Correlates of Neighborhood Environment With Walking Among Older Asian Americans

Abstract

Objective: There is a limited research and understanding regarding the physical activity (PA) of older Asian Americans. This study examined the associations between neighborhood factors and walking among older Asian Americans. Method: Drawing from the 2003 California Health Interview Survey, our sample included 1,045 older adults aged 55 and above representing five Asian groups: Chinese, Filipino, Japanese, Korean, and Vietnamese. Zero-inflated negative binomial regression models were used to test the association between neighborhood factors and walking. Results: The results showed that different from the less active health profile among Asian Americans when compared with White adults, Asian older adults overall walked considerably more than White seniors. Higher neighborhood cohesion was associated with more walking among some groups but not all. Association between other neighborhood factors and walking varied among the ethnic groups. Discussion: Health promotion policies and programs should be strategically tailored for specific ethnic groups to more effectively promote PA among older Asian Americans.

Keywords

neighborhood social cohesion walking Asian older adults

Regular physical activity (PA) has been recommended as possibly the single most effective means to reduce the burden of morbidity and mortality from many chronic conditions (World Health Organization, 1996). Older adults can gain substantial health benefits from regular PA, including reducing the risk of chronic diseases, managing active problems such as high blood pressure, diabetes, obesity, or high cholesterol, and improving the ability to function independently (Agency for Healthcare Research and Quality, 2002). However, few older adults meet the minimum recommended standards of moderate or vigorous PA: An estimated 28% to 34% of adults aged 65 to 74 are considered as “inactive” (Agency for Healthcare Research and Quality, 2002). Furthermore, research reveals that minority groups are less active compared with their White counterparts (Centers for Disease Control and Prevention, 2014). Asian adults, in particular, have the lowest rates of PA (Kandula & Lauderdale, 2005; Maxwell, Crespi, Alano, Sudan, & Bastani, 2012). However, research on PA among Asian Americans is scarce (Eyler et al., 2002) and more so among Asian older adults—a population that continues to be neglected in the health literature. As a consequence, little is known about the PA of Asian older adults and how their PA profiles may differ from other groups, including younger Asian adults and their counterparts from other racial/ethnic groups, as well as between Asian ethnic subgroups.

Of all types of PA, walking has been recommended as the first step to staying active due to its accessibility (The Surgeon General, 2001). For older adults, walking is the most popular exercise (Hootman, Macera, Ham, Helmick, & Sniezek, 2003; Yusuf et al., 1996) and has been shown to improve physical and mental health as well as quality of life (Fisher, Li, Michael, & Cleveland, 2004). Because of its easily accessibility and acceptability even among those who are typically sedentary (Siegel, Brackbill, & Heath, 1995), walking has become a focus for research and interventions to improve health among older adults.

Recent evidence highlights a critical need to examine the correlates of PA among Asian older adults. For example, qualitative study involving small number of Vietnamese older adults revealed that they faced barriers in engaging in physical activities such as lack of time (Mathews et al., 2010). Asian older adults, especially those living in ethnic minority neighborhoods which are often characterized by decreased safety, social cohesion, and neighborhood-based civic engagement, face even daunting challenges to engage in PA (Osypuk, Roux, Hadley, & Kandula, 2009).

A key environmental factor that may affect PA is social cohesion (McNeill, Kreuter, & Subramanian, 2006). Social cohesion refers to the “extent of connectedness and solidarity among groups in society” (Kawachi & Kennedy, 1997). When social cohesion among neighbors is combined with a willingness to intervene on behalf of the common good of communities, it can lead to collective response to local problems, such as efforts to reduce violence (Sampson, Raudenbush, & Earls, 1997). Fisher et al. (2004) conducted a study among older adults recruited from 56 predominantly White neighborhoods in Portland, Oregon, and found that neighborhood social cohesion was significantly associated with increased levels of PA. In addition, neighborhood safety (Shores, West, Theriault, & Davison, 2009; Tucker-Seeley, Subramanian, Li, & Sorensen, 2009) and accessibility of facilities and activities (Mathews et al., 2010; Shores et al., 2009; Taylor et al., 2012) have also been identified as important predictors of PA among older adults. The study conducted by Nagel, Carlson, Bosworth, and Michael (2008) suggested that built environment was associated with increased level of walking among older adults. Based on these findings, it is important to gain a better understanding of barriers and facilitators of PA among minority older adults to provide us with a contextual angle to identifying new approaches to promoting PA among minority seniors.

The goal of this study was to examine the role of neighborhood factors on the PA of Asian older adults, specifically comparing how these factors may differ by Asian ethnicity. Previous research tend to lump all the Asian subgroups together and treat them as one homogeneous group, ignoring the tremendous diversity within Asian Americans in terms of their culture, language, tradition, health perception, and behaviors (Kandula & Lauderdale, 2005; Yang, Bernstein, & Wu, 2003). As a consequence, the health or well-being of distinct ethnic groups may be masked by the characteristics of the larger, more established groups, which tends to result in a seemingly more positive outlook. In this study, we addressed the above-discussed limitations by analyzing five dominant Asian groups separately and focused on how neighborhood environmental factors imbedded in the community are related to Asian senior’s walking behavior.

