Skip to main content
Erschienen in: BMC Public Health 1/2019

Open Access 01.12.2019 | Research article

Brazilian Longitudinal Study of Adult Health (ELSA-Brasil): socio-occupational class as an effect modifier for the relationship between adiposity measures and self-rated health

verfasst von: Thaís Lopes de Oliveira, Rosane Harter Griep, Joanna Nery Guimarães, Luana Giatti, Dóra Chor, Maria de Jesus Mendes da Fonseca

Erschienen in: BMC Public Health | Ausgabe 1/2019

Abstract

Background

Little is known about the role of social class in the association between adiposity measures and self-rated health, and several studies have evaluated its influence as a confounder. The aim of the study is to investigate whether social class is an effect modifier in the association between adiposity measures and self-rated health in participants in the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil).

Method

Cross-sectional design, including 6453 men and 7686 women. Body mass index (kg/m2) and waist circumference (cms) were assessed. Self-rated health was categorized as good, fair and poor. Socio-occupational class was based on the participants’ occupation, education and per capita income. Multicovariate ordinal logistic model was used to evaluate the association between adiposity measures and self-rated health.

Results

For women, the low and medium socio-occupational class effects were higher for those with waist circumference between 80 and 88 cm or overweight. For men, the low and medium socio-occupational class effects were higher for those with adequate waist circumference or normal body mass index.

Conclusions

Social class is an effect modifier in the association between body mass index or waist circumference and self-rated health.
Hinweise

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Abkürzungen
BMI
Body mass index
DAG
Directed acyclic graph
ELSA-Brasil
Brazilian Longitudinal Study of Adult Health
OR
Odds Ratios
WC
Waist circumference

Background

Self-rated health is considered one of the most relevant indicators for health research and It has been widely used as an indicator of health conditions in different populations [1, 2]. Several studies have demonstrated that self-rated health is a good predictor of morbidity and mortality, even after controlling for risk factors such as gender, race, marital status, and education [1, 35]. In addition to these associations, some factors are predominant in determining poor self-rated health. Among them are advanced age, female sex, low income, low educational level, unemployment, low social class, being married, having low social capital and being obese [2, 3, 6].
Obesity is a public health concern in Brazil; which prevalence has been increasing over the years [7]. Some research has shown that high values of body mass index (BMI) and waist circumference (WC) are associated with poor self-rated health [810]. Moreover, social stratification has been considered a potential explanation for differences in health [11].
Social class is a determinant of health [12], and being in a more distal level has an influence on body mass index and on the perception of health. Some studies have shown that the prevalence of inadequate BMI varies according to social class, with an excess of weight being more prevalent in lower social classes [13, 14]. Other studies have noted that people of lower social class rate their own health worse [6, 14, 15].
Due to the importance of social class in BMI and self-rated health, some authors treat the social class as a confounder variable in the relationship between adiposity measures and self-rated health [3, 16]. However, it is possible that this relationship is not homogeneous in all strata of social class [6, 11, 15], suggesting that this variable may act as an effect modifier.
An effect modification, also known as an interaction, can be described as “a situation in which two or more risk factors modify the effect of each other with regard to the occurrence or level of a given outcome” [17]. The investigation of the presence of interactions has important implications for public health, including important implications for prevention, for the planning of intervention and to identify the most vulnerable population groups [17, 18]. Nevertheless, the interaction is a phenomenon not often explored in the epidemiological literature [19].
Moreover, some studies have investigated interactions involving self-rated health as the main outcome [2023], and, thus far, we have not found studies that evaluated the interaction between social class and adiposity measures in the association with self-rated health. In this way, the aim of this study was to investigate whether the social class is an effect modifier in the association between adiposity measures and self-rated health in participants from the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil).

Methods

Study population

This study used baseline data (2008–2010) from ELSA-Brasil. The ELSA-Brasil is a multicentric cohort of civil servants (35–74 years) conducted at six study research centres in three regions of the country, including the Northeast, South, and Southeast. These centres are located at five federal universities and the Oswaldo Cruz Foundation [24]. The present study included 6453 men and 7686 women; participants who did not have information about weight, height or waist circumference measurements (n = 7) or did not answer the question about self-rated health (n = 4), socio-occupational class (n = 239) or other variables of interest to the study (n = 1) were excluded.

Exposure

The exposures were body mass index (kg/m2) and waist circumference (cms). A stadiometer with a 0.1 cm scale was used to measure height. Weight was measured with the participant wearing a standardized uniform using a calibrated electronic scale with a capacity of 0 to 200 kg and divisions of 50 g. Waist circumference was measured using a measuring tape at the largest abdominal perimeter between the iliac crest and the last rib [25]. All measurements were performed using standardized techniques [26].
The following cut-off points adopted for BMI classification followed the recommendations of the World Health Organization [27]: ≤ 24.9 kg/m2 for underweight and normal weight, between 25 and 29.9 kg/m2 for overweight and ≥ 30 kg/m2 for obesity. The categories of underweight (≤ 18.5 kg/m2) and normal weight were grouped due to the small number of participants who were underweight (< 1%).
The WC cut-off points adopted for women were as follows: ≥ 80 cm and < 88 cm for increased risk of metabolic diseases and ≥ 88 cm for substantially increased risk. For males, the cut-off points were as follows: ≥ 94 cm and < 102 cm for increased risk of metabolic diseases and ≥ 102 cm for substantially increased risk [28].

