Skip to content

Empowering Latinas to Obtain Breast Cancer Screenings

Empowering Latinas to Obtain Breast Cancer Screenings

Status
Completed
Phases
NA
Study type
Interventional
Source
ClinicalTrials.gov
Registry ID
NCT02964234
Enrollment
145
Registered
2016-11-16
Start date
2017-03-01
Completion date
2021-08-31
Last updated
2021-11-18

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Breastcancer, Breast Diseases

Keywords

Mammography Screening, Latinas, Breast cancer disparities

Brief summary

The participatory-based project will quantify the 'added benefit' of an empowerment intervention relative to an education intervention for 150 Latinas on the following outcomes: women's adherence to breast cancer screening guidelines; women's psychosocial facilitators (self-efficacy, norms, support, and knowledge); and women's dissemination of breast health messages throughout their social network. The empowerment intervention will train Latinas in how to discuss breast health with their family and friends and volunteer in local breast health promotion programs. Academic, clinician, and community partners will work together throughout intervention development and evaluation.

Detailed description

Latinas suffer disproportionately from breast cancer relative to non-Latina Whites (NLWs), including late stage detection. While there have been controversies in breast cancer screening, non-adherence to guideline-concordant screening continues to be a major modifiable determinant of breast cancer outcome disparities. Thus, increasing participation in breast cancer screening among Latinas, especially care that corresponds with clinical and academic guidelines, is a public health priority. Participatory approaches are popular methods to improve screening within this group and have included approaches that 1) deliver education to non-adherent Latinas and 2) train community health advocates (community health workers, breast cancer survivors) to engage in breast health promotion. The second approach (empowerment interventions) concerns training participants to engage in social outreach (e.g., having conversations with family and friends about breast health) and volunteering (e.g., helping in health fairs, engaging in civic campaigns about breast cancer programs). Patient activation and volunteerism literature suggest that empowerment interventions may have 'added value' for participants themselves over delivering education in terms of preventive health psychosocial factors and practices. Relative to education interventions, empowerment interventions may also affect women's networks, as they may be more likely to disseminate evidence-based breast health promotion among their family and friends. To date, little research has compared interventions' effects on individual-level outcomes or used formal social network analysis to examine network effects. The proposed work adds to the literature through empirically comparing two approaches (education versus empowerment) on three sets of outcomes: 1) women's own screening, 2) women's own self-efficacy, norms, support, knowledge; and 3) women's networks (measured by egocentric analysis). I will lead this work and will benefit from the collective expertise and resources of my mentors (Drs. Ferrans, Mermelstein, Geller) and collaborators (Dr. Schneider, The Resurrection Project, Metropolitan Chicago Breast Cancer Task Force, Sinai Urban Health Institute, University of Illinois Cancer Center). Aim 1 intervention development will be accomplished through continuous stakeholder engagement and specifically through meetings and focus groups with a bilingual, bicultural community advisory engagement board (CEAB), UICC radiologists and target participants (non-adherent Latinas). We have already begun this process and are obtaining formative data. After we develop intervention materials, a pilot trial will be conducted with an area-level treatment control group design in Chicago. Participants will be 150 Latinas with no history of health volunteerism, residence in one of two targeted areas, and non-adherence to US Preventive Services Task Force screening guidelines. Aim 1 intervention evaluation will involve an analysis to compare differences in receipt of a medical record-confirmed screening within six months of participating in the study. Aim 2 will be an analysis to compare interventions' effects on self-efficacy, norms, support, knowledge across three time points - pre-intervention, immediately post-intervention, and six months post-intervention. Aim 3 will be a social network analysis, specifically egocentric, to compare interventions' effects on breast health, network size, and network density as well as will number of peers referred to the study.

Interventions

BEHAVIORALEmpowerment

Three group sessions (breast cancer education; communication; volunteerism) 1.5 hours 3 times across 3 weeks

BEHAVIORALEducation

Three group sessions (breast cancer education; diet; physical activity) 1.5 hours 3 times across 3 weeks

Sponsors

The Resurrection Project
CollaboratorOTHER
Juan Diego Centro Comunitario
CollaboratorOTHER
National Cancer Institute (NCI)
CollaboratorNIH
University of Illinois at Chicago
Lead SponsorOTHER

Study design

Allocation
NON_RANDOMIZED
Intervention model
PARALLEL
Primary purpose
SCREENING
Masking
NONE

