|
Home > Archive > Vision > October 2006 > Acuity readings plus/minus a few letters.
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
Acuity readings plus/minus a few letters.
|
|
|
| If one is having an Snellen test, and the reading comes out 20/100 minus
1, that means one letter has been lost in reading the 20/100 line. Truth?
And how is that reading should be converted to 20/? . Where ? is a
singular number. If so, how do we interpolate to the actual reading??
Now, if the reading is 20/100 plus one. Just exactly what does that mean?
Does that mean one is reading the 20/100 line totally and one more letter of
the next line? And does "plus" have any reasonable meaning since the letter
height of the next line might vary?
Don W.
| |
| William Stacy 2006-10-11, 9:29 pm |
|
Don W wrote:
> If one is having an Snellen test, and the reading comes out 20/100 minus
>1, that means one letter has been lost in reading the 20/100 line. Truth?
>
>
Approximately true. It can also indicate 20/100 but possibly with some
guesswork involved, or a very blurred 20/100.
There is no rigid standard in acuity since it it such a subjective and
necessarily variable measurement.
>And how is that reading should be converted to 20/? . Where ? is a
>singular number. If so, how do we interpolate to the actual reading??
>
>
You could convert 20/100- to 20/150 and nobody would get too excited,
since most snellen charts have 20/100 letters but many jump from there
20/200.
> Now, if the reading is 20/100 plus one. Just exactly what does that mean?
>Does that mean one is reading the 20/100 line totally and one more letter of
>the next line?
>
Yes. Usually, it means just one letter was seen on the next line. If
they read 2 letters on the 20/80 line, it would normally be written
20/20 ++ or 20/20 +2
>And does "plus" have any reasonable meaning since the letter
>height of the next line might vary?
>
>
>
See above. All Snellen acuity has reasonable but not necessarily
precise meaning.
| |
|
|
Don W wrote:
> If one is having an Snellen test, and the reading comes out 20/100 minus
> 1, that means one letter has been lost in reading the 20/100 line. Truth?
> And how is that reading should be converted to 20/? . Where ? is a
> singular number. If so, how do we interpolate to the actual reading??
>
> Now, if the reading is 20/100 plus one. Just exactly what does that mean?
> Does that mean one is reading the 20/100 line totally and one more letter of
> the next line? And does "plus" have any reasonable meaning since the letter
> height of the next line might vary?
>
> Don W.
Your a little better or worse than 20/100. Perhaps 20/80 if better,
perhaps 20/120 or 20/150 if worse. There needs to be at least the
20/150 line between 20/100 and 20/200, but many eyecharts dont! 
| |
|
| >>
> You could convert 20/100- to 20/150 and nobody would get too excited,
> since most snellen charts have 20/100 letters but many jump from there
> 20/200.
>
So would people get excited if vision went from 20/100 to 20/200?
In a lot of the clinical trials they are talking about the gain or loss of
so many _letters_ , not lines (e.g., Lucentis). Is that not cutting it
rather thin for "subjective"?
Don W.
| |
| serebel 2006-10-11, 9:29 pm |
|
Don W wrote:
>
> So would people get excited if vision went from 20/100 to 20/200?
>
> In a lot of the clinical trials they are talking about the gain or loss of
> so many _letters_ , not lines (e.g., Lucentis). Is that not cutting it
> rather thin for "subjective"?
>
> Don W.
Don, you're answering a retard who has no clue as to what he's talking
about.
| |
| otisbrown@pa.net 2006-10-11, 9:29 pm |
|
> Your a little better or worse than 20/100. Perhaps 20/80 if better,
> perhaps 20/120 or 20/150 if worse. There needs to be at least the
> 20/150 line between 20/100 and 20/200, but many eyecharts dont!
Otis> Or, just cut the distance in 1/2. Thus the 20/50 (size) becomes
20/100, 20/60 (size) is 20/120, etc.
Otis
Ace wrote:
> Don W wrote:
>
>
> Your a little better or worse than 20/100. Perhaps 20/80 if better,
> perhaps 20/120 or 20/150 if worse. There needs to be at least the
> 20/150 line between 20/100 and 20/200, but many eyecharts dont!
| |
|
|
otisbrown@pa.net wrote:
>
> Otis> Or, just cut the distance in 1/2. Thus the 20/50 (size) becomes
> 20/100, 20/60 (size) is 20/120, etc.
That wont work, the chart has to be read at optical infinity. At 10
feet there is 1/3 diopter accomodation. I can see more than twice as
well from 10 feet vs. 20 feet because the 1/3 diopter accomodation
"cancels" out -.33 diopters of myopia.
| |
| Anon E. Muss 2006-10-12, 2:31 am |
| On Wed, 11 Oct 2006 19:13:56 GMT, "Don W" <dwilgus@prodigy.net> wrote:
> If one is having an Snellen test, and the reading comes out 20/100 minus
>1, that means one letter has been lost in reading the 20/100 line. Truth?
>And how is that reading should be converted to 20/? . Where ? is a
>singular number. If so, how do we interpolate to the actual reading??
>
> Now, if the reading is 20/100 plus one. Just exactly what does that mean?
>Does that mean one is reading the 20/100 line totally and one more letter of
>the next line? And does "plus" have any reasonable meaning since the letter
>height of the next line might vary?
Some doctors don't like the minus/plus notation and refuse to use it
because it is ambiguous -- e.g., 20/40 +2/6 and 20/30 -4/6 are the
same?
They use the scoring system/chart used in the Early Treatment of
Diabetic Retinopathy Study (ETDRS) to eliminate this problem with the
Snellen and Sloan visual acuity.
| |
| Dr Judy 2006-10-12, 4:29 pm |
| Don W wrote:
> If one is having an Snellen test, and the reading comes out 20/100 minus
> 1, that means one letter has been lost in reading the 20/100 line. Truth?
