Skip to main content
. 2015 Jul;105(Suppl 3):e4–e15. doi: 10.2105/AJPH.2014.302490

TABLE 1—

Description of Included Studies

First Author Mean Age of Participants, Years Sample, No. % White Cancer Stage Medicaid or Medicare Data Source Type of ET Dependent Variable Analysis Racial Comparators Years of Data Collection Setting
Ashing-Giwa77 NR Patient, 102; provider, 20 Patient, 12; provider, 20 0–IV Neither Qualitative: interviews and focus groups Breast cancer care (including ET) NA Qualitative African American, Asian American, Latina, Caucasian NR UCLA community
Bhatta78 58 197 69 I–III Neither Survey and medical record review Tamoxifen, AIs Adherence, persistence Multivariable logistic regression White, Black NR University of Chicago Hospital
Bickell79 60 677 49 I–II Neither Medical record review Breast cancer care (including ET) Initiation Multivariable logistic regression White, Black, Hispanic, Asian, unknown 1999 and 2000 (diagnoses), 2002–2004 (abstraction) 6 NYC hospitals
Friese80 59 743 48 I–III Neither Survey and SEER Registry Tamoxifen, AIs Initiation, persistence Multivariable logistic regression White, Black, Latina Time of diagnosis, survey 4 y after diagnosis Multisite: LA and Detroit sites
Kimmick60 67 1 491 59 I–III Medicaid NC Central Tumor Registry and Medicaid claims Tamoxifen, AIs Initiation, adherence, persistence Multivariate logistic regression White, other 1998–2002 NC Medicaid
Liu81 51 303 34 I–III Medicaid Survey Tamoxifen, AIs Persistence Multivariate logistic regression Less acculturated Latina, more acculturated Latina, and other 2003–2005 enrollment CA Medicaid
Livaudais82 59 13 753 76 I–III Neither SEER Registry, KPNC claims, and US Census Tamoxifen, AIs Initiation Multivariable logistic regression Hispanic, African American, Chinese, Japanese, Filipino, and South Asian vs WNH 1996–2007 Multisite, KPNC
Livaudais83 69 3 575 92 I–III Neither WHI Extension Survey Tamoxifen, AIs Initiation, persistence Multivariable logistic regression Hispanic, African American, Asian/Pacific Islander vs WNH 2009–2010 Multisite, parent study (WHI)
Livaudais84 Phase 1 = 49; phase 2 = 50 Phase 1 = 731; phase 2 = 654 Phase 1 = 58; phase 2 = NA I–III Neither NCBCF Registry and self-report Tamoxifen, AIs Initiation Multivariable logistic regression African American, Hispanic, WNH vs Asian American 1995–1998 (phase 1) or 1998–2003 (phase 2) Multisite, NCBCF Registry
Partridge19 75 2 378 83 0–III Medicaid NJ cancer registry and Medicaid claims Tamoxifen Adherence Unconditional logistic regression White, non-White 1991–1995 NJ Medicaid and NJ PAAD program
Reeder-Hayes85 NR 2 640 89 I–III Neither NC Central Tumor Registry and Claims: ICISS Tamoxifen, AIs Initiation Generalized estimating equations; Cox proportional hazards White, Black 2004–2009 Multisite: NC Registry
Riley22 NR 15 542 81 I–IV Medicare SEER Registry and Medicare claims Tamoxifen, AIs Adherence Multivariable logistic regression WNH, WH, Black, other or unknown 2003–2005 (time of diagnosis), 2006–2007 (time of ET use) Multisite
Shelton86 NR 1 145 69 I–III Neither Survey and medical record review Breast cancer care (including hormonal therapy) Treatment decision-making χ2 test and analysis of variance White, Black, Hispanic, Asian, other 2006–2010 Multisite: Detroit, NYC, Northern CA (KPNC)
Wheeler87 49 222 47 0–II Medicaid NC Central Tumor Registry and Medicaid claims Tamoxifen, AIs Initiation Multivariable logistic regression White, Black, other 2004–2007 NC Medicaid

Note. AI = aromatase inhibitor; CA = California; ET = endocrine therapy; ICISS = Integrated Cancer Information Surveillance System; KPNC = Kaiser Permanente Northern California; LA = Los Angeles; NA = not applicable; NC = North Carolina; NCBCF = Northern California Breast Cancer Family Registry; NJ = New Jersey; NR = not reported; NYC = New York City; PAAD = Pharmaceutical Assistance to the Aged and Disabled; SEER = Surveillance, Epidemiology, and End Results; UCLA = University of California, Los Angeles; WH = White Hispanic; WHI = Women’s Health Initiative; WNH = White, non-Hispanic.