Medicare Claim Processors' Reimbursement and G-CSF Choice Among Non-Hodgkin Lymphoma Patients

Publication
Article
AJPB® Translating Evidence-Based Research Into Value-Based Decisions®July/August 2013
Volume 5
Issue 4

Using the SEER-Medicare database, we found negative association between Medicare claim processors“related physician's chemotherapy reimbursement and their prescription of granulocyte colony-stimulating factor for non-Hodgkin lymphoma patients.

Recent health policy efforts have targeted healthcare financing to reduce the escalating healthcare costs and improve patient care.1 One of the hallmarks of the healthcare reform is to provide patient-centric high quality care through innovative payment initiatives. For example, the Centers for Medicare & Medicaid Services (CMS) has proposed “bundled” payments for acute and post-acute care.2,3 Use of reimbursement policies to contain costs is not new. In a simulation analysis Welch suggested that bundling payments may not increase the financial risk to hospitals.4 However, the effect of reimbursement policies on services provided by physicians through augmentation of standard regimens with supplemental services has not been examined.

Medicare payments of chemotherapy for cancer patients are determined by the local Medicare claim processing agents. Such discretion often results in different payment polices across these claim processors. The claim processors often have the option of making a single, prospective bundled payment for all items and services provided, or making separate payments to the physician for each of the drugs and services delivered.5 As reimbursement amounts to physicians are linked to their profits,6 differences in reimbursement amounts can affect physicians’ income and can lead to variations in physician practices including cancer care.

However, to date there has been no research examining the association between Medicare claim processor reimbursement policies and physicians’ prescriptions of supplemental products among cancer patients. Due to lack of published literature in this area, we extrapolate the effect of differences in reimbursement policies on cancer care using studies that have analyzed the association between fee changes over time and the response of physicians in increasing or decreasing the volume of services. Reimbursement changes (ie, reduction in Medicare payments) on physician behavior have been investigated in many areas: cesarean delivery, major joint repair/replacement procedures, coronary artery bypass grafting, and physician consultation services.7-10 In these studies, no conclusive evidence was found in terms of linking reduction in Medicare payments to increased volume of services provided to the patients. Physician’s response to reduction in Medicare payments for surgical procedures including cataract, hip repair, hip replacement, knee replacement, and joint procedures was dependent on the type of procedures and physician specialties.8 In addition, physician’s response was dependent on what proportion of the income was affected by the fee change.7,11-15 If one applies these findings to assess the relationship between reimbursement policies and prescribing volume, one can conclude there is considerable uncertainty and there is a critical need for studies in this area.

In this context, potential variations in Medicare claim processors’ reimbursement policies for cancer care, specifically chemotherapy, can be considered a “natural experiment” that allows us to examine the association between reimbursement amounts and physicians’ prescribing practices. The perspective of the current study is different from studies that examined changes in Medicare fees on physicians’ volume of services for which fee changes were enacted. Rather, we focus on the relationship between reimbursement policies and change in provision of a supplemental product or service because physicians may respond to variations in reimbursement policies by offering a different mix of services including supplemental products or services.16

Therefore, the primary objective of the current study is to analyze the relationship between reimbursement policies and prescription of supplemental products among patients with cancer. For the purposes of this article, we consider use of granulocyte colony-stimulating factor (G-CSF) as the supplemental product during an episode of cancer care. We selected cancer patients with non-Hodgkin lymphoma (NHL) for several reasons. The main modality of treatment for patients with NHL is chemotherapy. Chemotherapy regimens are sometimes augmented with G-CSF as a prophylaxis to prevent febrile neutropenia,17 a life-threatening condition with low white blood cells and high fever18 among cancer patients on chemotherapy.19 In addition, G-CSF is listed as one of the top 10 most expensive Medicare Part B drugs, with average reimbursement amounts for 11 days of supply ranging from $2136 to $3442.20 Thus, augmenting chemotherapy with G-CSF may provide increased revenue for the physicians.

