Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

The association between diabetes and depressive symptoms varies by quality of diabetes care across Europe

The association between diabetes and depressive symptoms varies by quality of diabetes care... Downloaded from https://academic.oup.com/eurpub/article/28/5/872/4959186 by DeepDyve user on 13 July 2022 872 European Journal of Public Health Table A1 Gender inequality index, life expectancy at birth males and females and gender gap in life expectancy 2015 Life expectancy at birth 2015 Country GII 2015 Males Females Austria AT 0.078 78.8 83.7 4.9 Belgium BE 0.073 78.7 83.4 4.7 Bulgaria BG 0.223 71.2 78.2 7.0 Croatia HR 0.141 74.4 80.5 6.1 Cyprus CY 0.116 79.9 83.7 3.8 Czech Republic CZ 0.129 75.7 81.6 5.9 Denmark DK 0.041 78.8 82.7 3.9 Estonia EE 0.131 73.2 82.2 9.0 Finland FI 0.056 78.7 84.4 5.7 France FR 0.102 79.2 85.5 6.3 Germany DE 0.066 78.3 83.1 4.8 Greece GR 0.119 78.5 83.7 5.2 Hungary HU 0.252 72.3 79.0 6.7 Ireland IE 0.127 79.6 83.4 3.8 Italy IT 0.085 80.3 84.9 4.6 Latvia LV 0.191 69.7 79.5 9.8 Lithuania LT 0.121 69.2 79.7 10.5 Luxembourg LU 0.075 80.0 84.7 4.7 Malta MT 0.217 79.7 84.0 4.3 Netherlands NL 0.044 79.9 83.2 3.3 Poland PL 0.137 73.5 81.6 8.1 Portugal PT 0.091 78.1 84.3 6.2 Romania RO 0.339 71.5 78.7 7.2 Slovakia SK 0.179 73.1 80.2 7.1 Slovenia SI 0.053 77.8 83.9 6.1 Spain ES 0.081 80.1 85.8 5.7 Sweden SE 0.048 80.4 84.1 3.7 United Kingdom GB 0.131 79.2 82.8 3.6 GII: Gender inequality index. The European Journal of Public Health, Vol. 28, No. 5, 872–878 The Author(s) 2018. Published by Oxford University Press on behalf of the European Public Health Association. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/ 4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. doi:10.1093/eurpub/cky050 Advance Access published on 3 April 2018 ......................................................................................................... The association between diabetes and depressive symptoms varies by quality of diabetes care across Europe 1,2 3 3 Eva A. Graham , Katie H. Thomson , Clare L. Bambra 1 Department of Epidemiology, Biostatistics, and Occupational Health, Faculty of Medicine, McGill University, Montreal, QC, Canada 2 Douglas Mental Health University Institute, Montreal, QC, Canada 3 Institute of Health and Society, Faculty of Medical Sciences, Newcastle University, Newcastle upon Tyne, UK Correspondence: Eva A. Graham, Douglas University Mental Health Institute, Frank B. Common Pavilion, F-2116, 6875 Boul. LaSalle, Montreal, QC H4H 1R3, Canada, Tel: +1 5147616131 ext. 6140, e-mail: eva.graham@mail.mcgill.ca Background: Depressive symptoms are more common in adults with diabetes and may arise from the physical and psychosocial burden of disease. Better quality of diabetes care may be associated with a reduced disease burden and fewer depressive symptoms. Methods: This cross-sectional study included 34 420 participants from 19 countries in the European Social Survey Round 7 (2014–2015). Countries were grouped into quartiles based on their quality of diabetes care as measured in the Euro Diabetes Index 2014. Individual-level depressive symptoms were measured using the 8-item Center for Epidemiologic Studies—Depression Scale. Negative binomial regression was used to compare the number of depressive symptoms between adults with and without diabetes in each quartile of diabetes care quality. Analyses included adjustment for covariates and survey weights. Results: In countries with the highest quality of diabetes care, having diabetes was associated with only a 3% relative increase in depressive symptoms (95% CI 1.00–1.05). In countries in the second, third and fourth (lowest) quartiles of diabetes care quality, having diabetes was associated with a 13% (95% CI 1.08–1.17), 13% (1.08–1.19) and 22% (1.14–1.31) relative increase in depressive symptoms, respectively. Conclusion: The as- sociation between diabetes and depressive symptoms appears stronger in European countries with lower quality of diabetes care. Potential pathways for this association include the financial aspects of diabetes care, access to services and differential exposure to the social determinants of heath. Further research is needed to unpack these mechanisms and improve the quality of life of people with diabetes across Europe. ......................................................................................................... Downloaded from https://academic.oup.com/eurpub/article/28/5/872/4959186 by DeepDyve user on 13 July 2022 Diabetes and depressive symptoms across Europe 873 non-respondents. The mean response rate