Mental health in a gendered context: Gendered community effect on depression and problem drinking

ARTICLE IN PRESS Health & Place 15 (2009) 990–998

Contents lists available at ScienceDirect

Health & Place journal homepage: www.elsevier.com/locate/healthplace

Mental health in a gendered context: Gendered community effect on depression and problem drinking Lore van Praag a, Piet Bracke a,, Wendy Christiaens a, Katia Levecque a,b, Elise Pattyn a a b

Health & Demographic Research—HeDeRa, Department of Sociology, Ghent University, Korte Meer 5, 9000 Ghent, Belgium Research Foundation (FWO), Flanders, Belgium

a r t i c l e in f o

a b s t r a c t

Article history: Received 29 July 2008 Received in revised form 10 March 2009 Accepted 19 April 2009

Socio-economic features of a community influence people’s health. However, not all inhabitants are affected similarly. The present study explores gendered contextual effects on problem drinking and depression with the differential exposure, vulnerability and expression hypotheses of the social stress model in mind. Analyses are based on the pooled data of the Belgian Health Interview Survey 2001 and 2004 (N ¼ 21.367 respondents, N ¼ 589 municipalities). Results reveal that living in an area with high unemployment is more detrimental for women in terms of depression, but has the same impact on men and women when problem drinking is the outcome. & 2009 Elsevier Ltd. All rights reserved.

Keywords: Gender Context Unemployment Depression Problem drinking Belgium

1. Introduction Studies that focus on area effects are abundant. They interpret health variations in terms of effects of the social and physical environment to which all inhabitants of a community are exposed (Dalgard and Tambs, 1997; Driessen et al., 1998; Duncan et al., 1998; Kawachi et al., 1999; MacIntyre and Ellaway, 2000, 2003; MacIntyre et al., 2002; Pickett and Pearl, 2001; Robert, 1999). A diverse set of adverse conditions have been pinpointed as important community-level features affecting the health of local residents in previous research: income inequality (Lynch et al., 2004; Subramanian and Kawachi, 2004; Yen and Syme, 1999), unemployment (Hamilton et al., 1997; Lorant et al., 2008), fear of crime (Chandola, 2001), lack of social capital (Kawachi et al., 1999; Lomas, 1998) and social disorganisation (Faris and Dunham, 1960 cited in Yen and Syme, 1999; Silver, 2000). However, more insight into the mechanisms that connect the residential area and health outcomes is required (Cummins et al., 2005a; Tunstall et al., 2004; Weich, 2005; Yen and Syme, 1999). A good starting point for further research is the observation that area characteristics affect subpopulations differently (Diez-Roux, 2004; MacIntyre and Ellaway, 2000; Stafford and Marmot, 2003; Stafford et al., 2005). Numerous studies have

 Corresponding author. Tel.: +32 9 264 68 03; fax: +32 9 264 69 75.

E-mail address: [email protected] (P. Bracke). 1353-8292/$ - see front matter & 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.healthplace.2009.04.003

reported cross-level interaction effects (Diez-roux et al., 1997; Duncan et al., 1998; Stockdale et al., 2007; Waitzman and Smith, 1998; Weich et al., 2003). Specific gender differential contextual effects have also already received some attention (Blue, 2000; Kavanagh et al., 2006; Leclere et al., 1997; Molinari et al., 1998; Robert, 1999; Verheij, 1996). This study investigates the gendered community effects linked with the topic of mental health, taking as point of departure the assumption that men and women have a different relationship with their environment as well as with mental health. Some studies have found that place of residence affects women’s health more profoundly (Kavanagh et al., 2006; Monden et al., 2006; Poortinga et al., 2007; van Lenthe and Mackenbach, 2002), although the empirical underpinnings of this finding are inconsistent (Karvonen and Rimpela¨, 1996, 1997; Kelleher et al., 1992; Matheson et al., 2006; Molinari et al., 1998; Propper et al., 2005; Raleigh and Kiri, 1997; Skrabski et al., 2003; Stafford et al., 2005). We will focus on unemployment rates as a socio-economic feature of the community. Unemployment has been shown to have detrimental effects on mental health at the individual (Bracke and Wauterickx, 2003; Cummins et al., 2005b; Fone et al., 2007; Le´pine et al., 1997; Weich et al., 2003), as well as the household (Clark, 2003) and community level (Be´land et al., 2002; Catalano and Dooley, 1977; Cummins et al., 2005b; Fone et al., 2007; Hamilton et al., 1997; Lorant et al., 2008; van Lenthe et al., 2005). Most studies substantiate the gender differential effects of unemployment on the basis of role theory (De Goede and

ARTICLE IN PRESS L. van Praag et al. / Health & Place 15 (2009) 990–998

Maassen, 1998 cited in De Witte, 1999), finding more adverse effects for men, compared to women (Artazcoz et al., 2004; Paul, 2005). It is suggested that unemployed men experience more strain because they fail in their role of providing for the family as primary wage earner (Kulik, 2000). The role enhancement hypothesis implies that the existence of alternative family roles make it easier for women to cope with job loss (Artazcoz et al., 2004; Hunt and Annandale, 1993). Furthermore, other research emphasizes the gendered expression of distress, suggesting that this expression leads to health problems being more internalised in women, but more externalised in men (Kendler et al., 2003; Krueger et al., 2001; Rosenfield et al., 2000; Trickett and McBride-Chang, 1995). In brief, we will examine whether area-level unemployment rates have gendered effects irrespective of the mental health outcome used. In the present study, we chose depression to represent the internalising expression of mental distress and problem drinking to indicate externalising behaviour. With regard to gender differences in mental health, the social stress model provides a useful frame of reference, because it takes gender-specific effects of stressors, mediators and outcomes into account (Aneshensel et al., 1991; Pearlin, 1989). The social stress model covers three main hypotheses referring to exposure, vulnerability and expression. The differential exposure hypothesis assumes that women and men are exposed to various types of stressors with different intensities (Matheson et al., 2006), due to, for example, different experiences in the work environment (Molinari et al., 1998). The differential exposure hypothesis is usually complemented by the differential vulnerability hypothesis, which argues that vulnerability factors mediate the health consequences of stressors. As a result, women and men differ in responsiveness to stressors due to several biological, psychological and social factors (Bracke, 1993; De Silva et al., 2005; Ferlander, 2007; Lin and Ensel, 1989; Pearlin, 1989; Thoits, 1995), such as higher sensitivity (Turner et al., 1995), lower selfesteem or higher dependency on social support among women (Thoits, 1995; Wohlgemuth and Betz, 1991). But some studies refute this fact (Almeida-Filho et al., 2004; Bebbington, 1996; Piccinelli and Wilkinson, 2000). The third hypothesis of differential expression assumes that individuals express the strain they experience in different ways (Aneshensel et al., 1991; Aneshensel, 1992; Bracke, 1993; Dohrenwend and Dohrenwend, 1976; Pearlin, 1989). The expression of health problems is more internalised in women, while men express their distress in a more externalised way (Kendler et al., 2003; Krueger et al., 2001; Rosenfield et al., 2000; Trickett and McBride-Chang, 1995). In accord with both the differential exposure and the differential vulnerability hypotheses, a gendered community effect hypothesis states that the mental health of women is more affected by stressors stemming from their area of residence, compared to men (Kavanagh et al., 2006; Molinari et al., 1998; Monden et al., 2006; Poortinga et al., 2007; Propper et al., 2005; Stafford et al., 2005; van Lenthe and Mackenbach, 2002). Basically, because their social roles are more embedded in the local area (Robert, 1999), they spend more time at home, are more inclined to work part time (Kavanagh et al., 2005) and have more contact with the local neighbourhood when shopping or accompanying children to school (Floro and Miles, 2003). They are more inclined to be employed in the local community too, as a consequence of the gender gap in commuting. Moreover, their role as pivotal care provider (Bracke et al., 2008) also binds them to local informal care relationships with neighbours. Actually, when using unemployment rates as an environmental stressor, we allow for a very conservative test of the first hypothesis: even when we expect unemployment to have a more adverse effect on men’s mental health, unemployment rates at the community level are still

