I ask Mr. Purcell to introduce Chapter 14.1.
Chapter 14.1 of the report of the Comptroller and Auditor General reads:
14.1 National Treatment Purchase Fund
Introduction
The Waiting List Initiative (WLI) was introduced in 1993 as a short-term initiative to tackle the problem of significant numbers of public patients waiting excessively long periods for elective (i.e. non-emergency) hospital procedures. However, the initiative continued until 2003 by which time it had been funded on an annual basis to a total cost of €290 million. My Value for Money study on the Waiting List Initiative published in November 2003 pointed to the scope for co-ordinating waiting list funding more effectively, and to the fact that up to half of WLI funding was generating activity indistinguishable from activity funded through the normal budgetary processes applicable in the publicly funded health sector.
The National Treatment Purchase Fund (NTPF) was announced by the Minister for Health and Children in April 2002, as a key initiative of the Health Strategy, to treat patients who have been longest on hospital in-patient waiting lists. Funding for the initiative was provided in a distinct Subhead of the Vote for the Office of the Minister for Health and Children, from 2002 onwards. The amounts provided were, €5m (2002), €30m (2003), €44m (2004) and €64m (2005) and these were administered through the Department of Health and Children (the Department) and by the NTPF on an administrative basis until 1 May 2004 when the Minister for Health and Children, in exercise of the powers conferred on him by the Health (Corporate Bodies) Act, 1961, formally established the NTPF as a statutory Health Body.
The remit given to the Fund was to focus on those patients waiting longest for hospital procedures and to purchase treatment for them primarily in the private hospital system in Ireland, Northern Ireland and Britain. It may also make use of any capacity within public hospitals to arrange treatment for patients.
The 2001 Health Strategy set a target that by end of 2004 no public patient would wait longer than three months for treatment. The Strategy visualised the development of a national waiting time database by the proposed National Hospitals Agency. This database would help channel patients awaiting treatment to an appropriate hospital with sufficient capacity. The management and classification of waiting lists was to be reorganised in several important ways and used in the operation of the NTPF.
Waiting lists would
·be categorized by waiting times, broken down to sub-specialty/procedure level
·include the referring GP's name
·be available to GPs
·show consultants' names as an aid to decisions by GPs regarding referrals
·allow GPs to notify significant changes in the medical status of patients and to propose that the priority of a patient awaiting treatment be reviewed.
NTPF Operations
The NTPF arranges treatment for patients waiting longest for treatment. Hospitals have been informed that any NTPF work undertaken in public hospitals must be over and above core funded activity and should not displace the normal duties of these hospitals. While it is the Fund's policy to endeavour to ensure that consultants do not predominantly treat patients from their own public hospital waiting lists, there are exceptions in relation to the treatment of children and for instance, in certain cases for reasons of procedural or patient complexity.
Reference Prices
The Department has in place a programme to collect, categorise and interpret data related to the types of cases treated in the Public Hospital system. This programme — Casemix — categorises each hospital caseload and allows the comparison of activity and costs between different hospitals. One of the benefits of Casemix measurement is the extent to which it provides a common language for service planning, management and development that is meaningful to both clinicians and managers. Currently 37 hospitals participate in the Casemix programme.
NTPF state that Casemix is one tool used as a benchmark by the NTPF in price negotiation when appropriate to the treatments being procured. However, NTPF state that Casemix is not all-inclusive in the context of price negotiation with hospitals, as it does not comprehend the full service provided to NTPF patients in all cases, such as
·routine pre and post operative visits,
·tests required for specific procedures,
·capital costs and depreciation considerations which arise for private hospitals in some instances.
While the Casemix model has not been designed as a pricing benchmark, the Department has stated that it welcomes greater use of the data for monitoring and evaluation purposes and considers that it is best used to raise questions for discussion.
Referrals
Patients can be referred to the NTPF by their GP, hospitals or Consultant, or they can contact the NTPF directly. NTPF has funded the treatment of 23,379 patients up to end of 2004 — 1,920 in 2002, 7,832 in 2003 and 13,627 in 2004. Of the 13,627 patients treated in 2004, 12,762 were referred by public hospitals and 865 came through the lo-call line. Currently, the threshold for eligibility for NTPF treatment is three months waiting on an In-Patient or Day Case Waiting List. Patient permission is required before the NTPF organise treatment arrangements.
If it is necessary to arrange treatment in Britain or Northern Ireland, the NTPF will organise and pay for travel and accommodation for the patient and an accompanying person.
Audit Objectives
The examination sought to ascertain
·how the NTPF determined the price paid for procedures purchased
·the relevance of Casemix costs to NTPF activity
·the proportion of procedures purchased within and without the public health sector
·the contribution made by the NTPF to achieving the objective set out in the 2001 Health Strategy to reduce the numbers of persons waiting for treatment for an unacceptable length of time.
