Understanding adult vaccination in urban, lower-socioeconomic settings: influence of physician and prevention systems.

PURPOSE: Vaccination rates for pneumococcal polysaccharide vaccine (PPV) and influenza vaccine are relatively low in disadvantaged urban populations. This study was designed to assess which physician and practice characteristics might explain differences in rates across physicians. METHODS: PPV and influenza vaccination rates were determined for 2,021 patients aged 65 years and older receiving care from 30 physicians in 17 practices surveyed about their office systems for providing adult immunizations. Hierarchical linear modeling (HLM) analyses were used to examine the relationships among vaccination rates, patient-level characteristics, and physician variables. RESULTS: Overall, the weighted PPV vaccination rate was 60.0% and varied widely across physicians (range, 11%-98%). At the patient level in HLM, patient race (P=.01) and age (P = .02), but not neighborhood income, were associated with PPV status. By linking physician survey data with PPV rates, we found the best pair of physician variables to be "reported time spent with patients for a well visit" (P = .01) and "use of enhanced immunization documentation" (P=.10). The overall influenza vaccination rate was 51.9% (range, 22%-96%). Patient race (P=.003) and age (P = .002) were associated with influenza vaccination. The pair of physician variables with the strongest association with influenza vaccination was "use of standing orders" (P <.001) and "average observed physician examination room time," regardless of visit type (P=.02). CONCLUSIONS: Vaccination rates vary widely in urban settings and are associated with practice characteristics such as time spent with patients and, for influenza vaccine, use of standing orders.

Office manager and nurse perspectives on facilitators of adult immunization.

OBJECTIVE: To assess which characteristics of primary care practices serving low- to middle-income white and minority patients relate to pneumococcal polysaccharide vaccine (PPV) and influenza vaccination rates. METHODS: In an intentional sample of 18 primary care practices, PPV and influenza vaccination rates were determined for a sample of 2289 patients >or=65 years old using medical record review. Office managers and lead nurses were surveyed about their office systems for providing adult immunizations, beliefs about PPV and influenza vaccines, and their own vaccination status. Hierarchical linear modeling (HLM) analyses were used to account for the clustered nature of the data. RESULTS: Sampled patients were most frequently female (61%) and white (83%), and averaged 76 years of age. Weighted vaccination rates were 61.1% for PPV and 52.5% for influenza; rates varied by practice. Using HLM, with patient age and race entered as level 1 variables and office factors entered as level 2 variables, time allotted for an annual well visit was associated with a higher likelihood of influenza vaccination (odds ratio [OR] = 1.04; 95% confidence interval [CI] = 1.02, 1.07; P = .003). Nurse influenza vaccination status was associated with a higher likelihood of PPV vaccination (OR = 3.81; 95% CI = 1.49, 9.78; P = .009). CONCLUSIONS: In addition to race and age, visit length and the nurses' vaccination status were associated with adult vaccination rates. Quality improvement initiatives for adult vaccination might include strengthening social influence of providers and/or ensuring that adequate time is scheduled for preventive care.

Using quantitative and qualitative approaches to understand racial disparities in adult vaccination.

BACKGROUND: One proposed explanation for the persistence of racial disparities in adult immunizations is that minority patients receive primary care at practices that differ substantively from practices where white patients receive care. This study used both quantitative and qualitative methods to assess physician and practice factors contributing to disparities in a sample of inner-city, urban, and suburban practices in low to moderate income neighborhoods. METHODS: Pneumococcal polysaccharide vaccine (PPV) and influenza vaccination rates were determined from medical record review in a sample of 2021 elderly (aged > or = 65 years) patients. Their physicians were surveyed about office systems for adult immunizations and structured observations of practice physical features, and operations were conducted. Case studies of practices with lowest and highest rates and the largest racial disparities are presented. RESULTS: Overall, weighted PPV vaccination rate was 60%, but rates differed significantly by race (65.8% for whites vs 36.5% for minorities, P < .001 by stratified Cochran-Mantel-Haenszel test). Two of 6 minority panels had PPV rates less than 20%. Overall, weighted influenza vaccination rate, as measured by receipt of the vaccine in 3 of the 5 most recent seasons, was 51.9%, but rates also differed significantly by race (55.6% for whites vs 36.2% for minorities, P < .03, by stratified Cochran-Mantel-Haenszel test). CONCLUSIONS: Low rates in 2 minority panels, racial disparity between minorities and whites in mixed panels, and between-panel variation in rates contributed to the overall differences in vaccination rates by race.