991

assumed to be more detrimental for women. Specifically, we expect that the mental health of women is more affected by community unemployment rates in comparison with men’s mental health. The differential expression hypothesis implies that research on the gendered nature of community effects is flawed when only one indicator of mental health problems is taken into account. In other words, a more rigorous test of both aforementioned hypotheses suggests that cross-level effects on both internalising and externalising mental health problems need to be examined. Previously, several studies examined community-level effects on internalised expressions of distress, such as depression and anxiety (Kawachi and Berkman, 2001) or on externalised problem behaviour, such as problem drinking (Forcier, 1985, 1988; Hamilton et al., 1990; Hill and Angel, 2005; Lahelma et al., 1995) and antisocial behaviour (Ingoldsby and Shaw, 2002; Kendler et al., 2003), but few study examined gendered cross-level effects on both dimensions of mental health problems in one design (Parker et al., 1987; Silver et al, 2002). Hence, by taking both depression and problem drinking into consideration, our final hypothesis examines whether gender-specific differences in residential-area effects are, at least partially, due to a gender-specific expression of mental health problems. In sum, taking the differential exposure, differential vulnerability and differential expression hypotheses into account, the link between residential area, gender and both internalising and externalising dimensions of mental health will be targeted. More specifically, the effect of the cross-level interaction between gender and unemployment rate will be modelled with problem drinking and depression as gendered outcomes of mental health. In view of the aforementioned hypotheses, we expect residential unemployment rates to exert more of an effect on women. The effect will be perceptible especially in the effect on depression, because women are more inclined to internalise their mental problems. Additionally, previous research has outlined a series of risk factors for mental distress, which cannot be ignored in the analyses. Warr (1984 cited in De Witte, 1999), for example, emphasized that single women find unemployment as distressing as men, because they fulfill the role of sole wage earner. Findings in the literature on whether or not having children is beneficial or detrimental have been inconsistent (Artazcoz et al., 2004; Gutie´rrez-Lobos et al., 2000). But social ties in general have been well recognized as an advantage (Kawachi and Berkman, 2001; Pearlin, 1989; Ross and Mirowsky, 2006; Turner et al., 1995). The other SES indicators, education and income, should also be incorporated (Araya et al., 2003; Bebbington, 1998; Bracke and Wauterickx, 2003; Kawachi and Berkman, 2001; Lewis et al., 2003; Lorant et al., 2008; Ross and Mirowsky, 1999; Ross and Van Willigen, 1997; Stafford and Marmot, 2003). Regarding the area level, some studies have revealed that population density and mean area income are related to unemployment rates (McCulloch, 2001; Peterson et al., 2009; Pickett and Pearl, 2001; Pritchard and Evans, 1997; Sundquist et al., 2004). A high population density is linked with a lot of stress according to Taylor et al. (1997) and Greiner et al. (2004), while the median income is negatively associated with mental health (Galea et al., 2007; Weich et al., 2006). Therefore, the preceding determinants should be added to the conceptual model as controls.

2. Methods 2.1. Sample Individual and family level variables are obtained from the pooled data of the Belgian Health Interview Surveys (HISs) from

ARTICLE IN PRESS 992

L. van Praag et al. / Health & Place 15 (2009) 990–998

2001 and 2004. An extensive description of the methods, sampling frame and respondents is provided elsewhere (Scientific Institute of Public Health (SIPH), 2006). Data gathering was based on a multistage stratified cluster sample in which selected households were geographically clustered according to municipality (N ¼ 589); within each household a subsample of a maximum of four individuals was taken. The HIS (2001/2004) provides cross-sectional information that is representative of the population in private and collective households age 15 or older (N ¼ 10,092 men, N ¼ 11,284 women). For part of the questions, e.g., those concerning mental health, proxies were not allowed. The institutionalised population in psychiatric care was not contacted. Household characteristics were assessed using a household questionnaire; individual characteristics were assessed by means of a verbal and a written questionnaire administered to the selected household members. The response rate at the household level is 61.4% for both survey years. The data at the micro-level were merged with area-level data obtained from Statistics Belgium (2008) based on corresponding municipality codes. 2.2. Variables 2.2.1. Mental health Depression is measured with the self-report subscale of the Symptoms Checklist 90-Revised (Derogatis, 1977). This scale has been extensively used in international literature, and has been shown to have excellent reliability and validity in detecting depressive symptoms in the general population in Belgium (Levecque, 2006). The scale consists of 13 Likert-type items, with answering categories reflecting the frequency of symptoms during the last week. Total subscale scores were assessed when maximum three items were missing. Missing items were imputed by means of the item correlation substitution technique (Huisman, 1999), resulting in an item response rate of 91.1%. Problematic drinking is assessed with the CAGE test, which has been developed to screen for alcohol use in adults (Ewing, 1984). Several studies suggest the CAGE test to be a reliable and valid measurement instrument within a hospital, general practice and general population context (Smart et al., 1991; Alvarez and Del Rio, 1994). Cronbach’s a is 0.64, which can be regarded as satisfactory for a four-item scale (Netemeyer et al., 2003). The symptoms addressed in the CAGE are ‘cut down attempts’, ‘being annoyed at criticism of one’s drinking’, ‘feeling guilty about drinking’ and ‘early-morning drinking’. Respondents could answer with ‘no’ (1) or ‘yes’ (2). The scale is calculated with a maximum of two missing items (range 4–8) and missing items are imputed, again making use of the item correlation substitution (Huisman, 1999). The resulting response rate is 88.1%. 2.2.2. Individual and household characteristics The key variable gender is represented by a dummy variable: women receive score 1 and men are defined as the reference category. The current employment status is expressed by one of the following which are compared to people who have a job: ‘retired’, ‘disabled’, ‘unemployed’, ‘student and others’ and ‘househusband/wife’. On the individual level, the analyses are adjusted for the following control variables: age, education, household type, household income and social support. Age is measured in years and the education level is indicated by the highest diploma or degree (‘no diploma or primary education’, ‘lower secondary’, ‘higher secondary’ and ‘higher education’ as reference category). At the household level, the household type is categorised as one of the following types: ‘single’, ‘single with child(ren)’,

