Pharmacoeconomic Evaluation in Ireland

Pharmacoeconomic Evaluation in Ireland:

A review of the process

Lesley Tilson, Aisling O’Leary, Cara Usher, Michael Barry

National Centre for Pharmacoeconomics, Dublin 8, Ireland

Correspondence to:

Dr. Lesley Tilson,

National Centre for Pharmacoeconomics,

St James’s Hospital,

James’s St,

Dublin 8,

Ireland.

Tel: + 353 1 4103427

Fax: + 353 1 4730596

Email:

Abstract

Aim: The aim of this review is to describe the pharmacoeconomic assessment process in Ireland and to provide examples of recent appraisals and the subsequent impact on pricing and reimbursement decisions.

Method: The pharmacoeconomic appraisals conducted by the National Centre for Pharmacoeconomics (NCPE) between September 2006 and February 2009 were reviewed. The NCPE recommendations and subsequent reimbursement decisions by the Health Service Executive (HSE) were recorded. Recommendations made by the NCPE were compared with those of UK agencies. The duration of the NCPE pharmacoeconomic process and the time from marketing authorisation to reimbursement was estimated. The budget impact assessments from the pharmaceutical companies were reviewed and compared for consistency.

Results: The NCPE conducted twelve single technology appraisals during the study period. Eight of the medicines assessed were either recommended as a cost-effective use of resources or recommended with certain restrictions, and were funded by the HSE. Of the four medicines that were not considered cost-effective, two were reimbursed after a price reduction was negotiated and the remaining two were not reimbursed. The NCPE recommendations concurred with those of the UK agencies for the majority of appraisals, with the exception of sunitinib and lapatinib. The average duration of the NCPE process was 2.7 months. The average time from marketing authorisation to reimbursement was 7 months. The review of budget impact assessments highlighted a high degree of variability between submissions.

Conclusions: The findings of this review demonstrate the efficiency of the pharmacoeconomic process and the acceptance of the NCPE recommendations by the HSE for pricing and reimbursement decisions. NCPE recommendations broadly concurred with those of UK agencies for the majority of appraisals.

1.Background

The population of Ireland increased from 3.92 million to 4.24 million persons between 2002 and 2006 respectively, representing an 8.1% increase over a 4 year period (Central Statistics Office 2006). From a 10-year perspective, Ireland’s population increased at an annual average rate of 1.6% between 1996 and 2006. This is the largest population growth in the EU. Although the population is aging, in comparative terms it remains relatively young with 11.1% of the population over 65 years of age.

Healthcare policy and expenditure is governed by the Department of Health and Children (DoHC) and administered through the Health Service Executive (HSE). Healthcare funding is mainly derived through taxation (75%) with private funding via insurance agents accounting for 11% and patient co-payments the remainder. Total healthcare expenditure has increased considerably from €3.5 billion in 1997 to an estimated €15 billion in 2007. In 2007 expenditure on medicines under the Community Drugs Schemes (approximately 85% of total pharmaceutical expenditure) was €1.74 billion, a greater than 5 fold increase over the decade 1997 to 2007.

The National Centre for Pharmacoeconomics (NCPE) was established in Ireland in 1998 with funding from the DoHC. The aims of the centre are to promote expertise in Ireland for the advancement of the discipline of pharmacoeconomics through practice, research and education. The NCPE appraises the cost-effectiveness of technologies (medicines, diagnostics and devices)which may entail a high budget impact to the health system, in response to requests from the HSE or the DoHC. The NCPE also conducts pharmacoeconomic evaluations to inform public health policy (e.g. universal infant pneumococcal vaccination) and prescribing in primary care (e.g. statins for primary and secondary prevention of coronary heart disease)1-3.

Ireland is a small country with a different process for handling pharmacoeconomic evaluations. Similar to the UK, we are often one of the countries of first launch for new medicines in the EU and therefore the reimbursement decision could be of interest to other EU countries who may evaluate these medicines at a later stage. In this review we describe the pharmacoeconomic evaluation process in Ireland and provide examples of recent appraisals and the subsequent reimbursement decisions.

1.1 Reimbursement of medicines in Ireland

There are a number of different reimbursement schemes in Ireland [e.g General Medical Services (GMS), Drugs Payment (DP), Long term Illness (LTI) and High Tech Drugs Scheme (HTDS)] which are collectively referred to as the Community Drug Schemes. These four schemes account for 98% of prescriptions and 99% of expenditure in the community setting. Prior to reimbursement under the Community Drugs Schemes, a medicine must be included on a positive list.The GMS scheme is a means tested scheme and eligibility is dependent upon factors such as age, marital status, living alone or with family and allowances. Individuals covered by the GMS scheme are entitled to full reimbursement of prescription medications that are included on the GMS list of reimbursable items. Approximately one-third of the population are eligible for the GMS scheme and it accounts for about two-thirds of pharmaceutical expenditure in primary care. Pharmacists are paid a flat fee per item for dispensing medicines on the GMS scheme. The DP scheme applies to Irish residents who are not eligible for the GMS scheme. Individuals/ families are required to pay a deductible of up to €100 in any calendar month for approved prescribed medicines and the HSE pays for the remaining cost of medicines in any particular month. Individuals who suffer from fifteen chronic illnesses (e.g. diabetes mellitus and epilepsy) are also entitled to full reimbursement of drugs specific to that condition under the LTI scheme. Pharmacists are paid a percentage mark-up (50%) on the ingredient cost as well as a dispensing fee per item on the DP and LTI schemes. The HTDS was introduced in 1996 to facilitate the supply by community pharmacists of certain high cost drugs e.g. oral chemotherapy, beta-interferon and growth hormones. The price of these high cost drugs is paid directly to the wholesaler by the HSE and pharmacists are paid a standard monthly patient care fee. Therefore the reimbursement price of medicines in Ireland depend on patients’ eligibility for reimbursement of prescription drugs. Indeed, the difference in drug costs between the GMS and DP/LTI schemes can result in a medicine being cost-effective on one scheme, but not on the other. The Community Drug Schemes in Ireland are explained in greater detail in previous publications4, 5.

