NICE DSU latest one page summary for TSD 20

 

The latest of the TSD one page summaries is now live on the NICE Decision Support Unit website and it relates to Technical Support Document 20:

"Multivariate meta-analysis of summary data for combining treatment effects on correlated outcomes and evaluating surrogate endpoints"

For further information about the Multivariate meta-analysis TSD, please click here. 

For all other TSD one page summaries, please click here.

The Decision Support Unit (DSU) is commissioned by the National Institute for Health and Care Excellence (NICE) to provide:

• technical support to evaluation programmes;
• advanced methodological development for all types of health technologies, including pharmaceuticals, medical and diagnostic technologies and related products;
• educational support;
• advanced methodological, analytical and other ad hoc support to NICE and its independent advisory bodies on the assessment of medicines and medical technologies;
• quality assurance of economic models.

To visit the DSU website please visit http://nicedsu.org.uk/.

New HEDS DSU Report - Comparing the EQ-5D-3L and 5L versions.

The Decision Support Unit at ScHARR have published a new report. 

Comparing the EQ-5D-3L and 5L versions. What are the implications for model-based cost effectiveness estimates? 

HEDS Authors:

Becky Pennington

Monica Hernandez-Alava

Stephen Pudney

Allan Wailoo

You can download the report here

Real World Data: guidance from the DSU

The NICE Decision Support Unit, based in HEDS, has published a report that provides guidance on the use of real world data (RWD) for the estimation of treatment effects in NICE decision making.

It builds on the NICE Decision Support Unit (DSU) Technical Support Document (TSD17) “The use of observational data to inform estimates of treatment effectiveness in technology appraisal: methods for comparative individual patient data” (Faria et al, 2015), which focused on methods commonly used to estimate treatment effects from non-randomised studies, where individual patient data (IPD) is available.

This report expands on the TSD by considering how RWD has been used to inform decision making in seven of NICE’s programmes, how it could have been used and the guidance that NICE currently provides to those responsible for submitting evidence, critiquing evidence and making decisions based on those assessments.

What’s bothering NICE?

If you want to know which methodological issues are bothering NICE about their Technology Appraisal Programme, then having a look at what the DSU is doing is a good start.  The DSU, based in HEDS is currently involved in work in the following areas:

• New Technical Support Documents (TSDs) in partitioned survival analysis and in the calibration of treatment effects
• Quality assurance of models which inform NICE Technology Appraisals

Further information is available here.

New NICE Technical Support Document (TSD)

Well sort of…..it’s an updated version of TSD 2, “A general linear modelling framework for pair-wise and network meta-analysis of randomised controlled trials”.  Specifically, the coding to Example 6 (Psoriasis, from page 83 of the TSD) has been revised.

All TSDs are available from the DSU website here.

What are TSDs?  The TSDs are commissioned by NICE with the aim of providing further information about how to implement the approaches described in the current NICE Guide to the Methods of Technology Appraisal. They provide a review of the current state of the art in each topic area, and make clear recommendations on the implementation of methods and reporting standards where it is appropriate to do so. They aim to provide assistance to all those involved in submitting or critiquing evidence as part of NICE Technology Appraisals, whether companies, assessment groups or any other stakeholder type.

Image: nice by wallsdontlie

Appraisal methods for regenerative medicines

NICE have undertaken an exercise to assess whether changes to its methods and processes will be needed for regenerative medicines. Part of this work was to undertake a mock technology appraisal of CAR T-cell therapy for treating acute lymphoblastic leukaemia. Professor Allan Wailoo from HEDS was a member of the mock appraisal committee used in the project. 

NICE’s report of the exercise is available here, but a summary of their summary is:

“The overall findings of the exercise were that:

• The NICE appraisal methods and decision framework are applicable to regenerative medicines and cell therapies.
• Quantifying and presenting clinical outcome and decision uncertainty was key to the Expert Panel consideration of the hypothetical example products.
• Where there is a combination of great uncertainty but potentially very substantial patient benefits, innovative payment methodologies need to be developed to manage and share risk to facilitate timely patient access while the evidence is immature.”

Project update: NICE RA biologics appraisal finishes!

