings (see Table 4.1).
4
SCRGSP (2014) Report on Government Services 2014. Steering Committee for the Review of Government
Service Provision, Productivity Commission, Canberra.
5
Zaretzky, K., Flatau, P., Clear, A., Conroy, E., Burns, L. and Spicer, B. (2013) The cost of homelessness and
the net benefit of homelessness programs: a national study, Findings from the baseline survey. AHURI Final
Report No. 205. Australian Housing and Urban Research Institute, Melbourne.
6
Flatau, P., Conroy, E., Marchant, T., Burns, L., Spicer, B., Di Nicola, K., Edwards, R., Bauskis, A., Athanassios,
M. and Larsen, K. (2012), The Michael Project, 2007 – 2010. New Perspectives and Possibilities for Homeless
Men. Mission Australia, Sydney.
7
Conroy, E., Bower, M., Flatau, P., Zaretzky, K., Eardley, T. and Burns, L. (2014) The MISHA Project: From
Homelessness to Sustained Housing 2010-2013, Research Report, Mission Australia.
8
Ibid.
9
Zaretzky, K. and Flatau, P. (2013) The cost of homelessness and the net benefit of homelessness programs: a
national study, AHURI Final Report No. 218. Australian Housing and Urban Research Institute, Melbourne.
10
Flatau et al. (2012) Op Cit.
19
Table 4.1: Persons in South Australia identified as homeless on 2011 and 2006 Census nights,
by homeless operational group
Homeless operational
group
Number of
homeless
persons –
2006
Census
Proportion
of homeless
persons –
2006
Census
Number of
homeless
persons –
2011 Census
Proportion
of homeless
persons –
2011
Census
436
7.7%
257
4.3%
Persons in supported
accommodation for the
homeless
1,474
26.3%
1,620
27.1%
Persons staying temporarily
with other households
1,328
23.7%
1,389
23.2%
Persons staying in boarding
houses
977
17.4%
975
16.3%
Persons in other temporary
lodging
30
0.5%
27
0.4%
Persons living in ‘severely’
crowded dwellings
1,362
24.3%
1,714
28.7%
Total persons in South
Australia identified as
homeless in Census
5,607
100%
5,982
100%
Persons who are in
improvised dwellings, tents
or sleeping out
Homelessness can take different forms and is more likely to affect particular sub-groups of
the population, including Aboriginal and Torres Strait Islander people, people suffering from
mental illness, people who use drugs and alcohol at high levels, and people experiencing
domestic violence.11
A 2011-2013 Australian longitudinal study into homelessness, Journeys Home, found that
different population subgroups experience different durations of homelessness. Males, both
the relatively young and the relatively old, migrants and people from Aboriginal or Torres
Strait Islander backgrounds were found to experience longer periods of homelessness on
average. The study also found that people are less likely to exit homelessness the longer
they remain homeless.12 The 2010 ABS General Social Survey estimates that of the 1.1
million Australians who had experienced homelessness in the last ten years, 22% were
homeless for longer than six months during their most recent period of homelessness.13
4.3 Use of homelessness services
The 2010 ABS General Social Survey estimates that 40% of people who experienced
homelessness in the last ten years sought assistance from a service provider while they
were most recently homeless.14
In 2013–14, 23,916 persons in South Australian (41% male, 59% female) received support
from a government-funded specialist homelessness agency for a total 30,750 support
11
Scutella, R, Chigavazira, A., Killackey, E., Herault, N., Johnson, G, Moschion, J. and Wooden, M. (2014)
Journeys Home Research Report No. 4, August 2014, Melbourne Institute of Applied Economic and Social
Research, Melbourne.
12
Ibid.
13
Australian Bureau of Statistics (2011) General Social Survey: Summary Results, Australia, 2010, Australian
Bureau of Statistics, Catalogue No. 4159.0, Canberra.
14
Ibid.
20
periods. Around 27% of clients (6,478) were homeless at intake; the remaining clients
(17,438) were classified as at risk of homelessness.15
Many clients belong to a cohort identified as being particularly vulnerable to homelessness,
and many of these clients belong to more than one cohort or ‘client group’ (see Table 4.2).16
Table 4.2: Clients of Specialist Homelessness Services in South Australia, 2013-2014
17
Financial Year, by client group
Client group
Number of
clients
Proportion
of clients
Identified as Aboriginal or Torres Strait Islander
6,331
26.5%
Born overseas
1,870
7.8%
509
2.1%
Identified as having mental health issues
5,974
25.0%
Identified as experiencing domestic or family violence
9,082
38.0%
Aged under 18 years at intake
7,169
30.0%
Total clients of Specialist Homelessness Services in South
Australia, 2013-2014
23,916
100%
Identified as having a disability
Note: Categories do not add to total because individuals may be counted in more than one category
4.4 Government costs
The Department for Communities and Social Inclusion is responsible for the management
and implementation of housing and homelessness programs in South Australia.
