Description of a clinical stream of back-pain patients based on electronic medical records Mikel Aickin a, * , Anne McCaffery b , Guy Pugh b , Heather Tick a , Cheryl Ritenbaugh a , Paul Hicks a , Kenneth R. Pelletier a, c, d , Jennifer Cao b , Dan Himick b , Jennifer Monahan a Department of Family & Community Medicine, University of Arizona, Tucson, USA b The Marino Centers for Integrative Health, Cambridge and Wellesley, MA, USA c Department of Medicine & College of Public Health, University of Arizona, USA d University of California (UCSF) Schools of Medicine, USA Keywords: Integrative medicine CAM Event-stream Observational research Comparative effectiveness abstract Back pain consists of a spectrum of conditions, with no common etiology and therefore no dominant method of treatment. The purpose of this study is to describe the complexity of a collection of 8000 back pain patients who appeared in an integrative medicine clinic, as a prelude to conducing comparative effectiveness research on CAM alternatives to conventional therapy. Approximately 23% of all clinic patients were diagnosed at some time with back pain. Nearly half had treatment periods of less than one month, while more than 25% were treated for back pain for more than two years. Women were represented more than twice as often as men. The initial diagnosis categories that occurred most frequently were lumbar symptoms, cervical symptoms, and a general category, with smaller numbers having lumbar anatomic, thoracic symptom, brachial neuritis, or sciatica diagnoses. There were few strong relationships between initial diagnosis pattern and length of back pain treatment period. While 77% of back pain patients presented with diagnoses in only a single category, there were many composite categories each of which was sparsely represented. Between 50% and 75% of patients used some CAM service, depending on their initial diagnosis pattern. Patients with complex initial diagnosis patterns strongly tended to chose CAM, and among CAM-users those with complex diagnoses tended toward chiropractic, as opposed to acupuncture or bodywork. The CAM usage patterns of men and women were highly similar. Again among CAM users, 82% used only a single type of CAM service, and multiple service uses tend to be combined at random. Between two-thirds and three-quarters of multiple CAM service users had very simple temporal patterns of use, dominated by use of one type of CAM at a time. Ó 2013 Elsevier Ltd. All rights reserved. 1. Introduction Back pain is one of the conditions most frequently seen by primary care providers, and one which makes a substantial contribution to chronic health care utilization. The research literature on back pain consists primarily of relatively small studies of highly selected patient subgroups, directed toward testing the efficacy of specific therapies. It is unclear whether this research strategy is making much progress in reducing the back pain burden and its clinical costs. In its deliberations on how to use economic stimulus research funding for comparative effectiveness research (CER), the Federal Coordinating Council on CER emphasized the importance of obtaining medical effectiveness information from “real world” situations, as opposed to somewhat artificial clinical trials. It also pointed to the need for integrated electronic medical records as a potential source of such information. 11 There has, however, been little federal funding for the development of methods for extracting valid therapeutic information from clinical data where there are no research interventions. It is therefore not surprising that the literature on medical records-based comparative effec- tiveness studies is quite thin. Back pain seems to be a nearly optimal topic within which such methods can be developed, due to the large fraction of the patient population that presents with back pain, the complexity of the condition, and the relative lack of evidence about the multiple strategies that have been proposed to treat it. 2,7,9,10,12,13,16e19,22,23,25e27,29,30,33,35,38,39,43,45,49,50,52,53,55 * Corresponding author. Tel.: þ1 5208226762. E-mail address: maickin@comcast.net (M. Aickin). Contents lists available at SciVerse ScienceDirect Complementary Therapies in Clinical Practice journal homepage: www.elsevier.com/locate/ctcp 1744-3881/$ e see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ctcp.2013.02.002 Complementary Therapies in Clinical Practice 19 (2013) 158e176