-
[show abstract]
[hide abstract]
ABSTRACT: STUDY OBJECTIVE: To determine whether the hemoglobin level at which health care providers prescribed erythropoiesis-stimulating agent (ESA) therapy (trigger hemoglobin level) for their patients receiving chemotherapy was lower after the United States Food and Drug Administration (FDA) mandated a black-box warning in March 2007. DESIGN: Retrospective analysis. DATA SOURCE: U.S. Department of Veterans Affairs Healthcare System (VA) national databases. PATIENTS: A total of 7450 patients who were diagnosed with cancer between 2002 and 2009, were undergoing chemotherapy, and who received an ESA within 12 months after their cancer diagnosis. MEASUREMENTS AND MAIN RESULTS: Data were collected on patients' demographic, clinical, environmental, and treatment-related factors. After controlling for these factors, multivariable regression analyses were used to compare the trigger hemoglobin level before and after the FDA-mandated labeling change. The average trigger hemoglobin level was 0.73 g/dl lower after the labeling change (95% confidence interval [CI] -0.84 to -0.63). Moreover, the decline in trigger hemoglobin levels began in mid-2006, when the average trigger hemoglobin level fell from 10.50 g/dl in early 2006 (95% CI 10.36-10.63) to 9.30 g/dl by late 2009 (95% CI 9.10-9.49). CONCLUSION: Even before the 2007 FDA-mandated changes in ESA product labeling, hemoglobin levels that triggered ESA treatment began declining for patients receiving cancer care within the VA. This highlights the critical importance of dissemination of postmarketing safety data to impact shifts in ESA use for anemia management.
Pharmacotherapy 10/2012; · 2.90 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: BACKGROUND: Many veterans have dual Veterans Administration (VA) and Medicare healthcare coverage. We compared 3-year overall and cancer event-free survival (OS; EFS) among patients with non-metastatic colon cancer who obtained substantial portions of their care in both systems and those whose care was obtained predominantly in the VA or in the Medicare fee-for-service system. METHODS: We conducted a retrospective observational cohort study of patients older than 65 years with stages I-III colon cancer diagnosed 1999-2001 in VA and non-VA facilities. Dual use of VA and non-VA colon cancer care was categorized as predominantly VA use, dual use, or predominantly non-VA use. Extended Cox regression models evaluated associations between survival and dual use. RESULTS: VA and non-VA users (all stages) had reduced hazard of dying compared to dual users (for example, for stage I, VA HR 0.40, CI95 0.28-0.56; non-VA HR 0.54, CI95 0.38-0.78). For EFS, stage I findings were similar (VA HR 0.47, CI95 0.35-0.62; non-VA HR 0.64, CI95 0.47-0.86). Stage II and III VA users, but not non-VA users, had improved EFS (Stage II: VA HR 0.74, CI95 0.56-0.97; non-VA HR 0.92 CI95 0.69-1.22. Stage III: VA HR 0.73, CI95 0.56-0.94; non-VA HR 0.81 CI95 0.62-1.06). CONCLUSIONS: Improved survival among VA and non-VA compared to dual users raises questions about coordination of care and unmet needs. IMPACT Additional study is needed to understand why these differences exist, why patients use both systems and how systems may be improved to yield better outcomes in this population.
Cancer Epidemiology Biomarkers & Prevention 10/2012; · 4.12 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: In 2007, growing concerns about adverse impacts of erythropoiesis-stimulating agents (ESAs) in cancer patients led to an FDA-mandated black box warning on product labeling, publication of revised clinical guidelines, and a Medicare coverage decision limiting ESA coverage. We examined ESA therapy in lung and colon cancer patients receiving chemotherapy in the VA from 2002 to 2008 to ascertain trends in and predictors of ESA use.
A retrospective study employed national VA databases to "observe" treatment for a 12-month period following diagnosis. Multivariable logistic regression analyses evaluated changes in ESA use following the FDA-mandated black box warning in March 2007 and examined trends in ESA administration between 2002 and 2008.
Among 17,014 lung and 4,225 colon cancer patients, those treated after the March 2007 FDA decision had 65% (lung OR 0.35, CI(95%) 0.30-0.42) and 53% (colon OR 0.47, CI(95%) 0.36-0.63) reduced odds of ESA treatment compared to those treated before. Declines in predicted probabilities of ESA use began in 2006. The magnitude of the declines differed across age groups among colon patients (p = 0.01) and levels of hemoglobin among lung cancer patients (p = 0.04).
Use of ESA treatment for anemia in VA cancer care declined markedly after 2005, well before the 2007 changes in product labeling and clinical guidelines. This suggests that earlier dissemination of research results had marked impacts on practice patterns with these agents.