Raising adult vaccination rates over 4 years among racially diverse patients at inner-city health centers.

OBJECTIVES: To increase adult immunizations at inner-city health centers serving primarily minority patients. DESIGN: A before-after trial with a concurrent control. SETTING: Five inner-city health centers. PARTICIPANTS: All adult patients at the health centers eligible for influenza and pneumococcal vaccines. INTERVENTION: Four intervention sites chose from a menu of culturally appropriate interventions based on the unique features of their respective health centers. MEASUREMENTS: Immunization and demographic data from medical records of a random sample of 568 patients aged 50 and older who had been patients at their health centers since 2000. RESULTS: The preintervention influenza vaccination rate of 27.1% increased to 48.9% (P<.001) in intervention sites in Year 4, whereas the concurrent control rate remained low (19.7%). The pneumococcal polysaccharide vaccine (PPV) rate in subjects aged 65 and older increased from 48.3% to 81.3% (P<.001) in intervention sites in Year 4. Increase in PPV in the concurrent control was not significant. In logistic regression analysis, the likelihood of influenza vaccination was significantly associated with the intervention (odds ratio (OR)=2.07, 95% confidence interval (CI)=1.77-2.41) and with age of 65 and older (OR=2.0, 95% CI=1.62-2.48) but not with race. Likelihood of receiving the pneumococcal vaccination was also associated with older age and, to a lesser degree, with intervention. CONCLUSION: Culturally appropriate, evidence-based interventions selected by intervention sites resulted in increased adult vaccinations in disadvantaged, racially diverse, inner-city populations over 2 to 4 years.

Assessing disparities in adult vaccination using multimodal approaches in primary care offices: methodology.

Racial disparities in invasive pneumococcal disease and pneumococcal polysaccharide vaccination (PPV) persist despite significant progress. One reason may be that minority patients receive primary care at practices with fewer resources, less efficient office systems, and different priorities. The purposes of this paper are: (1) to describe the recruitment of a diverse array of primary care practices in Pittsburgh, Pennsylvania serving white and minority patient populations, and the multimodal data collection process that included surveys of key office personnel, observations of practice operations and medical record reviews for determining PPV vaccination rates; and (2) to report the results of the sampling strategy. During 2005, 18 practices participated in the study, six with a predominantly minority patient population, nine with a predominantly white patient population, and three with a racial distribution similar to that of this locality. Eight were solo practices and 10 were multiprovider practices; they included federally qualified health centers, privately owned practices and faculty and University of Pittsburgh Medical Center community practices. Providers represented several racial and ethnic groups, as did office staffs. PPV rates determined from 2,314 patients' medical records averaged 60.3 +/- 22.6% and ranged from 11% to 97%. Recruitment of practices with attention to location, patient demographics, and provider types results in a diverse sample of practices and patients. Multimodal data collection from these practices should provide a rich data source for examining the complex interplay of factors affecting immunization disparities among older adults.

Impact of the 2004 influenza vaccine shortage on patients from inner city health centers.

In the fall of 2004, the FDA and British authorities suspended the license of one of only two manufacturers that provided the US supply of inactivated influenza vaccine. With a 50% reduction in supply, a severe vaccine shortage resulted. This situation necessitated the development of priority groups for vaccination including those > or =65 years, when ordinarily, influenza vaccine is recommended for those > or =50 years old. A sample of patients > or =50 years old (n = 336), who had been seen at one of four inner-city health centers, was interviewed in summer 2005 using computer-assisted telephone interviewing. Associations of survey responses were examined for three groups: those vaccinated in the 2003-2004 and 2004-2005 influenza seasons (n = 142), those vaccinated in 2003-2004 but not vaccinated in 2004-2005 because of the shortage (n = 63), and those unvaccinated in both seasons (n = 83). Bivariate and multivariate logistic regression analyses were used to determine factors significantly influencing the likelihood of vaccination status. A significantly larger proportion of patients 50-64 years of age were unvaccinated due to the shortage (73%) compared to those who were vaccinated during both seasons (36%, P < 0.001), but there were no racial disparities in vaccination rates. Compared with patients who were vaccinated during both seasons, those who were unvaccinated due to the shortage were more frequently employed, self-reported their health positively, saw their physician less frequently, rated the US government's response to the shortage as "terrible," and blamed the US government for the shortage. Vaccination during the influenza vaccine shortage appears to have followed preferential vaccination of the CDC-established priority group (> or =65 years) and did not result in racial disparities in inner-city health centers.

