In this week’s guest post, researchers from the Meadows Mental Health Policy Institute (Meadows Institute) share the results of their analysis examining trends in death from suicide and associated policy opportunities to more accurately classify deaths related to mental illness.

One year ago, co-author Jennifer’s grandfather (hereafter, “Grandpa Decker”) broke his hip after slipping on ice in the parking lot of a Pennsylvania candy store. Being a stubborn man, he made the 3-hour drive home to New Jersey, where he laid in bed for days before calling his daughter to help. After seeing the severity of his injuries, his family immediately called 911. Grandpa Decker was transported to the hospital, and hip repair surgery was performed the next day.

Despite the successful surgery, his hospitalization continued. The doctors said that his blood pressure was “too high” to permit discharge, yet he had no history of cardiac issues. Prostate cancer, yes. But no high cholesterol, no high blood pressure, nothing. Five days post-surgery, Grandpa Decker passed away from sudden cardiac arrest. The physicians in charge of his care at the hospital had no reasonable explanation for his death, and his primary care provider was devastated – reiterating that he had no history of cardiac problems. The medical examiner had questions, too. In fact, he called Grandpa Decker’s daughter to review his medical history to help certify the death. What was the ultimate cause of death? We’ll revisit this later in the post.

Mental illness is one of the most common chronic diseases in America, with one in five adults living with a mental health condition.[1] Suicide is the ultimate, most severe consequence of under-treatment or ineffective treatment of mental health conditions.[2] Although previously described as “rare,” suicide now breaches the top 10 causes of death for all age groups from 10-64, with nearly 1 in 5 deaths among young people ages 10- 24 being caused by suicide.[3] According to the Centers for Disease Control and Prevention, 49,369 Americans died from suicide in 2022 - the highest number recorded since the CDC began routinely collecting mortality data in 1999.[4,5]

Rates of death from suicide are highest among adults ages 18-64. Between 2020 and 2022, the rate of death from suicide among older adults (ages 65+) accelerated, nearing the rate of deaths from suicide among adults under 65. This disturbing trend is often masked in the “Top 10 Causes of Death” (and similar summaries) because older age brings an increasing likelihood of death from other chronic diseases, like cancer or heart disease. This increasing rate of death from suicide among older adults is not always widely reported and deserves attention.

The Meadows Institute’s Research on Mortality Trends

The trends described above are troubling, but even more alarming is that our research suggests that the data reported above underestimate the actual number of deaths due to mental illness (and related substance use disorders – jointly ‘behavioral health conditions’). Mental health symptoms or substance use, especially when untreated, can cause death from physical health conditions like heart disease and cancer.[6] In our research, we found more than 70,000 death certificates listed a behavioral health condition as contributing to death from a physical health problem (or any cause besides suicide or overdose).[7] This under-reporting of all contributing causes limits our ability to measure the true impact of behavioral health conditions on death.[8]

This calls for an improvement in how deaths related to behavioral health conditions are classified on death certificates. An accurate representation of the magnitude of this problem (and examination of trends to identify who is affected and where) is essential to identify hotspots of these preventable deaths early for intervention deployment.

Here is where Grandpa Decker’s death becomes a useful case study. The only cause of death recorded on his death certificate was “complications of right femur fracture”; the remainder of the “Cause of Death” section was blank. Consequently, his death certificate presents only a snapshot of the cascade of death. There was no mention of cardiac arrest, hypertension, or any heart condition. Nor was there any mention of osteoporosis, arthritis, extreme frailty, or challenges he experienced standing shortly after the surgery – a practice recommended almost universally for patients following hip surgery.[9] Could this, or some other routine hospital procedures, have caused Grandpa Decker’s pulmonary arrest? Unfortunately, epidemiologists will not be able to answer this question and change practice guidelines unless all contributing causes of death are recorded on death certificates.

Policy Recommendations

There are hundreds of reasons that limit death certifiers’ ability to accurately report the circumstances surrounding a death on death certificates. Still, federal agencies, policymakers, and legislators can take action to remove the primary barriers. To enable death certifiers to present a complete picture of the cause of death (including behavioral health conditions when they played a role), we suggest the following actions:

1. Immediately, the CDC should provide examples of how behavioral health conditions can be reported in the instructions for completing Part II (“Contributing Causes of Death”) of death certificates.[10]
2. Legislators should enable the expansion of, and reimbursement for,[11] the collaborative care model to transform how mental health conditions are treated.[12] This major shift in care provision requires short-term investments to enable practices and large healthcare systems to adopt this care model. If undertaken, these investments would be returned by improving the quality of care, thus reducing behavioral health-related deaths; and, enabling the integration of health and behavioral health care records. Co-locating these records together will improve death certifiers’ likelihood of accessing and using them jointly to report the circumstances surrounding the cause of death more fully.   ​​​​​​
3. We also suggest that Congress and federal agencies such as the CDC work to expand the qualifications of entities able to access near real-time, individual-level surveillance data. This information has been exclusively accessible to a small group of [largely] underfunded, “covered entities,” which has prevented its use to proactively detect spikes in behavioral health-related death.