Outcome

Self-rated health [5] was obtained using the following question: “In general, compared to people of your age, how do you consider your state of health?”; the options were “very good, good, fair, poor, or very poor”. For the analyses, the answers were categorized as good self-rated health (very good and good), fair and poor self-rated health (poor and very poor).

Effect modifier

Socio-occupational class was used as an indicator of social class. This variable was obtained from the participants’ socioeconomic status based on their described occupation, expected income (based on the level of education - average market value) and observed income [29]. Socioeconomic status was calculated as an average between observed income (economic component) and expected income (educational component). Subsequently, occupational socioeconomic status was calculated for each occupational title as the average score of the socioeconomic status of the individuals with different occupations (Fig. 1). Seven socio-occupational strata were defined based on the occupational socioeconomic status scores. These strata were divided into seven levels (upper-upper, upper-lower, upper-middle, middle-middle, middle-lower, lower-upper, and lower-lower) and ordered according to educational level and income associated with occupations. In this study, the strata were categorized as follows: high socio-occupational class (upper-upper and upper-lower), medium (upper-middle, middle-middle, middle-lower) and low (lower-upper and lower-lower).

Covariables

Directed acyclic graph (DAG) were constructed to represent the theoretical-operational model and to understand the involvement of covariables in the relationship between adiposity measures and self-rated health. The DAG was constructed using the Dagitty tool (available at: http://​www.​dagitty.​net/​dags.​html). Additionally, the DAG was used to identify the minimum set of potential confounding variables. The covariates used in the DAG were age, self-declared colour or race, education, family income per capita, functional level, socio-occupational class, and marital status. The minimum set of potential confounding variables to explain the relationship between adiposity measures and self-rated health were socio-occupational class (low, medium and high), self-declared colour or race (white, brown and black), age and marital status (married/united, single, and separated/divorced/widowed/other) (Fig. 2).

Statistical analyses

Means and standard deviations, and proportions were used to describe population characteristics regarding self-rated health. Through DAG, the selected variables were analysed using multicovariate models. We estimated the crude and adjusted association measurements (Odds Ratios - OR), obtained by single and multicovariate ordinal logistic model, respectively. This type of modelling tested the multiplicative interactions by the inclusion of an interaction term in the full adjusted model (BMI or WC*socio-occupational class). The effects of interactions were also illustrated.
All analyses were stratified by sex, and 95% confidence intervals were considered. The analyses were performed in the R software, version 3.5.1, library “MASS”, “epiDisplay”, “VGAM” and “effects”. The ELSA-Brasil study was approved by the research ethics committees of each of the institutions involved and all participants signed informed consent forms.

Results

The mean age was similar for both sexes. The proportions showed that women with poor self-rated health were more likely to be married/united, brown self-declared colour, medium socio-occupational class, to have obesity and waist circumference above 88 cm. Men with poor self-rated health were more likely to be married/united, white self-declared colour, low socio-occupational class, overweight and waist circumference above 102 cm (Table 1).
Table 1
Characteristics of population regarding self-rated health - ELSA-Brasil baseline (2008–10)
 
