Part 5: Achieving nationally consistent clinical assessment

5.1
In this Part, we discuss the Ministry's and DHBs' progress in carrying out actions relevant to the sub-strategy "introducing nationally consistent clinical assessment" and related matters. The actions to implement the sub-strategy are to introduce nationally consistent assessment tools developed by the specialists who will use them. The tools are to be continuously improved using research. Data collected through the tools is to be used to improve national equity of access to scheduled surgery by targeting funding to patients with the highest needs.

5.2
In assessing progress on these matters, we discuss:

• the purpose of nationally consistent clinical assessment;
• the methods used to select patients for an FSA;
• whether patients are selected consistently;
• prioritisation tools and how they should work;
• whether patients are consistently prioritised for treatment;
• whether patients are treated in priority order;
• auditing selection decisions and priority scores;
• DHBs' and specialists' commitment to, and management of, prioritisation tools to achieve equity; and
• setting minimum service levels for access to scheduled services.

The purpose of nationally consistent clinical assessment

5.3
Nationally consistent clinical assessment is needed to ensure national equity of access to scheduled services. "National equity of access" means that patients with a similar level of need and ability to benefit from treatment will have similar access to scheduled services, regardless of where they live.

5.4
Certain conditions need to exist for this to be achieved. These are that:

• all patients referred to scheduled services are selected and prioritised for treatment using suitable national tools;
• clinically appropriate minimum thresholds for access to scheduled services are set so that patients do not reach a state of unreasonable distress, ill health, and/or incapacity before they receive scheduled services; and
• prioritisation decisions are regularly audited to check that they are consistent.

5.5
Two sets of national tools are needed – one set to select patients for FSAs and the other set to prioritise patients for treatment. These tools would help to ensure that:

• specialists make consistent decisions about patient care;
• patients most in need of services receive them first;
• futile or marginally beneficial scheduled procedures are not provided;
• any extra funding translates into tangibly improved scheduled services for patients; and
• accurate comparisons of access to scheduled services by district can be made and used to inform funding decisions that improve the national consistency of access.

5.6
For the sake of simplicity, in this Part, we discuss only how patients are selected for FSA and prioritised for treatment (even though, under the Strategy, access to specialist advice – not only access to FSAs – was to be increased). Virtual FSAs and specialist advice provided by other means, such as "same day" clinics¹¹ or out-reach services, may also increase patients' and GPs' access to specialist advice.

Methods used to select patients for a first specialist assessment

5.7
We asked specialists how they selected patients for an FSA, and they told us about three main methods. To aid their decision-making, specialists may use any referral guidelines that the DHB has issued, which may or may not have been prepared jointly with health professionals working in hospital and community care settings (see paragraph 4.16).

5.8
The first method involves a single individual, such as a GP liaison¹² or semi-retired surgeon, who assesses all referrals received by a specialty. This individual accepts or declines patients for an FSA based on any criteria that have been agreed. They might use a specific selection tool. The individual might telephone the GP to find out more information about the patient and discuss care or treatment options available in the hospital or the community. If this individual declines a referral, they might tell the GP when the hospital would accept any new referral for that particular patient and/or similar patients.

5.9
The second method is more commonly used when the number of referrals to a department is small. The specialists discuss all the referrals at a weekly meeting. They might use selection tools. The specialists jointly decide which patients they will accept, how quickly they will see each patient, and which specialist will care for the patient. Alternatively, the specialists may agree that one of them will discuss the patient with the GP before deciding how to proceed.

5.10
These first two methods offer similar benefits and disadvantages. Both methods are likely to promote consistent selection even if a selection tool is not used. They also create opportunities to:

• clarify or change any GP referral guidelines;
• provide general information or advice to GPs about managing conditions in primary care; or
• improve the effectiveness and efficiency of the scheduled services system by, for example, identifying circumstances when it would be useful to GPs to order diagnostic tests without involving the specialist.

5.11
However, if decisions about patient care are not recorded using a selection tool, data is not collected that can later be analysed to:

• help improve access to scheduled services;
• inform proposals to reduce scheduled services; or
• compare access to scheduled services between DHBs.

5.12
The third method involves hospital departments allocating the responsibility for selection referrals to the "on-call" specialist, who may or may not use selection tools. On-call specialists generally fit this selection work around their other duties, such as out-patients' clinics, ward rounds, operating theatre sessions, or assessing patients in the emergency department. On-call specialists do not usually set aside specific time to deal with GPs' referrals.

