 |
|
Predictors of Screening Mammography
Implications for Office Practice
Doyle M. Cummings, PharmD, FCP, FCCP;
Lauren Whetstone, PhD;
Amy Shende, MA;
David Weismiller, MD, ScM
Arch Fam Med. 2000;9:870-875.
ABSTRACT
| |
Background The completion of annual screening mammography and other preventive health services among women aged 50 years and older remains an important quality of care indicator.
Methods A biracial sample of 843 rural women (aged  50 years) from a population-based sample reported demographic and preventive health services utilization in the last year including the completion of screening mammography. Bivariate analysis and logistic regression were used to investigate the extent to which completion of other screening examinations, including Papanicolaou (Pap) smears and clinical breast examination, is associated with successful completion of mammography relative to demographic and health service variables.
Results The completion of mammography was associated with age, race, education, health insurance, and the presence of a regular primary care physician, but the strongest predictors were the completion of a clinical breast examination and/or a Pap smear.
Conclusions Women who receive a clinical breast examination and/or a Pap smear appear far more likely to receive screening mammography, suggesting a synergy in screening services. The relative efficacy of interventions to increase the completion of clinical breast examinations as well as other age-appropriate preventive services during routine office visits or during a single preventive services office visit should be further explored in primary care settings. Residency programs should provide training on the successful incorporation of such services into office practice patterns in an effort to continually improve quality of care.
INTRODUCTION
THE PROPORTION of women older than 50 years who receive annual screening mammography has been identified as an important quality indicator by several health insurance and health maintenance organizations.1-2 Mammography is chosen because breast cancer is the most commonly diagnosed cancer and the second most common cause of cancer death among women in the United States.3 Further, screening mammography has been shown to decrease breast cancer–related mortality rates in women between the ages of 50 and 69 years.4 Lower rates of screening mammography are associated with a more advanced stage at diagnosis of breast cancer among older African American women when compared with older white women.5 As reviewed in the article by Salive et al,6 mammography is recommended annually beginning at age 50 years by many policy groups. The US Preventive Services Task Force recommends screening mammography every 1 to 2 years for women aged 50 to 69 years, alone or in conjunction with a clinical breast examination.7
Several authors have explored the factors associated with successful completion of mammography. Among these factors have been the influence of demographic characteristics such as age and race,8-10 the perceptions and beliefs of women,11-12 and the practice patterns of physicians.13-16 Our interest is in the relative importance of these factors as effective predictors of which women are likely to successfully complete mammography. McCarthy et al8 described the relative importance of selected demographic and health care delivery factors on completion of screening mammography among women in a health maintenance organization in Michigan. Their findings suggest that factors surrounding health care delivery, such as having an office visit when due for a screening, the number of office visits, and the specialty of the provider, had a greater association with mammography use than did demographic characteristics. However, their study only included women who were already insured and enrolled in a health care delivery system. Whether these factors remain important in the broader population could not be elucidated from their report. In a population-based sample from the 1990 National Health Interview Survey, Pearlman et al9 carefully analyzed the impact of a variety of demographic and health care delivery variables on breast cancer screening practices among African American, Hispanic, and white women. These authors showed, in both stratified and unstratified models, that the completion of other preventive services, notably a clinical breast examination and Papanicolaou (Pap) smear, was an important predictor of who completed screening mammography. Further, demographic characteristics such as age and income, which appeared to be important correlates of mammography completion in unstratified models, were not well correlated in race-stratified models. Although population-based, these authors describe the use of uneven sample sizes by race as a limitation of their study, and recommend that additional research be done with more racially balanced samples. Further, data for the study were collected in 1990 and several programs have developed since then that are designed to increase screening mammography among targeted population subgroups, suggesting the need for reanalysis.
