Medical records are written by clinicians for clinicians, under time pressure, in a mix of templated text and free dictation, and then printed by a records vendor in whatever order the system exported them. Reading one for a case is a different skill from reading it for care. The clinician wants to know what to do next. The lawyer wants to know what happened, when, who said so, and whether the story holds together. This guide is about the second kind of reading.
A production usually contains a handful of record types, and each one is reliable for different things.
Emergency department records are the closest thing to a contemporaneous account of the incident. The triage note, the nursing assessment, and the physician's history of present illness are written within hours, before anyone has talked to a lawyer or an adjuster. The mechanism of injury recorded here tends to carry more weight than any later version, and discrepancies between the ED history and later histories are where cases turn.
Office visit notes from primary care and specialists document the course of treatment. Most are built on a template: chief complaint, history of present illness, review of systems, exam, assessment, plan. The templated sections are often copied forward from visit to visit and should be read with that in mind; a review of systems that is identical across twelve visits was not re-asked twelve times. The history of present illness and the assessment and plan are where the provider actually wrote something.
Imaging reports state what the radiologist saw. They do not state what caused it. A report that describes "degenerative changes" is describing appearance, not timing, and whether a finding is old or new is a question for a clinician comparing studies, not a fact the report itself supplies. Prior imaging, when it exists, is the most valuable comparison in the record.
Physical therapy records are dense, repetitive, and underrated. Every visit documents pain scores, measured range of motion, functional limits, and attendance. They are the best objective trend line in most injury cases and the first place to look for a treatment gap or a plateau.
Operative reports and procedure notes describe exactly what was done and what the surgeon found. They are short and precise, and the findings section sometimes says more about chronicity than any office note.
Billing records, pharmacy histories, and problem lists are indexes. They tell you what encounters and prescriptions existed, which is how you find the providers whose notes are not in the production.
A few entries do most of the work in an injury case, and they are rarely on the pages a records vendor puts first.
The first recorded mechanism of injury, usually in the ED triage or the first office visit. The first recorded complaint of each body part; a knee that appears for the first time at visit nine is a different fact from a knee documented on day one. Any reference to a prior injury, prior claim, prior surgery, or prior imaging of the same region. Work status and restrictions, which change over time and are often stated in one line. Pain scores and measured findings over time, from therapy. Referrals that were ordered, and whether they were completed. Statements attributed to the patient in quotation marks. And the discharge or maximum medical improvement note, if there is one, which marks the end of treatment.
Read in date order, not production order. Reordering is tedious and is the single change that most improves comprehension, because the story only makes sense in sequence.
Read every page, then decide what was important. The instinct to skip the nursing flowsheets and the duplicate fax covers is reasonable and is exactly how the nursing note recording "patient states neck pain since MVA three years ago" gets missed.
Distinguish what the provider observed from what the patient reported. "Patient reports" is history; "on exam" is a finding. Both matter, but they are different kinds of evidence and get cross-examined differently.
Notice copied text. Electronic records carry forward large blocks from prior visits. A plan that reads "continue PT, recheck in four weeks" at six consecutive visits may reflect six decisions or one decision and five clicks.
Keep a running list of what is missing. Every referral without a matching record, every "as previously discussed" without a previous discussion in the file, every prescriber in the pharmacy history with no office notes. That list is the discovery plan.
A short set covers most of the confusion. MVA is motor vehicle accident. HPI is history of present illness. ROM is range of motion, usually with degrees. LBP is low back pain. WNL is within normal limits, and NAD is no acute distress; both are default template entries that mean the provider saw nothing remarkable, not that the provider tested for anything specific. MMI is maximum medical improvement. PRN means as needed. Rx is a prescription; Dx is diagnosis; Tx is treatment; Hx is history; Sx is symptoms; Fx is fracture. "c/o" is complains of; "s/p" is status post, meaning after a procedure or event. Anything else is a two-second search, and the answer is more reliable than a guess.
A trained paralegal can build a competent timeline from a clean 400-page production. The reasons to hand the reading to a nurse practitioner, physician assistant, or physician are volume, ambiguity, and stakes. A clinician reads a chart the way it was written, catches the significance of a finding that a lay reader would pass over, and recognizes when a normal-looking note is not normal for that diagnosis. On a 3,000-page case with a contested causation question, that difference is the case.
Chronological, regardless of how they were produced. Sort by date of service first, then read. Records vendors export by facility or by document type, which separates records that belong together in time.
The history of present illness and the assessment and plan, because the provider composed them for that visit. Review of systems and exam templates are often carried forward from earlier visits and should be checked against the visit-specific text.
Only by comparison with prior imaging or by a clinician's reading of the description. The radiology report describes appearance, not timing. If prior studies exist, request the reports and, where the question is contested, the images.
Yes, for small and clean records. For large productions, contested causation, or records with clinical ambiguity, a clinician reader catches what a lay reader passes over, and the cost of a clinician chronology at $0.25 per page is usually less than the paralegal hours it replaces.
Every page, in order, with the entries that matter cited by page number.
Get Access