Gruschkus, S.K., Lairson, D., Dunn, J.K., Risser, J., & Du, X.L. (2010). Comparative effectiveness of white blood cell growth factors on neutropenia, infection, and survival in older people with non-Hodgkin's lymphoma treated with chemotherapy. Journal of the American Geriatrics Society, 58, 1885–1895.

To examine the effect of colony-stimulating factors (CSFs) used as primary and secondary prophylaxis on incidence of febrile neutropenia, infection, and survival in older adults.

The study used data from the Surveillance, Epidemiology, and End Results (SEER) Medicare database. ICD-9 codes were used to define inclusion diagnoses and definition of febrile neutropenia. 

Use of CSF, the type of chemotherapy administered, and the use of antibiotics were defined in terms of CPT codes. Regression analysis was used to analyze effects of primary and secondary CSF prophylaxis on outcomes of interest. Primary prophylaxis was defined as CSF during chemotherapy before occurrence of fever, infection, or neutropenia. Secondary prophylaxis was CSF administration that occurred after these events. Study used data from 1992–2002.

• Total cases numbered 13,203, with 5,266 receiving any CSF.
• Mean age was 74.9 years, with a range of 65–102.
• Women made up 53% of the sample; men made up 47%.
• All patients had non-Hodgkin lymphoma (NHL). 
• 62% had low comorbidity burden, as defined by ICD-9 coding.
• 44% had large B-cell lymphoma.
• 18% had follicular lymphoma. 
• 89.2% were Caucasian. 
• All socioeconomic groups were essentially equally represented.
   

Multi-site (SEER data)

There were mutliple phases of care

Application is for elder care

Retrospective cohort study

• Chemotherapy administration defined as ICD-9 code 9925; CPT codes 96400-96549, J9000-9999, and Q0083-0085; revenue codes 0331, 0332, and 0335; and ICD-9 V codes, k V58.1, V66.2, and V67.2.
• Use of CSF was identified by CPT codes J1440, J1441, and J2820.
• Febrile neutropenia was defined as the combination of neutropenia (ICD-9 288.0) and fever present (780.6).

Sixty percent of cases did not receive any CSF. Those who had 5–9 CSF administrations for primary prophylaxis has a 42% lower risk of febrile neutropenia (OR = 0.58, 95% confidence interval [CI] [0.41, 0.83]) and those with 10 or more administrations had a 48% lower risk after data were adjusted for age, marital status, stage, and other characteristics. Those with 5–9 administrations had a 27% lower incidence of infection, and those with 10 or more administrations had a 52% lower risk (OR = 0.48, 95% CI [0.35, 0.66]). Primary prophylactic CSF was not associated with longer overall survival. Secondary prophylaxis was associated with improved overall survival, with a strong dose-response effect. A range of 11–23 administrations was associated with a 23% lower risk of mortality (HR = 0.77, 95% CI [0.71, 0.84]), and those with more than 23 administrations had a 13% lower risk of mortality than others. Protective effects of primary prophylaxis was highest in those receiving the largest number of chemotherapy administrations and in those with large B-cell lymphoma.

Primary prophylaxis with CSF in older adults is effective in preventing febrile neutropenia and infection, but was not associated with improved survival. Secondary prophylaxis was associated with longer overall survival.

• Risk of bias (sample characteristics)
• Measurement validity/reliability questionable
• Findings not generalizable
• Measurement validity/reliability questionable
• The sample was limited to patients with NHL from 1992–2002, and findings would not necessarily be applicable to patients with other types of disease and those receiving newer and other treatment regimens. 
• Although data were stratified for classes of chemotherapeutic drugs there was no specific subgroup analysis based on different types of chemotherapy or other treatments.
• Reliability of medical records coding is across organizations is known to be questionable, so findings should be considered with this in mind.
• Definition of neutropenia by code, which depends upon physician documentation and does not specify actual lab results as criteria can be questionable.
• A higher proportion of patients who got primary prophylaxis had higher stage disease—it is not stated as to whether there were statistically significant differences in this factor or not and those who received CSF were more likely to have received radiation therapy as well.   
• Although data were said to be stratified for age, it is not clear if the factor of combined therapy was considered in analysis.

Findings support the use of primary prophylactic CSF for prevention of infection and febrile neutropenia, and secondary prophylaxis in improving survival in this group of patients. Limitation of retrospective statistical analysis using medical records code data only need to be considered in interpretation and application of these results.