Hospitalization at the End of Life in Myeloma Patients: Lessons from the National Inpatient Sample

INTRODUCTION:

Multiple myeloma (MM) remains a largely incurable disease, and despite the variety of treatment options available, duration of response decreases with each subsequent line of therapy resulting in refractory disease . In this setting, studies have shown most patients prefer to die in the comfort of home, yet hospitalizations remain frequent at the end of life. We explored the hospitalization burden of MM patients at the end of their life using the National Inpatient Sample (NIS).

METHODS:

The NIS is a database that provides information on all inpatient hospitalizations in the United States (US), including primary and secondary diagnoses, procedures, length of stay, and disposition. Approximately 20% of admissions are tracked and weighted estimates are provided regarding the total number of hospitalizations.

Using the NIS, we tracked hospital admissions for MM patients and inpatient mortality from 2002 to 2014 via procedural International Classification of Disease (ICD) 9 codes to gain insight into trends in transfusions, infectious complications, and cost of admission. Linear regression modeling was used for analysis. Overall annual number of deaths for MM in the United States was obtained from publicly available reports from the Centers for Disease Control (CDC) and Prevention and the National Cancer Institute (NCI).

RESULTS:

During the time period 2002-2014, the CDC and NCI reported a total of 144,105 deaths from MM, ranging from 10,913 in 2002 to 12,112 in 2014. The NIS identified a total of 233,932 (non-weighted) hospitalizations for MM during this time period. Amongst these, a total of 14,770 (non-weighted) hospitalizations resulted in death, thus 6.3% of all hospitalizations for myeloma patients resulted in death. A weighted sample of 69,825 hospitalizations resulting in deaths were identified. During our study time period, 48.4% of all deaths related to myeloma in the United States occurred in the hospital, ranging from 5,893 (54%) in 2002 to 5,035 (41.6%) in 2014, p<0.01.

We analyzed blood transfusion dependency in the hospitalization leading to death. There was a receipt of blood transfusions (35.8%) in 5,285 of the 14,770 (non-weighted) admissions leading to death.

Infection frequency was identified using the Clinical Classification Software. The Clinical Classifications Software (CCS) is a tool that allows for clustering patient diagnoses and procedures into clinically meaningful categories. A total of 6,644 infections were identified amongst the 14,770 (non-weighted) hospitalizations leading to death (45.0%).

We then analyzed palliative care/hospice involvement during the hospitalization leading to death over time. Palliative care/hospice was consulted in 67 of the 1260 (non-weighted) hospitalizations in 2002 (5.3%), and 338 out of the 1007 (non-weighted) hospitalizations in 2014 (33.57%), p<0.01.

Median cost of the hospitalization leading to death increased over time from $48,709 in 2002 to $104,115 in 2014, p<0.01.

CONCLUSIONS:

Despite a decrease in the percentage of inpatient deaths over time, greater than 40% of patients with myeloma continue to die in the hospital, with significant transfusion requirements and infections at the end of life. This comes with an increased cost to the health care system. Our analysis suggests that while palliative care involvement at the end of life has also increased over time, earlier involvement of palliative care and incorporation of transfusion support within hospice services may decrease the number of myeloma patients dying in the hospital and, therefore, the overall burden and cost of care.