‘couple’, ‘complex household’ and ‘couple with children’ (reference category). Information on the net household income equivalent is created by adopting the Organisation for Economic Co-operation and Development (OECD) scale, which gives a weight of 1 to the first adult of the household, 0.5 to all other adults (417 years old) and 0.3 to children. In this way, income varies along five categories: ‘less than h750’, ‘h750–h1000’, ‘h1000–h1500’, ‘h1500–h2500’ and ‘more than h2500’ (reference category). In addition, social support is assessed using the Medical Outcomes Study (MOS) Social Support scale, composed of 19 Likert-type items referring to perceived affective, emotional and instrumental support and positive interaction. Answering categories range from ‘never’ to ‘always’, and scores were calculated as suggested by Sherbourne and Stewart (1991). 2.2.3. Residential-area characteristics At the municipality level, unemployment rate is constructed by the ratio of the unemployed population (whether or not currently seeking a job), compared to the sum of all individuals aged 15–64 years old, or the potential working population. Population density and median area income are included as municipalitylevel controls. Population density is measured by means of the number of inhabitants per square kilometre. Median income indicates neighbourhood wealth and is based on the distribution of tax declarations of inhabitants in the municipality (see also Diez-Roux et al., 1997; Peterson et al., 2009; Pickett and Pearl, 2001). 2.2.4. Analysis procedure First, analyses based on the intra-class correlation explored whether a part of the variation in problematic drinking and depression could be attributed to the municipality level. The intra-class correlation was calculated by dividing the betweengroup variance of the dependent variable by the total variance of this variable, making use of the intercept-only model of the Mixed Models procedure in SPSS 15.0 (Hox, 2002). Results show that 3.4% of the variance in depression and 3.2% of the variance in problematic drinking can be attributed to the community level. Subsequently, multilevel regression techniques within the same SPSS module were used to examine the influence of residential area, over and beyond micro-level characteristics. In addition, cross-level interactions were tested (Goldsmith et al., 1998; MacIntyre and Ellaway, 2000; Pickett and Pearl, 2001). Before doing so, several of the variables were transformed. First, values were centred around the overall mean to enable an interpretation of the value ‘zero’ and to make it easier to interpret regression coefficients of interaction terms (Hox, 2002). Second, dependents were logarithmically transformed in order to address their skewed distribution and to render estimations more accurate (Moore and McCabe, 1999). To ensure that the sample represents the general population of interest, weights were applied to adjust for the probability of selection within the household, for municipality, province and region, and for interview timing (SIPH, 2006).

3. Results Before presenting the results of our analyses, descriptive statistics are given in Table 1. In Table 2, we report findings for multilevel analyses considering only main effects (Model 1), and considering both main and gender interaction effects (Model 2) on problem drinking and depression. Model 1 confirms that women report significantly more depressive complaints compared to men (B ¼ 0.039, SE ¼ 0.002, po0.001), whereas the latter show more

ARTICLE IN PRESS L. van Praag et al. / Health & Place 15 (2009) 990–998

993

Table 1 Sample characteristics (weighted sample). Continuous variables: number, mean and standard deviation (total, men and women) Total N

Men Mean

S.D.

Mental health outcomes (before log transformation) Depression 19,463 Problem drinking 18,842

1.339 4.223

0.528 0.634

Independent variables Social support Age

4.062 4.617

0.962 1.881

9.56 19,642 681

5.49 2002 1751

19,178 21,376

Area-level variables Unemployment rate (%) Median area income (EURO) Population density (inhabitants per km2)

589 589 589

N

Women Mean

S.D.

N

Mean

S.D.

9378 9193

1.260 4.317

0.432 0.749

10,085 9649

1.412 4.133

0.595 0.483

9220 10,362

4.053 4.496

0.979 1.809

9958 11,014

4.071 4.732

0.945 1.939

Categorical variables: number and valid percentage (total, men and women) N

%

N

%

N

%

Gender Women Mena

11,014 10,362

51.53 48.47

Labor market situation Retired Disabled Unemployed Student and others Housewife/husband Paid joba

4879 622 1150 379 1466 10,400

25.82 3.29 6.09 2.01 7.76 55.04

2304 318 463 124 26 5948

25.09 3.46 5.04 1.35 0.28 64.77

2575 304 687 255 1440 4452

26.51 3.13 7.07 2.62 14.82 45.83

Equivalent income of the household o750h 750–1000h 1000–1500h 1500–2500h 42500ha

991 1764 4175 6123 5199

5.43 9.67 22.87 33.55 28.49

394 667 2074 3030 2696

4.44 7.52 23.41 34.20 30.43

598 1097 2101 3093 2503

6.36 11.68 22.37 32.93 26.65

Education No diploma or primary education Lower secondary Higher secondary Higher educationa

3733 3815 5756 5200

20.17 20.62 31.11 28.10

1537 1816 3040 2566

17.16 20.27 33.93 28.64

2196 1999 2716 2634

23.01 20.94 28.45 27.60

Household type Single Single with child(ren) Couple Complex household Couple with child(ren)a

3842 935 6668 3166 6765

17.97 4.37 31.19 14.81 31.65

1686 231 3411 1603 3430

16.27 2.23 32.92 15.47 33.10

2156 704 3257 1563 3335

19.57 6.39 29.57 14.19 30.28

a

Reference category in multilevel regression models.

problem drinking (B ¼ 0.017, SE ¼ 0.001, po0.001). Age is positively related to depressive complaints (B ¼ 0.003, SE ¼ 0.001, p ¼ 0.004), but not to alcohol problems. Compared to individuals with paid employment, depression is more prevalent in people who are unemployed (B ¼ 0.028, SE ¼ 0.005, po0.001), retired (B ¼ 0.015, SE ¼ 0.004, po0.001), ill or disabled (B ¼ 0.128, SE ¼ 0.006, po0.001), and in students and others (B ¼ 0.017, SE ¼ 0.008, p ¼ 0.033). No significant association with depression is found for being a housewife/husband. The prevalence of problem drinking is also higher for the unemployed (B ¼ 0.010, SE ¼ 0.002, po0.001) in comparison with the employed, while people who are retired report less problem drinking than the employed (B ¼ 0.006, SE ¼ 0.002, p ¼ 0.003). With regard to educational level, a negative association is found with depression, but no association is found with problematic alcohol use. Social support has an impact on the amount of depressive complaints reported (B ¼ 0.033,

SE ¼ 0.001, po0.001) as well as on alcohol abuse (B ¼ 0.003, SE ¼ 0.001, po0.001). Household characteristics show that household type is associated with both mental health outcomes: significantly more depressive complaints are reported for singles, whether with children (B ¼ 0.015, SE ¼ 0.006, p ¼ 0.010) or without children (B ¼ 0.013, SE ¼ 0.004, po0.001), compared to couples with offspring. For problem drinking, only singles (B ¼ 0.009, SE ¼ 0.002, po0.001) risk more alcohol problems than couples with children. Finally, as concerns household income, lower income groups report more depressive complaints (oh750: B ¼ 0.014, SE ¼ 0.006, p ¼ 0.019; h750–h1000: B ¼ 0.014, SE ¼ 0.005, p ¼ 0.006) and alcohol problems (oh750: B ¼ 0.006, SE ¼ 0.003, p ¼ 0.021; h750–h1000: B ¼ 0.006, SE ¼ 0.002, p ¼ 0.007). When turning to the independent area-level variables in Model 1, results indicate that the unemployment rate is associated with

ARTICLE IN PRESS 994

L. van Praag et al. / Health & Place 15 (2009) 990–998

Table 2 Multilevel regression analyses: community effects on depression and problem drinking (respondents 15 years and older, Nmen ¼ 10,362; Nwomen ¼ 11,014; Nmunicipalities ¼ 589; weighted sample. Model 1 Dependent variables:

Depression B

Contextual effects Unemployment rate Unemployment rate*women Density Median area income Individual effects Intercept Women

