uruguayan women’s breast canceruruguayan women’s breast cancer knowledge: using novel modeling...
TRANSCRIPT
Uruguayan Women’s Breast Cancer Knowledge: Using Novel Modeling
Techniques
Deborah E. Linares M.A.1, Lydia P. Buki Ph.D.2,
Micaela Reich, Ph.D.3, Selva Sanabria4, & Dolores Larrosa4
1University of Illinois at Urbana-Champaign, 2University of Miami, 3Universidad Católica del Uruguay, 4Comisión Honoraria de Lucha Contra el Cáncer
• Highest breast cancer mortality rate in Latin America and continues to increase over time (Ferlay et al., 2010; WHO, 2012)
Breast Cancer in Uruguay
2
24.3
20.1
14.7
12.3
0
5
10
15
20
25
30 2008 Breast Cancer Standardized Age-Adjusted Mortality Rates by Country
per 100,000 Women
Uruguay Argentina United States Brazil
20.7
24.3
18
19
20
21
22
23
24
25
Uruguay Breast Cancer Standardized Age-Adjusted Mortality Rates by Year per 100,000
Women
2004 2008
http://www.veomapas.com/mapa-politico-de-los-departamentos-del-uruguay-m280.html
(Ferlay et al., 2010; WHO, 2012)
• Overall, research in this area is sparse
• In the U.S., women who have more breast cancer knowledge are more likely to:
• Be younger (Jimenez et al., 2011)
• Have children (Bird et al., 2010)
• Have higher levels of formal education (Ramirez et al., 2000)
• Obtain mammography screenings (Harris et al., 2003)
• Communities shape women’s breast cancer knowledge levels by hosting educational programs, providing access to information, and influencing the quality of resources available (Bigby, 2007)
Breast Cancer Knowledge in Uruguay
3
Health Literacy
4
Health Literacy
Cultural & Conceptual Knowledge
Oral Literacy Print Literacy Numeracy
(IOM, 2002)
Cultural & Conceptual Knowledge
Cultural and Conceptual Knowledge
5
Cultural & Conceptual Knowledge
Attitudes
Beliefs
Emotions
Knowledge
Cultural and conceptual knowledge represents the filter through which individuals obtain, process and understand
health information and options for diagnosis and treatment (IOM, 2002)
(Buki & Quintero Johnson, 2009)
Examine the psychometric properties of the breast cancer knowledge subscale and explain the influence of individual level variables (mammography screening status, age, education level, area of residence) on participant characteristics and item responses.
Aim of the Study
6
• Recruitment and data collection were done in collaboration with the Comisión Honoraria de Lucha Contra el Cáncer
• National sample of 410 women ages 40-65 years was obtained from 5 departments: Artigas, Flores, Maldonado, Montevideo, and Treinta y Tres
• Approximately half of participants were up to date for mammography screening (
2 years)
Sample
7
Mean %
Age 51 years -
Education level
10.43 years
-
Have Children
- 92%
• Within the Item Response Theory (IRT) family
• Innovative measurement approach using generalized linear and non-linear mixed models (De Boeck & Wilson, 2004)
• Models how properties of items and participant characteristics influence item responses
• Provides information about participants to help determine differences in knowledge levels
• Appropriate method for handling measurement error and dependency within data
• Protects against inflating standard error
Explanatory Item Response Modeling (EIRM)
8
• Within the Uruguayan version of the Cancer Health Literacy Measure – Breast and Cervical Cancer (Buki & Reich,
2012)
• 12 items
• Sample item: “Younger women have a higher risk of getting breast cancer than older women.”
