Racial Disparities in Intravenous Bisphosphonate Use Among Older Patients With Multiple Myeloma Enrolled in Medicare

PURPOSE:

Intravenous (IV) bisphosphonates reduce the risk of skeletal-related events in patients with multiple myeloma (MM). However, data describing racial differences in IV bisphosphonate utilization outside of clinical trial settings are limited. We evaluated population-level IV bisphosphonate initiation and discontinuation among patients of age ≥ 65 years with MM.

METHODS:

We conducted a retrospective cohort study of patients of age ≥ 65 years diagnosed with first primary MM between 2001 and 2011. Patients were identified using the SEER-Medicare linked database and followed through December 2013. Cumulative incidences of IV bisphosphonate initiation and time to discontinuation among users were compared between racial and ethnic groups. In Fine and Gray competing risk models, we estimated subdistribution hazard ratios (SHRs) and 95% CIs for initiation and discontinuation.

RESULTS:

We included 14,231 eligible patients with MM (median age, 76 years; 52% male). Over a median follow-up of 23.1 months, 54% of patients received at least one IV bisphosphonate dose. Our final analytical sample included 10,456 non-Hispanic (NH) Whites, 2,267 NH Blacks, 548 Asian and Pacific islanders, and 815 Hispanic and Latino patients. A higher proportion of White patients (56.1%) newly received IV bisphosphonates after MM diagnosis compared with NH Blacks (45.4%). Compared with White patients, NH Black patients were less likely to initiate IV bisphosphonates (SHR, 0.74; 95% CI, 0.70 to 0.79) and slightly more likely to discontinue treatment (SHR, 1.10; 95% CI, 1.01 to 1.19).

CONCLUSION:

Approximately half of the patients with MM of age ≥ 65 years did not receive IV bisphosphonates, with significant delay among racial minority groups. These findings highlight the need for improvement of IV bisphosphonate uptake in patients with MM of age ≥ 65 years.

INTRODUCTION

Multiple myeloma (MM) is a B-cell hematologic malignancy characterized by monoclonal plasma cell proliferation in the bone marrow.^1,2 Diffuse osteopenia and osteolytic lesions develop in 85% of patients with MM³ and can lead to skeletal-related events (SREs) such as bone pain, pathological fractures, and spinal cord compression. Given the substantial comorbidity and increased mortality associated with SREs among patients with MM, supportive therapy with intravenous (IV) bisphosphonates is recommended among most patients.⁴ Although MM remains incurable, the availability of more sensitive diagnostic tools, introduction of novel therapies, such as immunomodulatory drugs (IMiDs) and proteasome inhibitors (PIs), and wider adoption of autologous stem-cell transplantation (ASCT) in the upfront treatment of patients have significantly extended survival.⁵

The incidence of MM is two-fold higher among Black patients relative to their non-Hispanic (NH) White counterparts in the United States.⁶ Outcomes in MM also differ across racial groups and socioeconomic status.⁷ This is in part due to decreased access to care and timely administration of effective guideline-recommended anti-MM treatment, as well as the economic burden of these high-cost drugs that is imposed on patients.^8,9 Although differences in the use of PIs, IMiDs, and ASCT have been well-described by race and ethnicity, less is known about disparities in the use of supportive treatments, including IV bisphosphonates.¹⁰

Two highly effective nitrogen-containing bisphosphonate agents, pamidronate and zoledronic acid, are approved for the treatment of symptomatic bone lesions, in conjunction with antimyeloma treatment, in patients with MM.¹¹ Treatment with IV bisphosphonates for 2 years has been associated with a 26% reduction of risks of SRE in patients with MM.¹² Unfortunately, several epidemiologic studies have highlighted low rates and inappropriate duration of bisphosphonate use, contrary to what is recommended in guidelines.^13,14 Real-world barriers to adhering to parenteral treatment also apply for IV bisphosphonate use and therefore contribute to nonadherence to what is recommended in guidelines. For example, it requires the placement of a temporary peripheral IV catheter for administration of the 15-minute infusion and clinical monitoring for potential infusion–related reactions (ie, fever, chills, and myalgia). Additionally, for some patients, distance and time of travel to clinic visits are additive challenges, further contributing to untimely administration of IV bisphosphonates and increased risk for SREs.¹⁵

