当前位置：文档库 › 主客观测量的老人成功老龄化分布

主客观测量的老人成功老龄化分布

Colocation of older adults with successful aging based on objective and subjective measures

Ellen K.Cromley a,*,Maureen Wilson-Genderson b,1,Zachary Christman c,2,

Rachel A.Pruchno d,3

a Department of Community Medicine and Health Care,University of Connecticut School of Medicine,263Farmington Avenue,MC6325,Farmington, CT06030-6325,USA

b Department of Publi

c Health,Temple University,Health Sciences Campus3r

d Floor,Jones Hall,1316W.Ontario Street,Philadelphia,PA19140,USA

c Department of Geography an

d Environment,Rowan University,201Mullica Hill Road,Glassboro,NJ08028,USA

d New Jersey Institut

e for Successful Aging,Rowan University School o

f Osteopathic Medicine,Suite2300UDP,https://www.wendangku.net/doc/e016409425.html,urel Rd.,Stratford,NJ08084,USA

a r t i c l e i n f o

Article history:

Available online

Keywords:

Aging

Health geography Geographical gerontology Colocation quotient

Spatial analysis

New Jersey a b s t r a c t

Aging of the U.S.and world populations highlights the need to understand how and where people age successfully.Older adults with chronic conditions may rate themselves subjectively as aging successfully despite their objective limitations.A typology of successful aging combining objective and subjective criteria has been tested,but spatial patterns in these dimensions have not been widely studied.Our research explores patterns of successful and unsuccessful aging using the colocation quotient,a measure of spatial association among categories in a population.The colocation quotient assesses the degree to which older adults who age successfully are likely to live near other adults who do not age successfully. Data on5576participants in the ORANJ BOWL SM survey,a statewide survey of older adults in New Jersey, were geocoded to the Census block level.Each participant was scored as aging successfully or not on each of the two dimensions.Global and local patterns of colocation of successful and unsuccessful aging in individuals were calculated based on the objective and subjective measures separately and then compared.The analysis reveals a strong regional pattern.In northern New Jersey and along the southeast coast,successful older adults on both dimensions were more likely to be colocated with subjectively unsuccessful older adults.In southern New Jersey,especially in the southwest,successful older adults on both dimensions were more likely to be colocated with the objectively unsuccessful.Spatial analysis of colocation can inform needs assessment for the growing population of older adults by identifying where older people age successfully and where they are aging unsuccessfully.

Introduction

Older Americans numbered41.4million in2011,more than13% of the U.S.population(Administration on Aging,2012).Over the next30years,the number of people65or older is expected to double,accounting for20%of the total population.Similar de-mographic shifts are occurring in other countries(United Nations, Department of Economic and Social Affairs,Population Division,2013).Although population aging is a global concern,patterns and processes of aging vary from country to country and within countries(Davies&James,2011),highlighting the need to under-stand how and where people age successfully.Spatial analysis of colocation can inform needs assessment for the growing popula-tion of older adults by identifying where older people age suc-cessfully and where they are aging unsuccessfully at the regional and local levels.

The concept of successful aging has a long history.Cicero,in his treatise De Senectute,grappled with the extent to which later life could be a time of vitality and activity.Gerontologists have strug-gled to de?ne,measure,and predict successful aging for more than half a century(Cumming&Henry,1961;Havighurst,1963;Palmore, 1979).

The groundbreaking work of Rowe and Kahn(1987,1998) proposed that people aging successfully would show little or no

*Corresponding author.Tel.:t186********;fax:t186********.

E-mail addresses:cromley@https://www.wendangku.net/doc/e016409425.html,(E.K.Cromley),mwstat1@https://www.wendangku.net/doc/e016409425.html, (M.Wilson-Genderson),christmanz@https://www.wendangku.net/doc/e016409425.html,(Z.Christman),pruchnra@rowan. edu(R.A.Pruchno).

1Tel.:t19732222494;fax:t12157077819.

2Tel.:t185********;fax:t185********.

3Tel.:t185********;fax:t1856566

6874.Contents lists available at ScienceDirect Applied Geography

journal ho mep age:

https://www.wendangku.net/doc/e016409425.html,/locate/apgeog

https://www.wendangku.net/doc/e016409425.html,/10.1016/j.apgeog.2014.10.003

Applied Geography56(2015)13e20

age-related decrements in physiologic function,whereas those aging usually would show disease-associated decrements.Rowe and Kahn's model stimulated further research,even though they neither provided a precise de?nition of successful aging nor pro-posed speci?c ways of measuring it(Kahn,2003).As a result,the conceptual de?nitions and measures of success have varied widely across studies,and it is still unclear who should de?ne successful aging and what constitutes it(Bowling&Dieppe,2005).

