The overall population impact of chronic pain on work performance has been underestimated as it has often been described in terms of work-related absence, excluding more subtle effects that chronic pain may have on the ability to work effectively. Additionally, most studies have focussed on occupational and/or patient cohorts and treatment seeking, rather than sampling from the general population. We undertook a population-based random digit dialling computer-assisted telephone survey with participants randomly selected within households in order to measure the impact of chronic pain on work performance. In addition, we measured the association between pain-related disability and litigation. The study took place in Northern Sydney Health Area, a geographically defined urban area of New South Wales, Australia, and included 484 adults aged 18 or over with chronic pain. The response rate was 73.4%. Working with pain was more common (on an average 83.8 days in 6 months) than lost work days due to pain (4.5 days) among chronic pain participants in full-time or part-time employment. When both lost work days and reduced-effectiveness work days were summed, an average of 16.4 lost work day equivalents occurred in a 6-month period, approximately three times the average number of lost work days. In multiple logistic regression modelling with pain-related disability as the dependent variable, past or present pain-related litigation had the strongest association (odds ratio (OR)=3.59, P=0.001). In conclusion, chronic pain had a larger impact on work performance than has previously been recognised, related to reduced performance while working with pain. A significant proportion were able to work effectively with pain, suggesting that complete relief of pain may not be an essential therapeutic target. Litigation (principally work-related) for chronic pain was strongly associated with higher levels of pain-related disability, even after taking into account other factors associated with poor functional outcomes.