• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of amjphAmerican Journal of Public Health Web SiteAmerican Public Health Association Web SiteSubmissionsSubscriptionsAbout Us
Am J Public Health. 2004 October; 94(10): 1768–1774.
PMCID: PMC1448532

Local Area Deprivation and Urban–Rural Differences in Anxiety and Depression Among People Older Than 75 Years in Britain

Abstract

Objectives. We sought to determine the association of depression and anxiety with “area deprivation” (neighborhood socioeconomic deprivation) and population density among people older than 75 years in Britain.

Methods. Postal codes were used to link census area information to individual data on depression and anxiety in 13349 people aged 75 years and older taking part in a trial of health screening.

Results. Living in the most socioeconomically deprived areas was associated with depression (OR=1.4), but this relation disappeared after adjusting for individual deprivation characteristics. There was no association with anxiety. Living in the highest density and intermediate low-density areas was associated with depression (OR=1.6 and 1.5) and anxiety (OR=1.5 and 1.3) compared with the lowest density areas.

Conclusions. An association between area deprivation and depression in older people was explained by individual health, demographic, and socioeconomic factors. Higher population density was consistently associated with increased depression and anxiety.

Depression and anxiety are common problems in later life and are associated with considerable morbidity.1,2 Their origins are complex and multifactorial. A variety of potential risk factors, including female gender, low income, social isolation, loneliness, life events, absence of confiding relationships, and physical illness, have been reported in older people (≥55 years).2,3 There are complex overlapping pathways of effect for relations between socioeconomic factors and mental health. The relation between “area deprivation” (neighborhood socioeconomic deprivation) and psychiatric illness has been explored in general adult populations but not in older populations. In younger adults (<65 years), associations have been demonstrated between high scores on local area composite deprivation indices and overall psychiatric admission,4,5 suicide/parasuicide rates,5 and psychological distress.6,7 However, the relative importance of local area deprivation and individual deprivation is not clear.6,7

The social and physical environment in which people live is an important potential risk factor in depression and anxiety, but it has not been studied extensively in older populations. Past research has tended to focus on either rural or urban environments or younger adult populations and to be on relatively small samples, often not adequately controlled for confounding. This research has produced conflicting results.1,8–17 Historically, adults living in urban environments were thought to be at higher risk of depression9–11 and general psychological distress,12,13 although there was some evidence this may not apply to older populations.14 However, further research in younger adults in North America has found no difference in rates of psychiatric disorder between rural and urban areas15,16 and increased depression in men living in more rural areas.17 Rural living has been associated with stigmatized attitudes toward mental health care and reduced willingness to seek help,17 fewer visits to a mental health specialist,18,19 lower concordance with guidelines on treatment,18 increased risk of admission for mental health problems,19 and increased risk of suicide.20 Little research has specifically examined the relations between a finer categorization of population density and psychiatric morbidity. There is some evidence that the relationship between population density and mental illness is nonlinear and forms part of a more complex pathway or “system.”21 Other research has highlighted intrarural variation in rates of nonspecific, limiting long-term illness, suggesting a J-shaped curve with higher morbidity in both extreme rural and urban areas.22

The impact of socioeconomic and environmental risk factors on mental health may change in later life. For example, low job control, job insecurity, and unemployment are frequently cited risk factors that clearly have less relevance to older populations. The local living environment and deprivation may be more important to older people because they may spend more time in the immediate vicinity of their homes after retirement or because of other factors such as decreased mobility. They may have a heightened sense of vulnerability to factors associated with areas of deprivation, such as crime, which affects their confidence to leave their homes and hence increases social isolation.23 We evaluated for the first time the relation between local environment, deprivation, and mental well-being in a large representative sample of community-dwelling people older than 75 years.