A fundamental issue underlying the debate about de?ning successful aging has been whether it can be de?ned by objective criteria or subjective judgment.Perceived measures are important for a number of reasons(Kane,2000;Reviere,Berkowitz,Carter,& Ferguson,1996;Wright,Williams,&Wilkinson,1998).Older adults who perceive themselves to be aging well may not seek services to address age-related decreases in their objective abilities.Among older adults who score well on objective measures of successful aging,low perceived success may be a sign of psychological stress or the onset of undiagnosed conditions.

Rowe and Kahn's(1998)concept highlighted three objective criteria:the ability to maintain low risk of disease and disease-related disability,high levels of mental and physical health,and active engagement with life.Researchers adopting this approach use objective criteria,assessed from records or by self-report,such as diagnosis of chronic conditions such as arthritis,hypertension, and osteoporosis or functional disability such as dif?culty walking a speci?c distance,standing for2h,and stooping(Roos&Havens, 1991;Strawbridge,Cohen,Shema,&Kaplan,1996).

More recently,studies have examined the extent to which older adults perceive their own aging experience as successful(von Faber et al.,2001;Montross et al.,2006;Phelan&Larson,2002; Strawbridge,Wallhagen,&Cohen,2002).Subjective criteria include individuals'own assessments of how well they are aging and how they rate their lives at present(Montross et al.,2006). Tension is growing in the successful aging literature due to the marked discrepancy between the objective de?nitions used by clinicians and researchers and the subjective evaluations made by older people themselves.Not surprisingly,studies that have compared subjective and objective de?nitions of successful aging have found signi?cant differences in the proportion of people meeting criteria for success,with more people categorizing them-selves as successful according to their subjective perceptions than would be classi?ed as such according to the objective de?nitions (von Faber et al.,2001;Montross et al.,2006;Strawbridge et al., 2002).

The critical question at hand is whether people can experience chronic disease and functional disability and still feel that they are aging successfully.In one study,about a third of older adults with chronic conditions rated themselves as aging successfully while a similar proportion of those without chronic conditions believed they were not aging well(Strawbridge&Wallhagen,2003).Con-tending with the evidence that people with chronic disease and functional disability can believe they are aging successfully, Pruchno,Wilson-Genderson,and Cartwright(2010)developed a model of successful aging that includes both objective and sub-jective components.The objective component includes having few chronic diseases,ample functional ability,and little or no pain.The subjective component is an evaluation that individuals make of their own aging experience at one point in time.It includes how well they are aging,how successful their aging experience is,and the extent to which they rate their current life as positive.

In subsequent work,they examined how early in?uences and contemporary characteristics de?ne four groups of people e those successful according to both objective and subjective de?nitions, those successful according to neither de?nition,and those suc-cessful according to one,but not the other de?nition(Pruchno,Wilson-Genderson,Rose,&Cartwright,2010).These analyses found that although characteristics observable early in life pre-dicted group membership,their in?uence was modi?ed by current health behaviors and social support,helping to de?ne successful aging as a multidimensional construct having both objective and subjective dimensions.In an extension of this work,longitudinal follow up of the cohort permitted examining earlier life charac-teristics and later life behaviors as they predict stability and change in successful aging prospectively.Gender,race,education,never marrying and incarceration(earlier life characteristics)as well as midlife status(currently married,working),health behaviors (smoking,drinking,BMI,exercise),and social support distinguished people who continued to age successfully four years later from those who did not(Pruchno&Wilson-Genderson,2014).

Thus far,spatial patterns in its dimensions have not been widely studied.Such patterns are important for applied research in community-speci?c studies of older populations and their resource environments(Hirshorn&Stewart,2003;Liu&Engels,2012),as they provide important information about the needs of older people.Yet,despite the unprecedented demographic shift in the age structure of the population,“the connection between geographical inquiry and gerontology remains tenuous”(Cutchin, 2009,p.440).

This paper seeks to strengthen that connection by exploring spatial patterns of successful and unsuccessful aging using the colocation quotient,a measure of spatial association among cate-gories in a population.Speci?cally,the degree to which older adults who age successfully are more likely to live near other adults who do not is assessed.Spatial associations are summarized for the study region as a whole,using the global location quotient.In addition,spatial heterogeneity in colocation of adults who are successfully or unsuccessfully aging is revealed using local coloca-tion quotients,which describe patterns of colocation of these different aging factors experienced by older individuals in different areas within the study region.