Yes, the meaning is that one missed one letter of the 20/100 line. Not
everyone uses this notation, to be perfectly correct it should also
specify how many letters were in the 20/100 line, ie 20/100 -1/2 or
20/100 -1/5
> And how is that reading should be converted to 20/? . Where ? is a
> singular number. If so, how do we interpolate to the actual reading??
There is no need to "convert to 20/ ". The acuity stands as stated:
20/100 -1/2. You can't interpolate to say 20/120 or 20/150 if that is
what you were thinking. You need to use a chart with 20/120 line or
20/150 line to do that.
>
> Now, if the reading is 20/100 plus one. Just exactly what does that mean?
> Does that mean one is reading the 20/100 line totally and one more letter of
> the next line?
Yes.
And does "plus" have any reasonable meaning since the letter
> height of the next line might vary?
Again, to be correct one should specify the next line: 20/100 +1 20/80.
Some would define 20/100 as reading at least half the letters in the
line, so 20/100 -1 counts as 20/100 and 20/100 +1 is the same as 20/80
-3/4 so also counts as 20/100. Adding the + and - letters t is a way
to try and make a distinction between those two cases, but many do not
record that way due to the issues you have raised.
Snellen acuity at the higher end is a rough and ready approximation.
Useful clinically to monitor change if the same chart, same examiner
and same test lighting is used but not used much in low vision work or
for research purposes.
Better acuity measuring systems including LOGMAR, graded contrast
sensitivity charts and charts without alpha numberic characters are
available if you want precision.
Judy
>
> Don W.
| |
|
| Thanks to Dr. Judy (and others) for clearing up this confusion. I think
in going to the plus side, specifying the number of letters on that next
line (and the size) nails it down completely.
There is one other area in acuity reporting that is bothersome. I am
reading the results of the Lucentis trials. And therein is claimed a gain
of so many letters after so much usage. But they never specify in going
from what line to what other line to get the gain 6 letters. The gain of 6
letters is, of course, an average. (and maybe Snellen or not). But this
seems to be an extremely confusing way to state improvement, statistically.
Don W.
| |
| Mike Tyner 2006-10-13, 4:27 pm |
|
"Don W" <dwilgus@prodigy.net> wrote
> letters is, of course, an average. (and maybe Snellen or not). But this
> seems to be an extremely confusing way to state improvement,
> statistically.
Snellen charts are so simple and universal that they can't be tossed aside.
The interpolation of partial Snellen lines is mostly common sense -
20/20 -3/6 means three out of six. Accurate enough, because next time, with
new letters, it might be -1/6.
-MT
| |
|
| >
> Snellen charts are so simple and universal that they can't be tossed
> aside. The interpolation of partial Snellen lines is mostly common sense -
> 20/20 -3/6 means three out of six. Accurate enough, because next time,
> with new letters, it might be -1/6.
>
> -MT
>
OK, agreed as to the variability of results.
Let me see if I can clear up how I am confused. When they publish the
curves for the advantages of Lucentis, they are published as to the gain of
the number of letters as a function of time (monthly). These gains (I
presume), can be from 20/200, or from 20/100 or whatever. So gains are
really some kind of differential from some line in the Snellen chart (if
they used that) to some other line. And then these gains are then averaged
to show the average gain for that month. It seems they are averaging apples
and oranges. If someone is at 20/200 and gains 6 letters in the Snellen
chart, that is much more dramatic than someone who is 20/50 and then gains
6. At least this is what it seems like.
Don W.
| |
| Dr Judy 2006-10-14, 2:31 am |
|
Don W wrote:
> OK, agreed as to the variability of results.
>
> Let me see if I can clear up how I am confused. When they publish the
> curves for the advantages of Lucentis, they are published as to the gain of
> the number of letters as a function of time (monthly). These gains (I
> presume), can be from 20/200, or from 20/100 or whatever. So gains are
> really some kind of differential from some line in the Snellen chart (if
> they used that) to some other line. And then these gains are then averaged
> to show the average gain for that month. It seems they are averaging apples
> and oranges. If someone is at 20/200 and gains 6 letters in the Snellen
> chart, that is much more dramatic than someone who is 20/50 and then gains
> 6. At least this is what it seems like.
I'm surprised that these studies don't use LOGMAR charts or vernier
acuity charts.
JUdy
>
> Don W.
| |
| Mike Tyner 2006-10-14, 2:31 am |
|
"Don W" <dwilgus@prodigy.net> wrote
> and oranges. If someone is at 20/200 and gains 6 letters in the Snellen
> chart, that is much more dramatic than someone who is 20/50 and then gains
> 6. At least this is what it seems like.
It does, and it does seem strange. I see the inconsistency you're
describing.
-MT
| |
| William Stacy 2006-10-14, 4:28 pm |
| Don W wrote:
If someone is at 20/200 and gains 6 letters in the Snellen
> chart, that is much more dramatic than someone who is 20/50 and then gains
> 6. At least this is what it seems like.
Your example isn't clear. That is, snellen charts go from 20/200 to
20/100 or 20/150, and typically the 20/100 or 20/150 size letters have
only 4 letters or so on the chart. "gains 6 letters", without
specifying WHICH size letters is kind of meaningless, unless one has a
random letter generating chart and you really mean something like "gains
6/12 of 20/100" or 6 out of 12 presentations of that size.
Certainly 20/200 to 20/100 sounds more dramatic than 20/50 to 20/40, but
in the real world of seeing, the latter may be more significant than the
former.
w.stacy, o.d.
|
| |
|
|