METHODS

Our objective was to determine if there are significant associations between (a) living in either a more socioeconomically deprived or more heavily populated area and (b) depression and anxiety in community-dwelling older people in Britain. Our article describes a secondary analysis of existing data linked from 2 sources: The Medical Research Council (MRC) trial of the assessment and management of older people in the community and the 1991 population census data for Britain. The MRC trial was a cluster-randomized trial that investigated approaches to multidimensional screening for people aged 75 years and older.24 A representative sample of 106 family practices across diverse regions of Britain was selected, stratified by mortality experience (standardized mortality ratio) and Jarman index (an area deprivation measure indicator).25 Practices were randomized into universal and targeted screening arms. In the universal arm, all patients aged older than 75 years (excluding those resident in nursing homes or with terminal illnesses) were invited to complete a detailed screening needs assessment during 1995–1999, including symptoms of depression and anxiety and sociodemographic, physical, and functional variables. In the targeted arm, only selected patients completed a detailed assessment; therefore, only universal arm patients were included in this analysis as a representative sample of 15173 community-dwelling older people. Participants were linked by their residence zip code to the enumeration district (ED) in which they lived and hence to measures of area deprivation (Carstairs) and population density derived from the national population census.26 EDs are the smallest geographical units for which data are available, typically comprising 170 households or 400 people and reflecting the characteristics of the immediate vicinity of a persons’ home.

Measures

Depression was measured with the Geriatric Depression Scale, an extensively used and validated tool for community samples of older people.27–29 A case of depression was defined as a score of ≥ 6—a widely used threshold with a good balance of sensitivity and specificity.28,29 Anxiety was measured with the anxiety subscale of the General Health Questionnaire (GHQ-28), a measure of psychological distress in community samples extensively validated with good sensitivity and specificity.30,31 The 7-item anxiety subscale correlates highly with psychiatrists ratings of morbid anxiety on a structured diagnostic interview.30 No established thresholds determine caseness for the subscale because it is 1 of 4 scales in the GHQ-28. Anxiety symptoms were dichotomized to no symptoms or 1 or more symptoms.

The Carstairs deprivation score, a composite deprivation index that combines levels of unemployment, overcrowding, non–car ownership, and social class, was calculated from census data for each ED.32 Smoothed local area population densities over a standard radius (5 km) were calculated from census data based on the mean population density of all EDs whose centroid fell within a 5-km radius of the ED in which the participant resided. Spatially smoothed markers of population densities provided a clearer distinction of inner and outer city, town, and rural areas than unsmoothed markers, for example, distinguishing a built-up area within a small town from an area of similar population density within a city.33 We categorized the Carstairs deprivation score and the population density for each ED into 4 groups using quartiles for the entire British census data set. The 4 groups of population density represent the following: 1 = lowest density quartile (0–355 people/km), 2 = intermediate lower density (356–1069 people/km), 3 = intermediate higher density (1070–2466 people/ km), and 4 = highest density quartile ( ≥ 2467 people/km).

Analysis

Of 15173 subjects, we omitted 1824 from the main analyses because they had incomplete exposure or outcome data. Multiple logistic regression was performed with Stata 7 with adjustment for sampling design (clustering and stratification) with the survey commands, which are based on Generalized Estimating Equations with independent working correlation matrices and robust standard errors.34 A wide range of variables were available on each individual from the detailed needs assessment in the MRC trial. Many of these were conceptually similar and highly correlated. The inclusion of such highly correlated variables together would make parameter estimates unstable. We included age and gender as potential confounding factors. Housing status (owner-occupied, rented, or sheltered housing/residential care) and financial stress (difficulty making ends meet) were included as individual socioeconomic variables. Current physical illness, cognitive impairment (measured with the Mini Mental State Examination35), functional activity (unmet needs in activities of daily living), lifestyle (alcohol intake, smoking), life events in the past year, and social networks (confiding relationships, marital status, living alone, access to help) were considered as potential confounding or intermediate variables. To avoid model overload and potential colinearity, we omitted some variables after the crude analysis.