Method

Participants

Cross-sectional data from the 2003 California Health Interview Survey (CHIS) was used for this analysis. CHIS is a population-based telephone survey of civilian households, selected through random digit dialing. CHIS was designed to provide population-based estimates for California’s overall population and for its major racial/ethnic groups. To improve its Asian representation, CHIS oversampled for Vietnamese and Korean participants and conducted interviews in several Asian languages, that is, Chinese (both Mandarin and Cantonese dialects), Korean, and Vietnamese. For the CHIS adult sample, the adult interview response rate was 60% (CHIS, 2003), comparable with telephone surveys carried out by the National Center for Health Statistics. The final CHIS 2003 estimates were consistent with the 2003 California Department of Finance Population Projections (CHIS, 2003). The sample for this analysis was restricted to older Asian adults aged 55 years and above (n = 1,045) and was further differentiated into three age groups (55-65, 65-75, and 75 and above) for comparison. The Institutional Review Boards (IRB) at San Diego State University has approved this study.

Measures

Walking

Respondents were asked about the frequency and total duration of walking for transport and leisure in the past week. The total minutes each respondent walked in the past week were calculated by multiplying the number of times he or she walked and average minutes of each walk.

Neighborhood variables

The three main independent variables were social cohesion, the availability of recreational facility, and perceived neighborhood safety. These three neighborhood variables reflect respondents’ perceptions of their neighborhood social environment. A scale of social cohesion was constructed from five Likert-type items measuring the respondent’s level of agreement (1 = strongly agree; 2 = agree; 3 = disagree; 4 = strongly disagree). The five items are (a) “People in my neighborhood are willing to help each other”; (b) “People in this neighborhood generally do not get along with each other”; (c) “People in this neighborhood can be trusted”; (d) “People in this neighborhood do not share the same values”; and (e) “Most people in this neighborhood know each other.” Principal components analysis was used to compute the factor score. For our sample, the coefficient of alpha was .65, with higher factors scores indicating higher levels of social cohesion in the neighborhood.

Availability of recreational facility (i.e., access to nearby park or playground) was measured by a single yes/no item asking the participant to report whether they had a park, playground, or open space within walking distance of home.

Perceived neighborhood safety was a dichotomous variable created based on the response from two questions: (a) a Likert-type scale (4 = strongly agree; 3 = agree; 2 = disagree; and 1 = strongly disagree) response to “Many people in this neighborhood are afraid to go out at night” and (b) a yes or no response to “Has your home ever been broken into?” Respondents who strongly agreed or agreed that people were afraid or had their homes broken into were coded as living in an “unsafe” neighborhood. Respondents who reported no break-in and disagreed with the first question were coded as living in a “safe” neighborhood. Either they disagreeing or had their home broken into, was regarded as living in an “unsafe” neighborhood.

Sociodemographic variables

Demographic variables included race/ethnicity (Chinese, Filipino, Japanese, Korean, Vietnamese), marital status (married vs. unmarried), gender (female vs. male), age (in years), socioeconomic status, and immigration status. Individual-level socioeconomic status was measured by poverty income ratio (PIR), educational attainment, and employment status. The PIR is a ratio where the numerator is household income and the denominator is the appropriate federal poverty level (FPL) given the family’s size and composition. Poverty thresholds are revised each year by the Census Bureau. Thus, a PIR less than 100% indicates that the household is living below the poverty threshold. The PIR was categorized as less than 100% FPL versus 100% or more of the FPL. Education was categorized as less than high school, high school graduate, and college graduate. Employment status was categorized as employed vs. not employed. All of these were each entered as dummy variables in the regression models. Finally, immigration status was a dichotomous variable indicating whether a participant was born inside or outside the United States (U.S.-born vs. foreign-born).

Because social and environmental factors may affect body mass index (BMI), we also adjusted for BMI (weight/height² [kg/m²]) in the analysis. BMI was categorized as underweight (BMI < 18.5), normal weight (BMI = 18.5-24.99), overweight (BMI = 25-29.99), and obese (BMI ≥ 30). In addition, we also adjusted for whether the participant reported having asthma and heart disease (as told by a doctor), conditions that may pose restriction on the ability to walk for older adults. We also controlled for instrumental activities of daily living (IADLs), which was based on the following question: “For household chores, such as cooking, shopping, managing money, or cleaning, do you need special equipment or someone to help you because of a health problem or condition (yes/no)?”

