- Proposes 5 pain types:
1) Segmental nerve/cauda equina
2) Spinal cord
- Combines both mechanistic factors (e.g. slow fibre conduction from skin) and descriptive factors, such as time to onset post-injury, characteristics of pain (e.g. burning, stabbing, dull aching, etc), pain duration, and factors that make it worse or better.
Body Function – Subcategory: Sensory Function
Number of Items:
5 possible categories for each pain area.
Brief Instructions for Administration & Scoring
- This information is obtained through a semi-structured, pen and paper interview.
- Up to 40 minutes is required for those with complex issues.
MCID: not applicable
SEM: not applicable
MDC: not applicable
Presenting the information in a table format (like the one below) facilitates interpretation and clarity. The table below presents the descriptors for 5 different patients with each type of pain (examples for each category shown).
|Pain Type||Time of Onset Post Injury||Character||Duration||Aggravating factors||Diminishing factors||Possible causative factors|
|Days to weeks||Burning
|Seconds||Rest||Activity||Slow fibre conduction from skin|
|Spinal Cord||Weeks to months||Tingling
|Constant||Activity||Rest||All fibre conduction within cord|
|Visceral||Weeks to months||Burning||Constant||Variable||Variable||Slow fibre conduction from viscera|
|Mechanical||Weeks to months||Dull
|Variable||Activity||Rest||Slow fibre conduction from muscles or ligaments|
|Psychic||Variable||Variable||Variable||Variable||Variable||Preoccupation with unpleasant environmental stimuli|
Knowledge on the study of pain is recommended.
See the ‘How-to Use’ page of this tool.
- The assessment can be time-consuming for those patients with complex pain issues. However, this type of approach may be more suitable for difficult cases as it allows the patients to explain pain in their own language rather than being forced to pick specific descriptors for their pain.
Measurement Property Summary
# of studies reporting psychometric properties: 2
- Overall test-retest reliability is 78%; percentage agreement for segmental nerve/cauda equine was 67%, for visceral was 75%, for mechanical was 80% and for spinal cord was 84%.
- Intra-rater agreement ranged from 67-83%.
- Inter-rater agreement ranged from 62-73%.
[Richards et al. 2002, Putzke et al. 2003]
No values were reported for the validity of the Donovan SCI Pain Classification System for the SCI population.
No values were reported for the responsiveness of the Donovan SCI Pain Classification System for the SCI population.
No values were reported for the presence of floor/ceiling effects in the Donovan SCI Pain Classification System for the SCI population.
Dr. Vanessa Noonan, Matthew Querée
Date Last Updated:
Dec 1, 2016
Donovan W, Dimitrijevic M, Dahm L, Dimitrijevic M. Neurophysiological approaches to chronic pain following spinal cord injury. Paraplegia 1982;20:135-146.
Putzke JD, Richards JS, Ness T, Kezar L. Test-retest reliability of the Donovan spinal cord injury pain classification scheme. Spinal Cord 2003;41:239-241.
Richards JS, Hicken BL, Putzke JD, Ness T, Kezar L. Reliability characteristics of the Donovan spinal cord injury pain classification system. Arch Phys Med Rehabil 2002;83:1290-1294.