The Peptide CommonsEst. May 2024
Independent. We sell nothing and are affiliated with no manufacturer or pharmacy. Every moderation action is logged in public
Research Methods · N-of-1 designs · continued

Washout with a one-week half-life: the arithmetic posts 31–60

This is a continuation of a long topic, addressed by post number rather than by page. Start at post 1.

FH
f.haddadTL228 Nov 2024#31

The version of washout that circulates here is a simplification of a simplification. It is not wrong, but it has lost the conditions under which it holds, and those conditions are where the interesting cases live.

0 likes 20mo
ER
eire_readerTL228 Nov 2024#32
ZS
z.szaboTL228 Nov 2024#33
t.pereira, post #14: I read post #11 twice before replying, because I had assumed the opposite. Whatever the answer on washout turns out to be, the method for getting there is the same: state the assumption, do the arithmetic in public, invite the correction. Go to post

Building on post #30 rather than restating it.

Order effects matter. If you always try A before B, you cannot separate the treatment from the sequence, and alternating is cheap.

Adding it because I spent an afternoon working it out and nobody should have to twice.

9 likes in reply to #14 20mo
KR
k.redgraveTL2Member29 Nov 2024#34

An n of one tells you about one person, which is the person you are most interested in. That is the whole value and it is not nothing.

20 likes 20mo
SI
s.ivaturiTL229 Nov 2024#35

Adding a reference point for washout. Mine is a single case, collected without controls, and I am posting the method alongside it so it can be discounted appropriately.

29 likes 20mo
CP
citation_peakTL3Regular29 Nov 2024#36
v.baptista, post #12: One caution on washout: everything above assumes the underlying documentation is what it claims to be. That assumption is doing real work and is rarely stated. Go to post

Everything in post #33 holds. The case it does not cover is the one I have.

The failure mode on washout is boring rather than dramatic. It is almost always the step everyone assumes was done correctly because it is too simple to get wrong.

0 likes in reply to #12 20mo
MN
ma.nascimentoTL229 Nov 2024#37
f.fenwick, post #20: Post #18 describes the usual case. This is about the unusual one. Careful with the language on washout. "Not detected" and "not present" are different findings and the first is a statement about the method. Go to post

Stopping rules: decide in advance when you will stop measuring (after a defined duration, after a defined number of measurements, or after a defined condition is met). Not deciding in advance means stopping when the result satisfies you, which is bias.

5 likes in reply to #20 20mo
LP
l.parkinsonTL2Member29 Nov 2024#38

I would call the community position on washout likely rather than established, and I would be comfortable defending that hedge.

14 likes 20mo
NV
n.villalobosTL230 Nov 2024#39

Adding a data point of agreement rather than a data point.

0 likes 20mo
FP
forest_plotTL3Evidence synthesis30 Nov 2024 · edited#40
s.kimani, post #1: Washout with a one-week half-life: the arithmetic — setting out what I have, and where I think it stops being reliable. Washout: setting out the arithmetic in full, because I have had to do it twice and I would rather nobody else did. Every step is shown. If the answer is wrong the error will be visible, which is the point of writing it… Go to post

I read post #36 twice before replying, because I had assumed the opposite.

Summarising the washout thread so far, since it is long and the answer is buried: the first reply has the method, the fourth has the correction to it, and the rest is people agreeing at length.

5 likes in reply to #1 20mo
TI
trough_indexTL3Regular30 Nov 2024 · edited#41

A washout period between conditions is what stops one bleeding into the next, and its length should be set by the half-life rather than by convenience.

30 likes 20mo
AN
a.norgaardTL230 Nov 2024#42

This follows post #41 rather than contradicting it.

Two sentences on washout and then I will stop, because the rest is speculation and the thread is better without mine.

What is documented is narrow. What is inferred from it is broad. The gap between them is where every argument here lives.

15 likes 20mo
K
KTurkingtonTL3Regular30 Nov 2024#43
to.varga, post #25: Where I part company with post #23, and it is a narrow parting. It does not tell you about anybody else, which is why aggregating these accounts does not produce evidence of the kind people want it to. Go to post

Coming back to post #41, because the follow-up matters more than the original answer.

A definition problem is doing most of the work in this washout discussion. Once the term is pinned down I suspect the disagreement mostly goes away and what is left is small.

5 likes in reply to #25 20mo
FN
f.novakTL230 Nov 2024#44
e.lehtinen, post #16: Washout periods: after stopping a medication, how long does it take for the effect to wash out? For compounds with a week-long half-life, roughly a month is needed to reach baseline. Using that washout period in a before-after design strengthens the inference. Go to post

A rolling mean over several days is far more informative than any single reading for anything that varies day to day, which is nearly everything.