Self-rated health
Good
Fair
Poor
Female sex
 Agea
51.4 (8.8)
53.7 (8.9)
53.9 (8.6)
 Self-declared colourb
  Black
1023 (16.6)
375 (27.8)
47 (27)
  Brown
1617 (26.2)
468 (34.6)
64 (36.8)
  White
3521 (57.1)
508 (37.6)
63 (36.2)
 Marital statusb
  Married/united
3328 (54)
671 (49.7)
81 (46.6)
  Separated/divorced/widowed/other
1924 (31.2)
522 (38.6)
74 (42.5)
  Single
909 (14.8)
158 (11.7)
19 (10.9)
 Socio-occupational classb
  Low
1117 (18.1)
459 (34)
61 (35.1)
  Medium
2955 (48)
669 (49.5)
87 (50)
  High
2089 (33.9)
223 (16.5)
26 (14.9)
 BMIb
  Obesity
1309 (21.2)
517 (38.3)
87 (50)
  Overweight
2264 (36.7)
485 (35.9)
43 (24.7)
  Normal weight
2588 (42)
349 (25.8)
44 (25.3)
 Waist circumferenceb
   ≥ 88 cm
2502 (40.6)
853 (63.1)
121 (69.5)
   ≥ 80 cm and < 88 cm
1722 [28]
291 (21.5)
32 (18.4)
  Adequate
1937 (31.4)
207 (15.3)
21 (12.1)
Male sex
 Agea
51.6 (9.3)
53.9 (8.8)
54.7 (9.7)
 Self-declared colourb
  Black
691 (13.3)
219 (18.8)
21 (21.2)
  Brown
1547 (29.8)
420 (36.1)
35 (35.4)
  White
2953 (56.9)
524 (45.1)
43 (43.4)
 Marital statusb
  Married/united
4249 (81.9)
970 (83.4)
76 (76.8)
  Separated/divorced/widowed/other
653 (12.6)
146 (12.6)
21 (21.2)
  Single
289 (5.6)
47 (4)
2 (2)
 Socio-occupational classb
  Low
1312 (25.3)
494 (42.5)
46 (46.5)
  Medium
1823 (35.1)
409 (35.2)
31 (31.3)
  High
2056 (39.6)
260 (22.4)
22 (22.2)
 BMIb
  Obesity
955 (18.4)
347 (29.8)
37 (37.4)
  Overweight
2356 (45.4)
512 (44)
38 (38.4)
  Normal weight
1880 (36.2)
304 (26.1)
24 (24.2)
 Waist circumferenceb
   ≥ 102 cm
1233 (23.8)
435 (37.4)
43 (43.4)
   ≥ 94 cm and < 102 cm
1385 (26.7)
305 (26.2)
23 (23.2)
  Adequate
2573 (49.6)
423 (36.4)
33 (33.3)
amean (standard deviation); b n (%)
Tables 2 and 3 show the crude and adjusted ORs (obtained by single and multicovariate models, respectively) treating the socio-occupational class as an effect modifier (adjusted OR). In crude analyses, individuals with overweight, obesity or higher waist circumference were more likely to report worse health than individuals with adequate BMI and WC.
Table 2
Multicovariate ordinal logistic model of the association between waist circumference and self-rated health - ELSA-Brasil (2008–10)
 
Crude OR
Adjusted OR
Female sex
 Self-declared colour (white = 1)
  Brown
2.03 (1.78–2.31)
1.70 (1.48–1.95)
  Black
2.53 (2.19–2.92)
1.89 (1.63–2.20)
 Age (continuous variable)
1.02 (1.01–1.03)
1.02 (1.01–1.03)
 Marital status (single = 1)
  Married/united
1.16 (0.97–1.39)
1.24 (1.03–1.50)
  Separated/divorced/widowed/other
1.59 (1.33–1.92)
1.37 (1.13–1.67)
 Waist circumference (adequate = 1)
   ≥ 80 cm and < 88 cm
1.59 (1.33–1.91)
   ≥ 88 cm
3.31 (2.84–3.88)
 Socio-occupational class (high = 1)
  Medium
2.15 (1.84–2.51)
  Low
3.89 (3.29–4.60)
 Interactions*
  Adequate WC* high socio-occupational class
1
   ≥ 80 cm WC* high socio-occupational class
0.95 (0.62–1.46)
   ≥ 88 cm WC* high socio-occupational class
2.51 (1.81–3.53)
  Adequate WC* medium socio-occupational class
1.68 (1.19–2.39)
   ≥ 80 cm WC* medium socio-occupational class
2.62 (1.84–3.80)
   ≥ 88 cm WC* medium socio-occupational class
1.86 (1.52–2.29)
  Adequate WC* low socio-occupational class
2.77 (1.86–4.16)
   ≥ 80 cm WC* low socio-occupational class
4.07 (2.79–6.04)
   ≥ 88 cm WC* low socio-occupational class
2.62 (2.10–3.28)
Male sex
 Self-declared colour (white = 1)
  Brown
1.53 (1.33–1.75)
1.30 (1.12–1.51)
  Black
1.81 (1.52–2.15)
1.38 (1.14–1.65)
 Age (continuous variable)
1.02 (1.02–1.04)
1.03 (1.02–1.04)
 Marital status (single = 1)
  Married/united
1.46 (1.08–2.01)
1.14 (0.83–1.58)
  Separated/divorced/widowed/other
1.53 (1.09–2.19)
1.31 (0.92–1.89)
 Waist circumference (adequate = 1)
   ≥ 94 cm and < 102 cm
1.34 (1.14–1.56)
   ≥ 102 cm
2.19 (1.89–2.53)
 Socio-occupational class (high = 1)
  Medium
1.76 (1.49–2.07)
  Low
2.99 (2.56–3.52)
 Interactions*
  Adequate WC* high socio-occupational class
1
   ≥ 94 cm WC* high socio-occupational class
1.59 (1.13–2.23)
   ≥ 102 cm WC* high socio-occupational class
2.91 (2.15–3.96)
  Adequate WC* medium socio-occupational class
2.20 (1.64–2.98)
   ≥ 94 cm WC* medium socio-occupational class
1.70 (1.22–2.37)
   ≥ 102 cm WC* medium socio-occupational class
1.96 (1.51–2.56)
  Adequate WC* low socio-occupational class
3.93 (2.96–5.27)
   ≥ 94 cm WC* low socio-occupational class
3.15 (2.30–4.34)
   ≥ 102 cm WC* low socio-occupational class
2.22 (1.69–2.94)
OR = Odds Ratio, WC = Waist circumference; Ordinal logistic model was a proportional odds regression to model self-rated health (good, fair, poor). Crude and adjusted ORs were obtained by single and multicovariate models; Adjusted model: waist circumference + socio-occupational class + self-declared colour or race + age + marital status + waist circumference*socio-occupational class; * Effect of socio-occupational class status considering waist circumference
Table 3
Multicovariate ordinal logistic model of the association between body mass index and self-rated health- ELSA-Brasil (2008–10)
 