Materials and methods

Study area and data

Data used in the study were drawn from a panel of5688adults comprising the ORANJ BOWL SM(Ongoing Research on Aging in New Jersey e Bettering Opportunities for Wellness in Life)sample,who completed interviews between November2006and April2008.To be included,each participant must have been between50and74 years of age,residing in New Jersey,and able to participate in a1-h telephone interview in English.All participants were recruited by telephone cold calling using list-assisted random-digit-dialing(LA-RDD),with coverage or residential landline telephone numbers for this population estimated as95%.Previous research suggests that coverage loss due to cell phone-only households among this sam-ple population is very small(Blumberg&Luke,2007).Following American Association for Public Opinion Research standards,the response rate of the ORANJ BOWL sample was58.76%,with a Cooperation Rate of72.9%.These rates are consistent with or higher than average response rates in RDD efforts during this period.

In comparison to all persons aged50e74and living in New Jersey,those included in the ORANJ BOWL panel demonstrate similar racial composition,rates of being born in the state,and marital status,but there are slightly higher proportions of females (63.7e53.3%)and of individuals with advanced secondary degrees (18.5e14.8%).The ORANJ BOWL panel under-represents persons of Hispanic descent,due to the requirement of English language pro?ciency(Pruchno,Wilson-Genderson,Rose,et al.,2010).

E.K.Cromley et al./Applied Geography56(2015)13e20 14

For analyzing colocation of successful and unsuccessful aging, three variables were required:successful aging based on objective criteria,successful aging based on subjective criteria,and residen-tial location.Each participant was categorized on both dimensions of successful aging(1?not successful;2?successful)based on latent pro?le analysis of eight indicators for the objective dimen-sion and three indicators for the subjective dimension(Pruchno, Wilson-Genderson,Rose,et al.,2010).

Residential location was generalized to the census block level for the2000Census.The centroid of the census block was used to represent an individual's place of residence.Data for112partici-pants were excluded from the analysis because residential location was missing.Exclusion of these data did not affect the proportions of participants who aged successfully on objective or subjective measures(Table1).The residential distribution of the sample re-?ects the distribution of population across the state(Fig.1). Global and local colocation quotient calculations

Successful aging in this study is described by one of two cate-gorical data variables measured at point locations constituting the population of interest.Although the distribution of all older adults in the study population may be clustered,random,or uniform,the spatial associations of those who age successfully and unsuccess-fully are nested within this overall pattern(Fig.3).Older adults in different categories of aging based on different measurements may be more or less likely to be located near each other.

Several methods for analyzing spatial association have been

developed.Leslie and Kronenfeld(2011)review the join count statistic and the cross-k-function in their work introducing the Colocation Quotient(CLQ).The CLQ provides a measure of whether observations in pairs of categories are spatially associated or not. Unlike other measures of spatial association,the CLQ takes the geometric pattern of the joint categories,such as those shown in Fig.2,as a given.Colocation quotients can be used to analyze a wide range of phenomena in human and physical geography.The mea-sure helped to characterize agglomeration of different types of food establishments in research on food environments(Leslie, Frankenfeld,&Makara,2012).

In the original formulation(Leslie&Kronenfeld,2011),the CLQ is calculated as a ratio of observed versus expected points of one type among the set of?rst-order nearest neighbors of the same or another type in the entire population.CLQs can be derived for every pairwise combination of categories,including a category with itself. The CLQ recognizes that relationships among these combinations of categories may not be symmetric,especially when there are dif-ferences in the sizes of the categories.For example,if20percent of the population was less than20years of age and80percent20or older,young people could be dispersed among the older population such that80percent of young people's?rst order neighbors were older,as expected.From the perspective of individuals in the older category,however,there might be areas with few or no young people at all,resulting in higher colocation of old people than would be expected.

Cromley,Hanink,and Bentley(2014)generalized the formula-tion of Leslie and Kronenfeld's global CLQ by broadening the range of neighborhood de?nitions beyond?rst-order nearest neighbor and introducing a local measure of colocation.“Global”statistics, like the original CLQ,summarize data for entire study populations. Local statistics summarize data for individual observations within a larger study area(Lloyd,2011).In the local CLQ,the measure is calculated for each observation as the ratio of observed to expected points of its type among the set of neighboring points of the same or another type within a particular neighborhood.The neighbor-hood may be de?ned using a?xed?lter based on distance from each observation or a spatially adaptive?lter based on k-order nearest neighbors.

Objective and subjective measures of successful aging do not always agree at the individual level.To analyze differences in spatial patterns of successful and unsuccessful aging measured objectively and subjectively,global and local colocation quotients were?rst calculated for successful and unsuccessful aging based on the objective and subjective variables separately.We used a spatially adaptive?lter based on100nearest neighbors.