was 56% across all ESS7 Introduction countries and ranged from 31% in Germany to 69% in Lithuania. iabetes is highly prevalent in European adults, with a prevalence This study included all adult participants (age 18+) from ESS7 D of approximately 10.3% of men and 9.6% of women. countries except Estonia, as it did not collect information on Depression among people with diabetes is associated with poorer depressive symptoms, and Israel, as it is not located geographically diabetes control and treatment adherence, increased diabetes com- in Europe. All survey data and documentation are freely available at 3 4 plications, and a higher risk of mortality. Although it is generally www.europeansocialsurvey.org. accepted that depression or its symptoms are twice as prevalent 5,6 among people with diabetes, estimates vary across Europe. In a Measures UK study, 5.0% of participants with diabetes had possible cases of Self-reported depressive symptoms were measured using the 8-item depression compared with 3.8% without diabetes. In a large city in Center for Epidemiologic Studies—Depression Scale (CESD-8). the Netherlands, 29.7% of participants with diagnosed diabetes Depressive symptoms included feeling depressed, not feeling reported elevated depressive symptoms, in contrast to 19.4% of par- happy, feeling lonely, not enjoying life, feeling sad, feeling that ticipants with normal glucose levels. everything was an effort, feeling that you could not get going, and The relationship between diabetes and depression may be partially sleeping problems. Participants indicated how much of the time attributable to the physical and psychological burden of managing during the past week they endorsed each symptom on a 4-point diabetes and its complications. This burden may include financial scale (never, some of the time, most of the time or all or almost difficulties due to diabetes expenses, effects of social stigma or un- 10–12 all of the time). The CESD-8 supports a single factor structure, and employment as a result of ill health. People with diabetes may a final depressive symptom score was created by summing partici- also experience stress from the daily psychological burden of pants’ responses for each symptom. Possible scores ranged from 0 monitoring dietary intake, physical activity levels and blood sugar 9,13 (no symptoms) to 32 (all symptoms present nearly all the time). levels. The burden of disease hypothesis is supported by cross- Diabetes was measured in the ESS by asking participants whether sectional evidence that diagnosed diabetes is more strongly 8,14 they had or had experienced diabetes in the past 12 months. Self- associated with depression than undiagnosed diabetes. As well, reported diabetes has a high sensitivity in Europe when compared having diagnosed diabetes is associated with an approximate 27% with medical records, reported medication or repeated self-report of increased risk of new-onset depression compared with not having diabetes (sensitivity = 0.79). diabetes. Quality of diabetes care was measured using the Euro Diabetes Notably, the stress associated with diabetes may differ based on Index (EDI) 2014 published by Healthcare Consumer Powerhouse. quality of diabetes care. Better quality of care may reduce the The EDI was developed by an expert panel of clinicians and financial, social or psychological burden of diabetes and result in academics to compare European healthcare systems and services. reduced depressive symptoms. However, very little evidence is Indicators of quality of care included diabetes prevention initiatives, available to evaluate this hypothesis. One study conducted in the case finding and screening, range and reach of services (e.g. eye care, USA found that less availability or affordability of diabetes pharmaceuticals availability), access to treatment or care (e.g. patient medication and supplies was associated with increased diabetes- education, coverage of diabetes supplies), procedures available related psychological stress. Another study looked at differences (e.g. annual blood tests, eye exams) and diabetes outcomes. in depressive symptoms between people with diabetes living on Information for each indicator was obtained from national and either side of the USA/Mexico border, which have differing levels international surveys (e.g. WHO World Health Statistics), of quality of health care. No differences in depressive symptoms interviews with health care workers and public health experts, and were observed between groups. However, participants in this study national and regional diabetes registries, programs and reports. were selected from a