5.13
We found that this method was more commonly used by surgeons and less often by physicians. These surgeons were confident that they and their colleagues selected patients consistently, even when they were not using selection tools. We were not convinced. Because the specialists do not have dedicated time for considering the referrals, we consider that they may be less likely to contact the GP to discuss the patient and provide advice. If the specialists are selecting referrals in the evenings, GPs are less likely to be available to discuss the patients. We consider that this is likely to mean that patients may unnecessarily be accepted for FSAs because surgeons will be cautious about declining referrals, based only on a written referral.

Are patients selected consistently for first specialist assessment?

5.14
Even though different methods are used to select which patients will have an FSA, in our view, consistent access to scheduled services by specialty could be promoted if credible national tools were available. Analysing the data collected would help specialists to know whether consistent access was being achieved throughout the country and what needed to occur to improve consistency, such as refining the tool.

5.15
National tools for selecting patients for an FSA used to be publicly available. They were called "Access Criteria for First Specialist Assessment" and were available from the Ministry's website. The Ministry has:

… removed the guidelines for Primary Care Management; Patient Referral and Management; and Access Criteria for First Specialist Assessment from [its] website. These guidelines have not been reviewed for a number of years and are considered to be potentially out of date.¹³

5.16
To the best of our knowledge, there are currently no up-to-date selection tools that all DHBs use. Some DHBs told us that they had to create and use their own tools because the national ones were out of date. We consider that it is unlikely that patients are consistently selected for FSAs throughout the country. This means that one of the fundamental aspects of the Strategy is not in place. We consider that the Ministry and DHBs need to decide how they will address this matter with the relevant professional groups.

5.17
Some DHBs told us that they consider that all of the relevant tools (selection, prioritisation, and referral forms) should be available online and be easy to use. We are aware that some DHBs have started doing this. However, progress is not uniform throughout the country, and few of the DHBs are working with other DHBs to produce regional or national tools. We are concerned that this is another area where haphazard implementation may produce unnecessary variation for patients and where unnecessary duplication may increase costs for DHBs.

What are prioritisation tools for treatment, and how should they work?

5.18
At an appropriate time in a patient's care pathway, specialists assess a patient's need and ability to benefit from treatment. They are meant to do this using national prioritisation tools.¹⁴ The Ministry's contracts with the DHBs require that the systems to assign priorities are evidence-based, transparent, systematic, and procedurally fair.

5.19
Currently, three sets of prioritisation tools are used – older national tools, more recent national tools, and local (DHB-level) tools. Some prioritisation tools address a single procedure, such as varicose veins, and other tools are used for all patients seen by a specialty, such as the tool for General (internal) Medicine.

5.20
DHBs have to create their own prioritisation tool when one does not exist or the existing national prioritisation tools are out of date. However, DHBs are unable to use local prioritisation tools or processes to comply with ESPI 8 until the Ministry has agreed that they can be used for this purpose. DHBs must convince the Ministry that a proposed local variation to a national prioritisation tool improves outcomes for patients.

5.21
Any improvements at the local level do not lead to an automatic update and reissue of the national prioritisation tool to all DHBs, because the Ministry does not consider this to be practical. The Ministry says that a consensus of experts is needed to update a national tool. This can also involve re-weighting the criteria in each tool. Instead, the Ministry considers changes to a national tool when it is reviewed.

5.22
We consider that it would make more sense for the relevant professional groups to be responsible for the "life cycle" of national tools, including proposed changes to the tools, rather than the Ministry. The professional groups could recommend to the Ministry that it agree to allow DHBs to use a particular tool to comply with ESPI 8. (We discuss commitment to prioritisation in more detail in paragraphs 5.56-5.65.)

5.23
Local tools can take different approaches to assigning scores to patients with similar needs and ability to benefit from treatment, which means that scores cannot be used to compare access throughout the country. For example, some scores have a scale of one to 100, and others have a scale from one to five. In different DHBs, patients with a score of 70 may not have a similar need for, and ability to benefit from, scheduled services (in the same specialty or across specialties). This means that the data that the Ministry holds about patients' priority scores cannot be used to monitor progress towards national equity.

5.24
Specialists exercise their professional judgement in applying the prioritisation tools. They can override the priority score to consider circumstances particular to an individual patient. Because they are using their clinical judgement, specialists can apply tools differently. Research to examine how specialists make these decisions is under way.¹⁵ Nonetheless, the idea behind national prioritisation tools is that they will produce better consistency for patients than if each DHB used a different local prioritisation tool or no tool at all.

