Objective: To determine whether symptoms and clinical signs can predict persistent neck/shoulder pain, future medical care-seeking and sickness absence.
Design: A population-based cohort was followed prospectively over a 5—6 year period.
Setting: Subjects from the district of Norrtälje (Sweden).
Subjects: Subjects with self-rated neck/shoulder pain were included (n = 1471).
Main measures: Cox regression analyses were used to test the predictive value of single and combinations of symptoms and clinical signs obtained with questionnaires and simple tests concerning persistent neck/shoulder pain, future medical care-seeking and sickness absence.
Results: Several symptoms and clinical signs were associated with the outcomes of interest: the relative risk (RR) for persistent neck/shoulder pain was 1.38 (95% confidence interval (CI) 1.14—1.52) for subjects with pain for over three months at baseline and concerning future medical care-seeking RR was 2.10 (95% CI 1.73—2.54) for subjects who had previously sought medical care. An episode of sickness absence during the year of enrolment increased the risk for future sickness absence (RR = 2.42, 95% CI 1.95—3.00). Having five or seven concurrent symptoms and clinical signs was common and more strongly associated with persistent pain (RR = 1.77, 95% CI 1.39 — 2.27) and future medical care-seeking (RR = 4.51, 95% CI 2.54—9.94), respectively, but not concerning sickness absence.
Conclusion: By simply counting the number of concurrent symptoms and clinical signs, it is possible to predict persistent neck/shoulder pain and future medical care-seeking, but not sickness absence.