Crude OR
Adjusted OR
Female sex
 Self-declared colour (white = 1)
  Brown
2.03 (1.78–2.31)
1.74 (1.51–1.99)
  Black
2.53 (2.19–2.92)
1.88 (1.61–2.19)
 Age (continuous variable)
1.02 (1.01–1.03)
1.03 (1.02–1.03)
 Marital status (single = 1)
  Married/united
1.16 (0.97–1.39)
1.28 (1.06–1.54)
  Separated/divorced/widowed/other
1.59 (1.33–1.92)
1.41 (1.16–1.71)
 Body mass index (normal = 1)
  Overweigh
1.59 (1.33–1.91)
  Obesity
3.31 (2.84–3.88)
 Socio-occupational class (high = 1)
  Medium social class
2.15 (1.84–2.51)
  Low social class
3.89 (3.29–4.60)
 Interactions*
  Normal*high socio-occupational class
1
  Overweigh*high socio-occupational class
1.01 (0.73–1.40)
  Obesity*high socio-occupational class
2.51 (1.81–3.46)
  Normal*medium socio-occupational class
1.75 (1.34–2.30)
  Overweigh*medium socio-occupational class
2.36 (1.79–3.13)
  Obesity*medium socio-occupational class
1.74 (1.32–2.29)
  Normal*low socio-occupational class
2.68 (1.97–3.67)
  Overweigh*low socio-occupational class
3.47 (2.59–4.69)
  Obesity*low socio-occupational class
2.49 (1.86–3.36)
Male sex
 Self-declared colour (white = 1)
  Brown
1.53 (1.33–1.75)
1.26 (1.09–1.47)
  Black
1.81 (1.52–2.15)
1.32 (1.09–1.58)
 Age (continuous variable)
1.02 (1.02–1.04)
1.04 (1.03–1.04)
 Marital status (single = 1)
  Married/united
1.46 (1.08–2.01)
1.13 (0.83–1.57)
  Separated/divorced/widowed/other
1.53 (1.09–2.19)
1.29 (0.91–1.86)
 Body mass index (normal = 1)
  Overweigh
1.34 (1.14–1.56)
  Obesity
2.19 (1.89–2.53)
 Socio-occupational class (high = 1)
  Medium social class
1.76 (1.49–2.07)
  Low social class
2.99 (2.56–3.52)
 Interactions*
  Normal*high socio-occupational class
1
  Overweigh*high socio-occupational class
1.72 (1.25–2.41)
  Obesity*high socio-occupational class
3.21 (2.26–4.59)
  Normal*medium socio-occupational class
2.35 (1.68–3.32)
  Overweigh*medium socio-occupational class
1.72 (1.33–2.22)
  Obesity*medium socio-occupational class
1.94 (1.43–2.64)
  Normal*low socio-occupational class
4.08 (2.95–5.73)
  Overweigh*low socio-occupational class
2.89 (2.26–3.70)
  Obesity*low socio-occupational class
2.15 (1.57–2.96)
OR = Odds Ratio, BMI = Body mass index; Ordinal logistic model was a proportional odds regression to model self-rated health (good, fair, poor). Crude and adjusted ORs were obtained by single and multicovariate models; Adjusted model: body mass index + socio-occupational class + self-declared colour or race + age + marital status + body mass index*socio-occupational class; * Effect of socio-occupational class status considering body mass index
In multicovariate model, the association measurements included the interaction term (BMI or WC*socio-occupational class). It is possible to notice, for both sexes, that interactions terms, for WC and BMI, had similar values and were significant (Tables 2 and 3, respectively). For women, the low and medium socio-occupational class effects were higher for those with WC between 80 and 88 cm (Table 2) or overweight (Table 3). For men, the low and medium socio-occupational class effects were higher for those with adequate WC (Table 3) or normal BMI (Table 2). For example, for women, the effect of the lower socio-occupational class, considering WC between 80 and 88 cm, in worsening self-rated health was 307% (OR = 4.07) higher than those of high socio-occupational class with adequate WC (Table 2). In addition, for men, the effect of the lower socio-occupational class, considering adequate WC, in worsening the self-rated health was 293% (OR = 3.93) higher than those of high socio-occupational class with adequate WC (Table 2).
In Figs. 3 and 4, the effects of the BMI and WC and socio-occupational class interactions, based on model 2, are shown. For both sexes, the probability of good self-rated health decrease with the increase of waist circumference, or body mass index, and with the decrease of socio-occupational class. Consequently, with the decrease in good self-rated health, the higher were the probabilities of fair or poor self-rated health.