5.25
Prioritisation tools should contain standardised medical and social criteria that are systematically applied to give each patient a priority score that reflects their level of health need and ability to benefit from treatment. Although the details of each tool will differ, the tools' criteria should deal with such matters as:

• the severity of a patient's condition or disability,
• a patient's ability to benefit from treatment; and
• any consequences from delaying treatment.

5.26
Ideally, the priority scores would relate to a "clinically appropriate treatment period" so that patients are treated in priority order and are not in unreasonable distress by the time they receive treatment. A clinically appropriate treatment period can be set by considering the latest medical evidence. It can also reflect practical decisions that are necessary to meet policy requirements set by the Government, such as the requirement to treat patients within six months, and thresholds for access to scheduled services. (For example, some patients could wait nine months for surgery without undue distress, but current policy requires that scheduled services that have been offered are delivered within six months.) We acknowledge that it can be difficult to achieve universal agreement among specialists about a clinically appropriate treatment period (partly because, under the Strategy, resources have a role in setting the level of access to scheduled services). We do not consider that this is a reason not to try to promote national equity of access.

5.27
Cardiac surgeons and cardiologists have recently achieved universal agreement on their prioritisation tool for cardiac surgery (part of which is reproduced in Figure 9). These specialists have agreed that all patients who are offered valve surgery or coronary artery bypass grafts should receive their surgery within 90 days, which is well within the maximum period of six months. They use this tool for all patients being considered for cardiac surgery, not only those patients who could be offered scheduled treatment.

5.28
The patient's priority score determines the period in which the patient should or could be treated:

• Patients with a score of 50 or more should be treated as an unscheduled patient in 48 hours or less.
• Patients with scores of 40-49 could be treated as unscheduled or scheduled patients, but should receive their treatment in 10 days or less.
• Patients with scores of 25-39 could be treated in 30 days or less.
• Patients with scores of 24 or less could be treated at any time during the 90-day period.

Figure 9
Part of the prioritisation tool for cardiac surgery

The prioritisation tool that this chart is part of was endorsed for use in May 2010 for patients needing valve surgery or coronary artery bypass grafts. This tool was prepared by cardiac surgeons and cardiologists representing the New Zealand National Cardiac Surgery Clinical Network and the New Zealand Branch of the Cardiac Society of Australia and New Zealand. The Ministry provided specialist expertise to these groups.

5.29
Such a tool enables DHBs to schedule patients for treatment in priority order. The tool clearly demonstrates that there is no problem with treating patients with lower scores quickly as long as patients with higher scores are treated within the relevant period. Patients with lower scores often need treatment that takes less time to perform. Operating theatre sessions can be more efficiently used if patients needing shorter operations "fill up" any time that is left after longer operations.

5.30
Patients are entitled to certainty, so a patient should be told when a DHB cannot offer publicly funded treatment within the clinically appropriate treatment period. The Strategy assumes that the whole public health system is responsible for achieving its objectives, even though DHBs' legislative responsibilities are somewhat narrower. When one DHB cannot provide treatment within a clinically appropriate treatment period, it may be possible for another DHB to do so.

5.31
In practice, we expect that any offer of this sort would mainly be confined to patients with high priority scores who are at risk of suffering irreversible consequences of delayed treatment and who could be safely transferred to a Treating DHB (or a private hospital). Effective and efficient regional collaboration and planning would be needed to ensure that this works well for patients.

Whether prioritisation tools produce scores that have a matching treatment period

5.32
We examined each of the 30 currently available national prioritisation tools to assess whether they could produce a score that had a matching clinically appropriate treatment period. Figure 10 presents the results. Six tools produced both a score and a clinically appropriate treatment period. Another 16 tools produced scores, but did not have matching clinically appropriate treatment periods. Eight tools have a matching clinically appropriate treatment period but do not produce scores. Instead, patients are assigned to a category, such as urgent, semi-urgent, or routine.