Model 2 Problem drinking SE

SE

B

Problem drinking SE

B

SE

0.049 0.034 0.849 1.436

0.060** 0.013 0.007 1.875**

0.021 0.015 0.364 0.616

0.046

0.068***

0.020

0.047 1.410

0.848 1.435

0.000 1.889**

0.366 0.619

0.046 0.158*** 0.076 1.419

0.040*** 0.039***

0.003 0.002

0.614*** 0.017***

0.002 0.001

0.040*** 0.040***

0.003 0.002

0.613*** 0.015***

0.002 0.001

0.015*** 0.128*** 0.008 0.017* 0.028***

0.004 0.006 0.005 0.008 0.005

0.006** 0.001 0.002 0.002 0.010***

0.002 0.003 0.002 0.004 0.002

0.015*** 0.129*** 0.008 0.017* 0.020** 0.013

0.004 0.006 0.005 0.008 0.007 0.009

0.005** 0.001 0.001 0.002 0.025*** 0.026***

0.002 0.003 0.002 0.004 0.003 0.004

Equivalent income of the household (ref.: 4h2500) oh750 0.015* h750–h1000 0.016*** h1000–h1500 0.006 h1500–h2500 0.002 Social support 0.033*** Age 0.003**

0.006 0.005 0.004 0.003 0.001 0.001

0.006* 0.006** 0.003 0.002 0.003*** 0.000

0.003 0.002 0.002 0.001 0.001 0.000

0.015* 0.017*** 0.007 0.002 0.033*** 0.003**

0.006 0.005 0.004 0.003 0.001 0.001

0.007* 0.007** 0.003 0.001 0.003*** 0.000

0.003 0.002 0.002 0.001 0.001 0.000

Education (ref.: higher education) No diploma or primary education Lower secondary Higher secondary

0.015*** 0.010** 0.003

0.004 0.003 0.003

0.003 0.000 0.002

0.002 0.002 0.001

0.015*** 0.010** 0.003

0.004 0.003 0.003

0.003 0.000 0.002

0.002 0.002 0.001

Household type (ref.: couple with child(ren)) Single 0.013*** Single with child(ren) 0.015* Couple 0.003 Complex Household 0.004

0.004 0.006 0.003 0.004

0.009*** 0.003 0.002 0.002

0.002 0.003 0.001 0.002

0.013*** 0.013* 0.003 0.004

0.004 0.006 0.003 0.004

0.008*** 0.004 0.002 0.002

0.002 0.003 0.001 0.002

Employment status (ref.: paid labour) (Pre)pension Disabled Housewife/husband Student and others Unemployed Unemployed*women

0.127**

B

Depression

Significance: *po0.05; **po0.01; ***po0.001. Model 1: Baseline Model; Model 2: model with gender-specific interaction.

both depression (B ¼ 0.127, SE ¼ 0.046, p ¼ 0.005) and problem drinking (B ¼ 0.068, SE ¼ 0.020, po0.001). Median area income is associated with problem drinking only: more alcohol problems are noted in municipalities with a higher median area income (B ¼ 1.889, SE ¼ 0.0.619, p ¼ 0.003). Population density is not significantly linked with either health problem. In sum, our key variables gender, personal unemployment and area-level unemployment rates coincide with less favourable mental health outcomes. In Model 2 of Table 2, interaction effects between gender and unemployment are included, both at the micro-level and at the residential-area level. At the individual level, a gender differential expression effect is visible: the interaction effect shows that being unemployed is associated with problem drinking in men (B ¼ 0.026, SE ¼ 0.004, po0.001). With regard to depression, the personal experience of unemployment has the same impact on men and women. On the other hand, the unemployment rate of the area of residence seems to more strongly affect the amount of depressive complaints reported by women compared to men (B ¼ 0.158, SE ¼ 0.034, po0.001). No such interaction effect is found for problem drinking. Hence, a gender differential expression effect is present at the community level in case of depression, an internalised expression of mental health that is more pronounced in women.

4. Discussion In the present study, a gendered contextual effect on different mental health outcomes was explored through the incorporation of a cross-level interaction. Contextual effects were studied within the context of the general population in Belgium, using pooled, micro-level data from the Health Interview Surveys 2001 and 2004 and area-level data from Statistics Belgium 2001 (2008). On the micro-level, the results are similar to other research (Blue, 2000; Kawachi and Berkman, 2001; Levecque, 2006; Stockdale et al., 2007). Our results also indicate a higher risk of problem drinking in men and of depression in women. The most important finding of this study is that the area-level unemployment rates affect the mental health status of women and men in dissimilar ways and to a different extent. The impact of unemployment rates on depression in women is more profound, while the impact on problem drinking in women equals the impact on problem drinking in men. Before turning to our main findings, we want to draw attention to some limitations of our study. First, the validity of the CAGE alcohol-use screening test for general population samples has been challenged (Alvarez and Del Rio, 1994; Bisson et al., 1999; Bu¨hler et al., 2004; Cherpitel, 1998, 2002; Smart et al., 1991). Dhalla and Kopec (2007) argue that the questionnaire may lead to selective responding and socially desirable answers and that it

ARTICLE IN PRESS L. van Praag et al. / Health & Place 15 (2009) 990–998

indicates respondents’ feelings about alcohol use. Nevertheless, several studies indicate the validity of CAGE in a general population sample (Smart et al., 1991). Furthermore, an advantage of CAGE is that it does not incorporate information on effective alcohol consumption, because the amount of alcohol people can handle differs between men and women. Second, this study focuses on residential area, but the area one works in might be relevant as well. It is possible that municipality – as a politically demarked area – is not the most relevant area to take into consideration. In fact, it could be argued that one’s direct neighbourhood should be targeted to grasp the variation in mechanisms relating area to health (Dietz, 2002). Further investigation is needed to clarify what is the most appropriate spatial scale (Brady and Weitzman, 2007; MacIntyre et al., 2002). Third, due to the nature of the sample of the Health Interview Surveys, the data are not fully representative of all the municipalities in Belgium. The unemployment rate in our present sample is a little lower (9.6%) than the unemployment rate for the general population in Belgium (10.7%), the mean population density is a lot higher (681 inhabitants per km2) compared to the average of 336 inhabitants per km2 in Belgium, as is the mean median taxable income (h19,641 instead of h18,532) (Statistics Belgium, 2008). However, one positive characteristic of our sample is the exclusion of problems of spatial autocorrelation, since the sampled municipalities are seldom located next to each other (Lorant et al., 2001). Fourth, selection and causation processes co-exist and influence each other (Gary et al., 2007; Yen and Syme, 1999). Direct and indirect selection processes (Verheij, 1996) may explain a part of the health variations found, because they draw people with similar health-related characteristics to live together. But due to the cross-sectional design, no inferences could be made regarding causation. Finally, as indicated by the intra-class correlation, the identified contextual effects explain only a small proportion of the variance in problem drinking and depression. Similar results were found in several other empirical studies that have considered contextual effects on health (Pickett and Pearl, 2001; Weich, 2005). Although the explained variance was limited, it should be kept in mind that features of the area of residence affect a large number of people and complement the compositional effect of the individual characteristics of people living in the same community. Notwithstanding the fact that multilevel regression techniques control for individual characteristics, they cannot completely disentangle the micro-level from area-level effects on health. Because environmental influences also operate through individual characteristics, contextual effects could be underestimated (Diez-Roux, 2004; MacIntyre and Ellaway, 2003). Despite these shortcomings, our findings contribute to the study of mechanisms that connect place and mental health in several ways. In fact, the current study is the first to examine residential contextual effects on mental health in the general population in Belgium. Previous research within the Belgian context was limited to the risk factors and the prevalence of (common) mental health problems (Baruffol and Thilmany, 1993; Bracke, 1993, 1998; Bracke and Wauterickx, 2003; Levecque, 2006; Levecque et al., 2007; Levecque et al., 2008; Lorant et al., 2007), their occurrence during the lifecycle (Bonnewyn et al., 2007) and the persistence of depressive symptoms (Bracke, 2000). When we turn to the first main finding of this study, Model 1 shows that the unemployment rate of one’s residential area has an independent effect on both depression and problem drinking. This is in line with previous research in the United Kingdom, the Netherlands, Finland, Italy and Spain (Driessen et al., 1998; Duncan et al., 1995; Cummins et al., 2005b; Fone and Dunstan,

