Table 3.
HTA methods and techniques applied
| France (HAS/CEESPa) |
Germany (IQWiG) |
Sweden (TLV) |
England (NICE) |
Italy (AIFA) |
Netherlands (ZIN) |
Poland (AOTMiT) |
Spain (RedETS/ISCIII or ICP) |
|
|---|---|---|---|---|---|---|---|---|
| Analysis method | ||||||||
| Methods | Comparative efficacy/effectiveness (also CEA, CUA) | CBA but also CUA and CEA (not standard practice) | CUA (also CEA, CBA) | CUA (also CEA, CMA) | CMA, CEA, CUA, CBAb | CEA, CUA, no CMA | Cost-consequences analysis, CEA or CUA—obligatory, CMA (if applicable) | Comparative efficacy/effectiveness, CMA, CEA, CUA, CBAc |
| Preferred outcome measure | Final outcome, life years (QALY, if CUA; life years, if CEA) | Patient relevant outcome (can be multidimensional)—efficiency frontier | QALY (WTP, if CBA) | QALY (cost per life year gained, if CEA) | Final outcome, life years (QALY, if CUA or CEA; life years, if CEA) | Effectiveness by intention-to-treat principle, and expressed in natural units—preferably LYG or QALY | QALY or LYG | QALY in CUA |
| Utility scores elicitation technique | EQ-5D and HUI3, from general French population | Utility scores from patients, direct (e.g. TTO, SG), indirect | Utility scores from patients, direct (e.g. TTO, SG), indirect (EQ-5D) | Utility scores from general English population, direct (e.g. TTO, SG), indirect (EQ-5D), systematic review | Both direct and indirect (EQ-5D) elicitation techniques | Either direct (TTO, SG, VAS), or indirect (EQ-5D); selection should be justified | Direct or indirect utility scoresd | Utility scores from general Spanish population, direct (e.g. TTO, SG), indirect (EQ-5D)e |
| Comparator | Usually ‘best standard of care’ but can be more than onef | Usually ‘best standard of care’ but can be more than oneg | Usually ‘best standard of care’ but can be more than oneh | Usually ‘best standard of care’ but can be more than oneI | Usually ‘best standard of care’ but can be more than onej | Treatment in clinical guidelines of GPs; if not available, most prevalent treatment | ‘Best standard of care’ which is reimbursed in Polandk | Best standard of care, usual care and/or more cost-effective alternative |
| Perspective | Widest possible to include all health system stakeholdersl | Usually statutory health insurantm | Societal | Cost payer (NHS) or societal if justified | Italian National Health Servicen | Societal (report indirect costs separately) | The public payer’s perspective, public payer and patient (by law) | Cost payer (NHS) and societal (rarely used), and they should be presented separately |
| Subgroup analysis | Yes (when justified) | Yes | Yes | Yes | Yes | Yes | Yes (if needed, but decreases validity) | Yes |
| Clinical evidence | ||||||||
| Preferred study design | Head-to-head RCTs; other designs accepted if no RCTs available | Head-to-head RCTs; other designs accepted in the absence of RCTs | Head-to-head RCTs; other designs accepted if no RCTs available | Head-to-head RCTs; other designs accepted if no RCTs available | Head-to-head RCTs; other designs accepted if no RCTs available | Head-to-head RCTs | Head-to-head RCTs; other designs accepted if no RCTs available | Head-to-head RCTs; other designs accepted if no RCTs available |
| Systematic literature reviews for collecting evidence required/conducted by regulator | Yes, guidelines provided/yes, in French | Yes/no | Not mandatory | Yes/yes | Yes/yes | Yes/yes | Yes | Not alwayso |
| Meta-analysis for pooling evidence | Not specified | Not specified for new drugs | Not specified | Yes | Yes | Yes, encouraged | Yes | Nop |
| Data extrapolation | Qualitative only, in absence of effectiveness data form RCTs | No | Quantitative, both in absence of RCT effectiveness data and in absence of long-term effects | Qualitative and quantitative, both in absence of RCT effectiveness data and in absence of long-term effects |
Quantitative Qualitative in absence of RCT effectiveness data |
Qualitative, in the absence of RCTs and in absence of long-term effects | Possible if needed but not recommended | Quantitative, in the absence of effectiveness data |
| Resources/costs | ||||||||
| Types | Direct medical, direct non-medical, indirect (both for patient and carer) | Depending on perspective: direct medical, informal costs, productivity loss (as costs) | Direct medical, direct non-medical, indirect (both for patient and carer) | Direct medical, social services | Direct costs only; indirect costs can be taken into account in a separate analysis | Both direct and indirect costs inside and outside the healthcare system | Direct medical costs, direct non-medical costs | Direct and indirect costs (on rare occasions), costs of labour production losses or lost time, informal care costs |
| Data source/unit costs |
Direct: PMSI (Programme de Médicalisation des Systèmes d’Information) Indirect: human capital costing, friction costing |
Statutory health insurance, further considerations depending on perspective chosen |
Drugs: pharmacy prices Indirect: human capital costing |