Despite the efficacious role of IV bisphosphonate treatment in MM, few studies have described their real-world utilization patterns among older patients and minority groups, patient populations largely under-represented in clinical trials.¹⁶ Our objective was to examine racial differences in initiation and persistence of IV bisphosphonate treatment among Medicare beneficiaries with newly diagnosed MM.

METHODS

Data Source

We performed a retrospective cohort study using the SEER-Medicare linked database. The participating population–based cancer registries collect information on patient demographics, clinical characteristics, and cancer-related treatment modalities, as well as vital status and cause of death. The SEER Program of the National Cancer Institute (NCI) covers approximately 28% of the US population.¹⁷ The SEER-Medicare databases contain linkage to fee-for-service hospital, outpatient, physician, home health, and hospice Medicare claims administered by the Centers for Medicare & Medicaid Services. Medicare covers approximately 97% of patients over the age of 65 years, and more than 93% of subjects meeting Medicare eligibility in the SEER registry were linked to Medicare enrollment files.¹⁷ We evaluated data from patients with a primary MM diagnosed between the years 2001 and 2011, which are linked to Medicare claims through 2012 and with enrollment and follow-up through 2013.

The following SEER-Medicare linked database files were used to determine treatment and clinical characteristics of patients included in the present study: (i) Patient Enrollment and Diagnosis Summary file, (ii) Outpatient (institutional Medicare Part B claims), (iii) Medicare Provider Analysis and Review (inpatient Medicare Part A claims), (iv) National Claims History file (provider Medicare Part B claims), and Durable Medical Equipment.¹⁷

The University of Illinois at Chicago Institutional Review Board approved and determined this study to be exempt from human subjects' research requiring informed consent.

Cohort Selection

We included patients diagnosed with first primary MM defined by the International Classification of Diseases-O-3 code 9732 and microscopic confirmation of diagnosis from positive histology or cytology. Patients were required to have at least 180-day continuous Medicare part A and B fee-for-service enrollment prior to MM diagnosis. Because of potentially incomplete claims for patients enrolled in health management organizations, patients were excluded. In addition, patients were required to have at least one medical encounter within 30 days from the cancer diagnosis and were censored at the occurrence of a 90-day gap in coverage (Fig 1).

FIG 1.

Schematic overview of study design. BP, blood pressure; IV, intravenous; MM, multiple myeloma.

We used International Classification of Diseases, Ninth Revision, Clinical Modification and five-digit Healthcare Common Procedure Coding System codes from Medicare claims files to determine clinical characteristics and treatments received in the 180-day period prior to MM diagnosis (baseline) and from diagnosis to the end of follow-up (observation period). Since IV bisphosphonates are not recommended in patients with inadequate renal function,¹⁸ we excluded patients with end-stage renal disease at baseline (Appendix Fig A1, online only).

Exposure, Outcomes, and Follow-Up

We defined administration of IV bisphosphonates using Healthcare Common Procedure Coding System codes for pamidronate (C9411 and J2430) and zoledronic acid (J3487, J3489, J3488, Q4095, Q2051, and C9115) from MM diagnosis until death, disenrollment from Medicare coverage, or end of the study period (December 31, 2013), whichever came first. The index date from which patients were followed for IV bisphosphonate initiation was MM diagnosis. Similarly, for patients who initiated IV bisphosphonates, we assessed discontinuation with index time as the initiation date. In a conservative approach, we employed a 45-day gap assuming that an administration of IV bisphosphonates would cover for 28 days and allowing a 17-day grace period prior to the next dose. If patients did not receive a next dose by the end of 45 days from the previous administration, discontinuation was recorded by the end of 28-day period following the previous administration. To test the robustness of our findings, we additionally evaluated treatment discontinuation using 60- and 90-day allowable treatment gaps between two IV bisphosphonate administrations. Cumulative incidences of SRE were assessed using the number of subjects with at least one SRE following the approach of Gooley et al,¹⁹ and the numbers of days from incident MM diagnosis date to first occurrence of SREs were calculated and compared across racial and ethnic groups using the Wilcoxon rank-sum test, accounting for the non-normal distribution of the number of non-SRE survival.

