The Complete Guide to Tracking Lyme Disease

You're in your LLMD's office at 2:15 on a Thursday. They ask the question they always ask: "How have things been since we changed the protocol?" You open your mouth and nothing useful comes out. You know the last six weeks were rough. You think the joint pain got better around week three. You're pretty sure last week was bad, but you can't remember why. You had a herx in there somewhere. The doctor is patient. You sit there trying to reconstruct six weeks of your own body from memory, and the answer you give is some version of "up and down, mostly down, I think."

That moment is when most patients quietly decide tracking isn't optional anymore. Not because anyone told them to. Because the cost of not tracking got too expensive in the only currency that matters: actual treatment decisions made on actual evidence.

This guide walks through what to track, how often, what to skip, and how to turn four weeks of consistent data into something you and your doctor can use. It's long. Lyme is complicated and a five-minute version would be lying to you. Take it in pieces.

Why tracking beats memory

Your memory was never good enough for this, and Lyme makes it worse.

Three things stack against you. First, neurocognitive symptoms. Brain fog, word-finding trouble, and the kind of fatigue that flattens recall are common in chronic Lyme. The same symptoms that make you want to track are the symptoms that make you bad at remembering what to track. Asking yourself on Monday "how was last Tuesday" is asking the wrong brain on the wrong day.

Second, recall bias. Research on patient-reported outcomes has shown for decades that people remember symptoms more accurately within the same day than even one day later, and accuracy keeps degrading from there. People also distort memory based on how they feel right now. If today is bad, the last six weeks feel like they've all been bad. If today is unexpectedly good, you forget what last week was like.

Third, doctors need timelines, not snapshots. An LLMD with a 30-minute slot is trying to figure out whether the protocol is moving the needle. "I felt awful yesterday" tells them about yesterday. They need "fatigue averaged 3.5 the first two weeks on the new dose, dropped to 2.6 the next two, joint pain unchanged at about 3, herx around day 5 lasted four days." That second answer is something they can act on.

Real Lyme patterns also don't show up day to day. They show up across weeks. A flare cycle every four weeks. A symptom that worsens in the 36 hours before a pressure drop. A protocol change that kicks in slowly over a month. None of that is visible in a single bad afternoon. It's only visible when you have enough scored days lined up that the noise averages out.

What to track: the five buckets

A good system tracks five things. More than that and you'll quit. Less and you can't find patterns. Here's the working set.

Symptoms with severity, on a 1 to 5 scale

This is the foundation. Every symptom that matters to you, scored on the same scale, the same way, every day. Fatigue, brain fog, joint pain, headache, sleep quality, mood, plus whatever else is loud for you (chest tightness, nerve pain, GI stuff, palpitations, sound or light sensitivity).

Use 1 to 5, not 1 to 10. A 1 to 10 scale sounds more accurate but in practice adds noise. Most people can't reliably tell a 6 from a 7 in their own fatigue. They can tell mild from moderate from severe. The extra granularity ends up random, and random data washes out real patterns.

Anchor each level in plain language for each symptom. Write it down once. For fatigue: 1 is "normal energy for me," 2 is "a little dragging but functional," 3 is "noticeably tired, slowing me down," 4 is "couch-level, hard to do normal things," 5 is "can't get up, need help." For joint pain: 1 is "background ache, not limiting," 2 is "noticeable but I can ignore it," 3 is "can't ignore it, modifying activity," 4 is "limiting most movement," 5 is "can't put weight on it / can't grip." Different symptoms, different anchors. Without them, today's 3 and next month's 3 are different numbers in different moods.

Score per symptom, not "overall." Lyme doesn't move as one thing. Joint pain at 4 and brain fog at 2 average out to a meaningless 3.

Treatments

Drug or herb. Dose. Schedule. Start date. Stop date or change date. That's it.

The reason this matters is that without it, you can't connect cause to effect later. Six weeks from now, when you're trying to figure out whether the herbal added in week two helped, you need a clean record of when it started, at what dose, and whether anything else changed at the same time. "Started new protocol around mid-May" is not enough. "Started cryptolepis 2x daily on May 14, increased to 3x daily on May 28" is something you can correlate against your symptom data.

This applies to dose changes, brand changes, missed doses, and stopping something for any reason. Especially stopping something. You'd be amazed how often the question "when did the brain fog get worse" has a clean answer that lines up with "when did I run out of binder for four days."

Herxheimer reactions, separately

Herxes are not normal days. Treating them as normal days quietly contaminates your baseline data. If your average fatigue across a month includes a five-day herx where it spiked to a 5, you're not going to be able to tell whether your everyday fatigue is improving.

Log a herx as its own entity, not a string of high daily scores. What you want for each one: start date, what changed in the treatment, how bad it got at peak, how many days it lasted, which symptoms specifically flared, and what you tried. After three or four of these, real patterns start showing up. Maybe your herxes are getting shorter (often a good sign). Maybe one antibiotic consistently triggers worse reactions than another. Maybe they cluster on a four-week rhythm. None of that is visible if herxes are blended into your regular logs.