The formula for the global colocation quotient was:

GCLQ?PNeb j aT=Peb j aT(1) The numerator PN(b j a)is the probability,given the locations of the set of a-points(individuals who aged successfully),of?nding a b-point(individuals who aged unsuccessfully)with the set of a-point neighborhoods.The denominator P(b j a)is the probability, given any a-point,of?nding a b-point with the entire set of par-ticipants.This corresponds to:

Peb j aT?NB=eNà1T(2) Here,NB is the total number of b-points in the entire set of participants.If the colocation quotient measures spatial association

Table1

Frequency distribution of participants by successful aging.

All Participants Geocoded participants

Number Percent Number Percent

Successful on neither measure4748.34648.3 Successful on objective

measure only

4858.54808.6 Successful on subjective

measure only

57010.055610.0 Successful on both measures415973.1407673.1 Total5688100.05576

100.0Fig.1.Residential distribution of study participants by census block centroid.

E.K.Cromley et al./Applied Geography56(2015)13e2015

of members of a category with individuals of the same category,the denominator is:

P ea j a T?eNA à1T=eN à1T

(3)

The numerator is de ?ned as:

PN eb j a T?

X i 2A X

j 2D ej s i Tewij xij T

i 2A X

j 2D ej s i T

ewij T(4)

A is the set of all a -points.D is the set of all points in the data set.x ij is a binary variate with a value of 1if the j th point is a b -point or 0otherwise.w ij is the geographic weight.For this analysis,it was de ?ned using a Gaussian kernel function:

w ij ?exp h à0:5 d 2ij .d 2

(5)

d ij is th

e distance from the i th observation to the j th observation.d ik

is the distance from the i th observation to its k th nearest neighbor.d ik serves as the spatially adaptive bandwidth for the Gaussian function.At distances from a given observation to a neighboring observation less than the bandwidth,the ratio is less than 1and the weight is high.At distances from a given observation greater than the bandwidth,the weight formula ratio is greater than 1and the value of the weight declines rapidly as distance increases.A SAS program (Version 9.1)using proc iml was written to calculate the global and local CLQs for the objective and subjective variables.The output for the local CLQs was imported to ArcGIS 10.2and joined to database of participant residential locations.Local CLQs were mapped.One map shows local patterns of colocation of un-successful agers with successful agers based on objective measures;the other shows local patterns based on subjective measures.

Because a large number of participants were successful in aging on both measures,their local colocation with unsuccessful

agers

Fig.2.Different patterns of colocation of unsuccessful and successful agers.In each,10adults aging unsuccessfully are collocated with 40adults aging successfully.In Fig.2(a)and (b),adults aging unsuccessfully are more likely to have other adults aging unsuccessfully in their neighborhoods regardless of the distribution of the 50adults.In Fig.2(b),the overall population is clustered compared to its distribution in Fig.2(a).In Fig.2(c)and (d),adults aging unsuccessfully are less likely to have other adults aging unsuccessfully in their neighborhoods,whether population is distributed randomly (2c)or is clustered (2d).

E.K.Cromley et al./Applied Geography 56(2015)13e 20

based on subjective and objective measures can be compared.In the?nal stage of the analysis,a difference map of local CLQs was created to examine the relationship between objective and sub-jective measures of successful aging across the state.

Results

Older community-living adults in New Jersey were generally aging successfully whether measured by objective or subjective criteria.Almost three-quarters of the study subjects(73.1%)were successful on both dimensions while only8.3%were classi?ed as aging successfully on neither dimension(Table1).Nevertheless, 1036of the older adults included in the study(18.6%)scored differently on the two dimensions.

Patterns of global colocation of successful and unsuccessful agers are different for the two dimensions(Table2).For the objective measurement of successful aging,from the perspective of unsuccessful agers,there is evidence that they are somewhat more likely to be colocated with other objectively unsuccessful agers (CLQ?1.06137)than with objectively successful agers.Successful agers on the objective dimension have,looking at the state as a whole,almost exactly the degrees of colocation with other suc-cessful and unsuccessful agers that would be expected given the frequency distribution.The global CLQs are both close to1.00000.

For the subjective measurement of successful aging,the pattern is different.Statewide,unsuccessful and successful agers are both slightly more likely to be colocated with unsuccessful agers.The global CLQs are greater than1.00000.

