Reviews that survive appraisal
An HTA body will check how the evidence was found before it looks at what it shows. Search strategies, eligibility criteria, screening decisions and risk-of-bias judgements all need to be transparent and reproducible. We design every review so an assessor can follow it from protocol to results.
Types of review
- Clinical efficacy and safety: the evidence base for relative effectiveness and indirect comparisons
- Economic evaluations: published cost-effectiveness studies and models
- Health-related quality of life and utilities: inputs for the economic model
- Costs and resource use: local and published cost evidence
- Targeted literature reviews: faster, focused reviews where a full SLR is not required
How we work
- Protocol: PICO, eligibility criteria and search strategy, agreed before screening starts.
- Searching: Bibliographic databases such as MEDLINE, Embase and Cochrane CENTRAL, plus conference proceedings, trial registries and HTA agency websites.
- Screening and extraction: Two independent reviewers, with conflicts resolved and recorded.
- Quality assessment: Validated tools such as RoB 2 and ROBINS-I.
- Reporting: PRISMA flow diagram, evidence tables and a feasibility assessment for indirect comparisons.
One review, several markets
A global SLR can serve the EU JCA and national submissions if it is designed for that from the start. We set the scope to cover the comparators each agency is likely to expect, and plan for the updates you will need before each submission. See SLR updates.
Where AI tools help with screening, we use validated workflows with human review of every decision, in line with NICE's position statement on AI in evidence generation.
What HTA bodies expect of a review
HTA bodies set out clear expectations for how evidence is identified. NICE's technology appraisal and highly specialised technologies manual (PMG36) asks for a systematic review of the relevant evidence, done according to a previously prepared protocol, and describes a systematic review as one that assembles all available relevant evidence using explicit, valid and replicable methods. Where non-randomised studies are used, the manual asks for risk of bias to be assessed with a validated tool, such as ROBINS-I.
For the EU joint clinical assessment, the guidance on filling in the dossier template for medicinal products is specific. Searches should at least cover MEDLINE and the Cochrane Central Register of Controlled Trials, alongside study registries and study results registries, HTA reports on the medicine from EEA States applying the Regulation and from Australia, Canada, the United Kingdom and the United States, the clinical data in the EMA submission and patient registries. The search cut-off date should be a maximum of three months before dossier submission.
Reporting follows PRISMA 2020, which comprises a 27-item checklist, an abstract checklist and flow diagram templates for original and updated reviews.
Choosing the right review
| Review | Question it answers | Where the results go |
|---|---|---|
| Clinical efficacy and safety SLR | What is the relative effect of the technology and its comparators? | Clinical sections of the submission, the feasibility assessment and the indirect comparisons |
| Economic evaluation SLR | How have others modelled this decision, and with what results? | Model conceptualisation and the economic sections |
| HRQoL and utility SLR | Which health state utility values are available and appropriate? | Utility inputs and scenarios in the model (see NICE DSU TSD 9) |
| Cost and resource use SLR | What does care cost, and what resources does it use? | Cost inputs and the budget impact model |
| Targeted literature review | A focused question that does not need a full systematic review | Early strategy, landscaping and evidence planning |
NICE DSU TSD 13 sets out how evidence can be identified and reviewed to inform both the conceptualisation and the population of cost-effectiveness models, which helps decide where a systematic approach adds most value.
What a strong systematic review looks like
- A protocol written first: Questions, eligibility criteria and methods fixed before screening starts.
- Searches reported in full: Every strategy, date and source documented so the search can be rerun exactly.
- A scope that anticipates the comparators: Wide enough for the comparators NICE, national agencies and the JCA are likely to request.
- Recorded decisions: Screening and extraction decisions that can be traced from the PRISMA flow diagram back to individual records.
- Extraction designed for synthesis: Data captured in the form the network meta-analysis and the model will use.
- A clear route to updates: Search strategies kept ready to rerun before each submission; see when an HTA review needs updating.