Following lung cancer, breast cancer is the secondhighest cause of cancer death among African American women, and from 2008-2012African American women were 42% more likely to die compared to white women (“Report,” n.
d.). This gap isexpected to continue widening. Interventionsare needed to reduce racial disparities in breast cancer survival. For estrogenreceptor-positive tumors, African American women were 2.65 times as likely to diecompared to white women in the first twoyears (Warner et al., 2015).
Blacks were76% and 56% more likely to die as a result of luminal A-like and luminal B-liketumors, respectively (Warner et al., 2015). Even less-than-weekly religiousservice attendance was positively associatedwith annual mammograms (1.34), breast self-exams (1.14), PAP smears (1.22), non-smoking (1.
41) (Salmoirago-Blotcher et al., 2011). This study will address the researchquestion: Is religious affiliation(Muslim, Christian, Jewish, versus none) a protective factor for breast cancerdeath among African American women?The long-term goal is to determine ifreligious affiliation reduces breast cancer mortality in AfricanAmerican women. The specific objective of this proposal is todetermine if one particular religious affiliation (Muslim, Christian, Jewish,versus none) and a specific worship frequency (never, less than monthly,weekly-monthly, more than weekly) areduced risk of breast cancer death.
The conceptual hypothesis is that the affiliation with a religionincreases the probability of survival in African American women with breastcancer. Theoperational hypothesis is that in aWashington, DC, Prince Georges County and Montgomery County, MD hospital-basedsample of recently diagnosed AfricanAmerican women with breast cancer ages 18-60, those who attend religiousservice more than weekly will report decreased mortality over a 2 yearfollow-up period when compared to a comparable group of women with no religiousaffiliation.This hypothesiswas formulated based on preliminary data,which states that African American women who attend religious services engagein protective behaviors like mammograms and breast exams. There are racialdisparities in the stage of diagnosis between black women and white women, only 52% of black women are diagnosed at a local stage versus63% of white women (“Report,” n.d.). Late stage diagnosis has been attributed to a lowerfrequency of mammograms (“Report,” n.d.
). We are proposing a 2-yearprospective cohort study on newlydiagnosed African American women with breast cancer (within the previoussix months of study initiation) in Washington, DC, Prince Georges County andMontgomery County, MD. In2011, the 5-year survival rate among African American women with breast cancer was 80%, making incidence on deathnot rare (“Report,” n.d.). Cohort studies are not suitable for rare diseases.
Thusthis had to be ruled out when a picking study design. A cohort design was chosen to eliminatetemporal ambiguity. Also, if we did a retrospective design or case-control, we would have to rely onpotentially incomplete data or proxy answersbecause the outcome is death. Since religion can be personal, a proxy is notideal. Also, having a proxy complete aninterview for only the fatalitieswould create information bias due to differences in measurement. Becauseprevious studies noted that intervention is particularly vital in the first twoyears, the induction period is not an unrealistic study period for aprospective cohort and there should be a minimalloss to follow-up due to migration (Warner et al.
, 2015). The outcomeof interest is death due to breast cancer.The outcome will be measured by acategorical variable “Breast Cancer Mortality,” yes=1 and no=0. Thisinformation will be gathered from deathcertificates and confirmation calls to the participant’sphysicians.
The exposure is religiousaffiliation. Participants will beinterviewed at baseline, after screening. Religious affiliation will bemeasured by two categorical variables “Affiliation” and “WorshipFrequency.” Religious affiliation willprovide the following choices Muslim, Christian, Jewish, and None, andparticipants will be asked which belieffits them most. Then participants will be asked about “Worship Frequency,” and asked how often they attend religiousservices (never, less than monthly, weekly-monthly, more than weekly).
Modifiersare variables that enhance the effect of the exposure on the outcome.There are a few potential modifiers,including the stage of diagnosis, treatment method, and smoking. AJCCclinical stage of breast cancer will be notedat baseline interview, a categorical variable ofstages 0, 1, 2, 3, and 4. Treatment method will be noted during follow-up interviews,a categorical variable of surgery, radiation, chemotherapy, andother. Prevalence of smoking measured at baseline with a categorical variable of never smoked, quit smoking,social smoking, frequent smoking.
Previously we noted the increased protectiveeffect of religious affiliation on smokingand mammograms. The correlation of mammograms and early stage diagnosis and survivabilitywere also noted.Thus, these protective factors will probably modify the affect religiousaffiliation has on breast cancer survival.