Mr James Downie, CEO, presented on the topic 'IHPA 2017 and beyond' at the Enhancing Performance & Cost Effectiveness in Maternity & Women's Healthcare - Annual Benchmarking Meeting, hosted by Women's Healthcare Australasia on 26 May 2017.

1. IHPA 2017 and beyond
WHA Meeting
James Downie
CEO
Independent Hospital Pricing Authority

2. Strategic intent of ABF
• Transparency
• Value for money
• Independence
• National comparability
• Technical Efficiency
2

3. About IHPA
• Independent of all governments
‒Can not be directed on pricing
• Governed by a 9 member board
• 28 member clinical advisory committee
‒Senior medical, nursing and allied health
• 40 staff
‒Data management, statistical, classification, policy
and comms
• Strong consultation and transparency agenda
3

4. IHPA’s functions
•Set the National Efficient Price
•Classification systems
•Data standards
•Cross border and cost shifting disputes
4

5. Progress so far
• ABF
‒Admitted Acute
‒Subacute
‒Emergency
‒Non-admitted
• Block Funding
‒Community Mental Health
‒Teaching, training and
research
‒Small rural and remote
hospitals
5

6. Significant slowdown in costs
6
3664
3809
4023
4312
4400
4548 4549
4588
2006-7 2007-8 2008-9 2009-10 2010-11 2011-12 2012-13 2013-14
Cost per NWAU
Growth Rate: 4.2%
Growth Rate: 1.1%

7. Benchmarking portal
•ABF generates masses of data
‒ Cost data collection >1,000,000,000
records
•Used properly this data can help improve
the efficiency of hospitals by reducing
variation
•Have to make it accessible at the
hospital level!

8. 8www.ihpa.gov.au

9. 9www.ihpa.gov.au

10. 10www.ihpa.gov.au

11. Bundled pricing
• IHPA prices public hospital
services on an ‘activity
based funding’ (ABF) basis
wherever practicable.
• ABF separately prices
discrete episodes of care.
• Public hospitals may receive
multiple ABF payments for a single
patient in the course of their care.
• In contrast, a bundled pricing approach
involves a single price per patient which
reflects the average cost of care
across multiple episodes and settings.
11 www.ihpa.gov.au

12. intention of bundled pricing
• The intention of a bundled pricing approach:
‒ for resources and funding to be easier for
hospitals to manage
‒ to allow financial flexibility to experiment with
new models of care
‒ to provide transparency on the total cost of
maternity care
‒ to drive a long-term view of good practice.
• IHPA does not intend for bundled pricing to:
‒ prescribe a clinical care pathway
‒ reduce clinically necessary maternity care
‒ impact on care which is unrelated to the maternity episode
12 www.ihpa.gov.au
Activity based funding
Bundled pricing
$
$
$
$
$
$
$
$
$
How can each service be
delivered more efficiently?
What is the most effective
way to deliver care to
the patient?

13. Bundled payment schemes
• Governments are experimenting with novel payment systems to drive value.
• Bundled payments have been piloted for US Medicare/Medicaid since the 1980s.
• Implementation of bundled payments across other care settings is recent.
• In England, the Maternity Pathway Payment System provides a single
‘risk-adjusted’ payment per maternity patient for each stage of care.
• The US Medicare Bundled
Payments for Care
Improvement scheme spans
the admitted acute,
subacute and non-admitted
settings for many conditions.
• There is weak but consistent
evidence that bundled payments are effective
in cost containment without a major effect on quality.
13 www.ihpa.gov.au

14. Background
• IHPA consults with the public on its proposals
through the Consultation Paper on the Pricing
Framework for Australian Public Hospital Services
which is released in June.
• IHPA canvassed bundled pricing in 2015-16,
proposing bundled price weights for some
non-admitted chronic disease services.
• This involved setting a price weight for these
services equivalent to 28 days of service delivery.
• IHPA received strong support and the weights were
included in the National Efficient Price 2015-16.
• In the 2016-17 Consultation Paper, IHPA canvassed
bundled pricing for a broader range of conditions.
14 www.ihpa.gov.au

15. Starting with Maternity care
• In response to the 2016-17 and 2017-18 Consultation Papers, IHPA received stakeholder support for bundled
pricing for other conditions.
• Maternity care was identified as a good starting point given stakeholder support, its materiality to the public
hospital system and as it has a relatively predictable service delivery pathway with clear start and end points
to care.
• In early 2016, IHPA
convened the Bundled
Pricing Advisory Group
to oversee investigatory
work on bundled pricing
for maternity care.
15 www.ihpa.gov.au

