Have you spoken to Cheryl and clarified what you have said?
They spent a loooong time IMO processing data to get any possible positives and went to smallest possible subjects to get positives. You said "Don’t you think it strange that viraleze would have a greater effect for older less healthy individuals. That doesn’t make sense to me" No -it makes some sense- Assuming the older age group has a lesser immune response relatively use of viraleze by those participants should show a greater response over placebo . In a younger age cohort the simple flushing of nasal cavity by placebo and better immune response relatively and having lived through multiple covid reinfections so lesser reporting of symptoms relatively may make viraleze not look like it is outperforming to the required level
Well unless I am reading what you wrote incorrectly infact Viraleze should have a greater affect " relatively" in the older age groups than in younger age groups. As you state the study design and measurement point is possibly part of the issue with data . There is a couple of factors in this in regards to what age groups in UK were offered extra vaccine doses and also how often the older cohort were catching covid and gaining natural resistance. Younger cohort has greater immune response , also possibly in UK has been repeatably infected and has not has boosters as not required ( UK govt was only doing for at risk and pensioner age) Your theory is they cleared virus before measurement or didn't complete study and mine is that different strain of Covid that had higher viral load earlier and different time to symptom onsett that of course is noticed earlier / later by different cohorts. You also have to take this into account o understand how and where they were recruiting and who participated knowing who anticipated. I would be interested in knowing if slow recruitment was a particular age group - younger aged cohort for example as unless they were part of family group how / why would they get involved in study? The late recruitment of younger cohort could account for the ignorance of investigator not noticing low response of cohort as it happened late in study as they were prioritising only younger cohort. laggards , non completion at later stage to fill complete spread of cohorts across age groups and then close study rather than adding extra numbers etc etc.
Bottom line is that stats measure outright difference and relative difference and SPL even after reprocessing data down to minimum age cohorts couldn't get the numbers any lower than 45 for significance. They knew study design and knew for outset that if they had to break down to 5yr cohorts with only 8-12 participants per subgroup ( and of those 50% allocated to each arm) it doesn't take many participants to stuff it up ( as each participant in that smaller subgroup makes up a significant portion of that subgroup) . Looking at it the other way if you split study cohorts by age under and over 45 the number required in the next 5 year age cohort to cause failure is basically all of them between 40-45- maybe there was only 4 respondents in that age group and they were all on only 1 arm etc etc so couldn't drag age lower
Worse than that if SPL and investigator were so confident they may not have from the outset evenly allocated arms of study by agegroup making 5year subgroup analysis get pinched / crunched as not enough participants from each arm in each 5 year subgroup below 45. I tried multiple reverse engineering of data provide using assumptions like even spread of age and 50% spread in each arm per age and as I have noted graph would have to look ugly below 45 to end up with result .
What a shame that they didn't extend study to more participants in younger age cohorts if they noticed younger laggards on both arms etc etc as cost would have been minimal but I can only assume they didn't think they would have to reprocess and reprocess down to 5 year age cohorts to and end up with result they got
SPL Price at posting:
12.5¢ Sentiment: None Disclosure: Not Held
(20min delay)