Assessing the societal cost of chronic pain

Accurate estimates of the cost of disease are of major importance for scaling health services, prioritizing limited financial and personnel resources, and allocating research funding. Conversely, inaccurate, and biased estimates may inflate the importance of certain diseases at the cost of others, resulting in inequitable allocation of resources and funding.

Disease costs include both direct and indirect financial costs, as well disease burden expressed as years lived with disability (YLD), years of life lost due to premature mortality (YLL) and disability adjusted life years (DALY). These metrics are all dependent on incidence or prevalence estimates. Chronic pain was just recently introduced as a diagnosis in ICD-11 [1], and identifying cases through patient registries is not straight forward. A meta-analysis of epidemiological studies found a pooled prevalence estimate of 31%, which gives a rough benchmark, but huge variation between studies, suggests that prevalence estimates, and the cost estimates derived from these, must be treated with caution [2].

Studies of financial costs have relied on a variety of methods. One approach has been to study highly selected, well-characterized samples. For instance, a study of patients in an Irish pain management clinic, arrived at a cost estimate of $24,043 per patient-year [3]. However, extrapolating this finding is problematic, since patients in pain clinics are likely to have severe chronic pain conditions and the number comparable cases in the general population is not known. A second approach was taken by a Swedish study which defined cases in terms of pain-related diagnoses in health registries [4]. This study found an incremental cost of €6,400 per patient-year, corresponding to about one tenth of GDP. Indirect costs (e.g., sick leave) accounted for 59% of these costs. However, the study identified 840.000 cases out of a total population of 1.5 million (54%), indicating that this method most likely overestimates prevalence and thereby inflates cost estimates. A third approach was taken by a US study, that used survey data to identify cases and estimate costs [5]. This study found that 100 million US citizens (32%) were affected by chronic pain. Estimated incremental costs were $6,349 per patient-year, corresponding to 3.9% of GDP. Indirect costs accounted for 53% of this total. A major limitation with this study was that costs were estimated from questionnaire data, which is likely to be less reliable than health registries and reimbursement databases. A recent Norwegian study addressed this issue by linking questionnaire data on chronic pain to registry data on medical costs and work absence [6]. In this study, 36% of the participants reported having chronic pain. Total incremental costs were €8,211 per patient-year, corresponding to 4% of GDP. Indirect costs accounted for 80% of the total costs.

Disease burden is indexed by the Global Burden of Disease Study, under the auspices of the Institute for Health Metrics and Evaluation at the University of Washington. This study does not include chronic pain as a diagnostic category but includes some conditions where pain is the primary symptom, namely migraine, other headaches, neck pain, and low-back pain. These conditions together account for 22.9% of the total YLD world-wide [7]. By comparison all mental disorders account for 14.4% and cardiovascular diseases account for 8.5%. Though chronic pain has been associated with premature mortality [8], YLL estimates for chronic pain conditions (e.g., migraine) are not included in GBD estimates [9]. DALY estimates, which combine YLL and YLD measures, are likewise not meaingful for chronic pain until the increased mortality associated with chronic pain is factored into YLL estimates.