| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
EDITORIAL COMMENT |
Division of General Internal Medicine, Center for Research on Health Care (B.L.R.), and Department of Psychiatry (M.K.S.), University of Pittsburgh School of Medicine, Pittsburgh, PA 15123, Email: rollmanbl@msx.upmc.edu
According to recent figures from the Centers for Disease Control, suicide is responsible for over 29,000 deaths annually in the US, twice the number of HIV/AIDS deaths, and is the eighth leading cause of death among men (1). Up to two-thirds of the patients who commit suicide have seen a physician in the month before their death. Yet, although one-third of individuals report suicidal ideation at some point in their lives, few spontaneously report suicidal thoughts to their primary care physicians (PCPs) and PCPs, in turn, seldom detect these symptoms among their patients (2). Importantly, many patients, particularly women, who attempt suicide do so with medications prescribed by physicians who are unaware of their risk of suicidality (3).
Although certain actively suicidal patients are readily identified, it is important to identify clinical signals that can potentially assist PCPs in recognizing those at elevated risk. Not surprisingly, psychiatric illness has been identified as the strongest predictor of suicidality that is present among 80 to 90% of suicide completers on psychological autopsy (4). Of these various conditions, major depression is the most commonly identified psychiatric illness (5). Yet, because it is present among 6 to 10% of primary care patients (6), it is a fairly nonspecific risk factor. There remains a need to further enhance PCPs’ ability to accurately distinguish the patients at elevated risk of suicide from the larger number who present for care with emotional and behavioral problems.
Goodwin and colleagues’ (7) report in this issue of the Journal adds to our growing appreciation of both mental illness and comorbid physical illness as independent risk factors for suicidality. They examined data from 3000 primary care patients who completed the PRIME-MD Patient Health Questionnaire (PHQ) (8) case-finding instrument that classified respondents as experiencing either major depression (9.7%), anxiety (7.4%), or probable alcohol abuse or dependence (6.7%). PCPs were instructed to complete a brief checklist to identify such conditions as hypertension (prevalence 24.6%), cardiac problems (7.9%), diabetes (7.8%), and pulmonary disease (7.1%) among others. They assessed suicidality with the question "Have you had thoughts that you would be better off dead or of hurting yourself in some way for at least several days in the last 2 weeks?" Nine percent responded affirmatively; being younger; unmarried; less educated; and having depression, anxiety, or a probable alcohol use disorder were all associated with a higher rate of suicidality. Yet, the main finding is that suicidality was more common among those with pulmonary disease (11.5% vs. 6.7%, p = .003) even after adjustment for a variety of sociodemographic characteristics (70–90% increased risk). Although the investigators also found suicidality less common among those with hypertension in univariate analysis (18.6% vs. 25.2%, p = .02), this relationship disappeared in multivariate analysis, and patients experiencing one of the other six common medical conditions surveyed for did not experience an altered prevalence of suicidality.
The association of pulmonary disease with suicidal ideation is of interest. Investigators have identified an association between medical illness and suicide (9–12). One potential explanation is the two- to three-fold increased prevalence of depression among patients with heart and lung conditions, diabetes, and end-stage renal disease, among other common medical disorders, compared with typical primary care patients for causes that remain to be elucidated (9). Indeed, the combination of depression and pulmonary disease conferred nearly a 10-fold risk for suicidal ideation. The authors point out that the finding of pulmonary disease as an independent predictor of suicidality is consistent with a prior report in a community sample and with numerous observations of an association between pulmonary disease and depression. Yet, in contrast to the cited community report (9) or a recent Scandinavian study (12), they did not identify malignant disease as a predictor. This could have been due to its low prevalence in their primary care sample (2.3%). Moreover, other than hypertension, the conditions assessed on the checklist completed by PCPs were low in prevalence (eg. end-stage renal disease 1.1%). Thus their analyses may have been underpowered to detect a significant difference. Moreover, the investigators’ did not examine whether there was a correlation between the number of comorbid medical conditions or the severity of patients’ medical comorbidity and the risk of suicidal ideation. Still, the observation of an association of suicidality with pulmonary disease seems to be valid. The frequent co-occurrence of panic with pulmonary disorder is a possible explanation that is not mentioned by the authors (13). Panic disorder alone has been associated with suicidality and significantly increases the risk of suicide among patients with depression (14). Thus, in addition to recognizing and treating depression in medical settings, it is also important to identify and treat patients with panic disorder.
Few clinical trials have addressed the issue of treating depression among primary care patients with comorbid medical illness, and we are aware of just a single clinical trial specifically designed to prevent suicide by means of treating depression (15, 16). While awaiting further details from these studies, what are the practical implications of Goodman and colleagues’ report? This paper serves to remind PCPs of the association between mental illness and suicidal ideation and draws attention to the fact that suicidal ideation may occur in the presence of medical illness without depression. The 10-fold increase in suicide risk among patients with both depression and pulmonary disease confirms that clinicians not overlook the diagnosis of a mental disorder, as they so often do (17–19), particularly when a chronic physical illness is present. Because these patients are likely to visit their PCPs repeatedly, clinicians have multiple opportunities to detect suicidal ideation and to ensure these patients receive needed treatment. Although we recognize that PCPs can be reluctant to inquire about suicidal thoughts out of the belief that the question may exacerbate suicidality and thereby precipitate suicidal acts, data do not support this concern (20). Information from well-designed clinical trials remains necessary to confirm whether a strategy of targeted screening for suicidality among patients with "red flag" conditions and state-of-the-art care for mood (21–23) and anxiety disorders (24, 25) can improve clinical outcomes.
REFERENCES
This article has been cited by other articles:
 |
|
 |
|
  C. A. CLAASSEN and G. L. LARKIN Occult suicidality in an emergency department population The British Journal of Psychiatry, April 1, 2005; 186(4): 352 - 353. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Thompson, V. Henkel, J. C. Coyne, R. D. Goodwin, and K. Kroenke SUICIDAL IDEATION IN PRIMARY CARE: ASK A VAGUE QUESTION, GET A CONFUSING ANSWER * Response Psychosom Med, May 1, 2004; 66(3): 455 - 457. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
