Screening for Psychological Distress in Cancer Patients: Challenges and Opportunities
2007; Lippincott Williams & Wilkins; Volume: 25; Issue: 29 Linguagem: Inglês
10.1200/jco.2007.13.1367
ISSN1527-7755
Autores Tópico(s)Childhood Cancer Survivors' Quality of Life
ResumoA number of governmental and professional organizations have recommended that cancer patients be screened routinely for the presence of heightened psychological distress. For example, the National Comprehensive Cancer Network (NCCN) initially published guidelines in 1999, recommending that all patients be screened for distress at their initial visit and at appropriate intervals thereafter. Several arguments can be made for implementation of routine screening for distress. First, evidence suggests that heightened distress is associated with a number of negative outcomes that include poorer adherence to treatment recommendations, worse satisfaction with care, and worse quality of life. Relationships have also been reported between heightened distress (such as greater depressive symptomatology) and poorer survival in people with cancer. Second, heightened distress is treatable. Numerous randomized controlled trials show that psychological distress, including anxiety and depression, can be alleviated by pharmacologic and nonpharmacologic interventions. Third, heightened distress is common. Prevalence estimates derived from large-scale studies typically exceed 30%. Fourth, and perhaps most important, evidence indicates that heightened distress often goes unrecognized by oncology professionals. In one of the largest studies examining this issue, Fallowfield et al evaluated the ability of 143 physicians to detect heightened distress in 2,297 cancer patients. Only 29% of patients with scores exceeding the cutoff on an established screening measure were identified by their physicians as having heightened distress. Despite evidence that heightened distress is under-recognized, clinicians seem reluctant to use screening tools routinely. The format and length of many existing tools may be a barrier, in that the time required for administering, scoring, and interpreting these measures favors use of more informal methods. To address this issue, several ultrashort screening tools have been developed. The most widely studied of these is the Distress Thermometer, a single-item measure that asks respondents to rate their distress during the last week on an 11-point scale ranging from “no distress” (0) to “extreme distress” (10). Although they possess greater potential for routine clinical use than longer measures, the accuracy of these ultrashort tools in detecting heightened distress could be too low to recommend them. The article by Mitchell in the this issue of the Journal of Clinical Oncology represents the first comprehensive evaluation of the accuracy of ultrashort methods in detecting heightened distress in people with cancer. The study is notable for the use of systematic search and pooled analysis methods, similar to those used in a comprehensive evaluation of the results of randomized controlled trials. That is, the investigator used replicable methods to identify all relevant studies and then used quantitative methods to summarize findings across studies. The results of these efforts indicate that ultrashort methods possess both strengths and weaknesses that need to be considered when they are used in the clinic. The principal strength of these methods is their accuracy in ruling out the presence of heightened distress relative to longer methods; the principal weakness is their generally poor accuracy in confirming the presence of heightened distress. An example provided by the author illustrates these qualities. Findings summarized across studies suggest that if 100 people were screened for heightened distress using the Distress Thermometer and the cutoff currently recommended by NCCN, distress would be correctly ruled out (ie, patients identified as not distressed would be classified accurately) in 51 of 60 patients, but correctly confirmed (ie, patients identified as distressed would be classified accurately) in just 22 of 40 patients. Based on these findings, should ultrashort methods be recommended as screening tools? Their poor accuracy in ruling in heightened distress is a liability. If used consistent with NCCN guidelines to decide whether patients should be referred for mental health, social work, or pastoral services, implementation of ultrashort screening measures is likely to result in many patients who are not distressed receiving inappropriate referrals. The primary consequence of this would be the needless use of generally scarce resources available for psychosocial care. The mislabeling of some patients as distressed might also have the unintended consequence of causing distress where it was not present previously. Conversely, the generally good accuracy of ultrashort methods in ruling out heightened distress can be viewed as a major asset that outweighs the previously identified liability. As noted by Mitchell, this feature means that “. . . one simple screen could be used to exclude most cases of distress and mood disorder.” Given that ultrashort methods are unlikely to possess the same accuracy as longer methods, it seems preferable to have brevity come at the expense of making more false-positive errors than false-negative errors. Of the two types of errors, incorrectly identifying a patient as not distressed is likely to have more severe consequences for an individual’s quality of life and quality of care than incorrectly identifying him or her as distressed. JOURNAL OF CLINICAL ONCOLOGY E D I T O R I A L VOLUME 25 NUMBER 29 OCTOBER 1
Referência(s)