The epidemiology of cancer pain.

We identified 28 epidemiological surveys by applying a sensitive search in Medline and CancerLit databases, supplemented by hand searches. Only two surveys enrolled more than 10,000 patients with cancer. The remaining were hospital or clinic-based surveys of at most a few hundred patients. Fourteen surveys were conducted in the United States. The majority of the remaining studies were conducted in Europe (Finland, France, Germany, UK/Ireland). No single survey identified a prevalence of any type of pain below 14%. The prevalence of pain reported in these surveys varies with the specific type of pain (e.g., breakthrough pain) and/or population studied. Based on these surveys an aggregate statement could not be deduced regarding the correlation between the occurrence of pain and patient factors, disease characteristics, the setting in which care is provided (e.g., primary care or specialized oncology or pain treatment clinics), or specific treatments directed towards the underlying disease and its associate...