How to compare your records when you wonder if a supplement is working
What you can do
Pick one question, such as whether you wake up rested, and rate it the same way every day. If you decide to try a supplement, record a period without it first and a period of the same length with it, and avoid starting anything else in those weeks. Save an entry on each day you took it and note days you skipped or cannot remember. Then compare the two periods by date. A difference is a pattern in your own records, not proof that the supplement caused it, and a suspected deficiency is a question for a doctor, not for a daily rating.
Before you start
Write down why you are trying the supplement and what you hope will change. Choose one daily feeling that matches it, such as sleep or energy, and, if you already wear a device, one number such as sleep duration. If you take prescription medicine, are pregnant or breastfeeding, or have a health condition, ask a doctor or pharmacist before you start or stop a supplement. The US Office of Dietary Supplements notes that the FDA does not decide whether a supplement is effective before it is sold. The examples on this page are fictional and are not a suggestion to take anything.
Step by step
- 1.
Decide the question and how you will judge it before you begin, for example "Are more mornings rated better than usual?" If you decide after seeing the ratings, it is easy to pick whichever number looks best.
- 2.
Record a period without the supplement first, such as 14 days. Log what you eat and drink, any other supplements and the daily feeling. Without this period there is nothing to compare against.
- 3.
If you decide to start it, pick the start date in advance and keep everything else as steady as you can: the same other supplements, a similar bedtime and no new training plan in the same weeks. When two things start together, the records cannot tell them apart.
- 4.
Save an entry on each day you take it, with the amount you actually took and the time if you know it. On a day you skip it, say so in the day's note. If you cannot remember, write that instead of guessing. A day with no entry and no note does not tell you whether you took it.
- 5.
Rate the same feeling at about the same time each day, before looking back at earlier ratings. Note short nights, alcohol, illness, travel and unusually stressful days, because they move most feelings.
- 6.
At the end, read both periods by date rather than from memory: how many days have a rating, how the typical ratings compare, and where the colds or short nights fall. If you later stop taking it, recording the following days the same way shows whether the ratings move back.
Why a better week does not show that it worked
People often start something when they feel at their worst, and unusually bad days tend to be followed by more ordinary ones even if nothing changes. Statisticians call this regression to the mean. A period without the supplement, recorded the same way, shows what an ordinary run of days looks like for you.
Your ratings are also self-reports. A Cochrane review of placebo trials found no important placebo effects in general, but it found that self-reported outcomes such as pain and nausea could be influenced, and that this is hard to separate from biased reporting. Rating at a fixed time, before reading earlier days, limits how far expectations steer the number.
Researchers who test a treatment in one person, in what is called an N-of-1 trial, apply the same ideas more strictly: several periods on and off, sometimes a placebo that neither the person nor the clinician can tell apart, and outcomes chosen in advance. Useful records do not need all of that, but the more of it you use, the more a difference means.
A worked example
| Example | How to record or read it |
|---|---|
| Days 1 to 14: no supplement | Morning feeling rated on 13 days. Day 9 was forgotten and stays blank rather than being filled in later from memory. |
| Days 15 to 28: one capsule after dinner | An entry on 12 days. The two skipped evenings each have a note saying so. |
| Days 20 and 21: a cold | Noted on both days. Read them on their own instead of letting them pull the second period's ratings down unnoticed. |
| Reading the result | Mornings were rated a little higher on most days of the second period. That is a pattern in these records; the season, the cold or expectations could explain part of it. |
Check the result
In Feelments, open Views and create a view with a Records card for the supplement and a Feelings card for the feeling you chose, both set to 30 days so they cover the two periods. Check which dates hold both an entry and a rating, and open Timeline for any date you want to read in full. When Feelments compares days you logged an item with other days, it waits until both sides have enough days before showing anything. To read the dates in a spreadsheet, export your records as CSV.
Questions and troubleshooting
How long should each period be?
Long enough for the change you expect, and the same length for both periods. Some supplements are described as acting within days and others over weeks, so ask a pharmacist or doctor what is reasonable for the product you use. Choose the length before you start, not after seeing the ratings.
What if I start two supplements at the same time?
The records cannot tell them apart. Start dates that are separated by a period you read on its own keep each change distinguishable.
What if I notice no difference?
That is a result too. The supplement may not affect that feeling, the change may be too small to notice from day to day, the feeling you chose may not capture it, or the period may have been too short. A suspected deficiency is a question for a doctor, who can decide whether a test would help; a daily rating cannot answer it.
Should I stop the supplement to check?
Feelments does not suggest starting or stopping a supplement. If a clinician recommended or prescribed it, talk to them before changing anything. If you stop for your own reasons, recording the following days the same way shows whether the ratings move back.
Do device numbers make the comparison more reliable?
They add a measure that does not depend on how you remember the day, such as sleep duration or resting heart rate. They also change with alcohol, illness, training and the season, so read them alongside the same notes.
What this does and does not tell you
Your records can show whether a feeling or a number differed between periods you chose. They cannot show that the supplement caused the difference, rule out other causes, or tell you whether it is safe or suitable for you. Take medical questions, a suspected deficiency or a change to prescribed treatment to a qualified professional.
MHLW eJIM: evaluating health information (Japanese)