Suicidal Ideation and Risk Levels Among Primary Care Patients With Uncomplicated Depression

Herbert C. Schulberg; Pamela W. Lee; Martha L. Bruce; Patrick J. Raue; Jean J. Lefever; John W. Williams; Allen J. Dietrich; Paul A. Nutting

doi:10.1370/afm.377

Published eLetters

If you would like to comment on this article, click on Submit a Response to This article, below. We welcome your input.

Submit a Response to This Article

Compose eLetter

Title *

Author Information

Contributors
First Name and Middle Initial * First or given name, e.g. 'Peter'. Last Name * Your last, or family, name, e.g. 'MacMoody'. Email Address * Your email address, e.g. higgs-boson@gmail.com Occupation * Your role and/or occupation, e.g. 'Orthopedic Surgeon'. Affiliation * Your organization or institution (if applicable), e.g. 'Royal Free Hospital'.

Statement of Competing Interests

Competing interests? *

Yes

Please describe the competing interests

Vertical Tabs

Jump to comment:

The PHQ-9 and treatment implications
Herbert C. Schulberg, Ph.D.

Published on: 07 October 2008
Use of PHQ-9 for diagnosis of Depression
Ian M. Bennett

Published on: 27 September 2008
Suicidality and comorbid psychiatric conditions
Louise S. Acheson

Published on: 13 December 2005
More Inclusive Suicide Risk Assessment
Philip H I Lawson

Published on: 07 December 2005

Published on: (7 October 2008)
Page navigation anchor for The PHQ-9 and treatment implications
The PHQ-9 and treatment implications
Herbert C. Schulberg, Ph.D., White Plains, NY, USA
Other Contributors:
Dr. Bennett raises the meaningful issue of whether psychiatric assessment instruments such as the PHQ-9 which was developed for screening purposes, i.e. identifying patients at high risk for such disorders, can also serve diagnostic purposes, i.e. designating patients as true cases for whom treatment is indicated. Dr. Bennett correctly notes that confounding true and false positives may ensue when screening instruments a...
Show More

Dr. Bennett raises the meaningful issue of whether psychiatric assessment instruments such as the PHQ-9 which was developed for screening purposes, i.e. identifying patients at high risk for such disorders, can also serve diagnostic purposes, i.e. designating patients as true cases for whom treatment is indicated. Dr. Bennett correctly notes that confounding true and false positives may ensue when screening instruments are used inappropriately and can lead to the unnecessary prescribing of antidepressant drugs.
We acknowledge that our overly brief published paper could lead readers of the Annals incorrectly to conclude that the PHQ-9 alone generated the diagnosis of major depressive disorder or dysthymia in our study. We, therefore, wish to inform readers that our protocol followed the MacArthur Tool Kit principle which emphasizes that a high PHQ-9 score should lead physicians and investigators to clinically corroborate that a depressive syndrome is present. Thus, the primary care physicians in our study had already prescribed antidepressants for 90% of the study’s subjects based on such a clinical determination. The study’s Research Assistants similarly conducted clinical interviews to verify the presence of current symptoms consistent with the diagnosis of major depression and/or dysthymia. With regard to Dr. Bennett’s concern that we nevertheless possibly treated “false positive” cases, it is important to note that study subjects had to score at least 0.5 on the Hopkins Symptom Checklist-20 to be considered clinically eligible for the trial. In point of fact, the total sample’s baseline mean HSCL-20 score was 2.01 (see Table 1), which indicates that the subjects’ depressive severity was indeed of the moderate to severe type and for which treatment was clearly indicated.
With regard to Dr. Bennett’s more general concern about the validity of using the PHQ-9 alone as a diagnostic instrument in clinical practice or in other clinical trials, the reader might wish to consider the data reported earlier by Kroenke, et al (2001). They found that when a Mental Health Professional’s Validation Interview served as the criterion standard, a PHQ-9 score of 10 or greater had a sensitivity of 88% and a specificity of 88% for major depression. While these data might tempt the busy physician to intervene on the basis of a PHQ-9 score alone, we nevertheless continue to recommend clinical corroboration of this psychometric finding with a clinical interview which typically is structured upon the patient’s very responses to the PHQ-9.
Herbert C. Schulberg, Ph.D., Martha L. Bruce, Ph.D., Patrick J. Raue, Ph.D. Weill Cornell Medical College, White Plains, N.Y.
Kroenke K, Spitzer R, Williams J. The PHQ-9. Validity of a brief depression severity measure. J. Gen Intern Med 2001:16: 606-613.
Competing interests: None declared
Show Less

Competing Interests: None declared.
Published on: (27 September 2008)
Page navigation anchor for Use of PHQ-9 for diagnosis of Depression
Use of PHQ-9 for diagnosis of Depression
Ian M. Bennett, Philadelphia, PA
I was interested in the use of the PHQ-9 in this paper for identifying current depression. It is my impression that this instrument is meant to identify risk of depression (screening instrument) rather than provide a clinical diagnosis. In the methods section of this paper it seems to indicate that it was used to diagnose. This is a critical issue as I have come across many clinicians who are using the PHQ-9 as a diagn...
Show More

I was interested in the use of the PHQ-9 in this paper for identifying current depression. It is my impression that this instrument is meant to identify risk of depression (screening instrument) rather than provide a clinical diagnosis. In the methods section of this paper it seems to indicate that it was used to diagnose. This is a critical issue as I have come across many clinicians who are using the PHQ-9 as a diagnostic measure and initiating treatment based on the outcome. Given the high rate of inappropriate antidepressant use (patients do not have DSM-IV criteria for Major Depressive Disorder) it is important that these measures not be used incorrectly. Could the authors comment on their take on the use of this measure as part of the identification and diagnosis of depression? Thanks
Competing interests: None declared
Show Less

