Breaking Down the Barriers of Pain: Part 1
There is widespread recognition that musculoskeletal and chronic pain conditions are poorly diagnosed and treated, leading to chronic disability, risk for drug dependence and depression, dissatisfaction with the medical profession and epidemic costs. The reason is not a lack of sufficiently potent painkillers, nor undiagnosable or untreatable CNS pathology, but rather an over-reliance on analgesic and anesthetic palliation and a failure to make specific source-of-pain diagnoses.
The federal AHCPR (Agency for Health Care Policy and Research) guidelines for low back pain have been the only guidelines widely publicized.1 In the Initial Assessment it is recommended to rule out fracture, tumor, infection or cauda equina syndrome. While one cannot fault such a cautious approach, it nevertheless offers no guidance in diagnosing the other 99 percent of back pain conditions. Recommendations for care for the first month include no guidelines for identifying the source of pain, but rather emphasize palliation. The recommended treatment is limited to acetaminophen or NSAIDs and/or chiropractic. During this period, opioids, muscle relaxants, anti-depressants, physical therapy, modalities and injections are proscribed. No diagnostic testing is recommended. It is implied that the majority of patients will recover in this first month, but the guidelines offer little help for the approximately one-third who are not recovering and typically have significant pain and anxiety.
Care in the second and third month addresses one additional diagnosis: root compression. For patients with sciatica, the guidelines recommend EMG. This is surprising since EMG-NCV studies are not reliable in diagnosing root pathology and cannot diagnose root compression, the proper test for that is an MRI. And yet the guidelines recommend an MRI only if the EMG is positive. Because an MRI will document whether or not there is root compression in the distribution of the patient's sciatica, this test should be performed first. If the MRI is positive, an EMG is not necessary. The guidelines state that only if both tests are positive should a surgical consultation be recommended. The EMG is often negative in radiculopathy and therefore this recommendation is too limiting.
The guidelines recommend surgery if the "patient will consider surgery to speed recovery" and "physical limitations are not lessening." These recommendations are reasonable but carry the risk of permanent residual radiculopathy from on-going nerve root compression that is not decompressed. This risk can be avoided if decompression is recommended for the two to three percent of low back patients whose MRI shows root compression consistent with the radicular symptoms and those symptoms are not diminishing over this time frame. The guidelines for persistence at three months are to "recommend comfort options," "help patient consider options" and if the patient is not "seeking information about options" then address factors why not or arrange for a psychosocial evaluation. These recommendations are reasonable but not realistic. The guideline's three-month evaluation addresses less than five percent of the possible causes of back pain–1) sciatica, 2) fracture, 3) tumor, 4) infection and 5) cauda equina syndrome. It is not likely that a patient will accept a psychiatric referral after so minimal a consideration of physical causes. The guidelines offer no assistance to the clinician as to how one should broach this recommendation with the patient. What is causing the pain? The guidelines imply that if none of the five causes addressed in the first three months are discovered then there is no peripheral source of pain. Several other common sources of pain should be addressed before postulating unproven "central" or psychological theories.
New guidelines, developed by MyoPoint Medical Group, Inc., take a very different approach, based on addressing the question: Where is the pain coming from? MyoPoint is a group comprised of neurologists and occupational and physical therapists specializing in the diagnosis, treatment and management of musculoskeletal and chronic pain conditions. The group provides disease management programs to medical groups and health care insurers.
The MyoPoint program began in 1992 as the Pain Rehabilitation Service of Sharp Healthcare, a multi-hospital, multi-medical group organization, in San Diego, Calif., where MyoPoint maintains its clinic and training facility, as well as its call center and claims analysis services. MyoPoint has developed an algorithm for analyzing insurance claims databases, and a neurological approach to diagnosis, treatment and management of such conditions. The approach is based on determining the source of pain and focusing treatment and management specific to that source.
Incidence and Cost–The Insurance Claims Perspective
Although insurance claims are primarily used for billing, they can provide useful information on the incidence and costs of diagnoses and procedures in a defined population. An insurance claims database essentially groups together one or more ICD-9 (diagnosis) codes, and one or more CPT (evaluation and procedure) codes for a specific date and member (patient). Typically the provider and facility codes are also included. This type of data has significant limitations; the most obvious is it provides no clinical information other than the diagnosis and what was actually performed (or at least billed for). Another limitation is no relation between the reported ICD and the billed CPT. For example, the clinician may enter "low back pain" as the primary reason for the visit, but order tests such as urinalysis or EKG for unrelated symptoms. Thus in determining the true costs of low back pain in a database, and in the absence of asking the clinician directly, it is necessary to include only the CPT codes that are typically ordered to evaluate or monitor musculoskeletal pain conditions. Another limitation is errors in the ICD-9 coding.
To comprehensively analyze the incidence and costs of musculoskeletal and chronic pain conditions, the search algorithm includes more than 225 ICD-9 and 195 CPT codes. The algorithm calculates incidence and costs of each diagnosis and evaluation or treatment, as well as the time course from first presentation to final outcome, and patterns of care. This data can be used to identify patients who are high-utilizers or at risk.
In a previously reported algorithm analysis it was found that pain-related costs were 9.0 percent of all claims costs, and that patients with pain-related claims accounted for 39.69 percent of all claims, demonstrating that these patients are high-utilizers in general, not only for pain-related conditions.2 Table 1 is an analysis of a claims database of 155,000 lives in a major metropolitan area over a two-year period up to October 1999.