The state’s homelessness service sector provides a range of services and support to people
who are homeless or at risk of homelessness. The sector provides integrated support,
where agencies act as gateways for client referral to appropriate support. Client pathways
from intake and assessment to case plan may involve different agencies.
In 2012-13, the Department for Communities and Social Inclusion spent $56.5 million on
homelessness services ($54.0 million on service delivery and $2.5 million administrative
expenditure). This equates to $34.14 per person in the South Australian population, which is
slightly higher than the Australian average of $26.06 per person, and to a cost per client
accessing homelessness services of $2,647, which is slightly higher than the Australian
average of $2,421 per client.18 Funded services include specialist homelessness services
such as Domestic Violence, Aboriginal Family Violence, Aboriginal and Torres Strait Islander
Specific, Child, Youth and Generic, as well as programs such as Common Ground and
Street to Home. In addition, SA Health contributed $919,200 funding during 2013-14 to the
Street to Home outreach service, provided by a multidisciplinary team in the Central
Adelaide Local Health Network.
A recent AIHW report highlights the fact that while homelessness services can be effective,
the complex pattern of needs of clients has a strong influence on client outcomes and the
number of support days to arrive at outcomes.19
15
Department for Communities and Social Inclusion (2014) Homeless 2 Home Specialist Homelessness
Services Agency Contract Management Report, 01 July 2013 – 30 June 2014, Sector Total. Unpublished
report.
16
Australian Institute of Health and Welfare (2014) Housing outcomes for groups vulnerable to homelessness.
Cat. no. HOU 274. AIHW, Canberra.
17
Department for Communities and Social Inclusion (2014) Op Cit.
18
SCRGSP (2014) Op Cit.
19
Australian Institute of Health and Welfare (2014) Op Cit.
21
The cost of providing assistance to the homeless population extends beyond the provision of
homelessness programs. Several recent Australian studies have found that the government
costs associated with non-homeless services accrued by the homeless population are much
higher than that for the general population, particularly in the areas of health care and
justice. The cost estimates for services accessed are fairly consistent between studies; the
costs range from $15,168 to $22,080 higher annually for a homeless person than for an
average person in the general population. Inpatient costs (nights in hospital, mental health
facility or drug and alcohol centre) account for the majority of this difference, and are
estimated to be from $9,000 to $18,000 higher annually for a homeless person.20 21 22
Figure 4.1 (below) illustrates the average annual inpatient costs, other health costs and
justice costs reported in a study of a group of 183 men and women who were involved in
homeless case-managed programs, compared with the same types of government costs for
the general population.23 A summary of the cost differences found in three recent studies
can be found in Table 4.3.
Figure 4.1: Government costs for a group of homeless people and the general population, by
24
cost type
20
Conroy et al (2014) Op Cit.
Flatau et al (2012), Op Cit.
22
Zaretzky et al (2013) Op Cit.
23
Ibid.
24
Ibid.
21
22
Table 4.3: Average differences in annual government-funded costs per person between
homeless and general populations as reported in recent Australian studies
Study
Target cohort
Difference of health costs
between target cohort
and general population
Inpatient
costs
Other
health
costs
Difference of
justice costs
between target
cohort and
general
population
Total difference
of governmentfunded health
and justice
costs between
target cohort
and general
population
MISHA
project –
2010 to
25
2013
Men in NSW
aged 25 years
or older
$8,913
$2,973
$3,241
$15,127
Michael
project –
2007 to
26
2010
Men in Sydney
(NSW)
$17,999
$1,082
$3,000
$22,081
AHURI 2010 to
27
2012
Men and women
involved in
homeless
programs in
NSW, Vic, SA
and WA (a)
$12,002
$2,504
$5,905
$20,411 (b)
(a) Reported cost differences are for case managed clients only. A small number of Day Centre clients also
participated in the study.
(b) Additional costs were also measured but were excluded from table for consistency with other studies.
