Self-Report and Primary Care Medical Record Documentation of Mammography: METHODS
The study population included women 40 years of age and older who were established patients for at least one year (according to their medical record) attending two family practice health centers located in poor urban areas of Buffalo, NY. One health center served predominantly an African-American population and the other served a predominantly Puerto Rican population based on established practice demographics. Both sites provided a full spectrum of family practice, including gynecologic and obstetric care. The majority of mammograms were ordered by the primary care office but were completed at off-site locations.
Women presenting for either acute or preventive healthcare were consecutively asked to participate at each site. Five-hundred-ten women presented to the clinics. Forty-five women were ineligible due to mental deficiencies, speaking a language other than English or Spanish and/or past cancer diagnosis. Forty-one women, unable to be contacted at their visit due to visits occurring at the same time or readiness to be seen by their doctor, were unable to be contacted after their visit and were considered lost to follow-up. Eighty-one women refused to participate. Three-hundred-forty-three women completed interviews. These analyses were restricted to include data on 332 women that completed an interview and had a medical record available for review. Medical records were unavailable for 11 women. Women were asked to identify their race and ethnicity as separate questions. Women with an unknown race or self-identified “other” race, such as American Indian or Asian, were excluded from the analyses (N=18). Women who identified themselves as Hispanic were asked to report their country of origin. Only three women reported a country of origin other than Puerto Rico. Race and ethnicity variables were then combined to establish our final definition of race/ethnicity. In sum, analyses for this study were conducted on 314 women, self-identified as Puerto Rican, African-American or non-Latina white, who completed the face-to-face interview and had a medical record available for review for a response rate of 68%. In total, 79 non-Latina white women, 114 African-American women and 121 Puerto Rican women were included in this analysis. Eighty-six percent of the women received their gynecologic care from the family practice office.
buy kamagra tablets
A face-to-face survey was administered in the language preferred by the patient (English or Spanish). The survey was translated to Spanish and then back-translated to English to ensure accuracy of translation. Thirty-one percent of the surveys were administered in Spanish. This survey assessed screening practices and basic demographics. Demographic data reported was based on self-report from the face-to-face interview. Variables included age, employment status, household income level, marital status and education level. Place of care was also included to account for potential differences in site practices. All demographics were reported as categorical data. Demographics were compared across racial/ethnic groups. Chi-squared analyses were used to compare demographic characteristics across racial/ethnic groups. Percentages and p values based on chi-squared analyses were reported.
The analyses reported here focus on a series of questions that assessed screening habits from the face-to-face interview. Questions from the survey included:
“Have you ever had a Pap/mammogram?” (Yes/No)
“When was your last Pap/mammogram?” (<1 year ago/1-2 years ago/>2 years ago/<l year ago/1-3 years ago/>3 years ago)
“Did you have more than one Pap/mammogram?” (Yes/No), and
“When was your Pap smear/mammogram before the last one?” (<1 year ago/1-2 years ago l>2 years ago/<l year ago/1-3 years ago />3 years ago).
The responses to questions regarding the timing of Pap/mammograms were dichotomized as being up-to-date and not-up-to-date based on age and the 1998 American Cancer Society guidelines. All women who received Pap smears within the past three years were considered up-to-date. For mammography, women 50 and older who received a mammogram within the past year were up-to-date, and women 40 to 49 were up-to-date if they received a mammogram within the past two years. cialis canadian pharmacy
Dates and results of mammograms and Pap smears were recorded directly from laboratory reports found in the primary care record for up to five years past by trained medical students. Data collected from medical records was dichotomized as up-to-date or not up-to-date. If information was not available in the chart for mammography and/or Pap smears, it was assumed that the women never received the test.
Positive self-report and positive medical record documentation of being up-to-date with screening were reported for Pap smears and mammography use. The percent agreement between these two sources also was reported. Sensitivity, specificity, positive predictive value and negative predictive value were reported for each comparison to identify the ability of self-report to accurately reflect behavior as measured against medical record documentation. Sensitivity reflects the probability of chart documentation of screening among women who reported being screened. Specificity refers to the probability of no chart documentation of screening among women that reported not being screened. Positive predictive value reflects the probability of women that reported screening that did have chart documentation of screening. Negative predictive value is the probability of women that reported not being screened that did not have documentation of screening in their medical record. All analyses were stratified by racial/ethnic group.
buy antibiotics online without prescription
The Institutional Review Board of the School of Medicine and Biomedical Sciences, State University of New York granted human subjects approval for this study. All participants signed an informed consent for a verbal interview and access to their medical records prior to their participation.