Predictors of colorectal cancer screening in diverse primary care practices.

BACKGROUND: To explain why rates of colorectal cancer (CRC) screening including fecal occult blood testing (FOBT), flexible sigmoidoscopy (FS), colonoscopy (CS), and barium enema (BE), are low, this study assessed determinants of CRC screening from medical records. METHODS: Data were abstracted from patients aged > or =64 years selected from each clinician from 30 diverse primary care practices (n = 981). Measurements included the rates of annual FOBT, ever receiving FOBT, ever receiving FS/CS/BE under a combination variable, endoscopy/barium enema (EBE). RESULTS: Over five years, 8% had received annual FOBT, 53% had ever received FOBT and 22% had ever received EBE. Annual FOBT was negatively associated with female gender, odds ratio (OR) = .23; 95% confidence interval = .12-.44 and positively associated with routinely receiving influenza vaccine, OR = 2.55 (1.45-4.47); and more office visits: 3 to <5 visits/year, OR = 2.78 (1.41-5.51), and > or =5 visits/year, OR = 3.35 (1.52-7.42). Ever receiving EBE was negatively associated with age > or =75 years, OR = .66 (.46-.95); being widowed, OR = .59 (.38-.92); and positively associated with more office visits: 3 to <5 visits/year, OR = 1.83 (1.18-2.82) and > or =5 visits/year, OR = 2.01 (1.14-3.55). CONCLUSION: Overall CRC screening rates were low, but were related to the number of primary care office visits. FOBT was related to immunization status, suggesting the possible benefit of linking these preventive services.

Predictors of lower endoscopy use among patients at three inner-city neighborhood health centers.

Although colorectal cancer (CRC) is the second leading cause of cancer death in the U.S., screening rates are low. Understanding the predictors of CRC screening is needed. In 2003, a random sample of patients aged 50 and over from three inner-city health centers was surveyed by computer-assisted telephone interview concerning CRC screening. The questionnaire was based on the Transtheoretical Model and the Theory of Reasoned Action. Factor analysis with Varimax rotation and logistic regression analyses were conducted. Of 319 surveys with data about endoscopy, 148 (46%) met guidelines (19 reported sigmoidoscopy within 5 years, 105 reported colonoscopy within 10 years, and 24 reported both within 5 years). Factor analysis identified three factors associated with increased likelihood of lower endoscopy within guidelines: Social Influence for CRC Screening (Eigenvalue 1.73), Barriers to Lower Endoscopy (Eigenvalue 2.00), and Lower Endoscopy Benefit/Ease (Eigenvalue 1.19). Variables in logistic regression associated with a lower rate of endoscopy include being African American (Odds Ratio (OR) = 0.35, 95% confidence interval = 0.13-0.96), being a current smoker (OR = 0.13, CI = 0.03-0.60), and having a higher score on the Barriers to Lower Endoscopy factor (i.e., viewed the inconvenience and unpleasant aspects as more troubling, OR = 0.33, CI = 0.18-0.60). The perceived inconvenience and unpleasant aspects of lower endoscopy are substantial barriers to screening. Advances in colon preparation procedures and better educational campaigns might lessen this perceived barrier and may be particularly important in disadvantaged and African American communities.

Racial differences in beliefs about genetic screening among patients at inner-city neighborhood health centers.

BACKGROUND: Genetic testing has the potential to identify persons at high risk for disease. Given the history of racial disparities in screening, early detection and accessing treatment, understanding racial differences in beliefs about genetics is essential to preventing disparities in some conditions. METHODS: In 2004, a sample of older adult patients from four inner-city health centers was surveyed to assess beliefs about genetic determinants of disease, genetic testing and religion. Logistic regression determined which beliefs were associated with race. RESULTS: Of the 314 respondents, 50% were African Americans. Most respondents thought that sickle cell disease, cystic fibrosis and diabetes are primarily genetic. African Americans were more likely than Caucasians to believe that genetic testing will lead to racial discrimination (Odds ratio (OR): 3.02, 95% confidence interval (CI): 1.5-6.0) and to think that all pregnant women should have genetic tests (OR=3.8, 95% CI: 1.7-8.6). African Americans were more likely to believe that God's Word is the most important source for moral decisions (OR: 3.6, 95% CI :1.5-8.7). CONCLUSION: African Americans and Caucasians differ in beliefs about genetic testing and the basis for moral decision-making. Acknowledging and understanding these differences may lead to better medical care.