16. So what has the advisory group
learnt?

17. Reviewing Service delivery to
maternity patients
• The bundled price for maternity care will reflect the average
cost of service delivery for maternity patients.
• IHPA has reviewed national public hospital data to identify
service delivery patterns for maternity patients.
• In the absence of consistent patient identifiers in national data,
IHPA has linked patients to services based on patient characteristics which are shared across the admitted
and non-admitted data sets.
• This is an interim solution for analysis purposes.
• Jurisdictions will need to submit patient IDs for implementation.
17 www.ihpa.gov.au

18. The Admitted stay for birth
is the costly portion of care
• The main cost of maternity care is the admission for birth.
• Differences in admitted costs between patient groups (particularly caesarean vs. vaginal birth) are large.
• A maternity patient’s Diagnosis Related Group (DRG)
for birth was found to be the strongest predictor of
their admitted costs.
18 www.ihpa.gov.au
Vaginal birth
Vaginal birth, with
operating room procedures
Caesarean birth

19. Non-admitted costs
are relatively
consistent
• The main non-admitted maternity
clinics have relatively similar costs.
• Patients also have a similar
number of ante/postnatal visits.
• There is significantly
less difference in
non-admitted costs
between the most
and least complex
patient groupings
(using DRG).
19 www.ihpa.gov.au
Why do 8.5% of patients
have double the average?
35%
difference

20. Cost Drivers for
Non-Admitted care
• The DRG of a patient’s admission for birth appears to offer the greatest explanatory power for differences in
non-admitted costs.
• IHPA has also found that:
‒ A higher maternal age is associated with higher costs
‒ Greater socio-economic disadvantage, remoteness and being an
indigenous patient are not associated with higher costs… underservicing?
‒ Number of clinic visits and the types of clinics which patients access
are the strongest predictors, but are not patient-based
• IHPA has reviewed the impact on non-admitted costs due to additional diagnoses assigned to patients for
birth. The biggest factors were:
20 www.ihpa.gov.au
Good for patient
loadings!

21. Bundled pricing
options

22. Scope of the maternity bundle
Settings of care:
• Service delivery to maternity patients spans the non-admitted, admitted and
emergency settings of care.
• Only some patients have antenatal admissions, emergency presentations
or postnatal readmissions.
• IHPA does not intend for bundled pricing to financially penalise
hospitals for providing these services to complex patients.
• The Advisory Group has expressed a preference to include the non-admitted setting for antenatal and
postnatal care and the admitted acute setting for birth in the bundle as this covers routine services provided to
all patients.
22 www.ihpa.gov.au

23. Scope of the maternity bundle
Patient groups:
• The Advisory Group has considered which maternity patients (grouped
by DRG) should be included in a bundled pricing approach.
• The Advisory Group has expressed a preference for including all or most maternity patients in a bundled
pricing approach if possible.
• This increases the overall impact on the hospital system and extend the benefits towards service redesign to
all maternity patients.
• This increases the importance of risk adjustment!
23 www.ihpa.gov.au

24. Scope of the maternity bundle
Stages of maternity care:
• The Advisory Group has considered what stages of care should be included in the bundled pricing approach.
• The Advisory Group considers that including
all stages of care appears to offer the greatest
opportunity for service redesign.
24 www.ihpa.gov.au
Antenatal Birth Postnatal
• Opportunity to address
variance in the number
and type of antenatal visits
• Limited value on its own as the price
will continue to reflect DRG pricing.
• Its inclusion provides transparency
on the total cost of patient care and
allows for long-term hospital planning.
• Opportunity to address
underservicing in
postnatal care.

25. design of the maternity bundle
Pricing admitted care:
• The Advisory Group has considered whether it is feasible to bundle patients on the basis of patient factors
other than the DRG.
• The problem with using the DRG is that it splits patients based on
whether they had a vaginal or caesarean delivery.
• IHPA does not consider that this is technically feasible for 2018-19.
• Pricing the admitted portion of care on the basis of a
patient’s DRG for birth is preferred, in the first instance.
• DRGs have been refined over years and avoids the
financial losses which may arise if caesarean and
vaginal delivery patients received the same price.
25 www.ihpa.gov.au

26. design of the maternity bundle
Pricing non-admitted care:
• There are two approaches to pricing the non-admitted portion of care:
26 www.ihpa.gov.au
Too much cost
variation?
Too little cost
variation?

