Physical activity improves overall health and helps to prevent and reduce stress and many chronic diseases including cardiovascular diseases, type 2 diabetes, arthritis, and some forms of cancer (eg, breast and colon) (1). Getting people to be physically active is a major public health concern (2). As women age, their levels of physical activity decrease (3,4), and more than 60% of women older than 65 years do not participate in recommended levels of physical activity (5). Although general national health education plans exist for all adults aged 18 to 64 years, no national health education campaign specifically targets women in midlife to engage in physical activity before they enter their senior years.

Levels of participation in physical activity differ by race/ethnicity (6-10), age (3-5,11), and sex (12,13). Minority women participate less often in physical activity than do their white peers (14). Minority women are also at higher risk for several types of chronic diseases associated with being sedentary (eg, heart disease, type 2 diabetes, obesity) (15-17).

Physical activity may also improve health-related quality of life (HRQOL). For example, in one study of people with arthritis, those who were physically active reported fewer unhealthy days than those who were inactive (18). Physical activity also positively influences women with breast cancer after diagnosis, thus improving their HRQOL (19). Older women who were physically active and lived independently also reported better quality of life (5). In the US general adult population, those who were inactive were more likely to report experiencing poorer health than those who participated in the recommended amount of physical activity (20).

However, being physically active may have different effects on HRQOL for women of different racial/ethnic groups. Some women may be physically active because it helps prevent future disease and disability, while others may be active because it improves their immediate HRQOL. The association between physical activity among middle-aged women and self-reported HRQOL among ethnic groups in the United States has not been investigated in detail. We attempted to determine whether this association differs across 4 racial/ethnic groups after adjustment for potential confounders. If women in different racial/ethnic groups experience different degrees of HRQOL from the same amount of physical activity, this may explain why women in different racial/ethnic groups vary in their participation in leisure-time physical activity. This finding may also support new strategies for health promotion and intervention targeted to different racial/ethnic groups. The purpose of this research is to determine whether the association between physical activity and HRQOL differs among California's Latina, Asian, African American, and white women aged 40 to 64 years.

We used secondary data from a cross-sectional study, the 2005 California Health Interview Survey (CHIS), a random-digit–dialed telephone survey in 5 languages completed every 2 years to observe the overall health status and other health-related characteristics of California residents. The University of California, Los Angeles, Center for Health Policy Research (UCLA-CHPR) administers CHIS. CHIS is suitable for this study's purpose because it has more than 600 women aged 40 to 64 years in each of the 4 racial/ethnic groups of interest and includes measures of physical activity and HRQOL and information about several sociodemographic and individual characteristics associated with both physical activity and HRQOL. The CHIS sample arises from multistage sampling based on 44 primary geographic sampling units identified in the first stage. Within each of these units, CHIS randomly selects and dials household telephone numbers and, at each number, chooses 1 adult (aged 18 y or older). In the 2005 CHIS, the response rate was 29.5%; 43,020 adult participants completed the survey (21).

This study examined differences in the association between physical activity and HRQOL in women aged 40 to 64 years of different racial/ethnic groups. The findings that white women participated more often, and African American women less often, in regular physical activity agree with those of other studies (14). Because close to two-thirds of the African American women were not currently married, lack of available childcare for women with children, one of the main barriers to being more physically active (10,11,13,27), may account for the smaller percentage of African Americans participating in regular physical activity and the larger percentage of them being sedentary. However, lack of childcare would not explain the lack of physical activity among these older African American women, whose children may no longer live at home or may care for themselves, as well as among Latinas, who were more often currently married (28).

Lower educational levels and annual household incomes, another factor associated with less physical activity (10,29), did not affect physical activity levels among the women in this study because Asians and Latinas had similar levels of regular physical activity (26% vs 23%, respectively) despite having different educational levels (87% completing high school vs 50%, respectively) and annual household incomes (11% less than the federal poverty level vs 28%, respectively).

Asians reported fewer mentally unhealthy days and overall unhealthy days than did the other racial/ethnic groups. African Americans and Latinas reported more overall unhealthy days than did whites or Asians, and African Americans reported more recent activity limitation days than did the other groups. Yet only whites and Latinas who engaged in either regular or some physical activity reported better HRQOL than those who were sedentary. For whites, better HRQOL reflected all measures of unhealthy days, while for Latinas increasing physical activity was associated with better HRQOL only for overall unhealthy days. These findings are robust despite adjustment for self-rated health, a strong correlate for unhealthy days.

These study findings suggest that physical activity does not have the same effect on the HRQOL of middle-aged women in different racial/ethnic groups. Being physically active may not have the same meaning for each racial/ethnic group (8). Some groups may not consciously think of physical activity as part of their daily routine or may not consider it a separate act performed to achieve optimal health (30). Moreover, walking or riding a bike to work and working in factories or in the fields as many immigrants do may be perceived as a necessity, not a way to improve their mental and physical health. Even if such unintended and incidental physical activity improves mental and physical health physiologically, HRQOL, construed as perceived mental and physical health, may not improve. Finally, because women in different racial/ethnic groups may see different health care professionals, perhaps these professionals differ in their promotion of physical activity as important for health and HRQOL.

The study was subject to limitations. The response rate for the 2005 CHIS was only 30%, implying that this study's findings may not represent those of California adult women (21). For example, those who were more physically active may have been outside of their residence, engaging in physical activity, when CHIS attempted to call their home. Those with the poorest HRQOL may also have been less likely to participate in the survey. In addition, CHIS selects only landline telephone numbers for interview so that women who use only mobile telephones and who may have different physical activity habits would not have been sampled (21). Because more than two-thirds of the Latinas and Asian women in the survey were foreign-born, and because these women came from several countries with different languages, cultures, diet, and religions, the findings summarized for Latina and Asian groups may not reflect these country-specific cultural differences in how physical activity and HRQOL are perceived. Because CHIS is cross-sectional, determining whether physical activity affected HRQOL or HRQOL affected physical activity was not possible.

Despite these limitations, if physical activity levels truly affect HRQOL differently for women among different racial/ethnic groups, then programs that promote physical activity to women in different racial/ethnic groups may have to emphasize benefits to both health and HRQOL for some groups and benefits only to health for other groups. Even without considering HRQOL, health care professionals should discuss the concept of being physically active with all women. Further research is necessary to confirm this study's findings and to determine whether the concept of HRQOL is meaningful and relevant for all racial/ethnic groups. Finally, surveys may need to consider an expanded definition of physical activity beyond leisure-time activity to accommodate the physical activity involved in working.