5

Yet another surprising study on things that make a difference to the risk of covid has appeared.

This one claims, to quote the report in LiveScience:

People who wear eyeglasses may be at lower risk for catching COVID-19 than those who don't wear glasses, early research from China suggests.

The study researchers analyzed information from 276 patients at a hospital in China's Hubei province and found that only about 6% said they wore glasses for more than 8 hours a day, all of whom had myopia, or nearsightedness. That's much lower than the estimated rate of myopia in Hubei from previous research, which was 31.5%.

I'm skeptical as I am about many studies on things that affect covid. This is a small study and, even in this report, looks like it has problems.

Is there any credibility to the study? Does wearing glasses measurably protect from covid?

matt_black
  • 56,186
  • 16
  • 175
  • 373
  • Since you read the article... Does it ever make it clear if "31.5%" is for people wearing glasses "more than 8 hours a day" ? So far it sounds like just "only 1/5 of people with myopia wear glasses 8+ hours a day"... (Also it sound plausible there would be some correlation with wearing eyeglasses as touching face/eyes when one has glasses is harder) – Alexei Levenkov Sep 22 '20 at 05:43
  • 1
    @AlexeiLevenkov - confounding factor: wearing glasses and a mask leads to more frequent face contact while trying to re-find magic position that doesn't fog too much, an effect accentuated in 'flu season (entering warmer & damper atmospheres) – Tom Goodfellow Sep 22 '20 at 06:14
  • The study says - "may" so not giving any certainty. The study itself says it has a small sample size. So I think the study is creditable but the study does not claim that glasses measurably protect from covid – mmmmmm Sep 22 '20 at 15:57
  • @TomGoodfellow the reddit comments gave the opposite view, glass wearers touch their face less as one you have the mask set up you are less inclined to change its position - neither is covered in the study – mmmmmm Sep 22 '20 at 15:59
  • Correlation is not causation. On a different note, glasses are like small face shields. – Charlie Crown Sep 23 '20 at 01:03
  • Both myopia and hositalization rate due to coronavirus also correlate with age - have age effects been taken into consideration by the study? – Hulk Sep 25 '20 at 10:52
  • Your quote doesn't make any claim about causality yet the rest of your post does. Either find a different quote or change your post to match the claim. There might be a correlation because people who wear glases are more often nerds without much social contacts. If that correlation exists it's useful knowledge as it can be used to make decisions about where to sit in public transport. – Christian Sep 25 '20 at 13:07
  • @Christian That depends how you interpret casual language. To me the quote I already used does claim the possibility of causality. And that is how most readers will interpret it. *Why* the relationship exists and alternative possible causes is for answers to deal with. – matt_black Sep 25 '20 at 13:15
  • @matt_black : It makes a statement about the people that wear eyeglasses. It doesn't make any statement about whether or not that's causal. – Christian Sep 25 '20 at 13:17
  • @Christian "may be at lower risk" is not the same as "may–possibly for some other reason associated with spectacle wearing–be at lower risk". If we reject all claims in headlines with vague claims of association or causality, there would be few questions on this site. Again, this is something worth pointing out in an *answer* not a good reason to reject the question. Also, the paper is rubbish for much more important reasons than this. – matt_black Sep 25 '20 at 16:21
  • I don't see that other questions require reading someone as making claims that they didn't didn't explicitely make. If the correlation exists then a study that claims the correlation can be credible and useful for day to day actions even when there's no causation. – Christian Sep 25 '20 at 19:17
  • @TomGoodfellow That's not a confounder. From wikipedia: "In statistics, a confounder is a variable that influences both the dependent variable and independent variable, causing a spurious association." Wearing glasses is the independent variable. Getting COVID is the dependent. A confounder is something that influences both. I.e. it would have to cause the frequency of wearing glasses to change. – Acccumulation Sep 26 '20 at 08:03
  • @Acccumulation - so that which confounds interpreting the observations is not necessarily a confounder. Confound it all! - and thanks to you for the education. – Tom Goodfellow Sep 26 '20 at 22:15
  • @Acccumulation So a confounder would be for example an illness that makes your eyesight bad but also protects you from Covid. – gnasher729 Sep 30 '20 at 00:10
  • I kind of think glasses would decrease covid. Doctors, where protective glasses to avoid splatter going into there eyes. I have felt split hit me a number of times talking to people. Also. some amount of small water vapor is probably hitting your eyes all the time. I know glasses are not PPE, but they block a certain about of splatter, spit, and water vapor. – GC_ Nov 19 '20 at 16:46

0 Answers0