The to and fro on the virtues of policy focus on ‘remote’ versus ‘face-to-face’ GP appointments and the controversies around how to ‘incentivise’ GP to offer more face-to-face appointments made me think about the ‘default’ assumption that digital health is an ‘obvious’ choice and that it will make life easier for doctors and nurses.
Research shows otherwise.
For example, a study conducted using estimates from a systematic review of the literature between 2000 until 2019 to populate a model estimated the impact of various digital first consultation models on GP workload. The study showed that online, telephone, or video consultations are likely to (apart from in some very specific cases with minimal increased demand and where 50% of requests are resolved within one contact) increase GP workload by 25%, 3%, and 31%.
It was highlighted that patient messaging in the EPIC EHR increased by 150% during Covid-19. It was also interesting to note anecdotes that the complexity of the messages have increased (see reply on Twitter by Dr Walker below), meaning that the effort required to respond and resolve these requests will be higher.
There are many things to consider to get this right.
Assuming a design centric approach has been taken when developing the solution to ensure the core needs of the different user groups are taken into account, there are a further 3 key areas worth considering (they are all linked!):
Is the ‘right’ healthcare professional engaging with the patient or related activity / task? Does a doctor need to, or will a nurse, pharmacist or a health coach be a better option?
A recent study demonstrated that access to 24/7 digital first primary care (GP at Hand) had lower acute hospital care costs. Yes, the study was done by folk from Babylon and we can debate the robustness of the methodology, but there is something about thinking about which patients or which scenarios to focus digital health on (we need to be very careful about widening inequality, but digital health makes more sense in certain situations and not in others). In some cases, a ‘hybrid’ approach of digital and non-digital approaches works best.
Selected activities and interactions can be automated where appropriate – for example, chat bots or other ‘self-service’ options could reduce the burden on healthcare professionals (i.e. changing of appointments, basic information about a procedure such as vaccinations).
The other element is related to the previous targeting of the ‘right’ healthcare professional. If the doctor still has to engage with the patient or the activity before asking another healthcare professional to help, it’s not the best use of time. Where possible, the triage must happen automatically based on certain criteria to ensure safety. Taking this concept further and thinking of predictive capabilities, earlier interventions but different healthcare professionals (i.e. a health coach or a specialist nurse) may prevent worsening of a condition and the patient ultimately needing a higher level of care.
Another element of the ‘triage’ is the response time frames – does the healthcare professional need to respond within 1 hour or is safe to leave for 24-48 hours? Can the responses be done in ‘batches’. These are some elements which can be automated.
Digital solutions where a larger number of users can access services, more frequently and for longer periods during the day will, in some cases, inevitably lead to increased demand on healthcare professionals. As above, the most effective way to meet this demand will not simply be more doctors and nurses (if we can find them hidden somewhere) but will also include other healthcare professionals. There is also a need to develop the skills of all healthcare professionals to deliver health services digitally in a safe and efficient manner. All of this needs funding and business cases must provision for this.
Finally, I think we need to rehaul the way we pay for digital health solutions. Outcome based incentives (with upfront incentives to balance risk-benefits for innovators) is the way to go.
Very happy to hear your comments below or feel free to email me to share ideas – firstname.lastname@example.org