Chronic diseases and socioeconomic inequalities in quality of life among Brazilian adults: findings from a population-based study in Southern Brazil

Chronic diseases and socioeconomic inequalities in quality of life among Brazilian adults:... Abstract Background To evaluate the association between sociodemographic conditions and the quality of life (QoL) in adults and investigate whether these inequalities are greater among individuals with long-lasting chronic health conditions. Methods Cross-sectional analysis of the second wave (2012) of the EpiFloripa Study, a population-based cohort of 1720 adults living in Southern Brazil. QoL domains (physical, psychological, social relationships and environmental) were evaluated using the WHOQoL-BREF. Unadjusted and adjusted means of QoL according to socioeconomic and demographic variables were estimated and stratified by the presence of long-lasting chronic conditions (heart disease, stroke, diabetes, hypertension, chronic kidney disease, cirrhosis, tendinitis, arthritis, rheumatism and/or fibromyalgia) were peformed in 2016. Results Among 1222 interviewed adults (56.6% females, mean age 41.7 ± 11.4 years; follow-up rate 71.1%), the prevalence of 1+ long-lasting chronic disease was 37.3% (95%CI: 34.4–40.3). Their effect on the QoL was four times higher on the physical component (−9.6; 95%CI −12.1; −7.1) than on the other domains. Adults aged 40+ years with black skin colour or lower educational level had a lower physical QoL score only when any chronic disease was present. Among those with some chronic illness, the psychological domain was also lower among those aged 40+ years and with a lower family income. No interaction between sociodemographic variables and chronic diseases was observed for the other QoL domains. Conclusions The occurrence of long-lasting chronic diseases is associated with inequalities in QoL (physical and psychological domains), with stronger adverse effects among older adults, blacks and individuals with lower income or educational levels. Introduction Currently, chronic diseases account for 68% of all deaths worldwide,1 with 80% of them occurring in low-and-middle-income countries (LMIC), where they are fast replacing infectious diseases and malnutrition as the leading causes of disability and premature death.1,2 In Brazil, non-communicable chronic diseases (NCDs) also constitute a public health problem, accounting for 72% of all deaths, mainly among older individuals and those with lower income and schooling. NCDs are also responsible for decreased quality of life (QoL), higher degree of disability and impairment of daily life activities, which largely impact family, communities and society.2 As a result, there has been an increasing interest in the last decades for studies aiming to investigate the impact of specific diseases on QoL.3 According to the World Health Organization (WHO), QoL involves different relative dimensions, including individual perceptions, life position, culture and value systems about personal goals, expectations, living standards and concerns.4 This is a patient-centred outcome, deeply related to health conditions, used by policy decision-makers to understand the benefits for society of public health spending.5 NCDs are long-lasting and slowly progressive conditions, which can impact QoL in different ways: reducing well-being, restricting functional status, generating problems with obtaining and/or maintaining employment, affecting social relationships and involving psychosocial distress.6 At the same time, a lower QoL has a profound impact on functional capacity, adherence to treatment, disease severity and further complications among individuals with NCDs.1,6 Therefore, QoL has been suggested as a subjective instrument able to capture the impact of these illnesses on an individual’s life and evaluate their life-course progression.6,7 Furthermore, individuals belonging to socioeconomically disadvantaged groups tend to have a higher prevalence and/or severity of NCDs as a consequence of a higher frequency of unhealthy lifestyle, poor living and working conditions, limited access to health services, delays in the diagnosis of these conditions, and inappropriate knowledge of or compliance with disease management.2,8 Consequently, the coexistence of NCDs and socioeconomic disadvantage could interact to exacerbate QoL inequalities. Studies investigating this hypothesis are especially relevant for LMIC, considering the rapid increase in the prevalence of NCDs they are facing,1,2,9 the high levels of income inequalities10 and difficulties accessing health services.11 However, few studies have investigated the joint effect of socioeconomic disadvantage and occurrence of NCDs on QoL in LMIC, particularly studies using robust instruments to evaluate QoL and including population-based samples. Therefore, this paper aims to evaluate the association between sociodemographic variables and the QoL of adults living in Southern Brazil and investigate whether the presence of long-lasting chronic conditions increases QoL inequalities. Methods We analysed the second wave of the EpiFloripa Cohort study, a population-based cohort performed in Florianopolis (a state capital in Southern Brazil). In 2013, the city had an estimated population of 453 285 inhabitants (55% adults aged 20+ years) and a Human Development Index of 0.847 (the third highest in the country).12 Sample The baseline study conducted in 2009 included a representative sample of 1720 adults (20–59 years).The first follow-up occurred in 2012 when the QoL questions were incorporated into the study. Details about the methodology are available elsewhere.13 Briefly, the sampling process was performed in two stages: firstly, census tracts were systematically selected in each decile of household income (63/420), and, subsequently, the households (1134/16 755) were systematically selected. All adult residents in each household were considered eligible, except for those with some severe physical and/or neurological impairment that could affect their ability to answer the questions. Measures In 2012, all participants included in the baseline were traced. Household visits were phone scheduled or, when this failed, the interviewer directly visited the last available address of the participant. Personal data assistants were used in both waves to record participants’ data. Questionnaires were pre-tested, and a random sample of 15% of respondents in 2009 and 10% in 2012 answered a short-version of the original interview two weeks after the household interview. The κ value of these variables ranged from 0.6 (i.e. use of medicines) to 1.0 (age). The outcome was measured in 2012 using the Brazilian version of the short form of the World Health Organization QoL questionnaire (WHOQol-BREF).4 This instrument has excellent test–retest reliability, internal consistency and high construct and discriminant validity.4,14 TheWHOQol-BREF includes 26 questions, which provide information about four domains of QoL: physical (including daily living activities, energy, mobility, pain, sleep and work capacity); psychological (including body imagesatisfaction, feelings, self-esteem, personal beliefs, memory and concentration); social relationships (including personal relationships, social support and sexual activities); and environmental (including financial resources, safety and security, health/social care accessibility/quality, physical and home environment, recreation opportunities and transport).4 All domains were converted into discrete variables (0–100 scale), with higher values representing a better QoL.The WHOQol-BREF also includes a question regarding the respondents' QoL perception (very poor, poor, neither poor nor good, good or very good), which was analysed as a binary variable (good/very good QoL: yes or no). The exposure variables included demographic (gender, skin colour and age) and socioeconomic (schooling and monthly family income) characteristics. Self-reported skin colour was collected using the Brazilian Institute of Geography and Statistics classification (white, dark, black, Asiatic or indigenous).15 The age of the participants was collected as a numerical variable and divided into categories for analysis (20–29; 30–39; 40–49; 50+ years). In each wave, the monthly family income (1 USD = R$1.70 in 2009) was divided by the squared number of household members to obtain the equalised family income and then divided into tertiles (the first tertile representing a lower income). Schooling was measured as the total number of years successfully completed and then divided into categories (≤4; 5–8; 9–11; ≥12 years). The diagnosis of NCDs was self-reported and based on the question, ‘Has a physician or health professional told you that you have…’, followed by a list of conditions included in the Brazilian National Household Sample Survey.16 The long-lasting NCDs included in this study are (1) heart or cardiovascular disease; (2) stroke, cerebrovascular accident or brain ischemia; (3) diabetes mellitus; (4) systemic blood hypertension; (5) chronic kidney disease; (6) cirrhosis; (7) tendinitis; (8) arthritisor rheumatism and; (9) fibromyalgia. The number of affirmative answers for these conditions was summed up and subsequently transformed into a binary variable (presence of 1+ chronic conditions). Statistical analysis The statistical software Stata 13.0® (StataCorporation, College Station, United States) was used for all analyses. Characteristics of the sampling process, as well as sampling weights (probability of selection in 2009 and follow-up in 2012) were considered in all estimations. Mean and standard deviation or median and interquartile range (IQR) were used to describe continuous variables, depending on their symmetry. Absolute and relative frequencies were used for categorical variables. The QoL domain means were tested against sociodemographic variables through t-test or analysis of variance (ANOVA), as appropriate. Stratified analyses were performed according to the presence of any long-lasting NCDs. Linear regression models were used for crude and adjusted analyses. Wald tests for heterogeneity or trend were considered, depending on the nature of the independent variable. In multivariable analyses, demographic variables were first included in the models, and then the socioeconomic ones. An α of 5% was defined as indicative of statistical significance. In each model, marginal means of QoL in each category of the exposure variables were estimated. Interaction terms between the sociodemographic variables and the presence of long-lasting NCDs were included in the regression models and considered as indicative of effect modification when a P < 0.10 was obtained. The multicollinearity between the investigated sociodemographic variables was investigated considering the variance inflation factor (VIF) and the tolerance of the models. The EpiFloripa study was approved by the Ethics Committee on Human Research of the Federal University of Santa Catarina (no 351/08 and no 1772/11). All participants were informed about the objectives of the study and signed an informed consent form. Results In 2009, the mean age of the cohort was 37.5 ± 11.6 years (55.1% females), the median of the equalised family income was R$1443.4 (IQR R$866–2828; 1USD = R$1.70 in 2009), and the median of schooling was 11 years (IQR 9–15). The follow-up rate in 2012 was 71.1% (n = 1222; mean age 41.7 ± 11.4 years; 56.6% females). The only difference between the individuals located in 2012 and the original cohort was a lower proportion of individuals in the youngest age group in 2012 (Supplementary table S1). However, this difference was lower than 5% points. In 2009, the presence of at least one longstanding NCD was