Health tracking is offered as unambiguously good. For many people it is useful. For some it makes things worse, and the mechanisms are understood well enough to design around.
This is not an argument against keeping a record. It is the difference between a record that earns its place and a practice that becomes part of the problem.
How it backfires
Attention makes symptoms louder. Directing attention to a bodily or mental sensation reliably increases how intense it seems. Watch closely for word-finding pauses and you will find them — they were always there, at the same rate, unnoticed. The record then documents a deterioration that is an artefact of looking.
Rating requires introspection, and introspection affects the thing. Asking yourself how your mood is requires attending to your mood. For anxiety and low mood particularly, that attention is not neutral.
Normal variation looks like signal. Clarity, energy and mood vary substantially day to day for no reason at all. Plotted, that variation looks like a pattern, and looking for a cause for every dip generates a lot of false explanations and a lot of worry.
It supplies an outlet for checking. Where there is a tendency toward reassurance-seeking, a tracking app becomes the thing to check — which maintains anxiety rather than relieving it, in the same way that any checking behaviour does.
It can consolidate an identity. Assessing your health daily keeps being-unwell in the foreground. For a chronic condition that is being actively managed, that is sometimes necessary. As a permanent default it is a cost.
Who should be careful
Not a diagnosis, and worth recognising yourself in:
- If you tend toward health anxiety, or have in the past
- If you check things repeatedly for reassurance
- If a symptom becomes more noticeable when you think about it
- If a previous attempt at tracking left you worse
- If you find yourself testing whether something still works
- If the log is the first thing you reach for in the morning
The version that tends to be safe
Event-based, not scheduled. Write something down when it happens. No daily prompt, nothing to complete, no streak.
Instances, not ratings. "Could not remember Sunday's conversation with my brother" rather than a number. Instances are facts; ratings require self-assessment.
Include ordinary days, briefly. A log of only bad days looks like decline and reads as decline.
Time-limited. Two or three weeks for a specific question, then stop. An open-ended practice is the one that drifts.
One place, reviewed rarely. Write, close, do not reread. Rereading to compare is the behaviour that does most of the damage.
A question written at the top. "Is the fog worse in the afternoons?" When it is answered, stop.
Rules worth setting in advance
- No rereading until the review date.
- No more than once a day, and preferably less.
- No testing yourself. No recalling word lists, no checking whether you can still do arithmetic. These produce anxiety and no usable data.
- A stop date. Decide it when you start.
- A stop condition. "If I feel worse after logging for three days, I stop." Write it down.
If it is already making things worse
Stop. Then tell someone.
The second part matters. "I tried keeping a record and it made me much more anxious, so I stopped" is clinically useful information — it tells a clinician something about what is going on, and it prevents the same suggestion being made again.
There are alternatives that produce most of the benefit with less of the cost:
- Ask someone else to keep the record. Their observations are weighted more heavily in any assessment anyway, and it removes you from the monitoring loop entirely.
- Note only what a clinician specifically asked for. Two items, not twelve.
- Record function rather than sensation. What you did or could not do, rather than how you felt. Less introspective and often more useful.
The thing worth keeping
Even for someone who should not track, one practice is almost always worth it: when something notable happens, write a line with the date.
No schedule. No rating. No app. A note that says what happened and when.
That single habit produces the thing clinicians most need — specific, dated, contemporaneous observations — while creating none of the monitoring machinery that causes the harm. It is also what you will want in six months, when the question is whether things have changed and the honest answer would otherwise be "I am not sure."
The underlying point
A record is a tool for answering a question. It is not a practice, a discipline, or evidence of taking your health seriously.
If there is no question, there is no reason to collect data, and collecting it anyway has a cost. Knowing when to stop is part of using it well.