Guidance

UK NSC: horizon scanning

Published 29 September 2022

1. Overview

Many existing UK National Screening Committee (UK NSC) activities can be described as horizon scanning and use recognised best practice methods.

The expansion of the UK NSC’s remit, to include targeted and stratified screening as well as population screening, provides opportunities to increase and formalise these horizon scanning functions. The recommendations of the 4 UK chief medical officers (CMOs) and the updated UK NSC terms of reference shift the perspective to a further horizon.

This document highlights how the UK NSC can improve and expand its current practices, bringing them together into a consolidated horizon scanning system.

2. Background

2.1 Horizon-scanning organisations

Organisations that carry out horizon scanning activities include the Innovation Observatory which provides the national horizon scanning function for the National Institute for Health and Care Research (NIHR), and the European Network for Health Technology Assessment (EUnetHTA) and Health Technology Assessment International, which work collaboratively and internationally across regions.

2.2 Definitions

A systematic review by Hines and others ^{[footnote 1]} identified varying definitions for horizon scanning but said most could be captured with the following broad description:

A systematic examination of information sources to detect early signs of important developments.

The EUnetHTA definition ^{[footnote 2]} describes horizon scanning as:

The systematic identification of health technologies that are new, emerging or becoming obsolete and that have the potential to affect health, health services and/or society.

The definition in the UK government’s Futures toolkit for policy-makers and analysts focuses on strategic issues but the principles are the same:

Horizon scanning is a technique for looking ahead. Its focus is the future rather than the present and its purpose is to identify the strategic issues that will be important. Mostly, these will be different from the issues that are important today.

The literature refers to new technologies and conditions flagged during horizon scanning as signals, signs or topics. In this document we use the term ‘signals’.

3. Best practice

The EUnetHTA Horizon Scanning, Topic Identification, Selection and Prioritisation (TISP) recommendations and the steps reported in the only systematic review of horizon scanning methodologies (Hines and others, 2019) broadly agree on the following best practice steps for horizon scanning:

1. Identification/detection
2. Filtration
3. Prioritisation
4. Assessment
5. Dissemination
6. Follow-up/evaluation

These steps form a cyclical process and are required for a good systematic horizon scanning process. They also provide a set of principles we can use to assess the UK NSC’s horizon scanning practices.

The EUnetHTA recommends involving stakeholders at every stage and proactively asking independent stakeholders, including the public, to submit proposals for prioritisation. The Hines systematic review also reports best practice of stakeholder involvement occurring throughout the process.

The EUnetHTA report recognises that the identification, selection (filtration) and prioritisation steps may not be distinct. If all topics that are identified are to be assessed then prioritisation is not necessary.

Importantly, Hines and others identify transparency for all steps as a guiding principle for horizon scanning systems. This could be enhanced by making the horizon scanning criteria used to make decisions and recommendations publicly available ^{[footnote 1]}.

3.1 Identification/detection

The purpose of this step is to systematically identify and/or define sources used to detect signals in horizon scanning. It also sets out how to systematically search or browse those sources.

The 2019 systematic review identifies, and the EUnetHTA recommends, systematic and ad hoc searching of a wide variety of sources that are both reactive and proactive.

The systematic review identified the following sources used for signal detection in horizon scanning, with scientific literature being the most common:

• scientific/biomedical literature review
• patents
• input from industry and industry associations
• other observatories
• media
• international institutions and forums
• individuals, committees and expert groups
• surveys
• government bodies
• meetings and conferences
• other organisations
• grey literature

Systematically applied structured searches of scientific literature can be both sensitive and specific. These might be suitable for a broad remit or ‘deep dive’ respectively. More specific searches can be more useful once priorities have been identified.

In addition, monitoring investments in relevant industries and closer links, or consultation with, them can be useful ^{[footnote 1]}.

Social media, interviews, surveys and workshops with stakeholder input from experts and the public have also been reported as sources to identify signals. These approaches require a range of different methods for searching.

3.2 Filtration

Filtration is the triage stage for horizon scanning, when it is decided if signals detected from the sources are relevant. The Hines systematic review identified 2 important criteria for filtering out irrelevant topics:

• cost (and cost utility ratio)
• implications (impact on quality of life, burden of disease and patient safety)

The EUnetHTA states that filtration should focus on signals that are eligible for assessment.

Timescales are important. The Hines systematic review estimates horizon scanning to be most useful for technologies expected to be available within 2 to 15 years. Less than 2 years is not classed as horizon scanning, while longer than 15 years is classed as being too far in the future ^{[footnote 1]}.