Discussion

In the analyses, we observed that some characteristics that influence worse self-rated health were similar for men and women, such as BMI and WC, socio-occupational class, self-declared colour or race, and marital status. Others studies conducted on a general population and with a worker population also found similar results [1, 3, 6, 14, 15, 30]. Moreover, in this study, we found that socio-occupational class behaved as an effect modifier of the association between BMI or WC and self-rated health.
Our results show the presence of interactions with significant effects for men and women. These results demonstrate that individuals exposed to a low socio-occupational class and inadequate BMI or WC had greater chances of worse self-rated health. There are few studies investigating the presence of an interaction between exposures and self-rated health. Knol et al. [19] conducted a systematic review to examine how interactions were studied and reported results from cohort and case-control studies, including studies from 2001 to 2007. The authors demonstrated the small number of articles with this purpose and reported that the most frequent exposures were treatments, medical conditions, and lifestyle factors, and the most common outcomes were cardiovascular disease, cancer and all causes of mortality.
To date, we have not found studies that evaluated an interaction between social class and adiposity measures in the association with self-rated health. However, we found two studies that studied interactions involving socioeconomic factors and other variables, with self-rated health as the main outcome. Both found important results with significant interactions.
Ahnquist and collaborators [20] found an interaction between economic capital and social capital when studying poor self-rated health in Sweden. Their results show that individuals exposed to socioeconomic difficulties and low social capital are more likely to self-rate their health as poor. These results corroborate ours, although they do not use the same variable to test the interactions; both the results (by this study and ours) demonstrate the importance of social class and socioeconomic conditions when studying self-rated health. Both studies found interactions with effects involving socioeconomic issues, in which people of a lower social class, and consequently with less education and income, self-evaluate their health worse.
In Germany, Trachte et al. [23] studied the presence of an interaction between physical activity and socioeconomic status (measured by education and income) in relation to self-rated health. The authors found, for females, a significant synergistic interaction between education and physical activity. Women with higher levels of education and physical activity rate their own health better. Once more, these studies demonstrate the importance of verifying the influence of issues involving socioeconomic conditions and life habits, such as physical activity, on self-rated health. In this way, in addition to health promotion and nutritional education policies, policies to reduce social inequality and promote social advancement may have an important role in reducing the effect of social class in the association between adiposity measures and self-rated health.
One of the limitations of the present study is the cross-sectional design, as the variables used in this study were measured at the same moment in time during the baseline interview. However, the variables selected to compose the proposed DAG are ancestral variables of self-rated health, reinforcing that the sectional design, in this case, was adequate to study interactions that influence fair/poor self-rated health [31].
Another limitation is the generalization of our findings to the non-worker population, as our results are from a cohort of civil servants. Nevertheless, if an interaction between BMI or WC and the socio-occupational class was present in the ELSA population, which includes individuals with employment and income, perhaps this effect is even greater in the general population, which is composed of unemployed individuals with lower levels of education and income.
The large number of participants in the ELSA-Brasil study can be cited as an advantage of this study, given the need to have large sample sizes for the study of interactions [32, 33]. Kamangar [33] showed that a study with the aim of detecting interactions would require more than twice as many participants when compared to a study without this objective. Therefore, ELSA-Brasil had an adequate sample that allowed for the identification of variables that behave as effect modifiers.
Another advantage of this study was the use of two anthropometric measurements, BMI and WC. The association of different anthropometric methods assists in the nutritional diagnosis and can reduce the classification error associated with the use of just one anthropometric measurement [27]. Additionally, all equipment used was periodically checked; the scales were calibrated, the measuring tapes evaluated, and all interviewers were periodically recertified [34].

Conclusions

Therefore, the results show that the combined effects of social class and BMI or WC are more important than the independent effects of these factors on self-rated health. Our findings call attention to a more vulnerable population subgroup in relation to worse self-rated health, that is, those with overweight/obesity and low socioeconomic level. Nevertheless, these results can help in the formulation of public policies that involve adiposity measures, social class, social inequality, and other important issues when studying self-rated health.

Acknowledgements

We thank Raquel de Vasconcellos Carvalhaes de Oliveira for assistance with statistical analyses.
This study was submitted to the Ethics and Research Committee of the institution National School of Public Health, Oswaldo Cruz Foundation. Approval number: 1,630,132. All participants of the ELSA-Brasil study signed informed consent forms.
Not applicable.

