Healthy People 2010 is a national initiative that has set specific health and health care objectives (1). These performance targets include the delivery of adult clinical preventive services such as cancer screening and vaccinations. Studies of the effectiveness of these services have been reviewed by the U.S. Preventive Services Task Force (USPSTF) (2), and recommendations for providing the services have been developed by the Guide to Community Preventive Services (3). Both efforts emphasize the importance of providing routine screening for breast cancer, cervical cancer, and colorectal cancer as well as vaccinations against influenza and pneumococcal disease.  

The delivery rates of cancer screenings and adult vaccinations have typically been measured separately through population-based surveys, including the Behavioral Risk Factor Surveillance System (BRFSS) and the National Health Interview Survey (NHIS). Although this approach is well suited for assessing progress toward Healthy People 2010 objectives, it does not measure the extent to which individuals receive the full complement of recommended clinical preventive services.

There is currently no index that combines data on the use of all recommended clinical preventive services among adults. In this article, we present a composite measure of cancer screening services and vaccinations obtained by adults aged 50 years and older as a first step toward assessing overall clinical preventive service delivery. We chose to begin our assessment by combining these two sets of interventions because they are universally recommended and data on their use in all 50 states and the District of Columbia are available through the 2002 BRFSS. Composite indices have long been used for the surveillance of childhood vaccinations (4).

Our approach aims to provide a more meaningful and practical measure of the state of clinical preventive service delivery. Such a comprehensive measure could also enhance the ability of health departments and community groups to assess disparities in delivering preventive services, to better gauge progress toward measurable objectives, and to identify best practices for achieving prevention goals.

The median state response rate for the 2002 BRFSS was 58.3% (range 42.2%–82.6%) (11). Results for the individual preventive health services for each of the age and sex groups are presented in Appendix Tables A–D. All other results are for being up-to-date on cancer screening and vaccinations as defined above.

We combined data for 49 states and the District of Columbia to examine the percentage of adults who were up-to-date among demographic and risk-factor subgroups (Table 2). Black, Hispanic, and Asian older adults were significantly less likely than whites to be up-to-date in at least two of the four age and sex groups. On the other hand, although they did not reach statistical significance, rates for American Indians were consistently close to rates for whites in three of the four groups.

Among both men and women in all age groups, having more education was strongly related to being up-to-date. Respondents with less than a high school education were much less likely to be up-to-date (range 12.4%–29.5%) than those with a college degree (range 25.4%–43.8%). Similarly, higher income was associated with being up-to-date on cancer screening and vaccinations. Adults with health insurance were about twice as likely to almost three times as likely to be up-to-date as adults with no insurance. The same was true for adults with a personal physician compared with those without one. Overall, people who reported fair or poor health were more likely to be up-to-date than those who reported good, very good, or excellent health.

State-specific prevalence estimates for each of the age and sex groups were divided into quartiles and mapped (Figure 1). States in the western and southeastern United States were among those with the lowest percentage of men aged 50 to 64 who were up-to-date on cancer screening and vaccinations (California, Idaho, Nevada, Utah, and Wyoming in the West; Arkansas, Florida, Georgia, Louisiana, and Mississippi in the Southeast). There was a similar but less pronounced pattern for women aged 50 to 64 (Idaho, Nevada, and Wyoming in the West; Arkansas, Florida, Georgia, Louisiana, and Mississippi in the Southeast). The percentage of men and women aged 65 and older who were up-to-date was low in Indiana, Louisiana, Mississippi, Nebraska, and Texas. Arizona, California, Minnesota, North Dakota, Connecticut, Rhode Island, Wisconsin, Massachusetts, and Maine had higher percentages of people aged 65 and older who were up-to-date than most of the other states. Minnesota had the highest percentage of up-to-date adults in each of the four age and sex groups.

The percentage of each age and sex group that was up-to-date is presented by state in Table 3. The state medians for the percentage of respondents who were up-to-date were consistently less than 40% (with a range of 22.1% for women aged 50 to 64 to 38.2% for men aged 65 and older) (Figure 2). Median state values were similar to results for all adults reported in Table 2.

Table 4 shows the percentage of adults who had zero, one, two, three, four, or five cancer screenings or vaccinations. Most adults had one or more preventive services, with the percentage of adults who received none of the services ranging from 3.6% for women aged 65 and older to 38.3% for men aged 50 to 64. Thus, the results indicated that at least 61.7% of older men and more than 95% of older women have had some contact with the health care system within the past few years.