The Futures toolkit describes the Three Horizons model:

• horizon 1 (present and near future, within the next 2 years)
• horizon 2 (medium term)
• horizon 3 (long term)

The main purpose of horizon scanning would therefore be to identify medium and long term signals.

The EUnetHTA suggests that an important consideration is the estimated timeframe of when a signal might be expected to be available and approved for use in a screening programme.

The Hines systematic review provides more detail on additional criteria that might affect filtration, such as the technology’s stage of development and the level of evidence available. Methods reported for carrying this out may be a simple binary yes/no (as also suggested by the EUnetHTA) or may be ranked for example, using a traffic light system, or categories such as confirmed, likely, potential, unlikely and questionable. The EUnetHTA says processes should be efficient and simple.

The systematic review suggests using the following filtration criteria and methods to discard irrelevant signals:

Filtration criteria

Potential impact, size of affected population or global relevance, novelty, level of innovation, evidence, organisational impact, plausibility, levels of stakeholder and media interest.

Filtration methods

Classification criteria, automated text-mining tools, individual and group filtration, peer review, expert participation.

3.3 Prioritisation

This is the step when signals that have passed initial filtration are prioritised. This helps narrow down the number of signals evaluated according to available resources. Prioritisation criteria can be used to create more specific strategies to search or scan for signals relating to the organisation’s priorities rather than casting a wider, sensitive net.

As with filtration, the timeframe of when a signal might be expected to be available is an important consideration in the prioritisation step ^{[footnote 1]}.

The literature emphasises the importance of public consultation with a diverse group of experts from a range of locations and disciplines ^{[footnote 1]}.

The systematic review suggests the following prioritisation criteria and methods to assess signals:

Prioritisation criteria

Potential impact on outcomes; size and composition of the affected population; expected variation of impact; likely time frame; evidence of effectiveness; relevance to strategic and political priorities; effect on other related policies; desirability; factual basis; requirement of availability of expertise; novelty.

Prioritisation methods

Qualitative approach; quantitative or semi-quantitative approaches; rating and ranking; best-worst scaling; risk analysis; standardisation of signals; Delphi approach; public consultation; engagement of experts; mixed methods.

3.4 Assessment

The assessment stage highlights signals that have the most potential to:

• be integrated into an organisation’s objectives
• make a positive impact on outcomes and the target population

The outputs of horizon scanning within the EUnetHTA collaboration are full health technology assessments. Various methods of assessment are identified in the Hines systematic review.

The assessment stage also needs to consider practicalities such as whether further research is needed, time, specialised training required, resource implications, cooperation and acceptance from key stakeholders, ethics, whether introducing the new signal will be disruptive and if new regulations and guidelines are needed ^{[footnote 1] [footnote 2]}.

3.5 Dissemination

This step ensures the system is transparent by communicating to the public and stakeholders how horizon scanning is conducted and its outcomes.

3.6 Follow-up/evaluation

This step involves following up and evaluating outputs of horizon scanning and the processes underpinning it.

Outputs can be evaluated in the short, medium and long term.

4. UK NSC practice

Existing UK NSC proactive horizon scanning activities include the annual call for topics and the programme modification request process. Many current UK NSC activities could be described as near horizon scanning.

New evidence is brought to the committee’s attention via a literature scanning process which is cascaded on an almost daily basis.

Existing recommendations are updated on the basis of new evidence.

There is a proactive annual call for proposals for potential new screening programmes.

New developments are communicated directly to the committee via the wide range of expertise on the committee and its reference groups.

The UK NSC is receptive to working with researchers to develop studies that will effectively inform any form of screening and the committee has particularly strong links with the NIHR.

4.1 Current awareness

The UK NSC’s current awareness service (its literature scanning process) has been in place since 2011 to keep the UK NSC secretariat up to date with published research in screening. This has evolved to use different reactive and proactive search and scanning methods. It is not possible to provide fully comprehensive coverage, so there is an element of filtration to control the volume of sources searched and make it manageable. The informal criteria on what should be included in the alert has increased over time.

The service recently expanded to include population, targeted and stratified screening in line with the UK NSC’s expanded remit. The service also expanded to include ongoing primary and secondary research at an earlier stage using, for example, clinical trials databases and databases such as PROSPERO, Cochrane protocols and NIHR/HTA models.

Updates of international policy or recommendations from organisations across the world, such as the US Preventive Services Task Force (USPSTF), Canadian Task Force on Preventive Health Care, Canadian Agency for Drugs and Technologies (CADTH), Australian government websites, New Zealand National Screening Unit, Health Council of the Netherlands, and government websites for the Scandinavian countries, are already scanned at least weekly and incorporated into the current awareness article alert.