Figure 10
List of currently available national prioritisation tools

Specialty	Prioritisation tool	Tool number	Produces a priority score	Has a matching clinically appropriate treatment period	Assigns patients to a category	Tools in the implementation phase/live in NBRS*
Cardiac surgery	1. Aortic regurgitation	9074			-	-
Cardiac surgery	2. Aortic stenosis	9072			-	-
Cardiac surgery	3. Mitral regurgitation	9075			-	-
Cardiac surgery	4. Mitral stenosis	9073			-	-
Cardiac surgery	5. Coronary artery bypass graft	9076		-	-	-
Cardiac surgery	Universal coronary artery bypass graft and valve	**			-	Will be available in NBRS during 2011/12
Cardiology	6. Cardiac catheterisation	9082		-	-	-
Cardiology	7. Percutaneous coronary revascularisation	9081		-	-	-
Gastroenterology	8. Colonoscopy	9162	-	-		-
Gastroenterology	9. ERCP (endoscopic retrograde cholangio-pancreatography)	9163	-	-		-
Gastroenterology	10. Gastroscopy	9161	-	-		-
General (internal) medicine	11. All conditions	9170	-	-		-
General surgery	12. Varicose veins	9131		-	-	Was live in NBRS during 2007/08
Gynaecology	13. Infertility	9063		-	-	-
Gynaecology	14. Sterilisation	9062		–	-	-
Gynaecology	15. All other conditions	9065		–	-	Was live in NBRS during 2007/08
Hospital dentistry and oral maxillo-facial surgery	16. All conditions	9100		-	-	-
Neurosurgery	17. Spinal	9111		-	-	-
Neurosurgery	18. All other conditions	9110	-	-		-
Ophthalmology	19. Cataract	9041		-	-	Was live in NBRS during 2007/08
Ophthalmology	20. All other conditions	9040			-	-
Orthopaedics	21. Hip or knee joint replacement surgery	9011		-	-	Was live in NBRS during 2007/08
Orthopaedics	22. All other conditions	9010		-	-	-
Otolaryngology (ear, nose, and throat)	23. All conditions	9030		-	-	-
Paediatric medicine	24. All conditions	9220	-	-		-
Paediatric surgery	25. All conditions	9120			-	-
Plastic and reconstructive surgery	26. Skin lesion	9051		-	-	Was live in NBRS during 2008/09
Plastic and reconstructive surgery	27. All other conditions	9050		-	-	-
Respiratory medicine	28. All conditions	9240	-	-		-
Thoracic surgery	29. All conditions	9260	-	-		-
Urology	30. Prostatectomy	9090		-	-	-

Some of the tools are "integrated" or "general" tools. For the sake of consistency, we renamed these tools "all conditions" or "all other conditions" as appropriate.

* NBRS means the National Booking and Reporting System.

** A number will be allocated when the tool is live in NBRS.

5.33
When a tool does not produce a specific score, specialists can allocate standard scores to patients in each category. (Less sophisticated tools can also produce only a few standard scores.) For example, a specialty in a DHB organises the referrals that it accepts into urgent, semi-urgent A, and semi-urgent B groups. (It was not accepting routine referrals at the time of our audit fieldwork.) The booking clerks were instructed to allocate scores to patients in each group as follows:

• Urgent patients were given a score of 90 points.
• Semi-urgent A patients were given a score of 80 points.
• Semi-urgent B patients were given a score of 70 points.

5.34
We understand that this practice is commonplace. However, allocating scores in this way defeats the purpose of prioritisation because it means that specialists allocate scores for largely administrative, instead of clinical, purposes. They allocate a score simply because a score must be entered for each patient in the National Booking and Reporting System. We acknowledge that specialists may have taken pragmatic steps to compensate for deficits in the tools that some of their colleagues were involved in producing. However, when scores are allocated according to categories, it is not possible to:

• further prioritise patients within these categories;
• compare the scores with those of other DHBs (because the other DHBs might assign different values for each category); or
• use the scores to consider whether to alter the threshold for treatment.

5.35
Prioritisation tools that are not fit for the purpose they are used for are not likely to produce a good result. This means that:

• patients with a greater health need and ability to benefit from treatment could be given too low a score relative to other patients and potentially be denied access to scheduled services that they should have access to; and
• patients with a lesser health need and ability to benefit from treatment could be given too high a score relative to other patients and access scheduled services sooner than they should.

Are patients consistently prioritised for treatment?

5.36
Under the Strategy, DHBs are meant to use national prioritisation tools to assess each patient's level of health need and ability to benefit from treatment consistently throughout the country and to ensure that patients who most need services receive them soonest. Figure 10 lists the 30 currently available national prioritisation tools. The tools listed in Figure 10 do not cover all patients who are prioritised for scheduled services – for example, general surgery has a tool only for varicose veins¹⁶ – and some of the tools are not of good quality. The DHBs' local tools do not necessarily produce priority scores that reflect a similar level of need and ability to benefit from treatment for similar patients. This means that access cannot be fairly compared between DHBs. Therefore, it is not possible to determine whether patients are consistently prioritised for treatment.

Are patients treated in priority order?

5.37
Under the Strategy, patients needing scheduled services are meant to be treated in priority order. This is reinforced in the Ministry's contractual requirements with DHBs. In July 2010, the Ministry wrote to DHBs and emphasised that patients must be assigned an appropriate priority score and treated in priority order.

