NICE's medical technologies advisory committee considered evidence on digital technologies to manage non-specific low back pain (LBP) in people 16 years and over from several sources. These included an early value assessment report by the external assessment group (EAG), an overview of that report and an addendum to the report by the EAG. Full details are in the project documents for this guidance on the NICE website.
Provision of services for musculoskeletal-related pain varies across the NHS. Most non-specific LBP is managed in primary or community care settings, which have limited workforce capacity and resources to meet the growing demand for services. The clinical experts noted that there is often a long waiting list for referral to specialist services. They added that people are likely to try to resolve pain on their own before seeing a GP.
Digital technologies for managing non-specific LBP that are suitable for self-referral would provide people with the resource they need to self-manage. One clinical expert noted that there was no evidence about people who self-referred. But, in practice, healthcare professionals will want to use digital technologies to ease the strain on resource use. The committee noted the importance of safeguards. It advised that technologies suitable for self-referral need appropriate safeguards to accurately identify and escalate potential red flags.
Some of the technologies included in this assessment are used in the NHS. The committee stated that technologies designed to interact with GP systems ensure continuity of care by making important information accessible to healthcare professionals that need it. The companies for ACT for PAIN, getUBetter, Hinge Health and Pathway through Pain confirmed that their technologies are designed to be, or will be, able to interact with existing NHS systems.
Digital technologies can provide quicker access and increase management options for people with non-specific LBP. The patient experts said that quicker access to support can lead to faster improvement in symptoms, and can reduce the possibility of acute pain developing to chronic pain. They noted that following a personalised exercise and movement plan could improve mobility, ability to manage pain and mood, and give a sense of control over the condition.
The patient experts said that reassurance is needed that personal data will be secure. There also needs to be appropriate measures in place for reporting adverse events related to using the technologies. The patient experts noted that people less comfortable or skilled at using digital technologies, or unable to read or understand health-related information (including people who cannot read English) need considering. Appropriate alternative support should be provided for them.
The patient experts advised that patient choice should be a key consideration. They added that people should have the option to remain on a waiting list for a face-to-face appointment if they agree to engage with digital technologies. The committee noted that, for some of the technologies, people will be in the care of company-employed healthcare professionals. In these instances, it will be important that there is a referral pathway back into NHS care when the technology is no longer suitable.
The committee concluded that patient choice and preferences should be taken into consideration when deciding the suitability of digital technologies for managing non-specific LBP. It also noted the importance of codesigning digital technologies with people with LBP. This is to ensure that the content and management options are appropriate and relevant for the users. The companies for ACT for PAIN, getUBetter, Hinge Health and Pathway through Pain said that they have involved people with LBP and healthcare professionals in the development phase of their technologies.
The EAG prioritised 17 studies for assessment:
5 randomised controlled trials (RCTs)
1 prospective single-arm trial
1 prospective cohort study (providing non-comparative data)
3 retrospective cohort studies (providing non-comparative data)
7 retrospective case series.
Only 5 of the studies were done in the UK and 2 other studies included UK participants. The EAG noted that there was considerable uncertainty about the generalisability of the evidence to the UK NHS setting. It stated that there was limited clinical and economic evidence for acceptance and commitment (ACT) therapy in the UK, and no studies using ACT for PAIN were identified. The EAG also stated that there was economic evidence for Pathway through Pain, but no clinical evidence was identified. The committee noted that there was no evidence of harm or safety concerns, and that access to psychological therapies for chronic LBP is limited. So, it recommended using getUBetter, Hinge Health, Kaia, Pathway through Pain and SelfBack in the NHS while more evidence is being generated. For the other technologies (Ascenti Reach, Digital Therapist, Flok Health, Phio Engage and Joint Academy), there was no or limited evidence so the committee recommended their use only in research.
The evidence from the prioritised studies reported on 47 different outcomes, including function, pain self-efficacy, intervention adherence and adverse events. The EAG suggested the evidence showed that, when compared with standard care alone, digital technologies used with standard care may be effective in terms of improving pain and physical function outcomes. But the range of outcome measures used across the trials made it difficult to compare the digital technologies. Evidence was also limited to short-term effect, with no comparative data for outcomes beyond 3 months. The committee noted that there was no evidence to suggest that any technologies were unsafe. It advised that standardised data be collected for future evaluation.
The committee noted that clinical evidence showed variability in the way adherence was measured and the reported levels of engagement. The clinical experts stated that reported adherence levels were similar to that seen in clinical practice. The committee was informed that, because of the recurrent nature of LBP, people might stop using a technology when their symptoms improve but use it again if symptoms return. One patient expert said that using digital technologies might be fairly new to people, and that they might experience some challenges. The committee acknowledged that significant effort is needed from people with LBP to complete exercise programmes. It said that the companies should ensure that nudging features are in place to prompt people to engage with the technologies. The committee concluded that more evidence is needed on short-term (30 days) and long-term (6 to 12 months) adherence rates.
