Sharing the Pain

It finally happened. I finally had a chart audit that inspired me to tell others about it.

Most clinicians are accustomed to having their outpatient charts reviewed by a variety of representatives from managed care organizations, hospitals, and peer review organizations. When the medical records of patients with diabetes are reviewed, most surveys focus on several key quality measures. These usually include, as a minimum, the frequency of dilated eye exams and foot inspections, as well as HbA_1c results.

Our office is accustomed to these chart reviews, and our marks are usually very high. A recent chart review found that I was providing good clinical care for my patients with diabetes based on these quality measures. In one area, however, I received a score of zero. It seems that I had not assessed any of my patients for pain. This was the first time that our charts had been reviewed for this specific feature, as well as the first time, in my personal experience, that pain assessment had been given equal importance to the standard benchmarks of diabetes care.

It is never exactly clear which charts are used for random review, but I examined the chart of one patient that I believe had been included in the chart review. In the history section I found the following notation: “Neuropathy - present for 3 years, bilateral lower extremities, epicritic > protopathic, worse at night.” I felt that this was a reasonable assessment of the patient’s neuropathic discomfort, but it didn’t count. I had not used the “P word.”

The source of this new interest in pain stems from a revised set of standards issued by the Joint Commission on Accreditation of Healthcare Organizations (JCAHO). The standards were revised and implemented starting 1 January 2001. They state that screening for pain should be part of the initial assessment, and if pain is present, it should be acknowledged. The complete assessment and treatment of pain may be deferred to the practitioner or organization that is best able to manage the pain appropriately (1). While more accurate assessments of the care that has been provided can be achieved with better training of the medical auditors, I feel there are several specific areas in which implementation of these standards may be problematic for patients with diabetes. Before describing these areas, I need to state that chronic pain is a serious and expensive problem and that the standards from JCAHO are necessary. The goal of better detection and treatment of pain is not debated. It is also assumed here that the majority of chronic pain that can be directly attributed to diabetes is due to diabetic neuropathy.

One area of potential difficulty in implementing the new standards for pain in patients with diabetes lies in the measurement of pain. Significant concerns have been raised about all of the methods of pain self-reporting. In addition, there is no standard mechanism of measuring pain, and multiple methods of measuring pain exist. A small but consistent body of research suggests that patients with chronic pain may not always provide accurate self-reports of the severity of pain, the amount of opiates ingested, or the degree of impaired activity (2,3). Self-reporting of pain is also clearly influenced by the environment. One prospective study of predictors of pain found that psychological distress and job dissatisfaction were the most predictive of severe pain (4). This certainly does not represent all patients, and much of the literature supports the validity and usefulness of self-reported pain. Unfortunately, there is little information regarding diabetes.

In an attempt to create reproducible measures of neuropathic discomfort, numerous techniques have been applied. The clinical utility of any one method (test) can be questioned, however, because neuropathy may affect different types of nerve fibers (e.g., large sensory, small sensory, autonomic) that produce different signs and symptoms. In the Diabetes Control and Complications Trial (DCCT), a variety of different modalities were employed to measure neuropathy, resulting in large differences in the prevalence of neuropathy. For example, depending on what definition was used to define its presence, the prevalence of neuropathy in the conventional therapy cohort of the DCCT varied from 0.3 to 21.8%, representing a 73-fold difference (5). Traditionally, nerve conduction studies have been used as the gold standard for measuring the presence of neuropathy, but unfortunately an abnormal result correlates poorly with the self-reported severity of symptoms and with subsequent progression to ulceration or amputation. Tests to measure the severity of neuropathic discomfort, such as the Total Neuropathy Score, have been developed. They have been standardized and proved to be useful, but they are time consuming and cumbersome and are clearly best used in research settings (6).

Another area of difficulty involves treatment. If one succeeds in identifying pain in a diabetic patient and makes an attempt at measuring it, then what? To my knowledge, there are no pharmaceutical agents available to the clinician that have been specifically approved by the Food and Drug Administration for the management of diabetic neuropathy. We usually start by prescribing improved glycemic control. Numerous trials have confirmed an association between hyperglycemia and nerve disfunction, but the association starts to fall apart when the severity of neuropathic symptoms is the end point. There is a wealth of information concerning various pharmaceutical agents for the treatment of pain, such as tricyclic antidepressants, opiates, sodium channel blockers (e.g., mexiletine), antiseizure agents (e.g., gabapentin), and topical agents (e.g., capsaicin). Unfortunately, for most patients the treatment of painful neuropathy is a trial-and-error gamble, utilizing pharmaceutical agents without clear indications for the problem that they are being used to treat.

Finally, one of the biggest problems in adhering to the new guidelines is their apparent deficiency in predicting future health or happiness in patients with diabetes. It has been argued that “pain is a vital sign.” There is no question that untreated or undertreated pain of any kind, including diabetic neuropathy, is a useful predictor of future disability, hospitalization, and personal dissatisfaction. When it comes to diabetic neuropathy, however, the one thing worse than pain is the absence of pain. Anesthesia of the extremities can cause people to unknowingly injure themselves, and it is one of the better predictors of amputation. Patients with cardiac autonomic neuropathy due to diabetes frequently suffer silent ischemia, and they may have a life-threatening myocardial infarction with only minimal symptoms. When it comes to diabetic neuropathy, it seems that the true “vital sign” is actually an absence of pain.

As already stated, the pain guidelines from JCAHO are necessary and are an important step toward resolving a significant problem. There may be some difficulty in the utilization for the patient with diabetes, however, that could lead to confusion and frustration for the patient and the provider. Chronic pain is a serious problem, but it is also complex, difficult to classify and measure, and challenging to treat. Diabetic neuropathy may be viewed as the last jungle in diabetes complications, and clearly future research is needed. From the data available, it is hard to justify placing pain on the same level of importance as HbA_1c, blood pressure, or lipid level.