That has held every time I have looked, which is not the same as always.

1 like in reply to #16 20mo
L
LJankowiakTL3Regular1 Dec 2024#45

On post #41 — agreed on the reasoning, with one qualification.

An observation about washout that I cannot explain and am posting anyway, on the principle that unexplained observations are more useful public than private.

0 likes 20mo
MA
mi.amankwahTL21 Dec 2024#46

Picking up post #45: that is the part I would want checked first.

My understanding of washout is a few years old and may have been superseded. If it has been, I would genuinely like to know rather than keep repeating it.

21 likes 20mo
BR
buffer_reviewTL3Regular1 Dec 2024#47
eire_reader, post #32: That is the distinction I keep failing to hold on to. Written down now. Go to post

Helpful, and short, which on this subject is harder than long.

9 likes in reply to #32 20mo
NL
ne.laurentTL21 Dec 2024#48

Multiple outcomes measured without a primary one means something will move. Nominate the primary in advance and report the rest as secondary.

2 likes 20mo
IT
integrator_traceTL2Member1 Dec 2024#49
LJankowiak, post #45: On post #41 — agreed on the reasoning, with one qualification. An observation about washout that I cannot explain and am posting anyway, on the principle that unexplained observations are more useful public than private. Go to post

On washout, I would rather understate and be corrected upward than overstate and be quoted. That is a house style here and it is a good one.

0 likes in reply to #45 20mo
NK
n.kirchnerTL22 Dec 2024 · edited#50

Reading this washout thread as someone who came in with a fixed view: the third and seventh replies moved me and the confident ones did not.

29 likes 20mo
N
NorringtonTL3Regular2 Dec 2024#51
ne.laurent, post #48: Multiple outcomes measured without a primary one means something will move. Nominate the primary in advance and report the rest as secondary. Go to post

Adding the measurement that post #48 says would settle it.

Objective versus subjective measures: subjective measures (how you feel) are vulnerable to bias. Objective measures (weight, strength on a specific exercise) are less vulnerable but not immune.

0 likes in reply to #48 20mo
DA
d.achebeTL22 Dec 2024#52

Reading back through, this was answered upthread and I missed it. My fault.

1 like 20mo
N
NicolaidesTL3Regular2 Dec 2024#53

Distinguishing three things in the washout discussion that keep getting used interchangeably: the observation, the proposed mechanism, and the recommendation that gets attached to both.

6 likes 20mo
FP
f.piresTL22 Dec 2024#54
n.villalobos, post #39: Adding a data point of agreement rather than a data point. Go to post

The strongest argument against my own position on washout, stated as well as I can state it, since nobody else has yet.

16 likes in reply to #39 20mo
TF
taper_fileTL3Regular2 Dec 2024 · edited#55

The arithmetic in post #53 is right; the assumption feeding it is the part to check.

It does not tell you about anybody else, which is why aggregating these accounts does not produce evidence of the kind people want it to.

That is one dataset and I would not build a rule on it.

0 likes 20mo
AV
a.villalobosTL23 Dec 2024#56

Answering the question post #54 raises rather than the one it answers.

A note on scope: what I am saying about washout applies to the case in the first post and I would not extend it further without checking.

3 likes 20mo
D
DKwiatkowskiTL3Regular3 Dec 2024#57

A photograph or a device reading is a harder record than a recollection, and where one is available it should be the record.

10 likes 20mo
HK
h.krastevTL23 Dec 2024#58
citation_peak, post #36: Everything in post #33 holds. The case it does not cover is the one I have. The failure mode on washout is boring rather than dramatic. It is almost always the step everyone assumes was done correctly because it is too simple to get wrong. Go to post

I think the washout question is answerable and has not been answered, which is a more optimistic position than most of this thread.

23 likes in reply to #36 20mo
MH
m.haddadTL2Regular3 Dec 2024#59
KTurkington, post #43: Coming back to post #41, because the follow-up matters more than the original answer. A definition problem is doing most of the work in this washout discussion. Once the term is pinned down I suspect the disagreement mostly goes away and what is left is small. Go to post

This is the sort of exchange that makes the archive worth searching.

1 like in reply to #43 20mo
KM
k.marchandTL23 Dec 2024#60
DKwiatkowski, post #57: A photograph or a device reading is a harder record than a recollection, and where one is available it should be the record. Go to post

On washout, the part that usually goes wrong is that the question is asked as though it has one answer. It has a range, and the width of the range is the interesting bit.

If you can post the two or three numbers you are working from, several people here will check the arithmetic rather than argue about the conclusion.

6 likes in reply to #57 20mo