Competing interests

The authors declare that they have no competing interests.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://​creativecommons.​org/​licenses/​by/​4.​0/​), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://​creativecommons.​org/​publicdomain/​zero/​1.​0/​) applies to the data made available in this article, unless otherwise stated.

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Literatur
1.
Zurück zum Zitat Barros MBDA, Zanchetta LM, ECD M, Malta DC. Auto-avaliação da saúde e fatores associados, Brasil, 2006. Rev Saude Publica. 2009;43(suppl 2):27–37.CrossRef Barros MBDA, Zanchetta LM, ECD M, Malta DC. Auto-avaliação da saúde e fatores associados, Brasil, 2006. Rev Saude Publica. 2009;43(suppl 2):27–37.CrossRef
2.
Zurück zum Zitat Hosseinpoor AR, Stewart Williams J, Amin A, Araujo de Carvalho I, Beard J, Boerma T, et al. Social determinants of self-reported health in women and men: understanding the role of gender in population health. PLoS ONE. 2012;7(4):e34799.CrossRef Hosseinpoor AR, Stewart Williams J, Amin A, Araujo de Carvalho I, Beard J, Boerma T, et al. Social determinants of self-reported health in women and men: understanding the role of gender in population health. PLoS ONE. 2012;7(4):e34799.CrossRef
4.
Zurück zum Zitat Idler EL, Benyamini Y. Self-rated health and mortality: a review of twenty-seven community studies. J Health Soc Behav. 1997;38(1):21.CrossRef Idler EL, Benyamini Y. Self-rated health and mortality: a review of twenty-seven community studies. J Health Soc Behav. 1997;38(1):21.CrossRef
5.
Zurück zum Zitat Schnittker J, Bacak V. The increasing predictive validity of self-rated health. Noymer a, editor. PLoS One. 2014;9(1):e84933.CrossRef Schnittker J, Bacak V. The increasing predictive validity of self-rated health. Noymer a, editor. PLoS One. 2014;9(1):e84933.CrossRef
6.
Zurück zum Zitat Farkas J, Pahor M, Zaletel-Kragelj L. Self-rated health in different social classes of Slovenian adult population: nationwide cross-sectional study. Int J Public Health. 2011;56(1):45–54.CrossRef Farkas J, Pahor M, Zaletel-Kragelj L. Self-rated health in different social classes of Slovenian adult population: nationwide cross-sectional study. Int J Public Health. 2011;56(1):45–54.CrossRef
7.
Zurück zum Zitat Brasil M da S. Vigitel Brasil 2017: vigilância de fatores de risco e proteção para doenças crônicas por inquérito telefônico: estimativas sobre frequência e distribuição sociodemográfica de fatores de risco e proteção para doenças crônicas nas capitais dos 26 estados brasileiros e no Distrito Federal em 2017. Secretaria de Vigilância em Saúde. Departamento de Vigilância de Doenças e Agravos não Transmissíveis e Promoção da Saúde. 2018. Brasil M da S. Vigitel Brasil 2017: vigilância de fatores de risco e proteção para doenças crônicas por inquérito telefônico: estimativas sobre frequência e distribuição sociodemográfica de fatores de risco e proteção para doenças crônicas nas capitais dos 26 estados brasileiros e no Distrito Federal em 2017. Secretaria de Vigilância em Saúde. Departamento de Vigilância de Doenças e Agravos não Transmissíveis e Promoção da Saúde. 2018.
8.
Zurück zum Zitat Tang K, Zhao Y, Li C. The association between self-rated health and different anthropometric and body composition measures in the Chinese population. BMC Public Health. 2017;17(1):317.CrossRef Tang K, Zhao Y, Li C. The association between self-rated health and different anthropometric and body composition measures in the Chinese population. BMC Public Health. 2017;17(1):317.CrossRef
9.
Zurück zum Zitat Cullinan J, Gillespie P. Does overweight and obesity impact on self-rated health? Evidence using instrumental variables ordered Probit models: the impact of overweight and obesity on self-rated health. Health Econ. 2016;25(10):1341–8.CrossRef Cullinan J, Gillespie P. Does overweight and obesity impact on self-rated health? Evidence using instrumental variables ordered Probit models: the impact of overweight and obesity on self-rated health. Health Econ. 2016;25(10):1341–8.CrossRef
10.