convenience sample and diabetes care quality A total score for each country was calculated using a weighted was not measured in detail. Further cross-national research is warranted to better elucidate sum of these indicators. Scores ranged from 509 in Lithuania to 946 in Sweden, with a maximum possible score of 1000. associations between diabetes, depression and diabetes care quality in non-USA settings using large samples and robust measures of Importantly, a country’s relative position is more indicative of their quality of care than to their absolute score or ranking and quality of diabetes care. A large proportion of the research on depression and diabetes has been conducted in the USA and may therefore countries were divided into quartiles of quality of care for this analysis. not apply to European populations where healthcare systems generally provide more universal access, also termed higher levels Covariates included age in years, sex, education, marital status, current feeling about household income and employment (see of healthcare decommodification. This study aims to assess whether the association between diabetes and depressive symptoms table 1). Clinical covariates included the presence of heart/circula- tion problems (e.g. stroke) or high blood pressure in the past differs by quality of diabetes care across 19 European countries. It is hypothesized that this relationship will be weaker in countries with 12 months as well as body mass index (BMI) calculated from self- reported height and weight. The continuous covariates age and BMI higher standards of diabetes care. were modelled using fractional polynomial models to account for non-linearly, which select the best higher-order terms for each Methods covariate (e.g. age squared). Self-reported lifestyle covariates including smoking status, alcohol consumption and engagement in Data physical activity. Data come from the European Social Survey (ESS), a repeated cross- sectional survey of social attitudes and values in 21 countries across 18,19 Sample size Europe. This analysis used Round 7 of the ESS (ESS7), as this was the first time detailed information was collected on health A total of 34 467 adults participated in ESS7 from the 19 studies conditions, lifestyle behaviours and the social determinants of included in this analysis. Those with a self-reported height of 110 cm health. Each country in ESS7 used random probability sampling or below were excluded, as this may be due to measurement error to select a representative sample of individuals aged 15 and over in and obesity measures for these participants were extremely high private households, though specific sampling methods varied by (n = 17). Of the remaining sample, 91% had complete information country. Sampling was completed between August 2014 and on all variables. Most participants with incomplete information were December 2015 and several attempts were made to contact initial missing BMI values (n = 1113), presence of heart/circulation Downloaded from https://academic.oup.com/eurpub/article/28/5/872/4959186 by DeepDyve user on 13 July 2022 874 European Journal of Public Health Table 1 Demographic, clinical and lifestyle characteristics of participants in the European social survey 7 by quartile of diabetes care quality Quartiles of quality of Quartile 1 Quartile 2 Quartile 3 Quartile 4 diabetes care (highest) (lowest) N = 8719 N = 9362 N = 8211 N = 8128 Countries in order of quality of diabetes care Sweden, Slovenia, Austria, Ireland, Netherlands, Norway, Belgium, Czech Denmark, France, Portugal, Republic, Switzerland, Germany, Hungary, Poland, United Kingdom Finland Spain Lithuania Age mean (SD) 47.4 (0.30) 49.2 (0.32) 47.5 (0.29) 44.4 (0.32) % Female 51.6 51.9 51.9 52.4 Marital status Married/civil partnership/cohabitating 53.0 56.0 49.6 56.9 Divorced/separated/widowed 15.3 15.6 16.6 13.7 Never Married 31.7 28.4 33.8 29.4 Main activity Employed, retired, housework, other 91.3 91.1 86.9 92.3 Unemployed 4.9 6.3 10.0 6.7 Sick/disabled 3.7 2.6 3.0 0.9 Education Primary or lower secondary 30.3 30.4 55.2 35.6 Secondary 45.5 56.8 30.3 46.2 Tertiary 24.1 12.8 14.4 18.1 Income feeling Living comfortably 45.1 32.4 23.8 12.3 Coping on present income 39.6 50.9 46.7 61.5 Difficult on present income 11.8 13.5 21.7 22.3 Very difficult on present income 3.5 3.1 7.8 3.9 BMI mean(SD) 25.8 (0.08) 25.8 (0.09) 25.7 (0.07) 25.7 (1.00) % Heart/circulation problems or high blood pressure in past 12 months 3.9 6.1 5.0 7.1 % Diabetes in past 12 months 5.0 6.4 5.3 4.2 Total depressive symptoms mean(SD) 4.9 (0.06) 5.4 (0.07) 6.1 (0.07) 5.4 (0.09) Smoking Current 22.5 29.3 28.6 