Determinants of adult vaccination at inner-city health centers: a descriptive study.

BACKGROUND: Pneumococcal polysaccharide vaccination rates among adults 65 years and older or less than 65 years with high risk medical conditions are still below Healthy People 2010 recommended levels of 90%. This study was designed to: 1) assess self-reported pneumococcal vaccination rates following health center level interventions to increase adult vaccination rates; and 2) determine factors associated with vaccination. METHODS: Tailored interventions to increase immunizations were implemented at two inner-city health centers. We surveyed 375 patients 50 years of age and older. Multivariate logistic regression examines the predictors of 1) self-reported pneumococcal vaccination and 2) combined self-reported influenza and pneumococcal vaccination. Both of these models were stratified by age group (50-64 years and 65 years and older). RESULTS: Pneumococcal vaccination rates were 45% by self-report, 55% by medical record review, 69% for patients 65 years old and older, 32% for patients 50-64 years; they did not differ by race. Receipt of the previous season's influenza vaccine was significantly related to pneumococcal vaccination among both younger and older patients. Receiving both the pneumococcal vaccine and the most recent influenza vaccine compared with receiving neither, among younger patients was related to unemployment, more frequent physician visits, and belief that those who do not receive the flu shot are more susceptible to the flu. For older patients, receipt of both vaccines was related to nonsmoking status, believing that friends/family think the patient should be vaccinated, seeing posters advertising flu shot clinics, and belief that those who do not receive the flu shot are more susceptible to the flu. CONCLUSION: Our findings suggest that improving overall pneumococcal vaccination rates among eligible adults, has the potential to eliminate racial disparities. Interventions delivering vaccination messages specific to older and younger adult groups may be the best strategy for improving adult vaccination rates.

Impact of vaccine economic programs on physician referral of children to public vaccine clinics: a pre-post comparison.

BACKGROUND: The Vaccines for Children (VFC) Program is a major vaccine entitlement program with limited long-term evaluation. The objectives of this study are to evaluate the effect of VFC on physician reported referral of children to public health clinics and on doses administered in the public sector. METHODS: Minnesota and Pennsylvania primary care physicians (n = 164), completed surveys before (e.g., 1993) and after (2003) VFC, rating their likelihood on a scale of 0 (very unlikely) to 10 (very likely) of referring a child to the health department for immunization. RESULTS: The percentage of respondents likely to refer was 60% for an uninsured child, 14% for a child with Medicaid, and 3% for a child with insurance that pays for immunization. Half (55%) of the physicians who did not participate in VFC were likely to refer a Medicaid-insured child, as compared with 6% of those who participated (P < 0.001). Physician likelihood to refer an uninsured child for vaccination, measured on a scale of 0 to 10 where 10 is very likely, decreased by a mean difference of 1.9 (P < 0.001) from pre- to post-VFC. The likelihood to refer a Medicaid-insured child decreased by a mean of 1.2 (P = 0.001). CONCLUSION: Reported out-referral to public clinics decreased over time. In light of increasing immunizations rates, this suggests that more vaccines were being administered in private provider offices.

Predictors of pneumococcal polysaccharide vaccination among patients at three inner-city neighborhood health centers.