995

2006; Fone et al., 2007; van Lenthe et al., 2005). However, contradictory results were found in the west of Scotland (Ecob and MacIntyre, 2000) and the United States (Gary et al., 2007). The unemployment rate indicates the socio-economic conditions people live in and the deprivation of an area (Bellaby and Bellaby, 1999; Brenner, 1987; Hill and Angel, 2005; Kasl and Jones, 2000; Matheson et al., 2006; Novo et al., 2001). Besides, it is related to the general and future work conditions of inhabitants in the labour market. Therefore, it can be seen as a source of stress (Be´land et al., 2002). A second main finding in our study is the interaction effect between gender and area-level unemployment rates on depression. The gender differential expression hypothesis and the gendered community effects hypothesis are confirmed. Results indicate that women residing in places with high unemployment rates express more depressive complaints compared to men, in addition to women already having higher baseline rates of depression. But at the individual level, being unemployed is associated with depression without gender differences. With regard to problem drinking, the results show the opposite. Unemployment has no gendered effect on the area level; however, at the individual level, personal unemployment associated with problem drinking is more prevalent in men. These findings are in line with some empirical research relating place, health and gender. Contextual effects tend to have a greater impact on the health outcomes of women (Kavanagh et al., 2006; Poortinga et al., 2007; Propper et al., 2005; Stafford et al., 2005). Thus, the environment explains, in part, the gender differences in mental health, as stated by MacIntyre and Ellaway (2000). Women may be more exposed to stressors stemming from the immediate environment due to the social roles they fulfill (Matheson et al., 2006; Stafford et al., 2005). Women may also be more vulnerable, e.g., a lower sense of personal control may give them the feeling they cannot alter their environment (Ross and Mirowsky, 2002). Furthermore, an additional explanation for these gendered contextual effects may be the higher perception of stress experienced by women facing the same stressors as men (Day and Livingstone, 2003), with different health outcomes as a result. Finally, greater insight into gendered contextual effects was obtained through the inclusion of two mental health outcomes, as indicated by the differential expression hypothesis and the recurrent empirical findings of higher risks of depression in women and problem drinking in men (Dohrenwend and Dohrenwend, 1976; Nolen-Hoeksema, 2002; Prescott et al., 1999; Weissman, 1987). By incorporating both problem drinking and depression into our analyses as dependent variables, we prevent a gender bias due to the gender-specific expression of mental health outcomes. This gender-specific expression of emotional distress has been suggested in literature which has found more externalising behaviour in men and more internalising behaviour in women (Kendler et al., 2003; Krueger et al., 2001; Rosenfield et al., 2000; Trickett and McBride-Chang, 1995). Our findings validate the present strategy of using multiple outcomes to evaluate societal arrangements and their consequences for mental health, both at the individual and at the area level. These findings suggest that testing cross-level interactions might reveal important mechanisms linking area to health, and therefore should not be ignored in future research. Additionally, future research should pay attention to potential confounders such as employment status of the spouse (Penkower et al., 1988 cited in Catalano, 1991) and the commute distance (Matheson et al., 2006). Moreover, this study reveals some important policy implications. Because of the complexity of health, collaboration should be encouraged across discipline boundaries when approaching mental health (Moss, 2002). When interventions are

ARTICLE IN PRESS 996

L. van Praag et al. / Health & Place 15 (2009) 990–998

implemented, it should be kept in mind that different target groups require different residential prevention campaigns (Blue, 2000; Fone et al., 2007). References Almeida-Filho, N., Lessa, I., Magalha˜es, L., Arau´jo, M.J., Aquino, E., James, S.A., Kawachi, I., 2004. Social inequality and depressive disorders in Bahia, Brazil: interactions of gender, ethnicity and social class. Social Sciences & Medicine 59 (7), 1339–1353. Alvarez, F.J., Del Rio, C., 1994. Screening for problem drinkers in a general population survey in Spain by use of the CAGE scale. Journal of Studies on Alcohol 55 (4), 471–474. Aneshensel, C.S., 1992. Social stress: theory and research. Annual Review of Sociology 18, 15–38. Aneshensel, C.S., Rutter, C.M., Lachenbruch, P.A., 1991. Social structure, stress, and mental health: competing conceptual and analytic models. American Sociological Review 56 (2), 166–178. Araya, R., Lewis, G., Rojas, G., Fritsch, R., 2003. Education and income: which is more important for mental health? British Medical Journal 57 (7), 501–505. Artazcoz, L., Benach, J., Borrell, C., Corte`s, I., 2004. Unemployment and mental health: understanding the interactions among gender, family roles, and social class. American Journal of Public Health 94 (1), 82–88. Baruffol, E., Thilmany, M., 1993. Anxiety, depression, somatization and alcohol abuse prevalence rates in a general Belgian community sample. Acta Psychiatrica Belgica 93, 136–153. Bebbington, P.E., 1996. The origins of sex differences in depressive disorder: bridging the gap. International Review of Psychiatry 8 (4), 295–333. Bebbington, P.E., 1998. Sex and depression. Psychological Medicine 28 (1), 1–8. Be´land, F., Birch, S., Stoddart, G., 2002. Unemployment and health: contextual level influences on the production of health in populations. Social Science & Medicine 55 (11), 2033–2052. Bellaby, P., Bellaby, F., 1999. Unemployment and ill health: local labour markets and ill health in Britain 1984–1991. Work, Employment and Society 13 (3), 461–482. Bisson, J., Nadeau, L., Demers, A., 1999. The validity of the CAGE scale to screen for heavy drinking problems in a general population survey. Addiction 94 (5), 715–722. Blue, I., 2000. Individual and contextual effects on mental health status in Sa˜o Paulo, Brazil. Revista Brasileira de Psiquiatria 22, 116–123. Bonnewyn, A., Bruffaerts, R., Vilagut, G., Almansa, J., Demyttenaere, K., 2007. Lifetime risk and age-of-onset of mental disorders in the Belgian general population. Social Psychiatry and Psychiatric Epidemiology 42, 522–529. Bracke, P., 1993. Geslachtsverschillen in mentale gezondheid: enkele kritische bemerkingen omtrent het gebruik van het sociale stressmodel. Tijdschrift voor Sociologie 14 (2), 223–245. Bracke, P., 1998. Sex differences in the course of depression: evidence from a longitudinal study of representative sample of the Belgian population. Social Psychiatry and Psychiatric Epidemiology 33, 420–429. Bracke, P., 2000. The three-year persistence of depressive symptoms in men and women. Social Science & Medicine 51, 51–64. Bracke, P., Christiaens, W., Wauterickx, N., 2008. The pivotal role of women in informal care. Journal of Family Issues 29 (10), 1348–1378. Bracke, P., Wauterickx, N., 2003. Complaints of depression in a representative sample of the Belgian population. Archives of Public Health 61, 223–247. Brady, J.E., Weitzman, B.C., 2007. Inconsistencies in place definition: how different operational place definitions affect estimates of adolescent smoking and drinking risk. Health & Place 13, 562–568. Brenner, M.H., 1987. Economic instability, unemployment rates, behavioral risks, and mortality rates in Scotland, 1952–1982. International Journal of Health Services 17 (3), 475–487. Bu¨hler, A., Kraus, L., Augustin, R., Kramer, S., 2004. Screening for alcohol-related problems in the general population using CAGE and DSM-IV: characteristics of congruently and incongruently identified participants. Addictive Behaviors 29 (5), 867–878. Catalano, R., 1991. The health effects of economic insecurity. American Journal of Public Health 81 (9), 1148–1152. Catalano, R., Dooley, D., 1977. Economic predictors of depressed mood and stressful life events in a metropolitan community. Health and Social Behavior 18, 292–307. Chandola, T., 2001. The fear of crime and area differences in health. Health & Place 7 (2), 105–116. Cherpitel, C.J., 1998. Performance of screening instruments for identifying alcohol dependence in the general population, compared with clinical populations. Alcoholism: Clinical and Experimental Research 22 (7), 1399–1404. Cherpitel, C.J., 2002. Screening for alcohol problems in the US general population: comparison of the CAGE, RAPS4, and RAPS4-QF by gender, ethnicity, and service utilization. Alcoholism: Clinical and Experimental Research 26 (11), 1686–1691. Clark, A.E., 2003. Unemployment as a social norm: psychological evidence from panel data. Journal of Labor Economics 21 (2), 323–351. Cummins, S., MacIntyre, S., Davidson, S., Ellaway, A., 2005a. Measuring neighbourhood social and material context: generation and interpretation of ecological data from routine and non-routine sources. Health & Place 11, 249–260.