Treatments for the 13,627 patients funded in 2004 fall into 456 procedure groupings. The unaudited financial information recorded in the NTPF Annual Report for 2004 shows expenditure of €40,560,258 on direct patient care expenses in the year.
For the purposes of my examination the top eight procedures by volume were selected for detailed examination. Expenditure on these in 2004 was approximately €15.5 million or 38% of direct patient care expenditure (excluding ancillary costs). These 8 procedures accounted for 3,809 procedures paid for by the NTPF in 2004 or 28% of all treatments funded by the NTPF in that year.
Audit Findings
Price Negotiation
The price paid by the NTPF for treatments purchased from public and private hospitals is agreed by negotiation. The negotiation process culminates in a service agreement between hospitals and the NTPF whereby hospitals agree to provide the NTPF with agreed services. NTPF vets participating consultants or suitability. Only approved consultants are placed on a panel and are permitted to perform procedures or the NTPF.
NTPF stated that for the purposes of price-setting, NTPF uses the tools available to it as guides and benchmarks in seeking competitive prices. These are built into the NTPF comprehensive pricing database and include
·the Casemix system costs
·estimated insurers' prices
·consultant costs based on the insurers' Schedule of Fees
·prices proposed by peer hospitals.
The price negotiation process is a detailed exercise that is influenced by a number of benchmarks, prevailing prices, capacity requirements, complexity requirements and geographic considerations.
NTPF Negotiated Prices for FundedProcedures
I obtained details of the prices negotiated by NTPF for procedures with both private (Ireland and UK) and public sector hospitals. As well as using its other benchmarks in the negotiation of 2004 prices, NTPF had regard to the Department's Casemix Peer Group Report which classifies treatments carried out on patients into high level Diagnostic Related Groups (DRG). The precise treatments actually provided under the negotiated agreements cannot be ascertained until the patients are discharged from the treating hospitals.
Table 42 shows the percentage by which the highest negotiated prices exceeded the lowest prices for the 8 most common procedures arranged by the NTPF in 2004. The most costly procedure negotiated was €15,895 while the least costly amounted to €378. The Department has pointed out that prices agreed by the hospitals may vary according to the cost base of individual hospitals and the nature and age of patients being treated.
I have acceded to a request by the Accounting Officer of the Department not to disclose the prices paid for procedures by the NTPF on the basis that the publication of commercially sensitive information would affect NTPF's negotiating position and as a result its capacity to deliver a value for money service.
Table 42 Comparison of prices achieved by NTPF across the eight most common procedures
Procedure
|
% Highest exceeded Lowest
|
Cataracts
|
Inpatient 87%Day Case 76%
|
Varicose Veins (one leg)
|
Inpatient 61%Day Case 44%
|
Total Hip Replacement (excluding revisions)
|
Inpatient 72%
|
Skin Lesions
|
Inpatient 126%Day Case 217%
|
Coronary Angiogram
|
Inpatient 0%Day Case 20%
|
Total Knee Replacement (excluding revisions)
|
Inpatient 71%
|
Grommets (< 17 years)
|
Inpatient 54%
|
Laparascopic Cholecysectomy
|
Day Case 206%Inpatient 215%
|
Casemix Cost Comparison
In order to evaluate whether the NTPF has procured treatments at the most economically advantageous cost relative to that recorded in the Casemix programme it would be necessary to consider — at national or hospital level — how the price paid compared with the corresponding Casemix cost. This type of comparison has not yet been undertaken by the NTPF. An example of the variation that can occur is illustrated by the procedure labelled in Table 42 as Coronary Angiogram. An analysis of a sample of the discharge details for patients treated under this heading by the NTPF, carried out on its behalf by the ERSI, classified the patients into two discrete DRGs. The Casemix costs for these DRGs varied by some 84%.
It is worth noting that the average NTPF negotiated price for Coronary Angiogram was under half the Casemix cost of the lower of the two DRGs identified. In contrast, the average negotiated price for Grommets was more than twice the national average cost recorded in the Casemix 2005 model for the most likely corresponding DRG.
While acknowledging that Casemix was not designed as a benchmark for price-setting, these variations point to the necessity from a value for money perspective to ensure that the full potential of Casemix is exploited as an evaluation tool. This would help in any assessment of the relative cost effectiveness of NTPF funding of treatment as against other funding arrangements e.g. the allocation of the same funds directly to publicly funded hospitals whose patients have been treated by the NTPF.