Supportive Care in Cancer 09/2011; 20(8):1649-57. · 2.09 Impact Factor
-
Denise M Hynes, Elizabeth Tarlov,
Ramon Durazo-Arvizu,
Ruth Perrin,
Qiuying Zhang,
Thomas Weichle,
M Rosario Ferreira,
Todd Lee,
Al B Benson,
Nirmala Bhoopalam,
Charles L Bennett
[show abstract]
[hide abstract]
ABSTRACT: US veterans have been shown to be a vulnerable population with high cancer rates, and cancer care quality in Veterans Affairs (VA) hospitals is the focus of a congressionally mandated review. We examined rates of surgery and chemotherapy use among veterans with colon cancer at VA and non-VA facilities in California to gain insight into factors associated with quality of cancer care.
A retrospective cohort of incident colon cancer patients from the California Cancer Registry, who were > or = 66 years old and eligible to use VA and Medicare between 1999 and 2001, were observed for 6 months after diagnosis.
Among 601 veterans with colon cancer, 72% were initially diagnosed and treated in non-VA facilities. Among veterans with stage I to III cancer, those diagnosed and initially treated in VA facilities experienced similar colectomy rates as those at non-VA facilities. Stage III patients diagnosed and initially treated in VA versus non-VA facilities had similar odds of receiving adjuvant chemotherapy. In both settings, older patients had lower odds of receiving chemotherapy than their younger counterparts even when race and comorbidity were considered (age 76 to 85 years: odds ratio [OR] = 0.18; 95% CI, 0.07 to 0.46; age > or = 86 years: OR = 0.17; 95% CI, 0.04 to 0.73).
In California, older veterans with colon cancer used both VA and non-VA facilities for cancer treatment, and odds of receiving cancer-directed surgery and chemotherapy were similar in both systems. Among stage III patients, older age lowered odds of receiving adjuvant chemotherapy in both systems. Further studies should continue to explore potential health system effects on quality of colon cancer care across the United States.
Journal of Clinical Oncology 05/2010; 28(15):2571-6. · 18.37 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: The Department of Veterans Affairs (VA) provides nationwide access, lifetime coverage, and an integrated care structure to its enrollees. Those key aspects of VA healthcare-together with data contained in VA's electronic information systems supporting over eight million veterans-provide unique opportunities to study processes, outcomes, and costs of care. Recently, for example, VA data have been used to study outcomes associated with acute postoperative inpatient rehabilitation and care in specialized rehabilitation bed units after lower-limb amputation [1-2], medication adherence and relapse among patients discharged from a VA posttraumatic stress disorder (PTSD) treatment program [3], the provision and costs of assistive technology devices to veterans after stroke [4], and use of mental health services by veterans disabled by auditory disorders [5].
Journal of Rehabilitation Research and Development. 01/2010;
-
The Journal of Rehabilitation Research and Development 01/2010; 47(8):vii-xi. · 1.78 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: We evaluated the improvement in Department of Veterans Affairs (VA) race data completeness that could be achieved by linking VA data with data from Medicare and the Department of Defense (DOD) and examined agreement in values across the data sources. After linking VA with Medicare and DOD records for a 10% sample of VA patients, we calculated the percentage for which race could be identified in those sources. To evaluate race agreement, we calculated sensitivities, specificities, positive predictive values (PPVs), negative predictive values, and kappa statistics. Adding Medicare (and DOD) data improved race data completeness from 48% to 76%. Among older patients (≥65 years), adding Medicare data improved data completeness to nearly 100%. Among younger patients (<65 years), combining Medicare and DOD data improved completeness to 75%, 18 percentage points beyond that achieved with Medicare data alone. PPVs for white and African-American categories were 98.6 and 94.7, respectively, in Medicare and 97.0 and 96.5, respectively, in DOD data using VA self-reported race as the gold standard. PPVs for the non-African-American minority groups were lower, ranging from 30.5 to 48.2. Kappa statistics reflected these patterns. Supplementing VA with Medicare and DOD data improves VA race data completeness substantially. More study is needed to understand poor rates of agreement between VA and external sources in identifying non-African-American minority individuals.
The Journal of Rehabilitation Research and Development 01/2010; 47(8):781-95. · 1.78 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: We evaluated the improvement in Department of Veterans Affairs (VA) race data completeness that could be achieved by linking VA data with data from Medicare and the Department of Defense (DOD) and examined agreement in values across the data sources. After linking VA with Medicare and DOD records for a 10% sample of VA patients, we calculated the percentage for which race could be identified in those sources. To evaluate race agreement, we calculated sensitivities, specificities, positive predictive values (PPVs), negative predictive values, and kappa statistics. Adding Medicare (and DOD) data improved race data completeness from 48% to 76%. Among older patients (≥65 years), adding Medicare data improved data completeness to nearly 100%. Among younger patients (<65 years), combining Medicare and DOD data improved completeness to 75%, 18 percentage points beyond that achieved with Medicare data alone. PPVs for white and African-American categories were 98.6 and 94.7, respectively, in Medicare and 97.0 and 96.5, respectively, in DOD data using VA self-reported race as the gold standard. PPVs for the non-African-American minority groups were lower, ranging from 30.5 to 48.2. Kappa statistics reflected these patterns. Supplementing VA with Medicare and DOD data improves VA race data completeness substantially. More study is needed to understand poor rates of agreement between VA and external sources in identifying non-African-American minority individuals.