Additional costs were Welfare and tax forgone, Children placed in care and Eviction, and came to a total cost
difference of $9,037 between target cohort and general population.
According to SA Health data, there were 561 hospital separations for people with no fixed
address in 2012-13, which amounted to 2795 bed days and a total cost of $4,132,759. This
compares with 460 separations in 2011-12, for 3293 bed days and a total cost of
$4,374,813.
Over the course of a lifetime, cumulative costs to government incurred by people who have
cycled in and out of homelessness can be very significant. Baldry et al. (2012) undertook a
case study of 11 such people in New South Wales, using administrative data from a
range of sources, and estimated that lifetime costs for these individuals ranged from
around $900,000 to $5.5 million.28
Several program evaluations have demonstrated the potential for homelessness programs to
generate savings in areas such as health and justice. Table 4.4 lists the annual cost offsets
reported in several recent Australian homelessness program evaluations, broken down by
health (inpatient costs and other health costs), justice, and welfare payments. To calculate
the cost saving generated by the program, the costs incurred in the 12 months prior to the
start of the program (‘baseline’) are subtracted from the costs incurred in a 12-month period
after the program’s implementation. Therefore negative numbers indicate a saving in
government costs, while positive numbers indicate that an additional cost was incurred
compared to baseline.
25
Conroy et al (2014), Op cit
Flatau et al (2012), Op cit
27
Zaretsky et al (2013), Op cit
28
Baldry, E., Dowse, L., McCausland, R., and Clarence, M. (2012) Lifecourse institutional costs of
homelessness for vulnerable groups. Department of Families, Housing, Community Services and Indigenous
Affairs, Canberra.
26
23
Table 4.4: Annual cost offsets achieved by several recent Australian homelessness programs
Study
MISHA
project –
2010 to
2013
Michael
project –
2007 to
2010
AHURI
study 2010 to
2012
Target cohort
12 months
Men
postaged
program
25
years
24 months
or
postolder
program
Short and medium
term accommodation
clients (men)
Emergency
accommodation
clients (men)
Street-based
outreach clients
(men)
Single Men
Single Women
Tenancy Support
Total annual offsets
in health costs (a)(b)
Inpatient
Other
costs(c)
health
costs
Total
annual
offsets in
justice
costs
(a)(b)
Total
annual
offsets in
welfare
payments
(a)(b)
Total
annual
offsets
(a)(b)
-$697
-$481
-$1,064
$1,220
-$1,022
-$5,532
-$1,035
-$1,977
$542
-$8,002
-$12,739
$243
$231
n/a
-$12,265
-$1,014
-$581
$793
n/a
-$802
$4,687
$131
-$5,588
n/a
-$770
$3,773
-$7,220
$866
-$2,074
-$6,447
$146
$418
$229
-$1,390
-$8,919
$4,747
-$1,301
-$1,540
$26
$1,932
(a) Total offsets calculated comparing costs during 12 months prior to baseline to costs during the 12 months
post-program implementation unless stated otherwise.
(b) A negative figure represents a reduction in costs to government; a positive figure (shown in red text)
represents additional costs to government. In the original reports, cost savings were reported in varying
ways. Costs savings for individual studies may be reported differently in this table compared to the original
report for the purposes of consistency.
(c) Inpatient costs include nights in hospital, nights in mental health facility and nights in drug and alcohol centre
The particular cohort targeted by the program appears to have a strong influence on the
program’s likelihood of generating savings. For example, Zaretzky et al. (2013) found that a
cost saving of $8,919 was generated within the cohort of single women, while the cohort of
single men realised a much smaller cost saving of $1,390 and the tenancy support cohort
realised an additional cost to government of $1,932. In all three studies, a key driver of the
costs to government is nights spent in a hospital, mental health facility or drug and alcohol
centre (‘inpatient costs’). A significant saving on inpatient costs is found post-program in
some cohorts, while in others the inpatient costs are higher after the program’s
implementation. To explain this pattern, Zaretzky et al. (2013) report anecdotal evidence that
“a significant proportion of single men do not seek assistance for health-related issues, and
when they do obtain assistance the costs are often high.”
Furthermore, the size of cost offsets appears to change over time. Conroy et al. (2014)
found that annual cost saving at 24 months post-program implementation was $8,002, which
was almost eight times that achieved in the first 12 months post-program implementation,
and which was mostly driven by a large drop in inpatient costs (see Figure 4.2).29
The cost offsets reported below do not take into account the cost of running the program.