Accounting Officer's Response
Regarding apparent variations between the Casemix and NTPF data in relation to costs for particular procedures, the Department of Health and Children has advised that great care should be taken concerning the interpretation of the two sets of data. Casemix operates by classifying hospital patient data into over 600 Diagnosis Related Groups (DRGs). DRGs are the classification of patients into discrete groups which have similar attributes and resource intensity. The Casemix Peer Group Report generates a national aggregated, average cost per case by DRG and excludes capital and depreciation costs. The Department has therefore advised that Casemix provides the costs of treating patients with similar conditions rather than the cost of individual procedures or patients. Furthermore, information on diagnosis was not collected by the NTPF for referred patients. This is an essential variable required for Casemix classification and may affect conclusions drawn regarding prices negotiated by the NTPF and the Casemix costs.
He pointed out that the cost for Grommets in its Casemix model for 2004, based on 2002 costs, was radically reduced following the revision of the Casemix system for the 2005 model. The casemix tool available to the NTPF in 2004 was the 2002 cost data, i.e. 2004 casemix Peer Group Review. Using the 2004 casemix price for grommets would result in the NTPF price amounting to 62% of casemix costs. The Department has acknowledged this apparent anomaly and is actively reviewing it.
Hospital Referral Pattern Analysis
In 2002 the NTPF published a Patient Information Booklet and this was revised and reissued in 2004. Both booklets suggest that it is more likely that a patient will be treated in a private rather than a public hospital
· The NTPF will then proceed and arrange treatment for you in most cases in a private hospital
· In a small number of cases, you may receive treatment within a public hospital in Ireland.
My examination considered the extent to which NTPF procedures were carried out in private or public hospitals and the extent to which a patient was treated in the same hospital from which s/he was referred albeit via the NTPF. The results of this part of my examination are set out in Table 43.
Table 43 Referral Patterns of Hospitals utilising NTPF Services
|
TotalReferred
|
Public toPrivateReferrals
|
Public toPublicReferrals
|
Public toPublic%
|
SameHospital Referrals
|
SameHospitalReferrals%
|
Cataracts
|
1,467
|
950
|
517
|
35%
|
511
|
35%
|
Varicose Veins (one leg)
|
657
|
564
|
93
|
14%
|
79
|
12%
|
Total Hip Replacement (excluding Revisions)
|
460
|
80
|
380
|
83%
|
205
|
45%
|
Skin Lesions (Simple)
|
352
|
86
|
266
|
76%
|
260
|
74%
|
Coronary Angiogram
|
333
|
240
|
93
|
28%
|
89
|
27%
|
Total Knee Replacement (excluding Revisions)
|
205
|
76
|
129
|
63%
|
62
|
30%
|
Grommets (less than 17 years)
|
191
|
26
|
165
|
86%
|
128
|
67%
|
Laparascopic Cholecysectomy
|
144
|
113
|
31
|
22%
|
30
|
21%
|
Total
|
3,809
|
2,135
|
1,674
|
44%
|
1,364
|
36%
|
Of the 3,809 cases examined 1,674 or 44% were carried out in a public hospital. This is consistent with NTPF statistics for all treatments procured by it in 2004. Referrals to public hospitals for treatment ranged from 14% for varicose veins to 86% for Grommets.
An examination of referral patterns for the eight procedures sampled revealed that 36% of procedures (out of a total of 3,809) were carried out in the same public hospital from which the referral had been made. Same hospital referrals over the eight procedures ranged from 12% for varicose vein procedures to 74% for procedures to remove skin lesions.
It was also established that the documentation maintained by the NTPF did not systematically record information relating to the referring consultant and the consultant carrying out the surgical procedure to enable the NTPF to guard against the risk of excessive self-referral. However, the NTPF maintains that the referring and treating consultants are known to it. The NTPF informed me that its new patient management system, implemented in July 2005 incorporates processes to obtain this information in every case and thus strengthen the information base from which the NTPF can operate its monitoring activities.
Given the extent of same hospital referral of waiting list patients as outlined in Table 43, I asked the Accounting Officer to explain how it is possible for each of these hospitals to have a waiting list problem for the procedures in question and, at the same time, a capacity to undertake a substantial number of additional treatments requested by the NTPF.
Accounting Officer's Response
The Accounting Officer informed me that the predecessor of the NTPF Board had agreed that, for 2004, the use of public capacity could account for 30% of total NTPF activity, once public core service planned activity was not compromised. The Department has recently advised NTPF that use by the Fund of public facilities should be limited to 10% of its total referrals for treatment.
He pointed out that there were several reasons why it was imperative to use public capacity for shortening waiting times for surgery. It is acknowledged that minimal paediatrics capacity (in terms of both volume and expertise) exists in the private sector. In order to offer the benefits of NTPF to children there may be no other option but to utilise spare public capacity.
Other situations that compelled the use of public or "in-house" capacity were cases where for reasons of clinical or patient complexity it was clearly best practice to have certain patients treated by their own Consultant in the hospital where they were on the waiting list. Not to have used this facility would have effectively barred this cohort of patients from accessing the NTPF scheme. The NTPF considers that these activities should be excluded from the computation of the referral patterns in Table 43. This would have the effect of disregarding all public hospital to public hospital referrals for Total Hip Replacement, Total Knee Replacement and Grommets.