Eligibility

Sex/Gender
FEMALE
Age
52 Years to 75 Years
Healthy volunteers
Yes

Inclusion criteria

* Age 52-75 years old; * Identification as Latina/Hispanic/Chicana female; * Residence in Pilsen, Little Village, East Side or South Chicago; * No history of health volunteerism; * No history of breast cancer; and * Lack of a mammogram within the last two years

Exclusion criteria

* Not meeting all inclusion criteria; * Women will be excluded if they participated in formative focus groups

Design outcomes

Primary

MeasureTime frameDescription
Number of Participants Who Have Obtained Breast Cancer Screening6 monthsReceipt of mammogram based on medical records and self report within 6 months of baseline survey (Yes or No)

Other

MeasureTime frameDescription
Changes in Psychosocial Facilitators of Screening Survey MeasuresBaseline and 6 monthsTotal scores/ranges are used. No subscales used. Knowledge - 5 items (Williams et al., 2011). Range for the change scores (baseline to 6-months) is -5 to 5. Higher scores represent a better outcome. Cultural beliefs - Ferrans cultural beliefs scale (Ferrans et al., 2007). Range for the change scores (baseline to 6-months) is -17 to 17. Higher scores represent a worse outcome. Breast cancer-specific self-efficacy - Mammography-Specific Self-Efficacy Scale (Champion, Skinner, & Menon, 2005). Range for the change scores (baseline to 6-months) is -26 to 26. Higher scores represent a better outcome. Positive breast cancer screening norms - 6 items ( Molina et al., 2015). Range for the change scores (baseline to 6-months) is -12 to 12. Higher scores represent a better outcome. Breast cancer supportive social network size questions comprised an 8-item version of Berkman-Syme index. Range for the change scores (baseline to 6-months) is -25 to 25. Higher scores represent a better outcome.
Number of Individuals to Whom Participants Exchanged Information About Breast Cancer Screening.6 monthsThis was an open-ended questionnaire, based on the Burt Social Network Instrument. Higher numbers represent better outcomes.

Countries

United States

Participant flow

Participants by arm

ArmCount
Empowerment
Behavior: Empowerment Empowerment: Three group sessions (breast cancer education; communication; volunteerism) 1.5 hours 3 times across 3 weeks
76
Education
Behavior: Education Education: Three group sessions (breast cancer education; diet; physical activity) 1.5 hours 3 times across 3 weeks
69
Total145

Withdrawals & dropouts

PeriodReasonFG000FG001
Overall StudyLost to Follow-up2213

Baseline characteristics

CharacteristicEmpowermentEducationTotal
Age, Continuous61.05 years
STANDARD_DEVIATION 6.29
61.28 years
STANDARD_DEVIATION 6.36
61.16 years
STANDARD_DEVIATION 6.3
Ethnicity (NIH/OMB)
Hispanic or Latino
76 Participants69 Participants145 Participants
Ethnicity (NIH/OMB)
Not Hispanic or Latino
0 Participants0 Participants0 Participants
Ethnicity (NIH/OMB)
Unknown or Not Reported
0 Participants0 Participants0 Participants
Insurance24 Participants40 Participants64 Participants
Mammogram 4+ years ago22 Participants30 Participants52 Participants
Mammography intention within next year48 Participants40 Participants88 Participants
Region of Enrollment
United States
76 participants69 participants145 participants
Sex: Female, Male
Female
76 Participants69 Participants145 Participants
Sex: Female, Male
Male
0 Participants0 Participants0 Participants

Adverse events

Event typeEG000
affected / at risk
EG001
affected / at risk
deaths
Total, all-cause mortality
0 / 763 / 69
other
Total, other adverse events
0 / 760 / 69
serious
Total, serious adverse events
0 / 760 / 69

Outcome results

Primary

Number of Participants Who Have Obtained Breast Cancer Screening

Receipt of mammogram based on medical records and self report within 6 months of baseline survey (Yes or No)

Time frame: 6 months

Population: Complete case analysis used - only participants who were retained until 6 month follow-up.