Covariates

We collected information on patients' race and ethnicity, age, sex, and MM-specific treatment, as well as other clinical comorbid conditions at baseline. Patients were divided into five mutually exclusive racial-ethnic groups: NH White, NH Back, Hispanic and Latino, Asian and Pacific Islander (API), and other race and ethnicity. Additionally, we identified patients with a diagnosis that may increase the risk of fall or fracture, including Alzheimer's disease, history of fall, syncope, diabetes, heart failure, stroke, or chronic pulmonary disease during the index hospital admission and/or the prior six months using the NCI-Charlson Comorbidity Index.²⁰ Given that the established association between primary care and wellness follow-up coincides well with adherence to other medications,²¹ we described the utilization of bone mineral density tests over the precancer baseline period. The presence of SRE at baseline was also assessed and classified by anatomical site (ie, vertebral fracture, hip fracture, and other sites) (Appendix Tables A1 and A2, online only).

Statistical Analysis

Baseline demographic and clinical characteristics were compared between patients who did (users) and did not (nonusers) initiate IV bisphosphonates and by race and ethnicity using chi-square test for categorical variables and analysis of variance models for continuous variables. Wilcoxon rank-sum test and analysis of variance models were used assessing differences between medians and means across racial groups, respectively. Categories were combined, or the results were suppressed to avoid reporting any cell counts < 11.

We calculated cumulative incidences of treatment initiation and treatment discontinuation using Fine and Gray regression models²² that accounted for death without the event of interest as a competing risk. To evaluate the risk factors associated with treatment initiation and discontinuation, we evaluated both baseline and time-varying anti-MM treatments received in discontinuation analyses. Subdistribution hazard ratios (SHRs) and 95% CIs were estimated for associations between race and ethnicity and initiation and discontinuation of IV bisphosphonates. All statistical analyses were performed using SAS S 9.4 (SAS Institute Inc., Cary, NC) and open-source software R 3.4.1 (cmprsk package²³). Two-tailed P-values < .05 were considered statistically significant.

RESULTS

Study Cohort and Demographics

From an overall analytical cohort of 14,231 patients with MM, the median age was 76 years (interquartile range [IQR], 71 to 82) and 47% were female. We included 10,456 White (74%), 2,267 (16%) Black, 815 (6%) Hispanic and Latino, and 548 (4%) API patients; 145 (1%) had other or unknown race and ethnicity.

Demographic characteristics at MM diagnosis were significantly different between IV bisphosphonate users and nonusers (Appendix Table A3, online only) with a mean age of 78.0 years among users and 75.6 years among nonusers. A slightly higher proportion of nonusers were male (53%) compared with 51% among users, and a higher proportion of IV bisphosphonate users were NH White population among users compared with nonusers (77% v 70%).

Demographic and clinical characteristics across racial groups are summarized in Table 1. Compared with White patients, Black patients had a lower mean (median) age (75.8 [75] v 77.1 [77]) and higher rates of heart failure (20% v 16%), diabetes (31% v 21%), and stroke (9% v 5%). More Black patients had recent history of hospitalization (48% v 39%). Additionally, fewer Black patients received modern era MM treatment when compared with White patients (PIs [27.8% v 33.1%, P < .0001] and ASCT [2.5% v 4.9%, P < .0001]).

TABLE 1.

