FAQ for people in North London – which says this is an ‘independent’ study
see the evidence available for the app ;
“Our approach enables the reliable use of red-flags which leads to a high recall of 100% (all cases needing A&E will receive that outcome), at the cost of precision (some cases falsely receive A&E out- come), as shown in table 1..
This enables prediction with high precision at the cost of recall (potentially over-triaging for border-line cases).”
this is not a randomised controlled trial; the clinical vignettes are not included; there is no real life testing
one of the authors on this study is from Babylon
other research of note
Performance on appropriate triage advice across the 23 individual symptom checkers ranged from 33% (95% confidence interval 19% to 48%) to 78% (64% to 91%) of standardized patient evaluations.
“just over a third of users surveyed say they would have gone to primary and community services such as their GP if 111 hadn’t existed – but 111 actually sends around 60% of callers to these services
“There is potential that this type of service increases overall demand for urgent care.”