Clinical and financial ramifications of undiagnosed depression in SNFs
Dr. Robert Figlerski, Team Health
If I were to tell you that a virus was spreading throughout the world and more than 300 million people were infected — roughly the population of the United States — you would probably be alarmed. What are the symptoms? How is it transmitted? Is there a treatment?
Now what would you say if I told you the World Health Organization (WHO) announced this malady is already the leading cause of illness and disability in the world? Essentially, they just did.
The illness is depression, and on March 30 the WHO announced its latest estimate that more than 300 million people worldwide are living with it. The organization called the news “a wake-up call to rethink our approach to mental health and treat it with the urgency it deserves.”
Some estimates indicate up to 60% of skilled nursing residents exhibit some type of depressive symptoms. These numbers should actually not be surprising. Individuals admitted to a skilled nursing facility have just experienced a significant medical setback, which has left them, either temporarily or permanently, physically and emotionally vulnerable. In addition to their acute and chronic medical conditions, other stressors such as the uncertainty of their future, physical decline, multiple transitions and living in a challenging environment can all take an emotional toll. Undetected and untreated, depression can have devastating effects on a resident's quality of life and the outcome of their care.
The Centers for Medicare & Medicaid Services has been addressing the impact of depression over the years and have implemented a variety of approaches to assess and report depression in skilled nursing facilities. The current method of assessment and reporting involves Section D of the Minimum Data Set (MDS) 3.0, CMS's standard resident assessment tool.
Section D reports the results of the Patient Health Questionnaire-9, a commonly employed screening tool for depression. The PHQ-9 can be administered either through a resident interview or the PHQ-9 OV (staff interview) for residents determined to be unable to be interviewed. CMS, recognizing the importance of identifying residents with depressive symptoms, provides for higher rates of reimbursement based on PHQ-9 scores. The RUGS- IV patient classification system for determining reimbursement, takes into account depression scores in three categories: Special Care High, Special Care Low and Clinically Complex.
While this process has helped standardize the approach for assessing and reporting depression, there are significant drawbacks that limit the reliability and accuracy of the assessment and reporting process. The PHQ-9 is a self-report measure, making it a limited assessment tool because residents may distort their responses in reaction to the perceived demand characteristics of the situation and/or shape their responses to influence the perceptions of the interviewer. For example, think about common questions that occur during a regular medical check-up like, “What do you weigh?”, “How much do you exercise?” and “How much do you drink?” These are all self-report questions, the answers to which are often unreliable. Physicians can even be hard-pressed to get an honest answer to a simple question like, “Do you take your medication regularly?”
This type of problem shouldn't be a surprise. Individuals are not passive participants in any process, but rather have their own motivations and interests in shaping the outcome of an interaction. The individual can be motivated to “fake good” by minimizing symptoms to achieve an outcome or “fake bad” by exaggerating symptoms, often times for attention seeking or to manipulate staff. For example, a resident who does not want to be at a nursing home may minimize symptoms, since, in their mind, reporting depression is just another issue that may prevent them from returning home.
When a physical illness or disability is involved, concrete measures, like lab tests, are often available to verify a resident's status. In contrast, the way depression reporting is now structured, the resident's report is all that is considered, regardless of diagnosis, care plan, treatment, and behavioral documentation. If a resident reported they did not have a fever or their blood sugar was high, would we just accept that or choose to look for a more independent verification? In the skilled nursing setting there are multiple forms of assessment, observations, and documentation to verify a resident's self-report of mood. The MDS 3.0 reporting process marginalizes this critical information.
The PHQ-9 was developed as a screening tool to better identify depression in the community, and while those efforts are of significant value, residents of a skilled nursing facility are already part of a clinical population. These patients are a high-risk, vulnerable population, many of whom are already exhibiting a wide range of depressive symptoms. Perhaps a more robust procedure for assessing their mood state should be utilized, as opposed to an interview structured around a screening tool.
I have personally treated many residents who were clearly depressed, had been diagnosed with depression and were receiving psychological services in conjunction with antidepressants, but who denied depression and failed to score as depressed when assessed using the PHQ-9. How does a facility reconcile an active treatment plan for depression, but report a score that indicates the resident is not depressed? As a result of these challenges, CMS is getting bad information and facilities are losing reimbursement for important services they are providing to a vulnerable population. Depending on the RUGS-IV classification, the loss of reimbursement to a facility could range from approximately $750 to $2,200 a month.
Even if the PHQ-9 continues to be used, there are ways of adjusting the current guidelines for assessing depression that could significantly improve accuracy. First, the criteria to determine which residents are interviewed or assessed by the staff (PHQ-9OV—Staff Assessment) needs to be amended. Currently, residents' ability to make themselves understood (Section B, B0700) is rated by a staff member. This process means many residents who are not ideally suited for the interview are able to participate, despite having a limited communication range to fully engage in a discussion involving multiple self-reflections about emotional/behavioral states over an extended period of time. When a person's communication range falls short of what is necessary for an interview, they tend to keep their responses brief and resist more detailed responses. The person's motivation devolves into ending the interview as quickly as possible.
A simple adjustment could improve the interview process would be to revise the criteria for who qualifies for the interview process from its current standard to one that places more residents into the Staff Assessment category (PHQ-9OV). In addition, a resident's Ability to Understand Others (Section B, B0800) should likewise be taken into account. It makes sense that the interviewee should be able to reasonably understand the instructions and concepts needed to successfully complete the PHQ-9 interview.
Another area where the process could be improved is to incorporate the findings of the Brief Interview for Mental Status (BIMS, Section C) in the determination of who qualifies to be interviewed or who should be assessed by staff. The PHQ-9 interview is constructed around repetitive complex instructions involving temporal awareness and rating frequency of symptoms (For example, “Over the last two weeks have you been bothered by any of the following problems?”). The BIMS assesses key components necessary for adequately responding to the PHQ-9 interview, but the results of the BIMS are not taken into account when determining who is interviewed. If a resident is not temporally oriented, how is it possible for them to accurately report feelings and experiences over the last two weeks? Simply allowing the staff to take into account the BIMS score when deciding whether a resident should be interviewed or a staff assessment of mood should be conducted would go a long way to enhancing the accuracy of depression scores being reported.
Overall, the effort to standardize the assessment of depression is a great step forward. However, our current practices need to be adjusted, and staff should be provided more latitude about how to best assess their residents. The PHQ-9 and the PHQ-9OV can provide a structure for the focus of the assessment, but the process for who is interviewed or assessed by staff needs to be reconsidered, so there is greater confidence in the scores being reported. Adjusting the criteria for who is interviewed, taking into account a resident's BIMS score, and finding ways to include existing clinical diagnoses and documentation would achieve significant improvement in assessing, reporting, and ultimately, treating and monitoring depression in the skilled nursing setting.
Robert W. Figlerski, PhD, is the Director of Behavioral Health Services, New York Region, for TeamHealth.