Does Documenting a Dementia Diagnosis Improve Outcomes?

With the aging of the U.S. population, and in the absence of major breakthroughs in prevention and treatment, the numbers of Americans with dementia will dramatically increase over the coming decades [1]. Identifying opportunities to improve the care of this growing population is of paramount importance.

As reported in this issue of the Journal of the American Geriatrics Society, Burgdorf and colleagues studied the relationship between awareness of a dementia diagnosis by a home health clinician and subsequent care delivery and outcomes in these patients [2]. The research team used a complicated approach to investigate this issue in a large national sample of Medicare skilled home health patients. First, they identified home health patients with a Medicare claim associated with an ICD-10 code related to Alzheimer's disease or related/other dementias (ADRD) prior to home health admission and defined these patients as having dementia. Of this population, those with an ADRD ICD-10 code in their Outcomes and Assessment Information Set (OASIS) assessment were characterized as having “documented dementia.” The remaining home health patients with dementia, but who did not have an ADRD ICD-10 code in their OASIS assessment, were considered to have “undocumented dementia.” The investigators assumed that undocumented dementia reflected a lack of recognition by the home health clinician, which in turn would negatively impact care delivery and outcomes.

True enough, those patients characterized as having undocumented dementia had an increased risk of hospitalization and ED use, and a lower likelihood of discharge to self-care. The authors concluded that dementia documentation could be “protective against negative outcomes,” and that when home health agencies are aware of a dementia diagnosis, they more effectively tailor care. According to the authors, “documentation of a dementia diagnosis may indicate that clinicians are both aware of the patient's dementia status and have the willingness and capacity to incorporate this information in care planning.” Additionally, they stated that “there is likely an underlying set of constructs influencing both documentation of dementia and subsequent care delivery and outcomes, including elements such as staff training and experience related to dementia care, home health agency resources and culture regarding tailored care for those with cognitive impairment, clinical information available to staff upon referral, and even community level resources to address dementia-specific needs.”

Are these conclusions plausible? Maybe, but maybe not.

In many ways, the investigators may have been comparing apples to oranges. In this study, those with documented dementia, when compared with those having undocumented dementia, differed substantially across a range of important characteristics. Those with undocumented dementia were younger, had lower cognitive symptom severity, and less impairment in activities of daily living. However, compared with patients who had documented dementia, undocumented dementia patients were more likely to have received institutional acute or post-acute care preceding admission to home health, or to be living alone. In addition, they had higher overall clinical severity, were 30% more likely to have diabetes, and 50% more likely to have heart failure or COPD. While multivariable analytic techniques were employed to reduce bias, extensive baseline differences can make it challenging to “control away” differences. This is especially true when the severity of important individual comorbidities cannot be accurately characterized, or when the overall burden of multimorbidity is not assessed.

The reality is that what we most need is not evidence that awareness of a dementia diagnosis is helpful, but evidence on what best supports patients and caregivers after the diagnosis is made. That is where our real challenge lies. Unfortunately, randomized trials of dementia care programs have shown mixed results [3-5]. Some interventions have resulted in improved quality of life, reduced behavioral symptoms, and eased caregiver burden, but others have not. In addition, their impact on hospitalizations and ED use remains uncertain.

There is an urgent need to develop more effective, comprehensive dementia care programs—ones that strengthen care planning, improve management of comorbid conditions, enhance medication safety, reduce avoidable ED visits and hospitalizations, better align medical decisions with patients' values and goals (“what matters most”), and address caregiver burden.

Timely and accurate communication of a dementia diagnosis during referrals and care transitions is important. However, it is likely naïve to assume that merely documenting or sharing a diagnosis, by itself, will lead to better care or outcomes. If only it were that simple.

Dr. Gurwitz prepared the manuscript.

Dr. Gurwitz serves as a consultant to United Healthcare.

This publication is linked to a related article by Burgdorf et al. To view this article, visit https://doi.org/10.1111/jgs.19491.