Contribution of nonspinal comorbidity to low back pain outcomes

Objective: To determine the involvement of comorbidity to outcomes in a cohort of acute mechanical low back pain patients. Methods: Incident low back pain cases (n = 7077) in the acute or subacute phase assessed between January 1, 1999 and December 31, 2001 were included. Patients were categorized into 1 of 2 groups on the basis of their current medical history: (1) those with at least 1 of 7 medical histories considered (Comorbidity Group, n = 539), or (2) those with only low back pain (Back Pain Group, n = 6538). Main outcome measures were: change in perceived function and visual analog scale (VAS) pain rating from initial assessment to discharge, and total number of treatment days. Results: There were no baseline statistically significant differences in VAS pain rating, questionnaire score, or symptom duration between groups. Odds ratios (ORs) were adjusted to reflect age and sex differences between groups. Logistic regression analysis revealed no statistically significant difference for change in functional score (OR = 1.002) between groups; there were marginal differences in change in VAS pain rating (OR = 1.08) and total number of treatment days (OR = 1.006). chi(2) analysis revealed no statistically significant differences in medication use, global pain rating, or pain control ability posttreatment, between groups. Discussion: Significant ORs were barely greater than 1.00 and were likely the result of the large sample size. The clinical course for comorbid patients, who may seem more complicated at the start of treatment, is just as favorable.