We examined the correlations and univariate associations with the main exposures and outcomes and consequently based the analysis on a selected subset of the following individual factors: age, gender, housing status, financial stress, living alone, cognitive function, current physical symptoms (e.g., auditory or visual impairment, respiratory symptoms, reported diagnosed heart disease or cancer), and unmet needs in activities of daily living. Living alone was highly correlated with all other social support and network variables and was selected as the only 1 of these factors that showed an association with area deprivation and population density in the univariate analysis. Alcohol and life events in the last year were omitted as potential confounders because no association was demonstrated with either area deprivation or population density. Smoking was weakly associated with both depression and area deprivation (but not anxiety or population density) in the univariate analysis. However, smoking made no difference to adjusted odds ratios in the model, and we considered reverse causality to be highly likely; therefore, we excluded smoking from the models. The regression models were fitted sequentially in a forward-fitting fashion, assessing each factor in turn. We separately analyzed the effect of housing status (the best available proxy for individual deprivation) on the respective relationships of population density and deprivation with anxiety and depression. For clarity, we used the same models to explore associations with both outcomes (depression and anxiety) and both exposures (area deprivation and population density).

The shape of the associations between categories of environmental factors and the outcomes was assessed by inspecting for log-linear trends in odds across subcategories and assessing whether nonlinear and quadratic terms were a better fit. Consequently, deprivation score, population density, and age group were treated as nonlinear categorical variables. Sub-categories of population density and deprivation score were collapsed to give sufficient power to test for interactions, and subsequent tests for interaction were nonsignificant.

RESULTS

The detailed screening assessment of health and social needs was completed for 15173 of 21241 (71.4%) of the eligible people in the universal arm of the MRC trial. Nonresponders were more likely to be female (68% vs 62%) and older (aged ≥ 85 years, 27% vs 21%). There were no significant differences in the Jarman index or Standardized Mortality Ratio of the practice from which they were selected. Complete data were available on outcomes and main exposures for 13349 of 15173 (88%) of participants. Those with missing data were more likely to be older (aged ≥ 85 years, 28% vs 21%), female (65% vs 61%), living in sheltered accommodation or residential care (18% vs 8%), and cognitively impaired (38% vs 19%). They were similar in all other respects.

Study Population Characteristics

The median age of respondents was 80 years (interquartile range 77–84 years), with a range of 75–102 years. Just fewer than half the sample lived alone; 61% were female and 63% owned their accommodation. Nearly half had 3 or more current physical symptoms, 19% had cognitive impairment when the 24/30 threshold on the Mini Mental State Examination was used, and 7% had unmet needs in functional activity (i.e., difficulties performing activities of daily living, with no help available). Eight percent scored above the 5/6 Geriatric Depression Scale threshold for depression (median score 2/15, interquartile range 1–3). The majority had no symptoms of anxiety. The prevalence of anxiety was 18.3% at a 0/1 threshold (1 or more symptoms), 9.0% at a 1/2 threshold (2 or more symptoms), and 4.6% at a 2/3 threshold (3 or more symptoms). The sample was relatively concentrated in the less deprived areas (31% in lowest Carstairs deprivation quartile) and lowest density areas (31% in least densely populated quartile).

There were statistically significant associations between depression and increasing age, female gender, rented housing, financial stress, living alone, social support, life events, and physical and functional ability indicators (Table 1 [triangle]). These associations have been described in more detail elsewhere.36,37 Anxiety was significantly associated with female gender, financial stress, functional ability, physical health, lack of confiding relationship, access to help, and negative life events but not age, housing status, cognitive function, marital status, living alone, or high alcohol intake.

TABLE 1
Univariate Associations Between Variables and Depression and Anxiety

Multivariate Analysis

There was evidence of an association of area deprivation score and depression (adjusted by age and gender and population density, Table 2 [triangle]), with a small rising trend in point estimates and an odds ratio (OR) of 1.4 for depression in the most deprived compared with the least deprived areas. The association became nonsignificant after adjustment for housing status and disappeared completely when further adjusted for individual physical-, functional-, social-, and deprivation-related variables (P= .84). There was no statistically significant association between area deprivation score and anxiety, either crudely or after adjustment for potential confounding factors (P= .65), but the estimates showed a slight negative gradient. The results for anxiety were not sensitive to the choice of cutpoint (0/1, 1/2 or 2/3 thresholds).

