- Joined
- Dec 18, 2015
- Messages
- 3,216
- Reaction score
- 4,930
A human by the name of @wandering star came up with a neat way of looking at things yesterday estimating about 175 new starts per year per rad onc.
A while back I estimated 100 de novo patient new starts per rad onc each year. (It's very easy to prove this number.) Got criticism re: this number (too low) but some didn't understand these were the recently diagnosed "first line treats." (Even if palliative, and de novo diagnosed, would fit in this.) If the proportion of a service that's treating patients diagnosed >1y ago (palliatives, recurred, retreat etc) is 50%, this would mean ~200 new starts a year. If it's 40%, 167 new starts a year; if 33%, 150 new starts a year, etc.
A fellow named Graypeace did an even easier to cite/understand analysis that the number is 200 new starts per year.
Thus if we take the numbers (175, 200, 167, 150, 200) the mean is about 178 with a 99% CI of the mean of about 145-205.
Graypeace also showed that in 2003 rad oncs averaged about 350 new starts a year. If we do something (statistically questionable) and scale (175, 200, 167, 150, 200) about 350 we'd get 350, 400, 334, 300, 400.
[175, 200, 167, 150, 200] vs [350, 400, 334, 300, 400];
These two samples are different at p=0.01*
*Mann-Whitney
A while back I estimated 100 de novo patient new starts per rad onc each year. (It's very easy to prove this number.) Got criticism re: this number (too low) but some didn't understand these were the recently diagnosed "first line treats." (Even if palliative, and de novo diagnosed, would fit in this.) If the proportion of a service that's treating patients diagnosed >1y ago (palliatives, recurred, retreat etc) is 50%, this would mean ~200 new starts a year. If it's 40%, 167 new starts a year; if 33%, 150 new starts a year, etc.
A fellow named Graypeace did an even easier to cite/understand analysis that the number is 200 new starts per year.
Thus if we take the numbers (175, 200, 167, 150, 200) the mean is about 178 with a 99% CI of the mean of about 145-205.
Graypeace also showed that in 2003 rad oncs averaged about 350 new starts a year. If we do something (statistically questionable) and scale (175, 200, 167, 150, 200) about 350 we'd get 350, 400, 334, 300, 400.
[175, 200, 167, 150, 200] vs [350, 400, 334, 300, 400];
These two samples are different at p=0.01*
*Mann-Whitney