Minimum detectable and minimal clinically important changes for pain in patients with nonspecific neck pain.

Background: The minimal detectable change (MDC) and the minimal clinically important changes (MCIC) have been explored for nonspecific low back pain patients and are similar across different cultural settings. No data on MDC and MCIC for pain severity are available for neck pain patients. The objectives of this study were to estimate MDC and MCIC for pain severity in subacute and chronic neck pain (NP) patients, to assess if MDC and MCIC values are influenced by baseline values and to explore if they are different in the subset of patients reporting referred pain, and in subacute versus chronic patients.

Methods: Subacute and chronic patients treated in routine clinical practice of the Spanish National Health Service for neck pain, with or without pain referred to the arm, and a pain severity > or = 3 points on a pain intensity number rating scale (PI-NRS), were included in this study. Patients' own "global perceived effect" over a 3 month period was used as the external criterion. The minimal detectable change (MDC) was estimated by means of the standard error of measurement in patients who self-assess as unchanged. MCIC were estimated by the mean value of change score in patients who self-assess as improved (mean change score, MCS), and by the optimal cutoff point in receiver operating characteristics curves (ROC). The effect on MDC and MCIC of initial scores, duration of pain, and existence of referred pain were assessed.

Results: 658 patients were included, 487 of them with referred pain. MDC was 4.0 PI-NRS points for neck pain in the entire sample, 4.2 for neck pain in patients who also had referred pain, and 6.2 for referred pain. MCS was 4.1 and ROC was 1.5 for referred and for neck pain, both in the entire sample and in patients who also complained of referred pain. ROC was lower (0.5 PI-NRS points) for subacute than for chronic patients (1.5 points). MCS was higher for patients with more intense baseline pain, ranging from 2.4 to 4.9 PI-NRS for neck pain and from 2.4 to 5.3 for referred pain.

Conclusion: In general, improvements < or = 1.5 PI-NRS points could be seen as irrelevant. Above that value, the cutoff point for clinical relevance depends on the methods used to estimate MCIC and on the patient's baseline severity of pain. MDC and MCIC values in neck pain patients are similar to those for low back pain and other painful conditions.