Statistical Analysis

Approximately 23% of the respondents reported 0 min for walking, which is typical of PA engagement patterns among older adults in the current literature. This zero count on walking could be due to multiple reasons. Those who had 0 min for walking could be divided into two latent groups: one group with a very high propensity to not walk all the time, probably because of their poor health or limited physical function, and the other group also with a high probability of not walking, not because of their health but other non-health-related reasons such as living in an unsafe neighborhood or exercising in other ways. Zero-inflated negative binomial (ZINB) allows for these two different processes and accommodates this complexity by modeling zero counts differently.

The dependent variable was also assumed to have a negative binomial distribution due to observed overdispersion (i.e., the conditional variance was greater than the conditional mean). This further justified the use of a ZINB model, which addresses overdispersion by changing the mean structure to explicitly model the production of zero counts (Long, 1997). The ZINB model assumes two latent groups. One is the always-zero group, and the other is the not-always-zero or sometimes-zero group. Zero counts come from the former group and some of the latter group with a certain probability. The first part of the model, the negative binomial part, refers to the component predicting the minutes of walking among the walking subgroup. The second part of the model, the zero-inflated part, refers to the model component for predicting membership to the subpopulation with a high propensity to zero walking. Replicate weights were applied to account for the data’s complex survey design. We did not use random effects models in this research because all our neighborhood variables were based on individual survey responses and therefore were not group-level variables. We conducted all the statistical analysis using Stata 11.0.

In our analyses, we used White Caucasian as our reference group to put Asian groups in a large context and highlighted their differences from the mainstream. In addition, we also delved into the heterogeneity among the Asian groups. To do so, we used Japanese as the reference group given their highest number of native born and better socioeconomic status.

Results

Sample Characteristics

Sample characteristics by ethnicity are presented in Table 1. White participants are included in this table simply for comparison purposes. There were a total 1,045 older Asian adults aged 55 and above. On average, Asian adults reported walking 153.9 min per week, significantly more compared with their White counterparts (114.4 min/week). Among Asians, the Chinese adults reported walking the most (an average of 190.2 min/week), followed by the Vietnamese, Korean, Filipino and Japanese participants (183.6 min/week, 135.2 min/week, 126.6 min/week, 109.5 min/week, respectively).

Table 1.

Sample Description.

	By race			By Asian ethnicity
	White	Asian	p	Chinese	Filipino	Japanese	Korean	Vietnamese	p
Sample size	11,394	1,045		355	173	164	140	133
Age group			**						***
55-64 years	43%	43%		41%	44%	27%	56%	58%
65-74 years	28%	33%		36%	31%	33%	30%	27%
75 years and older	29%	23%		22%	25%	40%	14%	15%
Female	54%	56%	*	56%	53%	74%	58%	51%	***
Married	61%	72%	***	75%	74%	62%	77%	62%	*
Foreign-born	10%	82%	***	87%	94%	25%	99%	100%	***
Education
Less than high school	11%	22%	***	32%	8%	6%	32%	47%	***
High school graduate	52%	40%		37%	33%	66%	32%	39%
College	37%	39%		32%	59%	28%	36%	14%
Employed	35%	33%		26%	44%	24%	48%	27%	***
Income < 100% FPL	5%	22%	***	37%	12%	3%	19%	39%	***
Health and disability
Asthma	12%	11%		9%	16%	11%	8%	9%
Heart disease	19%	15%	**	15%	20%	11%	10%	14%
Needs help with IADLs	10%	11%		12%	14%	5%	5%	14%
Body mass index (m/kg²)
Underweight (0-18.49)	2%	4%	***	5%	3%	3%	2%	7%	*
Normal (18.5-24.99)	39%	59%		66%	48%	58%	63%	69%
Overweight (25.0-29.99)	39%	29%		22%	40%	33%	31%	16%
Obese (30.0 or higher)	20%	8%		7%	9%	7%	4%	8%
Neighborhood variables
Social cohesion (M)	15.1	14.1	***	13.8	14.2	14.8	13.8	13.6	***
Nearby park/playground	74%	69%	*	66%	74%	66%	79%	55%	*
Neighborhood safety	72%	58%	***	52%	61%	50%	62%	62%	*
Minutes walked/week (M)	114.4	153.9	***	190.2	126.6	109.5	135.2	183.6	***

Note. FPL = federal poverty level; IADLs = instrumental activities of daily living.

p < .05. **p < .01. ***p < .001.

Thirty-seven percent of older Asian adults were overweight or obese, which is significantly lower than their White counterparts (59%). Among older Asian adults, Filipinos (49%) and Japanese (40%) have the highest proportion of overweight and obese adults, significantly exceeding the total Asian average of 37%. In contrast, the Vietnamese and Chinese were much less likely to be overweight or obese. In fact, they were more likely to be underweight. Compared with Whites, older Asian adults have high blood pressure and diabetes but lower proportion of heart disease. Among Asian subgroups, older Filipino adults reported the highest rate of high blood pressure (64%) and diabetes (23%). Older Korean adults fared the best in almost all categories of health indicators with low rate of overweight or obese, diabetes, high blood pressure, heart disease, asthma, and limited physical functions.