Cummins, S., Stafford, M., MacIntyre, S., Marmot, M., Ellawy, A., 2005b. Neighbourhood environment and its association with self rated health: evidence from Scotland and England. Journal of Epidemiology and Community Health 59, 207–213. Dalgard, O.S., Tambs, K., 1997. Urban environment and mental health. British Journal of Psychiatry 171, 530–536. Day, A.L., Livingstone, H.A., 2003. Gender differences in perceptions of stressors and utilization of social support among university students. Canadian Journal of Behavioural Science-Revue 35 (2), 73–83. Derogatis, L.R., 1977. SCL-90-R, Administration, Scoring and Procedures Manual II. National Computer Systems, Minneapolis, MN. De Silva, M.J., McKenzie, K., Harpham, T., Huttly, S.R., 2005. Social capital and mental illness: a systematic review. Journal of Epidemiology and Community Health 59, 619–627. De Witte, H., 1999. Job insecurity and psychological well-being: review of the literature and exploration of some unresolved issues. European Journal of Work and Organizational Psychology 8 (2), 155–177. Dhalla, S., Kopec, J.A., 2007. The CAGE questionnaire for alcohol misuse: a review of reliability and validity studies. Clinical and Investigative Medicine 30 (1), 33–41. Dietz, R.D., 2002. The estimation of neighborhood effects in the social sciences: an interdisciplinary approach. Social Science Research 31, 539–575. Diez-Roux, A.V., 2004. Estimating neighborhood health effects: the challenges of causal inference in a complex world. Social Science & Medicine 58, 1953–1960. Diez-Roux, A.V., Nieto, F.J., Muntaner, C., Tyroler, H.A., Comstock, G.W., Shahar, E., Cooper, L.S., Watson, R.L., Szklo, M., 1997. Neighborhood environments and coronary heart disease: a multilevel analysis. American Journal of Epidemiology 146 (1), 48–63. Dohrenwend, B.P., Dohrenwend, B.S., 1976. Sex differences and psychiatric disorders. American Journal of Sociology 81 (6), 1447–1454. Driessen, G., Gunther, N., van Os, J., 1998. Shared social environment and psychiatric disorder: a multilevel analysis of individual and ecological effects. Social Psychiatry and Psychiatric Epidemiology 33 (12), 606–612. Duncan, C., Jones, K., Moon, G., 1995. Psychiatric morbidity—a multilevel approach to regional variations in the UK. Journal of Epidemiology and Community Health 49 (3), 290–295. Duncan, C., Jones, K., Moon, G., 1998. Context, composition and heterogeneity: using multilevel models in health research. Social Science & Medicine 46 (1), 97–117. Ecob, R., MacIntyre, S., 2000. Small area variations in health related behaviours: do these depend on the behaviour itself, its measurement, or on personal characteristics? Health & Place 6, 261–274. Ewing, J.A., 1984. Detecting alcoholism: the CAGE questionnaire. Journal of the American Medical Association 252, 1905–1907. Ferlander, S., 2007. The importance of different forms of social capital for health. Acta Sociologica 50 (2), 115–128. Floro, M.S., Miles, M., 2003. Time use, work and overlapping activities: evidence from Australia. Cambridge Journal of Economics 27 (6), 881–904. Fone, D., Dunstan, F., 2006. Getting strategic about the environment and health. Public Health 120 (10), 904–906. Fone, D., Dunstan, F., Williams, G., Lloyd, K., Palmer, S., 2007. Places, people and mental health: a multilevel analysis of economic inactivity. Social Science & Medicine 64, 633–645. Forcier, M.W., 1985. Labor force behavior of alcoholics: a review. International Journal of the Addictions 20 (22), 253–268. Forcier, M.W., 1988. Unemployment and alcohol abuse: a review. Journal of Occupational Medicine 30, 246–251. Galea, S., Ahern, J., Tracy, M., Vlahov, D., 2007. Neighborhood income and income distribution and the use of cigarettes, alcohol, and marijuana. American Journal of Preventive Medicine 32 (6), S195–S202. Gary, T.L., Stark, S.A., LaVeist, T.A., 2007. Neighborhood characteristics and mental health among African Americans and whites living in a racially integrated urban community. Health & Place 13, 569–575. Goldsmith, H.F., Holzer, C.E., Manderscheid, R.W., 1998. Neighborhood characteristics and mental illness. Evaluation and Programming Planning 21, 211–225. Greiner, K.A., Li, C., Kawachi, I., Hunt, D.C., Ahluwalia, J.S., 2004. The relationships of social participation and community ratings to health and health behaviors in areas with high and low population density. Social Science & Medicine 59 (11), 2303–23212. Gutie´rrez-Lobos, K., Wo¨lfl, G., Scherer, M., Anderer, P., Schmidl-Mohl, B., 2000. The gender gap in depression reconsidered: the influence of marital and employment status on the female/male ratio of treated incidence rates. Social Psychiatry and Psychiatric Epidemiology 35, 202–210. Hamilton, V., Broman, C., Hoffman, W., Renner, D., 1990. Hard times and vulnerable people: initial effects of plant closing on autoworkers’ mental health. Journal of Health and Social Behavior 31, 123–140. Hamilton, V.H., Merrigan, P., Dufresne, E., 1997. Down and out: estimating the relationship between mental health and unemployment. Health Economics 6, 397–406. Hill, T.D., Angel, R.J., 2005. Neighborhood disorder, psychological distress, and heavy drinking. Social Science & Medicine 61, 965–975. Hox, J., 2002. Multilevel Analysis. Techniques and Applications. Lawrence Erlbaum Associates, London. Huisman, M., 1999. Item Non Response: Occurrence, Causes, and Imputation of Missing Answers to Test Items. DSWO Press, Leiden.