METHODSStudy Design

We used a retrospective longitudinal design using linked cancer registry data from 2002 through 2007 and Medicare claims data from 2001 through 2008. Patients were followed for the fi rst 5 months after diagnosis of NHL to define the fi rst course of chemotherapy.21

Data

The data for the current study come from the Surveillance, Epidemiology, and End Results (SEER)-Medicare linked database. The SEER data contain 13 cancer registries with demographic information (age, gender, race, county of residence), area-level socioeconomic status, Medicare enrollment status, and clinical information (diagnosis date, histology, stage, lymph node involvement, and grade). The current study uses the “Patient Entitlement and Diagnosis Summary File” (PEDSF), a customized file of SEER data. SEER data were linked with Medicare fi les to allow us to analyze Medicare-covered healthcare services.

Study Population

Patients With NHL. Cancer registry data were used to identify patients over age 66 years and diagnosed with NHL anytime between 2002 and 2007 using primary diagnosis (N = 33,017). Other exclusion criteria were: patients without a reliable death date (N = 45), not continuously enrolled in Medicare Part A and/or Part B, enrolled in Medicare HMO (N = 9908), cancer site was any part of the central nervous system (N = 2322), did not have the fi rst course of chemotherapy and hospitalized or had emergency department (ED) visits during the fi rst cycle of chemotherapy (N = 13,366), and patients who had missing information of area socioeconomics, node, stage, and histology type (N = 127). Thus, the fi nal study population consisted of 7249 patients with NHL.

Key Variable: Medicare Claim Processors. This variable was derived using information on patients’ county of residence within SEER programs and Medicare claim processors recorded in Medicare Part B claims. For each county, we used the most frequently billed Medicare claim processor. When 2 or more Medicare claim processors served a particular county we had a combination code. For example, the highest percentage of physician visit claims in Johnson County, Iowa, were submitted to the Medicare claim processor Iowa Wellmark Inc, and the highest percentage of outpatient claims in Johnson county were submitted to Nebraska Blue Cross. Therefore, for Johnson County, we combined both claim processors: Iowa Wellmark and Nebraska Blue Cross. Because claim processors serving a county might change across years (eg, some claim processors were terminated in a particular year), the claim processors were assigned annually to each county. After grouping of claim processors by each county, there were 108 Medicare claim processor groups in our study population.

Determination of Reimbursement Policies. Reimbursement policies were identifi ed in 2 steps using the average physician reimbursement amount for the first cycle of chemotherapy and the number of chemotherapy administration codes. In the fi rst step, for each patient, we defined the first cycle of chemotherapy as a fixed period of 21 days following the initiation of chemotherapy. Wechose 21 days as the cutoff period because most chemotherapy regimens last for 21 days.22 Chemotherapy was identified from the outpatient and physician visit claims. Healthcare Common Procedures Classifi cation System (HCPCS) codes (J8999-J9999, Q0083, Q0084, Q0085, J7150, 964XX, and 965XX) and revenue center codes (0331, 0332, and 0335) were used to identify chemotherapy regimens and administration procedures. Only for year 2005, the following codes were also used to identify chemotherapy claims: 51720, G0355, G0356, G0357, G0358, G0359, G0360, G0361, G0362, and G0363.23 In this step, total physician reimbursement amount derived from outpatient and physician visit claims during the first cycle of chemotherapy was used to define reimbursement policies. As our study period covered multiple years, total physician reimbursement amount was converted to 2008 dollars based on the Consumer Price Index for medical services published by the bureau of labor statistics.24 Medicare payments also include the Geographical Practice Cost Index (GPCI) component and refl ect costof-living adjustments based on geographical region. To standardize the different price levels for healthcare inputs across location and time, we divided the reimbursement amount by the GPCI from Federal Register. We estimated average reimbursement amount for each Medicare claim processor after controlling for clinical and non-clinical factors using regression techniques.