Sociodemographic characteristics also varied substantially between the racial/ethnic groups. Compared with their White counterparts, older Asians adults were more likely to be younger, female, and married. They were also more likely to have less than a high school education and to fall below the federal poverty line. Within the Asian subgroups, older Chinese and Vietnamese adults were the least educated and the most financially impoverished. The majority of the older Asian adults (more than 80%) were foreign-born, with the exception of the older Japanese adults (25%). These older Japanese adults were also better educated and have the lowest rate of poverty.

The findings suggest that older White and older Asian adults perceived their neighborhoods differently. In general, older Asian adults reported living in neighborhoods with less social cohesion, poorer access to nearby park/playground, and worse neighborhood safety. There was a significant difference in the perceived level of social cohesion across the Asian subgroups. Among the five groups, the older Vietnamese participants reported the lowest level of social cohesion while the Japanese participants reported the highest levels. Perceived access to park/playground was lowest among the older Vietnamese group and perceived neighborhood safety was lowest among the Japanese and then the Chinese groups.

Neighborhood Characteristics and Walking

Table 2 presents the ZINB regression models predicting minutes walked per week and the likelihood of being a non-walker for the full Asian sample. In general, after all the health and sociodemographic variables were controlled for, the results showed no significant differences across the Asian subgroups in the number of minutes walked. However, when compared with the Japanese group, the older Chinese adults were less likely to be non-walkers (or in other words, more likely to be walkers).

Table 2.

Predicting Minutes Walked per Week and Non-Walkers Among Older Asian Americans (Zero-Inflated Negative Binomial Models).

	Model 1		Model 2
Predicting minutes walked/week	IRR	95% CI	IRR	95% CI
Ethnicity (ref = Japanese)
Chinese	1.17	[0.88, 1.55]	1.16	[0.87, 1.54]
Filipino	0.88	[0.63, 1.22]	0.86	[0.61, 1.20]
Korean	0.93	[0.65, 1.33]	0.98	[0.67, 1.44]
Vietnamese	1.12	[0.77, 1.62]	1.12	[0.77, 1.62]
Other Asian	1.13	[0.72, 1.80]	1.06	[0.68, 1.64]
Age group (ref = 55 to 64 years)
65 to 74 years	0.95	[0.77, 1.16]	0.95	[0.78, 1.16]
75 and older	1.00	[0.80, 1.24]	0.98	[0.78, 1.23]
Social cohesion (continuous)	—	—	1.14*	[1.02, 1.26]
Nearby park/playground	—	—	1.04	[0.85, 1.28]
Safe neighborhood	—	—	1.07	[0.90, 1.26]
Predicting non-walkers	OR	95% CI	OR	95% CI
Ethnicity (ref = Japanese)
Chinese	0.39*	[0.17, 0.89]	0.38*	[0.17, 0.84]
Filipino	0.57	[0.25, 1.29]	0.55	[0.25, 1.24]
Korean	0.63	[0.27, 1.48]	0.60	[0.26, 1.41]
Vietnamese	0.51	[0.20, 1.32]	0.52	[0.20, 1.34]
Other Asian	0.46	[0.17, 1.24]	0.46	[0.17, 1.21]
Age group (ref = 55 to 64 years)
65 to 74 years	0.51**	[0.31, 0.84]	0.52**	[0.31, 0.85]
75 and older	0.81	[0.44, 1.50]	0.82	[0.45, 1.49]
Social cohesion (continuous)	—	—	0.97	[0.77, 1.23]
Nearby park/playground	—	—	1.25	[0.84, 1.86]
Safe neighborhood	—	—	0.81	[0.53, 1.22]
Overdispersion (ln alpha)	−0.25***		−0.27***
Adjusted Wald Test: F(df1, df2)	1.37 (21, 59)		1.35 (24, 56)

Note. Both models controlled for the following variables: gender, immigration status, marital status, poverty level, educational attainment, employment status, health conditions (asthma and heart diseases), instrumental activities of daily living, and body mass index groups. IRR = incidence rate ratios; CI = confidence interval; OR = odds ratios.

p < .05. **p < .01. ***p < .001.

When the neighborhood variables were added in Model 2 (after the sociodemographic and health variables have been controlled), social cohesion emerged as a significant predictor of minutes walked per week. Social cohesion was associated with increased minutes of walking for the older Asian sample (incidence rate ratios [IRR] = 1.14). As for the other two neighborhood factors, neither availability of recreational facility (park or playground) nor living in a safe neighborhood were significant predictors of walking for our sample. We also found that those aged 65 to 74 years old were more likely to walk compared with the younger group (55-64 years old).

