We identified factors protective of all-cause sickness absence (SA) among subjects with multisite musculoskeletal pain (MSP). The nationally representative source sample comprised 3420 actively working Finns aged 30 to 55 in year 2000 and alive at follow-up. Pain in 18 body locations was combined into four sites (neck, low back, upper limbs, and lower limbs). The baseline prevalence of MSP (pain in ≥ 2 sites) was 32%. Baseline data on sociodemographic factors, work ability, work, health, and lifestyle were gathered by questionnaire, interview, and clinical examination and linked with national registers on all-cause SA (periods lasting ≥10 workdays) for 2002 to 2008. Based on trajectory analysis, 74% of those with MSP had a low and 26% a high probability of SA. In logistic regression analysis, younger age, male sex, and professional occupational group were inversely associated with SA. Allowing for these, good physician-assessed work ability, physically light work, possibility to adjust workday length, encouraging workplace atmosphere, no problems with working community or mental stress, normal weight, and no sleep disorders were predictive of lower SA rates (odds ratios between 0.47 and 0.70). In a final stepwise model adjusted for age, sex, and occupational group, no exposure to lifting (odds ratio 0.58, 95% confidence interval 0.39-0.85) and to repetitive hand movements (0.57, 0.39-0.83), possibility to adjust workday length (0.73, 0.53-0.99), and normal weight (0.59, 0.40-0.87) were inversely associated with SA. In conclusion, several modifiable factors related to work and lifestyle were found as predictive of lower rates of longer SA among occupationally active subjects with MSP.