5.38
We asked the Ministry for information about selected procedures to check whether DHBs are treating patients in priority order. We chose to report on waiting times by priority score for adult cardiac surgery,¹⁷ general gynaecology surgery,¹⁸ and hip and knee replacement surgery.¹⁹ Because DHBs prioritise patients differently and have different treatment thresholds, we decided to use data from only one DHB, which we have not named. We chose to use one DHB on the assumption that practices are consistent within each specialty, which may not be the case. We have also included only those patients who were scored using a national prioritisation tool.²⁰

5.39
Figures 11-16 are scatter graphs that compare changes in performance from 2005/06 to 2009/10. If more than one patient has the same score and waited the same length of time to get treatment, it will show up as one diamond, triangle, or square on the graphs. Also, the thresholds for treatment sometimes changed during both years. For these reasons, the scatter graphs cannot stand alone. We report some statistical information about the data used to produce these graphs after each pair of graphs. We discuss the implications of the findings from all of the scatter graphs in paragraphs 5.46-5.51.

Figure 11
Days waited for treatment by patient priority score (adult cardiac surgery), at one DHB during 2005/06

Data was extracted from the National Booking and Reporting System on 14 October 2010 and reports each patient's latest priority score. Patients were scored using the national tools for cardiac surgery, which were CABG (9071), Aortic Stenosis (9072), Mitral Stenosis (9073), Aortic Valve (9074), Mitral Valve (9075), and the National Tool Cardiac Surgery – CABG – Cardiac Society of Australia and New Zealand Clinical Priority Score 2005 (Revised July 2005) (9076).

Figure 12
Days waited for treatment by patient priority score (adult cardiac surgery), at one DHB during 2009/10

Data was extracted from the National Booking and Reporting System on 14 October 2010 and reports each patient's latest priority score. Patients were scored using the national tools for cardiac surgery, which were Aortic Stenosis (9072), Mitral Stenosis (9073), Aortic Valve (9074), Mitral Valve (9075), and the National Tool Cardiac Surgery – CABG – Cardiac Society of Australia and New Zealand Clinical Priority Score 2005 (Revised July 2005) (9076).

5.40
In Figure 12, Patient A had a score of 95 and had their surgery in 60 days. Patient B had a score of 85 and had their surgery after waiting 229 days (about seven and a half months). The data for cardiac surgery also shows that in 2009/10:

• seven out of 18 patients (39%) with scores of 80 or higher were treated in 30 days or less, compared with one out of eight patients (13%) in 2005/06;
• 14 out of the 18 patients (78%) with scores of 80 or higher were treated in 90 days or less, compared with three out of eight patients (38%) in 2005/06; and
• 30 out of the 35 patients (86%) who exited the booking system without treatment exited within 180 days (about six months), compared with seven out of 20 (35%) in 2005/06.

5.41
The problems in ensuring that patients waiting for coronary artery bypass graft are treated in priority order are longstanding. In 2006, an article in The New Zealand Medical Journal reported on patients waiting for coronary artery bypass grafts who were assigned a priority score from June 2002 to September 2004. It stated:

Some patients with low [priority] scores were assigned an emergency category, while others with much higher [priority] scores were sent home to wait. These findings are similar to previous studies in this area.²¹

5.42
In 2008, the Cardiac Surgery Service Development Working Group reported that:

… there is little correlation between assigned priority and time to treatment. … Provision of surgery in accordance with assigned priority will require resolution of a number of factors which currently make this difficult.²²

5.43
We expect the updated "cardiac surgery urgency score by urgency time frame" shown in Figure 9 to rapidly result in observable improvements for patients. Given the longstanding nature of this problem, we plan to report on it when we monitor the Ministry's and DHBs' progress in responding to our recommendation.

Figure 13
Days waited for treatment by patient priority score (general gynaecology surgery), at one DHB during 2005/06

Data was extracted from the National Booking and Reporting System on 14 October 2010 and reports each patient's latest priority score. The data excludes patients treated for infertility or who were sterilised. Patients were scored using the National Tool Gynaecology – General (9065).

Figure 14
Days waited for treatment by patient priority score (general gynaecology surgery), at one DHB during 2009/10

Data was extracted from the National Booking and Reporting System on 14 October 2010 and reports each patient's latest priority score. The data excludes patients treated for infertility or who were sterilised. Patients were scored using the National Tool Gynaecology – General (9065).

