Payment by Results for chemotherapy—challenges and solutions

Dear Colleagues

Establishing and effectively maintaining a tariff for chemotherapy, within our current financial restrictions, is going to be key in managing costs and ensuring prescription appropriateness in of one of the most costly areas of medicine.

In this month’s issue of Cancer Services Forum, Anne Hines modestly describes her tour through the highly complex and challenging process of defining this tariff as ‘merely a cursory look’. In essence, Anne provides an excellent overview of a long and drawn out project that many of us have only a partial understanding of. A process that has been beset with difficulties, including poor Trust data collection, inter and intra Trust prescribing variability and within a specialty that is constantly changing in response to research advances.

Also in this issue, Nicola Redfern provides a response to the discussion featured in Issue 13 on patient access schemes from an industry perspective. The challenges faced by the industry in being asked to offer these schemes appear to be every bit as challenging as those faced by the provider in administering them. Despite the ongoing work from the Patient Access Schemes Liaison Unit (PASLU), the burden of the schemes continues to increase, and the move towards encouraging less variability in the scheme models to reduce the administrative workload is not yet evident.

Libby Hardy
Lead Pharmacist, Peninsular Cancer Network

Introduction

Payment by Results (PbR) has been in use within the National Health Service (NHS) since 2003–04.
At first, its application was generally restricted to inpatient surgical services, but the Department of
Health (DH) is working to introduce PbR to other clinical areas—including chemotherapy.¹ This article considers the progress that has been made so far towards a chemotherapy tariff for PbR, the challenges still faced, and the plans in place to tackle them.

What is Payment by Results?

Unlike block contracts or cost-and-volume contracts, PbR means that payment is made (by NHS purchasers) for services actually performed (by NHS providers). It offers several advantages over traditional contracts, but is also associated with some risks (Panel 1).

INSERT PANEL 1

Coding, classification, and Health Resource Groups

The Office of Population Censuses and Surveys (OPCS) organises the coding and classification system that defines clinical activity undertaken by the NHS. Clinical coders record the OPCS code for every clinical activity, plus the international diagnostic code (ICD-10),² so that the diagnosis is linked to the procedure. The use of the two sets of codes is overseen by NHS Connecting for Health (NHS CfH).

Health Resource Groups (HRGs)³ are OPCS codes that are clustered together because they all relate to clinical activities that use similar amounts of NHS resources to perform a specified activity, and they are also similar from a clinical perspective. It is for each HRG that is recorded, that the tariff for a particular activity is paid. HRGs are overseen and maintained by the NHS Information Centre (NHS IC) Casemix team. Every year, the Casemix team releases a number of software tools known as HRG Groupers, which track the OPCS codes recorded by clinical coders and allocate them to an HRG.⁴

HRGs have been used by the NHS to cost their activity since 2006–07, and for funding services from April 2009. Currently, the NHS is operating under the latest version of HRGs—HRG4 (and the most up-to-date version of the OPCS, i.e. OPCS-4.5).³ Prior to HRG⁴, chemotherapy could be coded, but it was incorporated into other procedure-related HRGs. The HRG was applied at the beginning of an entire course of chemotherapy, and there was no distinction made for the cost or complexity of chemotherapy delivery.

Chemotherapy was, and still is, excluded from PbR. However HRG4 has ‘unbundled’ the chemotherapy component from the original all-encompassing HRG, and made it a stand-alone entity. One consequence of the unbundling of the chemotherapy HRGs is an increase in the level of detail that is defined and recorded for each activity (known as increased granularity). The number of OPCS and HRG codes has significantly increased, as has the complexity of coding. However, the increased granularity should facilitate the coding and recording of chemotherapy activity, and hence the introduction of a tariff.⁵

Table 1 shows the current chemotherapy HRGs. The 10 procurement HRGs6 relate to the purchase cost of chemotherapy regimens (banded 1–10 according to price). Procurement codes are applied to the first attendance in each chemotherapy cycle, and cover the costs of the entire cycle. Additionally, there are five delivery HRGs. These are also applied at the first attendance, and contain resources needed for the entire cycle; however, a code for subsequent attendances is applicable to some regimens. Inpatient chemotherapy does not attract a delivery code, but a procurement code should be applied. The procurement and delivery codes all map directly to corresponding OPCS codes. (Note that unlike some other clinical areas, where several procedures, and hence OPCS codes, are grouped into each HRG code, the OPCS and HRG codes for chemotherapy correspond one-to-one.)