It is safe to say that it appears to have been one of the more complex and controversial appraisals.  HEDS was involved throughout as the evidence review group; we can’t list everyone due to space constraints.  If you want to examine the voluminous documentation and high ICERs then look here.

From the NICE website:

“The guidance recommends adalimumab (Humira, AbbVie), etanercept (Enbrel, Pfizer), infliximab (Remicade, Merck Sharp & Dohme; Inflectra, Hospira UK; Remsima,Napp Pharmaceuticals ) ii, certolizumab pegol (Cimzia, UCB Pharma), golimumab (Simponi, Merck Sharp & Dohme), tocilizumab (RoActemra, Roche) and abatacept (Orencia, Bristol-Myers Squibb), each in combination with methotrexate.

Adalimumab, etanercept, certolizumab pegol or tocilizumab are also recommended as monotherapy for people who cannot take methotrexate.

In the case of certolizumab pegol, golimumab, abatacept and tocilizumab the recommendation is subject to the companies providing them as agreed in their patient access schemes.

The guidance states that treatment should be started with the least expensive drug (taking into account administration costs, dose needed and product price per dose).”

Image: My Arthritis

Assessing Managed Entry Agreements

The NICE Decision Support Unit has just published a report titled “Framework for analysing risk in Health Technology Assessments and its application to Managed Entry Agreements”.  Written by Sabine Grimm, Mark Strong, Alan Brennan and Allan Wailoo, an abridged abstract is below…..

Background: Recent changes to the regulatory landscape of pharmaceuticals may require reimbursement authorities to issue guidance on technologies with an evidence base that is less mature than has previously been the case. The greater uncertainty regarding the clinical and cost-effectiveness of new technologies at the point of decision making in a Health Technology Assessment (HTA) translates into a larger risk to the health-care payer. Decision makers need to be aware of the magnitude of those risks and the potential to manage it through assessment of a broad range of decision options, including so-called Managed Entry Agreements (MEAs).

Objective: The aim of this work was to present an analytical framework that can both quantify the need for an MEA, and assess the value of different MEAs for their reduction in the risk to the payer.

Methods: We developed the MEA risk analysis framework, an updated taxonomy of MEA schemes and the MEA design guidance questionnaire. Within this framework, we developed the concepts of Payer Uncertainty Burden (PUB), a measure of the risk associated with decision uncertainty in a HTA; and the Payer Strategy Burden (PSB) which quantifies the additional risk linked to each strategy in the HTA, given the proposed price and available evidence. We called the sum of the two the Payer Strategy and Uncertainty Burden (P-SUB). Both can be calculated with commonly used cost-effectiveness models, probabilistic sensitivity analyses (PSA) and the Sheffield Accelerated Value of Information (SAVI) online tool. We applied the MEA risk analysis framework to eight past NICE technology appraisals.

Results: The application of the framework in past NICE technology appraisals confirmed its feasibility within relatively short timelines. The value of different price reduction and evidence collection schemes depended on the uncertainties present in the appraisal and the magnitude of the PSB.

Conclusion: This report concludes that coherent, consistent and transparent assessments of proposed MEA schemes are critical. The MEA risk analysis framework proposed to routinely evaluate the decision risk in terms of Payer Uncertainty Burden and Payer Strategy Burden in HTA offers a consistent and transparent method of assessing the need for and the value of MEA schemes that is feasible within standard HTA timelines.

New staff in HEDS

New staff that have started work in HEDS over the last month or so are:

• Sabine Grimm, who is a research association working primarily within the NICE DSU.  Her work has included the development of methods for assessing Managed Entry Agreements and involvement in a Single Technology Appraisal (STA).
• Liam Wright, who is Research Assistant funded by a NIHR Fellowship.  He is currently undertaking our MSc Health Economics and Decision Modelling.
• Suzannah Bridge and Rachel Walker, who are both Information Officers working in ScHARR library.

Rheumatoid arthritis technology appraisal finishes!

After two and a half years, the NICE RA appraisal has been completed and is available in a handy 92 page FAD.  So here’s a summary, yes, in severe disease, if other treatments have failed, if discounts are available and only continued after 6 months if there is a moderate response, and if patients are started on the cheapest drug (including infliximab biosimilars).