5.44
In Figure 14, Patient A had a score of 100 and had their surgery in 11 days. Patient B had a score of 86 and had their surgery after waiting 327 days (about 11 months). The data for general gynaecology surgery also shows that in 2009/10:

• 322 out 434 patients (74%) with scores of 80 or higher were treated in 30 days or less, compared with 319 out of 484 patients (66%) in 2005/06;
• 418 out of 434 patients (96%) with scores of 80 or higher were treated in 90 days or less, compared with 454 out of 484 patients (94%) in 2005/06; and
• 216 out of 258 patients (84%) who exited the booking system without treatment exited within 180 days (about six months), compared with 181 patients out of 271 (67%) in 2005/06.

Figure 15
Days waited for treatment by patient priority score (hip and knee replacement surgery), at one DHB during 2005/06

Data was extracted from the National Booking and Reporting System on 14 October 2010 and reports each patient's latest priority score. Patients were scored using the National Tool Orthopaedics (9010).

Figure 16
Days waited for treatment by patient priority score (hip and knee replacement surgery), at one DHB during 2009/10

Data was extracted from the National Booking and Reporting System on 14 October 2010 and reports each patient's latest priority score. All except five patients were scored using the New Zealand Orthopaedics Association Hip/Knee Replacement/Revision Clinical Priority System 2006 (9011). The other five patients were scored using the earlier tool – National Tool Orthopaedics (9010).

5.45
In Figure 16, Patient A had a score of 100 and had their hip replacement surgery in 16 days. Patient B had a score of 100 and had their hip replacement surgery after waiting 240 days (eight months). The data for hip and knee replacement surgery also shows that in 2009/10:

• 29 out 99 patients (29%) with scores of 80 or higher were treated in 30 days or less, compared with 33 out of 272 patients (12%) in 2005/06;
• 80 out of 99 patients (81%) with scores of 80 or higher were treated in 90 days or less, compared with 134 out of 272 patients (49%) in 2005/06; and
• 58 out of 75 patients (77%) who exited the booking system without treatment exited within 180 days (about six months), compared with 51 patients out of 76 (53%) in 2005/06.

Some overall conclusions about these graphs

5.46
It is clear from the scatter graphs that more patients were treated overall and that more of them were treated within six months in 2009/10 than in 2005/06. These graphs support the information that we report in Parts 4 and 6. In this Part, we focus on whether patients were treated in priority order.

5.47
We expected the scatter graphs to show the pattern of treatment set out in Figure 9 – that is, that the high priority patients cluster in the top left-hand side of the graph and the other patients spread out in "steps" that increase in width as the priority scores decrease. The scatter graph that comes closest to this is the one for general gynaecology surgery in 2009/10 (Figure 14), which has not been the focus of a national initiative.

5.48
We have taken a fairly blunt approach to assessing whether patients have been treated in priority order. We chose to analyse how many patients with scores of 80 or more were treated within 90 days and 180 days.

5.49
We consider it reasonable to expect that all patients with a score of 80 or more will receive treatment within three months and that none of these patients will wait longer than six months. (The new cardiac surgery tool requires all patients with a score of 50 or more to be treated within 48 hours, not 90 days.) The example DHB treated most patients within these periods, but not all patients were treated in priority order.

5.50
We have no reason to believe that these results cannot be generalised to other specialties in our example DHB or to all specialties in all the DHBs.

5.51
In our view, the scatter graphs provide rich information about whether DHBs are treating patients in priority order. We suggest that DHBs publish this information for selected specialties or procedures to tell their communities about any improvements that they make. We consider that the Ministry should regularly report on this aspect of DHBs' performance, which might involve introducing an ESPI.

5.52
The Ministry periodically provides each DHB with scatter graphs about their untreated patients who have been given a commitment to treatment within six months. The DHBs are meant to use these graphs to treat patients in priority order.

Auditing selection decisions and priority scores

5.53
The Ministry expects DHBs to evaluate the effectiveness of their prioritisation processes and systems, and to improve their quality where required. This includes audits. We were told that audits of selection decisions and priority scores for treatment are not part of the routine management of scheduled services within a DHB or between DHBs.

5.54
DHBs occasionally complete one-off audits. This is usually in response to a complaint or because a new manager or clinical director wants to examine whether changes to a service are needed. A DHB's management of a waiting list can also be externally audited, which results in a published report, such as Report on Patients Awaiting Cardiac Surgery: Capital and Coast District Health Board (1 October 2008). The report is available from www.moh.govt.nz.

5.55
In our view, DHBs should have continuous quality assurance and improvement programmes for their scheduled services, and those programmes should use a range of methods. National prioritisation tools should be audited to validate the prioritisation of access to treatment, because such audits are fundamental to good clinical practice. We also consider that the Ministry and DHBs need to share the information produced and update the national prioritisation tools as needed.

