Trends in neutropenia-related inpatient events

Abstract

Purpose: Neutropenic complications (NCs) after myelosuppressive chemotherapy are associated with significant morbidity and mortality. We described NC rates by using US hospital discharge data.

Materials and methods: This cross-sectional analysis used data from the US National Inpatient Sample database. Hospital discharges with cancer diagnoses (International Classification of Diseases, Ninth Revision, Clinical Modification [ICD-9-CM] code) from 1989 to 2007 were analyzed for the ICD-9-CM neutropenia code. NC rates per 10,000 discharges were calculated for all adult discharges without radiation therapy (study population, all cancers); lung cancer, breast cancer, and non-Hodgkin's lymphoma (NHL); and all three combined. The use of growth factors and myelosuppressive chemotherapy from 1994 to 2008 was estimated by using the IMS Health Drug Distribution Database.

Results: Estimated lung cancer and breast cancer discharges remained relatively steady, whereas NHL discharges increased. NC rates for each study cancer increased two-fold until the late 1990s before stabilizing and/or declining. The average hospital stay for all three cancers decreased from 10.4 days to 7.1 days. The mortality rates for NCs for the three cancers combined decreased at a fairly constant rate from 10% in 1989 to 5.4% in 2007. Estimated discharges for NCs from 1989 to 2007 ranged from 111,000 to 169,000 for the study population, from 57,000 to 103,000 for all cancers, and from 21,000 to 40,000 for the three study cancers. The use of growth factors and myelosuppressive chemotherapy increased from 1994 to 2008.

Conclusion: Whereas the number of hospitalizations with cancer diagnoses has remained steady since 1989, hospitalizations for NCs increased approximately two-fold from 1989 to 1997 and then stabilized.

Figure 1.

Absolute number and rate per 10,000 hospital discharges with neutropenic complications (left panels) and absolute number of hospital discharges (International Classification of Diseases, Ninth Revision, Clinical Modification code: 288; right panels) for (A) lung cancer, (B) breast cancer, (C) non-Hodgkin's lymphoma (NHL), and (D) the three cancers combined, from the Nationwide Inpatient Sample (1989-2007).

Figure 2.

(A) Sex, (B) age, (C) length of stay, and (D) mortality as a discharge status among hospital discharges with neutropenic complications (International Classification of Diseases, Ninth Revision, Clinical Modification code: 288) for lung cancer, breast cancer, and non-Hodgkin's lymphoma (NHL), by individual cancer type and for all three types combined, from the Nationwide Inpatient Sample (1989-2007).

Figure 3.

Estimated relative changes in the use of filgrastim/pegfilgrastim and myelosuppressive chemotherapy. The IMS Health Drug Distribution Database (1994-2008) and dosing data were used to calculate the number of days receiving filgrastim/pegfilgrastim on the basis of WHO-defined daily doses. The numbers of doses of chemotherapeutic agents were calculated by using recommended dosing data from the package inserts.