TABLE 2
Multivariate Analysis—Associations of Carstairs Deprivation Score and Depression/Anxiety

After adjustment for age, gender, area deprivation, and population density, depression was associated (P< .001) with living in rented (adjusted OR = 1.24) and sheltered accommodation/residential care (adjusted OR = 1.62) compared with those who owned their housing. There was, however, no association between housing status and anxiety (adjusted OR = 1.06 for rented and 1.16 for sheltered accommodation/residential care, P= .27).

There was evidence for associations of higher population density with both depression (P=.006) and anxiety (P=.019) (Table 3 [triangle]). The results were not sensitive to choice of cutpoint for anxiety. Adjusted odds ratios for both depression and anxiety were greatest in the highest density quartile (depression OR = 1.6, anxiety OR = 1.5) and intermediate lower density quartile (depression OR = 1.5, anxiety OR = 1.3) compared with the lowest density quartile. The confidence intervals were wide and overlapped for intermediate lower and intermediate higher density quartiles; therefore, these results should be interpreted with caution. The associations were consistent for both outcomes and after adjusting for the full range of variables previously mentioned, including both area deprivation and housing status.

TABLE 3
Multivariate Analysis—Associations of Population Density and Depression/Anxiety

DISCUSSION

In this study, living in a more deprived area was associated with depression. This association became nonsignificant after adjustment for housing status, a proxy for individual deprivation, and entirely disappeared after adjustment for individual physical health, functional ability, and social variables. This pattern is consistent with some previous work in younger populations, in which most of the association of deprivation and psychological distress was explained by individual rather than local area factors.6,7 This is to be expected, because area-based and individual measures of deprivation are highly correlated. The pathways are likely to be complex, but local area deprivation may have an indirect impact on depression through these intermediate individual physical, functional, social, and deprivation factors. Area deprivation may be associated with higher crime rates and lower social cohesion, which may lead to increased fear of crime, and social isolation, which may lead to increased depression.23 Lower expectations and a lower sense of control may be associated with living in a more deprived area,38 which again in turn is associated with increased depression.

Contrary to our hypothesis, local area deprivation showed no significant association with anxiety score both crudely and after adjustment for potential confounding factors; instead it showed a slight negative gradient. However, the confidence intervals were wide and the trend was nonsignificant. Our study also found no association between anxiety and housing status, and other work has shown no association between individual socioeconomic status and anxiety in late life.2,3 We found no comparable studies reporting on the associations between anxiety symptoms and local area deprivation in late life or younger populations.

The results appear to indicate that living in both the highest population density and intermediate lower density areas is associated with increased depression and anxiety compared with the lowest density areas. The confidence intervals for the increased depression and anxiety in the intermediate lower density quartile are wide, and for anxiety in particular, the observed association in this subgroup may have occurred by chance. Associations of psychological disorders and urban living have been found in a number of previous studies in younger populations,9–13 although this finding is not consistent.14–17 Because most studies have assessed a crude rural/urban split, associations with areas of intermediate population density generally have not been reported. One study in the United States showed higher rates of depression in small villages compared with large towns and cities,17 and other work has described nonlinear associations, suggesting that various alternative factors such as social organization, adaptation, previous group experience, and environment also contribute.22 The effects of population density on mental health are complex, because they interweave with other aspects such as the local physical environment (e.g., the nature of local housing or availability of green spaces) and sociocultural and economic factors. There is some previous evidence that the physical environment, including population density, has equivalent importance to the sociocultural environment in explaining regional variations in psychosocial health.39

No single factor is likely to explain the association between population density and mental health. Explanations may include factors such as variations in the provision of health care, social and leisure services, ethnicity and culture, attitudes toward mental health problems, crime rates, or suitability of available housing in different areas. For example, in New Zealand and Australia, distance from a health center was a significant but nonlinear predictor of mental health service use.40,41 It may be that living in an area of higher population density, often associated with higher population mobility, means people are less likely to know their neighbors (i.e., the anonymity of city living), which might in turn lead to social isolation and loneliness in later life, increasing the risk of depression and anxiety.