Few investigators have compared the influence of demographic and health system variables on mammography screening rates in a racially balanced, population-based sample with a variety of physicians who are not part of the same health care delivery system. Because mammography completion often requires the recommendation and referral of a provider as well as adherence by the patient,15 a population-based sample provides the unique opportunity to concurrently assess the impact of breast cancer screening information directed at both providers and the public at large. We examined the completion of screening mammography during a 1-year period among a representative and racially balanced sample of women older than 50 years (average age, 66 years) residing in 4 predominantly rural counties.
The purpose of the study was to investigate the extent to which completion of screening examinations, including Pap smear and clinical breast examination, is associated with completion of mammography relative to demographic and health service variables. Based on the preliminary findings of Pearlman et al,9 we hypothesized that completion of mammography would occur less often among women who were older, African American, uninsured, and without a regular physician and more often among women who completed a Pap smear or clinical breast examination. Such data can provide critical information for primary care physicians in the structuring of office practice and outreach programs to facilitate completion of recommended screening. Specifically, office systems as well as community efforts that encourage regular and on-time clinical breast examinations and Pap smears in eligible women can be promulgated and linked to screening mammography referral and follow-up.
SUBJECTS AND METHODS
The data for this study were obtained from the Rural Eastern Carolina Health (REACH) project, which has been previously described.17 This project was reviewed and approved by the University Medical Center Institutional Review Board.
RESPONDENTS
The REACH survey used population-based sampling methods to identify 2500 randomly selected households in 4 rural counties representative of the coastal plain of North Carolina. Respondents, typically the female head of the household, provided detailed health information for themselves and other household members. Surveys were administered face-to-face by trained interviewers. Only women aged 50 years and older who responded to the REACH survey were included in the present analysis.
DATA
Demographic data included age, race, education, county of residence, and a history of breast cancer in parents. Health care access and utilization variables included the presence of health insurance, the presence of a regular primary care provider, the number of physician visits in the last year, and the self-reported completion of clinical breast examination and/or Pap smear in the last year.
ANALYSIS
The self-reported completion of a mammogram in the last year was the outcome variable of interest. The accuracy of self-reports regarding mammography in the last year has been previously studied and demonstrated to be reliable.18 We sought to identify the extent to which the completion of a Pap smear and/or completion of a clinical breast examination in the last year was associated with the completion of a mammogram in the last year. The relationship between completion of mammography and the above-listed variables was initially examined using contingency tables and  2 tests. These analyses were followed by multiple logistic regression models with completion of mammography as the outcome variable to assess the independent effects of the demographic, health care access, and utilization variables listed above. The predictive value of the completion of clinical breast examinations and Pap smears was compared with that of demographic variables such as race and insurance coverage, which have been associated with mammography completion in earlier studies. One of the counties (n = 206) was not included in the logistic regression analyses as the health care access questions were not asked in the survey for that county.
RESULTS
A total of 843 respondents were included in the present analysis. The characteristics of the study population are presented in Table 1. The sample was well balanced racially. Of the 843 women older than 50 years, 429 (51%) reported having had a mammogram in the last year. A similar percentage (51%) reported having had a Pap smear, while 48% reported having had a clinical breast examination.
|
|
|
|
Table 1. Characteristics of Study Respondents
|
|
|
BIVARIATE ANALYSIS
There was a strong and significant relationship between having had either a Pap smear or a clinical breast examination and having had a mammogram in the last year (Pap smear 2 = 278.86, P<.001 and clinical breast examination, 2 = 330.59, P<.001). Figure 1 shows the percentage of women who reported having had a mammogram in the last year in relation to whether they also reported having had a Pap smear and/or a clinical breast examination.
|
|
|
|
Likelihood of mammography with other preventive screening. Pap indicates Papanicolaou smear; CBE, clinical breast examination.
|
|
|
There was no significant difference in mammography completion by race overall, while modest differences existed by age, with a larger percentage of women aged 50 to 64 years reporting completion of a mammogram than women aged 65 years or older (2 = 5.16, P<.05). African American women aged 65 years or older had the lowest reported mammography rates (42%), while white women aged 50 to 64 years had the highest rates (58%). The percentage of women who reported having completed a mammogram increased with each level of education (2 = 13.88, P<.01). There was also a significant difference among the 4 counties, with 3 counties having similar rates and 1 county substantially lower rates (2 = 9.14, P<.05). Finally, a parental history of breast cancer had only a small and nonsignificant effect on mammography screening.
