RETURN TO TABLE OF CONTENTS
Spring 2002

CFS Research: The Need for Better Standards

By Nancy Klimas, MD,
University of Miami

The scientific literature on chronic fatigue syndrome as an entity begins in 1988 with the first case definition that coined the controversial term “CFS” and established a set of criteria for researchers to use in selecting study subjects. Over the past 14 years, the field has grown in both the number of researchers and disciplines represented. Yet for all that’s been learned since 1988, much remains unknown about this enigmatic illness. Each research finding seems to raise more questions than it answers.

The bulk of the blame, of course, lies with the vagaries of CFS itself. But the research effort also is hampered by poorly conceived, constantly changing — even non-existent — standards. Tighter research methodology could improve individual studies and enable greater comparability of research findings across studies. What follows is an overview of the areas most in need of clarification and consensus, and a summary of the efforts to achieve overarching agreement among CFS researchers.

Case definition
Authors of the 1988 case definition set out to identify a group of patients sharing similar symptoms and clinical signs, but problems using the definition quickly became apparent. A revision in 1994 by an international consensus group (see back cover) attempted to address some of the difficulties, but the resulting guidelines are rife with ambiguity and vagueness. Symptoms are counted either as present or absent, without regard to severity or frequency. Terms such as “substantial reduction” and “not lifelong” are subject to varied interpretation. The use by some groups of outdated case criteria developed in England and Australia obscures comparability as well.

Thus, the case definition itself contributes to the heterogeneity of the patient population, creating a shaky foundation for CFS research. The U.S. Centers for Disease Control and Prevention (CDC) has convened an international group of CFS investigators to develop a new case definition. The group has a paper in press that outlines ways in which the case definition could be strengthened.

Subgrouping
Another weakness in CFS research stems from the lack of information about illness course and long-term effects. Patients are often enrolled in studies solely on the basis of currently meeting CFS criteria, without regard to the length of their illness, which could account for differences in biological markers (e.g., cell subsets, viral titers) and functional status. Researchers have discussed the need for “staging criteria” as a means of grouping patients by length and severity of illness, but none have been established. Variations among study participants in age at onset, present age, co-morbid conditions, menstrual cycle, body weight, medication use, deconditioning, diurnal rhythms and other factors may also limit study conclusions.

Control groups
The selection of controls is a particularly tricky element of study design. If the study hypothesis calls for “healthy” or “normal” controls, recruitment techniques can present unintended biases. For instance, healthy controls recruited from a health club near the investigator’s site would probably differ from control subjects obtained from an urban hospital clinic or patients’ families. Unrecognized medical issues in “healthy controls” may taint study results. In some studies, the researcher may seek to compare CFS subjects with people with other illnesses — if so, defining appropriate disease groups can be problematic, in part because the symptoms of CFS overlap with so many conditions. Selecting disease control subjects who have fibromyalgia, multiple sclerosis, migraine, orthostatic intolerance or depression may make scientific sense, but investigators must consider the potential for overlap, as individual CFS study subjects may meet criteria for one or more of the comparison conditions. Although the study hypothesis may dictate the type of control subjects to use, it is important that recruitment methods and selection criteria be thoroughly described, so that reviewers fully understand group definition, and investigators hoping to duplicate the study conditions are capable of performing confirmatory studies.

Lab methods
Laboratory studies of cell subsets, cytokines, immune activation, hormones, etc. often suffer from the lack of agreed-upon protocols for sample collection, handling and storage. At a recent CFS research conference, one reviewer noted that clinical samples are often “treated more carelessly than last night’s leftovers.” Variability in lab processing standards can also contribute to differences in test results. Academic research laboratories are typically held to a high standard in establishing methodologies and quality control measures not necessarily seen in lower cost send-out laboratories. Yet published papers in CFS that have been peer reviewed often neglect to disclose the site of the laboratory or quality control measures used in determining the results of the study. Similarly, differing procedures and conditions for tilt table tests, sleep studies and exercise tolerance may account for discrepancies in the literature. Standardization of procedures is a lofty goal that has evaded most disciplines, but better documentation of protocols in research reports is an important first step.

Outcome measures
In 1992, the federal Food and Drug Administration convened an advisory meeting on outcome measures used in CFS research. Participants tried to agree on standard measures of improvement in treatment studies. No consensus was reached, and the field still lacks a widely endorsed instrument by which to judge clinical improvement. This leaves the decision to individual investigators, some of who rely on simplistic self-report questionnaires (e.g., “Do you feel better than you did before the study began?”), while others conduct exhaustive batteries that evaluate progress across numerous domains (e.g., sleep, pain, fatigue, memory, self-care, leisure activity). As part of its effort to revise the case definition, CDC is evaluating numerous instruments and piloting use of a panel developed for CFS studies.