The local colocation quotients reveal whether the patterns observed at the state level hold for different localities within the state(Fig.3).In Fig.4,the individual residential locations of the participants who were aging successfully are shaded based on the participant's level of colocation with unsuccessful agers.The darkest shaded points represent individuals who are much more likely to have unsuccessful agers in their vicinities than we would expect by chance given the overall split in the state as a whole.The lightest shaded points are successful agers who are much less likely

to have unsuccessful agers in their vicinities.

Based on objective criteria,4556older adults aged successfully (Fig.4(a)).Older individuals living near Newark,Trenton,and the southwestern counties of the state were more likely to have older adults who were not aging successfully based on objective criteria in their midst.In Bergen County in northeastern New Jersey and three counties(Warren,Hunterdon,and Somerset)in the north-western part of the state,older adults who aged successfully on objective measures were much less likely to have neighbors who were aging unsuccessfully based on these measures.This pattern was also apparent among adults in Red Bank borough in Mon-mouth County and near Marlton in Burlington County.

Based on s ubjective criteria,4632older adults aged successfully (Fig.3(b)).Individuals living near Newark were more likely to have older adults who were not aging successfully based on subjective criteria in their local areas.A similar pattern of high local colocation of successful agers with unsuccessful agers is apparent near Toms River in Ocean County.

To compare the two perspectives on successful aging,the dif-ference in colocation for the4076participants who aged success-fully on both dimensions with unsuccessful agers was mapped (Fig.5).We subtracted the local colocation quotient measuring their spatial association with subjectively unsuccessful agers from the local colocation quotient measuring their spatial association with objectively unsuccessful agers to explore spatial differences in the two perspectives.This map shows a strong regional pattern.In northern New Jersey and along the southeast coast,successful agers on both dimensions were more likely to be colocated with unsuc-cessful subjective agers because the difference is negative.In southern New Jersey,especially in the southwest in Gloucester, Salem,and Cumberland counties,successful agers on both di-mensions were more likely to be colocated with unsuccessful objective agers because the difference is positive.

In both main regions,however,there are local pockets where the patterns of colocation are different from the pattern in the surrounding region.Near Bayonne,Paterson,and Orange,in northern New Jersey,successful agers on both dimensions are more likely to be colocated with unsuccessful objective neighbors because the difference is positive.Near Marlton in Burlington County and Atlantic City along the coast in southern New Jersey, successful agers on both dimensions are more likely to be colocated with unsuccessful subjective agers because the difference is negative.

Table2

Global colocation quotients for participants in different categories of aging.

Objective unsuccessful Objective successful

Objective unsuccessful 1.061370.98627

Objective successful0.99422 1.00129

Subjective unsuccessful Subjective successful

Subjective unsuccessful 1.034020.99307

Subjective successful 1.01317

0.99732Fig.3.Counties and selected cities in New Jersey.Each county in New Jersey has a designated Area Agency on Aging,the primary organization responsible for developing community-based services for older adults in the U.S.

E.K.Cromley et al./Applied Geography56(2015)13e2017

Discussion

Consistent with research conducted in other settings,more of the older adults in New Jersey considered themselves aging suc-cessfully than would be categorized as such based on objective measures.Most of the research conducted to date,however,has been aspatial.In this study,the global CLQs reveal whether or not there are particular geographic associations between self-rated successful agers and others in the state as a whole.The global CLQs,all close to the expected level of 1.0,suggest that there is no major overall association in New Jersey between subjectively and objectively successful older adults.

Local spatial statistical methods allow us to uncover patterns in successful aging at levels between those of the individual and the study area as a whole.For this research,a spatially adaptive ?lter was used to de ?ne communities based on the number of nearest neighbors.This approach is widely used in health geographic research because it establishes a ?xed denominator population or spatial basis of evidence for exploring local variability (Beyer &Rushton,2009).The size of the ?lter or number of nearest neigh-bors can be changed to explore spatial patterns of successful aging at different scales.Based on the size selected for this analysis,both community-level and broad regional patterns in successful aging were observed.

Community-dwelling adults who were not aging successfully on objective measures were more likely than expected in counties in northeastern and southwestern New Jersey.Such a pattern sug-gests that some spatial process might be important.In terms of meeting objective criteria,neighborhood conditions have been found to affect outcomes on objective measures of aging including physical function in the U.S.and other countries (Balfour &Kaplan,2002;King et al.,2011;Li,Fisher,Brownson,&Bosworth,2005;Nagel,Carlson,Bosworth,&Michael,2008;Sun,Cenzer,Kao,Ahalt,&Williams,2012

Fig.4.Local colocation coef ?cients.Fig.4(a)shows the locations of all older adults who aged successfully on the objective measure of aging and their degree of colocation with older adults (unmapped)who aged unsuccessfully on the objective measure.Fig.4(b)shows the locations of all adults who aged successfully on the subjective measure of aging and their degree of colocation with older adults (unmapped)who aged unsuccessfully on the subjective

measure.