Table 1: Chemotherapy HRGs⁶

To facilitate coding of chemotherapy regimens, the PbR team at the DH has commissioned the production of a national chemotherapy regimens list, which is currently published by NHS CfH and updated annually.⁷ As shown in Table 2, the list states the regimen name, the major components of the regimen, the usual cycle length and the procurement and delivery codes. It does not, however, state whether any subsequent attendances are associated with the regimen. For the regimen shown in Table 2, the procurement and delivery codes should be applied every 14 days. The subsequent attendance code should not be used.

Table 2: OPCS chemotherapy regimens coding list for ABVD (doxorubicin + vinblastine + bleomycin + dacarbazine)⁷

Formulating a tariff

A comprehensive description of how the DH calculates the tariff for PbR each year is beyond the scope of this article, and can be found in key documents on the department’s website.⁸ However, it is useful to consider some of the basic principles behind tariff calculation, especially those that will influence a future chemotherapy tariff.

Calculating a tariff involves much more than just adding up costs. The process (Figure 19) must start with accurate recording of patient-level data through reliable clinical coding. Costs are allocated, ideally at patient level, and the data are grouped together according to HRGs and casemix (i.e. all the chemotherapy regimens within the particular band). Finally, a tariff can be calculated.

Figure 1: Pyramid of data required for accurate tarriff calculation⁹-one wrong step and the whole pyramid collapses

Deficiencies in any of the steps can lead to an inaccurate tariff calculation, which may have catastrophic consequences for both providers and purchasers. For this reason, the tariff is always ‘road tested’ before being implemented. The importance of testing was one of the lessons learnt after the development of the final 2006–07 tariff, which was published just 1 week before the start of the financial year and led to considerable financial instability across the whole of the NHS.¹⁰ The ensuing chaos led to the Lawlor report,¹¹ which made several recommendations, including better pre-release testing of tariffs, publication by mid-December at the latest and the suggestion that newly unbundled clinical areas, such as chemotherapy and radiotherapy, continue to be excluded from the tariff until problems have been resolved.

Despite the changes instigated by Lawlor, changes to PbR are always controversial—and HRG4 is no exception.¹⁰ However, on a brighter note, a review published in 2009, found that implementation of PbR had brought many positive benefits, including a reduction in healthcare costs with no evidence of a corresponding reduction in quality.¹²

Reference costs

At its most fundamental level, the tariff is calculated from costs collected and submitted each year by provider trusts. These reference costs represent the entire costs attributed to delivering a particular HRG, including staffing and overheads. Each year, the DH PbR team provides guidance on how to collect reference costs and what should be included.¹³ Reference cost returns are published annually on the DH website.¹³

Each Trust that submits returns has a reference cost index (RCI) calculated and published. An index of 100 indicates that the Trust’s costs per HRG are average compared with all other Trusts. If the index is greater than 100, the Trust’s costs are higher than average, and if the index is less than 100 then its costs are lower than average. Publication of the RCI provides a great deal of incentive for Trusts to record and count their activity accurately. Over-costing, under-recording or both can lead to a high RCI.

Provided the range of reference cost returns for each specific HRG is not too great, the tariff is, in chemotherapy regimen details. This has now been superseded by a portal hosted by the PbR website.¹⁵

Collecting reference costs

Despite advice from the DH on how to collect reference costs,¹³ considerable confusion remains over how to assign costs to the chemotherapy HRGs. The guidance makes it clear that costs to be assigned to the procurement HRGs must include acquisition costs, all pharmacy costs involved in procurement and preparation and the costs of drugs used to treat side effects. In practice, regimen-specific information on chemotherapy is hard to collate because traditional pharmacy computer systems do not store information by regimen but by drug. Furthermore, there is a dearth of prescribing information in secondary care in general, and on chemotherapy in particular.¹⁶

