Guide for HTA submissions

Real-World Evidence for Nordic HTA: A Guide to Sweden, Norway and Denmark

Global real-world evidence strategies often require substantial adaptation for Nordic health technology assessment. This guide explains how TLV in Sweden, NOMA in Norway and Medicinrådet in Denmark use real-world data, identifies common sources of delay, and shows how a shared analytical core can support three country-specific submissions.

This guide focuses on pharmaceutical assessment pathways in Sweden, Norway and Denmark. Finland, Iceland and medical-device-specific pathways are outside its main scope.

Why Nordic data is different

Personal identifiers enable individual-level linkage across national health, prescription and mortality registers, creating unusually comprehensive opportunities for longitudinal population-based research. Coverage, data quality, access conditions and reporting lag nevertheless differ between registers and must be evaluated for each study. Coverage is register-specific. Sweden's National Patient Register covers specialised care but not primary care, while the Prescribed Drug Register covers prescribed medicines dispensed through pharmacies.

Three assessment contexts

Sweden assesses societal cost-effectiveness within an ethical framework where severity matters. Norway weighs benefit, resource use and severity, with severity quantified through absolute shortfall. Denmark assesses comparative effect and safety alongside a QALY-based cost-utility analysis, budget impact and uncertainty. The same product therefore requires three locally adapted decision analyses.

One reusable core

A shared, well-documented analytical core can reduce duplication and improve consistency, provided that it feeds country-specific models and dossiers reflecting local comparators, populations, costs, value sets, perspectives and submission requirements.

What real-world evidence means to an HTA assessor

Regulators determine whether a product's quality, safety and benefit-risk profile support market authorisation. HTA bodies then assess its comparative value and relevance to local clinical practice. That is where real-world data can contribute by describing who is treated, which treatments they receive, for how long, at what resource use and with what outcomes.

An assessor reading a submission is not evaluating the data source in the abstract. They are asking whether this analysis, on this population, with these adjustments, supports the specific input it is being used to justify in the economic model or the clinical comparison. A large dataset cannot compensate for an unclear design, inappropriate comparison or uncontrolled bias.

In practice, real-world evidence is commonly used to inform epidemiology and disease burden, patient characteristics, treatment patterns and comparator selection, healthcare resource use, adherence and persistence, natural history, long-term outcomes and the external validation of extrapolations. Official unit-cost sources and national mortality tables are often used alongside real-world evidence. Real-world evidence receives the greatest scrutiny when used to estimate comparative treatment effects.

Requirements by country

Sweden

TLV (Tandvårds- och läkemedelsförmånsverket)

How value is judged

TLV applies value-based pricing and evaluates cost-effectiveness within Sweden's ethical platform, which comprises the human-dignity, need-and-solidarity and cost-effectiveness principles. Health-economic analyses generally take a societal perspective, although TLV now applies greater caution when including productivity effects. Sweden does not apply one universal published cost-per-QALY threshold, and the level considered reasonable depends partly on the severity and circumstances of the condition.

Where real-world evidence fits

Swedish health registers and disease-specific quality registers can inform epidemiology, patient characteristics, treatment patterns, comparator selection, resource use and long-term outcomes. Their suitability is assessed for the specific purpose, and national coverage should not be assumed to imply complete capture of every relevant clinical variable. Post-launch follow-up matters throughout the medicine lifecycle, particularly where uncertainty exists about utilisation or treatment effects.

Norway

NOMA (Norwegian Medical Products Agency)

How value is judged

NOMA assesses the benefit, resource use and severity criteria collectively, together with uncertainty and budget consequences. In cost-utility analyses, severity is quantified as absolute shortfall: the expected QALYs lost under current standard care compared with an age-matched general population. NOMA conducts pharmaceutical single-technology assessments. For medicines financed by the specialist health service, Sykehusinnkjøp conducts negotiations after the assessment and the Decision Forum makes the adoption decision.

Where real-world evidence fits

Real-world data may inform epidemiology, demographics, treatment duration, real-world utilisation, effectiveness, safety and long-term follow-up. Registry data can also inform the age and prognosis used in absolute-shortfall calculations and help assess the plausibility of long-term extrapolations. When real-world data is used for relative efficacy, NOMA requires transparent and reproducible methods, explicit assessment of bias and a discussion of relevance to Norwegian practice.

Denmark

Medicinrådet (Danish Medicines Council)

How value is judged

Medicinrådet assesses whether a medicine's effect, measured in QALYs, and its safety are reasonably balanced against the costs of introducing it compared with current Danish standard care. The economic evaluation is normally a cost-utility analysis reporting an ICER and takes a limited societal perspective. Budget impact and uncertainty are also considered.

Where real-world evidence fits

Real-world evidence may be used to describe Danish patient populations, medicine utilisation and disease natural history, support the assessment of treatment effects, and externally validate economic-model extrapolations. Randomised trials, preferably direct head-to-head trials against the relevant Danish comparator, remain the preferred source for comparative effects. When real-world evidence is used comparatively, Medicinrådet requires a protocol and transparent, reproducible reporting, including assessment of data quality and bias. For comparative-effect studies, it recommends target-trial emulation.