Fig.5.Difference in local colocation coef ?cients.The locations of all adults who aged successfully on both objective and subjective measures are shown with the difference in coef ?cients measuring their colocation with other adults (unmapped)who aged unsuccessfully on the objective or the subjective measure.The local colocation quo-tient for subjective unsuccessful aging was subtracted from the local colocation quo-tient for objective unsuccessful aging.

E.K.Cromley et al./Applied Geography 56(2015)13e 20

Local patterns were also observed for adults who were not aging successfully on subjective measures.These adults were more likely than expected primarily in counties in northeastern New Jersey. With respect to subjective assessments of aging,local community characteristics also contribute to perceived health and quality of life among the elderly(Greenberg,2009;Luz,C e sar,Lima-Costa,& Proietti,2011).Equally intriguing is the possibility that shared so-cial and behavioral norms affect individuals'views on how suc-cessfully they are aging.In their innovative work on social networks and health,Christakis and Fowler(2013)found that geographic proximity among social contacts did not explain correlation in in-dividual health behaviors,but affective states such as happiness, loneliness,and depression were positively associated with physical proximity.They also investigated whether these results were driven by contextual effects,given that nearby older adults may be exposed to similar neighborhood environments,and found that associations were signi?cant only for next-door neighbors.

These spatial associations between objective and subjective measures are only apparent from the local analysis,not for the state-level analysis in this case.The differences across regions in New Jersey average out for the state as a whole.Local spatial as-sociations will always be more apparent than associations across larger areas,underscoring the need for state-level planners and policymakers to move beyond aggregate assessments that mask local conditions.

Comparing local spatial associations between objective and subjective measures reveals a strong regional pattern.Across six counties in southwestern New Jersey,older adults aging success-fully on both objective and subjective measures are much more likely than expected to have neighbors who are objectively un-successful agers but much less likely to have neighbors who are subjectively unsuccessful.This?nding is evidence of spatial het-erogeneity in the relationship between subjective and objective measures used to assess need.For practitioners conducting local needs assessments,the discrepancy between perceived need and objective need may be much greater in some local communities than in others.

It is unlikely that the patterns of difference are related to regional differences in the composition of the population.Such differences,if found,might be important because of gender and age differences in successful aging.In the six southern counties where the difference was positive and large,the percent of respondents who were women was63.9.In the study population as a whole,it was63.8.The percent of participants in the six counties who were younger(from the Baby Boom generation)was55.3.In the study population as a whole,it was54.5.

The purpose of this research was to apply spatial methods to investigate spatial associations among measures used in needs assessment and clear local patterns of association were revealed. The next step,given these patterns,is to investigate the factors which might explain why older adults in some communities differ on perceived and objective measures of successful aging.The relative lack of older adults perceiving themselves to be aging un-successfully among adults with objective problems of aging in the southwestern counties of New Jersey suggests that a given objec-tive measure of aging means something different in different en-vironments.In places like this region,where there are few large cities and people do not rely on public transit,a functional limita-tion such as the ability to walk a certain distance may be acknowledged but may not be perceived as a problem because the environment does not create a high demand for walking.

A limitation of this study is its cross-sectional design.Future longitudinal work is needed to understand the relationship be-tween successful aging on different measures across geographic areas.For the individuals who are aging successfully on both dimensions,longitudinal analysis would reveal where that success is maintained and where older adults experience declines, including whether the declines on perceived measures precede objective decrements or not.

These?ndings provide novel information to local service pro-viders and policy analysts(Kane,2000;Reviere et al.,1996). Incorporating objective assessments and subjective perceptions of successful aging where people live advances understanding of the needs of older people within a spatial perspective.In New Jersey, county-based Area Agencies on Aging serve as the primary orga-nizations responsible for older adults.Realizing the differential needs of people living in different regions and communities in New Jersey may help service providers match local resources to local needs.

Conclusions

The aging of the population in the U.S.and other countries presents a wide range of challenges to practitioners and re-searchers.Because this research relied on a geographically repre-sentative sample of community-dwelling older adults,it yields a less biased view of patterns of successful aging than research based on service utilization data.The geography of service delivery often acts as?lter,masking our understanding of underlying patterns of population needs.If,for example,services are less available in some areas and utilization is low,the needs of the underserved popula-tion will not be captured in utilization data.