We also assessed the relationship between mammography completion and a series of health care access and utilization variables (Table 2). Not surprisingly, a larger percentage of women with health insurance had completed mammograms than women without health insurance (2 = 17.19, P<.001). Similarly, a larger percentage of women who reported having a regular primary care physician had had a mammogram in the last year than women who reported not having a regular source of primary care (2 = 10.83, P<.01). We examined the relationship between mammography completion and combinations of health insurance and having a primary care physician. Among women who reported having both a primary care physician and health insurance, 54% reported having had a mammogram in the last year. Among those with a regular physician but without health insurance, only 34% reported having had a mammogram in the last year. Among those women without a regular primary care physician but with health insurance, 33% reported having had a mammogram in the last year. Among women without either a regular primary care physician or health insurance, only 23% reported having had a mammogram in the last year. A smaller percentage of women making no office visits in the last year reported having had a mammogram than women who had one or more office visits in the last year (2 = 330.59, P<.001). We found no consistent pattern between the number of preventive health screenings completed (clinical breast examination, Pap smear, and mammography) and the number of times an individual reported visiting the primary care physician's office in the last year.
|
|
|
|
Table 2. Bivariate Analysis of the Association Between Demographic, Access, and Screening Variables and the Completion of Mammography
|
|
|
LOGISTIC REGRESSION
To determine the relative importance of these health services and demographic variables in terms of their predictive value for the completion of mammography in this population-based sample, we constructed 2 logistic regression models with completion of mammography as the outcome variable (Table 3). In the initial model (model 1) we included the following variables: age, race, educational level, county of residence, parental history of breast cancer, health insurance coverage, presence of a regular primary care physician, and the number of office visits in the last year. We found that age, educational level, health insurance, and having 1 or more office visits were significant predictors of having completed a mammogram in the last year. In the second model (model 2), we also included having completed a Pap smear and having received a clinical breast examination (Table 3). In this analysis, 3 of the previous predictive variables (age, educational level, health insurance) were no longer significantly associated with mammography completion. As noted in Table 3, having had a clinical breast examination was the strongest single predictor of whether a mammogram was done. Having had a Pap smear and having made 1 or more office visits to a physician in the last year were also significant predictors of completing a mammogram.
|
|
|
|
Table 3. Relative Predictive Value of Demographic, Health Care Access, and Screening Variables for the Completion of Mammography
|
|
|
COMMENT
BIVARIATE ANALYSIS
Much research has been undertaken on the completion of screening mammography as a function of age and race. Previous research19 has demonstrated that African American women are less likely to receive mammography than white women. While we found no significant difference by race overall, African American women aged 65 years or older had the lowest reported mammography rates. Burns et al19 demonstrated that more frequent primary care visits were a less important predictor of mammography completion for older African American women than for older white women. This issue remains a perplexing problem in that older African American women are at higher risk for diagnosis of late-stage breast cancer than older white women.20 Moreover, this late-stage disease apparently results in higher mortality rates in women older than 65 years,21 which might be prevented by appropriate screening. McCarthy et al5 demonstrated that mammography was an important factor in explaining this difference in the stage of disease at diagnosis. The reasons why older African American women do not obtain mammography have been and continue to be explored by various methods.5, 11, 22 Clearly, older African American women represent a high-risk group at which culturally relevant efforts should continue to be targeted.23 Such efforts should include an attempt to identify these women in primary care practices and to promote routine screening.