Conroy et al. (2014) estimated that it would take 4.36 years of cost offsets for the MISHA
program to become cost neutral, after which time cost savings could be generated.30
29
30
Conroy et al (2014) Op Cit
Ibid.
24
It should be noted that measures of average costs incurred by homeless people can be
strongly influenced by a small number of ‘outliers’ who have a much greater use of services
than the majority of people in that cohort.
Figure 4.2: Average annual cost offset achieved per person by type of cost, at 12 months and
31
24 months post program implementation, MISHA project
4.5 A current homelessness social impact bond example
The London Homelessness Social Impact Bond, launched in November 2012, is the first
social impact bond that has been trialled in the area of homelessness. The bond targets
rough sleepers identified using a database for organisations who work with rough
sleepers in London. The program involves a personalised approach by workers who
help the program participants access existing provisions and achieve sustained longterm positive outcomes.
The London Homelessness Social Impact Bond measures outcomes based on
achievements beyond the baseline for the target cohort and payments are linked to the
savings generated from achievements above baseline levels. The social impact bond
has five goals or outcomes for which metrics have been identified and an associated
payment structure determined (see Table 4.5).
Goals were determined through detailed analysis and modelling using the rough
sleeper database and through consultation with relevant parties. Despite recognition
that reduction in crime was another important potential outcome, this was not
included due to issues in accessing appropriate measurement data.
31
Ibid.
25
Table 4.5: Goals, metrics and payment structure for London Homelessness Social Impact
32
Bond.
Goal
Metric
Reduced rough
sleeping.
Sustained stable
accommodation.
Sustained
reconnection.
Employability and
employment.
Better managed
health.
Reduced number of
individuals rough sleeping
each quarter.
Entry to non-hostel
tenancy, and sustained
for 12 and 18 months.
Confirmed reconnection
outside of the UK.
Level 2 qualification
achieved
Sustained volunteering
Sustained part-time
employment
Sustained full-time
employment.
Reduction in Accident and
Emergency episodes.
Payment Mechanism
Payment Proportion
Payments based on progress
beyond expected baseline.
25%
Payment on entry to
accommodation, and at 12 and
18 month points.
40%
Payment on reconnection and
at 6 month point.
25%
Payment for achievement.
Payments when volunteering
or employment sustained for
13 and 26 weeks.
5%
Payments for reduction in
episodes against baseline data
from Department of Health.
5%
4.6 Measurement framework
Particular outcome measures could include:
•
an increase in the sustainability of tenancies
•
a reduction in episodes of repeat homelessness
•
a reduction in hospital admissions and presentations
•
a reduction in justice services.
4.7 Program requirements
The Government is interested in developing opportunities to fund service delivery through
social impact investment where possible.
Proposals for interventions that target this problem should clearly identify:
32
•
the particular cohort that will be targeted
•
the nature of the intervention and how this will demonstrate it will meet the
outcomes and potential savings to Government.
UK Department for Communities and Local Government (2014) Qualitative Evaluation of the London
Homelessness Social Impact Bond: First Interim Report, September 2014. Department for Communities and
Local Government, London.
26
5. Out-of-Home Care
5.1 Objective
The government is seeking to reduce the number of children and young people entering the
out-of-home care system.
5.2 Background
South Australia’s child protection system is administered by Families SA in the Department
of Education and Child Development, with statutory powers and obligations under the
Children’s Protection Act 1993.
Families SA relies on the community or people who are legally responsible (these people are
called mandated notifiers’ and they include police, teachers, doctors and priests) to tell
Families SA when they suspect children or young people are at risk of or experiencing abuse
or neglect. These reports are called ‘notifications’.
Families SA will then look into the notification and decide what course of action to take.
There are times when it is decided that there is no need to take action at all, other times
advice might be given to the family or families guided to more formal training or advised to
seek assistance from organisations that specialise in helping families cope with day-to-day
issues.
If it is suspected that someone outside the family is responsible for the abuse the relevant
authorities (such as South Australia Police) will be notified.
Sometimes Families SA will need to be intensively involved with a family. In the more
extreme circumstances they will have to remove children from their families because it is not
safe for them. Where they can, they will work with those families to improve the home
environment enough for the children to remain in or return home. During the time children
are not in the care of their parents, Families SA will make sure the child is in a safe and
stable home, preferably with another family member.