Even with all of these provisos, it looks like an ICER of well over £30K per QALY:

“The Committee considered that the most plausible incremental cost-effectiveness ratio (ICER) for biological DMARDs used in severe active rheumatoid arthritis previously treated with methotrexate, was likely to lie between the Assessment Group’s base-case ICER (that is, £41,600 per quality-adjusted life year [QALY] gained) and the Assessment Group’s ICER for the severe group with the fastest Stanford Health Assessment Questionnaire (HAQ) progression (that is, £25,300 per QALY gained).”

The ERG for this appraisal was ScHARR-TAG, with the research team being Matt Stevenson, Rachel Archer, Jon Tosh, Emma Simpson, Emma Everson-Hock, John Stevens, Allan Wailoo, Monica Hernandez, Suzy Paisley and Kath Williams.  The work also spawned additional work for the NICE DSU, principally around HAQ progression, by Laura Gibson, Mónica Hernández-Alava and Allan Wailoo.

All documents are available here.

Not cost-effective at zero price

NICE DSU published a report assessing technologies that are not cost-effective at a zero price recently.  The abstract is below:

Photo by Leo-seta via Flickr CC BY 2.0

"In a National Institute for Health and Care Excellence (NICE) appraisal of a new drug (pertuzumab)

in metastatic breast cancer the appraisal consultation document (ACD) concluded that pertuzumab, when used in accordance with its licensed indication, did not represent a cost-effective use of NHS resources. The manufacturer had indicated in their comments on the ACD that when using plausible assumptions (those preferred by the evidence review group) there was no price at which pertuzumab would be cost-effective (it was not cost-effective at zero price). The issue driving this relatively high incremental cost-effectiveness ratio (ICER) appeared to be that the drug was given in combination with another drug (also the comparator) and any additional progression-free survival (PFS) was accompanied by the costs of both pertuzumab and the comparator drug. In view of the fact that the technology was associated with substantial benefits in terms of both progression-free and overall survival, the Institute's Guidance Executive decided not to issue the Final Appraisal Documents (FAD) pending further exploration of the issue.

The Decision Support Unit (DSU) was asked to explore the circumstances in which clinically effective technologies are not cost-effective even at a zero price. In the light of this exploration, the DSU was asked to consider the usual rules for assessing cost-effectiveness and their appropriateness or otherwise in these circumstances."

Current work at the DSU

For an idea of the methodological issues troubling NICE, the current work programme for the NICE Decision Support Unit, can be informative.  Current topics listed on their website are:

• Assessing technologies that are not cost-effective at a zero price
• A review of utilities evidence used in Technology Appraisals
• Technical Support Document (TSD) on observational data (link to further information and published work)

DSU publications

Two new publications from the NICE DSU.

• Latimer N, Abrams K.  Adjusting survival time estimates in the presence of treatment switching.
• Poku E, Rathbone J, Everson-Hock E, Essat M, Wong R, Pandor A, Wailoo A.  The safety of intravitreal bevacizumab monotherapy in adult ophthalmic conditions: systematic review.

Cost-effectiveness modelling using patient-level simulation…

…a Technical Support Document (TSD) from the NICE Decision Support Unit.  The aims of this TSD are to:

• Increase awareness of patient-level modelling approaches and highlight the key differences between patient-level and cohort-level modelling approaches.
• Provide guidance on situations in which patient-level modelling approaches may be preferable to cohort-level modelling approaches.
• Provide a description of how to implement a patient-level simulation using either a patient-level state-transition or discrete event simulation (DES) framework in a variety of software environments (Microsoft Excel®, R, TreeAge Pro®).
• Provide example models illustrating how to implement a simple cost-effectiveness model using a patient-level approach. The example models cover both DES (R, Visual Basic module within Microsoft Excel®, TreeAge Pro®) and patient-level state-transition (Microsoft Excel® and TreeAge Pro®) frameworks.
• Provide guidance on good practice when developing and reporting a patient-level simulation to inform a NICE TA.

The focus of this document is on the application of patient-level simulation to systems where patients can be assumed to be independent as systems incorporating patient interaction are not a common feature with Technology Appraisal.

Details of this guidance are available here.  The other TSDs, covering evidence synthesis, utilities, reviewing model parameters and survival analysis are available here.  TSDs on treatment switching and observational data will be available shortly.