Descriptive Baseline Demographic, Clinical, and Treatment Characteristics of Eligible Patients With Multiple Myeloma, Stratified by Racial and Ethnic Status

IV Bisphosphonate Initiation by Racial Groups

IV bisphosphonate utilization patterns differed significantly across racial and ethnic groups (Table 2). Overall, 54% of patients with MM initiated IV bisphosphonates after diagnosis with a median time to first bisphosphonate dose of 50 days (IQR, 25-140). White patients had higher rates of IV bisphosphonate initiation (56%), compared with NH Blacks (45%), Hispanic (50%), and API (47%) (Table 2). We described the timing from MM diagnosis to IV bisphosphonate initiation, and 47% of patients received first dose within the first year, and among patients who remained at risk at the beginning of year two, 14% initiated IV bisphosphonates during the second year postdiagnosis.

TABLE 2.

Patterns of Intravenous Bisphosphonate Treatment on Initiation and Discontinuation, Stratified by Racial and Ethnic Status

NOTE. Categories were combined or suppressed to avoid reporting a count of < 11.

Abbreviations: API, Asian and Pacific Islander; IQR, interquartile range; MM, multiple myeloma; NH, non-Hispanic; Q4w, once every 4 weeks; Q8w, once every 8 weeks; Q12w, once every 12 weeks; SD, standard deviation.

^aWilcoxon rank-sum test for median comparison across racial groups, chi-square test for categorical variable comparison between racial groups, and analysis of variance for mean comparison across racial and ethnic groups. Everitt B, Skrondal A: The Cambridge Dictionary of Statistics (ed 4). Cambridge, UK, Cambridge University Press, 2010.

The cumulative incidence of initiation of IV bisphosphonate therapy following MM diagnosis was 52% during year 1, 57% in year 2, and 60% in year 3 postdiagnosis (Fig 2A).

FIG 2.

Cumulative incidence of intravenous bisphosphonate therapy initiation (A) and discontinuation (B), stratified by racial and ethnic status. NH, non-Hispanic.

IV Bisphosphonate Discontinuation and Cumulative Dose and Frequency

Associations between treatment persistence by race are presented in Table 2. The median treatment duration was 290 days (IQR, 135 to 577). Black patients had a lower duration of IV bisphosphonate treatment compared with other racial groups (median 267 days v 294 [White], 293.5 [API], and 282.5 [Hispanic and Latino]). Overall, only 47% of patients received IV bisphosphonates for longer than 1 year; no statistically significant differences in persistence were observed across racial groups among those who initiated treatment (48% in White, 44% in Black, 50% in API, and 45% in Hispanic and Latino patients) (Fig 2B).

The median number of doses administered over the duration of IV bisphosphonate treatment, defined as no interval between two administrations for more than 45 days, was 12 (IQR, 4 to 23). Black patients had a slightly lower mean number of cumulative IV bisphosphonate doses (14.0), compared with White (15.9 doses), API (15.6 doses), and Hispanic and Latino patients (14.7 doses) (Table 2). Average administration also differed by race. Although overall 33% of patients received guideline-recommended IV bisphosphonate frequency (at least every 4 weeks), fewer Black patients received optimal IV bisphosphonate frequency (29% v 33% in White, 31% in API, and 35% in Hispanic and Latino patients). Treatment discontinuation after accounting for death without occurrence of event as a competing risk was 44% by the end of year 1, 67% by year 2, and 81% by year 3.

Cumulative SRE During Follow-Up

Overall, 6,703 (47.1%) patients experienced at least one SRE following MM diagnosis. Among patients with at least one SRE during the follow-up period, 4,743 (70.8%) had vertebral SREs and 1,371 (20.5%) experienced hip SREs. The postindex SRE incidences were higher among patients who initiated IV bisphosphonate treatment compared with nonusers (52.9% v 35.7%, P < .0001). Additionally, the SRE rates differed across racial and ethnic groups. The SRE rates were highest among Hispanic patients (49.5%), followed by White (49.1%) and API (45.1%), whereas Black patients had lowest rates of SREs (38.2%). Moreover, the median duration of follow-up time without SRE was longer among Black patients (79 days) compared with NH White (50 days) and Hispanic patients (49 days) (P = .0391).

