Use serial examinations, imaging, mechanism, prior condition, and function as a chain—not competing shortcuts.

Collect prehospital run sheets, trauma flowsheets, nursing examinations, physician notes, consultation, operative records, rehabilitation assessments, and follow-up. Extract the exact time, examiner, body side, muscle group or sensory area, cooperation, sedation, pain, splinting, fracture, intoxication, language, and other limitation. A copied-forward phrase can look like a new examination; compare source fields and note authorship.

Preserve original imaging and reports, not just selected screenshots. Computed tomography can show fractures and alignment; magnetic resonance imaging can show cord, disc, ligament, blood, swelling, and compression findings in appropriate circumstances. The treating radiologist and specialists should explain timing, quality, sequence, and significance. The legal team should not claim that a single sequence proves trauma timing or rules out every condition without expert support.

Track diagnosis and classification language over time. Emergency concern, radiology impression, surgical finding, discharge code, rehabilitation classification, and later specialist opinion may differ because information changed. Identify neurological level, completeness, central or other pattern if clinically supported, related vertebral or nerve injury, and differential diagnoses. The goal is not to force uniform language; it is to explain why the final opinion is reliable.

Reconstruct the event with measurements and physical evidence. For a crash: vehicle damage, intrusion, restraint geometry, seat position, occupant movement, event data, scene marks, and post-impact events. For a fall: height, path, landing surface, body orientation, obstructions, and immediate symptoms. For equipment or product events: motion, controls, energy, guard, component condition, maintenance, and alteration. Witness adjectives should be tested against the record.

A biomechanical opinion may address force direction, body movement, and consistency with an injury mechanism. Medical experts address anatomy, diagnosis, differential causes, aggravation, prognosis, and treatment. They should not simply adopt each other's conclusion. Each expert should state materials reviewed, methods, assumptions, uncertainty, and whether the opinion changes if a disputed fact is resolved differently.

The differential should include relevant prior conditions and non-traumatic explanations. Prior spine disease, surgery, symptoms, neuropathy, inflammatory or vascular conditions, infection, tumor, and other neurological causes may matter depending on the case. Temporal association supports a question but is not a complete method. Conversely, preexisting degeneration does not establish that the event caused no new cord injury or aggravation.

Collect standardized rehabilitation measures when used, therapy goals, transfers, mobility, wheelchair skills, walking trials, upper-extremity function, self-care, bowel and bladder routines, respiratory independence, skin management, pain, spasticity, endurance, and discharge recommendations. Explain each scale and its limits. Do not translate a score into a legal percentage without foundation.

Obtain contemporaneous observations from family, coworkers, first responders, therapists, and others who saw function before and after. Ask for concrete tasks and dates, not medical labels. Photographs and videos may show transfers, equipment, gait, or activity. Note context: off-frame assistance, fatigue, pain, surface, device, and whether the clip reflects a typical day.

Separate injury proof from damages. Diagnosis and causation support one part of the case. Future care requires medical recommendations and probability. Work loss requires actual demands and vocational evidence. Pain, impairment, assistance, and participation require individual records. Liability requires duty, breach or defect, control, and causal proof under governing law.

Preserve physical objects and native files. Coordinate inspection of vehicles, restraints, seats, equipment, components, scene, original imaging, device data, and photographs. Record acquisition tools, versions, checksums when appropriate, time zones, conversions, custody, and destructive-testing protocol. An exported PDF or screenshot should not silently replace the underlying file.

Select experts by disputed proposition. Radiology, neurology, neurosurgery, rehabilitation, biomechanics, reconstruction, product or mechanical engineering, human factors, life-care, vocational, and economic expertise may matter. More experts are not always better. Overlap may create inconsistency and cost; an issue map should define each assignment.

Test opinions before relying on them. Confirm credentials, reviewed records, complete history, accepted methods, alternatives, missing data, demonstrative accuracy, future-care foundation, and admissibility standards. Treating opinions may need clarification, but the legal team should not draft medical conclusions for signature. Meaningful uncertainty is more credible than certainty beyond the evidence.