If a Trust cannot distinguish between different indications for the same drug, it is very difficult to accurately assign the costs to the appropriate band. Figure 2 shows the extent of this problem for the 2007–08 reference costs returns for chemotherapy procurement HRGs.¹⁷ There is the expected bandby-band increase, except that the submitted costs for Band 3 and Band 4 regimens are skewed, as are those for Band 8 and Band 9. This anomaly is a reflection of providers’ varying ability to report accurate information and the accuracy of the allocation of the band in the OPCS listing. principle, an average of all the costs submitted. Cost weights—i.e. various other market forces factors, for example to account for higher staff and overhead costs in London—are included in the final tariff and RCI calculations.¹³

Figure 2: Reference cost returns for chemotherapy procurement HRGs for 2007-2008¹⁷
This graph shows an average of the reference costs for each delivery type across inpatients, outpatients, daycases and other categories

Note that there is always a 3-year lag between reference cost submission and tariff calculation, so the tariff for 2010–11 will be based on 2007–08 reference costs.¹⁴

Where is the tariff for chemotherapy?

As explained above, a viable tariff calculation relies on sound patient-level data captured through reliable clinical coding, casemix and accurate costing. Unfortunately, so far, every step of this pathway for chemotherapy has proved problematic.

Patient level data and clinical coding

With the advent of HRG4, more clinical areas now need to be coded; for example, the new areas of coding for chemotherapy include outpatient activity. Accurate coding of chemotherapy regimens requires the clinical coders to know a vast amount of specific clinical information about the patient and the chemotherapy regimen, specifically:
• Is the patient seen as an inpatient, an outpatient or a daycase?
• Is it the patient’s first or subsequent attendance for the chemotherapy cycle?
• What is the expected cycle length?
• Is the chemotherapy regimen on the CfH list?
• Is the trust chemotherapy regimen the same as that on the CfH list but just called by a different name?
• Are there any subsequent attendances attached to the regimen?

However, coders do not always have access to the best information sources, and are often not considered part of the clinical team. Also, the CfH chemotherapy regimens list has significant weaknesses.⁷ For example, lists of paediatric and clinical trial regimens are either out of date or non-existent. Formerly, the only mechanism for getting new regimens onto the list was via the CfH portal, which was not ideally suited for transmitting chemotherapy regimen details. This has now been superseded by a portal hosted by the PbR website.¹⁵

Collecting reference costs

Despite advice from the DH on how to collect reference costs,¹³ considerable confusion remains over how to assign costs to the chemotherapy HRGs. The guidance makes it clear that costs to be assigned to the procurement HRGs must include acquisition costs, all pharmacy costs involved in procurement and preparation and the costs of drugs used to treat side effects. In practice, regimenspecific information on chemotherapy is hard to collate because traditional pharmacy computer systems do not store information by regimen but by drug. Furthermore, there is a dearth of prescribing information in secondary care in general, and on chemotherapy in particular.¹⁶

If a Trust cannot distinguish between different indications for the same drug, it is very difficult to accurately assign the costs to the appropriate band. Figure 2 shows the extent of this problem for the 2007–08 reference costs returns for chemotherapy procurement HRGs.¹⁷ There is the expected bandby-band increase, except that the submitted costs for Band 3 and Band 4 regimens are skewed, as are those for Band 8 and Band 9. This anomaly is a reflection of providers’ varying ability to report accurate information and the accuracy of the allocation of the band in the OPCS listing.

The picture for the delivery HRGs in 2007–08 (Figure 3¹⁷) shows that the submitted costs for simple chemotherapy regimens were more than for more complex regimens. Again, the likely explanation is a lack of understanding of the codes at Trust level. Also, the OPCS delivery band for a regimen is allocated based on chair time, which, in some cases, may not accurately reflect the resources actually needed to deliver the treatment. Because all the costs for a regimen are supposed to be allocated to the first attendance, it would be reasonable to expect to see very few costs attached to the subsequent attendance element of the HRG. However, in 2007–08, the costs associated with subsequent elements of chemotherapy were roughly the same as those for first attendance for simple or more complex chemotherapy (Figure 3¹⁷).This finding probably reflects general confusion about what constitutes a subsequent element of a chemotherapy cycle, and what the purpose of the HRG is. A simple amendment to the current OPCS list structure, indicating whether or not subsequent attendances are associated with a specific regimen, would improve coding and thus, hopefully, cost allocation.