Factors

The things that move your symptoms even when nothing else changed. The honest list:

Sleep duration and quality. The biggest single lever for most chronic Lyme patients. A six-hour night versus an eight-hour night isn't a small difference; it can be the difference between a 2 and a 4 the next day.

Weather. Pressure changes especially. A lot of patients flare 24 to 48 hours before a storm rolls in. Some flare with humidity. Some with cold. You don't know which kind you are until you have a few months of data lined up against the local weather.

Stress. Not the abstract kind. The specific kind: a fight with a family member, a hard deadline, a financial scare, a medical appointment. Score it on the same 1 to 5.

Activity. Both directions. Overdoing it on a good day pays a tax 24 to 72 hours later (the post-exertional crash that anyone with PEM knows well). Sitting still for a week also makes things worse for most people. Track the actual amount, roughly.

Diet. Not every meal. The big stuff: alcohol, gluten, sugar binges, eating out, anything you suspect for yourself. Patterns show up when you have the data.

Mood. Anxiety and depression both feed back into symptom severity, and the brain inflammation in Lyme makes the connection messier than usual. Track it as its own number, not lumped into "fatigue."

You don't have to track all six factors from day one. Start with sleep and weather. Add the rest in week two or three.

Notable events

The non-medical stuff that affects your body. A cross-country flight. A funeral. A new job. A bad cold or flu (Lyme tends to flare for weeks after any other infection). A move. A fight. A surgery, even a minor one. A vaccination. These don't fit on a severity scale and they don't fit in the factor list, but they often explain a flare nobody can otherwise explain. A short text note dated to the day is enough.

The cadence

The rhythm matters more than the depth. A perfect tracking system you abandon after two weeks is worse than a sloppy one you keep.

The daily check-in should take 60 to 90 seconds. Score your active symptoms. Note the basics on factors (sleep, anything notable). Add a one-sentence note if anything stood out. That's it. Anyone telling you to spend 15 minutes a day on this is selling you something. The whole point is that tired, foggy, in-pain you can do it without thinking. Pick a time of day and stick with it. Most patients land on early evening, after the day has played out but before sleep blurs the memory. Mornings work too if you're consistent.

Herx logs happen as needed. When you feel a flare coming on that has the herx signature (started or changed treatment recently, all your existing symptoms got louder, plus the chills/sweats/aches/fog stack), log it as a herx, not as bad regular days. Update it as the herx evolves. A herx log might span six entries over five days.

The weekly review is five minutes, on the same day every week. Sunday evening tends to work. Open your last seven days, look at the pattern. Anything obvious? A bad stretch? A factor that lined up with worse days? You're not analyzing yet. You're just looking. The act of looking is what catches things you'd otherwise miss.

The monthly summary is ten minutes, ideally the day before any doctor appointment. Pull the last 30 days into a chart or report. Note the trend lines. Average severity by symptom. Any herxes that happened, with the treatment context. Factors that correlated. A summary with three or four bullet points is what your LLMD actually needs to see.

That's the cadence. 60 to 90 seconds daily, 5 minutes weekly, 10 minutes monthly. Five hours total over a year. That's the real budget.

The pitfalls

Most tracking attempts fail in predictable ways. Knowing the failure modes ahead of time saves the project.

Over-engineering early is the most common one. People come in motivated, set up a 40-field spreadsheet, build color-coded charts, and quit by week two. The system has to be simple enough that you'll do it on your worst day, not your best day. If your worst-day version of yourself can't open the app and tap five things in 90 seconds, the system is too heavy. Cut it.

Vague entries quietly destroy the dataset. "Kinda bad today" and "rough day" can't be compared to anything. Six months in, when you're trying to find out if you're better than you were in March, prose entries are unusable. Numbers you can compare. Words you mostly can't. Use the scale. Save the prose for the one-line note next to the number.

Abandoning after two weeks is the failure mode for the medium-motivated. You start strong, life gets in the way, you miss a day, then three, then a week, and the gap feels too big to come back from. The trick is that there's no "coming back from" anything. Just start logging again from today. Don't try to backfill. The data starts where it starts.

Scoring from memory is worse than skipping a day. If you didn't log Tuesday, leave Tuesday blank. Do not score it on Wednesday from memory. Same-day scoring is the only reliable kind. Backfilled data introduces error you can't separate from real data later. A blank day is honest. A guessed day is misleading and you can't tell which is which a month later.

A 1 to 10 scale will hurt you. People feel like 10 points are more accurate than 5, but the noise from telling a 6 apart from a 7 swamps any extra signal. 1 to 5 with anchored language wins every time. The granularity isn't the goal; consistency is.

