Cynthia Kay1,2,3, Erica Wozniak2, Joanne Bernstein3. 1. Department of Medicine, Clement J. Zablocki VA Medical Center, Milwaukee, Wisconsin. 2. Center for Patient Care and Outcomes Research, Medical College of Wisconsin. 3. Department of Medicine, Medical College of Wisconsin, Milwaukee, Wisconsin, USA.
Abstract
OBJECTIVE: Examine traditional and uncompensated health care utilization associated with chronic noncancer pain. DESIGN: Retrospective chart review. SETTING: Tertiary academic medical center. SUBJECTS: Internal medicine patients on long-term opioids for chronic noncancer pain with or without a pain agreement between April 1, 2014, and April 1, 2015 (N = 834). Patients without pain served as controls (N = 782). METHODS: Univariate statistics were used to compare health care utilization by the presence of chronic pain, pain agreement status, opioid dose, and schedule. Logistic regression was used to assess predictors of health care utilization, with emergency room visits, hospitalizations, and after-hour calls as binary outcomes and office visits and phone/e-mail contacts as ordinal outcomes. RESULTS: Patients with chronic pain used significantly more health care resources compared with patients without pain (all P < 0.001). Patients on a pain agreement had more telephone and secure messages than patients without an agreement ( P = 0.002). Controlling for demographics and other factors, patients with chronic pain had 2.6 (95% confidence interval [CI] = 2.1-3.4) times the odds of an emergency room visit, 5.0 (95% CI = 3.6-7.0) times the odds of a hospitalization, and 2.3 (95% CI = 1.7-3.0) times the odds of an after hour call, compared with nonpain controls. Ordinal logistic regression yield adjusted common odds ratios of 3.4 (95% CI = 2.7-4.2) and 2.9 (95% CI = 2.3-3.6) for total clinic visits and telephone or secure messages, respectively, indicating higher utilization for chronic pain patients. CONCLUSIONS: Patients with chronic noncancer pain utilized more traditional and uncompensated health care resources compared with patients without chronic pain. 2017 American Academy of Pain Medicine. This work is written by US Government employees and is in the public domain in the US.
OBJECTIVE: Examine traditional and uncompensated health care utilization associated with chronic noncancer pain. DESIGN: Retrospective chart review. SETTING: Tertiary academic medical center. SUBJECTS: Internal medicine patients on long-term opioids for chronic noncancer pain with or without a pain agreement between April 1, 2014, and April 1, 2015 (N = 834). Patients without pain served as controls (N = 782). METHODS: Univariate statistics were used to compare health care utilization by the presence of chronic pain, pain agreement status, opioid dose, and schedule. Logistic regression was used to assess predictors of health care utilization, with emergency room visits, hospitalizations, and after-hour calls as binary outcomes and office visits and phone/e-mail contacts as ordinal outcomes. RESULTS: Patients with chronic pain used significantly more health care resources compared with patients without pain (all P < 0.001). Patients on a pain agreement had more telephone and secure messages than patients without an agreement ( P = 0.002). Controlling for demographics and other factors, patients with chronic pain had 2.6 (95% confidence interval [CI] = 2.1-3.4) times the odds of an emergency room visit, 5.0 (95% CI = 3.6-7.0) times the odds of a hospitalization, and 2.3 (95% CI = 1.7-3.0) times the odds of an after hour call, compared with nonpain controls. Ordinal logistic regression yield adjusted common odds ratios of 3.4 (95% CI = 2.7-4.2) and 2.9 (95% CI = 2.3-3.6) for total clinic visits and telephone or secure messages, respectively, indicating higher utilization for chronic pain patients. CONCLUSIONS: Patients with chronic noncancer pain utilized more traditional and uncompensated health care resources compared with patients without chronic pain. 2017 American Academy of Pain Medicine. This work is written by US Government employees and is in the public domain in the US.