During consultation, information was provided for 4 new technologies (Digital Therapist, Flok Health, Joint Academy and Physitrack) to be considered as part of the assessment. Also, further evidence was submitted for Phio Engage and SelfBack. The committee concluded that the previous evidence provided for SelfBack was enough to support the recommendation for use while further evidence is being generated. Two retrospective case series and 1 retrospective cohort study were considered as part of the assessment for Phio Engage. The committee concluded that the information was limited and there was inadequate evidence to recommend Phio Engage for use in the NHS while further evidence is generated. It suggested that further research be done through company, research or non-core NHS funding. One retrospective case series study for Joint Academy was considered relevant for the assessment by the EAG. There was no clinical evidence for Digital Therapist and Flok Health. The committee concluded that further evidence is needed for these 3 technologies.
The committee did not make a recommendation for Physitrack. It considered the technology to be out of scope for inclusion in this assessment. The clinical experts noted that Physitrack is an exercise prescription tool used in practice as a digital information leaflet, supporting the company's claim that it is not a medical device.
Digital technologies for managing non-specific LBP may not be suitable for everyone, including people who:
have limited access to devices or an internet connection
are less comfortable or skilled at using digital technologies
are unable to read or understand health-related information (including people who cannot read English)
have a visual impairment
have problems with manual dexterity.
The committee concluded that face-to-face treatment options should be available when digital technologies are not suitable, and that companies should consider providing translations.
Early economic modelling using a simple cost-utility model suggested that digital technologies for managing non-specific LBP may be cost effective when used alongside standard care. Base-case results showed the technologies alongside standard care were cost saving by an estimated £84 per person, with a quality-adjusted life year (QALY) gain of 0.01 compared with standard care alone. This was using a threshold of £20,000 per QALY gained. The analyses were done using an NHS and personal social services perspective. The base-case results were supported by sensitivity and scenario analyses. The economic model used a 1‑year time horizon because of uncertainty about the long-term treatment benefits and the risk of pain relapses, particularly for people with chronic LBP.
The EAG acknowledged that, because of limited evidence, the model did not have a specific placement in the clinical pathway and that different placement may lead to different reported outcomes. The model included costs of the technologies, healthcare professional time, other health services use and medication use. The details of the assumptions used in the model are outlined in table 8.2 of the assessment report on the NICE website. The EAG noted that the main drivers of the model were:
the cost of the technology
incremental utility
proportion of people engaged with the technology
reduction in physiotherapy referral
number of physiotherapy appointments after referral.
The EAG excluded training and implementation costs from the model because of uncertainty in the level of resource use needed.
The committee noted that the evidence informing the cost model was limited but that there was plausibility of cost effectiveness. It concluded that it was appropriate to recommend some technologies for conditional use within the NHS while more evidence is being generated.
No clinical or economic evidence was identified by the EAG for Ascenti Reach, ACT for PAIN, Digital Therapist or Flok Health. For the remaining technologies, evidence gaps were identified in population demographics, clinical effectiveness, treatment adherence and healthcare resource use. The committee concluded that there is potential benefit and some evidence to support recommending 5 of the digital technologies for managing non-specific LBP in the NHS with evidence generation, once appropriate regulatory approval is in place. The key evidence gaps were:
Population: the EAG noted that some clinical studies were excluded because they had an unspecified population. More evidence generation should clearly report population characteristics, particularly type of LBP (chronic or acute, specific or non-specific) and pain severity at baseline. This will ensure that data collected captures people with sciatica and back-related leg pain, so that this data can be easily extracted and excluded from the analysis.
Outcomes: comparative evidence identified a wide range of key outcome measures used across the trials, making comparison of the digital technologies difficult. Also, published evidence was not available for most of the outcomes in the scope of this evaluation. Evidence generation should include using consistent measures for key outcomes, such as pain score and health-related quality of life (Musculoskeletal Health Questionnaire and EQ‑5D‑5L) to enable comparison of different technologies.
Adherence: studies that did report adherence varied in the definition of adherence and in the methods used to measure it, making comparisons difficult. More evidence generation should report consistent measures of adherence and reasons for stopping should be recorded.
Referral: evidence is lacking on the referral setting (referred or self-referred) and the place of the digital technologies in the clinical pathway. It is unclear whether this affects the effectiveness of the digital technologies. There is scarce evidence on the effect of digital technologies on referral rates for other services such as imaging, physiotherapy, surgery or emergency department attendances.
Resource use: more evidence generation is needed on healthcare resource use, including training and implementation associated with different types of digital technologies, especially those providing psychological treatment. Additional information on resource use is included in the evidence generation plan.