1.2 Pharmacoeconomic evaluation in Ireland

The requirement for pharmacoeconomic evaluation for reimbursement of technologies in Ireland is outlined in the agreement between the Irish Pharmaceutical Healthcare Association (IPHA) and the HSE (IPHA/HSE Agreement)6. The requirement for pharmacoeconomic data is by agreement, rather than by statutory regulation. The Agreement came into effect on the 1st September 2006 and replaced the previous Agreement which commenced in 1993, was renewed in 1997 and extended in 2001. It outlines the supply terms, conditions and pricing of medicines supplied to the health service in Ireland. The Agreement covers all medicines reimbursed under the Community Drug Schemes and all medicines supplied to the HSE, State funded hospitals and State agencies whose functions normally include the supply of medicines.

Under the terms of the 2006 IPHA/HSE Agreement, the HSE reserves the right to assess new and existing technologies (medicines, diagnostics and devices) that may incur a high cost or have a significant budget impact. What constitutes “high cost” or “high budget impact” has not been explicitly stated or defined. In the case of new medicines, assessments may be conducted prior to the reimbursement application but must be completed within 90 days. Evaluations are conducted in accordance with the existing agreed Irish Healthcare Technology Assessment Guidelines (2000)7. Prior to the current Agreement, pharmacoeconomic data was used on an informal basis in Ireland.

The pharmacoeconomic evaluation process includes a number of key steps (Figure 1):

Products are formally referred to the NCPE by the Products Committee of the Corporate Pharmaceutical Unit of the HSE, whose responsibility it is to approve medicines for reimbursement under the Community Drug Schemes. The Committee meet on a monthly basis.

Once the NCPE receives a request to conduct a pharmacoeconomic assessment, representatives of the relevant pharmaceutical company are invited to attend a meeting at which the details of the process, as well as the scope and requirements for the evaluations are outlined by the NCPE.

The pharmaceutical company then prepares a pharmacoeconomic submission which includes evidence of the cost-effectiveness and an estimate of the potential budget impact of the product in the Irish setting. The budget impact estimate is not part of the decision making process on cost-effectiveness but it is intended to give the HSE an indication of the financial implications should the product be reimbursed.

The NCPE Review Group critically appraise the company submission. The Review Group is a multidisciplinary team, which comprises a clinician, pharmacists, research scientist, health economist and statistician.This process includes an assessment of the modelling and data inputs provided by the manufacturer (as opposed to the NCPE developing a model and preparing an assessment report themselves). Local expert clinical opinion is sought by the Review Group regarding the place in therapy of the new medicine.

The NCPE prepares a critical appraisal of the company submission with the systematic application of a ten-point checklist. The checklist was devised by Drummond et al.8 and it is used to ensure a consistent systematic approach is adopted to critically appraise each submission. The NCPE invite comments from the pharmaceutical company prior to completion of the final appraisal.

The final appraisal, which includes a recommendation of whether the drug may be considered cost-effective from the HSE perspective, is completed by the NCPE and submitted to the Products Committee (HSE).

A summary of the final appraisal document is posted on the NCPE website. This includes a recommendation as to whether the technology would be considered a cost-effective use of resources.

Figure 1. Summary of the Pharmacoeconomic Process in Ireland (see end of paper for Figure 1).

Products subject to an assessment are reimbursed within 40 days of a positive reimbursement decision. Should reimbursement be refused, an appeal may be made to a designated expert committee whose final decision may be made within a further 90 days and is accepted as binding6.

1.3 Pharmacoeconomic evaluation in other jurisdictions

The processes of selecting and appraising new drugs varies between jurisdictions both within the EU and other international settings. In this paper, we compare our recommendations to those of the three UK bodies which evaluate cost-effectiveness of drugs and medical interventions: the National Institute for Health and Clinical Excellence (NICE), the Scottish Medicines Consortium (SMC) and the All Wales Medicine Strategy Group (AWMSG). NICE, the independent agency that provides guidance to the UK NHS on the use of technology, has two approaches to completing technology appraisals: a single technology (STA) and multiple technology (MTA) appraisal process. NICE MTAs take an average of 14 months and STAs are expected to take between 6 and 8 months9. One of the main differences between the STA and MTA processes is that under the STA process the cost-effectiveness model provided by the manufacturer is reviewed, whereas under the MTA process an independent model is developed9, 10. NICE assesses both drugs and medical interventions, but it does not consider all drugs, only those that are referred to it by the Secretary of State for Health.