DHBs' and specialists' commitment to, and management of, prioritisation tools to achieve equity

5.56
Many DHBs told us that improving equity within their DHB (by this, they mean that the resources made available for each specialty closely matches the needs of patients in their district) and throughout the country is important to them. They also want to be able to measure and acknowledge the level of unmet need in the population. Effective prioritisation can help with this. DHBs consider that national prioritisation tools enable decisions to accept or decline patients for scheduled services to be transparent and make it possible to compare the scheduled services offered by each DHB.

5.57
Half of the DHBs told us that progress towards achieving national equity had been slow and needed to improve. One DHB said that achieving equity of access objectives remains less than ideal, largely because of the need to further mature the prioritisation tools and their acceptance. Another DHB considered that there is a clear difference in views between nonspecialists and specialists about using nationally consistent prioritisation tools. This DHB considered that the nonspecialists' view is that it is unclear whether there is a commitment to using the tools nationally because they are not being used by all services. The specialists' view is that the usefulness of the tools is limited – the tools become either too complex to be embraced by specialists or too simple to provide enough stratification of priority.

5.58
We are not surprised that there are a wide range of views, given that there is not a full suite of national prioritisation tools (or processes if tools are not practical) that properly meet all of the Strategy's objectives. We understand that none of the national prioritisation tools have been audited to verify that they produce the results that were intended. Until this is done, the information that the tools produce cannot be used to assess whether the thresholds for access to scheduled treatment are appropriate. The survey responses we received from DHBs are consistent with our conclusion.

5.59
We also consider that no-one is clearly responsible for controlling the tools and their use. In our view, this has weakened the incentives to prepare and introduce national tools and use them effectively.

5.60
Prioritisation is always a specialist medical judgement. The resulting priority score determines whether the patient meets the DHB's threshold for access to scheduled services within six months. The specialist is responsible for overriding a patient's score, as needed, to respond to a patient's circumstances. (Audits would soon discover if this responsibility had been abused or whether the tool has a "blind spot" that disadvantages certain types of patients.) We expect specialists and DHBs to decide about access together.

5.61
The New Zealand Medical Association's Code of Ethics (2008) sets out doctors' wider role in prioritising care. One of the four clauses on this topic (clause 65) states that:

Patients must be able to trust their doctor to deal with their needs fairly and honestly. Doctors should, within reason, provide adequate information to their patients about their assessment and available treatments, including those not readily available.

5.62
We understand that, in earlier years, the professional groups did not formally lead the preparation of the tools, but individual members were involved. Professional groups are increasingly taking on a formal leadership role, which is appropriate. We consider that these groups should be responsible for the "life cycle" of each national prioritisation tool. They could:

• lead the preparation and testing of each tool;
• design and implement a change management process to effectively communicate a new or updated tool to the relevant health professionals and managers at each DHB, so that they are correctly used;
• apply to the Ministry recommending that a tool be used to comply with ESPI 8;
• review the tools regularly;
• consider how best to collect patients' views about the benefits of the treatment compared to their pre-treatment expectations; and
• audit the use of the tools to check whether patients were dealt with consistently.

5.63
In our view, the DHB's role is to provide the administrative and operational systems and resources to provide scheduled services within the time limits determined by the patient's priority score and the Government's maximum time limits.

5.64
The Ministry's role may be more flexible. At a minimum, it should set the operational policy parameters for the DHBs and professional groups to work within, to ensure that the tools created meet the Strategy's objectives and to share lessons learned by the professional groups. The Ministry could be responsible for the processes used to approve the use of the tools to comply with ESPI 8. Other than that, we consider that its role in preparing and introducing tools may be to provide support – such as funding, co-ordination, and expertise – to the professional groups.

5.65
We consider that a full suite of tools needs to be prepared, introduced, and used to improve the national consistency of clinical assessment and prioritisation for treatment. Once they are introduced, we expect the Ministry to withdraw its agreement for specialists and DHBs to use local tools to comply with ESPI 8. We consider that there should be as few national prioritisation tools as practicable within a specialty, so that:

• a DHB can ensure equitable access to treatment within a specialty;
• access to treatment can be more easily compared throughout the country;
• the effect of proposed increases or decreases in funding allocated to a specialty or procedure can be assessed more easily;
• the tools are used with enough patients to collect enough data to improve the tool and/or services; and
• the cost of preparing and updating the tools is minimised.