As noted in our findings, the presence of both a regular primary care physician and health insurance may be important enabling factors. The input of the primary care physician cannot be understated. Several studies have shown the effect the primary care physician has when he or she recommends screening mammography.24-25 This makes structuring the office practice environment to be conducive to preventive health screening all the more important.
Because cost is a consideration in mammography screening, Medicare and most other insurance carriers now provide some coverage for annual screening mammography costs for eligible women. Despite these efforts to address the financial access barrier, previous research has demonstrated a continuing deficit in mammography screening in older women who do not have supplemental insurance or who must make copayments.26 This factor may explain, in part, the lower mammography rates in older African American women in our sample. The presence of insurance alone is insufficient to ensure mammography completion, as our data and the work of Lannin et al11 demonstrate.
We found no consistent pattern between the number of preventive health screenings completed and the number of times a woman reported visiting the primary care physician's office. This again suggests that it is not the number of visits that is important, but rather that preventive screening such as mammography is systematically incorporated into routine practice patterns in primary care.
LOGISTIC REGRESSION
The regression models demonstrate a role for sociodemographic variables in influencing mammography completion; however, these variables are far less predictive when other health screening variables are included. This suggests that, when both the provider and the patient address the need and complete selected screenings, the likelihood that other screenings will be completed is greatly enhanced. This argues strongly that office practice and patient education should be configured to encourage the completion of all age-appropriate screenings.
Although perhaps dependent on socioeconomic status or other factors, the patient's agenda in any given office visit is often unrelated to preventive health screening. Further, there may be volume and quality tradeoffs in busy primary care practices. Data by Zyzanski et al27 suggest that, in high-volume practices, increased efficiency may come at the cost of lower rates of completion of preventive services. In addition, the periodic preventive services visit has not been widely accepted by physicians.6 We recommend that mammography and other preventive services be systematically incorporated into routine visits, which center around other priorities or into one or more preventive services visits. However, this approach was attempted with the "Put Prevention Into Practice" program, and results were less than favorable.28
The limitations of this study include those associated with the use of self-reported data. Because patients were seen in practices throughout a 4-county region and had a variety of insurance plans, review of medical records or administrative claims data was not feasible. As noted above, self-reported data on annual screening mammography completion has been demonstrated to be accurate. Further, this is a study undertaken among African American and white respondents in a predominately rural region and generalization to urban areas and other ethnic groups cannot be assumed. Data on mammography screening are for the previous 1-year period. Because the US Preventive Services Task Force recommends screening every 1 to 2 years in women aged 50 to 69 years, it is possible that some women were being routinely screened but were not screened during the study period. The recommendations differ, however, from those of the American Cancer Society, which recommends annual screening in this age group. Finally, no data are available on the physicians providing care for these women.
While progress has been made in increasing screening mammography, it is clear that additional efforts are required to minimize the all-too-common scenario in which the diagnosis of breast cancer is delayed until the disease is advanced and premature mortality occurs. Because of the substantial influence of primary care physicians on completion of screening mammography, it is incumbent on health care delivery systems to develop patterns of care that encourage the recommendation and completion of preventive health screening such as mammography.
Our and others' data show that women who receive a clinical breast examination or Pap smear are much more likely to receive screening mammography. However, of concern to us are the women in our sample who had one or more office visits, but who did not report completing mammography screening. Eligible women who are not being regularly screened should be targeted by culturally relevant programs that promote completion of screening. Although not always feasible, we believe these data support the proposition that there is synergy in bundling of preventive services. Our findings suggest the need for continued research on the best methods to incorporate preventive health screening into routine primary care practice. Additional research might assess the relative effectiveness of incorporating preventive health services such as mammography into a single visit, in which all age-appropriate services are bundled, or incorporating them into a series of routine visits that occur for other purposes. Further, we would recommend that primary care training programs emphasize the importance of preventive health screenings and the adoption of practice patterns that encourage their completion.
AUTHOR INFORMATION
Accepted for publication June 6, 2000.