Figure 3: Reference cost returns for delivery HRGs in 2007-08, based on an average of the reference costs for inpatients, outpatients, daycases and other categories¹⁷

Previously, Trusts were allowed not to submit reference costs for the chemotherapy HRGs by listing them as a known area of difficulty. It remains to be seen whether the accuracy of returns will improve now submission of reference costs for chemotherapy is mandatory.

Band 10 regimens

The costs of chemotherapy regimens in Band 10 vary widely, and it is difficult to see how a single tariff will ever adequately recoup the spending on specific very high-cost treatments such ibritumomab tiuxetan (Zevalin®), at approximately £13,000 for the single cycle given.^18,19 Possible solutions are an increase in the number of bands (i.e. increased granularity) or, once we have a tariff for chemotherapy, continuing to exclude Band 10 regimens so that these are subject to local negotiation—or, more likely, both.

Overcoming the problems

The hurdles facing the chemotherapy tariff may suggest a picture of gloom and doom, but all is not lost. There are specific plans and programmes in place to address all of these issues. The Cancer Network Pharmacist Forum has long argued that the current mechanisms for updating and maintaining the chemotherapy regimens list are not robust. The NHS IC Casemix team is leading on specific work programmes, with the aim of producing a fully up-to-date regimen list including all paediatric and adult clinical trial regimens.

To this end, there are plans for a new mechanism for requesting inclusion of regimens on the list and a consistent method of allocating them to a procurement and delivery band. Furthermore, the new PbR-hosted portal is expected to improve the addition of new regimens to the list.¹⁵ The specific Band 10 issues are also being addressed. These important developments are being undertaken with the engagement of the DH, NHS CfH, the National Cancer Action Team (NCAT) and a number of NHS provider and commissioner organisations across England.

The continued development of electronic prescribing systems for chemotherapy will help to improve the recording of regimen-specific chemotherapy activity. However the benefits of e-prescribing will be seen only if systems are set up to record and then report the required information. A mandated chemotherapy dataset and specific guidance to Primary Care Trusts (PCTs) to insist that their acute trusts code chemotherapy in accordance with HRG4, will achieve greater accuracy in the recording of chemotherapy information.²⁰ It is also encouraging to see that clinical coders are increasingly regarded as part of the clinical chemotherapy service.

One important development is an NCAT project using the Capacity Planning Oncology Resource Tool (C-PORT)²¹ with the aim of describing how the costs of procuring and delivering chemotherapy may be allocated along the chemotherapy pathway. This initiative should eventually lead to much clearer guidance for trusts on the allocation of reference costs.

Furthermore, a simple guide to PbR for chemotherapy and radiotherapy, produced by the DH PbR team, the NHS IC Casemix team, NHS CFH and NCAT, is now published.²² It is expected to raise the profile and understanding of PbR for chemotherapy services among Trust, clinical and managerial staff alike.

Conclusion

The road to a chemotherapy tariff has been a challenging one, and we have still not reached our destination. Current Trust information systems are insufficient to accurately count, record and cost chemotherapy activity. Clinical coding of chemotherapy will not be complete until the chemotherapy regimens list is up to date and includes all the regimens that need to be coded. The current financial turmoil of the NHS is also adding to the uncertainty, leading some PCTs to reconsider the implications of funding services by PbR.

But there is light at the end of the tunnel. Work is underway to resolve the specific problems, and reference cost returns will become more reliable. However, with the 3-year delay between reference costs and tariff setting, it is unlikely that we will see a chemotherapy tariff before 2012.