There are few sources of measurement bias that could explain the differences found. Data on individual variables were collected 4 or more years later than the population census. Area sociodemographic characteristics may have changed over this time, causing nondifferential misclassification bias and an underestimation of the effect. This study was limited to self-reported data through nurseled assessments, and there was no access to any other corroborating sources. However, the instruments used (particularly the Geriatric Depression Scale) have been shown to have good interrater reliability and validity in a number of settings. The GHQ-28 anxiety subscale may not correspond to a definitive diagnosis of anxiety; however, previously, it has correlated highly with a clinical diagnosis of anxiety.30 Some variables were reported relatively infrequently, such as high alcohol intake (2%), which may reflect underreporting because of social desirability, poor sensitivity of the questions used, or less drinking by older people.

Response bias was unlikely to be large enough to affect the observed odds ratios substantially, because response rates were high. The relative underrepresentation of those living in sheltered accommodation/residential care or who were cognitively impaired (because of exclusions for missing data in this analysis) may mean that the results are not applicable to the most frail or cognitively impaired section of the population. The exclusion of those in nursing homes in the main trial may have introduced some selection bias from geographical variability in the availability of nursing homes. In this study, however, although the proportion of people living in sheltered accommodation and residential care was higher in the 2 intermediate quartiles of population density, adjustment for housing type did not affect the relationship between population density and mental health.

There may be residual confounding, in which variables in the model do not truly reflect the dimension they were intended to measure or data were not available, for example, income, access to health care, or other aspects of the local environment. It has been suggested that composite measures of deprivation are good explanatory variables in urban but not in rural areas, where other factors such as accessibility to public services are important and adjustment based on these may not be adequately controlling for area deprivation in rural areas.22,42 This analysis was based on cross-sectional data; therefore, temporal relationships were not established, and reverse causality cannot be excluded (e.g., depressed or anxious older people could have changed their locations to be closer to relatives).

Our study supports arguments that there is some association of local area deprivation with depression but not anxiety in older people; this association appears to be explained by individual socioeconomic, physical, functional, and social factors. Although older people may have different needs, such as more physical and functional problems, the relationship of local area deprivation to psychological well-being appears similar to that found in younger populations. Further work is needed to establish whether the higher prevalence of depression and anxiety in older people in higher population density areas found in this study can be replicated and is explained by other aspects of the environment, such as access to services, crime rates, or intolerance to mental health problems. There are potential implications for the provision of health and social services and more generally for policies affecting community cohesiveness.

Acknowledgments

The MRC trial of Assessment and Management of Older People in the Community was funded by the Medical Research Council, Department of Health and Scottish Office (grant G9223939). K. Walters was supported by a research fellowship from the Medical Research Council.

We thank nurses, practice staff, and patients in the participating practices for the MRC trial. We also thank Chris Grundy at the London School of Hygiene and Tropical Medicine for providing postcode linked deprivation data.

Human Participant Protection

Ethical approval was obtained from the relevant local research ethics committees for all family practices participating in the MRC trial.

Notes

Contributors
K. Walters devised and completed the data analysis and wrote the article. E. Breeze was involved in administering the MRC trial, preparing the data for analysis, and advising on this analysis and interpretation. P. Wilkinson supervised linkage of the census data and environmental aspects of the data analysis and interpretation. G. M. Price advised on aspects of the statistical analyses. A. Fletcher is the principal investigator and C. J. Bulpitt is a coinvestigator of the MRC trial; they were responsible for its design and implementation. All authors commented on drafts of the article.