reported by 37.3% (95%CI: 34.4–40.3) of the sample, with 12.9% affected by two or more conditions. In decreasing order of frequency, the prevalence of these conditions were tendinitis (17.8%), hypertension (14.1%), arthritis or rheumatism (7.5%), heart disease (6.6%), diabetes mellitus (3.7%), fibromyalgia (2.6%), chronic kidney disease (2.3%), stroke (0.6%) and cirrhosis (0.1%). The QoL in 2012 was classified as good or very good by 86.3% of the respondents (95%CI: 81.0–92.2%), with no differences between those with or without a long-lasting NCD (87.8% and 84.9%, respectively). Mean QoL scores were 72.9 points (95%CI: 71.7–74.2) for the physical domain, 71.1 (CI95%: 70.2–72.1) for the psychological domain, 75.2 (95%CI: 74.2–76.3) for the social relationship domain, and 62.0 (95%CI: 60.5–63.6) for the environmental domain. The presence of at least one long-lasting NCD had a stronger association with the physical domain (mean difference −9.6; 95%CI: −12.1;−7.1) and a smaller but significant association with the psychological (−2.5; 95%CI: −4.4;−0.5) and environmental domains (−2.6; 95%CI: −4.4;−0.8).However, the association with the social domain was not significant (−2.2; 95%CI: −4.5; 0.1; data not shown in tables). Table 1 shows unadjusted associations between the sociodemographic variables and the four QoL domains, stratified by the presence of long-lasting NCDs. Women showed lower QoL than men in all domains, regardless of the presence of chronic conditions. Age was inversely associated with the physical and psychological domains of QoL only among those with long-lasting NCDs, while the environmental score was higher in the oldest age category only among those free of these illnesses. Table 1 Unadjusted means and standard errors (SE) of quality of life in adults (2012), according to sociodemographic variables (N = 1222)   Physical domain   Psychological domain   Social domain   Environmental domain   Long-lasting chronic diseaseb   Long-lasting chronic diseaseb   Long-lasting chronic diseaseb   Long-lasting chronic diseaseb   Variables (n)a  No   Yes   No   Yes   No   Yes   No   Yes   Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Gender  P<0.001*  P = 0.002*  P = 0.039*  P = 0.007*  P = 0.199*  P = 0.267*  P = 0.034*  P = 0.034*  Male (n = 518)  78.8(0.64)  70.3(1.35)  73.1(0.68)  72.1(0.87)  76.8(0.78)  72.9(1.16)  64.4(0.87)  62.1(1.09)  Female (n = 694)  74.8(0.95)  65.3(1.25)  71.3(0.79)  68.3(1.05)  75.4(0.88)  73.3(1.19)  61.8(1.23)  59.4(1.02)  Age  P = 0.765†  P = 0.002†  P = 0.450†  P = 0.005†  P = 0.381†  P = 0.055†  P = 0.005†  P = 0.528†  20–29 years (n = 325)  77.6(0.94)  76.4(2.17)  73.0(0.95)  74.1(1.91)  78.3(1.10)  73.8(2.75)  62.2(1.15)  60.4(2.89)  30–39 years (n = 276)  75.3(1.21)  70.4(1.96)  71.2(1.37)  72.3(1.62)  73.9(1.17)  74.6(1.63)  61.1(1.21)  60.4(2.08)  40-49 years (n = 344)  78.4(1.08)  65.3(1.65)  73.6(0.95)  68.4(1.15)  76.8(1.28)  72.8(2.03)  63.9(1.26)  58.5(1.30)  50+ years (n = 267)  76.0(1.34)  64.9(1.75)  70.9(1.13)  68.1(1.34)  75.6(1.29)  73.0(1.46)  65.6(1.48)  61.5(1.08)  Skin colour  P = 0.853**  P = 0.031**  P = 0.003**  P = 0.052**  P = 0.715**  P = 0.885**  P < 0.001**  P = 0.002**  White (n = 996)  76.7(0.70)  67.9(1.10)  72.9(0.59)  70.4(0.81)  76.3(0.63)  73.8(1.04)  64.0(1.00)  61.5(0.85)  Dark (n = 131)  77.2(1.72)  65.3(3.46)  68.7(1.50)  67.6(2.49)  74.7(1.80)  77.0(2.73)  60.6(1.48)  56.3(2.50)  Black (n = 59)  76.6(1.85)  60.5(3.94)  70.2(1.76)  66.1(3.10)  77.2(2.43)  71.4(2.80)  53.6(2.06)  52.6(3.25)  Family income (tertiles)c  P < 0.001†  P < 0.001†  P < 0.001†  P < 0.001†  P = 0.852†  P = 0.865†  P < 0.001†  P < 0.001†  First (<R$1429R$) (n = 409)  73.5(1.10)  60.1(1.68)  68.4(1.03)  64.3(1.39)  75.8(1.16)  72.9(2.14)  53.8(0.93)  52.3(1.35)  Second (1429–2887R$) (n = 387)  76.6(0.93)  70.1(1.46)  73.1(0.91)  70.9(0.95)  75.6(1.16)  75.6(1.33)  63.7(0.90)  61.6(0.93)  Third (>R$2, 887) (n = 398)  79.3(0.86)  71.4(1.37)  74.6(0.87)  74.0(0.93)  76.1(1.20)  73.2(1.19)  70.4(1.10)  68.0(1.04)  Schooling  P = 0.017†  P < 0.001†  P < 0.001†  P < 0.001†  P = 0.984†  P = 0.984†  P < 0.001†  P < 0.001†  0–4 years (n = 107)  71.8(3.12)  55.8(3.23)  63.9(1.66)  62.6(2.81)  78.2(3.14)  75.2(3.25)  54.3(2.36)  54.6(2.13)  5–8 years ( n = 172)  75.7(2.07)  61.0(2.49)  67.0(1.70)  63.9(2.06)  76.8(1.59)  72.3(2.62)  55.1(1.59)  52.7(1.67)  9–11 years (n = 392)  75.9(1.17)  68.9(1.43)  70.9(1.12)  70.7(1.13)  74.5(1.14)  74.1(1.16)  59.7(1.05)  59.5(1.10)  12+ years (n = 538)  78.2(0.66)  71.5(1.26)  75.4(0.58)  73.2(0.86)  76.8(0.80)  74.0(1.31)  68.5(0.79)  65.8(1.01)    Physical domain   Psychological domain   Social domain   Environmental domain   Long-lasting chronic diseaseb   Long-lasting chronic diseaseb   Long-lasting chronic diseaseb   Long-lasting chronic diseaseb   Variables (n)a  No   Yes   No   Yes   No   Yes   No   Yes   Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Gender  P<0.001*  P = 0.002*  P = 0.039*  P = 0.007*  P = 0.199*  P = 0.267*  P = 0.034*  P = 0.034*  Male (n = 518)  78.8(0.64)  70.3(1.35)  73.1(0.68)  72.1(0.87)  76.8(0.78)  72.9(1.16)  64.4(0.87)  62.1(1.09)  Female (n = 694)  74.8(0.95)  65.3(1.25)  71.3(0.79)  68.3(1.05)  75.4(0.88)  73.3(1.19)  61.8(1.23)  59.4(1.02)  Age  P = 0.765†  P = 0.002†  P = 0.450†  P = 0.005†  P = 0.381†  P = 0.055†  P = 0.005†  P = 0.528†  20–29 years (n = 325)  77.6(0.94)  76.4(2.17)  73.0(0.95)  74.1(1.91)  78.3(1.10)  73.8(2.75)  62.2(1.15)  60.4(2.89)  30–39 years (n = 276)  75.3(1.21)  70.4(1.96)  71.2(1.37)  72.3(1.62)  73.9(1.17)  74.6(1.63)  61.1(1.21)  60.4(2.08)  40-49 years (n = 344)  78.4(1.08)  65.3(1.65)  73.6(0.95)  68.4(1.15)  76.8(1.28)  72.8(2.03)  63.9(1.26)  58.5(1.30)  50+ years (n = 267)  76.0(1.34)  64.9(1.75)  70.9(1.13)  68.1(1.34)  75.6(1.29)  73.0(1.46)  65.6(1.48)  61.5(1.08)  Skin colour  P = 0.853**  P = 0.031**  P = 0.003**  P = 0.052**  P = 0.715**  P = 0.885**  P < 0.001**  P = 0.002**  White (n = 996)  76.7(0.70)  67.9(1.10)  72.9(0.59)  70.4(0.81)  76.3(0.63)  73.8(1.04)  64.0(1.00)  61.5(0.85)  Dark (n = 131)  77.2(1.72)  65.3(3.46)  68.7(1.50)  67.6(2.49)  74.7(1.80)  77.0(2.73)  60.6(1.48)  56.3(2.50)  Black (n = 59)  76.6(1.85)  60.5(3.94)  70.2(1.76)  66.1(3.10)  77.2(2.43)  71.4(2.80)  53.6(2.06)  52.6(3.25)  Family income (tertiles)c  P < 0.001†  P < 0.001†  P < 0.001†  P < 0.001†  P = 0.852†  P = 0.865†  P < 0.001†  P < 0.001†  First (<R$1429R$) (n = 409)  73.5(1.10)  60.1(1.68)  68.4(1.03)  64.3(1.39)  75.8(1.16)  72.9(2.14)  53.8(0.93)  52.3(1.35)  Second (1429–2887R$) (n = 387)  76.6(0.93)  70.1(1.46)  73.1(0.91)  70.9(0.95)  75.6(1.16)  75.6(1.33)  63.7(0.90)  61.6(0.93)  Third (>R$2, 887) (n = 398)  79.3(0.86)  71.4(1.37)  74.6(0.87)  74.0(0.93)  76.1(1.20)  73.2(1.19)  70.4(1.10)  68.0(1.04)  Schooling  P = 0.017†  P < 0.001†  P < 0.001†  P < 0.001†  P = 0.984†  P = 0.984†  P < 0.001†  P < 0.001†  0–4 years (n = 107)  71.8(3.12)  55.8(3.23)  63.9(1.66)  62.6(2.81)  78.2(3.14)  75.2(3.25)  54.3(2.36)  54.6(2.13)  5–8 years ( n = 172)  75.7(2.07)  61.0(2.49)  67.0(1.70)  63.9(2.06)  76.8(1.59)  72.3(2.62)  55.1(1.59)  52.7(1.67)  9–11 years (n = 392)  75.9(1.17)  68.9(1.43)  70.9(1.12)  70.7(1.13)  74.5(1.14)  74.1(1.16)  59.7(1.05)  59.5(1.10)  12+ years (n = 538)  78.2(0.66)  71.5(1.26)  75.4(0.58)  73.2(0.86)  76.8(0.80)  74.0(1.31)  68.5(0.79)  65.8(1.01)  a Number of observations per categories. b Positive for at least one of heart disease, stroke, diabetes, hypertension, chronic kidney disease, cirrhosis, tendinitis, arthritis, rheumatism, fibromyalgia. c Equalised, using the monthly family income collected in 2012. * t-test. ** ANOVA test for heterogeneity. † ANOVA test for trend. Table 1 Unadjusted means and standard errors (SE) of quality of life in adults (2012), according to sociodemographic variables (N = 1222)   Physical domain   Psychological domain   Social domain   Environmental domain   Long-lasting chronic diseaseb   Long-lasting chronic diseaseb   Long-lasting chronic diseaseb   Long-lasting chronic diseaseb   Variables (n)a  No   Yes   No   Yes   No   Yes   No   Yes   Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Gender  P<0.001*  P = 0.002*  P = 0.039*  P = 0.007*  P = 0.199*  P = 0.267*  P = 0.034*  P = 0.034*  Male (n = 518)  78.8(0.64)  70.3(1.35)  73.1(0.68)  72.1(0.87)  76.8(0.78)  72.9(1.16)  64.4(0.87)  62.1(1.09)  Female (n = 694)  74.8(0.95)  65.3(1.25)  71.3(0.79)  68.3(1.05)  75.4(0.88)  73.3(1.19)  61.8(1.23)  59.4(1.02)  Age  P = 0.765†  P = 0.002†  P = 0.450†  P = 0.005†  P = 0.381†  P = 0.055†  P = 0.005†  P = 0.528†  20–29 years (n = 325)  77.6(0.94)  76.4(2.17)  73.0(0.95)  74.1(1.91)  78.3(1.10)  73.8(2.75)  62.2(1.15)  60.4(2.89)  30–39 years (n = 276)  75.3(1.21)  70.4(1.96)  71.2(1.37)  72.3(1.62)  73.9(1.17)  74.6(1.63)  61.1(1.21)  60.4(2.08)  40-49 years (n = 344)  78.4(1.08)  65.3(1.65)  73.6(0.95)  68.4(1.15)  76.8(1.28)  72.8(2.03)  63.9(1.26)  58.5(1.30)  50+ years (n = 267)  76.0(1.34)  64.9(1.75)  70.9(1.13)  68.1(1.34)  75.6(1.29)  73.0(1.46)  65.6(1.48)  61.5(1.08)  Skin colour  P = 0.853**  P = 0.031**  P = 0.003**  P = 0.052**  P = 0.715**  P = 0.885**  P < 0.001**  P = 0.002**  White (n = 996)  76.7(0.70)  67.9(1.10)  72.9(0.59)  70.4(0.81)  76.3(0.63)  73.8(1.04)  64.0(1.00)  61.5(0.85)  Dark (n = 131)  77.2(1.72)  65.3(3.46)  68.7(1.50)  67.6(2.49)  74.7(1.80)  77.0(2.73)  60.6(1.48)  56.3(2.50)  Black (n = 59)  76.6(1.85)  60.5(3.94)  70.2(1.76)  66.1(3.10)  77.2(2.43)  71.4(2.80)  53.6(2.06)  52.6(3.25)  Family income (tertiles)c  P < 0.001†  P < 0.001†  P < 0.001†  P < 0.001†  P = 0.852†  P = 0.865†  P < 0.001†  P < 0.001†  First (<R$1429R$) (n = 409)  73.5(1.10)  60.1(1.68)  68.4(1.03)  64.3(1.39)  75.8(1.16)  72.9(2.14)  53.8(0.93)  52.3(1.35)  Second (1429–2887R$) (n = 387)  76.6(0.93)  70.1(1.46)  73.1(0.91)  70.9(0.95)  75.6(1.16)  75.6(1.33)  63.7(0.90)  61.6(0.93)  Third (>R$2, 887) (n = 398)  79.3(0.86)  71.4(1.37)  74.6(0.87)  74.0(0.93)  76.1(1.20)  73.2(1.19)  70.4(1.10)  68.0(1.04)  Schooling  P = 0.017†  P < 0.001†  P < 0.001†  P < 0.001†  P = 0.984†  P = 0.984†  P < 0.001†  P < 0.001†  0–4 years (n = 107)  71.8(3.12)  55.8(3.23)  63.9(1.66)  62.6(2.81)  78.2(3.14)  75.2(3.25)  54.3(2.36)  54.6(2.13)  5–8 years ( n = 172)  75.7(2.07)  61.0(2.49)  67.0(1.70)  63.9(2.06)  76.8(1.59)  72.3(2.62)  55.1(1.59)  52.7(1.67)  9–11 years (n = 392)  75.9(1.17)  68.9(1.43)  70.9(1.12)  70.7(1.13)  74.5(1.14)  74.1(1.16)  59.7(1.05)  59.5(1.10)  12+ years (n = 538)  78.2(0.66)  71.5(1.26)  75.4(0.58)  73.2(0.86)  76.8(0.80)  74.0(1.31)  68.5(0.79)  65.8(1.01)    Physical domain   Psychological domain   Social domain   Environmental domain   Long-lasting chronic diseaseb   Long-lasting chronic diseaseb   Long-lasting chronic diseaseb   Long-lasting chronic diseaseb   Variables (n)a  No   Yes   No   Yes   No   Yes   No   Yes   Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Gender  P<0.001*  P = 0.002*  P = 0.039*  P = 0.007*  P = 0.199*  P = 0.267*  P = 0.034*  P = 0.034*  Male (n = 518)  78.8(0.64)  70.3(1.35)  73.1(0.68)  72.1(0.87)  76.8(0.78)  72.9(1.16)  64.4(0.87)  62.1(1.09)  Female (n = 694)  74.8(0.95)  65.3(1.25)  71.3(0.79)  68.3(1.05)  75.4(0.88)  73.3(1.19)  61.8(1.23)  59.4(1.02)  Age  P = 0.765†  P = 0.002†  P = 0.450†  P = 0.005†  P = 0.381†  P = 0.055†  P = 0.005†  P = 0.528†  20–29 years (n = 325)  77.6(0.94)  76.4(2.17)  73.0(0.95)  74.1(1.91)  78.3(1.10)  73.8(2.75)  62.2(1.15)  60.4(2.89)  30–39 years (n = 276)  75.3(1.21)  70.4(1.96)  71.2(1.37)  72.3(1.62)  73.9(1.17)  74.6(1.63)  61.1(1.21)  60.4(2.08)  40-49 years (n = 344)  78.4(1.08)  65.3(1.65)  73.6(0.95)  68.4(1.15)  76.8(1.28)  72.8(2.03)  63.9(1.26)  58.5(1.30)  50+ years (n = 267)  76.0(1.34)  64.9(1.75)  70.9(1.13)  68.1(1.34)  75.6(1.29)  73.0(1.46)  65.6(1.48)  61.5(1.08)  