BACKGROUND: Despite the availability of pneumococcal polysaccharide vaccine (PPV), vaccination rates are modest, and racial disparity in these rates is known to occur. OBJECTIVE: The purpose of this study was to identify determinants of patient-reported receipt of PPV among patients from 3 inner-city health centers that serve large numbers of minority and disadvantaged patients. METHODS: In 2003, a random sample of patients from 3 inner-city health centers was surveyed by computer-assisted telephone interview concerning their vaccination status and their attitudes and beliefs about PPV. The questionnaire was based on the theory of reasoned action, which emphasizes the contributions to behavioral intention of cognitive beliefs (including the likelihood of consequences), the value of those consequences, and normative beliefs. Factor analysis was conducted to examine whether survey items specific to attitudes about PPV would group together; the variance of the components resulting from the factor analysis was expressed in eigenvalues. Logistic regression analyses were conducted to determine which factors, demographic characteristics, and health habits were independently associated with PPV status. RESULTS: Of 674 patients sent introductory letters (13 of whom were ineligible), 325 (49%) completed the interview with usable data. These patients were primarily aged > or = 65 years (52% [168/325]), female (69% [223/325]), and white (57% [174/306]), and had an annual household income < 20,000 dollars (61% [174/287]). Vaccination status or intention to be vaccinated was unknown for 25 persons, reducing the sample to 300. Of these respondents, 59% (178) reported vaccination, 9% (26) intended to be vaccinated within 6 months, and 32% (96) had no intention of being vaccinated. Vaccination rates were 76% (119/157) for responders aged > or = 65 years and 41% (59/143) for responders aged 50 to 64 years (P < 0.001). Logistic regression analyses found 3 variables that were significantly associated with the likelihood of vaccination: age > or = 65 years; frequency of physician visits; and timing of the last complete physical examination. The factor analysis with varimax rotation identified 2 factors that increased the likelihood of reporting ever receiving PPV: attitudes about PPV and social influences (eigenvalue = 2.69), and beliefs about pneumonia risk and vaccine efficacy (eigenvalue = 1.24). When these factors were added to the model, the effects of age and frequency of seeing a physician became nonsignificant. Compared with those having a complete physical examination < 1 year ago, those having an examination 1 to 2 years ago (odds ratio [OR] = 0.16; 95% CI = 0.04-0.69) or > 2 years ago (OR = 0.10; 95% CI = 0.01-0.78) were less likely to be vaccinated. Those agreeing with the attitudes and social influence factor were more likely to be vaccinated (OR = 8.24; 95% CI = 3.97-17.12), as were those agreeing with the pneumonia risk and vaccine efficacy factor (OR = 2.63; 95% CI = 1.20-5.80). CONCLUSIONS: Attitudes about vaccination and a clinician's recommendation are the most powerful predictors of PPV status. Efforts to raise rates of vaccination and reduce racial disparities should emphasize communication of clinicians' recommendations to patients.

What predicts influenza vaccination status in older Americans over several years?

OBJECTIVES: To examine the correlates of repeat influenza vaccination and determine whether there are age-group (50-64, > or =65) differences in decision-making behavior. DESIGN: Longitudinal survey study. SETTING: Two community health centers in Pittsburgh, Pennsylvania. PARTICIPANTS: Two hundred fifty-three patients aged 50 and older in 2001 who visited one of the health centers and completed telephone surveys in 2002 and 2003 after the respective influenza seasons. MEASUREMENTS: Influenza vaccination status, demographic characteristics, and decision-making behavior were self-reported. Vaccination status was identified for three seasons: 2000-2001, 2001-2002, and 2002-2003. A three-level outcome was defined as unvaccinated all 3 years, vaccinated one to two times over 3 years, and vaccinated all 3 years. Factor analysis identified three decision-making behaviors. RESULTS: Predictors of being vaccinated across 3 years included being older, the belief that social forces influence vaccination behavior, and disagreement with the view that vaccine is detrimental. CONCLUSION: National educational efforts should be intensified to dispel the myths about alleged adverse events, including contracting influenza from inactivated influenza vaccine. Physicians should continue to share their personal experiences of treating patients with influenza, including the incidence of hospitalization and death.

The 2002 United States varicella vaccine shortage and physician recommendations for vaccination.

BACKGROUND: The US experienced a shortage of varicella vaccine in 2002, leading to the concerns about its impact. METHODS: 204 Minnesota and Pennsylvania physicians, most (164) of whom were interviewed in 1999 on the topic of varicella vaccine, responded to a 2003 survey. RESULTS: Although 67% were aware of the 2002 varicella vaccine shortage, 24% experienced it and only 45% were aware of the 2002 temporary change in national vaccination recommendations. In response, more vaccinated until the supply was exhausted (59%) than postponed vaccination as recommended (41%). Most (91%) reported that the shortage did not change their likelihood of recommending vaccine. From 1999 to 2003, the percentage of physicians highly likely to recommend vaccination increased from 73% to 82% for children 12-18 months old (P < 0.01). In 2003, more physicians believed that it was likely for secondary skin infections to occur following varicella disease and for parents to request vaccination than in 1999 (P < 0.01). Almost all (93%) physicians in both years believed that serious side effects were unlikely. CONCLUSIONS: Over half of physicians were unaware of the change in vaccine recommendations due to the shortage and many did not follow that change, suggesting the need for a different strategy.

Barriers and facilitators of colon cancer screening among patients at faith-based neighborhood health centers.

