Do digital-first consultation models reduce or increase GP workload?

Problem

All general practices will be required to offer online and video consultations by April 2021, and the use of telephone consultations is also increasing. Some practices have introduced a digital-first model, whereby patients are expected to use these consultation modes before having a face-to-face consultation only when necessary. These developments are justified by claims that they will both increase access for patients and help GPs to manage their workload, which could allow them to give more time to patients with complex needs. But are these aims compatible or contradictory?

Approach

We conducted a modelling study using data from published studies of online, telephone or video consultations in general practice. Through a systematic literature review we identified studies providing quantitative data about use of digital-first access models for unselected patients requesting a general practice consultation for any problem. We sought estimates about the following variables: the proportion of consultations managed digitally, the proportion of digital consultations completed without a subsequent consultation, the proportion of subsequent consultations conducted by telephone rather than face-to-face, consultation duration, and changes in demand after introduction of a digital-first model. We used the data from the review as inputs to a dynamic model in Excel which enabled us to estimate the overall impact on GP workload. The model provided a base-case estimate using evidence from the published studies and graphically demonstrates how workload varies under different plausible scenarios.

Findings

Under most plausible scenarios, digital-first approaches to accessing primary care are likely to increase GPs workload. They could decrease workload if their duration is shorter, and a higher proportion of digital consultations are managed without a subsequent face-to-face consultation, than observed in most published studies. An important determinant of workload is whether digital-first approaches increase or reduce overall demand for GP, but there is little robust evidence about this question.

Consequences

Digital-first consultations to primary care are likely to increase GP workload unless stringent conditions are met, so how they are implemented is crucial. If they do increase workload they may still be justified, but this should be based on evidence about the benefits in relation to the costs, rather than assumptions about reductions in workload. Improvements in access from digital-first consultations could mean GPs devoting more of their time to younger and healthier patients, which could reduce the time available for older and sicker patients who need face-to-face care. Given this potential for increasing health inequalities, and increased workload at a time of GP shortages, these initiatives should be rolled out in a staged way alongside careful evaluation. In particular, we need long term studies from natural experiments on how changes in access to care are associated with changes in general practice workload.