Table 3 presents the ZINB regression models for specific Asian subgroups and highlights the interethnic group differences in the significant predictors of walking. Neighborhood factors were associated with walking among the Chinese, Filipinos, Koreans, and Japanese but not so among Vietnamese. Although higher social cohesion was associated with minutes walked across all Asian subgroups, it was only significant for the Chinese (IRR = 1.25). As for the availability of recreational facility, the older Chinese and Korean adults who lived near a park or playground tended to walk more than their Asian counterparts who did not live near a park or playground. Most notably, living near a park/playground nearly doubled the amount of walking among the Korean sample (IRR = 1.99). Surprisingly, we found that Japanese adults who lived near a park or playgroup walked less compared with their counterparts who did not live near a park or playgroup. Finally, for older Filipino adults, living in a safe neighborhood was associated with more minutes walked per week. No significant neighborhood effects were observed on the likelihood of being a non-walker.

Table 3.

Predicting Minutes Walked per Week and Non-Walkers by Asian Ethnicity (Zero-Inflated Negative Binomial Models).

	Chinese		Filipino		Japanese		Korean		Vietnamese
Predicting minutes walked/week	IRR	95% CI	IRR	95% CI	IRR	95% CI	IRR	95% CI	IRR	95% CI
Age group (ref = 55 to 64 years)
65 to 74 years	1.33*	[1.06, 1.67]	0.51**	[0.32, 0.81]	1.96	[0.88, 4.38]	1.02	[0.56, 1.85]	0.80	[0.51, 1.26]
75 and older	1.11	[0.73, 1.69]	0.76	[0.44, 1.31]	0.97	[0.51, 1.85]	1.39	[0.87, 2.240	0.90	[0.39, 2.07]
Social cohesion (continuous)	1.25**	[0.07, 1.45]	1.16	[0.84, 1.61]	1.01	[0.78, 1.30]	0.97	[0.73, 1.30]	1.26	[0.62, 2.59]
Nearby park/playground (vs. no)	1.25*	[1.03, 1.51]	1.17	[0.74, 1.83]	0.54*	[0.31, 0.96]	1.99*	[1.05, 2.77]	1.39	[0.85, 2.28]
Safe neighborhood (vs. no)	1.13	[0.94, 1.36]	1.99**	[1.27, 3.12]	0.91	[0.51, 1.63]	0.77	[0.49, 1.23]	0.78	[0.47, 1.30]
Predicting non-walkers	OR	95% CI	OR	95% CI	OR	95% CI	OR	95% CI	OR	95% CI
Age group (ref = 55 to 64 years)
65 to 74 years	0.50	[0.21, 1.16]	0.45	[0.11, 1.79]	0.26*	[0.07, 0.97]	0.83	[0.15, 4.58]	0.39	[0.06, 2.47]
75 and older	0.75	[0.29, 1.97]	0.85	[0.13, 5.37]	0.67	[0.13, 3.35]	0.20	[0.00, 20.44]	0.56	[0.05, 6.48]
Social cohesion (continuous)	0.86	[0.51, 1.46]	1.00	[0.55, 1.81]	1.18	[0.64, 2.18]	1.55	[0.59, 4.10]	0.46	[0.10, 2.20]
Nearby park/playground (vs. no)	1.15	[0.46, 2.88]	1.56	[0.41, 5.90]	0.96	[0.30, 3.09]	1.65	[0.30, 9.16]	3.00	[0.81, 11.19]
Safe neighborhood (vs. no)	0.95	[0.46, 1.94]	0.60	[0.17, 2.11]	0.37	[0.12, 1.13]	3.59	[0.55, 23.66]	2.52	[0.55, 11.60]
Overdispersion (ln alpha)	−0.60**		−0.53**		−0.47**		−0.41*		−0.60*
Adjusted Wald Test: F(df1, df2)	6.38 (18, 62)***		4.08 (18, 62)***		2.82 (18, 62)**		6.47 (16, 64)***		1.94 (17, 63)*

Note. All models controlled for the following variables: age, gender, immigration status, marital status, poverty level, educational attainment, employment status, health conditions (asthma and heart diseases), instrumental activities of daily living, and body mass index groups. IRR = incidence rate ratios; CI = confidence interval; OR = odds ratios.

p < .05. **p < .01. ***p < .001.

In addition, we also noted that older Chinese and Japanese American (65-74 years old) reported more walking or higher likelihood of walking compared with the younger group (55-64 years old) after all the demographic and health variables were controlled for. On the contrary, Filipino older adults (65-74 years old) reported fewer minutes walked compared with their younger peers (55-64 years old).