ARTICLE IN PRESS L. van Praag et al. / Health & Place 15 (2009) 990–998

Hunt, K., Annandale, F., 1993. Just the job? Is the relationship between health and domestic and paid work gender-specific? Sociology of Health and Illness 15 (5), 632–664. Ingoldsby, E.M., Shaw, D.S., 2002. Neighborhood contextual factors and earlystarting antisocial pathways. Clinical Child and Family Psychology Review 5 (1), 21–55. Karvonen, S., Rimpela¨, A.H., 1996. Socio-regional context as a determinant of adolescents’ health behaviour in Finland. Social Science & Medicine 43 (10), 1467–1474. Karvonen, S., Rimpela¨, A.H., 1997. Urban small area variation in adolescents’ health behaviour. Social Science & Medicine 45 (7), 1089–1098. Kasl, S.V., Jones, B.A., 2000. The impact of job loss and retirement on health. In: Berkman, L.F., Kawachi, I. (Eds.), Social Epidemiology. Oxford University Press, Oxford, pp. 118–136. Kavanagh, A.M., Bentley, R., Turrell, G., Broom, D.H., Subramanian, S.V., 2006. Does gender modify associations between self rated health and the social and economic characteristics of local environments? Journal of Epidemiology and Community Health 60, 490–495. Kavanagh, A.M., Goller, J.L., King, T., Jolley, D., Crawford, D., Turrell, G., 2005. Urban area disadvantage and physical activity: a multilevel study in Melbourne, Australia. Journal of Epidemiology and Community Health 59 (11), 934–940. Kawachi, I., Berkman, L.F., 2001. Social ties and mental health. Journal of Urban Health 78 (3), 458. Kawachi, I., Kennedy, B.P., Glass, R., 1999. Lack of social capital and self-rated health: a contextual analysis. American Journal of Public Health 89 (8), 1187–1193. Kelleher, K.J., Rickert, V.I., Hardin, B.H., Pope, S.K., Farmer, F.L., 1992. Rurality and gender: effects on early adolescent alcohol use. Archives of Pediatrics & Adolescent Medicine 146 (3), 317–322. Kendler, K.S., Prescott, C.A., Myers, J., Neale, M.C., 2003. The structure of genetic and environmental risk factors for common psychiatric and substance use disorders in men and women. Archives of General Psychiatry 60 (9), 929–937. Krueger, R.F., McGue, M., Iacono, W.G., 2001. The higher-order structure of common DSM mental disorders: internalization, externalization and their connections to personality. Personality and Individual Differences 30 (7), 1245–1259. Kulik, L., 2000. Jobless men and women: a comparative analysis of job search intensity, attitudes toward unemployment and related responses. Journal of Occupational and Organizational Psychology 73 (4), 487–500. Lahelma, E., Kangas, R., Manderbacka, K., 1995. Drinking and unemployment: contrasting patterns among men and women. Drug and Alcohol Dependence 37 (1), 71–82. LeClere, F.B., Rogers, R.G., Peters, K.D., 1997. Ethnicity and mortality in the United States: individual and community correlates. Social Forces 76 (1), 169–198. Le´pine, J.-P., Gastpar, M., Mendlewicz, J., Tylee, A., 1997. Depression in the community: the first pan-European study DEPRES (Depression Research in European Society). International Clinical Psychopharmacology 12, 19–29. Levecque, K., 2006. Generalized anxiety and depression in the general population: risk factors according to the Belgian health interview survey 2001. Depression and Anxiety 23, 509–511. Levecque, K., Lodewyckx, I., Bracke, P., 2008. Psychological distress, depression and generalized anxiety in Turkish and Moroccan immigrants in Belgium. Social Psychiatry and Psychiatric Epidemiology 44, 188–197. Levecque, K., Lodewyckx, I., Vranken, J., 2007. Depression and generalised anxiety in the general population in Belgium: A comparison between native and immigrant groups. Journal of Affective Disorders 97 (1), 229–239. Lewis, G., Bebbington, T., Brugha, M., 2003. Socio-economic status, standard of living and neurotic disorder. International Review of Psychiatry 15, 91–96. Lin, N., Ensel, W.M., 1989. Life stress and health: stressors and resources. American Sociological Review 54 (3), 382–399. Lomas, J., 1998. Social capital and health: implications for public health and epidemiology. Social Science & Medicine 47 (9), 1181–1188. Lorant, V., Croux, C., Weich, S., Delie`ge, D., Mackenbach, J., Ansseau, M., 2007. Depression and socio-economic risk factors: 7-year longitudinal population study. British Journal of Psychiatry 190, 293–298. Lorant, V., Thomas, I., Delie`ge, D., Tonglet, R., 2001. Deprivation and mortality: the implications of spatial autocorrelation for health resources allocation. Social Science & Medicine 53, 1711–1719. Lorant, V., Van Oyen, H., Thomas, I., 2008. Contextual factors and immigrants’ health status: double jeopardy. Health & Place 14 (4), 678–692. Lynch, J., Smith, G.D., Harper, S., Hillemeier, M., Ross, N., 2004. Is income inequality a determinant of health? Part 1. A systematic review. Milbank Quarterly 82 (1), 5–99. MacIntyre, S., Ellaway, A., 2000. Ecological approaches: rediscovering the role of the physical and social environment. In: Berkman, L.F., Kawachi, I. (Eds.), Social Epidemiology. Oxford University Press, Oxford, pp. 332–348. MacIntyre, S., Ellaway, A., 2003. Neighborhoods and health: an overview. In: Kawachi, I., Berkman, L.F. (Eds.), Neighborhoods and Health. Oxford University Press, Oxford. MacIntyre, S., Ellaway, A., Cummins, S., 2002. Place effects on health: how can we conceptualise, operationalise, and measure them. Social Science & Medicine 55 (1), 125–139. Matheson, F.I., Moineddin, R., Dunn, J.R., Creatore, M.I., Gozdyra, P., Glazier, R.H., 2006. Urban neighborhoods, chronic stress, gender and depression. Social Science & Medicine 63 (10), 2604–2616.