Please note, this has, by necessity, been merely a cursory look at what is an extremely complex and technical subject. For more information, please refer to the following websites:

NHS IC

http://www.ic.nhs.uk/

NHS Connecting for Health

http://www.connectingforhealth.nhs.uk/

DH PbR website

http://www.dh.gov.uk/en/Managingyourorganisation/Financeandplanning/NHSFinancialReforms/index.htm

References

1. Hines A. The National Tariff: how chemotherapy may be funded. Cancer Services Forum 2007; Issue 2.
2. Department of Health. Clinical coding instruction manual – ICD-10 and OPCS-4. Available at: http://www.connectingforhealth.nhs.uk/systemsandservices/data/clinicalcoding/codingstandards/publications/ccim (accessed June 2010).
3. Department of Health. NHS. The Information Centre. What are Healthcare Resource Groups (HRGs)? Available at: http://www.ic.nhs.uk/services/the-casemix-service/newto-this-service/what-are-healthcare-resourcegroups-hrgs (accessed June 2010).
4. Department of Health. NHS. The Information Centre. Casemix Service HRG4. Report on the Casemix Service online survey for the HRG4 2007/08 reference cost grouper. London: DH, 2009.
5. Department of Health. NHS. The Information Centre. Healthcare Resource Groups 4 (HRG4). Available at: http://www.ic.nhs.uk/services/the-casemix-service/new-to-thisservice/healthcare-resource-groups-4-hrg4 (accessed June 2010).
6. Department of Health. NHS. The Information Centre. Casemix Service HRG4. Guide to unbundling. London: DH, 2007.
7. Department of Health. OPCS-4.5 chemotherapy regimens list and guidance for clinical coders version 1.0. London: DH, 2009.
8. Department of Health. What is payment by results? Available at: http://www.dh.gov.uk/en/Managingyourorganisation/Financeandplanning/NHSFinancialReforms/DH_4065236 (accessed June 2010).
9. Moss T, Gibbin S. National Cancer Action Team. Delivering payment by result – an overview. Available at: http://www.rcr.ac.uk/docs/hosns/ppt/SpecialistCommissioningPbRHRGs.ppt (accessed June 2010).
10. Gainsbury S. Trusts on edge as draft payment by results tariff runs into trouble. HSJ, 25 September, 2008. Available at: http://www.hsj.co.uk/trusts-on-edge-as-draft-payment-byresults-tariff-runs-into-trouble/1871928.article (accessed June 2010).
11. Department of Health. Report on the tariff setting process for 2006/07. London: DH, 2006.
12. Farrar S, Yi D, Sutton M, Chalkley M, Sussex J, Scott A. Has payment by results affected the way that English hospitals provide care? Difference-in-differences analysis. BMJ 2009; 339: b3047.
13. Department of Health. NHS reference costs 2008/09: collection guidance. London: DH, 2009.
14. Department of Health. Reference costs 2007-08 publication. London: DH, 2009.
15. Department of Health. Payment by results. New chemotherapy regimen area. Available at: http://www.dh.gov.uk/en/Managingyourorganisation/Financeandplanning/NHSFinancialReforms/DH_109534 (accessed June 2010).
16. Department of Health. Variations in usage of cancer drugs approved by NICE. Report of the Review undertaken by the National Cancer Director. London: DH, 2004.
17. Gibbin S, Levy D. National Cancer Action Team. Making ready for a national tariff in chemotherapy. Available at: http://www.northtrentcancernetwork.nhs.uk/media/14602/susan_gibbin__making_ready_for_a_national_tariff_in_chemotherapy.pdf (accessed June 2010).
18. Bayer Schering Pharma. January 2010 price list. Available at: http://www.bayer.co.uk/html/pdf/BSP_Price_List_Jan_2010.pdf (accessed June 2010).
19. Zevalin. Summary of Product Characteristics. Bayer Schering Pharma AG, 2009.
20. National Chemotherapy Advisory Group. Chemotherapy services in England: ensuring quality and safety. London: NCAG, 2009.
21. Chemotherapy Planning Oncology Resource Tool (C-PORT). Pharmaceutical Oncology Initiative Partnership (POIP). Version 3, March 2008.
22. Department of Health. Payment by result. Chemotherapy and radiotherapy. A simple guide. Available at: http://www.dh.gov.uk/prod_consum_dh/groups/dh_digitalassets/documents/digitalasset/dh_110175.pdf (accessed June 2010).