ArmMeasureValue (COUNT_OF_PARTICIPANTS)
EmpowermentNumber of Participants Who Have Obtained Breast Cancer Screening49 Participants
EducationNumber of Participants Who Have Obtained Breast Cancer Screening29 Participants
Comparison: We conducted logistic regressions, clustering by cohort, given both interventions used a group format, and adjusting for age, education, income, insurance status, baseline mammography history, and baseline mammography intention. The null hypothesis was there would be no study arm differences. A priori power analysis suggested that a sample size of 150, assuming alpha = .05, power = .80 would lead us to detect medium/large effects (OR = 2.8).p-value: 0.00595% CI: [2.01, 56.31]Regression, Logistic
Other Pre-specified

Changes in Psychosocial Facilitators of Screening Survey Measures

Total scores/ranges are used. No subscales used. Knowledge - 5 items (Williams et al., 2011). Range for the change scores (baseline to 6-months) is -5 to 5. Higher scores represent a better outcome. Cultural beliefs - Ferrans cultural beliefs scale (Ferrans et al., 2007). Range for the change scores (baseline to 6-months) is -17 to 17. Higher scores represent a worse outcome. Breast cancer-specific self-efficacy - Mammography-Specific Self-Efficacy Scale (Champion, Skinner, & Menon, 2005). Range for the change scores (baseline to 6-months) is -26 to 26. Higher scores represent a better outcome. Positive breast cancer screening norms - 6 items ( Molina et al., 2015). Range for the change scores (baseline to 6-months) is -12 to 12. Higher scores represent a better outcome. Breast cancer supportive social network size questions comprised an 8-item version of Berkman-Syme index. Range for the change scores (baseline to 6-months) is -25 to 25. Higher scores represent a better outcome.

Time frame: Baseline and 6 months

Population: This is a complete case analysis for outcome data, with all available data being used for participants who completed the 6-month follow-up.

ArmMeasureGroupValue (MEAN)Dispersion
EmpowermentChanges in Psychosocial Facilitators of Screening Survey MeasuresSupportive Network size1.38 units on a scaleStandard Deviation 4.66
EmpowermentChanges in Psychosocial Facilitators of Screening Survey MeasuresCultural Beliefs-1.36 units on a scaleStandard Deviation 2.94
EmpowermentChanges in Psychosocial Facilitators of Screening Survey MeasuresSelf-Efficacy1.02 units on a scaleStandard Deviation 4.67
EmpowermentChanges in Psychosocial Facilitators of Screening Survey MeasuresSocial Norms0.25 units on a scaleStandard Deviation 2.42
EmpowermentChanges in Psychosocial Facilitators of Screening Survey MeasuresKnowledge0.98 units on a scaleStandard Deviation 1.6
EducationChanges in Psychosocial Facilitators of Screening Survey MeasuresSelf-Efficacy-1.00 units on a scaleStandard Deviation 5.99
EducationChanges in Psychosocial Facilitators of Screening Survey MeasuresKnowledge1.37 units on a scaleStandard Deviation 1.69
EducationChanges in Psychosocial Facilitators of Screening Survey MeasuresCultural Beliefs-2.41 units on a scaleStandard Deviation 3.13
EducationChanges in Psychosocial Facilitators of Screening Survey MeasuresSocial Norms0.32 units on a scaleStandard Deviation 2.94
EducationChanges in Psychosocial Facilitators of Screening Survey MeasuresSupportive Network size5.85 units on a scaleStandard Deviation 2.11
Comparison: Analyses were conducted using GEE with an exchangeable correlation structure and a gamma distribution with log link. All models included study arm, time, and the study arm\*time. Covariates included age, education, and mammography history. With 150 Latinas, alpha = 0.05, power = 0.80, ICC = 0.0-0.05, we expected to detect small/medium interaction effects.p-value: 0.0395% CI: [-0.34, -0.04]Regression, Linear
Other Pre-specified

Number of Individuals to Whom Participants Exchanged Information About Breast Cancer Screening.

This was an open-ended questionnaire, based on the Burt Social Network Instrument. Higher numbers represent better outcomes.

Time frame: 6 months

Population: Complete case analysis used, wherein we focus on participants who completed the 6-month follow-up surveys.

ArmMeasureValue (MEAN)Dispersion
EmpowermentNumber of Individuals to Whom Participants Exchanged Information About Breast Cancer Screening.8.27 family/friends engagedStandard Deviation 6.17
EducationNumber of Individuals to Whom Participants Exchanged Information About Breast Cancer Screening.5.81 family/friends engagedStandard Deviation 12.58
Comparison: We used ordinal regression, given the non-normal distribution revealed by preliminary analyses. Analyses clustered by cohort and adjusted for age, SES (education, income, insurance), mammography history, mammography intention, and baseline supportive breast cancer social network size. Power analyses suggested that with 150 Latinas, alpha = .05, power = .80, we would be able to detect a small effect (Cohen's f = 0.05).p-value: <0.000195% CI: [2.82, 13.32]Ordinal regression

Source: ClinicalTrials.gov · Data processed: Feb 4, 2026