Discussion and Conclusion

Most studies on determinants of active living lifestyle have not included Asians, even less so among Asian older adults. Thus, there is limited evidence on how neighborhood factors are linked to PA among older Asian Americans. Using a multi-ethnic, population-based sample from California, this study examined the relationship between neighborhood environmental factors and walking behavior among older Asian adults. Instead of treating older Asian adults as one homogeneous group, we conducted separate analyses for each of the five largest Asian subgroups in California. The results revealed tremendous heterogeneity among the Asian subgroups in terms of socioeconomic and health status, the amount of walking, and the association of neighborhood factors with their walking behavior.

In this study, older Asian adults as a whole walked significantly more compared with their White counterparts. Although there are no comparable studies of older adults, this finding was contradictory to current literature on nonelderly Asian adults and PA. For example, Wen, Kandula, and Lauderdale (2007) found no significant difference between Asian and White adults in terms of walking whereas Kandula and Lauderdale (2005) found that Asians reported lower rates of PA compared with other racial groups. These discrepancies may be due to the fact that the current study is focused on older Asian adults who were more likely to be female, younger, but more impoverished than the older White adults. For many older Asians with limited incomes, walking as a mode of transportation may be a necessity and walking for exercise may be the only economically viable option for leisure-time PA. It is also typical that in Asian cultures recreational walking is regarded as a safe and affordable way to stay healthy among older adults (Cerin, Sit, Barnett, Cheung, & Chan, 2013). In addition, in Asian cultures, extended cross-generation living arrangements are common; for example, many older Asian adults live with their adult children. In these households, older Asians, many of them women, usually take-up the majority of grocery shopping and childcare responsibilities and may have to walk to do so. However, there were some differences within the Asian subgroups. In this sample, older Chinese and Vietnamese adults walked the most whereas the Japanese seniors walked the least. Again, this may be due to economical reasons. Older Japanese adults, who were less likely to be foreign-born and who enjoyed higher socioeconomic status on average, may have more transportation or PA options like driving a personal car or going to a gym to exercise rather than taking a walk around their neighborhood.

In this study, we included those aged 55 to 64 as the younger peer for comparison purposes. We found that Asian older adults (aged 65-74) in general reported more walking compared with their younger peers after the other health and demographic variables were controlled, most notably among Chinese and Japanese older adults. Such increases in the level of PA with age have also been reported in studies conducted among Chinese from Mainland China (Bauman et al., 2011; Muntner et al., 2005) and Taiwan (Ku, Fox, McKenna, & Peng, 2006). In contrast, we also noted that Filipino older adults reported fewer minutes walked compared with their younger counterparts, a pattern more similar to previous research conducted among the U.S. population (Schoenborn & Barnes, 2002). Future research should further explore the reasons behind such discrepant patterns across different ethnic groups.

Neighborhood characteristics presented differential relationships with walking behaviors among older Asian adults. Our results showed that social cohesion was related to increased minutes of walking among Asian older adults as a whole, after individual sociodemographic and health characteristics were controlled. However, the significant association between social cohesion and walking is most notable among the Chinese older population. This finding is consistent with other studies among White older sample (Fisher et al., 2004) but different from findings in another study by Wen et al. (2007) in which they found that social cohesion was not associated with walking among Asians as compared with other racial groups. We speculate that compared with their younger counterparts, seniors tend to live in their neighborhood longer and thus, may have more long-term exposure to any adverse environmental factors (Balfour & Kaplan, 2002; Diez Roux, 2002). In addition, due to limited physical and transportation mobility, Asian older adults have limited opportunities to venture outside of their immediate neighborhoods and thus, are more susceptible to neighborhood environmental influences. It is then possible that the neighborhood effects on Asian elders may be more manifested than on nonelderly adult populations as previously reported (Wen et al., 2007). Older Asian adults who perceived high social cohesion in their neighborhood would more likely be better connected to others in their neighborhood. They may socialize more with each other by walking to each other’s home or their gathering places. The end result would be more walking for leisure.

We also found that access to park/playground was associated with increased amounts of walking among older Chinese and Korean adults, similar to results from previous studies (Mathews et al., 2010; Shores et al., 2009). But such relationship did not emerge for some other groups, and it is even negative for Japanese adults. It could be due to the measure of park accessibility in this study is a crude dichotomous indicator of park presence. Additional features of parks, such as the quality, the types of amenities, and size, may also influence the utilization of parks, and ultimately, PA.

The impact of neighborhood safety was mixed. Living in a safe neighborhood was related to increased minutes of walking only among old Filipino adults and not for other Asian subgroups. This mixed result is consistent with previous findings on the link between perceived neighborhood safety and PA (Pichon et al., 2007; Tucker-Seeley et al., 2009). It is possible neighborhood safety only matters to PA engagement under certain circumstances and different measures of safety and PA may also be the source of such inconsistency. Additional work needs to be done to further investigate this association.