Skin colour  P = 0.853**  P = 0.031**  P = 0.003**  P = 0.052**  P = 0.715**  P = 0.885**  P < 0.001**  P = 0.002**  White (n = 996)  76.7(0.70)  67.9(1.10)  72.9(0.59)  70.4(0.81)  76.3(0.63)  73.8(1.04)  64.0(1.00)  61.5(0.85)  Dark (n = 131)  77.2(1.72)  65.3(3.46)  68.7(1.50)  67.6(2.49)  74.7(1.80)  77.0(2.73)  60.6(1.48)  56.3(2.50)  Black (n = 59)  76.6(1.85)  60.5(3.94)  70.2(1.76)  66.1(3.10)  77.2(2.43)  71.4(2.80)  53.6(2.06)  52.6(3.25)  Family income (tertiles)c  P < 0.001†  P < 0.001†  P < 0.001†  P < 0.001†  P = 0.852†  P = 0.865†  P < 0.001†  P < 0.001†  First (<R$1429R$) (n = 409)  73.5(1.10)  60.1(1.68)  68.4(1.03)  64.3(1.39)  75.8(1.16)  72.9(2.14)  53.8(0.93)  52.3(1.35)  Second (1429–2887R$) (n = 387)  76.6(0.93)  70.1(1.46)  73.1(0.91)  70.9(0.95)  75.6(1.16)  75.6(1.33)  63.7(0.90)  61.6(0.93)  Third (>R$2, 887) (n = 398)  79.3(0.86)  71.4(1.37)  74.6(0.87)  74.0(0.93)  76.1(1.20)  73.2(1.19)  70.4(1.10)  68.0(1.04)  Schooling  P = 0.017†  P < 0.001†  P < 0.001†  P < 0.001†  P = 0.984†  P = 0.984†  P < 0.001†  P < 0.001†  0–4 years (n = 107)  71.8(3.12)  55.8(3.23)  63.9(1.66)  62.6(2.81)  78.2(3.14)  75.2(3.25)  54.3(2.36)  54.6(2.13)  5–8 years ( n = 172)  75.7(2.07)  61.0(2.49)  67.0(1.70)  63.9(2.06)  76.8(1.59)  72.3(2.62)  55.1(1.59)  52.7(1.67)  9–11 years (n = 392)  75.9(1.17)  68.9(1.43)  70.9(1.12)  70.7(1.13)  74.5(1.14)  74.1(1.16)  59.7(1.05)  59.5(1.10)  12+ years (n = 538)  78.2(0.66)  71.5(1.26)  75.4(0.58)  73.2(0.86)  76.8(0.80)  74.0(1.31)  68.5(0.79)  65.8(1.01)  a Number of observations per categories. b Positive for at least one of heart disease, stroke, diabetes, hypertension, chronic kidney disease, cirrhosis, tendinitis, arthritis, rheumatism, fibromyalgia. c Equalised, using the monthly family income collected in 2012. * t-test. ** ANOVA test for heterogeneity. † ANOVA test for trend. Independent of the presence of NCDs, black skin colour participants had lower psychological and environmental QoL scores than their white peers. Nevertheless, for the physical domain, the score was lower only among blacks compared to whites when affected by long-lasting NCDs. Family income and schooling showed a positive trend association with the physical, psychological, and environmental QoL scores. Although the same direct-trend association was observed in the healthy group, the differences in the physical and psychological scores between the extreme categories of income and schooling were at least twice as high among those with these diseases. Figures 1–3 show the predicted adjusted means (and their 95%CI) for each of the four QoL domains, according to sociodemographic variables, and stratified by the presence of long-lasting NCDs. Even after adjustment for confounding, most of the associations remained relatively stable. Except for the social component, all QoL domains were lower in females, regardless of whether the participants had a NCD or not (Supplementary figure S1). Age remained negatively associated with the physical and psychological domains of QoL only among those with a long-lasting NCD, while an inverse-trend association with the social domain became apparent in the same strata (figure 1). Even after adjustment for the other sociodemographic variables, black skin colour participants had a lower score in the physical domain than whites. This effect was evident only among those with some chronic illness (figure 2). Figure 1 View largeDownload slide Adjusted means and 95% confidence intervals of quality of life domain scores according to age groups adjusted for gender and skin colour, stratified by longstanding chronic disease status. P-values for interaction test: Physical domain: 0.140; Psychological domain: 0.005; Social relations domain: 0.246; Environmental domain: 0.520. **: Wald's test for trend. Figure 1 View largeDownload slide Adjusted means and 95% confidence intervals of quality of life domain scores according to age groups adjusted for gender and skin colour, stratified by longstanding chronic disease status. P-values for interaction test: Physical domain: 0.140; Psychological domain: 0.005; Social relations domain: 0.246; Environmental domain: 0.520. **: Wald's test for trend. Figure 2 View largeDownload slide Adjusted means and 95% confidence intervals of quality of life domain scores according to skin colour adjusted for gender and age groups, stratified by longstanding chronic disease status. P-values for interaction test: Physical domain: 0.233; Psychological domain: 0.772; Social relations domain: 0.118; Environmental domain: 0.748. *: Wald's test for heterogeneity. Figure 2 View largeDownload slide Adjusted means and 95% confidence intervals of quality of life domain scores according to skin colour adjusted for gender and age groups, stratified by longstanding chronic disease status. P-values for interaction test: Physical domain: 0.233; Psychological domain: 0.772; Social relations domain: 0.118; Environmental domain: 0.748. *: Wald's test for heterogeneity. Figure 3 View largeDownload slide Adjusted means and 95% confidence intervals of quality of life domain scores according to per capita family income adjusted for demographic and socioeconomic variables, stratified by longstanding chronic disease status. Florianopolis (SC), Brazil, 2012. P-values for interaction test: Physical domain: 0.140; Psychological domain: 0.005; Social relations domain: 0.510; Environmental domain: 0.481. **: Wald's test for trend. Figure 3 View largeDownload slide Adjusted means and 95% confidence intervals of quality of life domain scores according to per capita family income adjusted for demographic and socioeconomic variables, stratified by longstanding chronic disease status. Florianopolis (SC), Brazil, 2012. P-values for interaction test: Physical domain: 0.140; Psychological domain: 0.005; Social relations domain: 0.510; Environmental domain: 0.481. **: Wald's test for trend. Family income remained positively associated with the physical and environmental scores of QoL, with no evidence of heterogeneity according to the presence of NCDs. Nevertheless, the psychological domain was adversely affected by lower family income only among those with some chronic illness (figure 3). No evidence of multicollinearity was identified in the study (VIF between 1.00–1.39 and tolerance 0.72–0.99). Discussion Our original hypothesis of socioeconomic inequalities in QoL related to the presence of long-lasting chronic conditions was confirmed for the physical and psychological domains. Adults over the age of 40 years, with black skin colour, and with lower educational levels were more negatively associated with the physical domain of QoL than their peers only when a chronic condition was present. Furthermore, when affected by some chronic illness, individuals aged 40+ years or with a lower family income showed a reduced score in the psychological domain of QoL. According to our findings, long-lasting NCDs affected more than one-third of adults, which is consistent with the global estimates by the WHO.17 Several studies have demonstrated the impact of NCDs on the QoL, principally on the physical domain.18 Socioeconomic and demographic variables (lower educational level, unemployment and female gender) have been also associated with a lower QoL.3,19–21 There is also substantial evidence in the scientific literature suggesting that NCDs (CVD, hypertension, dyslipidaemia and diabetes) predominantly affect individuals from lower socioeconomic groups.22–24 The few studies investigating the joint effect of socioeconomic disadvantage and the presence of long-lasting health conditions as a source of QoL inequalities20,21,25corroborate our findings. However, they were conducted in Europe and used theEQ-5 D, an instrument that does not allow the separation of the effects of these conditions on the different QoL domains.7 In agreement with our findings, various authors have suggested that socioeconomically disadvantaged groups experience a ‘double suffering,’ characterised by a higher prevalence of NCDs and impaired QoL (especially the physical domain).20,26 The higher burden of NCDs among individuals with a lower socioeconomic status has been related to a higher frequency of unhealthy lifestyle, poor living and working conditions, difficulties in utilising health services, late diagnosis, inappropriate disease management, and higher frequency of complications.23,27 Nonetheless, the assessment of the double suffering requires to investigate whether the QoL differs across the same levels of morbidity, which is beyond the scope of this paper. In this sense, Brazilian health disparities remain despite all progress achieved by the Unified Brazilian Health System (SUS), especially when more than 50% of the population use private health services. These private plans are more affordable for the wealthiest, and they are related to a higher health service utilization.28,29 Furthermore, health expenses associated with illness management increase psychological distress,30 which is consistent with our findings of a lower psychological score of QoL among poor individuals affected by NCDs.The proportion of the family budget used for purchasing medicines is 2.7 times higher among the poorest than the wealthiest.31 Additionally, chronic illnesses are also associated with decreased workforce participation, early retirement, working limitations, absence due to sickness, and lower access/return to work,32 with a consequent impact on socioeconomic status and QoL.3 All these factors could explain the lower physical and psychological scores of QoL among adults aged 40+ years affected by NCDs, although these negative outcomes were already apparent at the age of 30–39 years. These individuals belong to economically active groups, which in turn leads to worse consequences from an economic perspective.32 Additionally, chronic illnesses are not only associated with socioeconomic inequalities, but they also aggravate QoL disparities as a consequence of the underlying vicious poverty-disease cycle.33 Black skin colour participants showed a reduced score in the physical domain only when affected by a NCD. Brazil has the largest population of African descendants among all Latin American countries, and racial discrimination is considered a foundation for Brazilian social inequalities.34 Perceived discrimination and socioeconomic disadvantage have been associated with refraining from seeking medical treatment, even after adjustment for confounders.35 Diverse patient-centred interventions have been proposed to improve QoL among people with NCDs, which can potentially reduce socioeconomic inequalities. They include the increase in the number of health consultations, a better assessment of risk factors, the encouragement of regular physical activity, and enhancing health literacy.3,25,36 However, further longitudinal studies would be necessary to elucidate the real benefit of these possible interventions. Finally, although the social and/or the environmental QoL domains were also lower among women, older individuals, those with lower income or schooling, or those with black skin colour, the presence of long-lasting NCDs did not modify the effect of these associations. Similar results have also been demonstrated in other studies.3,19–21 This study has some important strengths, such as the quality control of the interview-based assessments, the robust instrument used to investigate QoL, and the evaluation of a population-based sample of adults in a middle-income setting. One of the possible limitations is the percentage of losses of participants for follow-up. However, the individuals located in 2012 were similar to the original cohort according to most of the baseline characteristics. A second possible limitation is