997

McCulloch, A., 2001. Ward-level deprivation and individual social and economic outcomes in the British Household Panel Study. Environment and Planning 33, 667–684. Molinari, C., Ahern, M., Hendryx, M., 1998. The relationship of community quality to the health of women and men. Social Science & Medicine 47 (8), 1113–1120. Monden, C.W.S., van Lenthe, F.J., Mackenbach, J.P., 2006. A simultaneous analysis of neighbourhood and childhood socio-economic environment with self-assessed health and health-related behaviours. Health & Place 12, 394–403. Moore, D.S., McCabe, G.P., 1999. Statistiek in de praktijk. Academic Service, Den Haag. Moss, N.E., 2002. Gender equity and socioeconomic inequality: a framework for the patterning of women’s health. Social Science & Medicine 54 (5), 649–661. Netemeyer, R.G., Baerden, W.O., Sharma, S., 2003. Scaling Procedures. Sage, Thousand Oaks. Nolen-Hoeksema, S., 2002. Gender differences in depression. In: Gotlib, I.H., Hammen, C.L. (Eds.), Handbook of Depression, second ed. Guilford Press, New York, pp. 386–404. Novo, M., Hammarstro¨m, A., Janlert, U., 2001. Do high levels of unemployment influence the health of those who are not employed? A gendered comparison of young men and women during boom and recession. Social Science & Medicine 53, 293–303. Parker, D.A., Parker, E.S., Harford, T.C., Farmer, G.C., 1987. Alcohol use and depression symptoms among employed men and women. American Journal of Public Health 77 (6), 704–707. Paul, K.I., 2005. The negative mental health effect of unemployment: meta-analyses of cross-sectional longitudinal data. (online) Available at: http://www.opus.ub. uni-erlangen.de/opus/volltexte/2006/473/pdf/KarstenIngmarPaulDissertation.pdf (Accessed: 11/02/2009). Pearlin, L.I., 1989. The sociological study of stress. Journal of Health and Social Behavior 30, 241–256. Peterson, L.E., Tsai, A.C., Petterson, S., Litaker, D.G., 2009. Rural–urban comparison of contextual associations with self-reported mental health status. Health & Place 15 (1), 125–132. Piccinelli, M., Wilkinson, G., 2000. Gender differences in depression. Critical review. British Journal of Psychiatry 177, 486–492. Pickett, K.E., Pearl, M., 2001. Multilevel analyses of neighbourhood socioeconomic context and health outcomes: a critical review. Journal of Epidemiology and Community Health 55, 111–122. Poortinga, W., Dunstan, F., Fone, D., 2007. Perceptions of the neighbourhood environment and self rated health: a multilevel analysis of the Caerphilly Health and Social Needs Study. BMC Public Health 7, 284. Prescott, C.A., Aggen, S.H., Kendler, K.S., 1999. Sex differences in the sources of genetic liability to alcohol abuse and dependence in a populationbased sample of US twins. Clinical and Experimental Research 23 (7), 1136. Pritchard, C., Evans, B., 1997. Population density and cancer mortality by gender and age in England and Wales and the Western World 1963–93. Public Health 111 (4), 215–220. Propper, C., Jones, K., Bolster, A., Burgess, S., Johnston, R., Sarker, R., 2005. Local neighbourhood and mental health: evidence from the UK. Social Science & Medicine 61, 2065–2083. Raleigh, V.S., Kiri, V.A., 1997. Life expectancy in England: variations and trends by gender, health authority, and level of deprivation. Journal of Epidemiology and Community Health 51, 649–658. Robert, S.A., 1999. Socioeconomic position and health: the independent contribution of community socioeconomic context. Annual Review of Sociology 25 (1), 489–516. Rosenfield, S., Vertefuille, J., McAlpine, D.D., 2000. Gender stratification and mental health: an exploration of dimensions of the self. Social Psychology Quarterly 63 (3), 208–223. Ross, C.E., Mirowsky, J., 1999. Refining the association between education and health: the effects of quantity, credential, and selectivity. Demography 36 (4), 445–460. Ross, C.E., Mirowsky, J., 2002. Age and the gender gap in the sense of personal control. Social Psychology Quarterly 65 (2), 125–145. Ross, C.E., Mirowsky, J., 2006. Sex differences in the effect of education on depression: resource multiplication or resource substitution? Social Science & Medicine 63 (5), 1400–1413. Ross, C.E., Van Willigen, M., 1997. Education and the subjective quality of life. Journal of Health and Social Behavior 38 (3), 275–297. Scientific Institute of Public Health (SIPH), 2006. Health Interview Survey Belgium Manual 2004. SIPH, Unit of Epidemiology, Brussels. Sherbourne, C.D., Stewart, A.L., 1991. The MOS Social Support Survey. Social Science & Medicine 32, 705–714. Silver, E., 2000. Extending social disorganization theory: a multilevel approach to the study of violence among persons with mental illnesses. Criminology 38 (4), 1043–1074. Silver, E., Mulvey, E.P., Swanson, J.W., 2002. Neighborhood structural characteristics and mental disorder: Faris and Dunham revisited. Social Science & Medicine 55 (8), 1457–1470. Skrabski, A., Kopp, M., Kawachi, I., 2003. Social capital in a changing society: cross sectional associations with middle aged female and male mortality. Journal of Epidemiology and Community Health 57, 114–119. Smart, R.G., Adlaf, E.M., Knoke, D., 1991. Use of the CAGE scale in a population survey of drinking. Journal of Studies on Alcohol 52 (6), 593–596.

ARTICLE IN PRESS 998

L. van Praag et al. / Health & Place 15 (2009) 990–998

Stafford, M., Cummins, S., MacIntyre, S., Ellaway, A., Marmot, M., 2005. Gender differences in the associations between health and neighbourhood environment. Social Science & Medicine 60, 1681–1692. Stafford, M., Marmot, M., 2003. Neighbourhood deprivation and health: does it affect us all equally? International Epidemiological Association 32, 357–366. Statistics Belgium. Nationaal Instituut voor Statistiek. (Online) Available at: http:// statbel.fgov.be. 2008 (Accessed 1 April 2008). Stockdale, S.E., Wells, K.B., Tang, L., Belin, T.R., Zhang, L., Sherbourne, C.D., 2007. The importance of social context: neighborhood stressors, stress-buffering mechanisms, and alcohol, drug, and mental health disorders. Social Science & Medicine 65, 1867–1881. Subramanian, S.V., Kawachi, I., 2004. Income inequality and health: what have we learned so far? Epidemiologic Reviews 26 (1), 78–91. Sundquist, K., Frank, G., Sundquist, J., 2004. Urbanisation and incidence of psychosis and depression. A follow-up study of 4.4 million women and men in Sweden. British Journal of Psychiatry 184, 293–298. Taylor, S.E., Repetti, R.L., Seeman, T., 1997. Health psychology: what is an unhealthy environment and how does it get under the skin? Annual Review of Psychology 48, 411–447. Thoits, P.A., 1995. Stress, coping, and social support processes – where are we – what next. Journal of Health and Social Behavior, 53–79. Trickett, P.K., McBride-Chang, C., 1995. The developmental impact of different forms of child abuse and neglect. Developmental Review 15 (3), 311–337. Tunstall, H.V.Z., Shaw, M., Dorling, D., 2004. Places and health. Journal of Epidemiology and Community Health 58 (1), 6–10. Turner, R.J., Wheaton, B., Lloyd, D.A., 1995. The epidemiology of social stress. American Sociological Review 60 (1), 104–125.

van Lenthe, F.J., Borrell, L.N., Costa, G., Diez-Roux, A.V., Kauppinen, T.M., Marinacci, C., Martikainen, P., Regidor, E., Stafford, M., Valkonen, T., 2005. Neighbourhood unemployment and all cause mortality: a comparison of six countries. Journal of Epidemiology and Community Health 59, 231–237. van Lenthe, F.J., Mackenbach, J.P., 2002. Neighbourhood deprivation and overweight: the GLOBE study. International Journal of Obesity 26, 234–240. Verheij, R.A., 1996. Explaining urban-rural variations in health a review of interactions between individual and environment. Social Sciences and Medicine 42 (6), 923–935. Waitzman, N.J., Smith, K.R., 1998. Phantom of the area: poverty-area residence and mortality in the United States. American Journal of Public Health 88 (6), 973–976. Weich, S., 2005. Absence of spatial variation in rates of the common mental disorders. Journal of Epidemiology and Community Health 59, 254–257. Weich, S., Twigg, L., Holt, G., Lewis, G., Jones, K., 2003. Contextual risk factors for the common mental disorders in Britain: a multilevel investigation of the effects of place. Journal of Epidemiology and Community Health 57, 616–621. Weich, S., Twigg, L., Lewis, G., 2006. Rural/non-rural differences in rates of common mental disorders in Britain. Prospective multilevel cohort study. British Journal of Psychiatry 188, 51–57. Weissman, M.M., 1987. Advances in psychiatric epidemiology: rates and risks for major depression. American Journal of Public Health 77 (4), 445–451. Wohlgemuth, E., Betz, N.E., 1991. Gender as a moderator of the relationships of stress and social support to physical health in college-students. Journal of Counselling Psychology 38 (3), 367–374. Yen, I.H., Syme, S.L., 1999. The social environment and health: a discussion of the epidemiological literature. Annual Review of Public Health 20, 287–308.