> William Stacy wrote:
> > Don W wrote:
> > > > > > Well, when a major drug company enlists symptomatic patients with an
> > > acuity of at least 20/320, (Snellen equivalent, of course), and then
> > > announces the result of a 6.6 letter gain for everyone, would you still
> > > call this all meaningless??
> > > > > > > > > > > > > > Yes I would. the 20/320 is a completely arbitrary number and doesn't
> > correspond to any acuity charts in common use.
> > > a 6.6 letter gain for EVERYONE is an absurdity beyond description.
> > > It's kind of like a drug company saying that people complaining of a
> > fever that measured less than 102.543 F. took their drug and every one
> > of them had a reduction of their fever by 0.249 deg. F.
> > > Make that "expected 6.6 letter gain".
> Don W.

> It would seems to me that it would not matter too much where a
> patient is in acuity but how well you could get him to 20/20 (with
> precision).

> Although slightly off topic, how do you refract someone that
> primarily has just peripheral vision?

> Don W wrote:
> > Well, when a major drug company enlists symptomatic patients with an
> > acuity of at least 20/320, (Snellen equivalent, of course), and then
> > announces the result of a 6.6 letter gain for everyone, would you still
> > call this all meaningless??
> > > > > Yes I would. the 20/320 is a completely arbitrary number and doesn't
> correspond to any acuity charts in common use.
> a 6.6 letter gain for EVERYONE is an absurdity beyond description.
> It's kind of like a drug company saying that people complaining of a
> fever that measured less than 102.543 F. took their drug and every one
> of them had a reduction of their fever by 0.249 deg. F.

> Well, when a major drug company enlists symptomatic patients with an
> acuity of at least 20/320, (Snellen equivalent, of course), and then
> announces the result of a 6.6 letter gain for everyone, would you still
> call this all meaningless??
Yes I would. the 20/320 is a completely arbitrary number and doesn't

> Are you an engineer?
> --

> You are correct, the chart jumps from 20/200 to 20/100.
> But, you can always make your own using a simple ratio:
> 0.375 inches / 20 feet = x / distance
> or, for 20/40 letters:
> Letter Size = 0.375 ( 40 / 20 )
> Letter Size = 0.75 inches.
> For 20/180
> Letter Size = 0.375 ( 180 / 20 )
> Letter Size = 3.36 inches.
> Best,
> Otis

> "Don W" <dwilgus[at]prodigy.net> wrote
> > Well, when a major drug company enlists symptomatic patients with an
> > acuity of at least 20/320, (Snellen equivalent, of course), and then
> > announces the result of a 6.6 letter gain for everyone, would you still
> > call this all meaningless??
> Stacy and I use it with individual patients and decimal/letter precision is
> absurd in our setting.
> But drug companies always choose units and comparisons that paint their
> results in the best possible light. "6.6 letters" sounds much better than
> "about one line."
> -MT

> To me, it is unfortunate that the established scales differences (the
> steps) get larger as the acuity declines. For example, a person with
> the acuity of 20/200 would like very much to know if they have achieved
> the improvement of 20/180 instead of waiting to the next logMAR value
> of 20/160. If one is receiving treatments for the 20/200, any
> improvement is most welcome instead of the next step that fits the log
> scale best. Like would you accept a fever thermometer that has the
> high end in a log scale?
> Don W.

> Well, when a major drug company enlists symptomatic patients with an
> acuity of at least 20/320, (Snellen equivalent, of course), and then
> announces the result of a 6.6 letter gain for everyone, would you still
> call this all meaningless??

> In fact, the Snellen numbers become increasingly meaningless as you go
> beyond 20/100. Unlike a thermometer, where the accuracy is pretty level
> across the entire scale. I suppose once you exceed 1000 degrees, a
> similar problem with accuracy also occurs, where such a log scale might
> be appropriate as well.
> w.stacy, o.d.