Competing Interests: None declared.
Published on: (13 December 2005)
Page navigation anchor for Suicidality and comorbid psychiatric conditions
Suicidality and comorbid psychiatric conditions
Louise S. Acheson, Cleveland, OH, USA
Schulberg et al found low rates of active suicidal ideation among primary care patients referred to a study of depression management who were determined to have “uncomplicated” major depression or dysthymia. Depression was defined as ‘complicated’ if participants met criteria for schizophrenia, bipolar disorder, PTSD, alcohol or substance abuse. As the authors acknowledge in their Discussion, excluding patients with the...
Show More

Schulberg et al found low rates of active suicidal ideation among primary care patients referred to a study of depression management who were determined to have “uncomplicated” major depression or dysthymia. Depression was defined as ‘complicated’ if participants met criteria for schizophrenia, bipolar disorder, PTSD, alcohol or substance abuse. As the authors acknowledge in their Discussion, excluding patients with these comorbid psychiatric conditions excludes a subset likely to be at higher suicide risk. In my experience, such patients are frequently cared for in primary care practice, and do not always end up in the care of psychiatrists.
Therefore it may be of interest to compare population-based data on the prevalence of suicidal ideation, plans, gestures, and attempts recently published from the National Comorbidity Survey (NCS) of English- speaking U.S. adults age 18-54.[Kessler RC, Berglund P, Borges G, Nock M, Wang PS. Trends in suicidal ideation, plans, gestures, and attempts in the United States, 1990-1992 to 2001-2003. JAMA 2005; 293:2487-2495.] Overall, during the preceding 12 months (in 2001-2003), 3% of adults sampled had suicidal ideation, 1% had a plan, 0.2% had made a suicide gesture, and 0.6% had attempted suicide and survived. Of those with any suicidal ideation, 61% met DSM-IV criteria for an anxiety disorder, including 20% with PTSD; 47% had major depression or dysthymia; 22% met criteria for bipolar disorders; 28% had impulse-control disorders such as ADHD; and 19% had a substance use disorder. Based on these population- based data, it appears possible that the eligibility criteria for the study by Schulberg et al. of uncomplicated depression in primary care exclude the majority of people with suicidal thoughts, a) because they are not depressed (39% in the NCS), b) because of comorbid conditions (61% of those with suicidal ideation in the NCS had PTSD, bipolar disorder, or substance abuse; 16% of depressed patients in Schulberg’s study were excluded for one of these conditions), or c) because they did not visit the family practice. This resonates with experience, in that not all emotionally-distressed or existentially challenged patients whose struggles I witness have a diagnosable mood disorder; many have substance abuse and are not under a mental health professional’s care, and many are anxious.
In the NCS, 59 people (out of 4320 sampled) had survived a suicide gesture or attempt in the past year. More than ¾ of the attempters met diagnostic criteria for a mood disorder, but 27% of these had bipolar disorders and might require specialized treatment. Thirty percent of people who attempted suicide had PTSD and 29% a substance use disorder. Those with substance abuse/ dependence were more likely to have survived a gesture than a serious suicide attempt. One third of people in the NCS who said they had had a plan for suicide during the past year had attempted suicide; so had 10% of those who had suicidal ideation without a plan. The prevalence of suicide gestures and attempts, at each stratum of risk, was not lower among those who were receiving treatment than those who were not.
These data speak to the importance of inquiring about suicidality whenever patients are emotionally distressed, whether or not they are depressed. While screening for mood disorders will identify the majority of suicide attempters, more research should encompass suicidality of people who are not depressed. The data also suggest that quite a few serious suicide attempts are impulsive and unplanned. And finally, the NCS data do not provide evidence that current treatments for mental health disorders are effective for preventing suicide attempts. Perhaps it is not only that we do not treat depression intensively enough, but also that so many people with comorbid conditions require a different approach.
Competing interests: None declared
Show Less

Competing Interests: None declared.
Published on: (7 December 2005)
Page navigation anchor for More Inclusive Suicide Risk Assessment
More Inclusive Suicide Risk Assessment
Philip H I Lawson, Bethlehem, NH, USA
As a recent patient of mine commited suicide 2 weeks ago, I was interested in Schulberg et al's article on prevalence of suicide risk in the primary care population. The often stated 10% figure (for risk in psychiatric patients) seemed somewhat high from my experience.
We follow a very similar process of care for treatment of depression at our facility (2 question screen, use of the PHQ-9, structured suicide...
Show More

As a recent patient of mine commited suicide 2 weeks ago, I was interested in Schulberg et al's article on prevalence of suicide risk in the primary care population. The often stated 10% figure (for risk in psychiatric patients) seemed somewhat high from my experience.
We follow a very similar process of care for treatment of depression at our facility (2 question screen, use of the PHQ-9, structured suicide assessment...) Unfortunately the low prevalence stated (@1-3%) glosses over the huge percentage of excluded patients due to lack of being "uncomplicated" (PTSD, alcohol or substance use, bipolar or schizophrenia, immediate high risk of suicide in one practice, and those not agreeing to take part (47%)). It is not a simple matter to make these diagnoses and to "exclude" these patients into a higher risk group as we assess presenting psychiatric complaints. I would be much more interested in the suicide risk including these patients, as that is more representative of what I see day to day.
Competing interests: None declared
Show Less

Competing Interests: None declared.