In the second step, the average estimated reimbursement amount per claim processor computed from the regression was grouped into top and bottom deciles. Medicare claim processors in these deciles were further examined for chemotherapy administration codes to determine single bundled payment and separate payments for services and drugs.

Prescription of G-CSF. We constructed an indicator variable representing presence or absence of any G-CSF prescription during the fi rst cycle of chemotherapy (ie, within 21 days following the initiation of chemotherapy). G-CSF drug was identified from outpatient and physician visits based on HCPCS codes (J1440, J1441, Q0453, S0135, and J2505).

Other Independent Variables. Demographic variables were: patient’s age at diagnosis (66-70, 71-75, 76-80, 81-84, >85 years), gender (female/male), race (white, African American, other). Patients’ clinical characteristics were: histology type, stage of cancer, lymph node involvement, prior-diagnosis inpatient and outpatient comorbidities, and grade. We used both physician and inpatient claims to identify comorbid conditions 1 year prior to NHL diagnosis. Using methodology developed by Klabunde et al,25 comorbidity index was computed using non-cancer Charlson condition26 from inpatient claims and physician or outpatient claims. Our comorbidity index may be limited in scope because it did not include all co-occurring conditions present in an individual. Area-level socioeconomic variables such as income level, education level, and percentage of white were used as independent variables as well.

Type of chemotherapy (anthracycline-based chemotherapy [ABC] with rituximab, ABC without rituximab, rituximab with other non-ABC chemotherapy, other non-ABC chemotherapy only, and rituximab only) was used as an independent variable because reimbursement amounts may vary by type of chemotherapy. Site of visit (outpatient vs office-based physician visits) was also distinguished and used as one of the independent variables. Year of first chemotherapy treatment was included as a time trend variable to account for differences in treatment patterns over time.

Statistical Techniques. Reimbursement policies were measured using estimates from an ordinary least square regression (OLS) model. The dependent variable was total physician reimbursement amount derived from outpatient and physician visit claims during the first cycle of chemotherapy. Indicator variables for each of the 108 Medicare claim process or groups were the key explanatory variables in the model. The reference group for Medicare claim processor variables was the claim processor group with the lowest estimated chemotherapy reimbursement amount. In addition to Medicare claim processor variables, this regression also included patient-level demographic and clinical factors and arealevel socioeconomic status listed in the measures section. Adjusted differences in average total reimbursement amounts by Medicare claim processor groups were tested by Chow F-statistics. The average estimated chemotherapy reimbursement amount by claim processor group was computed using the regression coefficients and intercept.

To evaluate the association between G-CSF prescription and average estimated reimbursement amounts, we conducted logistic regressions. In the first model, we included only patient-level demographic, clinical factors, and area socioeconomic status to describe the association between these variables and G-CSF prescription. In the second model, we additionally included policy-level variable (ie, average chemotherapy reimbursement levels by claim processor groups estimated from OLS regression).

RESULTS

The study population consisted of 7249 patients; 38.45% had prophylactic G-CSF prescription during the first cycle of chemotherapy (data not presented in tabular form). The mean age was 76 years. An overwhelming majority of patients were white (92%). Fifty-one percent were females. The most common histology type at diagnosis was diffuse large cell lymphoma (DLC) (43%). A majority of patients were diagnosed with B symptom and others (54%) with stage I (26%), stage II (17%), stage III (18%), and stage IV (32%). Approximately 40% had at least 1 comorbid condition within the 12-month period before diagnosis. Forty-fi ve percent received ABC with rituximab and 9% patients received ABC without rituximab.

Determination of Reimbursement Policies

Table 1 describes the average estimated reimbursement amounts across all 108 Medicare claim processor groups and also estimated reimbursement amount at the patient level. The estimated average reimbursement amount at the patient level was $6239 and median was $6014. For ease of presentation, in this table we also grouped average reimbursement amounts in 4 categories: 1) less than $5669; 2) $5669 to $6901; 3) $6901 to $8094; and 4) greater than $8094. The top panel of the table presents the number and percentage of Medicare claim processor groups at different estimated average reimbursement levels.