The 4-week minimum

Here is the part nobody wants to hear. Tracking for one week tells you almost nothing. Tracking for two weeks tells you a little. Real patterns in chronic Lyme need about four weeks of consistent data before anything reliable shows up.

The reason is the noise floor. Lyme symptoms move around on their own. A bad Wednesday doesn't mean anything. Three bad Wednesdays in a row doesn't mean much either, statistically. You need enough data points that the random day-to-day variation averages out and the underlying trend shows up. Four weeks is roughly the floor. Eight is better. Twelve is when correlation with weather, treatment cycles, and life events starts becoming clearly readable.

This is also why "I tracked for ten days and didn't see anything" isn't evidence the tracking isn't working. It's evidence you don't have enough data yet. Keep going.

The flip side: once you have four weeks of clean data, the patterns that emerge are usually obvious. Not subtle. People are routinely shocked at how clearly their flare cycle, their post-exertional crash window, or their sleep-to-symptom lag time shows up once enough days are stacked together. The information was always there. You just couldn't see it day by day.

How LymeTrack handles all of this

LymeTrack is built around the cadence above. Here's what each piece looks like in the app.

The daily check-in is five steps. Severity goes on a 1 to 5 scale with descriptive levels for every symptom, and the descriptions are anchored to that specific symptom (joint pain has joint-pain anchors, brain fog has brain-fog anchors, not generic numbers). On a normal day, the whole flow takes about a minute and a half. On a bad day, you can tap through it in under a minute.

You can do multiple check-ins per day. This matters more than people realize. Mornings and evenings in Lyme are often very different bodies. A single daily score averages over that, which hides the pattern. Patients running a morning and evening check-in routinely discover that their fatigue is a 2 in the morning and a 4 by 6 PM, every day. You'd never see that from one score.

The Herxheimer reaction screen is separate from the daily check-in on purpose. Herxes get logged with their own start date, peak severity, duration, treatment context, and which specific symptoms flared. They sit in their own bucket so they don't contaminate baseline averages. Later, you can see a clean history of every herx you've had and which medications triggered which kinds.

Factor tracking lives alongside symptoms in the same flow. Sleep, weather, stress, activity, diet, mood. The data needed for pattern detection is captured next to whatever you're feeling, day one, rather than something you'd have to add later.

The Compass and Insights view is where the cross-week work happens. After two to four weeks of data, it starts pulling together severity scores, factors, and treatment changes to show what's correlating. Whether your sleep is dragging your symptoms. Whether the new protocol is moving the needle. Whether the weather pattern other patients talk about is real for you specifically. Treatment-symptom correlation is built in: change a treatment, and the app tracks how each of your symptoms moves over the days and weeks afterward.

Symptoms, treatments, and factors are customizable. The defaults cover what most Lyme patients deal with, but you can add your own (a specific nerve pain location, a non-standard supplement, a factor that matters to you that nobody else tracks) in user preferences. The 1 to 5 anchored scale stays the same; the labels are yours.

The doctor-shareable report is the last piece. A clean summary your LLMD can read in two minutes, with severity trends per symptom, the herx history with treatment context, and the factor correlations. Not a screen-share of your phone. An actual export designed to make a 15-minute appointment go further.

Getting started: the first four weeks

A realistic plan, by week.

Week one is baseline. Pick three to five symptoms that matter most. Don't try to track everything. Daily check-in only, same time every day. Sleep is the only factor; ignore the rest for now. The goal is seven days without missing one. That is the only goal. Resist looking at the data yet; there isn't enough to look at.

Week two, add factors. Keep the daily check-in. Add weather (most apps pull this automatically) and stress. If you're on treatment, log any dose changes. If a herx hits, switch to the Herx screen for it. End of the week, spend five minutes reviewing. Are you logging consistently? Are your severity numbers actually consistent, or is "3" drifting? Re-read your anchors.

Week three is review. Two weeks of data is enough to start seeing the shape of things. Open the insights view. Look for anything that stands out. Don't trust correlations yet (the dataset is too small), but get familiar with the views. Add one or two more factors if you want (activity, mood, diet). Resist adding more symptoms.

Week four, share with your doctor. Generate the monthly report. Look at it before the appointment. The trends won't be deeply correlated yet, but you'll have something concrete: average severity per symptom, which days were worst, any herx events with their treatment context, and the factor data you've been collecting. Bring it. Even a rough four-week summary changes the conversation. Your doctor sees a timeline, not a snapshot, and the next treatment decision gets to be made on actual evidence.

After four weeks, you have the habit and the foundation. Months three and beyond are when this starts paying off in ways that surprise you. The cycle nobody could explain becomes visible. The mystery flares stop being mysteries. The "is this working" question stops being a guess.

That's the whole point. Not tracking for tracking's sake. Tracking so the next time someone asks how the last six weeks have gone, you have a real answer.