In summary, the findings in this study suggest the following: (a) walking for transport or leisure among older Asian adults is greater than their White counterparts; (b) both neighborhood social (e.g., cohesion) and built (e.g., presence of park) environmental attributes are important for some groups, therefore, it is important to capture the overall picture of a neighborhood instead of focusing only on one single aspect; and (c) neighborhood effects on walking vary greatly across different Asian subgroups. Given that walking is common activity among older Asian adults, practitioners can readily incorporate more walking into the daily lives of older Asian Americans. However, given the mixed findings within the Asian subgroups, practitioners should also take into consideration of the older adult’s cultural background, acculturation experience, and individual sociocultural and health factors. Overall, interventions that address both neighborhood and individual-level factors may be more fruitful in increasing PA when interventions are tailored to the needs of individuals and groups in their neighborhood context.

This study has several limitations. First, cross-sectional data limit our ability to make causal inferences about the neighborhood environment and walking behavior. Even longitudinal observational data would not resolve the direction of causality because of the selection bias. Second, the study is based in California and these results are not necessarily generalizable to other places where Asians are proportionally less in number and geographically widely distributed. Third, our measure of walking was based on subjective recall and was subject to response bias especially among those older adults who may experience memory deterioration. Fourth, indigenous measures of neighborhood in minority neighborhoods may illicit interesting findings and shed most light on the unexpected findings we found in this study.

Besides addressing these limitations, future studies would benefit from teasing apart the purposes for walking. Walking for transport and walking for leisure have different social and economic implications. Knowing the reasons why older Asian adults engage in walking would provide better information for practitioners to incorporate walking into exercise routines. In addition, future studies would greatly benefit from having improved measurement of the surrounding environment, such as population composition of ethnic members in the area and the “walkability” of the neighborhood. Finally, future studies should focus more on cultural factors and influences. A more in-depth understanding of cultural values, beliefs, and purposes for walking would help provide better information to tailor more effective interventions for older Asian adults.

Footnotes

Acknowledgements

We would like to thank Ming Wen, professor of sociology at University of Utah for reviewing the earlier version of this article.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was supported by University Grant Program (2012-2013) at San Diego State University to Yawen Li, PhD.

References

Agency for Healthcare Research and Quality. (2002). Physical activity and older Americans: Benefits and strategies. Retrieved from http://www.innovations.ahrq.gov/content.aspx?id=991

Balfour

J. L.

Kaplan

G. A.

(2002). Neighborhood environment and loss of physical function in older adults: Evidence from the Alameda County Study. American Journal of Epidemiology, 155, 507-515. doi:10.1093/aje/155.6.507

Bauman

Cuevas

Omar

Waqanivalu

Phongsavan

. . . Bhushan

(2011). Cross-national comparisons of socioeconomic differences in the prevalence of leisure-time and occupational physical activity, and active commuting in six Asia-Pacific countries. Journal of Epidemiology & Community Health, 65, 35-43. doi:10.1136/jech.2008.086710

California Health Interview Survey. (2003). The 2003 California Health Interview Survey response rates. Los Angeles: Center for Health Policy Research, University of California, Los Angeles.

Centers for Disease Control and Prevention. (2014). Behavioral risk factor surveillance system prevalence data for physical activity. Retrieved from http://www.cdc.gov/physicalactivity/data/facts.html

Cerin

Sit

C. H.

Barnett

Cheung

M. C.

Chan

W. M.

(2013). Walking for recreation and perceptions of the neighborhood environment in older Chinese urban dwellers. Journal of Urban Health, 90, 56-66.

Diez Roux

A. V.

(2002). Invited commentary: Places, people, and health. American Journal of Epidemiology, 155, 516-519. doi:10.1093/aje/155.6.516

Eyler

A. E.

Wilcox

Matson-Koffman

Evenson

K. R.

Sanderson

Thompson

Rohm-Young

(2002). Correlates of physical activity among women from diverse racial/ethnic groups. Journal of Women’s Health & Gender-Based Medicine, 11(3), 239-253. doi:10.1089/152460902753668448

Fisher

K. J.

Michael

Cleveland

(2004). Neighborhood-level influences on physical activity among older adults: A multilevel analysis. Journal of Aging and Physical Activity, 12, 45-63. doi:10.1136/jech.2004.028399

10.

Hootman

J. M.

Macera

C. A.

Ham

S. A.

Helmick

C. G.

Sniezek

(2003). Physical activity levels among the general US adult population and in adults with and without arthritis. Arthritis Care & Research, 49, 129-135.

11.

Kandula

N. R.

Lauderdale

D. S.

(2005). Leisure time, non-leisure time, and occupational physical activity in Asian Americans. Annals of Epidemiology, 15, 257-265. doi:10.1016/j.annepidem.2004.06.006

12.

Kawachi

Kennedy

B. P.

(1997). Socioeconomic determinants of health: Health and social cohesion: Why care about income inequality? British Medical Journal, 314, 1037-1040.

13.