We determined the barriers to and facilitators of colorectal cancer (CRC) screening among two faith-based, inner city neighborhood health centers in Southwestern Pennsylvania. Data from a random sample of patients 50 years and older (n = 375) were used to estimate logistic regression equations to compare and contrast the predictors of four different CRC screening protocols: (1) fecal occult blood test (FOBT) < or = 2 years ago, (2) colonoscopy < or = 10 years ago, (3) lower endoscopy (colonoscopy or sigmoidoscopy) < or = 10 years ago, and (4) any of these screening measures. Racial differences (between African Americans or Caucasians) in type of colon cancer screening were not found. Controlling for covariates, logistic regression equations showed that a physician's support of colon cancer screening was positively associated with the receipt of colonoscopy (OR: 19.47, 95% CI: 5.45-69.54), lower endoscopy (OR: 10.96, 95% CI: 3.77-31.88) and any colon cancer screening (OR: 10.12, 95% CI: 3.36-30.46). Patients who see their physicians more frequently were also more likely to be screened for CRC. Unlike other studies, the faith-based environment in which these patients are treated may explain the lack of racial disparity specific to our measures of CRC screening.

Tailored interventions to increase influenza vaccination in neighborhood health centers serving the disadvantaged.

OBJECTIVES: We designed and evaluated interventions to increase adult immunizations within inner-city health centers. METHODS: Interventions included reminders, standing orders, and walk-in "flu shot clinics." Patients were surveyed and records evaluated. RESULTS: Records from 1 center showed that immunization rates increased from 24% to 30% (P <.001) for patients aged 50 to 64 years and from 45% to 53% for patients aged 65 years and older (P <.001). Self-reported vaccination rates did not increase. In logistic regression analyses, the strongest predictor of vaccination among patients aged 50 to 64 years was the belief that unvaccinated persons will contract influenza (odds ratio [OR] = 5.4; 95% confidence interval [CI] = 2.4, 12.0). Among patients aged 65 years and older, the strongest predictor of vaccination was the belief that friends/relatives thought that they should be vaccinated (OR = 9.7; 95% CI = 4.2, 22.3). CONCLUSIONS: Tailored interventions can improve immunization rates at inner-city health centers.

Correspondence among patients' self-reports, chart records, and audio/videotapes of medical visits.

This pilot study examined the covariation of patients' self-reports of instrumental and affective aspects of communication during physician-patient visits with 2 other sources of data: medical chart records and audio/videotapes. Participants were 17 community-based (nonuniversity) primary-care physicians and 77 of their patients, ages 50 to 80. Patients were interviewed by telephone within 1 week after their medical visits. Thirty-five of these visits were audio- and videotaped. Patients were asked to report on their receipt of specific cancer screening in the previous 2 years, the occurrence of instrumental communication events during the visit (e.g., recommendations), their affect, and their visit experiences and communication with their physicians. Results showed (a) noteworthy disagreements between patients' self-reports and medical charts regarding cancer screening; (b) better agreement of patients' self-reports with videotape records than with chart records regarding physicians' recommendations; (c) accurate recognition of patients' self-reported affect, communication, and visit experiences by third-party raters of both audiotapes and videotapes; and (d) similar correlations of audio- and videotape ratings with patients' self-reports as well as substantial correlations between audio and video ratings. The implications of these findings are discussed, and recommendations are made for future research.

The relationship between cognitive and physical performance: MacArthur Studies of Successful Aging.

BACKGROUND: The relationship between change in cognitive and physical performance has yet to be fully understood. Because aging decreases the ability to learn new information while preserving more established knowledge, this article examines whether the association between change in cognitive and physical performance depends on the nature of the physical task. METHODS: Data from the MacArthur Research Network on Successful Aging Community Study--a longitudinal three-site, cohort study of high-functioning, disability-free Americans aged 70 to 79 in 1988 (reinterviewed in 1991 and 1995)--are used for this investigation. We examine the association between change in cognitive performance and two categories of physical performance: novel/attentional demanding physical tasks (e.g., standing on a single leg) or routine physical tasks (e.g., walking at a normal pace). Change in physical performance (over 7 years) is regressed on change in cognitive performance (over the same period) controlling for baseline cognitive ability, demographic factors, health status, and behavioral characteristics. RESULTS: The findings suggest that declines in cognitive performance are associated with declines in both novel/attentional demanding and routine physical tasks. In addition to decline in cognition, gender, prevalent health conditions (e.g., cancer, high blood pressure, and the fracture of a hip), and smoking behavior are associated with decline in performance on some physical tasks. CONCLUSIONS: The findings suggest that the successful execution of physical tasks demands cognitive processes regardless of the nature of the task. Researchers using performance-based measures of physical functioning should pay particular attention to the cognitive capacities of their subjects and how these might influence their assessment.