The results from the OLS regressions on reimbursement amounts at patient level are summarized in the bottom panel of Table 1. F-test from the regression indicated that there were statistically significant variations in the estimated chemotherapy reimbursement amount across Medicare claim processor groups (P <.0001).

To identify reimbursement policies of the claim processors (bundled vs separate payments), we also examined chemotherapy administration codes across different Medicare claim processor groups. The average number of chemotherapy administration codes used in the first cycle of chemotherapy was 5 in top deciles, while the average number of chemotherapy administration codes was 3 in bottom deciles. We further identifi ed administration codes in top deciles and not in bottom deciles: Current Procedural Terminology (CPT) code 96408. The code indicates “more than once per day for each drug the oncologist provides.” Administration codes in bottom decile and not in top decile included CPT code 96523, indicating “only payable when not billed with other service.” Therefore, we speculated that bundling policy is more likely to be the source of variations in estimated chemotherapy reimbursement. It is plausible that Medicare claim processors with low estimated average reimbursement amounts were using bundled payments and those with high estimated average reimbursement amounts were using separatepayments for services and drugs delivered.

Association Between Reimbursement Policies and G-CSF Prescription. The adjusted odds ratios (AORs) and the associated 95% confidence intervals (CIs) from 2 logistic regressions are presented in Table 2. In the first model (without average chemotherapy reimbursement variable), the results showed that chemotherapy type, race, age group at diagnosis, histology, and time-trend were signifi cantly associated with G-CSF prescription. Results from the second model (with average chemotherapy reimbursement variable) revealed that while patient-level factors continued to have a statistically signifi cant association with G-CSF prescription, the average chemotherapy reimbursement level variable also contributed to G-CSF prescription. For ease of interpretation, estimated chemotherapy reimbursement amount from the OLS regression was divided by $1000. After controlling for patients’ demographic, clinical factors, and area socioeconomic status, we found that, for every $1000 increase in estimated physician reimbursement amounts, the likelihood of G-CSF prescription decreased. The AOR was 0.91 with 95% CI (0.85-0.97). If the reimbursement amounts were to be expressed in units of $100, this result would translate to a 0.9% decrease in G-CSF prescription likelihood for an average increase of $100 in chemotherapy reimbursement amounts.

DISCUSSION

Our study set out to examine the relationship between Medicare claim processors’ reimbursement policies and G-CSF prescription among patients with NHL. Reimbursement policies of Medicare claim processors were identified with an indirect approach using average estimated reimbursement amount for the fi rst cycle of chemotherapy (21 days) after initiation and the number of drugs and services delivered during this period. We found that Medicare claim processors with lower reimbursement amounts used bundled payment polices and those with higher average reimbursement amounts used separate payments for services and drugs. Our study did not assess the actual payment polices of each of the claim processors. Future studies are needed to relate the actual payment polices to physician reimbursement amounts for chemotherapy.

Our study findings revealed that G-CSF prescription was less likely in counties with higher average estimated physician reimbursement amount compared with those with lower average estimated physician reimbursement amount. This finding along with variations in average reimbursement amount across Medicare claim processors suggest that the likelihood of G-CSF prescription is influenced by reimbursement policies of the Medicare claim processors. Our study did not survey the physicians to elicit information on reasons for adding or not adding G-CSF prescriptions to chemotherapy regimens. However, we can speculate as to why G-CSF prescription rates varied across Medicare carries. It is plausible that G-CSF prescriptions in addition to standard chemotherapy regimens may increase a physician’s revenue. Prior empirical research suggests that higher average physician reimbursement amounts closely reflect physicians’ profi t or income.6 Thus, physicians contracting with Medicare claim processors with higher average reimbursement amounts may have had higher incomes compared with their counterparts contracting with Medicare claim processors with low reimbursement amounts. Based on economic theory,11 one could postulate that substitution effect could be responsible for observing lower likelihood of G-CSF prescription among physicians contracting with Medicare claim processors with higher reimbursement. Under substitution-effect theory, these physicians may derive smaller marginal benefi ts from additional income due to prescription of G-CSF drugs. Alternatively, onecould speculate that income effect may have been responsible for the observed pattern. Under the incomeeffect theory, physicians contracting with Medicare claim processors with low reimbursement amounts may have low income and hence may need to increase their income by providing supplemental services and products. Thus in our case, physicians contracting with Medicare claim processors with low reimbursement amount may augment standard chemotherapy regimens with G-CSF for patients with NHL and therefore may be more likely to prescribe G-CSF.