Peer Reviewed

References

1. Lindesay J, Briggs K, Murphy E. The Guy’s/Age Concern survey. Prevalence rates of cognitive impairment, depression and anxiety in an urban elderly community. Br J Psychiatry. 1989;155:317–329. [PubMed]
2. Beekman AT, De Beurs E, Van Balkom AJ, Deeg DJ, Van Dyck R, Van Tilburg W. Anxiety and depression in later life: co-occurrence and communality of risk factors. Am J Psychiatry. 2000;157:89–95. [PubMed]
3. De Beurs E, Beekman A, Geerlings S, Deeg D, Van Dyck R, Van Tilburg W. On becoming depressed or anxious in late life: similar vulnerability factors but different effects of stressful life events. Br J Psychiatry. 2001;179:426–431. [PubMed]
4. Koppel S, McGuffin P. Socio-economic factors that predict psychiatric admissions at a local level. Psychol Med. 1999;29:1235–1241. [PubMed]
5. Gunnell D, Peters T, Kammerling R, Brooks J. Relation between parasuicide, suicide, psychiatric admissions, and socio-economic deprivation. BMJ. 1995; 311:226–230. [PMC free article] [PubMed]
6. Reijneveld SA, Schene AH. Higher prevalence of mental disorders in socio-economically deprived urban areas in the Netherlands: community or personal disadvantage? J Epidemiol Community Health. 1998;52:2–7. [PMC free article] [PubMed]
7. Duncan C, Jones K, Moon G. Psychiatric morbidity: a multi-level approach to regional variations in the UK. J Epidemiol Community Health. 1995;49:290–295. [PMC free article] [PubMed]
8. Kirby M, Bruce I, Radic A, Coakley D, Lawlor BA. Mental disorders among the community-dwelling elderly in Dublin. Br J Psychiatry. 1997;171:369–372. [PubMed]
9. Dohrenwend B. Socio-cultural and social psychological factors in the genesis of mental disorders. J Health Soc Behav. 1975;16:365–392. [PubMed]
10. Neff JA. Urbanicity and depression reconsidered: the evidence regarding symptomatology. J Nerv Ment Dis. 1983;171:546–552. [PubMed]
11. Prudo R, Brown GW. Psychiatric disorder in a rural and an urban population: 1. aetiology of depression. Psychol Med. 1981;11:581–599. [PubMed]
12. Vazquez-Barquero JL, Munoz PE, Jauregui V. The influence of the process of urbanisation on the prevalence of neurosis: a community survey. Acta Psychiatrica Scandinavica. 1982;65:161–170. [PubMed]
13. Hodiamont P, Peer N, Syben N. Epidemiological aspects of psychiatric disorder in a Dutch health area. Psychol Med. 1987;17:495–505. [PubMed]
14. Crowell BA, George LK, Blazer D, Landerman R. Psychosocial risk factors and urban/rural differences in the prevalence of major depression. Br J Psychiatry. 1986;149:307–314. [PubMed]
15. Kessler RC, McGonagle KA, Zhao S, et al. Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the United States: results from the National Comorbidity Survey. Arch Gen Psych. 1994;51: 8–15. [PubMed]
16. Parikh SV, Wasylenki D, Goering P, Wong J. Mood disorders: rural/urban differences in prevalence, health care utilisation, and disability in Ontario. J Affect Disord. 1996;38(1):57–65. [PubMed]
17. Hoyt DR, Conger RD, Valde JG, Weihs K. Psychological help seeking in rural America. Am J Community Psychol. 1997;25(4):449–470. [PubMed]
18. Fortney J, Rost K, Zhang M, Warren J. The impact of geographic accessibility on the intensity and quality of depression treatment. Med Care. 1999;37:884–893. [PubMed]
19. Rost K, Zhang M, Fortney J, Smith J, Smith GR. Rural-urban differences in depression treatment and suicidality. Med Care. 1998;36:1098–1107. [PubMed]
20. Singh GK, Siahpush M. Increasing rural-urban gradients in US suicide mortality, 1970–1997. Am J Public Health. 2002;92:1161–1167. [PMC free article] [PubMed]
21. Sengel RA. A graph analysis of the relationship between population density and social pathology. Behav Sci. 1978;23:213–224. [PubMed]
22. Barnett S, Roderick P, Martin D, Diamond I. A multi-level analysis of the effects of rurality and social deprivation on premature limiting long term illness. J Epidemiol Community Health. 2001;55:44–51. [PMC free article] [PubMed]