Factors Associated With Initiation and Persistence

In multivariable competing risk regression models, non-White patients had significantly delayed or no initiation of IV bisphosphonates compared with White patients (Black SHR, 0.74; 95% CI, 0.70 to 0.79; API SHR, 0.72; 95% CI, 0.64 to 0.82; and Hispanic and Latino SHR, 0.79; 95% CI, 0.72 to 0.88). Younger age at MM diagnosis was associated with higher rates of treatment initiation (65-75 years: SHR, 1.74; 95% CI, 1.59 to 1.91; 76-85 years: SHR, 1.44; 95% CI, 1.32 to 1.58), compared with patients of age 86 years and older at diagnosis. We found that greater comorbidities including Alzheimer's disease (SHR, 0.73; 95% CI, 0.56 to 0.95), heart failure (SHR, 0.77; 95% CI, 0.72 to 0.83), and renal impairment (SHR, 0.63; 95% CI, 0.58 to 0.69) were associated with lower likelihood of initiating IV bisphosphonates. In contrast, patients with a baseline history of SRE were more likely to receive IV bisphosphonates after MM diagnosis (SHR, 1.52; 95% CI, 1.44 to 1.62) (Fig 3A).

FIG 3.

Forest plots of risk factors associated with IV bisphosphonate therapy initiation (A) and discontinuation (B). BP, blood pressure; CCI, Charlson Comorbidity Index; CFI, claims-based frailty index; COPD, chronic obstructive pulmonary disease; ESRD, end-stage renal disease; IV, intravenous; NCI, National Cancer Institute; NH, non-Hispanic; SHR, subdistribution hazard ratio; SRE, skeletal-related event.

In multivariable-adjusted models, we found a modest increase in risk of IV bisphosphonate discontinuation among Black patients (SHR, 1.10; 95% CI, 1.01 to 1.19) compared with White patients. Younger age was suggestive to be associated with longer duration of IV bisphosphonate therapy. Both more comorbidities at baseline (NCI-Charlson Comorbidity Index scores ≥ 3) and the presence of renal disease at baseline were associated with early discontinuation. In addition, exposure to cytotoxic chemotherapy agents prior to IV bisphosphonate initiation was also associated with greater likelihood of treatment discontinuation (SHR, 1.24; 95% CI, 1.15 to 1.34) (Fig 3B).

DISCUSSION

In the present study, we found that Black, Hispanic and Latino, and API patients were less likely to receive IV bisphosphonate therapy after MM diagnosis even after accounting for differences in comorbidity. Other studies have documented disparities in MM therapy that tend to indicate that nonguideline concordant care affects racial and ethnic minorities.¹⁰ However, we found that although there appear to be differences in initiation of therapy, only modest differences in persistence on IV bisphosphonate therapy by race were observed.

The International Myeloma Working Group and National Comprehensive Cancer Network practice guidelines recommend initiation of IV bisphosphonates in all patients receiving first-line antimyeloma therapy, irrespective of the presence of osteolytic bone lesions.^3,24 The ASCO guidelines recommend the use of bisphosphonates among patients with MM with lytic destruction of bone or compression fracture of the spine from osteopenia and without contraindications for up to 2 years following MM diagnosis.²⁵ In addition, both pamidronate²⁶ and zoledronic acid²⁷ have been found to improve survival in randomized clinical trials, with a reduced hazard ratio (HR) for the mortality risk up to 18% (HR, 0.82; 95% CI, 0.70 to 0.95) associated with IV bisphosphonates.²⁸ More recently, in a retrospective analysis of older patients with MM in the SEER-Medicare linked database, IV bisphosphonate use was associated with a lower risk of all-cause mortality (HR, 0.70; 95% CI, 0.56 to 0.88), although the study did not account for immortal time prior to IV bisphosphonate initiation.²⁹