the cross-sectional analysis of the available data, which does not allow the evaluation of the temporality of the associations. Nevertheless, it could be assumed that the diagnosis of NCDs preceded the assessment of the QoL. Thirdly, mental health problems were not included in the list of chronic health conditions investigated in this study, and they are a potential source of bias37,38 to be considered in further studies, either as confounders or moderators of the associations.37,38 Finally, even though the NCDs were self-reported, some studies have demonstrated the excellent reliability of this information.39,40 Besides, it is unlikely that this information bias might explain our results, as underreporting these conditions would reduce the effect magnitude of the associations. This study showed that age, skin colour, income, and schooling interact with the occurrence of NCDs, reducing the physical and mental domains of QoL among adults. These results reinforce the relevance of developing health policies with a special focus on socioeconomically disadvantaged groups, considering the adverse effect of a reduced QoL for health management and disease progression.18 However, deciding between a universal care model and one targeting specific groups is one of the further challenges faced by the Brazil's health system.29 A multidisciplinary approach is necessary to reduce QoL inequalities, considering the increase in life expectancy and the prevalence of long-lasting NCDs. Acknowledgements This article is part of the post-doctoral studies of D.A. Höfelmann at the Post-Graduate Program in Nutrition at the Federal University of Santa Catarina, Florianopolis, Brazil. We would like to thank the Brazilian Institute of Geography and Statistics (IBGE) and the Florianópolis Health Authority staff for their useful help with the practical aspects of the EpiFloripa Cohort Study. Funding This paper is based on the EpiFloripa – Florianópolis Adult Health Study, Brazil. The Project was sponsored by the Brazilian National Council for Scientific and Technological Development (CNPq - Grant No. 485327/2007-4 and 508903/2010-6) and the Brazilian Coordination for the Improvement of Education Personnel (CAPES – Grant No. PVE-A020/2013). This research was developed by the Federal University of Santa Catarina, Brazil. D.A. Höfelmann received a post-doctoral scholarship from the Brazilian Coordination for the Improvement of Education Personnel (CAPES). Supplementary data Supplementary data are available at EURPUB online. Conflicts of interest: None declared Key points Chronic diseases impact more the physical than other domains of quality of life. Chronic diseases are associated with inequalities in physical and psychological QoL. The effects of chronic conditions are stronger among lower socioeconomic groups. References 1 World Health Organization. WHO Library Cataloguing-in-Publication Data Health in 2015: from MDGs, Millennium Development Goals to SDGs, Sustainable Development Goals . In: WHO, ed. Geneve: WHO, 2015. 2 Hunter DJ, Reddy KS. Noncommunicable diseases. N Engl J Med  2013; 369: 1336– 43. http://dx.doi.org/10.1056/NEJMra1109345 Google Scholar CrossRef Search ADS PubMed  3 Gonzalez-Chica DA, Mnisi Z, Avery J, et al.   Effect of health literacy on quality of life amongst patients with ischaemic heart disease in Australian general practice. PLoS One  2016; 11: e0151079. doi: 10.1371/journal.pone.0151079. eCollection 2016. Google Scholar CrossRef Search ADS PubMed  4 World Health Organization. WHOQOL-BREF: Introduction, Administration, Scoring and Generic Version of the Assessment . Geneve, Switzerland: WHO, 1996. 5 Fayers PM, Machin D. Quality of Life: Assessment, Analysis and Interpretation  West Sussex, England: John Wiley & Sons Ltd, 2000. 6 Megari K. Quality of life in chronic disease patients. Health Psychol Res  2013; 1: e27. Google Scholar CrossRef Search ADS PubMed  7 Mark DB. Assessing quality-of-life outcomes in cardiovascular clinical research. Nat Rev Cardiol  2016; 13: 286– 308. http://dx.doi.org/10.1038/nrcardio.2016.10 Google Scholar CrossRef Search ADS PubMed  8 Blank N, Diderichsen F. Social inequalities in the experience of illness in Sweden: a “double suffering”. Scand J Soc Med  1996; 24: 81– 9. Google Scholar CrossRef Search ADS PubMed  9 Malta DC, Stopa SR, Szwarcwald CL, et al.   A vigilância e o monitoramento das principais doenças crônicas não transmissíveis no Brasil - Pesquisa Nacional de Saúde, 2013. Revista Brasileira de Epidemiologia  2015; 18: 3– 16. Google Scholar CrossRef Search ADS PubMed  10 World Bank. GINI index (World Bank estimate) . World Bank, 2016. https://data.worldbank.org/indicator/SI.POV.GINI. PubMed PubMed  11 Boccolini CS, de Souza Junior PR. Inequities in Healthcare utilization: results of the Brazilian National Health Survey, 2013. Int J Equity Health  2016; 15: 150. http://dx.doi.org/10.1186/s12939-016-0444-3 Google Scholar CrossRef Search ADS PubMed  12 Brazilian Institute of Geographic and Statistics. Cidades. Rio de Janeiro: IBGE, 2010. 13 Boing AC, Peres KG, Boing AF, et al.   EpiFloripa Health Survey: the methodological and operational aspects behind the scenes. Rev Bras Epidemiol  2014; 17: 147– 62. http://dx.doi.org/10.1590/1415-790X201400010012ENG Google Scholar CrossRef Search ADS PubMed  14 Fleck MP, Louzada S, Xavier M, et al.   [Application of the Portuguese version of the abbreviated instrument of quality life WHOQOL-bref]. Rev Saude Publica  2000; 34: 178– 83. Google Scholar CrossRef Search ADS PubMed  15 Brazilian Institute of Geographic and Statistics. Censo Brasileiro 2010. IBGE, 2010. 16 Brazilian Institute of Geographic and Statistics. PNAD - Pesquisa Nacional por Amostra de Domicílios. Metodologia. Brasília: IBGE, 2016. 17 WHO. WHO maps noncommunicable disease trends in all countries: country profiles on noncommunicable disease trends in 193 countries. Cent Eur J Public Health  2011; 19: 130. 138. PubMed  18 Fortin M, Lapointe L, Hudon C, et al.   Multimorbidity and quality of life in primary care: a systematic review. Health Qual Life Outcomes  2004; 2: 51. http://dx.doi.org/10.1186/1477-7525-2-51 Google Scholar CrossRef Search ADS PubMed  19 Mello A. d C, Engstrom EM, Alves LC. Health-related and socio-demographic factors associated with frailty in the elderly: a systematic literature review. Cad Saude Publica  2014; 30: 1143– 68. http://dx.doi.org/10.1590/0102-311X00148213 Google Scholar CrossRef Search ADS PubMed  20 Mielck A, Vogelmann M, Leidl R. Health-related quality of life and socioeconomic status: inequalities among adults with a chronic disease. Health Qual Life Outcomes  2014; 12: 58. http://dx.doi.org/10.1186/1477-7525-12-58 Google Scholar CrossRef Search ADS PubMed  21 Stafford M, Soljak M, Pledge V, Mindell J. Socio-economic differences in the health-related quality of life impact of cardiovascular conditions. Eur J Public Health  2012; 22: 301– 5. http://dx.doi.org/10.1093/eurpub/ckr007 Google Scholar CrossRef Search ADS PubMed  22 Hawkins NM, Jhund PS, McMurray JJ, Capewell S. Heart failure and socioeconomic status: accumulating evidence of inequality. Eur J Heart Fail  2012; 14: 138– 46. http://dx.doi.org/10.1093/eurjhf/hfr168 Google Scholar CrossRef Search ADS PubMed  23 Schroder SL, Richter M, Schroder J, et al.   Socioeconomic inequalities in access to treatment for coronary heart disease: a systematic review. Int J Cardiol  2016; 219: 70– 8. http://dx.doi.org/10.1016/j.ijcard.2016.05.066 Google Scholar CrossRef Search ADS PubMed  24 Sommer I, Griebler U, Mahlknecht P, et al.   Socioeconomic inequalities in non-communicable diseases and their risk factors: an overview of systematic reviews. BMC Public Health  2015; 15: 914. http://dx.doi.org/10.1186/s12889-015-2227-y Google Scholar CrossRef Search ADS PubMed  25 Ludt S, Wensing M, Szecsenyi J, et al.   Predictors of health-related quality of life in patients at risk for cardiovascular disease in European primary care. PLoS One  2011; 6: e29334. Google Scholar CrossRef Search ADS PubMed  26 Mielck A, Reitmeir P, Vogelmann M, Leidl R. Impact of educational level on health-related quality of life (HRQL): results from Germany based on the EuroQol 5D (EQ-5D). Eur. J Public Health  2013; 23: 45– 9. 27 Terraneo M. Inequities in health care utilization by people aged 50+: evidence from 12 European countries. Soc Sci Med  2015; 126: 154– 63. Google Scholar CrossRef Search ADS PubMed  28 Malta DC, Bernal RTI, Lima MG, et al.   Noncommunicable diseases and the use of health services: analysis of the National Health Survey in Brazil. Rev Saude Publica  2017; 51: 4s. Google Scholar PubMed  29 Paim J, Travassos C, Almeida C, et al.   The Brazilian health system: history, advances, and challenges. Lancet  2011; 377: 1778– 97. http://dx.doi.org/10.1016/S0140-6736(11)60054-8 Google Scholar CrossRef Search ADS PubMed  30 Holmes AM, Deb P. The effect of chronic illness on the psychological health of family members. J Ment Health Policy Econ  2003; 6: 13– 22. Google Scholar PubMed  31 Boing AC, Bertoldi AD, Peres KG. Socioeconomic inequalities in expenditures and income committed to the purchase of medicines in Southern Brazil. Rev Saude Publica  2011; 45: 897– 905. Google Scholar CrossRef Search ADS PubMed  32 Dray-Spira R. [Chronic diseases and employment: impact on social health inequalities]. Rev Epidemiol Sante Publique  2013; 61: S146– 51. Google Scholar CrossRef Search ADS PubMed  33 World Health Organization. Noncommunicable disease and poverty: the need for pro-poor strategies in the Western Pacific Region: a review . Geneva: WHO, 2006. 34 Bailey SR, Loveman M, Muniz JO. Measures of “Race” and the analysis of racial inequality in Brazil. Soc Sci Res  2013; 42: 106– 19. Google Scholar CrossRef Search ADS PubMed  35 Wamala S, Merlo J, Bostrom G, Hogstedt C. Perceived discrimination, socioeconomic disadvantage and refraining from seeking medical treatment in Sweden. J Epidemiol Community Health  2007; 61: 409– 15. http://dx.doi.org/10.1136/jech.2006.049999 Google Scholar CrossRef Search ADS PubMed  36 Diaz-Toro F, Verdejo HE, Castro PF. Socioeconomic inequalities in heart failure. Heart Fail Clin  2015; 11: 507– 13. http://dx.doi.org/10.1016/j.hfc.2015.07.012 Google Scholar CrossRef Search ADS PubMed  37 DE Hert M, Correll CU, Bobes J, et al.   Physical illness in patients with severe mental disorders. I. Prevalence, impact of medications and disparities in health care. World Psychiatry  2011; 10: 52– 77. http://dx.doi.org/10.1002/j.2051-5545.2011.tb00014.x Google Scholar CrossRef Search ADS PubMed  38 Lund C, Breen A, Flisher AJ, et al.   Poverty and common mental disorders in low and middle income countries: a systematic review. Soc Sci Med  2010; 71: 517– 28. http://dx.doi.org/10.1016/j.socscimed.2010.04.027 Google Scholar CrossRef Search ADS PubMed  39 Muggah E, Graves E, Bennett C, Manuel DG. Ascertainment of chronic diseases using population health data: a comparison of health administrative data and patient self-report. BMC Public Health  2013; 13: 16. http://dx.doi.org/10.1186/1471-2458-13-16 Google Scholar CrossRef Search ADS PubMed  40 Huerta JM, Tormo MJ, Egea-Caparros JM, et al.   Accuracy of self-reported diabetes, hypertension and hyperlipidemia in the adult Spanish population. DINO study findings. Rev Esp Cardiol  2009; 62: 143– 52. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2017. Published by Oxford University Press on behalf of the European Public Health Association. All rights reserved. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The European Journal of Public Health Oxford University Press

Chronic diseases and socioeconomic inequalities in quality of life among Brazilian adults: findings from a population-based study in Southern Brazil

Loading next page...
