Outcomes
What we would publish, if we had it. And what we have published instead, in advance.
The position
We have none to publish yet
The programme has not run long enough to produce results worth reporting. Rather than show you published research about a category of treatment and let it read as our own record, this page is empty of numbers on purpose.
If you find a weight-loss service in India quoting a success rate, it is worth asking who was counted and who was left out.
Methodology, published first
How we will report them
Published now, before there are any numbers. A methodology that appears alongside a favourable result is worth nothing, because nobody can tell whether it was chosen to fit.
- Everyone who started is counted
- Including people who stopped, people a doctor took off treatment, and people we lost contact with. A figure computed only over people still enrolled measures who stayed, not what worked.
- No single headline number
- An average across everyone hides the groups doing worse. We will report outcomes broken down, or not at all.
- Groups reported separately
- By sex, by whether someone is in a metro, and by the language they were served in. If one group does worse, that will be visible rather than averaged away.
- Groups under 20 people are not reported
- Not because the number is inconvenient, but because a handful of people is not a statistic and is identifying. Where that happens we will say a group was suppressed, not silently omit it.
- The denominator is stated
- Every figure will say how many people it is computed over and how many started. A percentage without a denominator is a decoration.
- Weight lost is not the only measure
- Rapid loss is a risk, not a success — above about 1% of body weight a week, what is going is disproportionately muscle. Any outcome report will say what happened to lean mass and to the things weight is a proxy for.
Related