In contrast, the SMC assesses and produces advice on all new drugs but does not assess other healthcare interventions. The AWMSG appraises new high cost medicines for which no NICE guidance is expected for at least 12 months. NICE guidance takes precedence over AWMSG guidance if the two differ.

The SMC and AWMSG processes put the burden of proof on the manufacturers and provide more rapid recommendations based on a review of the manufacturers’ own models compared to the NICE MTA process. The NICE STA process is broadly comparable, in terms of its timelines, with the Scottish and Welsh processes. SMC aim to issue advice on all newly licenced medicines within 12 weeks of products being made available. The SMC process does not include the time for companies to prepare their submission, whereas the NICE STA appraisal includes this period9. The AWMSG process takes approximately six months and can be timed so that the medicine can be endorsed for NHS Wales as expediently as possible. This results in their use and funding prior to NICE guidance or in its absence.

The objectives of this paper are to assess the recommendations made by the NCPE and subsequent reimbursement decisions by the HSE. Furthermore, in cases where the NCPE have evaluated the same drugs as NICE, SMC and AWMSG, the recommendations are compared and reasons for any disagreements discussed. Additional elements of this paper include a review of the time from marketing authorisation to reimbursement and the duration of the pharmacoeconomic appraisal process itself. Common issues raised by the Review Group in the critical appraisal of company submissions are highlighted. Furthermore, ancillary analysis of the estimated budget impact analyses submitted by the pharmaceutical companies was completed.

2. Methods

The pharmacoeconomic appraisals conducted by the NCPE between September 2006 and February 2009 were reviewed. Details of the therapeutic indication of the medicine at the time of the evaluation, the chosen comparator for the cost-effectiveness evaluation and the setting/drug reimbursement scheme were collated for each appraisal. All medicines appraised by the NCPE during the study period were categorised according to the NCPE recommendation, whether cost-effectiveness was demonstrated and whether they were subject to an appeal. Subsequent reimbursement decisions by the HSE were recorded.

The evaluations of the drugs which were also appraised by NICE, SMC and AWMSG were reviewed and compared with the NCPE recommendations. The date of publication of all appraisals was recorded.

The time from first Marketing Authorisation (MA) to reimbursement was estimated. The list of licensed human medicines on the Irish Medicines Board website ( ) provides access to the date of first MA in Ireland and is a publicly available source of the data.The date of inclusion of the medicine on the list of reimbursable items was obtained from the product files of the Primary Care Reimbursement Service (PCRS). This list is updated on a monthly basis.

The duration of the pharmacoeconomic process was estimated for each appraisal.The “start date” is regarded as the date the NCPE receives the submission from the pharmaceutical company. The “finish date” is regarded as the date that the NCPE submit their final appraisal of the company submission to the Product Committee of the HSE. According to the IPHA-HSE Agreement the process should take no longer than 90 days. The NCPE Review Group may request the company to undertake certain changes to the pharmacoeconomic submission. For example, more extensive sensitivity analysis or alterations to key data inputs may be required. The time taken by the company to prepare and make amendments to the submission is not included in the 90 day time limit (i.e. there is a “stop-clock”). The duration of the pharmacoeconomic process recorded in this review included the time for the company to make amendments to their submission.

Common issues raised by the Review Group in the critical appraisal of company submissions were highlighted and the budget impact analyses (BIAs) from the pharmaceutical companies were reviewed and compared for consistency. The following information was extracted from the six BIAs that were reviewed:

Estimated budget impact: The submissions were categorised according to whether by year five the intervention would be cost saving, low budget impact (< €0.5 million), medium budget impact (€0.5-2 million), high budget impact (>€2 million) or unclear. These ranges were adopted arbitrarily from a study conducted by the Scottish Medicines Consortium (SMC)11 solely for the purpose of this review and are not representative of any classification used by the HSE. The budget impact estimate was based on assumptions regarding the number of eligible persons that would be prescribed the new drug and the drug acquisition cost.
Method of calculating budget impact: The submissions were categorised according to whether they considered the impact on the drug budget only or the overall healthcare budget. If limited to a drug budget perspective, the BIA will not answer the broader health system questions concerning overall healthcare spending in the context of cost offsets. The drug budget perspective may include reduced utilisation of drugs due to replacement of current treatments, potential increase in certain drugs to manage side-effects and impact on hospital day cases (e.g. for intravenous infusions).
Treatment replacement assumptions – The inclusion of potential cost offsets due to products that will be replaced by the new medicine was considered in the review.
Time horizon – The time horizon for BIAs was recorded. A survey of drug plan managers in Canada indicated that 83% of respondents would find a time horizon of 3 to 5 years valuable12.Time horizons that go beyond a few years are subject to considerable assumptions.
Sensitivity analysis – There is considerable uncertainty in a BIA. A range of results should be presented based on realistic scenarios regarding inputs and assumptions.