P. W.

Fox

K. R.

McKenna

Peng

T. L.

(2006). Prevalence of leisure-time physical activity in Taiwanese adults: Results of four national surveys, 2000-2004. Preventive Medicine, 43, 454-457. doi:10.1016/j.ypmed.2006.04.011

14.

Long

J. S.

(1997). Regression models for categorical and limited dependent variables (Vol. 7). Sage.

15.

Mathews

Laditka

Wilcox

Corwin

Liu

. . .Logsdon

(2010). Older adults’ perceived physical activity enablers and barriers: A multicultural perspective. Journal of Aging and Physical Activity, 18, 119-140.

16.

Maxwell

A. E.

Crespi

C. M.

Alano

R. E.

Sudan

Bastani

(2012). Health risk behaviors among five Asian American subgroups in California: Identifying intervention priorities. Journal of Immigrant and Minority Health, 14, 890-894. doi:10.1007/s10903-011-9552-8

17.

McNeill

L. H.

Kreuter

M. W.

Subramanian

S. V.

(2006). Social environment and physical activity: A review of concepts and evidence. Social Science & Medicine, 63, 1011-1022. doi:10.1016/j.socscimed.2006.03.012

18.

Muntner

Wildman

R. P.

Chen

Qan

Whelton

P. K.

(2005). Prevalence of physical activity among Chinese adults: Results from the International Collaborative Study of Cardiovascular Disease in Asia. American Journal of Public Health, 95, 1631-1636. doi:10.2105/AJPH.2004.044743

19.

Nagel

C. L.

Carlson

N. E.

Bosworth

Michael

Y. L.

(2008). The relation between neighborhood built environment and walking activity among older adults. American Journal of Epidemiology, 168(4), 461-468. doi:10.1093/aje/kwn158

20.

Osypuk

T. L.

Roux

A. V.

Hadley

Kandula

N. R.

(2009). Are immigrant enclaves healthy places to live? The multi-ethnic study of atherosclerosis. Social Science & Medicine, 69, 110-120. doi:10.1016/j.socscimed.2009.04.010

21.

Pichon

L. C.

Arredondo

E. M.

Roesch

Sallis

J. F.

Ayala

G. X.

Elder

J. P.

(2007). The relation of acculturation to Latinas’ perceived neighborhood safety and physical activity: A structural equation analysis. Annals of Behavioral Medicine, 34, 295-303. doi:10.1007/BF02874554

22.

Sampson

R. J.

Raudenbush

S. W.

Earls

(1997). Neighborhoods and violent crime: A multilevel study of collective efficacy. Science, 277, 918-924. doi:10.1126/science.277.5328.918

23.

Schoenborn

C. A.

Barnes

P. M.

(2002). Leisure-time physical activity among adults: United States, 1997–98. Hyattsville, MD: National Center for Health Statistics.

24.

Shores

K. A.

West

S. T.

Theriault

D. S.

Davison

E. A.

(2009). Extra-individual correlates of physical activity attainment in rural older adults. The Journal of Rural Health, 25, 211-218. doi:10.1111/j.1748-0361.2009.00220.x

25.

Siegel

P. Z.

Brackbill

R. M.

Heath

G. W.

(1995). The epidemiology of walking for exercise: Implications for promoting activity among sedentary groups. American Journal of Public Health, 85, 706-710.

26.

The Surgeon General. (2001). The Surgeon General’s call to action to prevent and decrease overweight and obesity. Washington, DC: Office of the Surgeon General, US, Office of Disease Prevention and Health Promotion, Centers for Disease Control and Prevention, & National Institutes of Health.

27.

Taylor

V. M.

Liu

Yasui

Talbot

Sos

Coronado

Bastani

(2012). Physical activity among Cambodian Americans: An exploratory study. Journal of Community Health, 37, 1040-1048.

28.

Tucker-Seeley

R. D.

Subramanian

S. V.

Sorensen

(2009). Neighborhood safety, socioeconomic status, and physical activity in older adults. American Journal of Preventive Medicine, 37, 207-213. doi:10.1016/j.ame-pre.2009.06.005

29.

Wen

Kandula

N. R.

Lauderdale

D. S.

(2007). Walking for transportation or leisure: What difference does the neighborhood make? Journal of General Internal Medicine, 22, 1674-1680. doi:10.1007/s11606-007-0400-4

30.

World Health Organization. (1996). The Heidelberg guidelines for promoting physical activity among older persons. Geneva, Switzerland: Author.

31.

Yang

Bernstein

A. H.

(2003). Physical activity and breast cancer risk among Asian-American women in Los Angeles: A case-control study. Cancer, 97, 2565-2575. doi:10.1002/cncr.11364

32.

Yusuf

H. R.

Croft

J. B.

Giles

W. H.

Anda

R. F.

Casper

M. L.

Caspersen

C. J.

Jones

D. A.

(1996). Leisure-time physical activity among older adults: United States, 1990. Archives of Internal Medicine, 156, 1321-1326.