Our findings are consistent with previous studies on supportive care in older patients with MM. One study by Giri et al³⁰ of 1996 older patients with MM in the SEER-Medicare linked database showed that 64% receiving anti-MM therapy used bone-modifying drugs within 12 months postdiagnosis. NH Black patients were less likely to receive bone-modifying drugs compared with White patients (adjusted odds ratio, 0.63; 95% CI, 0.46 to 0.88), and patients with baseline renal impairment (adjusted odds ratio, 0.43; 95% CI, 0.34 to 0.54) had lower odds of receiving bone-modifying drugs. They also found substantial underutilization of guideline-recommended influenza vaccination and antiviral prophylaxis, suggesting that barriers in access to care exist among older patients with MM.³⁰

Several reasons could explain the observed racial disparity in IV bisphosphonate utilization, including differential healthcare seeking behavior, access to care, and costs. Previous studies describe less frequent stem-cell transplantation and the use of bortezomib among Black patients with MM and that these differences contributed to a 12% increased risk of all-cause mortality.³¹ Many older patients with cancer encounter issues with transportation and logistical challenges with travel and attending multiple office visits and outpatient infusion centers for IV bisphosphonate administration every 3-4 weeks, and this may disproportionately affect racial and ethnic minorities and lower-income patients.³² Zoledronic acid can be administered via 15-minute IV infusion, whereas pamidronate is administered over 2 hours. The relative ease of use in administering zoledronic acid compared with pamidronate could explain zoledronic acid's greater uptake during our study period. Another alternative explanation to disparities in the utilization of IV bisphosphonates in patients with MM is the concern over the risks of bisphosphonate-related osteonecrosis.³³ The prevalence of this adverse effect in patients undergoing treatment for bone involvement because of cancer varies by tumor type with reports indicating higher rates occurring in patients with MM, between 5% and 8%.^34-37 In general, patients with cancer with poor oral health (eg, decayed and broken teeth, periodontal disease, and active inflammation and infection) receiving chemotherapy, comorbidities, and use of other medications can contribute to a poor environment for tissue and wound healing, which can promote the development of osteonecrosis and lead to osteomyelitis.³⁴ Furthermore, disparities exist by race and ethnicity in terms of unmet dental care needs³⁸; thus, barriers to appropriate oral health screening for the safe initiation of IV bisphosphonates in patients with MM could lead to our observed disparity in initiation and persistence.

We found that the incidence of SREs was lower among Black patients compared with White and Hispanic patients with MM despite lower utilization of IV bisphosphonate therapy. Several reasons could have contributed to the observed difference in SRE rates. First, patients with severe disease might have died prior to the occurrence of SREs. Second, Black patients were younger at MM diagnosis compared with other racial and ethnic groups and had lower prevalence of osteoporosis compared with White patients (6.7% v 14.0%). Likewise, Black patients had lower rates of pre-existing SREs (13.7%) compared with White (21.5%) and Hispanic patients (22.5%). Lower prevalence of SRE risk factors in Black patients could have attributed to inferior SRE rates in this racial group. Studies have indicated that comparable treatment outcomes can be regardless of racial and ethnic group when patients receive appropriate, timely care and treatment.³⁹ Although there exists disease heterogeneity by race, Black patients with MM are reported to have a better survival than Caucasian patients, despite a two- to three-fold increase in MM incidence rates.⁴⁰ Thus, it has been hypothesized that even more pronounced survival benefit could be achieved among Black patients after improvement of inequalities in access to modern care, such as newer therapy (PIs, iMiDs, and transplantation) and supportive care as bone-modifying agents.^41,42

We also identified characteristics associated with delayed IV bisphosphonate initiation and early discontinuation; older age and greater comorbidity were associated with delayed initiation and shorter duration of IV bisphosphonate therapy. In other studies, persistence on zoledronic acid was associated with lower risks of SREs and fracture.⁴³ In addition, patients with history of Alzheimer's disease, heart failure, and stroke were at risk of delayed initiation of treatment, after accounting for death as a competing risk. We also found that patients with a history of SRE were more likely to receive IV bisphosphonates and patients with renal disease were less likely. This reflects concordance with the indications and contraindications for IV bisphosphonate therapy similar to other reports.^13,44

Our study has several strengths. First, we included a population-based sample of patients from over a 10-year period in the SEER-Medicare linked database. Second, we used a robust approach to account for competing risks of death without treatment initiation or discontinuation and identify characteristics associated with IV bisphosphonate utilization in clinical practice. Finally, we were able to measure a wide range of baseline clinical characteristics and adjusted for other MM treatment when determining the likelihood of IV bisphosphonate initiation and discontinuation.