This guide reflects guidance current as of August 2026, including TLV's current health-economics guidance, NOMA's July 2026 submission guidelines, the Nye metoder process description, Medicinrådet's 2026 methodological guideline and Medicinrådet's current real-world evidence guidance, version 1.0 from 2023. Requirements are updated regularly, so always confirm the applicable methods and submission guidance before finalising an analysis plan.

Five gaps between global RWE strategy and local Nordic submissions

1. One global RWE package, three different questions

A single international evidence deck rarely answers what each Nordic agency actually asks. Sweden assesses societal cost-effectiveness within its ethical platform, Norway weighs benefit, resource use and severity together, and Denmark assesses comparative effect and safety alongside a QALY-based cost-utility analysis. Map every deliverable to the specific question it answers before analysis starts.

2. Comparators that do not reflect local practice

Treatment sequencing differs across Sweden, Norway and Denmark. A comparator that is standard of care in a pivotal trial may be a minority regimen locally, which weakens the incremental analysis. Register-based treatment-pattern studies can quantify the extent to which candidate comparators reflect local practice. Where new data access is required, this work must be planned well in advance.

3. Unadjusted external controls

External controls constructed from registry data are subject to particularly close scrutiny because of confounding, selection, time-related bias and differences in outcome measurement. Pre-specify the causal question, time zero, eligibility criteria, estimand, confounders, adjustment methods and sensitivity analyses. Where appropriate, structure the analysis as a target-trial emulation.

4. Underestimating data access timelines

Data access, governance approvals, contracting and delivery frequently take months. The precise approvals depend on the country, purpose and data source. RWE feasibility work should therefore begin during clinical development rather than after regulatory approval. In Sweden, Socialstyrelsen currently reports approximately three months before assignment to a case officer, followed by another two to four months for processing.

Socialstyrelsen data-order processing times

5. Models that assessors cannot audit

Health-economic models must be transparent, auditable and sufficiently flexible to test alternative assumptions and scenarios. Inputs, calculations and outputs should be traceable, and relevant parameters should be editable. Norway explicitly prefers Microsoft Excel and requires prior contact if alternative software is proposed. Interactive applications can support internal scenario testing and communication, but they should complement rather than replace the agency's required model format.

Submission readiness checklist

Define the decision problem separately for each Nordic agency, including perspective, comparator and time horizon.
Run a data feasibility assessment before committing to an analysis plan, covering coverage, lag, linkage and sample size.
Pre-specify the protocol and statistical analysis plan, including confounder set, time zero and missing data strategy.
Document data provenance and validation for every register variable used in the economic model.
Build background mortality from relevant national life tables where appropriate. Use locally representative prognosis data when available and justify the applicability of international sources when local evidence is unavailable.
Prepare a reproducible analysis pipeline so that assessor questions can be answered within the review timeline.
Where post-launch evidence collection or reassessment is anticipated, plan it early and assess whether the original data sources and definitions can be reused.

Frequently asked questions

What is the difference between real-world data and real-world evidence?

Real-world data is the raw information collected outside a randomised trial, such as national registers, electronic health records, claims and quality registries. Real-world evidence is the clinical or economic conclusion generated when that data is analysed with a defined design and statistical method. HTA bodies assess the evidence, so the design and its transparency matter as much as the data source.

Can real-world evidence replace a randomised controlled trial in a Nordic HTA submission?

Generally not. Randomised trials remain the preferred basis for relative treatment effects, across all three countries. Real-world evidence commonly strengthens the submission by informing epidemiology, comparators, resource use, adherence, long-term extrapolation and local prognosis. It may also contribute to comparative-effect estimation when randomised evidence is unavailable, but this requires substantially stronger justification and bias assessment.

Which Nordic data sources are used most often?

Common sources include national patient registers, prescribed-drug registers, cause-of-death registers, population registers and disease-specific clinical quality registers. These sources can often be linked at individual level, subject to legal approval and data-access requirements. Coverage remains source-specific. Patient registers may omit parts of primary care, prescription registers generally measure dispensing rather than ingestion, and quality-register coverage varies by disease, geography and time.

How early should real-world evidence planning start?

Ideally during clinical development. Data access, governance approvals, contracting and delivery frequently take months, and the evidence needed for Nordic submissions often differs from the trial endpoints. Early feasibility work avoids launching a submission with evidence gaps that cannot be closed in time.

Is real-world evidence used the same way for medtech as for pharma?

No. The principles of fit-for-purpose data, transparent design and local relevance apply to both medicines and medical technologies, but the assessment pathways and evidence requirements are not interchangeable. Medicinrådet's methods described in this guide apply to medicines. Sweden and Norway also have separate processes and guidance for medical-device assessments. Real-world evidence can be particularly valuable for studying implementation, learning effects, organisational consequences and post-launch performance, but the requirements must be checked against the relevant medtech pathway.

Turning this into a submission ready evidence package

We combine pharmacoepidemiology, health economics and application development, which means the same team can design the registry study, build the economic model in the format the agency requires, and add interactive tools for scenario testing and internal communication. Talk to us about your Nordic evidence plan.

Nordic reimbursement and market access Custom analytical tools