 
/lp/ou_press/chronic-diseases-and-socioeconomic-inequalities-in-quality-of-life-106LyuTD5m
Publisher
Oxford University Press
Copyright
© The Author(s) 2017. Published by Oxford University Press on behalf of the European Public Health Association. All rights reserved.
ISSN
1101-1262
eISSN
1464-360X
D.O.I.
10.1093/eurpub/ckx224
Publisher site
See Article on Publisher Site

Abstract

Abstract Background To evaluate the association between sociodemographic conditions and the quality of life (QoL) in adults and investigate whether these inequalities are greater among individuals with long-lasting chronic health conditions. Methods Cross-sectional analysis of the second wave (2012) of the EpiFloripa Study, a population-based cohort of 1720 adults living in Southern Brazil. QoL domains (physical, psychological, social relationships and environmental) were evaluated using the WHOQoL-BREF. Unadjusted and adjusted means of QoL according to socioeconomic and demographic variables were estimated and stratified by the presence of long-lasting chronic conditions (heart disease, stroke, diabetes, hypertension, chronic kidney disease, cirrhosis, tendinitis, arthritis, rheumatism and/or fibromyalgia) were peformed in 2016. Results Among 1222 interviewed adults (56.6% females, mean age 41.7 ± 11.4 years; follow-up rate 71.1%), the prevalence of 1+ long-lasting chronic disease was 37.3% (95%CI: 34.4–40.3). Their effect on the QoL was four times higher on the physical component (−9.6; 95%CI −12.1; −7.1) than on the other domains. Adults aged 40+ years with black skin colour or lower educational level had a lower physical QoL score only when any chronic disease was present. Among those with some chronic illness, the psychological domain was also lower among those aged 40+ years and with a lower family income. No interaction between sociodemographic variables and chronic diseases was observed for the other QoL domains. Conclusions The occurrence of long-lasting chronic diseases is associated with inequalities in QoL (physical and psychological domains), with stronger adverse effects among older adults, blacks and individuals with lower income or educational levels. Introduction Currently, chronic diseases account for 68% of all deaths worldwide,1 with 80% of them occurring in low-and-middle-income countries (LMIC), where they are fast replacing infectious diseases and malnutrition as the leading causes of disability and premature death.1,2 In Brazil, non-communicable chronic diseases (NCDs) also constitute a public health problem, accounting for 72% of all deaths, mainly among older individuals and those with lower income and schooling. NCDs are also responsible for decreased quality of life (QoL), higher degree of disability and impairment of daily life activities, which largely impact family, communities and society.2 As a result, there has been an increasing interest in the last decades for studies aiming to investigate the impact of specific diseases on QoL.3 According to the World Health Organization (WHO), QoL involves different relative dimensions, including individual perceptions, life position, culture and value systems about personal goals, expectations, living standards and concerns.4 This is a patient-centred outcome, deeply related to health conditions, used by policy decision-makers to understand the benefits for society of public health spending.5 NCDs are long-lasting and slowly progressive conditions, which can impact QoL in different ways: reducing well-being, restricting functional status, generating problems with obtaining and/or maintaining employment, affecting social relationships and involving psychosocial distress.6 At the same time, a lower QoL has a profound impact on functional capacity, adherence to treatment, disease severity and further complications among individuals with NCDs.1,6 Therefore, QoL has been suggested as a subjective instrument able to capture the impact of these illnesses on an individual’s life and evaluate their life-course progression.6,7 Furthermore, individuals belonging to socioeconomically disadvantaged groups tend to have a higher prevalence and/or severity of NCDs as a consequence of a higher frequency of unhealthy lifestyle, poor living and working conditions, limited access to health services, delays in the diagnosis of these conditions, and inappropriate knowledge of or compliance with disease management.2,8 Consequently, the coexistence of NCDs and socioeconomic disadvantage could interact to exacerbate QoL inequalities. Studies investigating this hypothesis are especially relevant for LMIC, considering the rapid increase in the prevalence of NCDs they are facing,1,2,9 the high levels of income inequalities10 and difficulties accessing health services.11 However, few studies have investigated the joint effect of socioeconomic disadvantage and occurrence of NCDs on QoL in LMIC, particularly studies using robust instruments to evaluate QoL and including population-based samples. Therefore, this paper aims to evaluate the association between sociodemographic variables and the QoL of adults living in Southern Brazil and investigate whether the presence of long-lasting chronic conditions increases QoL inequalities. Methods We analysed the second wave of the EpiFloripa Cohort study, a population-based cohort performed in Florianopolis (a state capital in Southern Brazil). In 2013, the city had an estimated population of 453 285 inhabitants (55% adults aged 20+ years) and a Human Development Index of 0.847 (the third highest in the country).12 Sample The baseline study conducted in 2009 included a representative sample of 1720 adults (20–59 years).The first follow-up occurred in 2012 when the QoL questions were incorporated into the study. Details about the methodology are available elsewhere.13 Briefly, the sampling process was performed in two stages: firstly, census tracts were systematically selected in each decile of household income (63/420), and, subsequently, the households (1134/16 755) were systematically selected. All adult residents in each household were considered eligible, except for those with some severe physical and/or neurological impairment that could affect their ability to answer the questions. Measures In 2012, all participants included in the baseline were traced. Household visits were phone scheduled or, when this failed, the interviewer directly visited the last available address of the participant. Personal data assistants were used in both waves to record participants’ data. Questionnaires were pre-tested, and a random sample of 15% of respondents in 2009 and 10% in 2012 answered a short-version of the original interview two weeks after the household interview. The κ value of these variables ranged from 0.6 (i.e. use of medicines) to 1.0 (age). The outcome was measured in 2012 using the Brazilian version of the short form of the World Health Organization QoL questionnaire (WHOQol-BREF).4 This instrument has excellent test–retest reliability, internal consistency and high construct and discriminant validity.4,14 TheWHOQol-BREF includes 26 questions, which provide information about four domains of QoL: physical (including daily living activities, energy, mobility, pain, sleep and work capacity); psychological (including body imagesatisfaction, feelings, self-esteem, personal beliefs, memory and concentration); social relationships (including personal relationships, social support and sexual activities); and environmental (including financial resources, safety and security, health/social care accessibility/quality, physical and home environment, recreation opportunities and transport).4 All domains were converted into discrete variables (0–100 scale), with higher values representing a better QoL.The WHOQol-BREF also includes a question regarding the respondents' QoL perception (very poor, poor, neither poor nor good, good or very good), which was analysed as a binary variable (good/very good QoL: yes or no). The exposure variables included demographic (gender, skin colour and age) and socioeconomic (schooling and monthly family income) characteristics. Self-reported skin colour was collected using the Brazilian Institute of Geography and Statistics classification (white, dark, black, Asiatic or indigenous).15 The age of the participants was collected as a numerical variable and divided into categories for analysis (20–29; 30–39; 40–49; 50+ years). In each wave, the monthly family income (1 USD = R$1.70 in 2009) was divided by the squared number of household members to obtain the equalised family income and then divided into tertiles (the first tertile representing a lower income). Schooling was measured as the total number of years successfully completed and then divided into categories (≤4; 5–8; 9–11; ≥12 years). The diagnosis of NCDs was self-reported and based on the question, ‘Has a physician or health professional told you that you have…’, followed by a list of conditions included in the Brazilian National Household Sample Survey.16 The long-lasting NCDs included in this study are (1) heart or cardiovascular disease; (2) stroke, cerebrovascular accident or brain ischemia; (3) diabetes mellitus; (4) systemic blood hypertension; (5) chronic kidney disease; (6) cirrhosis; (7) tendinitis; (8) arthritisor rheumatism and; (9) fibromyalgia. The number of affirmative answers for these conditions was summed up and subsequently transformed into a binary variable (presence of 1+ chronic conditions). Statistical analysis The statistical software Stata 13.0® (StataCorporation, College Station, United States) was used for all analyses. Characteristics of the sampling process, as well as sampling weights (probability of selection in 2009 and follow-up in 2012) were considered in all estimations. Mean and standard deviation or median and interquartile range (IQR) were used to describe continuous variables, depending on their symmetry. Absolute and relative frequencies were used for categorical variables. The QoL domain means were tested against sociodemographic variables through t-test or analysis of variance (ANOVA), as appropriate. Stratified analyses were performed according to the presence of any long-lasting NCDs. Linear regression models were used for crude and adjusted analyses. Wald tests for heterogeneity or trend were considered, depending on the nature of the independent variable. In multivariable analyses, demographic variables were first included in the models, and then the socioeconomic ones. An α of 5% was defined as indicative of statistical significance. In each model, marginal means of QoL in each category of the exposure variables were estimated. Interaction terms between the sociodemographic variables and the presence of long-lasting NCDs were included in the regression models and considered as indicative of effect modification when a P < 0.10 was obtained. The multicollinearity between the investigated sociodemographic variables was investigated considering the variance inflation factor (VIF) and the tolerance of the models. The EpiFloripa study was approved by the Ethics Committee on Human Research of the Federal University of Santa Catarina (no 351/08 and no 1772/11). All participants were informed about the objectives of the study and signed an informed consent form. Results In 2009, the mean age of the cohort was 37.5 ± 11.6 years (55.1% females), the median of the equalised family income was R$1443.4 (IQR R$866–2828; 1USD = R$1.70 in 2009), and the median of schooling was 11 years (IQR 9–15). The follow-up rate in 2012 was 71.1% (n = 1222; mean age 41.7 ± 11.4 years; 56.6% females). The only difference between the individuals located in 2012 and the original cohort was a lower proportion of individuals in the youngest age group in 2012 (Supplementary table S1). However, this difference was lower than 5% points. In 2009, the presence of at least one longstanding NCD was reported by 37.3% (95%CI: 34.4–40.3) of the sample, with 12.9% affected by two or more conditions. In decreasing order of frequency, the prevalence of these conditions were tendinitis (17.8%), hypertension (14.1%), arthritis or rheumatism (7.5%), heart disease (6.6%), diabetes mellitus (3.7%), fibromyalgia (2.6%), chronic kidney disease (2.3%), stroke (0.6%) and cirrhosis (0.1%). The QoL in 2012 was classified as good or very good by 86.3% of the respondents (95%CI: 81.0–92.2%), with no differences between those with or without a long-lasting NCD (87.8% and 84.9%, respectively). Mean QoL scores were 72.9 points (95%CI: 71.7–74.2) for the physical domain, 71.1 (CI95%: 70.2–72.1) for the psychological domain, 75.2 (95%CI: 74.2–76.3) for the social relationship domain, and 62.0 (95%CI: 60.5–63.6) for the environmental domain. The presence of at least one long-lasting NCD had a stronger association with the physical domain (mean difference −9.6; 95%CI: −12.1;−7.1) and a smaller but significant association with the psychological (−2.5; 95%CI: −4.4;−0.5) and environmental domains (−2.6; 95%CI: −4.4;−0.8).However, the association with the social domain was not significant (−2.2; 95%CI: −4.5; 0.1; data not shown in tables). Table 1 shows unadjusted associations between the sociodemographic variables and the four QoL domains, stratified by the presence of long-lasting NCDs. Women showed lower QoL than men in all domains, regardless of the presence of chronic conditions. Age was inversely associated with the physical and psychological domains of QoL only among those with long-lasting NCDs, while the environmental score was higher in the oldest age category only among those free of these illnesses. Table 1 Unadjusted means and standard errors (SE) of quality of life in adults (2012), according to sociodemographic variables (N = 1222)   Physical domain   Psychological domain   Social domain   Environmental domain   Long-lasting chronic diseaseb   Long-lasting chronic diseaseb   Long-lasting chronic diseaseb   Long-lasting chronic diseaseb   Variables (n)a  No   Yes   No   Yes   No   Yes   No   Yes   Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Gender  P<0.001*  P = 0.002*  P = 0.039*  P = 0.007*  P = 0.199*  P = 0.267*  P = 0.034*  P = 0.034*  Male (n = 518)  78.8(0.64)  70.3(1.35)  73.1(0.68)  72.1(0.87)  76.8(0.78)  72.9(1.16)  64.4(0.87)  62.1(1.09)  Female (n = 694)  74.8(0.95)  65.3(1.25)  71.3(0.79)  68.3(1.05)  75.4(0.88)  73.3(1.19)  61.8(1.23)  59.4(1.02)  Age  P = 0.765†  P = 0.002†  P = 0.450†  P = 0.005†  P = 0.381†  P = 0.055†  P = 0.005†  P = 0.528†  20–29 years (n = 325)  77.6(0.94)  76.4(2.17)  73.0(0.95)  74.1(1.91)  78.3(1.10)  73.8(2.75)  62.2(1.15)  60.4(2.89)  30–39 years (n = 276)  75.3(1.21)  70.4(1.96)  71.2(1.37)  72.3(1.62)  73.9(1.17)  74.6(1.63)  61.1(1.21)  60.4(2.08)  40-49 years (n = 344)  78.4(1.08)  65.3(1.65)  73.6(0.95)  68.4(1.15)  76.8(1.28)  72.8(2.03)  63.9(1.26)  58.5(1.30)  50+ years (n = 267)  76.0(1.34)  64.9(1.75)  70.9(1.13)  68.1(1.34)  75.6(1.29)  73.0(1.46)  65.6(1.48)  61.5(1.08)  Skin colour  P = 0.853**  P = 0.031**  P = 0.003**  P = 0.052**  P = 0.715**  P = 0.885**  P < 0.001**  P = 0.002**  White (n = 996)  76.7(0.70)  67.9(1.10)  72.9(0.59)  70.4(0.81)  76.3(0.63)  73.8(1.04)  64.0(1.00)  61.5(0.85)  Dark (n = 131)  77.2(1.72)  65.3(3.46)  68.7(1.50)  67.6(2.49)  74.7(1.80)  77.0(2.73)  60.6(1.48)  56.3(2.50)  Black (n = 59)  76.6(1.85)  60.5(3.94)  70.2(1.76)  66.1(3.10)  77.2(2.43)  71.4(2.80)  53.6(2.06)  52.6(3.25)  Family income (tertiles)c  P < 0.001†  P < 0.001†  P < 0.001†  P < 0.001†  P = 0.852†  P = 0.865†  P < 0.001†  P < 0.001†  First (<R$1429R$) (n = 409)  73.5(1.10)  60.1(1.68)  68.4(1.03)  64.3(1.39)  75.8(1.16)  72.9(2.14)  53.8(0.93)  52.3(1.35)  Second (1429–2887R$) (n = 387)  76.6(0.93)  70.1(1.46)  73.1(0.91)  70.9(0.95)  75.6(1.16)  75.6(1.33)  63.7(0.90)  61.6(0.93)  Third (>R$2, 887) (n = 398)  79.3(0.86)  71.4(1.37)  74.6(0.87)  74.0(0.93)  76.1(1.20)  73.2(1.19)  70.4(1.10)  68.0(1.04)  Schooling  P = 0.017†  P < 0.001†  P < 0.001†  P < 0.001†  P = 0.984†  P = 0.984†  P < 0.001†  P < 0.001†  0–4 years (n = 107)  71.8(3.12)  55.8(3.23)  63.9(1.66)  62.6(2.81)  78.2(3.14)  75.2(3.25)  54.3(2.36)  54.6(2.13)  5–8 years ( n = 172)  75.7(2.07)  61.0(2.49)  67.0(1.70)  63.9(2.06)  76.8(1.59)  72.3(2.62)  55.1(1.59)  52.7(1.67)  9–11 years (n = 392)  75.9(1.17)  68.9(1.43)  70.9(1.12)  70.7(1.13)  74.5(1.14)  74.1(1.16)  59.7(1.05)  59.5(1.10)  12+ years (n = 538)  78.2(0.66)  71.5(1.26)  75.4(0.58)  73.2(0.86)  76.8(0.80)  74.0(1.31)  68.5(0.79)  65.8(1.01)    Physical domain   Psychological domain   Social domain   Environmental domain   Long-lasting chronic diseaseb   Long-lasting chronic diseaseb   Long-lasting chronic diseaseb   Long-lasting chronic diseaseb   Variables (n)a  No   Yes   No   Yes   No   Yes   No   Yes   Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Gender  P<0.001*  P = 0.002*  P = 0.039*  P = 0.007*  P = 0.199*  P = 0.267*  P = 0.034*  P = 0.034*  Male (n = 518)  78.8(0.64)  70.3(1.35)  73.1(0.68)  72.1(0.87)  76.8(0.78)  72.9(1.16)  64.4(0.87)  62.1(1.09)  Female (n = 694)  74.8(0.95)  65.3(1.25)  71.3(0.79)  68.3(1.05)  75.4(0.88)  73.3(1.19)  61.8(1.23)  59.4(1.02)  Age  P = 0.765†  P = 0.002†  P = 0.450†  P = 0.005†  P = 0.381†  P = 0.055†  P = 0.005†  P = 0.528†  20–29 years (n = 325)  77.6(0.94)  76.4(2.17)  73.0(0.95)  74.1(1.91)  78.3(1.10)  73.8(2.75)  62.2(1.15)  60.4(2.89)  30–39 years (n = 276)  75.3(1.21)  70.4(1.96)  71.2(1.37)  72.3(1.62)  73.9(1.17)  74.6(1.63)  61.1(1.21)  60.4(2.08)  40-49 years (n = 344)  78.4(1.08)  65.3(1.65)  73.6(0.95)  68.4(1.15)  76.8(1.28)  72.8(2.03)  63.9(1.26)  58.5(1.30)  50+ years (n = 267)  76.0(1.34)  64.9(1.75)  70.9(1.13)  68.1(1.34)  75.6(1.29)  73.0(1.46)  65.6(1.48)  61.5(1.08)  Skin colour  P = 0.853**  P = 0.031**  P = 0.003**  P = 0.052**  P = 0.715**  P = 0.885**  P < 0.001**  P = 0.002**  White (n = 996)  76.7(0.70)  67.9(1.10)  72.9(0.59)  70.4(0.81)  76.3(0.63)  73.8(1.04)  64.0(1.00)  61.5(0.85)  Dark (n = 131)  77.2(1.72)  65.3(3.46)  68.7(1.50)  67.6(2.49)  74.7(1.80)  77.0(2.73)  60.6(1.48)  56.3(2.50)  Black (n = 59)  76.6(1.85)  60.5(3.94)  70.2(1.76)  66.1(3.10)  77.2(2.43)  71.4(2.80)  53.6(2.06)  52.6(3.25)  Family income (tertiles)c  P < 0.001†  P < 0.001†  P < 0.001†  P < 0.001†  P = 0.852†  P = 0.865†  P < 0.001†  P < 0.001†  First (<R$1429R$) (n = 409)  73.5(1.10)  60.1(1.68)  68.4(1.03)  64.3(1.39)  75.8(1.16)  72.9(2.14)  53.8(0.93)  52.3(1.35)  Second (1429–2887R$) (n = 387)  76.6(0.93)  70.1(1.46)  73.1(0.91)  70.9(0.95)  75.6(1.16)  75.6(1.33)  63.7(0.90)  61.6(0.93)  Third (>R$2, 887) (n = 398)  79.3(0.86)  71.4(1.37)  74.6(0.87)  74.0(0.93)  76.1(1.20)  73.2(1.19)  70.4(1.10)  68.0(1.04)  Schooling  P = 0.017†  P < 0.001†  P < 0.001†  P < 0.001†  P = 0.984†  P = 0.984†  P < 0.001†  P < 0.001†  0–4 years (n = 107)  71.8(3.12)  55.8(3.23)  63.9(1.66)  62.6(2.81)  78.2(3.14)  75.2(3.25)  54.3(2.36)  54.6(2.13)  5–8 years ( n = 172)  75.7(2.07)  61.0(2.49)  67.0(1.70)  63.9(2.06)  76.8(1.59)  72.3(2.62)  55.1(1.59)  52.7(1.67)  9–11 years (n = 392)  75.9(1.17)  68.9(1.43)  70.9(1.12)  70.7(1.13)  74.5(1.14)  74.1(1.16)  59.7(1.05)  59.5(1.10)  12+ years (n = 538)  78.2(0.66)  71.5(1.26)  75.4(0.58)  73.2(0.86)  76.8(0.80)  74.0(1.31)  68.5(0.79)  65.8(1.01)  a Number of observations per categories. b Positive for at least one of heart disease, stroke, diabetes, hypertension, chronic kidney disease, cirrhosis, tendinitis, arthritis, rheumatism, fibromyalgia. c Equalised, using the monthly family income collected in 2012. * t-test. ** ANOVA test for heterogeneity. † ANOVA test for trend. Table 1 Unadjusted means and standard errors (SE) of quality of life in adults (2012), according to sociodemographic variables (N = 1222)   Physical domain   Psychological domain   Social domain   Environmental domain   Long-lasting chronic diseaseb   Long-lasting chronic diseaseb   Long-lasting chronic diseaseb   Long-lasting chronic diseaseb   Variables (n)a  No   Yes   No   Yes   No   Yes   No   Yes   Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Gender  P<0.001*  P = 0.002*  P = 0.039*  P = 0.007*  P = 0.199*  P = 0.267*  P = 0.034*  P = 0.034*  Male (n = 518)  78.8(0.64)  70.3(1.35)  73.1(0.68)  72.1(0.87)  76.8(0.78)  72.9(1.16)  64.4(0.87)  62.1(1.09)  Female (n = 694)  74.8(0.95)  65.3(1.25)  71.3(0.79)  68.3(1.05)  75.4(0.88)  73.3(1.19)  61.8(1.23)  59.4(1.02)  Age  P = 0.765†  P = 0.002†  P = 0.450†  P = 0.005†  P = 0.381†  P = 0.055†  P = 0.005†  P = 0.528†  20–29 years (n = 325)  77.6(0.94)  76.4(2.17)  73.0(0.95)  74.1(1.91)  78.3(1.10)  73.8(2.75)  62.2(1.15)  60.4(2.89)  30–39 years (n = 276)  75.3(1.21)  70.4(1.96)  71.2(1.37)  72.3(1.62)  73.9(1.17)  74.6(1.63)  61.1(1.21)  60.4(2.08)  40-49 years (n = 344)  78.4(1.08)  65.3(1.65)  73.6(0.95)  68.4(1.15)  76.8(1.28)  72.8(2.03)  63.9(1.26)  58.5(1.30)  50+ years (n = 267)  76.0(1.34)  64.9(1.75)  70.9(1.13)  68.1(1.34)  75.6(1.29)  73.0(1.46)  65.6(1.48)  61.5(1.08)  Skin colour  P = 0.853**  P = 0.031**  P = 0.003**  P = 0.052**  P = 0.715**  P = 0.885**  P < 0.001**  P = 0.002**  White (n = 996)  76.7(0.70)  67.9(1.10)  72.9(0.59)  70.4(0.81)  76.3(0.63)  73.8(1.04)  64.0(1.00)  61.5(0.85)  Dark (n = 131)  77.2(1.72)  65.3(3.46)  68.7(1.50)  67.6(2.49)  74.7(1.80)  77.0(2.73)  60.6(1.48)  56.3(2.50)  Black (n = 59)  76.6(1.85)  60.5(3.94)  70.2(1.76)  66.1(3.10)  77.2(2.43)  71.4(2.80)  53.6(2.06)  52.6(3.25)  Family income (tertiles)c  P < 0.001†  P < 0.001†  P < 0.001†  P < 0.001†  P = 0.852†  P = 0.865†  P < 0.001†  P < 0.001†  First (<R$1429R$) (n = 409)  73.5(1.10)  60.1(1.68)  68.4(1.03)  64.3(1.39)  75.8(1.16)  72.9(2.14)  53.8(0.93)  52.3(1.35)  Second (1429–2887R$) (n = 387)  76.6(0.93)  70.1(1.46)  73.1(0.91)  70.9(0.95)  75.6(1.16)  75.6(1.33)  63.7(0.90)  61.6(0.93)  Third (>R$2, 887) (n = 398)  79.3(0.86)  71.4(1.37)  74.6(0.87)  74.0(0.93)  76.1(1.20)  73.2(1.19)  70.4(1.10)  68.0(1.04)  Schooling  P = 0.017†  P < 0.001†  P < 0.001†  P < 0.001†  P = 0.984†  P = 0.984†  P < 0.001†  P < 0.001†  0–4 years (n = 107)  71.8(3.12)  55.8(3.23)  63.9(1.66)  62.6(2.81)  78.2(3.14)  75.2(3.25)  54.3(2.36)  54.6(2.13)  5–8 years ( n = 172)  75.7(2.07)  61.0(2.49)  67.0(1.70)  63.9(2.06)  76.8(1.59)  72.3(2.62)  55.1(1.59)  52.7(1.67)  9–11 years (n = 392)  75.9(1.17)  68.9(1.43)  70.9(1.12)  70.7(1.13)  74.5(1.14)  74.1(1.16)  59.7(1.05)  59.5(1.10)  12+ years (n = 538)  78.2(0.66)  71.5(1.26)  75.4(0.58)  73.2(0.86)  76.8(0.80)  74.0(1.31)  68.5(0.79)  65.8(1.01)    Physical domain   Psychological domain   Social domain   Environmental domain   Long-lasting chronic diseaseb   Long-lasting chronic diseaseb   Long-lasting chronic diseaseb   Long-lasting chronic diseaseb   Variables (n)a  No   Yes   No   Yes   No   Yes   No   Yes   Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Mean (SE)  Gender  P<0.001*  P = 0.002*  P = 0.039*  P = 0.007*  P = 0.199*  P = 0.267*  P = 0.034*  P = 0.034*  Male (n = 518)  78.8(0.64)  70.3(1.35)  73.1(0.68)  72.1(0.87)  76.8(0.78)  72.9(1.16)  64.4(0.87)  62.1(1.09)  Female (n = 694)  74.8(0.95)  65.3(1.25)  71.3(0.79)  68.3(1.05)  75.4(0.88)  73.3(1.19)  61.8(1.23)  59.4(1.02)  Age  P = 0.765†  P = 0.002†  P = 0.450†  P = 0.005†  P = 0.381†  P = 0.055†  P = 0.005†  P = 0.528†  20–29 years (n = 325)  77.6(0.94)  76.4(2.17)  73.0(0.95)  74.1(1.91)  78.3(1.10)  73.8(2.75)  62.2(1.15)  60.4(2.89)  30–39 years (n = 276)  75.3(1.21)  70.4(1.96)  71.2(1.37)  72.3(1.62)  73.9(1.17)  74.6(1.63)  61.1(1.21)  60.4(2.08)  40-49 years (n = 344)  78.4(1.08)  65.3(1.65)  73.6(0.95)  68.4(1.15)  76.8(1.28)  72.8(2.03)  