Official DoH listing | Variety of sourcesq | Reference prices list should be used | Variety of sourcesr | Official publications, accounts of health care centres, and the fees applied to NHS service provision contracts |
| Discounting | ||||||||
| Costs | 4% (up to 30 years) and 2% after | 3% | 3% | 3.5% | Not available (update in progress) | 4% | 5% | 3% |
| Outcomes | 4% (up to 30 years) and 2% after | 3% | 3% | 3.5% | Not available (update in progress) | Under review—will probably be set at same level as costs discounting | 3.5% | 3% |
| Sensitivity analysis | 0%, 3% (6% max) | 0–5% | 0–5% | 0–6% | Not available (update in progress) | Not obligatory |
5 and 0% for costs and outcomes 0% for outcomes 5% for costss |
0–5% |
| Time horizon | ||||||||
| Time horizon | Long enough so that all treatment outcomes can be included | At least the average (clinical) study duration; longer for chronic conditions, especially if lifetime gains are expected; same horizon for costs and benefits | Time needed to cover all main outcomes and costs | Long enough to reflect any differences on outcomes and costs between technologies compared | Duration of the trial is consideredt | Primarily based on duration of RCTsu | Long enough to allow proper assessment of differences in health outcomes and costs between the assessed health technology and the comparators | Should capture all relevant differences in costs and in the effects of health treatments and resourcesv |
| Thresholds | ||||||||
| Thresholds | No threshold (only eligibility threshold to conduct economic evaluation) | Efficiency frontier (Institute’s own approach) | No official threshold; 50% likelihood of approval for ICER between €79,400 and €111,700 | £20,000–£30,000 per QALY; Empirical: £12,936 per QALY | No threshold in use | No official threshold | 3 × GDP per capita for ICUR(QALY) or ICER(LYG) | Unofficial: €21,000–€24,000/QALY (recently provided by SESCSw to the Spanish MoH) |
Source The authors (based on literature review findings and expert consultation)
CEA Cost-effectiveness analysis, CUA cost utility analysis, CMA cost minimization analysis, QALY quality adjusted life year, LYG life year gained, TTO time trade off, SG standard gamble
aIn France, economic evaluations are undertaken only for selected drugs with expected significant budget impact
bA template for the submission of the pricing and reimbursement (P&R) dossier to AIFA is in progress
cFor the case of drugs at central level carried out by ICP, comparative efficacy/effectiveness is taken into account. The ICP receives the so called “Informe de Posicionamiento Terapéutico” (Therapeutic Positioning report), a therapeutic assessment conducted by the Spanish Medicines Agency (Agencia Española del Medicamento) based on which confidential discussions around the appraisal of the drugs takes place but which does not take into consideration cost-effectiveness. Economic evaluations are mainly taking place for the case of non-drug technologies under the scope of RedETS
dIt is recommended to use indirect methods for preferences measurement—validated questionnaires in Polish. While measuring preferences with the EQ-5D questionnaire, it is advised to use the Polish utility standard set obtained by means of TTO
eSurveys or previously validated HRQOL patient surveys
fIncluding most cost-effective, least expensive, most routinely used, and newest
gIncluding most cost-effective, least expensive, and most routinely used. If the efficiency frontier approach is used as part of CBA, then “all relevant comparators within the given indication field” must be considered
hIncluding most cost-effective, least expensive, and most routinely used
IIncluding most cost-effective, least expensive, most and routinely used
jIncluding most cost-effective, and most routinely used
kThese might include (1) most frequently used; (2) cheapest; (3) most effective; and (4) compliant to the practical guidelines
lNeeds justification (especially if societal)
mAlso community of statutorily insured, perspective of individual insurers, or the societal perspectives are possible
nSocietal perspective is not mandatory, but can be provided in separate analysis
oFor non-drugs under RedETS, a systematic literature review is always conducted
pFor non-drugs under RedETS, a meta-analysis may be conducted
qPrices available in the Official Journal of the Italian Republic (Gazzetta Ufficiale), accounts of health care centres, the fees applied to NHS service, scientific literature/ad hoc studies
rIncluding (1) list of standard costs, (2) formerly published research, (3) local scales of charges, (4) direct calculation
sIt is currently under revision (AOTMiT HTA Guidelines updating process) and may change soon
tAdditional long-term evidence collected through monitoring registries
uSecondary horizons include any longer needed depending on the context of interest
vIn some cases, the time horizon will have to be extended to the individual’s entire lifespan
wServicio de Evaluación y Planificación, Islas Canarias