When the breakdown of the family is beyond help and children will not be safe no matter
what, an application for the child to be placed under the Guardianship of the Minister is
made. That child will be legally cared for by a responsible person, either another family
member or a foster carer.
5.3 Size of child protection in South Australia
Families SA received 39,733 child protection concern reports in 2012-13, of which 19,120
33
were ‘screened-in’ notifications. Of the 5,333 (children at risk and imminent risk of
significant harm) notifications that were investigated, 2,221cases were substantiated.34
Substantiation occurs when an investigation concludes that there is reasonable cause to
believe that the child has been, is being or is likely to be abused, neglected or otherwise
harmed. Where a notification of abuse is substantiated but a child is not separated – and
also in some cases where an investigation occurs but the abuse is not substantiated –
children will receive intensive family support services. These services are currently provided
by non-government providers on behalf of Families SA.
33
SCRGSP (2014) Report on Government Services 2014 Volume F: Community Services, Steering Committee
for the Review of Government Service Provision, Productivity Commission, Canberra.
34
Department of Education and Child Development (2014), Annual Report 2013, Government of South Australia,
Adelaide.
27
Where Families SA recommends – and court orders35 – that a child needs to be separated,
various out-of-home care types are available, depending on the needs of the child. The
majority of children (86.5% in 2012-13) are placed in family-based care, such as living with
relatives or kin, with foster carers, or other home-based care arrangement.
Figure 5.1: Number of children in out-of-home care by placement type (as at 30 June)
36
(a) 2014 estimated numbers provided by the Department of Education and Child Department
Some children are not able to be accommodated in a family-based type of care, for example
children with special needs or children with complex and challenging behaviour. For these
children, different types of care – at significantly higher cost – are needed.
Aboriginal children continue to be over-represented in all child protection categories in South
Australia and Australia. Aboriginal children represent approximately 3.6% of the total
population of young people in South Australia but accounted for more than 20% of children
in screened-in notifications and more than 30% of children in substantiations.
Analysis of data reported by the Productivity Commission37 and the Australian Institute of
Health and Welfare38 also shows the following in respect to out-of-home care in South
Australia:
•
As at 30 June 2013, there were 2,657 children in out-of-home care.39
•
Between 2008-09 and 2012-13, the rate per 1,000 children in South Australia
(aged 0-17 years old) in such care placements increased from 7.1 to 7.4. Over
the same period, the remaining Australian jurisdictions (except Tasmania) all
reported decreases in their rate per 1,000 children placed in out-of-home care
placements.
•
There is an upward trend in the number of children placed in out-of-home care
placements, with an annual growth rate of 7% over the past five years due to a
35
The Youth Court. Children’s Protection Act
SCRGSP (2014) Op Cit.
37
Ibid.
38
Annual reports published by the Australian Institute of Health and Welfare.
39
Department of Education and Child Development (2014) Op Cit.
36
28
number of systematic factors (i.e. increasingly complex cases and greater public
awareness).
•
Indigenous children comprise 29.6% of all children in such care placements.
•
Typically, about three quarters of children in out-of-home care have been placed
continuously for more than two years and many more have been placed for more
than five years.
•
On average, the majority of children admitted into out-of-home care are under
the age of 10 – 21% under the age of one, 24% for children aged 1 to 4 and 22%
for children aged 5 to 9. The remaining 33% consist of children aged above 10.
•
In contrast, the age distribution of children exiting out-of-home care is higher
than that of children admitted into out-of-home care. On average, 61% of
children discharged from out-of-home care were above the age of 10. This may
indicate that children are being admitted into out-of-home care at a younger age
and staying in these care arrangements for a longer duration.
•
Foster care and relative/kinship care remain the main placement types for
children in out-of-home care arrangements. On average, 86% of all children in
out-of-home care arrangements are placed in these two placement types – 44%
in foster care and 41% in relative/kinship care. The majority of the remaining
children (around 10%) are generally placed in residential care.
Further information about the children and young people in care can be found in the annual
Report on Government Services.
5.4 Cost of the problem
South Australia’s total real recurrent expenditure on out-of-home care services in 2012-13
was $156 million. Since 2003-04, the State’s expenditure on these services has shown an
average annual increase of 19.5%, equating to an increase of around $125 million since
2003-04.40
For every child aged 0-17 in the South Australian population, real recurrent expenditure on
out-of-home services was approximately $435 in 2012-13. Since 2003-04, the real
expenditure per child on out-of-home services in South Australia has shown an average
annual increase of 19%, which is the second highest (below Northern Territory) among all
Australian jurisdictions.