Journal of Rehabilitation Research and Development. 01/2010;
-
[show abstract]
[hide abstract]
ABSTRACT: In the United States, despite substantial investment in public health initiatives to promote early detection of breast cancer through screening mammography, the proportion of female breast cancers that have advanced beyond the localized stage by the time of diagnosis remains high. Our objective in this exploratory study was to investigate whether stage of breast cancer at diagnosis among Chicago residents is associated with characteristics of the neighborhoods in which proximate mammography facilities are located. Those characteristics may influence likelihood of utilizing the service routinely and partly explain differences in stage at diagnosis. We used a retrospective cohort design and combined 3 years of data from the Illinois State Cancer Registry (ISCR) with information on locations of mammography facilities, public transportation service, crime, and area demographic and economic characteristics. Using a Geographic Information System (GIS), we identified the five facilities located nearest to each case's residence. Estimates of the association between characteristics of mammography facility locations and breast cancer stage at diagnosis were obtained using the partial proportional odds regression model. We found that the number of homicides in areas in which the nearest mammography facilities were located was associated with increased odds of later stage diagnosis. This effect was independent of age, race, and residential area education and income. We found no effect on stage of distance, public transportation service, or measures of neighborhood social similarity. The "spatial dynamics" of health may involve geographies beyond the immediate neighborhood. The results of our study suggest that areas in which the nearest mammography facilities are located may be one such geography. We hope that this study will spark research interest in the impact of health service locations on utilization.
Journal of Urban Health 11/2008; 86(2):196-213. · 2.13 Impact Factor
-
Kevin T Stroupe,
Bridget M Smith,
Todd A Lee, Elizabeth Tarlov,
Ramon Durazo-Arvizu,
Zhiping Huo,
Tammy Barnett,
Lishan Cao,
Muriel Burk,
Francesca Cunningham,
Denise M Hynes,
Kevin B Weiss
[show abstract]
[hide abstract]
ABSTRACT: In February 2002, the Department of Veterans Affairs (VA) raised medication copayments from $2 to $7 per 30-day supply of medication for certain veteran groups. We examined the impact of the copayment increase on medication acquisition from VA.
This was a retrospective cohort study using data from national VA databases from February 2001 through February 2003. We took a random sample of over 5% of male VA users in 2001. Of 149,107 veterans sampled, 19,504 (13%) had copayments for no drugs, 101,410 (68%) had copayments for some drugs, and 28,193 (19%) had copayments for all drugs. We used multivariable count models to examine changes in the number of 30-day medication supplies after the increase.
After the copayment increase, veterans subject to copayments for all drugs received 8% fewer 30-day supplies of medication annually relative to veterans with no copayments (P < 0.001). The effect of the copayment increased as the number of different medications veterans received increased. Among veterans subject to copayments for all drugs, acquisition of lower-cost drugs fell by 36%, higher-cost medications fell by 6%, over-the-counter medications fell by 40%, and prescription-only medications fell by 4% relative to veterans with no drug copayments.
The number of medications veterans obtained from VA decreased after the copayment increase. There were relatively larger impacts on veterans with higher medication use and on lower-cost and over-the-counter medications.
Medical Care 12/2007; 45(11):1090-7. · 3.41 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: Recent research suggests living in an economically disadvantaged neighborhood is associated with decreased likelihood of undergoing mammography and increased risk of late-stage breast cancer diagnosis. Long distances and travel times to facilities offering low- or no-fee mammography may be important barriers to adherence to mammography screening recommendations for women living in economically disadvantaged urban neighborhoods, in which African-Americans are disproportionately represented. The purpose of this study was to examine whether the spatial distribution of facilities providing low- or no-fee screening mammography in Chicago, Illinois, is equitable on the basis of neighborhood socioeconomic and racial characteristics. We found that distance and travel times via automobile and public transportation to facilities generally decrease as neighborhood poverty increases. However, we also found that the strength of the association between neighborhood poverty level and two of the spatial accessibility measures-distance and public transportation travel time-is less strong in African-American neighborhoods. Among neighborhoods with the greatest need for facilities (i.e., neighborhoods with the highest proportions of residents in poverty), African-American neighborhoods have longer travel distances and public transportation travel times than neighborhoods with proportionately fewer African-American residents. Thus, it appears that the spatial accessibility of low- and no-fee mammography services is inequitable in Chicago. In view of persistent social disparities in health such as breast cancer outcomes, these findings suggest it is important for researchers to examine the spatial distribution of health resources by both the socioeconomic and racial characteristics of urban neighborhoods.
Journal of Urban Health 04/2006; 83(2):195-210. · 2.13 Impact Factor
