Learn how migraine baseline tracking over 90 days uncovers patterns, monthly migraine days, and the data your doctor needs to measure treatment progress.
The first thing most people discover when they start migraine baseline tracking is that their memory is wrong. Not wrong in a dramatic way, just wrong in the quiet, cumulative way that makes you realize you have been averaging six migraine days a month in your head when the actual number is nine. That gap matters. It affects the conversation you have with your doctor, the treatment decisions that follow, and whether any of those decisions can be objectively evaluated later.
A 90-day baseline is the foundation of all of that. Here is what the process looks like, what the data reveals, and why the time investment is worth it.
One month of data is almost useless on its own. A single month can be a fluke in either direction: unusually calm because you were on vacation, unusually severe because of a respiratory infection or a stressful deadline. Two months is better but still susceptible to a paired outlier.
Three months gives you a stable average. It captures at least one full hormonal cycle for most people, accounts for seasonal variation, and smooths out single-month anomalies. When your doctor asks, "How many migraines do you have per month?" and you answer with a number, that number should come from a 90-day average, not a rough estimate.
The 90-day window also matters for treatment evaluation. Before any preventive can be assessed, there needs to be a pre-treatment reference point. Without a documented baseline, you and your doctor are comparing your current state to a number you invented from memory. That comparison is unreliable and can lead to undervaluing a treatment that is actually working or staying on one that is not.
The core metric is your monthly migraine day count: the number of calendar days per month on which you had a qualifying migraine attack. A migraine day is any day with a migraine-type headache lasting at least four hours, or any day on which you treated a headache with a migraine-specific acute medication regardless of duration.
Log every single day. Headache-free days must be explicitly recorded, not implied by the absence of an entry. Without them, your count is meaningless. A log that only shows attack days cannot tell you your monthly frequency.
For a full breakdown of what to record during each attack, see the guide on what to log in a migraine diary.
Frequency is only one dimension. Ninety days of data also shows you:
These details tell a different story than frequency alone. Someone with eight migraine days a month where six resolve within two hours of treatment is in a different clinical situation than someone with eight migraine days where most last 24 hours and resist treatment.
Triggers are notoriously hard to identify from memory, because the gap between trigger and attack can span 24 to 48 hours. A 90-day log gives you enough paired observations to start seeing real patterns: hormonal triggers that appear on predictable cycle days, weather-related clusters, or sleep disruption links that only show up when you can compare entries side by side.
This is also where you start seeing what is not a trigger, which can be equally useful. Many people carry trigger assumptions that their data does not support.
Frequency numbers alone do not capture impact. Two people with the same monthly migraine day count can have vastly different levels of disability depending on attack severity, timing, and recovery time. Tools like the MIDAS and HIT-6 scales quantify that impact in a standardized way.
You can calculate your current MIDAS score at /tools/midas-calculator and your HIT-6 score at /tools/hit-6-calculator. For a full explanation of what these scores mean and how clinicians use them, see the guide on MIDAS and HIT-6 explained. Running these at the start of your baseline gives you a disability snapshot that can be directly compared after treatment.
Baselines need a defined start date. Pick one, record it, and count 90 days forward. Knowing the end date helps with consistency: you are collecting data for a specific, finite period rather than indefinitely.
This is the most common failure point. People log during attacks and skip headache-free days. After a few weeks, the log looks like an attack diary instead of a complete picture. Set a daily reminder if needed. A 30-second entry on a headache-free day takes less effort than a single email.
For detailed guidance on avoiding the errors that quietly corrupt tracking data, see how to track migraines accurately.
If you are establishing a pre-treatment baseline, starting or stopping a preventive during the 90 days invalidates the data. You need a stable period that reflects your untreated (or stably-treated) frequency. If a treatment change is necessary for medical reasons, restart the baseline clock after things stabilize, in consultation with your clinician.
A 90-day summary changes the clinical conversation. Instead of "I think I get migraines pretty often," you arrive with:
For more on how clinicians interpret this kind of data, see how doctors use migraine data. If you are preparing to share your log directly, the guide on sharing your migraine log with your doctor covers the most useful formats and what to highlight.
Once a preventive is in place, the 90-day baseline becomes the reference for measuring reduction. You can run those numbers at /tools/migraine-reduction-calculator to see the percentage change in monthly migraine days from baseline to current.
Most people report the same thing when they finish a 90-day baseline: the number was higher than they thought, the pattern is clearer than they expected, and the log gives them a sense of agency they did not have before. It is not that the data changes the migraines. It is that having the data changes what you can do about them.
Migraine Tracker: CGRP Log builds your 90-day baseline entry by entry, calculates your monthly migraine day averages automatically, and generates a renewal-ready summary report that packages your frequency, severity trends, and disability scores in a format your clinician can actually use at your next appointment.
Educational, not medical advice. Migraine Tracker: CGRP Log is a personal tracking tool, not a medical device. It does not diagnose, treat, or provide medical advice. Always talk to your clinician.
Most headache specialists consider 90 days (three full calendar months) the minimum for a reliable baseline. A single month can be skewed by illness, travel, or an unusual stressor. Three months averages out those fluctuations and gives a stable monthly migraine day count to compare against once treatment begins.
Learn how to track migraine aura episodes with precision: which details to capture, when to log them, and how that data helps your clinician.
A migraine pattern recognition diary turns raw attack logs into clear trends your doctor can act on. Learn what the data reveals and how to read it.
The headache day vs migraine day difference shapes how doctors assess your condition and adjust care. Learn what each term means and why tracking both counts.