Zurück zum Zitat Souza LSPLD, Brunken GS, Segri NJ, Malta DC. Tendências da autoavaliação de saúde em relação ao excesso de peso na população adulta residente nas capitais do Centro-Oeste do Brasil. Rev Bras Epidemiol. 2017;20(2):299–309.CrossRef Souza LSPLD, Brunken GS, Segri NJ, Malta DC. Tendências da autoavaliação de saúde em relação ao excesso de peso na população adulta residente nas capitais do Centro-Oeste do Brasil. Rev Bras Epidemiol. 2017;20(2):299–309.CrossRef
11.
Zurück zum Zitat Stanwell-Smith R. Health inequality – is it all about social class? Perspect Public Health. 2014;134(6):306.CrossRef Stanwell-Smith R. Health inequality – is it all about social class? Perspect Public Health. 2014;134(6):306.CrossRef
12.
Zurück zum Zitat Buss PM, Pellegrini FA. A saúde e seus determinantes sociais. Physis: Revista de Saúde Coletiva. 2007;17(1):77–93.CrossRef Buss PM, Pellegrini FA. A saúde e seus determinantes sociais. Physis: Revista de Saúde Coletiva. 2007;17(1):77–93.CrossRef
13.
Zurück zum Zitat Bann D, Johnson W, Li L, Kuh D, Hardy R. Socioeconomic inequalities in body mass index across adulthood: coordinated analyses of individual participant data from three British birth cohort studies initiated in 1946, 1958 and 1970. PLoS Med. 2017;14(1):e1002214.CrossRef Bann D, Johnson W, Li L, Kuh D, Hardy R. Socioeconomic inequalities in body mass index across adulthood: coordinated analyses of individual participant data from three British birth cohort studies initiated in 1946, 1958 and 1970. PLoS Med. 2017;14(1):e1002214.CrossRef
15.
Zurück zum Zitat Rodrigues CG, Maia AG. Como a posição social influencia a auto-avaliação do estado de saúde? Uma análise comparativa entre 1998 e 2003. Cad Saude Publica. 2010;26(4):762–74.CrossRef Rodrigues CG, Maia AG. Como a posição social influencia a auto-avaliação do estado de saúde? Uma análise comparativa entre 1998 e 2003. Cad Saude Publica. 2010;26(4):762–74.CrossRef
16.
Zurück zum Zitat Granström F, Molarius A, Garvin P, Elo S, Feldman I, Kristenson M. Exploring trends in and determinants of educational inequalities in self-rated health. Scand J Public Health. 2015;43(7):677–86.CrossRef Granström F, Molarius A, Garvin P, Elo S, Feldman I, Kristenson M. Exploring trends in and determinants of educational inequalities in self-rated health. Scand J Public Health. 2015;43(7):677–86.CrossRef
17.
Zurück zum Zitat Szklo M, Nieto FJ. Epidemiology: beyond the basics. 3rd ed. Burlington: Mass: Jones & Bartlett Learning; 2014. p. 515. Szklo M, Nieto FJ. Epidemiology: beyond the basics. 3rd ed. Burlington: Mass: Jones & Bartlett Learning; 2014. p. 515.
18.
Zurück zum Zitat VanderWeele T. Explanation in causal inference: methods for mediation and interaction. 1st ed. New York: Oxford University Press; 2015. p. 729. VanderWeele T. Explanation in causal inference: methods for mediation and interaction. 1st ed. New York: Oxford University Press; 2015. p. 729.
19.
Zurück zum Zitat Knol MJ, Egger M, Scott P, Geerlings MI, Vandenbroucke JP. When one depends on the other: reporting of interaction in case-control and cohort studies. Epidemiology. 2009;20(2):161–6.CrossRef Knol MJ, Egger M, Scott P, Geerlings MI, Vandenbroucke JP. When one depends on the other: reporting of interaction in case-control and cohort studies. Epidemiology. 2009;20(2):161–6.CrossRef
20.
Zurück zum Zitat Ahnquist J, Wamala SP, Lindstrom M. Social determinants of health – a question of social or economic capital? Interaction effects of socioeconomic factors on health outcomes. Soc Sci Med. 2012;74(6):930–9.CrossRef Ahnquist J, Wamala SP, Lindstrom M. Social determinants of health – a question of social or economic capital? Interaction effects of socioeconomic factors on health outcomes. Soc Sci Med. 2012;74(6):930–9.CrossRef
21.
Zurück zum Zitat Burr H, Pohrt A, Rugulies R, Holtermann A, Hasselhorn HM. Does age modify the association between physical work demands and deterioration of self-rated general health? Scand J Work Environ Health. 2017;43(3):241–9.CrossRef Burr H, Pohrt A, Rugulies R, Holtermann A, Hasselhorn HM. Does age modify the association between physical work demands and deterioration of self-rated general health? Scand J Work Environ Health. 2017;43(3):241–9.CrossRef
22.