This study also had limitations. Using data from administrative health claims, we lacked information on certain prognostic factors important for determining MM treatment decisions and survival in patients with MM including prognostic risk scores, cytogenetics, and performance status. Thus, we cannot rule out confounding by unmeasured covariates. Furthermore, we lacked information on outpatient pharmacy records for all patients to comprehensively assess the patterns of their oral anticancer medication use; Medicare part D data were only available for enrolled patients from 2007 forward. Moreover, because of inherent limitation of our data source, we were unable to identify osteonecrosis of jaw as potential factors associated with initiation and discontinuation of IV bisphosphonate therapy.⁴⁵ Finally, our sample was restricted to older Medicare beneficiaries and these findings may not be directly extrapolated to younger patients and patients with MM lacking health insurance coverage.

In conclusion, we found substantial underutilization of supportive care with IV bisphosphonate therapy among older adults with MM overall. We also found significantly lower initiation of IV bisphosphonates among Black, Hispanic and Latino, and Asian and Pacific Islander patients compared with White patients. Further studies should evaluate patient preferences and identify barriers to timely initiation and adherence to IV bisphosphonate therapy guidelines to provide optimal supportive cancer care in patients with MM.

APPENDIX

FIG A1.

Patient selection flow diagram. ESRD, end-stage renal disease; ICD, International Classification of Diseases; HMO, health maintenance organization; IV, intravenous; MM, multiple myeloma; PEDSF, Patient Entitlement and Diagnosis Summary File.

TABLE A1.

Codes Used for Determining Antimyeloma Parenteral Treatment

TABLE A2.

Codes Used for Ascertaining Clinical Conditions During Baseline and Follow-Up Periods

TABLE A3.

Baseline Demographic and Clinical Characteristics of Eligible Patients With Multiple Myeloma, Stratified by Bisphosphonate Exposure Status

AUTHOR CONTRIBUTIONS

Conception and design: Jifang Zhou, Karen Sweiss, Edith A. Nutescu, Jin Han, Todd A. Lee, Gregory S. Calip

Financial support: Gregory S. Calip

Administrative support: Gregory S. Calip

Provision of study materials or patients: Gregory S. Calip

Collection and assembly of data: Jifang Zhou, Karen Sweiss, Gregory S. Calip

Data analysis and interpretation: Jifang Zhou, Karen Sweiss, Edith A. Nutescu, Pritesh R. Patel, Naomi Y. Ko, Todd A. Lee, Brian C.-H. Chiu, Gregory S. Calip

Manuscript writing: All authors

Final approval of manuscript: All authors

Accountable for all aspects of the work: All authors

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

Racial Disparities in Intravenous Bisphosphonate Use Among Older Patients With Multiple Myeloma Enrolled in Medicare

The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated unless otherwise noted. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or ascopubs.org/op/authors/author-center.

Open Payments is a public database containing information reported by companies about payments made to US-licensed physicians (Open Payments).

Pritesh R. Patel

Consulting or Advisory Role: Celgene

Speakers' Bureau: Janssen, Amgen, Celgene

Naomi Y. Ko

Honoraria: Pfizer

Consulting or Advisory Role: Pfizer

Research Funding: Pfizer

Todd A. Lee

Consulting or Advisory Role: EMD Serono

Expert Testimony: Cook Medical, AstraZeneca

Gregory S. Calip

Employment: Flatiron Health

Research Funding: Pfizer

No other potential conflicts of interest were reported.