27. design of the maternity bundle
Exclusions:
• The Advisory Group has a preference for excluding ‘very sick patients’, whereby they would be priced under
the existing ABF arrangements.
• Factors under consideration for exclusion include:
‒ Maternal Fetal Medicine: Patients accessing this clinic have a very different clinical profile to other patients.
Removes ~1% of patients.
‒ Clinic visits: The top 10% of costly patients have at least double
the ante/postnatal visits. This may reflect their clinical profile.
‒ Private and ‘shared care’ patients: IHPA cannot include GP
or private obstetrician/midwife services in the bundle.
 IHPA is awaiting MBS data which will identify
the materiality of ‘shared care’ arrangements
27 www.ihpa.gov.au

28. Next Steps

29. Work For the Advisory Group
• At future meetings, the Advisory Group will:
‒ Finalise the design of the bundled pricing approach;
‒ Assess the impact of the bundled pricing options at the state and territory, Local Hospital Network, peer
group and hospital levels;
‒ Address implementation issues, including issues arising from:
o maternity patients with care spanning multiple financial years
o movement of patients between health services
o patients who are treated under ‘shared care’ arrangements
o patients who leave the pathway.
29 www.ihpa.gov.au

30. How would it look?
• Single payment, risk adjusted by DRG, plus other
factors:
‒Diabetes
‒Anaemia
‒Multiple births
• Single patient identifier critical
• Requires good patient level non-admitted data
30

31. 31

32. Pricing for Safety
and Quality
32 Footer appears here

33. Premise
• Australian and international costing studies estimate that adverse
events explain between 12.0% and 16.5% of total costs
• ICD-10-AM data is a rich source of safety and quality data,
currently underutilised
• Literature review
‒ Good evidence that the provision of timely clinical information to
clinicians & managers leads to improvements in patient
outcomes
• Pricing signals:
‒ Provide clear sign that government values safety and quality
‒ Promote discussion of safety and quality systems amongst
clinicians AND managers
33

34. Approach
• Three areas of focus:
‒Sentinel Events
‒Hospital Acquired Complications
‒Avoidable readmissions
• Data provision to clinicians and managers a critical
component of work
34

35. Sentinel Events
35
1. Procedures involving the wrong patient or body part resulting in death or major
permanent loss of function
2. Suicide of a patient in an inpatient unit
3. Retained instruments or other material after surgery requiring re-operation or
further surgical procedure
4. Intravascular gas embolism resulting in death or neurological damage
5. Haemolytic blood transfusion reaction resulting from ABO incompatibility
6. Medication error leading to the death of a patient reasonably believed to be due
to incorrect administration of drugs
7. Maternal death associated with pregnancy, birth and the puerperium
8. Infant discharged to the wrong family

36. Sentinel Events
• From 1 July 2017 no funding for episodes of care with
a sentinel event
• ~100 events per annum (public hospitals)
• Funding impact ~$5 million per annum
36

37. Hospital Acquired Complications
• Coded data differentiates between conditions present on
admission, and those arising during admission
• Measured using CHADx system:
‒ Too much noise
‒ No measure of preventability
• Hospital Acquired Complication:
‒ Developed by clinicians
• Clear criteria:
‒ Preventability
‒ Patient Impact
‒ Cost Impact
‒ Clinical priority
37

38. 38
Pressure injury Gastrointestinal bleeding
Falls resulting in fracture and intracranial
injury
Medication complications
Healthcare associated infection Delirium
Surgical complications requiring
unplanned return to theatre
Persistent incontinence
Unplanned Intensive Care Unit
admission
Malnutrition
Respiratory complications Cardiac complications
Venous thromboembolism Third and fourth degree perineal
laceration during delivery
Renal failure Birth trauma
Hospital Acquired Complications

39. HACs add cost
39
HAC
Incremental
cost
All HACs 8.6%
Pressure injury 13.8%
Falls resulting in fracture or other intracranial injury 1.7%
Healthcare associated infection 8.8%
Surgical complications requiring unplanned return to theatre 10.9%
Unplanned intensive care unit admission
Respiratory complications 15.9%
Venous thromboembolism 12.4%
Renal failure 21.7%
Gastrointestinal bleeding 10.0%
Medication complications 8.2%
Delirium 9.8%
Persistent incontinence 2.3%
Malnutrition 7.4%
Cardiac complications 11.3%
Perineal laceration 23.2%
Neonatal birth trauma 10.8%

40. Rates vary
40

41. Principle referral hospitals
41
0
2
4
6
8
10
12
14
frequency
Raw HAC rate per 100 episodes

42. Age is a driver
42

43. Risk Adjustment Critical
• Patient risk factors:
‒Age
‒DRG
‒Charlson Complexity Score (predicts the one year
mortality for a patient with a range of specific
comorbidities)
‒ICU admission
‒Emergency admission
• Can calculate risk score for every patient
43

44. 44

45. 45

46. 46

47. Preventable Readmissions
• All admissions are currently paid for
• Some evidence of preventable readmissions in system
• Currently no nationally agreed, clinically acceptable list
of readmission causes
• List currently being developed – clinically led, data
driven project
• Possible inclusions:
‒Preventable hospitalisations
‒Readmission for HACs
‒Constipation
47

48. 48 www.ihpa.gov.au

49. www.ihpa.gov.au
49