Background

Knee pain is a major source of distress and disability, with pain progression highly variable between individuals. Previous studies defining pain trajectories have all used a single measure of pain, and these differ across studies. Different measures reflect diverse pain mechanisms. To ascertain the clinical utility of pain trajectories we explored associations between opioid and non-steroidal anti-inflammatory drug (NSAID) use.

Methods

We model pain trajectories using two measures - Intermittent and Constant Osteoarthritis Pain (ICOAP) and the painDETECT, in 2141 participants, across 3 waves (the baseline, 1- and 3-year assessments) of the Knee Pain In the Community (KPIC) cohort.

Results

Latent class growth analysis identified 6 trajectories using ICOAP subscales (High Stable, Low Stable, Moderate Worsening, Moderate Recovering, Worsening, and Recovering), and 4 trajectories using painDETECT (High stable, Low stable, Moderate Worsening, and Moderate Recovering). There was a high degree of correspondence between people assigned to pain trajectories between ICOAP intermittent and constant subscales, but less so using painDETECT. Opioid use was associated with ICOAP trajectories only (e.g., High Stable and Worsening intermittent ICOAP trajectories) and in women.

Conclusion

Different measures of pain produce different patterns of pain progression and these are differentially related to medication use. Opioid use is linked to trajectories of pain based on the impact of pain on behaviour and not pain symptoms. Thus, managing pain’s behavioural impact is more central to understanding opioid use than managing pain symptoms. These findings support more in-depth questioning about the type of pain and its progression in clinical practice.