63.9(1.26)  58.5(1.30)  50+ years (n = 267)  76.0(1.34)  64.9(1.75)  70.9(1.13)  68.1(1.34)  75.6(1.29)  73.0(1.46)  65.6(1.48)  61.5(1.08)  Skin colour  P = 0.853**  P = 0.031**  P = 0.003**  P = 0.052**  P = 0.715**  P = 0.885**  P < 0.001**  P = 0.002**  White (n = 996)  76.7(0.70)  67.9(1.10)  72.9(0.59)  70.4(0.81)  76.3(0.63)  73.8(1.04)  64.0(1.00)  61.5(0.85)  Dark (n = 131)  77.2(1.72)  65.3(3.46)  68.7(1.50)  67.6(2.49)  74.7(1.80)  77.0(2.73)  60.6(1.48)  56.3(2.50)  Black (n = 59)  76.6(1.85)  60.5(3.94)  70.2(1.76)  66.1(3.10)  77.2(2.43)  71.4(2.80)  53.6(2.06)  52.6(3.25)  Family income (tertiles)c  P < 0.001†  P < 0.001†  P < 0.001†  P < 0.001†  P = 0.852†  P = 0.865†  P < 0.001†  P < 0.001†  First (<R$1429R$) (n = 409)  73.5(1.10)  60.1(1.68)  68.4(1.03)  64.3(1.39)  75.8(1.16)  72.9(2.14)  53.8(0.93)  52.3(1.35)  Second (1429–2887R$) (n = 387)  76.6(0.93)  70.1(1.46)  73.1(0.91)  70.9(0.95)  75.6(1.16)  75.6(1.33)  63.7(0.90)  61.6(0.93)  Third (>R$2, 887) (n = 398)  79.3(0.86)  71.4(1.37)  74.6(0.87)  74.0(0.93)  76.1(1.20)  73.2(1.19)  70.4(1.10)  68.0(1.04)  Schooling  P = 0.017†  P < 0.001†  P < 0.001†  P < 0.001†  P = 0.984†  P = 0.984†  P < 0.001†  P < 0.001†  0–4 years (n = 107)  71.8(3.12)  55.8(3.23)  63.9(1.66)  62.6(2.81)  78.2(3.14)  75.2(3.25)  54.3(2.36)  54.6(2.13)  5–8 years ( n = 172)  75.7(2.07)  61.0(2.49)  67.0(1.70)  63.9(2.06)  76.8(1.59)  72.3(2.62)  55.1(1.59)  52.7(1.67)  9–11 years (n = 392)  75.9(1.17)  68.9(1.43)  70.9(1.12)  70.7(1.13)  74.5(1.14)  74.1(1.16)  59.7(1.05)  59.5(1.10)  12+ years (n = 538)  78.2(0.66)  71.5(1.26)  75.4(0.58)  73.2(0.86)  76.8(0.80)  74.0(1.31)  68.5(0.79)  65.8(1.01)  a Number of observations per categories. b Positive for at least one of heart disease, stroke, diabetes, hypertension, chronic kidney disease, cirrhosis, tendinitis, arthritis, rheumatism, fibromyalgia. c Equalised, using the monthly family income collected in 2012. * t-test. ** ANOVA test for heterogeneity. † ANOVA test for trend. Independent of the presence of NCDs, black skin colour participants had lower psychological and environmental QoL scores than their white peers. Nevertheless, for the physical domain, the score was lower only among blacks compared to whites when affected by long-lasting NCDs. Family income and schooling showed a positive trend association with the physical, psychological, and environmental QoL scores. Although the same direct-trend association was observed in the healthy group, the differences in the physical and psychological scores between the extreme categories of income and schooling were at least twice as high among those with these diseases. Figures 1–3 show the predicted adjusted means (and their 95%CI) for each of the four QoL domains, according to sociodemographic variables, and stratified by the presence of long-lasting NCDs. Even after adjustment for confounding, most of the associations remained relatively stable. Except for the social component, all QoL domains were lower in females, regardless of whether the participants had a NCD or not (Supplementary figure S1). Age remained negatively associated with the physical and psychological domains of QoL only among those with a long-lasting NCD, while an inverse-trend association with the social domain became apparent in the same strata (figure 1). Even after adjustment for the other sociodemographic variables, black skin colour participants had a lower score in the physical domain than whites. This effect was evident only among those with some chronic illness (figure 2). Figure 1 View largeDownload slide Adjusted means and 95% confidence intervals of quality of life domain scores according to age groups adjusted for gender and skin colour, stratified by longstanding chronic disease status. P-values for interaction test: Physical domain: 0.140; Psychological domain: 0.005; Social relations domain: 0.246; Environmental domain: 0.520. **: Wald's test for trend. Figure 1 View largeDownload slide Adjusted means and 95% confidence intervals of quality of life domain scores according to age groups adjusted for gender and skin colour, stratified by longstanding chronic disease status. P-values for interaction test: Physical domain: 0.140; Psychological domain: 0.005; Social relations domain: 0.246; Environmental domain: 0.520. **: Wald's test for trend. Figure 2 View largeDownload slide Adjusted means and 95% confidence intervals of quality of life domain scores according to skin colour adjusted for gender and age groups, stratified by longstanding chronic disease status. P-values for interaction test: Physical domain: 0.233; Psychological domain: 0.772; Social relations domain: 0.118; Environmental domain: 0.748. *: Wald's test for heterogeneity. Figure 2 View largeDownload slide Adjusted means and 95% confidence intervals of quality of life domain scores according to skin colour adjusted for gender and age groups, stratified by longstanding chronic disease status. P-values for interaction test: Physical domain: 0.233; Psychological domain: 0.772; Social relations domain: 0.118; Environmental domain: 0.748. *: Wald's test for heterogeneity. Figure 3 View largeDownload slide Adjusted means and 95% confidence intervals of quality of life domain scores according to per capita family income adjusted for demographic and socioeconomic variables, stratified by longstanding chronic disease status. Florianopolis (SC), Brazil, 2012. P-values for interaction test: Physical domain: 0.140; Psychological domain: 0.005; Social relations domain: 0.510; Environmental domain: 0.481. **: Wald's test for trend. Figure 3 View largeDownload slide Adjusted means and 95% confidence intervals of quality of life domain scores according to per capita family income adjusted for demographic and socioeconomic variables, stratified by longstanding chronic disease status. Florianopolis (SC), Brazil, 2012. P-values for interaction test: Physical domain: 0.140; Psychological domain: 0.005; Social relations domain: 0.510; Environmental domain: 0.481. **: Wald's test for trend. Family income remained positively associated with the physical and environmental scores of QoL, with no evidence of heterogeneity according to the presence of NCDs. Nevertheless, the psychological domain was adversely affected by lower family income only among those with some chronic illness (figure 3). No evidence of multicollinearity was identified in the study (VIF between 1.00–1.39 and tolerance 0.72–0.99). Discussion Our original hypothesis of socioeconomic inequalities in QoL related to the presence of long-lasting chronic conditions was confirmed for the physical and psychological domains. Adults over the age of 40 years, with black skin colour, and with lower educational levels were more negatively associated with the physical domain of QoL than their peers only when a chronic condition was present. Furthermore, when affected by some chronic illness, individuals aged 40+ years or with a lower family income showed a reduced score in the psychological domain of QoL. According to our findings, long-lasting NCDs affected more than one-third of adults, which is consistent with the global estimates by the WHO.17 Several studies have demonstrated the impact of NCDs on the QoL, principally on the physical domain.18 Socioeconomic and demographic variables (lower educational level, unemployment and female gender) have been also associated with a lower QoL.3,19–21 There is also substantial evidence in the scientific literature suggesting that NCDs (CVD, hypertension, dyslipidaemia and diabetes) predominantly affect individuals from lower socioeconomic groups.22–24 The few studies investigating the joint effect of socioeconomic disadvantage and the presence of long-lasting health conditions as a source of QoL inequalities20,21,25corroborate our findings. However, they were conducted in Europe and used theEQ-5 D, an instrument that does not allow the separation of the effects of these conditions on the different QoL domains.7 In agreement with our findings, various authors have suggested that socioeconomically disadvantaged groups experience a ‘double suffering,’ characterised by a higher prevalence of NCDs and impaired QoL (especially the physical domain).20,26 The higher burden of NCDs among individuals with a lower socioeconomic status has been related to a higher frequency of unhealthy lifestyle, poor living and working conditions, difficulties in utilising health services, late diagnosis, inappropriate disease management, and higher frequency of complications.23,27 Nonetheless, the assessment of the double suffering requires to investigate whether the QoL differs across the same levels of morbidity, which is beyond the scope of this paper. In this sense, Brazilian health disparities remain despite all progress achieved by the Unified Brazilian Health System (SUS), especially when more than 50% of the population use private health services. These private plans are more affordable for the wealthiest, and they are related to a higher health service utilization.28,29 Furthermore, health expenses associated with illness management increase psychological distress,30 which is consistent with our findings of a lower psychological score of QoL among poor individuals affected by NCDs.The proportion of the family budget used for purchasing medicines is 2.7 times higher among the poorest than the wealthiest.31 Additionally, chronic illnesses are also associated with decreased workforce participation, early retirement, working limitations, absence due to sickness, and lower access/return to work,32 with a consequent impact on socioeconomic status and QoL.3 All these factors could explain the lower physical and psychological scores of QoL among adults aged 40+ years affected by NCDs, although these negative outcomes were already apparent at the age of 30–39 years. These individuals belong to economically active groups, which in turn leads to worse consequences from an economic perspective.32 Additionally, chronic illnesses are not only associated with socioeconomic inequalities, but they also aggravate QoL disparities as a consequence of the underlying vicious poverty-disease cycle.33 Black skin colour participants showed a reduced score in the physical domain only when affected by a NCD. Brazil has the largest population of African descendants among all Latin American countries, and racial discrimination is considered a foundation for Brazilian social inequalities.34 Perceived discrimination and socioeconomic disadvantage have been associated with refraining from seeking medical treatment, even after adjustment for confounders.35 Diverse patient-centred interventions have been proposed to improve QoL among people with NCDs, which can potentially reduce socioeconomic inequalities. They include the increase in the number of health consultations, a better assessment of risk factors, the encouragement of regular physical activity, and enhancing health literacy.3,25,36 However, further longitudinal studies would be necessary to elucidate the real benefit of these possible interventions. Finally, although the social and/or the environmental QoL domains were also lower among women, older individuals, those with lower income or schooling, or those with black skin colour, the presence of long-lasting NCDs did not modify the effect of these associations. Similar results have also been demonstrated in other studies.3,19–21 This study has some important strengths, such as the quality control of the interview-based assessments, the robust instrument used to investigate QoL, and the evaluation of a population-based sample of adults in a middle-income setting. One of the possible limitations is the percentage of losses of participants for follow-up. However, the individuals