We would like to acknowledge the financial support of the Duke Endowment, Charlotte, NC, the Health Resources and Services Administration, Rockville, Md, and Pitt County Memorial Hospital, Greenville, NC.
We also acknowledge the in-kind support of many communities, individuals, and agencies in a 4-county region who have contributed to our community health assessment program. We also thank Kristen Borre, MPH, PhD, for helping to design the sampling frame for our study, Christopher Mansfield, PhD, for reviewing the manuscript, and Linda Wagener for helping to prepare the manuscript for submission.
Corresponding author: Doyle M. Cummings, PharmD, FCP, FCCP, Department of Family Medicine, East Carolina University School of Medicine, 600 Moye Blvd, Greenville, NC 27858 (e-mail: cummingsd{at}mail.ecu.edu).
From the Department of Family Medicine, East Carolina University School of Medicine, Greenville, NC (Drs Cummings, Whetstone, and Weismiller), and the Research Triangle Institute, Triangle Park, NC (Ms Shende).
REFERENCES
| |
1. Rimer BK, Ross E, Balshem A, Engstrom PF. The effect of a comprehensive breast screening program on self-reported mammography use by primary care physicians and women in a health maintenance organization. J Am Board Fam Pract. 1993;6:443-451.
2. Himmelstein DU, Woolhandler S, Hellander I, Wolfe SM. Quality of care in investor-owned vs. not-for-profit HMOs. JAMA. 1999;282:159-163.
FREE FULL TEXT
3. National Cancer Institute. Racial/Ethnic Patterns of Cancer in the United States, 1988-1992. Bethesda, Md: National Cancer Institute; 1998.
4. Fletcher SW, Black W, Harris R, Rimer BK, Shapiro S. Report on the international workshop on screening for breast cancer. J Natl Cancer Inst. 1993;85:1644-1656.
FREE FULL TEXT
5. McCarthy EP, Burns RB, Coughlin SS, et al. Mammography use helps to explain differences in breast cancer stage at diagnosis between older African-American and white women. Ann Intern Med. 1998;128:729-736.
FREE FULL TEXT
6. Salive ME, Guralnik JM, Brock D. Preventive services for breast and cervical cancer in US office-based practices. Prev Med. 1996;25:561-568.
FULL TEXT
|
ISI
| PUBMED
7. Frame PS, Berg AO, Woolf S. US preventive services task force: highlights of the 1996 report. Am Fam Physician. 1997;55:567-576.
ISI
| PUBMED
8. McCarthy BD, Yood MU, MacWilliam CH, Lee MJ. Screening mammography use: the importance of a population perspective. Am J Prev Med. 1996;12:91-5.
9. Pearlman DN, Rakowski W, Ehrich B, Clark MA. Breast cancer screening practices among African-American, Hispanic, and white women: reassessing differences. Am J Prev Med. 1996;12:327-337.
ISI
| PUBMED
10. Makuc DM, Freid VM, Kleinman JC. National trends in the use of preventive health care by women. Am J Public Health. 1989;79:21-26.
FREE FULL TEXT
11. Lannin DR, Mathews HF, Mitchell J, Swanson MS, Swanson FH, Edwards MS. Influence of socioeconomic and cultural factors on racial differences in late-stage presentation of breast cancer. JAMA. 1998;279:1801-1807.
FREE FULL TEXT
12. Taylor VM, Taplin SH, Urban N, White E, Peacock S. Repeat mammography use among women ages 50-75. Cancer Epidemiol Biomarkers Prev. 1995;4:409-413.
ABSTRACT
13. Anderson LA, Janes GR, Jenkins C. Implementing preventive services: to what extent can we change provider performance in ambulatory care? a review of the screening, immunization, and counseling literature. Ann Behav Med. 1998;20:161-167.
ISI
| PUBMED
14. Murata PJ, Li JE. Relationship between Pap smear performance and physician ordering a mammogram. J Fam Pract. 1992;35:644-648.