Setting minimum service levels for access to scheduled services

5.66
The Strategy anticipated that methods for funding scheduled services could shift from payments made for activities (such as an FSA or treatment) to funding DHBs based on their meeting an agreed service level for the population, measured in clinical and human terms. The Strategy gave two examples:

• Ophthalmology – all patients clinically assessed as requiring a cataract operation to keep their driver licence will be provided with cataract surgery within six months of assessment (about X operations per 1000 population).
• Orthopaedics – all patients clinically assessed as requiring a hip replacement to comfortably walk a flight of stairs will be provided with hip replacement surgery within six months of assessment (about X operations per 1000 population).

5.67
We understand that no moves have been made to introduce such methods. Instead, the minimum service levels have been set or managed in two main ways. The first is that DHBs take more patients into the system than they have the resources to deliver scheduled services to within six months, as long as they comply with the ESPIs. Secondly, the Ministry has recently introduced target intervention rates for each DHB's population and for certain operations.²³ The intervention rates are one way to substitute for consistent national data about priority scores. Because the rates are standardised, they also address changes in the size of the population and changes in demographic profile. The rates also ensure that DHBs increasingly make a proportional contribution to meeting the Health Target (see paragraph 6.10).

5.68
Managers and funders of scheduled services need to be mindful that minimum service levels for scheduled services or for a specific operation should not be set in isolation from other services that DHBs deliver or provide access to. One DHB told us:

Horizontal equity/prioritisation issues are not well addressed – that is, as a nation or region how do we decide whether greater resources should be targeted into a surgical specialty at the expense of another [hospital or service] because [Quality Adjusted Life Years], or some other outcome measure, would be greater?

---

11: Some DHBs operate clinics where GPs can send a patient to the hospital for a same-day assessment without sending the patient through the emergency department. These patients may be returned to the GP's care or admitted to hospital.

12: A GP liaison is a GP who works part-time for a DHB to improve the way that DHBs and GPs work together, and part-time as a GP. Their roles vary considerably; some are mainly to provide a channel for the DHB to distribute information to GPs, and others are more deeply involved in changing the way that scheduled services are delivered.

13: As at 12 April 2011, www.electiveservices.govt.nz.

14: These are called Clinical Priority Assessment Criteria (CPAC) or Clinical Prioritisation System (CPS) tools. The terms are interchangeable.

15: More information is available from the Health Services Research Centre, Victoria University of Wellington, at www.victoria.ac.nz/hsrc.

16: General surgery deals with conditions such as common gallbladder conditions and hernia repairs. We expect that access to surgery for these and other conditions would be assessed using one or more national prioritisation tools. General surgeons increasingly focus in sub-specialty areas, such as upper intestinal surgery, lower intestinal surgery, breast surgery, and vascular surgery. It may be practical to have a national prioritisation tool for each of these sub-specialties.

17: Access to cardiac surgery (heart valve procedures and coronary artery bypass grafting) is an area of concern to the public, and delays in receiving treatment can be life-threatening. Five of the 20 DHBs provide this surgery. The Clinical Cardiac Network was established to help improve services to cardiac patients.

18: General gynaecology surgery was not the focus of a national initiative to increase the number of operations performed. All DHBs provide this surgery.

19: Hip and knee replacement surgery was the focus of a special initiative to increase the number of operations performed. As part of the initiative, DHBs had to agree a plan with time frames with the Ministry, such as complying with ESPI 7 and ESPI 8, before each DHB could access any additional funding. All DHBs provide this surgery.

20: DHBs may find this technical information useful. Data is for "normal" patients only, which means that patients identified as staged, planned, and surveillance were excluded. Because the exit categories changed substantially between 2005/06 and 2009/10, the Ministry sorted the data as shown in our graphs. For 2005/06 data, patients coded as "treated other hospital" are included in "Treated as a scheduled patient", and patients coded as "treated privately" are included in "Exited from NBRS untreated".

21: Seddon, M et al (March 2006), "Coronary artery bypass graft surgery in New Zealand's Auckland region: a comparison between the clinical priority assessment criteria score and the actual clinical priority assigned", The New Zealand Medical Journal, www.nzma.org.nz/journal/119-1230/1881/.

22: Cardiac Surgery Service Development Working Group (September 2008), Cardiac Surgery Services in New Zealand, www.moh.govt.nz.

23: In 2010/11, the target rate for each DHB is to treat 292 patients for every 10,000 population. From 2009/10, target rates have been set for cataract, hip and knee replacement, and cardiac surgery. DHBs need to explain to the Ministry if they cannot meet the targets.

page top