Zurück zum Zitat Cho S-S, Ju Y-S, Paek D, Kim H, Jung-Choi K. The combined effect of long working hours and low job control on self-rated health: an interaction analysis. J Occup Environ Med. 2018;60(5):475–80.CrossRef Cho S-S, Ju Y-S, Paek D, Kim H, Jung-Choi K. The combined effect of long working hours and low job control on self-rated health: an interaction analysis. J Occup Environ Med. 2018;60(5):475–80.CrossRef
24.
Zurück zum Zitat Schmidt MI, Duncan BB, Mill JG, Lotufo PA, Chor D, Barreto SM, et al. Cohort profile: longitudinal study of adult health (ELSA-Brasil). Int J Epidemiol. 2015;44(1):68–75.CrossRef Schmidt MI, Duncan BB, Mill JG, Lotufo PA, Chor D, Barreto SM, et al. Cohort profile: longitudinal study of adult health (ELSA-Brasil). Int J Epidemiol. 2015;44(1):68–75.CrossRef
26.
Zurück zum Zitat Mill JG, Pinto K, Griep RH, Goulart A, Foppa M, Lotufo PA, et al. Afericoes e exames clinicos realizados nos participantes do ELSA-Brasil. Rev Saude Publica. 2013;47(suppl 2):54–62.CrossRef Mill JG, Pinto K, Griep RH, Goulart A, Foppa M, Lotufo PA, et al. Afericoes e exames clinicos realizados nos participantes do ELSA-Brasil. Rev Saude Publica. 2013;47(suppl 2):54–62.CrossRef
27.
Zurück zum Zitat WHO. Physical Status: The use and interpretation of anthropometry. Geneva: World Health Organization; 1995. p. 463. WHO. Physical Status: The use and interpretation of anthropometry. Geneva: World Health Organization; 1995. p. 463.
28.
Zurück zum Zitat WHO. Waist circumference and waist-hip ratio: report of a WHO expert consultation. Geneva: World Health Organization; 2011. WHO. Waist circumference and waist-hip ratio: report of a WHO expert consultation. Geneva: World Health Organization; 2011.
30.
Zurück zum Zitat Rocha KB, Muntaner C, Solar O, Borrell C, Bernales P, González MJ, et al. Clase social, factores de riesgo psicosocial en el trabajo y su asociación con la salud autopercibida y mental en Chile. Cad Saude Publica. 2014;30(10):2219–34.CrossRef Rocha KB, Muntaner C, Solar O, Borrell C, Bernales P, González MJ, et al. Clase social, factores de riesgo psicosocial en el trabajo y su asociación con la salud autopercibida y mental en Chile. Cad Saude Publica. 2014;30(10):2219–34.CrossRef
31.
Zurück zum Zitat Kleinbaum DG, Klein M, Pryor ER. Logistic regression: a self-learning text. 3rd ed. New York: Springer; 2010. p. 701. (Statistics in the health sciences)CrossRef Kleinbaum DG, Klein M, Pryor ER. Logistic regression: a self-learning text. 3rd ed. New York: Springer; 2010. p. 701. (Statistics in the health sciences)CrossRef
32.
Zurück zum Zitat Corraini P, Olsen M, Pedersen L, Dekkers O, Vandenbroucke J. Effect modification, interaction and mediation: an overview of theoretical insights for clinical investigators. Clin Epidemiol. 2017;9:331–8.CrossRef Corraini P, Olsen M, Pedersen L, Dekkers O, Vandenbroucke J. Effect modification, interaction and mediation: an overview of theoretical insights for clinical investigators. Clin Epidemiol. 2017;9:331–8.CrossRef
33.
Zurück zum Zitat Kamangar F. Effect modification in epidemiology and medicine. Arch Iran Med. 2012;15(9):575–82.PubMed Kamangar F. Effect modification in epidemiology and medicine. Arch Iran Med. 2012;15(9):575–82.PubMed
34.
Zurück zum Zitat Schmidt MI, Griep RH, Passos VM, Luft VC, Goulart AC, de Menezes GM S, et al. Estrategias e desenvolvimento de garantia e controle de qualidade no ELSA-Brasil. Rev Saude Publica. 2013;47(suppl 2):105–12.CrossRef Schmidt MI, Griep RH, Passos VM, Luft VC, Goulart AC, de Menezes GM S, et al. Estrategias e desenvolvimento de garantia e controle de qualidade no ELSA-Brasil. Rev Saude Publica. 2013;47(suppl 2):105–12.CrossRef
Metadaten
Titel
Brazilian Longitudinal Study of Adult Health (ELSA-Brasil): socio-occupational class as an effect modifier for the relationship between adiposity measures and self-rated health
verfasst von
Thaís Lopes de Oliveira
Rosane Harter Griep
Joanna Nery Guimarães
Luana Giatti
Dóra Chor
Maria de Jesus Mendes da Fonseca
Publikationsdatum
01.12.2019
Verlag
BioMed Central
Erschienen in
BMC Public Health / Ausgabe 1/2019
Elektronische ISSN: 1471-2458
DOI
https://doi.org/10.1186/s12889-019-7072-y

Weitere Artikel der Ausgabe 1/2019

BMC Public Health 1/2019 Zur Ausgabe