located in 2012 were similar to the original cohort according to most of the baseline characteristics. A second possible limitation is the cross-sectional analysis of the available data, which does not allow the evaluation of the temporality of the associations. Nevertheless, it could be assumed that the diagnosis of NCDs preceded the assessment of the QoL. Thirdly, mental health problems were not included in the list of chronic health conditions investigated in this study, and they are a potential source of bias37,38 to be considered in further studies, either as confounders or moderators of the associations.37,38 Finally, even though the NCDs were self-reported, some studies have demonstrated the excellent reliability of this information.39,40 Besides, it is unlikely that this information bias might explain our results, as underreporting these conditions would reduce the effect magnitude of the associations. This study showed that age, skin colour, income, and schooling interact with the occurrence of NCDs, reducing the physical and mental domains of QoL among adults. These results reinforce the relevance of developing health policies with a special focus on socioeconomically disadvantaged groups, considering the adverse effect of a reduced QoL for health management and disease progression.18 However, deciding between a universal care model and one targeting specific groups is one of the further challenges faced by the Brazil's health system.29 A multidisciplinary approach is necessary to reduce QoL inequalities, considering the increase in life expectancy and the prevalence of long-lasting NCDs. Acknowledgements This article is part of the post-doctoral studies of D.A. Höfelmann at the Post-Graduate Program in Nutrition at the Federal University of Santa Catarina, Florianopolis, Brazil. We would like to thank the Brazilian Institute of Geography and Statistics (IBGE) and the Florianópolis Health Authority staff for their useful help with the practical aspects of the EpiFloripa Cohort Study. Funding This paper is based on the EpiFloripa – Florianópolis Adult Health Study, Brazil. The Project was sponsored by the Brazilian National Council for Scientific and Technological Development (CNPq - Grant No. 485327/2007-4 and 508903/2010-6) and the Brazilian Coordination for the Improvement of Education Personnel (CAPES – Grant No. PVE-A020/2013). This research was developed by the Federal University of Santa Catarina, Brazil. D.A. Höfelmann received a post-doctoral scholarship from the Brazilian Coordination for the Improvement of Education Personnel (CAPES). Supplementary data Supplementary data are available at EURPUB online. Conflicts of interest: None declared Key points Chronic diseases impact more the physical than other domains of quality of life. Chronic diseases are associated with inequalities in physical and psychological QoL. The effects of chronic conditions are stronger among lower socioeconomic groups. References 1 World Health Organization. WHO Library Cataloguing-in-Publication Data Health in 2015: from MDGs, Millennium Development Goals to SDGs, Sustainable Development Goals . In: WHO, ed. Geneve: WHO, 2015. 2 Hunter DJ, Reddy KS. Noncommunicable diseases. N Engl J Med  2013; 369: 1336– 43. http://dx.doi.org/10.1056/NEJMra1109345 Google Scholar CrossRef Search ADS PubMed  3 Gonzalez-Chica DA, Mnisi Z, Avery J, et al.   Effect of health literacy on quality of life amongst patients with ischaemic heart disease in Australian general practice. PLoS One  2016; 11: e0151079. doi: 10.1371/journal.pone.0151079. eCollection 2016. Google Scholar CrossRef Search ADS PubMed  4 World Health Organization. WHOQOL-BREF: Introduction, Administration, Scoring and Generic Version of the Assessment . Geneve, Switzerland: WHO, 1996. 5 Fayers PM, Machin D. Quality of Life: Assessment, Analysis and Interpretation  West Sussex, England: John Wiley & Sons Ltd, 2000. 6 Megari K. Quality of life in chronic disease patients. Health Psychol Res  2013; 1: e27. Google Scholar CrossRef Search ADS PubMed  7 Mark DB. Assessing quality-of-life outcomes in cardiovascular clinical research. Nat Rev Cardiol  2016; 13: 286– 308. http://dx.doi.org/10.1038/nrcardio.2016.10 Google Scholar CrossRef Search ADS PubMed  8 Blank N, Diderichsen F. Social inequalities in the experience of illness in Sweden: a “double suffering”. Scand J Soc Med  1996; 24: 81– 9. Google Scholar CrossRef Search ADS PubMed  9 Malta DC, Stopa SR, Szwarcwald CL, et al.   A vigilância e o monitoramento das principais doenças crônicas não transmissíveis no Brasil - Pesquisa Nacional de Saúde, 2013. Revista Brasileira de Epidemiologia  2015; 18: 3– 16. Google Scholar CrossRef Search ADS PubMed  10 World Bank. GINI index (World Bank estimate) . World Bank, 2016. https://data.worldbank.org/indicator/SI.POV.GINI. PubMed PubMed  11 Boccolini CS, de Souza Junior PR. Inequities in Healthcare utilization: results of the Brazilian National Health Survey, 2013. Int J Equity Health  2016; 15: 150. http://dx.doi.org/10.1186/s12939-016-0444-3 Google Scholar CrossRef Search ADS PubMed  12 Brazilian Institute of Geographic and Statistics. Cidades. Rio de Janeiro: IBGE, 2010. 13 Boing AC, Peres KG, Boing AF, et al.   EpiFloripa Health Survey: the methodological and operational aspects behind the scenes. Rev Bras Epidemiol  2014; 17: 147– 62. http://dx.doi.org/10.1590/1415-790X201400010012ENG Google Scholar CrossRef Search ADS PubMed  14 Fleck MP, Louzada S, Xavier M, et al.   [Application of the Portuguese version of the abbreviated instrument of quality life WHOQOL-bref]. Rev Saude Publica  2000; 34: 178– 83. Google Scholar CrossRef Search ADS PubMed  15 Brazilian Institute of Geographic and Statistics. Censo Brasileiro 2010. IBGE, 2010. 16 Brazilian Institute of Geographic and Statistics. PNAD - Pesquisa Nacional por Amostra de Domicílios. Metodologia. Brasília: IBGE, 2016. 17 WHO. WHO maps noncommunicable disease trends in all countries: country profiles on noncommunicable disease trends in 193 countries. Cent Eur J Public Health  2011; 19: 130. 138. PubMed  18 Fortin M, Lapointe L, Hudon C, et al.   Multimorbidity and quality of life in primary care: a systematic review. Health Qual Life Outcomes  2004; 2: 51. http://dx.doi.org/10.1186/1477-7525-2-51 Google Scholar CrossRef Search ADS PubMed  19 Mello A. d C, Engstrom EM, Alves LC. Health-related and socio-demographic factors associated with frailty in the elderly: a systematic literature review. Cad Saude Publica  2014; 30: 1143– 68. http://dx.doi.org/10.1590/0102-311X00148213 Google Scholar CrossRef Search ADS PubMed  20 Mielck A, Vogelmann M, Leidl R. Health-related quality of life and socioeconomic status: inequalities among adults with a chronic disease. Health Qual Life Outcomes  2014; 12: 58. http://dx.doi.org/10.1186/1477-7525-12-58 Google Scholar CrossRef Search ADS PubMed  21 Stafford M, Soljak M, Pledge V, Mindell J. Socio-economic differences in the health-related quality of life impact of cardiovascular conditions. Eur J Public Health  2012; 22: 301– 5. http://dx.doi.org/10.1093/eurpub/ckr007 Google Scholar CrossRef Search ADS PubMed  22 Hawkins NM, Jhund PS, McMurray JJ, Capewell S. Heart failure and socioeconomic status: accumulating evidence of inequality. Eur J Heart Fail  2012; 14: 138– 46. http://dx.doi.org/10.1093/eurjhf/hfr168 Google Scholar CrossRef Search ADS PubMed  23 Schroder SL, Richter M, Schroder J, et al.   Socioeconomic inequalities in access to treatment for coronary heart disease: a systematic review. Int J Cardiol  2016; 219: 70– 8. http://dx.doi.org/10.1016/j.ijcard.2016.05.066 Google Scholar CrossRef Search ADS PubMed  24 Sommer I, Griebler U, Mahlknecht P, et al.   Socioeconomic inequalities in non-communicable diseases and their risk factors: an overview of systematic reviews. BMC Public Health  2015; 15: 914. http://dx.doi.org/10.1186/s12889-015-2227-y Google Scholar CrossRef Search ADS PubMed  25 Ludt S, Wensing M, Szecsenyi J, et al.   Predictors of health-related quality of life in patients at risk for cardiovascular disease in European primary care. PLoS One  2011; 6: e29334. Google Scholar CrossRef Search ADS PubMed  26 Mielck A, Reitmeir P, Vogelmann M, Leidl R. Impact of educational level on health-related quality of life (HRQL): results from Germany based on the EuroQol 5D (EQ-5D). Eur. J Public Health  2013; 23: 45– 9. 27 Terraneo M. Inequities in health care utilization by people aged 50+: evidence from 12 European countries. Soc Sci Med  2015; 126: 154– 63. Google Scholar CrossRef Search ADS PubMed  28 Malta DC, Bernal RTI, Lima MG, et al.   Noncommunicable diseases and the use of health services: analysis of the National Health Survey in Brazil. Rev Saude Publica  2017; 51: 4s. Google Scholar PubMed  29 Paim J, Travassos C, Almeida C, et al.   The Brazilian health system: history, advances, and challenges. Lancet  2011; 377: 1778– 97. http://dx.doi.org/10.1016/S0140-6736(11)60054-8 Google Scholar CrossRef Search ADS PubMed  30 Holmes AM, Deb P. The effect of chronic illness on the psychological health of family members. J Ment Health Policy Econ  2003; 6: 13– 22. Google Scholar PubMed  31 Boing AC, Bertoldi AD, Peres KG. Socioeconomic inequalities in expenditures and income committed to the purchase of medicines in Southern Brazil. Rev Saude Publica  2011; 45: 897– 905. Google Scholar CrossRef Search ADS PubMed  32 Dray-Spira R. [Chronic diseases and employment: impact on social health inequalities]. Rev Epidemiol Sante Publique  2013; 61: S146– 51. Google Scholar CrossRef Search ADS PubMed  33 World Health Organization. Noncommunicable disease and poverty: the need for pro-poor strategies in the Western Pacific Region: a review . Geneva: WHO, 2006. 34 Bailey SR, Loveman M, Muniz JO. Measures of “Race” and the analysis of racial inequality in Brazil. Soc Sci Res  2013; 42: 106– 19. Google Scholar CrossRef Search ADS PubMed  35 Wamala S, Merlo J, Bostrom G, Hogstedt C. Perceived discrimination, socioeconomic disadvantage and refraining from seeking medical treatment in Sweden. J Epidemiol Community Health  2007; 61: 409– 15. http://dx.doi.org/10.1136/jech.2006.049999 Google Scholar CrossRef Search ADS PubMed  36 Diaz-Toro F, Verdejo HE, Castro PF. Socioeconomic inequalities in heart failure. Heart Fail Clin  2015; 11: 507– 13. http://dx.doi.org/10.1016/j.hfc.2015.07.012 Google Scholar CrossRef Search ADS PubMed  37 DE Hert M, Correll CU, Bobes J, et al.   Physical illness in patients with severe mental disorders. I. Prevalence, impact of medications and disparities in health care. World Psychiatry  2011; 10: 52– 77. http://dx.doi.org/10.1002/j.2051-5545.2011.tb00014.x Google Scholar CrossRef Search ADS PubMed  38 Lund C, Breen A, Flisher AJ, et al.   Poverty and common mental disorders in low and middle income countries: a systematic review. Soc Sci Med  2010; 71: 517– 28. http://dx.doi.org/10.1016/j.socscimed.2010.04.027 Google Scholar CrossRef Search ADS PubMed  39 Muggah E, Graves E, Bennett C, Manuel DG. Ascertainment of chronic diseases using population health data: a comparison of health administrative data and patient self-report. BMC Public Health  2013; 13: 16. http://dx.doi.org/10.1186/1471-2458-13-16 Google Scholar CrossRef Search ADS PubMed  40 Huerta JM, Tormo MJ, Egea-Caparros JM, et al.   Accuracy of self-reported diabetes, hypertension and hyperlipidemia in the adult Spanish population. DINO study findings. Rev Esp Cardiol  2009; 62: 143– 52. Google Scholar CrossRef Search ADS PubMed  © The Author(s) 2017. Published by Oxford University Press on behalf of the European Public Health Association. All rights reserved.

Journal

The European Journal of Public HealthOxford University Press

Published: Dec 25, 2017

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off