Supporting value based assessment

The DSU has been working to support NICE with the implementation of Value Based Pricing (now Value Based Assessment) in 2013 /14.

Image: The University of Sheffield

Based in HEDS, the DSU have produced reports dealing with implementing the Department of Health (DH) estimated values for Wider Societal Benefits into cost effectiveness models (reports 1 and 2). They have also provided commentary and critical review of the DH proposals around "Burden of Illness" (report 3) and "Wider Societal benefits" (report 4).

• DSU report 1. Incorporating wider societal benefits into estimates of cost per QALY: Implications of value based pricing for NICE (Nov 2012)
• DSU report 2. Incorporating wider societal benefits into estimates of cost per QALY: Case studies (Dec 2012)
• DSU report 3. Department of Health proposals for including burden of illness into value based pricing: a description and critique (July 2013)
• DSU report 4. Department of Health proposals for including wider society benefits into value based pricing: a description and critique (August 2013)

The NICE consultation process Value Based Assessment, which will form an addendum to the Guide to the methods of Technology Appraisals, is from 27 March to 20 June 2014.

HEDS at ISPOR

6 workshops, 3 issues panels, 3 research podiums and 13 posters.  Or see us at Booth 129.  This is all neatly summarized in in the slides below.

If this isn’t friendly to your device, details are below: 

Workshops

Monday (W3) 16:45-17:45
Navigating the waters of economic evaluations of medical devices
Praveen Thokala

Tuesday (W7) 08:45-09:45
Statistical challenges in HTA
Nick Latimer

Tuesday (W8) 08:45-09:45
The economic evaluation of diagnostics: challenges and methods for assessing value
Ron Akehurst

Tuesday (W11) 08:45-09:45
Multiple challenges in capturing utilities for paediatric conditions
John Brazier

Wednesday (W24) 13:34-14:45
Methods for dealing with treatment switching in clinical trials
Nick Latimer

Wednesday (W25) 13:45-14:45
Navigating the pitfalls around progression-free survival estimation
John Stevens

Issues panels

Tuesday (IP7) 13:45 - 14:45
Making value-based pricing a reality
John Brazier (panelist)

Tuesday (IP8) 13:45-14:45
How are the results from HTA methodology research implemented in the updates of HTA guidelines?
Allan Wailoo (panelist)

Wednesday (IP17) 10:00 - 11:00
How can health economic modellers win the trust of decision makers?
Paul Tappenden (panelist)

Research Podiums

Monday (M01) 14:15-15:15
A guide to adjusting survival time estimates to account for treatment switching in randomised controlled trials
Nick Latimer

Monday (PP2) 14:15-15:15
Multinational consistency of a discrete choice model in quantifying health states for the extended 5-level EQ-5D
Ben van Hout

Monday (CL3) 15:30-16:30
Validation of surrogate endpoints in advanced solid tumours: systematic review of statistical methods, results and implications for policy makers.
Paul Tappenden

Posters

Monday (PRS6) Discussion: 13:15-14:15
Systematic review of colistimethate sodium dry powder and tobramycin dry powder antibiotics for pseudomonas aeruginosa lung infection in cystic fibrosis
Sue Harnan

Monday (PRS38) Discussion: 13:15-14:15
The cost-effectiveness of dry powder antibiotics for the treatment of pseudomonas aeruginosa in patients with cystic fibrosis
Paul Tappenden

Monday (PCN89) Discussion: 18:30-19:30
Economic evaluation of an electrical impedance spectoscopy (EIS) device used as an adjunct to colposcopy
Alice Bessey

Monday (PCN97) Discussion: 18:30-19:30
A systematic review and critical appraisal of economic evaluations of radiotherapy for cancer
Roberta Ara

Monday (PCN107) Discussion: 18:30-19:30
Economic evidence of surgical procedures in cancer:
a systematic literature review
Hasan Basarir

Monday (PCN121) Discussion: 18:30-19:30
Evaluating the cost effectiveness of gene expression profiling and immunohistochemistry tests
Rachid Rafia

Tuesday (PHP216) Discussion: 12:45-13:45
Comparing the value of different HTA decision making process: evaluating the evaluators
Allan Wailoo

Tuesday (PCV90) Discussion: 18:00-19:00
Issues with cost-effectiveness modelling of diagnostic tests – case study of ischaemic cardiomyopathy
Praveen Thokala