• Response format of “yes,” “no,” and “I don’t know”
• “I don’t know” responses collapsed into “no” responses for the analyses
• Alpha and Kuder-Richardson coefficients of reliability:
• α =.61, KR-20 = .62
Breast Cancer Knowledge Subscale
9
Y1 Y2 Y3 Y4 Y5 Y6 Y7 Y8 Y9 Y10 Y11 Y12
Age x1
Mammography
status x2
Residency x3
Education x4
Having Children x5
BC Knowledge
• A 2PL EIRM was fit to the data using SAS 9.3 PROC NLMIXED (SAS Institute, 2011)
• -2 Log Likelihood = 27551, AIC = 27593, BIC = 27725
• Likelihood ratio test indicated the model is significantly different, p <.0001
• Items 2, 7, 10, and 12 were dropped which improved item fit
• Women who were up to date for a mammogram, had children, and were from Flores and Artigas were more likely to have higher BC knowledge levels, p <.0001
• Women from Maldonado were more likely to have lower levels of BC knowledge, p <.0001
Results
10
a1 b1 a3 b3 a4 b4 a5 b5 a6 b6 a8 b8
Mammography
status x4
(OR = 2.364)
Having Children x2
(OR = 1.978)
a9 b9 a11 b11
Maldonado x1
(OR =.794)
Artigas x5
(OR = 1.612)
Flores x3
(OR = 1.612)
= decrease in BC knowledge
• Having children, mammography status, and women’s residency, explained breast cancer knowledge item responses
• Demonstrate utility of EIRM for health literacy measurement development
• Provide information on screening barriers and facilitators within a universal healthcare system
Implications and Future Directions
• Develop educational interventions to increase BC knowledge targeting women (40-65 years) who have not obtained a mammogram screening within 2 years, do not have children, and reside in Maldonado
Conclusion
11
Bigby, J. A. (2007). The role of communities in eliminating healthcare disparities: Getting down to the grassroots. In R. A. Williams (Ed.), Eliminating healthcare disparities in America: Beyond the IOM report. Totowa, NJ: Humana Press Inc.
Bird, Y., Moraros, J., Banegas, M.P., King, S., Prapasiri, S., & Thompson, B. (2010). Breast cancer knowledge and early detection among Hispanic women with a family history of breast cancer along the U.S.-Mexico border. Journal of Health Care for the Poor and Underserved, 21(2), 475-488.
Buki, L. P., & Reich, M. (2012). Adaptation and validation of the CLM-BCC for Uruguayan women. Unpublished manuscript. Buki, L. P., & Quintero Johnson, J. M. (2009, October). Cultural and conceptual knowledge: Conceptual and measurement
advances in health literacy. Poster presented at the meeting of the Health Literacy Annual Research Conference, Washington, D.C.
De Boeck, P., & Wilson, M. (2004). Explanatory item response models: A generalized linear and nonlinear approach. In P. De Boeck & M. Wilson (Ed.). New York City: Springer.
Ferlay, J., Shin, H. R., Bray, F., Forman, D., Mathers, C., & Parkin, D. M. (2010). Cancer incidence and mortality worldwide: IARC CancerBase no 10. [GLOBOCAN 2008]. International Agency for Research on Cancer. Retrieved from http://globocan.iarc.fr
Harris, D., Miller, J., Davis, D. (2003). Racial differences in breast cancer screening, knowledge and compliance. Journal of the National Medical Association, 94(8), 693-701..
Institute of Medicine. (2002). Health literacy: A prescription to end confusion. Washington, D.C.: The National Academies Press. Jimenez, E. A., Xie, Y., Goldsteen, K., & Chalas, E. (2011). Promoting knowledge of cancer prevention and screening in an
underserved Hispanic women population: A culturally sensitive education program. Health Promotion Practice, 12(5), 698-695.
Ramirez, A., Suarez, L., Laufman, L., Barroso, C., & Chalela, P. (2000). Hispanic women’s breast and cervical cancer knowledge, attitudes, and screening behaviors. American Journal of Health Promotion, 14(5), 292-300.
Ratzan, S. C., & Parker, R. M. (2000). Introduction. In C. R. Selden, M. Zorn, S. C. Ratzan, & R. M. Parker (Eds.), National library of medicine current bibliographies in medicine: Health literacy. Bethesda, MD: National Institutes of Health, US Department of Health and Human Services: NLM Pub. No. CBM 2000-1.
SAS Institute. (2011). In Baxter A. (Ed.), Base SAS® 9.3 procedures guide. Cary, NC: SAS Institute Inc. World Health Organization (2012). Mortality database. Retrieved from
http://www.who.int/whosis/mort/download/en/index.html
References
12
Parameters Est. Std
Error df t value p value
b1/a1 -1.62 0.09 4007 -18.05 <.0001
b3/a3 4.07 0.68 4007 5.95 <.0001
b4/a4 -2.80 0.19 4007 -14.61 <.0001
b5/a5 -2.75 0.19 4007 -14.77 <.0001
b6/a6 -1.58 0.05 4007 -29.50 <.0001
b8/a8 -2.30 0.12 4007 -18.81 <.0001
b9/a9 -1.92 0.12 4007 -15.92 <.0001
b11/a11 -2.48 0.32 4007 -7.81 <.0001
Final Model Parameter Estimates
13