It has to be noted that we did not consider other factors that may discourage the addition of G-CSF. For example, the physician may need to consider the patient’s burden as well. Adding G-CSF to a chemotherapy regimen may involve high out-of-pocket costs for the patient,20 greater travel time to get the injections,27 and impaired quality of life due to possible side effects.27 However, such factors may not explain the systematic variation in G-CSF prescriptions by reimbursement amounts.

Although we focused on the relationship between reimbursement policies and G-CSF prescription, it has to be noted that G-CSF prescription also involves clinical decision making. We found compared with patients in the age group 66 to 70 years, older patients were more likely to receive G-CSF. A plausible explanation for this fi nding could be the increased risk of neutropenia and neutropenia-associated hospitalization with increasing age.28-31 Therefore, physicians may consider age a highrisk factor and prescribe prophylactic G-CSF to older patients.

Furthermore, we found that patients with less toxic chemotherapy regimens such as non-ABC chemotherapy and rituximab with non-ABC chemotherapy were less likely to receive G-CSF compared with highly toxic chemotherapy regimen (ie, ABC chemotherapy). We also found patients with ABC without rituximab were less likely to be treated with G-CSF compared with those with ABC and rituximab. Taken together these fi ndings suggest that greater toxicity of chemotherapy regimens may be associated with greater risk of neutropenia and physicians may prescribe G-CSF as a prophylactic agent.

Diffuse large B-cell lymphoma was found to be significantly associated with G-CSF use. The result conforms to American Society of Clinical Oncology (ASCO) 2006 guideline recommendation “Prophylactic CSF for patients with diffuse aggressive lymphoma aged 65 years and older treated with curative chemotherapy (the regimen—cyclophosphamide, doxorubicin, vincristine, and prednisone [CHOP] or more aggressive regimens) should be given to reduce the incidence of febrile neutropenia and infections.”32 The study showed there was an increased trend of G-CSF prescription across years. The odds ratio for time trend variable was 1.354. The increased likelihood of prescribing G-CSF over time may be because of innovation diffusion, dissemination, and increasing adoption of ASCO guidelines.

The study findings need to be interpreted in light of the strengths and limitations. The strengths were large study population, availability of clinical variables, longitudinal design, and ability to identify chemotherapy and GCSF prescription from claims (not subject to recall bias). The study was potentially limited by its selection criteria.

The study population consisted of Medicare elderly NHL patients without inpatient or ED claims in the fi rst cycle of chemotherapy. We did not control for eligibility for G-CSF (ie, patients who are at high risk of neutropenia). We used surrogate variables for reimbursement policies. Therefore, our fi ndings could at best be considered an indirect evidence of the relationship between reimbursement policies and G-CSF choice. In addition, our grouping of Medicare claim processors may have affected our findings.

Despite these weaknesses, our study is the first to explore Medicare reimbursement policies across different claim processors and its association with prescribing volume for patients with cancer. While our study only measured bundled payment policies through an indirect approach, it also provided some useful insights as to the association between bundled payment policies and provision of care for patients with NHL. Future studies need to obtain the actual payment policies at the claim processor level to further elucidate the relationship between bundled payments and physician practice patterns.

Related Videos
Practice Pearl #1 Active Surveillance vs Treatment in Patients with NETs
© 2024 MJH Life Sciences

All rights reserved.