This analysis of state BRFSS data indicates that among the four age and sex groups, the state medians for adults who are up-to-date with recommended cancer screenings and vaccinations range from 22.1% (women aged 50 to 64) to 38.2% (men aged 65 and older). Although the delivery rates of individual clinical preventive services may be relatively high — cervical cancer screening, for example, is at a national median of 81.0% — overall levels of protection provided by cancer screenings and vaccinations are low.

This study did not include data on cholesterol screening or tetanus–diphtheria vaccination, or data on clinical preventive services that are also recommended according to a specific schedule. Other excluded measures included screening for abnormalities of blood pressure, height, weight, obesity, vision, and hearing; chemoprevention for cardiovascular disease (aspirin); and counseling on calcium intake, folic acid, tobacco cessation, drug and alcohol use, sexually transmitted diseases, human immunodeficiency virus (HIV), nutrition, physical activity, sun exposure, oral health, injury prevention, and polypharmacy.

In addition to low absolute levels, the data indicate important differences among states. For each of the four age and sex groups, Minnesota consistently had the highest up-to-date percentages. Several factors may explain why this state has done so well in preventive service delivery: a high proportion of the population is enrolled in managed care; the state has a low uninsurance rate; and state-sponsored outreach promotes colorectal cancer screening and adult vaccinations. Additional research is necessary to fully explore the reasons for success in Minnesota. Many of the lowest measurements were found in the southeastern and south-central United States; these states have low rates of health insurance among adults (12).

Each of the four age and sex groups has its own set of recommended clinical preventive services. Figure 2 illustrates that men and women within each of the two age groups have similar proportions that are up-to-date, despite the fact that more tests are required of women. When compared with women in the same age group, men are much more likely to have had none of the recommended services. The reasons for these differences are not clear from the data and require further analysis.

There were significant differences in being up-to-date among racial and ethnic categories. Based on consistency across age and sex groups, it appears that blacks, Hispanics, and Asians may be less likely than whites to be up-to-date for cancer screenings and adult vaccinations. There may be higher rates of screening among American Indians because many American Indians receive health care through the Indian Health Service, which emphasizes preventive health practices. Further investigation of these racial and ethnic disparities is warranted.

Not only are reasons for racial and ethnic disparities not clear from the data but these disparities also are confounded by disparities in being up-to-date according to educational, socioeconomic, and insurance status. Respondents who have a college degree are 1.5 to 2 times more likely to be up-to-date than respondents who do not have a high school degree. Receipt of clinical preventive services is also consistently associated with higher income level for each of the four age and sex groups. Large differences exist between people who have medical insurance and people who do not. People with health insurance or personal physicians have approximately two to three times the rate of clinical preventive service delivery as people without either of them. According to the U.S. Census, 11.1% of whites, 19.6% of blacks, 18.8% of Asians, and 32.7% of Hispanics lacked insurance in 2003 (13). These findings are consistent with those of other studies that have found significant disparities in the delivery of cancer screening (14-16) and adult vaccinations (17,18) when measured by race/ethnic group, education, and income.

More than 95% of adults aged 65 and older have health insurance through Medicare, which pays for cancer screening and vaccinations (19). Nonetheless, fewer than 40% of this age group are up-to-date on all of the recommended cancer screening and vaccinations. Having health insurance (and a personal physician) — although necessary — is not in itself sufficient for achieving high levels of being up-to-date with cancer screenings and vaccinations.

Our results suggest that most respondents have received at least one cancer screening or vaccination. Therefore, most adults aged 50 and older have had some contact with the health care system within the past few years, but they do not receive the preventive services recommended for their age and sex group. At the same time, the analysis indicates that up-to-date rates are not low because adults lack only a single service.

This work builds on the analysis of others (20). The BRFSS has followed state rates for clinical preventive service delivery for more than 20 years, and a recent study has provided a comprehensive overview of trends (21). Combined measurements of mammography, clinical breast exam, and Pap testing have been made using data from the 1990 NHIS of Health Promotion and Disease Prevention (22). A methodology for building a prevention index has also been developed using electronic medical records. This index can serve as a combined quality assessment measure and be compared with consensus measures or with selected Health Plan Employer Data and Information Set (HEDIS) scores for health management organizations (23).