The Accounting Officer added that according to the Health Strategy, the NTPF might make use of spare capacity in public hospitals and pointed out that elective activity in hospitals does not take place 24 hours per day and 7 days a week. Therefore using theatres and beds outside of normal working hours is one way of creating extra elective capacity. Allowing public hospitals to undertake work under the NTPF initiative also incentivises hospitals to perform extra work and to treat more patients over and above core funded activity.
Same hospital referrals are necessary where the level of expertise provided is not readily available in other hospitals. This expertise is required for complex surgery, in the case of elderly patients and where children are involved. This activity is carried out often by staff working overtime, who come in at weekends or who extend theatre time on occasions.
Waiting Lists
The Department last published quarterly waiting list data to 31 December 2003. Waiting list data for 2003 is summarised in Table 44.
Table 44 Public In-patient Waiting List 2003 Target Specialties Summaries
|
3 to 6 Months
|
6 to 12 Months
|
12 to 24 Months
|
24 Montha Plus
|
Total
|
Number of Adults Waiting for Target Specialties
|
|
|
|
|
|
March 2003
|
3,780
|
4,394
|
2,488
|
2,294
|
12,956
|
June 2003
|
4,023
|
4,327
|
2,308
|
1,944
|
12,602
|
September 2003
|
4,145
|
4,238
|
2,165
|
1,493
|
12,041
|
December 2003
|
4,245
|
4,472
|
2,080
|
1,257
|
12,054
|
|
|
|
|
|
|
Number of Children Waiting for Target Specialties
|
|
|
|
|
|
March 2003
|
485
|
376
|
204
|
117
|
1,182
|
June 2003
|
554
|
368
|
195
|
113
|
1,230
|
September 2003
|
672
|
490
|
168
|
76
|
1,406
|
December 2003
|
483
|
522
|
129
|
61
|
1,195
|
While no waiting list figures have been published in respect of 2004, the numbers of patients awaiting treatment at the end of 2004 will undoubtedly have improved and stabilised relative to what might have been expected as a result of
·The 23,000 treatments purchased by the NTPF
·The residual impact of the final tranche of WLI funding of €44 million in 2003
·Other initiatives implemented by Ministers since 2001.
In May 2004, the Minister announced the transfer of responsibility for the collation and publication of surgical waiting list data to the National Treatment Purchase Fund (NTPF). At the same time the NTPF indicated that in excess of 4,000 patients could be removed from the Department's reported figure and that it was expected that additional removals would result from further validation of the data by the NTPF. In October 2004, the NTPF conducted an analysis of waiting list data from hospitals, which indicated that
·Data focuses on volumes, not length of time patients are waiting
·Statistics had not been validated and were not reconcilable from one period to the next
·Data did not capture changes in patients' status, i.e. treated, temporary unavailability, no longer in need of treatment
·Data were not treated in a consistent manner and could be up to 6 months out of date.
As I was concerned that access to reliable and independently verified data on patients awaiting treatment is essential to be able to assess the impact of the NTPF initiative on the treatment backlog, I sought the Accounting Officer's observations.
Accounting Officer's Response
In his response, the Accounting Officer pointed out that over the period of the Waiting List Initiative, the Department had sought to improve waiting lists reporting by Health Boards and Voluntary Hospitals. The 2001 Health Strategy specifically acknowledged the need to further improve the management of waiting lists and as a result responsibility for waiting lists was transferred to the NTPF.
He confirmed that as a result of NTPF analysis of waiting list data, NTPF decided in December 2004 not to publish waiting list figures, but instead opted to develop a National Patient Treatment Register which would focus on the waiting times of individual patients rather than statistically based waiting lists. It was announced in May 2005 that it was intended that the register would be implemented on a phased basis in 2005. Funding of €1 million has been provided from within the NTPF's allocation for this in 2005.
Waiting list data for this register will be supplied to the NTPF by individual hospitals. The Register will contain patient specific details (including name and contact details for the first time) as forwarded by hospitals. Patients waiting for treatment the longest can now be identified and will be contacted by the NTPF with an offer of treatment. It is intended that the register will
·Track the progress achieved in reducing waiting times
·Be accessible to patients and General Practitioners
·Show patient status
·Provide the healthcare system with an accessible and accurate tool for waiting list data, and a tool that reconciles changes in patient status
·Inform patients and GPs about prospective waiting times and referral choices
·Assist in reducing waiting times to achieve the Health Strategy commitment that all patients on inpatient and day cases waiting lists will be treated within three months.
NTPF point out that the register will be a national on-line system and will capture current patient status. Hospitals will continue to be responsible for validating and changing patient status on the system. NTPF will operate an audit process of the system.