Although many older adults maintain high levels of physical and mental function as they age,others face disease,disability,and dissatisfaction with their lives as older members of society.Mea-sures used to describe health,function,and satisfaction in needs assessments are often assumed to be equally reliable across all communities,and methods for testing this assumption,particularly from a geographic perspective,are not widely used.

Spatial analysis,using methods such as the colocation quotient, can enhance understanding of how different measures of success in aging are related and whether the relationships are the same across all communities.This understanding is of practical importance to gerontologists who need to assess the status of older adults in the communities they serve.It may lead to a better understanding of how neighborhood environments and social connections in?uence successful aging.

Acknowledgments

The authors thank the UMDNJ-SOM New Jersey Institute for Successful Aging whose generous support funded the data collec-tion efforts of the ORANJ BOWL(“Ongoing Research on Aging in New Jersey e Bettering Opportunities for Wellness in Life”) research panel,and the Swedish Council on Working Life and Social Research(FAS dnr2012-1932)for funding a Visiting Professorship in the Department of Occupational and Environmental Medicine at Lund University during which methods used in this research were developed.

References

Administration on Aging.(2012).A pro?le of older Americans:2012.Available www.

https://www.wendangku.net/doc/e016409425.html,/Aging_Statistics/Pro?le/2012/docs/2012pro?le.pdf Accessed06.01.14. 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(6),507e515.

Beyer,K.M.M.,&Rushton,G.(2009).Mapping cancer for community engagement.

Preventing Chronic Disease,6(1),A03.

Blumberg,S.J.,&Luke,J.V.(2007).Coverage bias in traditional telephone surveys of low-income and young adults.Public Opinion Quarterly,71(5),734e749.

E.K.Cromley et al./Applied Geography56(2015)13e2019

Bowling,A.,&Dieppe,P.(2005).What is successful ageing and who should de?ne it?British Medical Journal,331(7531),1548e1551.

Christakis,N.A.,&Fowler,J.H.(2013).Social contagion theory:examining dynamic social networks and human behavior.Statistics in Medicine,32(4),556e577. Cromley,R.G.,Hanink,D.M.,&Bentley,G.C.(2014).Geographically weighted colocation quotients:speci?cation and application.The Professional Geographer, 66(1),138e148.

Cumming,E.,&Henry,W.(1961).Growing old:The process of disengagement.New York:Basic Books.

Cutchin,M.P.(2009).Geographical gerontology:new contributions and spaces for development.The Gerontologist,49(3),440e444.

Davies,A.,&James,A.(2011).Geographies of ageing:Social processes and the spatial unevenness of population ageing.Farnham,Great Britain:Ashgate.

von Faber,M.,Bootsma-van der Wiel,A.,van Exel,E.,Gussekloo,J.,Lagaay,A.M.,van Dongen,E.,et al.(2001).Successful aging in the oldest of old:who can be charac-terized as successfully aged?Archives of Internal Medicine,161(22),2694e2700. Greenberg,M.R.(2009).Neighborhood quality and the older elderly:theory and two pilot tests.Human Ecology Review,16(2),184e194.

Havighurst,R.J.(1963).Successful aging.In R.H.Williams,C.Tibbits,&W.Donohue (Eds.),Process of aging(Vol.1,pp.299e320).New York:Atherton Press. Hirshorn, B. A.,&Stewart,J. E.(2003).Geographic information systems in community-based gerontological research and practice.Journal of Applied Gerontology,22(1),134e151.

Kahn,R.L.(2003).Successful aging:intended and unintended consequences of a concept.In L.W.Poon,S.H.Gueldner,&B.M.Sprouse(Eds.),Successful aging and adaptation with chronic diseases(pp.55e69).New York:Springer.

Kane,R.L.(2000).Mandated assessment.In R.L Kane,&R.A.Kane(Eds.),Assessing older persons:Measures,meaning,and practical applications(pp.458e482).New York:Oxford University Press.

King,A.C.,Sallis,J.F.,Frank,L.D.,Saelens,B.E.,Cain,K.,Conway,T.L.,et al.(2011).

Aging in neighborhoods differing in walkability and income:associations with physical activity and obesity in older adults.Social Science&Medicine,73(10), 1525e1533.

Leslie,T.F.,Frankenfeld,C.L.,&Makara,M.A.(2012).The spatial food environment of the DC metropolitan area:clustering,co-location,and categorical differen-tiation.Applied Geography,35(1e2),300e307.

Leslie,T.F.,&Kronenfeld,B.J.(2011).The colocation quotient:a new measure of spatial association between categorical subsets of points.Geographical Analysis, 43(3),306e326.