This study has a number of limitations. First, the BRFSS relies on self-reported data. Depending on the measure, self-reports can result in overestimates or underestimates compared with other data sources, such as health care records. Unfamiliarity with medical terms and telescoping (24) (recalling events more recently than they actually occurred) can affect the validity of self-reported clinical preventive services and probably result in overestimates (20,25). A review article by Nelson et al rates the validity of self-reports for mammography, Pap test, colorectal cancer screening, and pneumococcal vaccination as moderate and the validity for influenza vaccination as high (26). Second, households without telephones, which are more likely to have adults with low socioeconomic status, were excluded from the survey. This omission may have resulted in an overestimation of delivery rates (27). Third, the survey excludes people who rely only on cellular telephones for their telephone service. Finally, the BRFSS questions limited our ability to adequately determine compliance with Pap test recommendations for women aged 65 and older. The USPSTF recommends against routine screening for cervical cancer for women in this age group if they have had sufficient recent Pap tests with no abnormal results and are not otherwise at increased risk for cervical cancer (28). Without Pap test histories, we could not ascertain which of these older women might not need further Pap testing, and thus we were unable to determine how this anomaly in the recommendations might have affected our estimates of being up-to-date on the Pap test among women aged 65 and older. 

The results of this study suggest several potential next steps. From a surveillance perspective, we recommend that state-based surveys such as the BRFSS consider periodically collecting in the same year information about clinical preventive services with specific schedules (i.e., the services analyzed in this study as well as cholesterol screening and tetanus–diphtheria vaccination). In this way, a more complete measure of delivered adult clinical preventive services recommended by age and sex can be calculated. One of the strengths of a composite measure of clinical preventive services is that it emphasizes the overall protection of individuals by combining measurements across disease categories. We also recommend that surveys such as the BRFSS examine the feasibility and implications of routinely reporting a composite measure such as the one suggested by this study.

The building blocks of composite measures will change as recommendations for individual interventions evolve. For example, colorectal cancer screening guidelines for Healthy People 2010 objectives have not yet been updated to match USPSTF recommendations. (Healthy People 2010 objectives for Pap test, mammography, and adult vaccinations are consistent with USPSTF recommendations for women aged 50 and older.)

The results of this study also suggest that the delivery of clinical preventive services in physicians' offices needs continued attention. Among respondents who indicated having a personal physician, the group with the highest proportion of up-to-date was men aged 65 and older at only 41.5%; the group with the lowest proportion was women aged 50 to 64 at 22.8%. Despite important initiatives designed to improve the provision of clinical preventive services in physicians' offices, such as the Agency for Healthcare Research and Quality's Put Prevention Into Practice (PPIP) (29), more work is necessary to enhance chart-flagging systems and improve reminder protocols aimed at patients. Physicians have begun to systematically consider various approaches to addressing multiple behavioral risk factors in primary care (30), and we hope that this commitment can be extended to providing multiple clinical preventive services.

From a public health standpoint, new kinds of programs are needed to address the shortfall identified by this study. Access to each of the recommended clinical preventive services can be facilitated inside and outside of the clinical setting. New initiatives must focus on populations, not patient panels. A community-wide strategy has been developed by the Sickness Prevention Achieved through Regional Collaboration (SPARC) program in Connecticut, Massachusetts, and New York, and has achieved promising results (31,32) by expanding delivery at clinical sites and creating new points of access in nonclinical settings. A similar approach has been taken by New York State's Healthy Women Partnerships, which facilitates the delivery of multiple cancer screenings for underserved populations. Community-wide assurance of clinical preventive services has been a priority for the Group Health Cooperative, a nonprofit health care system based in Seattle, Wash, which covers a large proportion of its region's residents (33).

Additional analytic work is necessary. An examination of data from states that collect information on cholesterol screening and tetanus–diphtheria vaccinations will yield a more complete — and perhaps lower — composite measure of up-to-date use of recommended services. A multivariable regression analysis will provide a fuller understanding of factors that are most strongly and independently associated with failure to be up-to-date, and a longitudinal analysis of trends in the composite measure will yield useful information for targeting state preventive health efforts.

This study reports low composite rates of adult clinical preventive service delivery based on 2002 BRFSS data for adults aged 50 and older and highlights significant disparities among groups. We must redouble our efforts to develop more effective approaches to delivering these basic adult clinical preventive services.