How often do/should we test?

[ElectLive]

Each of us probably would interpret the graphic a little differently. I definitely think the individual patient factor will always come into play. Some people will do very well at any testing frequency, some people will do poorly at any testing frequency.

So, as it sounds like you're doing, I would probably ignore then all those outlying dots, and focus more on that cluster in the middle. If you look at the cluster, at the 30 day mark, range is therapeutic anywhere between 35% or so and 90% or so. At the 1 week mark of the cluster, range is therapeutic anywhere between 45% and 70%, so much more narrow overall and closer to the mean. What this suggests, to me, is that the hypothesis of this thread may just be right, that weekly testing gives a more defined, or perhaps "less fuzzy", picture of what overall INR control actually is than 30 day testing. Again, my interpretation. Of course, it actually kind of looks a little like a Rorschach test too, though, so maybe we'll all read something a little different into it!

More importantly, something this graph does not show is the total range, it only shows the goal of INR 2-3. Much more revealing, and a little more reassuring, is another set of data in this same study: "The overall mean percent of reported INR measurements greater than 4 was ~2%; the overall mean percent of INR measurements <2.0 was ~22 to 23% and < 1.5 was ~5%. Compared to later months, during the first month of therapy, a greater percentage of INR measurements were greater than 4 (~5 vs. ~2%), less than 2 (32% vs. 23-24%) or less than 1.5 (~11% vs. ~5%)" Also, the report that accompanied this graphic mentions: "There did appear to be a numerical increase in the number of events (deaths and strokes) in subjects undergoing the most frequent monitoring, possibly representing the subset of subjects with more difficult to control or variable INRs."

These again were not patients intentionally experimenting with testing frequency to improve control. They were patients following standard medical guidance at different facilities and likely only adjusting frequency of testing in response to events (being out of range, surgical interventions, etc). So the weekly testing I'm guessing was more representative of difficult periods of INR control.

 

[Protimenow]

Where weekly testing can be better is a) determining earlier that there were periods that were out of range, or b) showing out of range periods that resolve before monthly tests are taken. There's a c) here, too -- a person who has a stroke or embolic event before the NEXT montly test is taken may drop out of the subject pool. If EVERYONE tested once monthly, and people who had gone out of range had strokes or hemorraghes between tests, then the MONTHLY test would show everyone in range (because those who had gone out of range between tests had already died or not been candidates for continued monthly testing). This self-limits the population.

Testing weekly also assumes that testers (or their medical advisers) don't overreact to INRs near the top or bottom of the range by making major dosing changes that could, conceivably, move the person out of range. Conservative management of weekly results should keep the person doing the testing in range better than knee-jerk reactions when nearing the range extremes. (These are all my personal opinions, but I'm here to stand behind them)

 

[chaconne]

Interesting comments about the graph. I think there are a lot of parameters to this and it's difficult to know how to interpret the data. For one thing, the points are so clustered together, it's hard to know what shape they are. Getting the raw data would be helpful.

An interesting un-related side note:
I had my preop last Friday for my surgery next Monday. I raised the question about bridging with Lovenox if INR goes below 2.0 and the anti-coagulation clinic at Kaiser doesn't do it. They said bridging is only needed for surgeries, certain procedures etc... and that they just gently increase the dose to compensate.

 

[Protimenow]

I appreciate the side note about bridging. It's consistent with the material I've read and quoted. The new opinions about reduced need for bridging - for most of us - is probably fairly recent (the last ten years or so), and I suspect that some of us who have been on warfarin for decades are still believing what they were told was 'known' years ago.

And yes, that chart wasn't clear or easily read. I'm not sure that many people - other than the paper's authors, who had the raw data - were able to read it accurately.

(On a personal note -- I've had a toothache since last week, and started taking PenVK (which I had in the house) to try to keep infection down, and an occasional ibuprofen for the discomfort. As a result of one or both, I tested my INR after four days and it was 3.7. On Monday, I dropped my warfarin dose from 6 to 4, anticipating that my INR would probably have increased because of the antibiotic and NSAID. I actually was able to see a dentist on Tuesday when I formed an abscess, and now I'm on Amoxicillin, which I was told by the pharmacy, didn't have an effect on my INR. I took 4 mg last night, and will probably return to 6 tonight (my current usual dosage), and retest on Friday. It's really nice to have my own meter - I can test more frequently than once a week (and certainly a lot more often than some here, who do monthly testing) to be sure of my INR when changes like antibiotics and pain relievers are added to the picture)