ISI
| PUBMED
15. May DS, Kiefe CI, Funkhouser E, Fouad MN. Compliance with mammography guidelines: physician recommendation and patient adherence. Prev Med. 1999;28:386-394.
FULL TEXT
|
ISI
| PUBMED
16. Desnick L, Taplin S, Taylor V, Coole D, Urban N. Clinical breast examination in primary care: perceptions and predictors among three specialties. J Womens Health. 1999;8:389-397.
ISI
| PUBMED
17. Whetstone LM, Cummings DM. Population-based health assessment. Soc Sci Med. In press.
18. Zapka JG, Bigelow C, Hurley T, Ford LD, Egelhofer J, Cloud WM, Sachsse E. Mammography use among sociodemographically diverse women: the accuracy of self-report. Am J Public Health. 1996;86:1016-1021.
FREE FULL TEXT
19. Burns RB, McCarthy EP, Freund KM, et al. African-American women receive less mammography even with similar use of primary care. Ann Intern Med. 1996;125:173-182.
FREE FULL TEXT
20. Satariano WA, Belle SH, Swanson GM. The severity of breast cancer at diagnosis: a comparison of age and extent of disease in African-American and white women. Am J Public Health. 1986;76:779-782.
FREE FULL TEXT
21. SEER Program. SEER Cancer Statistics Review, 1973-1991. Bethesda, Md: US Dept of Health and Human Services; 1994. NIH publication 94-2789.
22. Mickey RM, Durski J, Worden JK, Danigelis NL. Breast cancer screening and associated factors for low-income African-American women. Prev Med. 1995;24:467-476.
FULL TEXT
|
ISI
| PUBMED
23. Grana G. Ethnic differences in mammography use among older women: overcoming the barriers. Ann Intern Med. 1998;128:773-775.
FREE FULL TEXT
24. Grady KE, Kemkau JP, McVay JM, Reisine ST. The importance of physician encouragement in breast cancer screening of older women. Prev Med. 1992;21:766-780.
FULL TEXT
|
ISI
| PUBMED
25. Fox SA, Murata PJ, Stein JA. The impact of physician compliance on screening mammography for older women. Arch Intern Med. 1991;151:50-56.
FREE FULL TEXT
26. Blustein J. Medicare coverage, supplemental insurance, and the use of mammography by older women. N Engl J Med. 1995;332:1138-1143.
FREE FULL TEXT
27. Zyzanski SJ, Stange KC, Langa D, Flocke SA. Trade-offs in high-volume primary care practice. J Fam Pract. 1998;46:397-402.
ISI
| PUBMED
28. McVea K, Crabtree BF, Medder JD, et al. An ounce of prevention? evaluation of the "Put Prevention into Practice" program. J Fam Pract. 1996;43:361-369.
ISI
| PUBMED
RELATED ARTICLE
The Archives of Family Medicine Continuing Medical Education Program
Arch Fam Med. 2000;9(9):887-891.
FULL TEXT
THIS ARTICLE HAS BEEN CITED BY OTHER ARTICLES
Beliefs, recommendations and intentions are important explanatory factors of mammography screening behavior among Muslim Arab women in Israel
Soskolne et al.
Health Educ Res 2007;22:665-676.
ABSTRACT
| FULL TEXT
The Influence of Breast Self-Examination on Subsequent Mammography Participation
Jelinski et al.
AJPH 2005;95:506-511.
ABSTRACT
| FULL TEXT
On-Schedule Mammography Rescreening in the National Breast and Cervical Cancer Early Detection Program
Bobo et al.
Cancer Epidemiol. Biomarkers Prev. 2004;13:620-630.
ABSTRACT
| FULL TEXT
Impact of Medicare Coverage on Basic Clinical Services for Previously Uninsured Adults
McWilliams et al.
JAMA 2003;290:757-764.
ABSTRACT
| FULL TEXT
|