40
SCRGSP (2014) Op Cit.
29
Figure 5.2: Expenditure on out-of-home care per child, 2012-13 and average annual increase
41
since 2003-04
The cost to the state government per child in care in 2013-14 is estimated to be:42
•
$63,600 for foster care;
•
$50,860 for relative and kinship care; and
•
From $174,906 for community residential care to $279,000 for government
residential care, and up to $353,250 for emergency accommodation.
5.5 Indicative benefit
Successful early intervention program could be expected to result in better outcomes for
families and children notified to Families SA, and at the same time increase Families SA’s
capacity to respond to high risk notifications.
Given the costs to government of out of home care identified above, a successful program
could also be expected to result in real savings to government through costs avoided of
families and children escalating to higher risk (and higher cost) investigations and
interventions.
The government benefits to be used in calculating reward payments will be determined by
negotiation with the proponents selected through the EOI process.
There are other economic and social benefits associated with reducing time for children in
OOHC. These may include benefits from reduced future involvement in the child protection
system, increased educational engagement, reduction in offending, increased employment
opportunities and establishment of a more secure home for younger siblings and future
generations.
Final costs and benefits will be determined through collaborative work with the preferred
proponents selected through the EOI process. Intangible benefits may include the value of
41
42
Source: SCRGSP (2014) Op Cit.
Estimated costs provided by the Department of Education and Child Department, Adelaide.
30
the pilot for testing investor willingness and the practicalities of setting and measuring
outcomes for reward payments, as well as the opportunity to test innovative service delivery
approaches in this area.
5.6 Target cohort
The intervention group for a trial social impact investment intervention could be chosen from
a range of cohorts of children who are in or facing the prospect of being placed in out-ofhome-care. In order to meet the selection criteria, some factors to consider in selecting a
cohort would include – but not limited to:
•
families who have been notified to the Child Abuse Report Line (CARL) and
referred to a Families SA office, where a priority response to children at low or
moderate to high risk of harm is required or
•
children or young people who are at risk of entering the child protection system,
but who are not currently on a Supervision, Custody or Guardianship of the
Minister Order
•
the majority of families will have at least one child aged under five years old
•
one third of children and young people will identify as Aboriginal or Torres Strait
Islander.
In considering other cohorts like young people in high cost residential care or young people
transitioning from care, consideration needs to be given to the issue that in South Australia,
Families SA currently case manages all children and young people in care.
5.7 Outcome and measurement
As noted in the government’s discussion paper on social impact investment, the following
outcomes are sought:
•
reduction in the number of families that are the subject of a child protection renotification
•
reduction in the number of children and young people entering out-of-home care
•
preserve and strengthen family relationships to ensure children and young
people reside in a safe and stable home environment.
Specific targets that could be sought include:
•
a reduction in the number of days in care
•
families not receiving a confirmed child protection re-notification for a defined
period after the service has ceased involvement with a family
•
children and young people have been maintained with their family as a result of
the service
•
families have reported that the service has assisted them to achieve their Case
Plan goals.
Data could be collected for an intervention group and a comparison group for the
measurement period, following entry to the intervention. There may be multiple
measurement periods as part of one social impact investment, which the Government
envisages will have a term of five to seven years.
31
6. Recidivism
6.1 The issue
The Government is seeking to reduce recidivism – reoffending by offenders after they are
released from prison or forensic treatment facilities – and in particular, the rate of return to
custody, within a 12 month period, for offenders on short sentences.
6.2 Size of the problem
At 30 June 2013, the overall prison population in South Australia was 2,266 (2,119 males
and 147 females), a 9% increase (189 prisoners) from the previous year.43 Prisoners were
located in one of ten different facilities – nine of which are correctional facilities, as well as
James Nash House which is a forensic mental health facility (see Table 6.1).
Table 6.1: Prisoners in South Australia at 30 June 2013 by location (ABS, 2013)
Males
Yatala Labour Prison
497
James Nash House
13
Cadell Training Centre
158
Port Augusta Prison
459
Port Lincoln Prison
139
Mount Gambier Prison
187
Adelaide Remand Centre
283
Mobilong Prison
326
Adelaide Pre-Release Centre
57
Total
2,119
Females
James Nash House