Although depression is one of the most common mental health disorders in the United States (U.S.) only 8.6 million (65.5%) received mental health treatment among the 13.1 million U.S. adults with serious mental illness in 2019.2 Additionally, 70 percent of people who die of suicide visit a doctor in the month before their death and about 40% see a doctor within one week of their death. Many practices, including ours, have adopted the nine-item Patient Health Questionnaire (PHQ-9) as an easy, validated, evidence- based tool for screening patients who might have depression. Despite the implementation of validated screening tools for depression such as the PHQ-9 questionnaire, it is our opinion that reliance on these tools as the “end all, be all” of depression screening—without genuine discussion and inquiry—could lead to over- and underdiagnosis, particularly among vulnerable populations.
To be clear, the validity of the questionnaire across multiple populations is not, in our opinion, the main issue. The problem comes when we (1) stop using the questionnaire as a tool to aid in a diagnosis, but instead use it as our only diagnostic tool; and (2) assume that it has been validated across all patient types and populations. To that end, we deeply caution national quality reporting metrics and healthcare systems from relying on the PHQ-9 as the quintessential tool for which it was never designed to be.
In our clinic, the PHQ-9, a self-administered patient questionnaire, is administered in the exam room while the patient waits to see their doctor. This setup offers a quiet, private setting for patients, but may have under recognized shortcomings. Was this questionnaire designed for the patients we serve at our safety net clinic? A screening questionnaire completed with all “no answers” marked should serve as a red flag for patients with limited English proficiency or low health literacy, those with cultural and/or individual concerns related to stigma about mental health, patients whose caregivers respond on their behalf because the patient has a chronic disability, or simply forgot their glasses, with limited vision, etc. Myriad other scenarios are possible, all of which could lead to false negatives causing possible depression to go undiagnosed. Further, because depression plays into a patient’s ability to manage other chronic conditions, the impact of underdiagnoses can have broad implications for the mental and physical well-being of affected patients.
PHQ-9 scores may also be elevated, for example, among those with conditions such as chronic pain affecting their desire to participate in fun activities, or among those who experiencing social distress related to homelessness, severe grief from loss of a child, food insecurity and social isolation, or domestic violence. The PHQ-9 questionnaire could lead to ‘false positives’ for major depression in our most vulnerable populations—people who seem depressed in their answers to the questions, but who are really just dealing with life’s many challenges.
Algorithmic Treatment based on PHQ-9
Primary care settings, national reporting metrics, and health systems use the PHQ-9 as their main quality measure for managing depression. Patients with positive scores are put into categories of mild, moderate, or severe depression based on their total score. Once a patient screens positive, their score is then monitored at each visit, to ensure that score is improving. From a national quality reporting standpoint, significant reductions in the total score indicates that the health professional is appropriately treating the depression. If health professionals don’t achieve depression remission in the form of “better” PHQ-9 scores, they may receive lower reimbursements. This places physicians in the difficult position of “improving” their patient’s moderate and severe depression with medications, in ways that often overlook cause.
Psychosocial factors, such as trauma, housing, education, and income, impact health; in fact, the U.S. Centers for Disease Control and Prevention estimates these factors make up 70% of a persons’ physical and emotional well-being.1 If a patient is depressed because a relative died, they have no housing, or because they have a substance use disorder, behavioral therapy may help with coping skills, but this does not resolve the underlying issues. Using the PHQ-9 as our mainstay is seriously problematic because it promotes a system that encourages health professionals to rely on numbers and circled answers instead of listening to patients and understanding the background, context, and etiology of their symptoms. We need time to diagnose situational depression, complicated grief, comorbid substance use disorders, and bipolar disorder. To the degree that the score stimulates such conversations and inquiries to occur, it may prove useful. If we fail to explore these factors, we ignore our patients.
The challenges of our healthcare system make this type of dialogue difficult at best. Ethnic minorities, in particular, often encounter additional barriers to treatment, including mental illness stigma, lack of insurance or underinsurance, lack of culturally competent care, and inadequate mental health services in safety net settings. Assuming such barriers can be overcome, how can we then juggle—in a time-constrained clinic visit—five chronic conditions, yet still find time to have meaningful conversations lending insight into what’s really on patients’ minds? Issues with time likely also underlie the fact that while mental health disorders are the third leading cause of disability in the United States,2 less than half of patients receive any pharmacotherapy or psychotherapy treatment.3, 4
A Way Forward
Given these many barriers, delivering care for mental health problems in primary care offices, with the assistance of embedded counselors, psychologists, and/or psychiatrists, has become an alternative model of care.5 As primary care physicians, we are grateful for the integrated behavioral health services recently implemented in our safety net hospital primary care center. When our patients become tearful—from, for example, grief or stress—we are someone they can go to and talk to in real time. Our behavioral health team is keenly aware of the additional resources available and give us real-time feedback on management plans—whether it’s cognitive behavioral therapy, group support, goal setting and action planning, or linkage to community resources and social worker support.
To the extent that the PHQ-9 has increased focus and enthusiasm around implementing integrated behavioral health, we are grateful. We look forward to continued funding and reimbursement of longitudinal integrated behavioral health across healthcare systems. Particularly important are capitated payment models that truly assess and reimburse based on social risk; such payment models might allow behavioral counselors, licensed psychologists, and psychiatrists the time needed for patients to feel heard. However, it is important to note that health care alone does not have the resources or capacity to provide psychosocial solutions for all our patients’ needs. In truth, reducing depression rates starts long before the doctor ever sees a patient—it lies in part within the social constructs, pooling of resources, and political decisions made by local, state, and federal governments. To that end, we embrace community-engaged approaches, government advocacy efforts, and public health collaborations focused on improving the many social determinants that afflict patients and building resilience/empowerment within vulnerable communities, all of which can promote mental health and well-being.