23. Klinenberg E. Dying alone. The social production of urban isolation. Ethnography. 2001;2:501–531.
24. Fletcher AE, Jones DA, Bulpitt CJ, Tulloch AJ. The MRC Trial of assessment and management of older people in the community: objectives, design and interventions [ISRCTN23494848] BMC Health Services Research 2002, 2:21. Available at: http://www.biomedcentral.com/1472-6963/2/21. Accessed July 6, 2004. [PMC free article] [PubMed]
25. Jarman B. Underprivileged areas: validation and distribution of scores. BMJ. 1984;289:1587–1592. [PMC free article] [PubMed]
26. Dale A, Marsh C (eds). 1991 Census Users Guide, London, HMSO; 1993.
27. Arthur A, Jagger C, Lindesay J, Graham C, Clarke M. Using an annual over 75 health check to screen for depression: validation of the short Geriatric Depression Scale (GDS15) in general practice. Int J Geriatr Psychiatry. 1999;14:431–439. [PubMed]
28. Lyness JM, Noel TK, Cox C, King DA, Conwell Y, Caine ED. Screening for depression in elderly primary care patients. A comparison of the centre for epidemiological studies scale and the geriatric depression scale. Arch Intern Med. 1997;157:449–454. [PubMed]
29. Almeida OP, Almeida SA. Short versions of the geriatric depression scale: a study of their validity for the diagnosis of a major depressive episode according to ICD-10 and DSM-IV. Int J Geriatr Psychiatry. 1999; 14:858–865. [PubMed]
30. Goldberg D, Hillier VF. A scaled version of the General Health Questionnaire. Psychol Med. 1979;9: 139–145. [PubMed]
31. Goldberg D. The General Health Questionnaire Users Manual. Windsor, Berkshire, U.K., NEFR Nelson; 1978.
32. Carstairs V, Morris R. Deprivation and health in Scotland. Aberdeen University Press, Aberdeen, Scotland; 1992.
33. Grundy C, Wilkinson P, Fletcher P. Spatial smoothing of population and road density as markers for urbanisation for epidemiological studies. Proceedings of the 8th Symposium in Medical Geography, July 13–17, 1998, Baltimore, Md; 1998.
34. Stata 7 Reference Manual, College Station, Tex: Stata Corp; 2001:15–101.
35. Folstein M, Folstein S, McHugh, P. “Mini-Mental State” A practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res. 1975; 12:189–198. [PubMed]
36. Osborn DPJ, Fletcher AE, Smeeth L, et al. Geriatric Depression Scale Scores in a representative sample of 14,545 people aged 75 and over in the United Kingdom: results from the MRC trial of assessment and management of older people in the community. Int J Geriatr Psychiatry. 2002;17:1–8. [PubMed]
37. Osborn DPJ, Fletcher AE, Smeeth L, et al. Factors associated with depression in a representative sample of 14217 people aged 75 and over in the United Kingdom: Results from the MRC trial of assessment and management of older people in the community. Int J Geriatr Psychiatry. 2003;18:623–630. [PubMed]
38. Thompson E, Krause N. Living alone and neighbourhood characteristics as predictors of social support in late life. J Gerontol. 1998;53:S354–S364. [PubMed]
39. Brogan DR, James LD. Physical environment correlates of psychosocial health among urban residents. Am J Community Psychol. 1980;8:507–522. [PubMed]
40. Hall G. Monitoring and predicting community mental health centre utilisation in Auckland, New Zealand. Soc Sci Med. 1988;26:55–70. [PubMed]
41. Stampfer H, Reymond J, Burvill P, Carlson J. The relationship between distance from inpatient facilities and rate of psychiatric admissions I Western Australia. Soc Sci Med. 1984;19:879–884. [PubMed]
42. Haynes R, Gale S. Deprivation and poor health in rural areas: inequalities hidden by averages. J Epidemiol Community Health. 2000;54:778. [PubMed]

Articles from American Journal of Public Health are provided here courtesy of American Public Health Association

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

  • MedGen
    MedGen
    Related information in MedGen
  • PubMed
    PubMed
    PubMed citations for these articles

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...