Tuesday (PCV105) Discussion: 18:00-19:00
Telemonitoring after discharge from hospital with heart failure – cost-effectiveness modelling of alternative service designs
Praveen Thokala

Tuesday (PMS61) Discussion: 18:00-19:00
Health economic modelling of sequential therapies for rheumatoid arthritis: A systematic review
Jon Tosh

Wednesday (PRM32) Discussion: 12:45-13:45
Development of the ScHARR Health Utilities Database (HUD)
Angie Rees

Wednesday (PRM178) Discussion: 12:45-13:45
Modelling the relationship between the Womac osteoarthritis index and EQ-5D
Allan Wailoo

Wednesday (PRM230) Discussion: 12:45-13:45
Should there be an option to “unrefer” NICE single technology appraisals: case study of aripiprazole for bipolar I disorder in adolescents
Lesley Uttley

NICE ranibizumab decision

Ranibizumab (Lucentis, Novartis) for treating visual impairment caused by macular oedema secondary to retinal vein occlusion has now been published.  This is after three appraisal committee meetings and a big piece of work for the DSU looking at issues related to quality, use, efficacy and safety of bevacizumab in eye conditions.

Ranibizumab is recommended as an option for treating visual impairment caused by macular oedema following central retinal vein occlusion or following branch retinal vein occlusion only if treatment with laser photocoagulation has not been beneficial, and if the manufacturer provides ranibizumab with the discount agreed in the patient access scheme revised in the context of NICE technology appraisal guidance.

NICE DSU updates

Three of the technical support documents (TSDs) were updated in March.  They are:

TSD 2: A general linear modelling framework for pair-wise and network meta-analysis of randomised controlled trials and the associated WinBUGS system(.odc) files.

TSD 4: Inconsistency in networks of evidence based on randomised controlled trials
WinBUGS system(.odc) files.

TSD 14: Survival analysis for economic evaluations alongside clinical trials - extrapolation with patient-level data.

DSU report on societal preferences for end of life

The NICE Supplementary Guidance (2009) specifies conditions under which 'end of life' treatments may be given higher priority than other health care interventions. This project seeks to examine whether a social preference exists for giving higher priority to life-extending, end of life treatments than to other treatments, using preference data elicited from the general public.

NICE commissioned the DSU to undertake two studies: (i) a small scale 'preference validation' study designed to develop a better understanding of the reasons and principles underpinning people's priority setting preferences; and (ii) a large scale discrete choice experiment designed to provide a thorough and robust investigation of the preferences of a representative sample of the general public.

A literature review, the validation study and associated presentations have been previously published on the DSU web site.  The results of the main DCE are now available, in full or as a summary.  The key findings are:

• There is no evidence that respondents on average are willing to sacrifice aggregate health gains in order to give priority to the treatment of end of life patients.
• Most respondents choose to treat the patient who is closest to their end of life only when the benefits of treating that patient (in terms of QALYs gained) are similar to or greater than the benefits of treating the non-end of life patient.
• Whilst both types of gain appear to be important, the results from the regression analysis suggest that life-extending treatments are valued more highly than quality of life-improving treatments that offer similar gains in terms of QALYs.
• All else being equal, respondents are more likely to choose to treat a patient who has just found out about their illness than one who has known about their illness for some time.

DSU report on estimating utilities from clinical outcomes

This reports is a detailed case study using  a large observational database of patients diagnosed with Rheumatoid Arthritis (n=100,398 observations). Direct estimation of UK EQ-5D scores as a function of Health Assessment Questionnaire (HAQ), pain and age was performed using a limited dependent variable mixture model. And indirect modelling (“response maping”) was undertaken using a set of generalized ordered probit models with expected tariff scores calculated mathematically. Linear regression was reported for comparison purposes.

The report concludes that modelling of EQ-5D from clinical measures is best performed directly using the bespoke mixture model. This substantially outperforms the indirect method in this example. Linear models are inappropriate, suffer from systematic bias and generate values outside the feasible range.

The report, including link is:
M Hernandez Alava, A Wailoo, F Wolfe, K Michaud. A comparison of direct and indirect methods for the estimation of health utilities from clinical outcomes.