NHS patients who get referred to private hospitals for two very common surgeries enjoy better results, Newsnight analysis has shown.
Looking at knee and hip replacements, the only two procedures for which there is enough data, Newsnight found patients referred to private providers did better than patients of the same age, gender and prior health who were referred to NHS hospitals.
The analysis uses a dataset called "Proms", which stands for "patient reported outcome measures".
This is an unusual set of records which sets out to record how well patients feel. They respond to a detailed questionnaire about their health before an operation, then afterwards.
By comparing the two sets of results, we can benchmark - in detail - how well providers paid similar sums of money did in improving the lot of patients.
Proms also contains results for two other categories: groin hernia repair and varicose veins.
Analysis of both of these categories of care found that the private providers did better, too, but that the quantities of data are too small for the analysis to be reliable.
After conducting regression analyses of the survey data, we can demonstrate how this affects patients.
Take, for example, a hypothetical 70 year-old who reported that they had a low score on the general wellbeing measure on Proms - 50 out of 100 - because they need a hip replacement.
If they were to go to a public hospital they could expect a score after surgery of 78.9 out of 100.
At a private one, it would be 83.9 out of 100 - a gap of five points. For a similar patient seeking a knee replacement, the private-public gap would be around 3.4 points.
'Puzzle on provision'
In both cases, the results are highly statistically significant.
Since we have adjusted for gender, age and prior condition of the patient, we can say the difference is not explained by the slightly simpler set of cases that private providers tend to take on.
But remember that while, en masse, private providers may tend to do better than public ones, individual NHS providers may beat individual private ones.
And this is only a small part of the puzzle on provision inside the NHS: it covers just two operations for a single year (albeit common ones).
There is, furthermore, existing academic research that suggests that specialist providers doing a few elective treatments often do beat the pack. This may explain it.
Still: Watch this space.
As the amount of NHS provision that is outsourced has drifted up since the early 2000s, we will probably find other areas where new (and that might often mean private) providers tend to do better than legacy NHS hospitals.
Conversely, we will also confirm that there areas where patients report that traditional NHS providers are best. They have unique variety of services on-site.
If we can get more data which gives us good information on care quality we can, one hopes, work out which is which.