Li,F.,Fisher,K.J.,Brownson,R.C.,&Bosworth,M.(2005).Multilevel modelling of built environment characteristics related to neighbourhood walking activity in older adults.Journal of Epidemiology and Community Health,59(7),558e564. Liu,G.-J.,&Engels,B.(2012).Accessibility to essential services and facilities by a spatially dispersed aging population in suburban Melbourne,Australia.In

G.Gartner,&F.Ortag(Eds.),Advances in location-based services(pp.327e348).

Berlin:Springer-Verlag.

Lloyd,C.D.(2011).Local models for spatial analysis(2nd ed.).Boca Raton,FL:CRC Press.

Luz,T.C.B.,C e sar,C.C.,Lima-Costa,M.F.,&Proietti,F.A.(2011).Satisfaction with the neighborhood environment and health in older elderly:cross-sectional

evidence from the Bambuícohort study of aging.Cadernos de Saúde Pública, 27(Suppl.3),S390e S398.

Montross,L.P.,Depp,C.A.,Daly,J.,Reichstadt,J.,Golshan,S.,Moore,D.,et al.(2006).

Correlates of self-rated successful aging among community-dwelling older adults.American Journal of Geriatric Psychiatry,14(1),43e51.

Nagel,C.L.,Carlson,N.E.,Bosworth,M.,&Michael,Y.L.(2008).The relation be-tween neighborhood built environment and walking activity among older adults.American Journal of Epidemiology,168(4),461e468.

Palmore,E.(1979).Predictors of successful aging.The Gerontologist,19(5Part1), 427e431.

Phelan,E.A.,&Larson,E.B.(2002).Successful aging:where next?Journal of the American Geriatrics Society,50(7),1306e1308.

Pruchno,R.A.,&Wilson-Genderson,M.(2014).A longitudinal examination of the effects of early in?uences and midlife characteristics on successful aging.

Journals of Gerontology:Psychological Sciences.https://www.wendangku.net/doc/e016409425.html,/10.1093/geronb/ gbu046.

Pruchno,R.A.,Wilson-Genderson,M.,&Cartwright,F.P.(2010).A two-factor model of successful aging.Journal of Gerontology:Psychological Sciences,65(6), 671e679.

Pruchno,R.A.,Wilson-Genderson,M.,Rose,M.,&Cartwright,F.P.(2010).Successful aging:early in?uences and contemporary characteristics.The Gerontologist, 50(6),821e833.

Reviere,R.,Berkowitz,S.,Carter,C.C.,&Ferguson,C.G.(1996).Needs assessment:A creative and practical guide for social scientists.New York:Routledge.

Roos,N.P.,&Havens,B.(1991).Predictors of successful aging:a twelve-year study of Manitoba elderly.American Journal of Public Health,81(1),63e68.

Rowe,J.W.,&Kahn,R.L.(1987).Human aging:usual and successful.Science, 237(4811),143e149.

Rowe,J.W.,&Kahn,R.L.(1998).Successful aging.New York:Pantheon. Strawbridge,W.J.,Cohen,R.D.,Shema,S.J.,&Kaplan,G.A.(1996).Successful aging: predictors and associated activities.American Journal of Epidemiology,144(2), 135e141.

Strawbridge,W.J.,&Wallhagen,M.I.(2003).Self-rated successful aging:correlates and predictors.In L.W.Poon,S.H.Gueldner,&B.M.Sprouse(Eds.),Successful aging and adaptation with chronic diseases in older adulthood(pp.1e24).New York:Springer.

Strawbridge,W.J.,Wallhagen,M.I.,&Cohen,R.D.(2002).Successful aging and well-being:self-rated compared with Rowe and Kahn.The Gerontologist,42(6), 727e733.

Sun,V.K.,Cenzer,I.S.,Kao,H.,Ahalt,C.,&Williams,B.A.(2012).How safe is your neighborhood?Perceived neighborhood safety and functional decline in older adults.Journal of General Internal Medicine,27(5),541e547.

United Nations,Department of Economic and Social Affairs,Population Division.

(2013).World population prospects:The2012revision,key?ndings and advance tables.Working paper no.ESA/P/WP.227.Available https://www.wendangku.net/doc/e016409425.html,/wpp/ Documentation/pdf/WPP2012_%20KEY%20FINDINGS.pdf Accessed12.01.14. Wright,J.,Williams,R.,&Wilkinson,J.R.(1998).Development and importance of health needs assessment.British Medical Journal,316(7140),1310e1313.

E.K.Cromley et al./Applied Geography56(2015)13e20 20