The current awareness article alert is disseminated to the UK NSC secretariat and evidence team, members of the committee and reference groups. A record is maintained of the conditions and technologies included in the alerts.

4.2 Regular recommendation updates

The evidence team uses evidence maps and evidence summaries to assess the relevant literature before commissioning full systematic reviews and cost effectiveness analyses.

4.3 Annual call for topics, early updates, and programme modifications

The annual call for topics enables anybody to submit a proposal for a new screening programme.

Stakeholders can also suggest modifications to current programmes, for example new tests or new implementation methods, as well as early updates for conditions on the regular review list if significant new evidence is published before the next scheduled review.

Signals submitted by stakeholders through these processes are followed up in the first instance by an evidence map, which might be considered a filtration document. An evidence map documents the volume and type of evidence available and can be used to prioritise what may be taken forward for more in-depth assessment. The assessment of any signal is ultimately measured against the criteria for a screening programme, either for population or targeted screening.

5. Examples of the UK NSC’s horizon scanning activities in practice

5.1 Example 1: expert input/request

Following approaches from experts in the field, an artificial intelligence (AI) task group was convened. Outputs from this group have so far included recommendations for developing test sets for studying the test accuracy of AI, recommendations for evaluation processes for the application of AI in screening, and guidance on evidence requirements for the application of AI in breast screening and diabetic eye screening specifically. UK NSC reviews summarising the current evidence of AI in breast screening and diabetic eye screening have also been produced, as well as a review of social and ethical issues.

5.2 Example 2: annual call for topics

A proposal was submitted in 2019, suggesting that all pregnant women, irrespective of risk status, should be screened using a point of care handheld ultrasound device (POCUS) to detect fetal presentation during routine antenatal appointments at around 36 weeks’ gestational age. An evidence map concluded there was insufficient direct evidence to recommend screening at the time. But the evidence found was promising enough for the topic to be added to the list of conditions that the UK NSC regularly reviews.

There was no direct evidence on the performance of POCUS devices compared to palpation (usual care). The UK NSC evidence team therefore put together a proposal for a HTA call for a diagnostic accuracy study on the use of POCUS to detect breech presentation at term. The application was successful.

5.3 Example 3: programme modifications

The UK NSC evidence team has carried out numerous programme modification requests, including from NHS screening programmes and through the annual call for topics. The NHS Sickle Cell and Thalassaemia (SCT) Screening Programme asked the committee to consider using cell-free DNA (cf-DNA) testing to identify sickle cell and thalassaemia during pregnancy. Although the evidence map found there was insufficient evidence to recommend the proposed modification at that time, it was highlighted that 2 UK-based research groups were collaborating on validating the test for sickle cell disease.

A workshop with expert stakeholders was held to address the problem and define the scope for proposed modelling work. A mathematical model aiming to determine the consequences and costs of introducing cf-DNA testing into the antenatal SCT programme at different levels of testing accuracy (sensitivity and specificity) was subsequently commissioned.

5.4 Example 4: early updates

TreatSMA (a UK charity that campaigns to improve the diagnostics, standard of care, social support, and access to treatments for spinal muscular atrophy (SMA)) requested an early update of the UK NSC’s SMA recommendation following the approval by the National Institute for Health and Care Excellence (NICE) of a treatment (Nursinersen) under the Managed Access Agreement. The UK NSC organised a workshop with a wide range of stakeholders to plan the next steps to review the case for screening for SMA. A proposal for a scoping exercise to inform a decision analytic model and cost effectiveness evaluation of newborn screening for SMA was drafted following the stakeholder workshop and is in the process of being commissioned.

6. Future developments

The revised UK NSC terms of reference state that the committee:

works with partners to ensure it keeps abreast of scientific developments in screening, including screening research, screening policy in other countries and emerging technologies with a particular focus on stratified screening.

The CMOs’ recommendations include the following wording:

Horizon scanning function of the committee will be key in order to keep up to date with national and international research in screening. To optimise horizon scanning and reduce fragmented and duplicate efforts, the CMOs recommend that an annual forum could be held with researchers to identify emerging research and technologies as well as activity which may not be on the radar of the committee.

The UK NSC secretariat will establish a consolidated horizon scanning system to help the committee take these recommendations forward.

7. References

1. Hines P, Hiu Yu L, Guy RH, and others. Scanning the horizon: a systematic literature review of methodologies. BMJ Open 2019;9:e026764 ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶ ↩⁷

2. Lauvrak V, Arentz-Hansen H, Di Bidino R, and others. Recommendations for Horizon Scanning, Topic Identification, Selection and Prioritisation for European Cooperation on Health Technology Assessment. EUnetHTA WP4 Deliverable 4.10, Oslo, 2020. ↩ ↩²