28.3 Former 25.4 23.6 19.9 18.9 Never 52.1 47.1 51.4 52.7 Alcohol use in past 12 months Never 17.3 16.1 27.6 21.7 Once a month or less 20.4 22.5 19.7 34.5 Once a week or less 31.2 34.1 26.3 31.9 Every day or several times a week 31.0 27.3 26.4 11.8 Number of past 7 days did sports or other physical activity mean(SD) 3.4 (0.04) 3.1 (0.05) 2.8 (0.04) 3.0 (0.05) problems (n = 594) or physical activity levels (n = 383). Missing incorporated clustering by country as well as variance from demographic, clinical and lifestyle variables were imputed using multiply imputed datasets. Sampling weights were used in all ordered logistic regression (education level, satisfaction with analyses to account for (i) selection probabilities of each participant current income, number of days of physical activity), multinomial based on each country’s sampling design and (ii) sampling error and logistic regression (sex, presence of heart problems or diabetes, non-response based on age, gender, education and region. marital status, smoking status, alcohol consumption) or linear Population weights were applied to account for differences in the regression (age, BMI) regression with all other variables as population size of each country. All analyses were conducted using predictors. As well, approximately 2.2% of the sample had missing Stata version 12 and used a significance level of 0.05. information on at least one depressive symptom (n = 752). Depressive symptoms were imputed using ordered logistic Results regression on all other depressive symptoms as well as the demo- graphic and clinical characteristics listed above. Twenty complete The highest quality of diabetes care was observed in Nordic imputed datasets were created. The final sample included34 420 countries, while lower quality of care was more prevalent in participants who had complete information following multiple Eastern Europe (table 1). This is consistent with trends observed imputation. in overall healthcare in European countries in the same year, albeit with some discrepancies. Country rankings are also Statistical analysis somewhat consistent with Bohm et al.’s classification of health care systems based on health care regulation, financing and service The outcome of total number of depressive symptoms was modelled provision. Countries with National Health Service or National using negative binomial regression in order to reflect the positively skewed distribution of depressive symptoms in the European Health Insurance systems (healthcare regulation and financing by the state) include Denmark, Finland, Norway, Sweden, Portugal, population. Presence of diabetes, diabetes quality of care quartile and an interaction term were included as predictors. A non-zero Spain, the UK and Ireland. While the majority of these countries interaction term would indicate that the association between rank at top of quality of diabetes care, Spain, Portugal and Ireland diabetes and depression varies depending on the quartile of have lower quality of diabetes care. Countries with Social Health diabetes care quality. Sensitivity analyses examined the same associ- Insurance systems, or regulation and financing by non-governmen- ations by sex and education level. Standard error estimates tal organizations, include Austria, Germany, Switzerland and Downloaded from https://academic.oup.com/eurpub/article/28/5/872/4959186 by DeepDyve user on 13 July 2022 Diabetes and depressive symptoms across Europe 875 Table 2 Rate ratio of diabetes on depressive symptoms by diabetes quartile Quality of diabetes care Model 1 Model 2 Model 3 Quartile 1 1.19 1.16–1.22 1.05 1.03–1.09 1.03 1.00–1.05 a a a Quartile 2 1.37 1.33–1.41 1.14 1.11–1.17 1.13 1.08–1.17 a a a Quartile 3 1.33 1.25–1.42 1.16 1.10–1.21 1.13 1.08–1.19 a,c a,b a,b Quartile 4 1.53 1.36–1.71 1.26 1.15–1.39 1.22 1.14–1.31 Rate ratios estimated from a negative binomial regression model of diabetes, quartile of diabetes care and their interaction on number of depressive symptoms. Model 1: unadjusted. Model 2: adjusted for individual differences in age, education, education  diabetes care quartile, satisfaction with income, gender, marital status, employment status, BMI and heart/circulation problems or high blood pressure. Model 3: additionally adjusted for smoking, alcohol use, alcohol use  diabetes care quartile and number of days doing physical activity. a: Significantly different from Quartile 1. b: Significantly different from Quartile 2. c: Significantly different from Quartile 3. Figure 1 Percentage (%) increase in number of depressive symptoms in people with diabetes versus without diabetes. Estimates adjusted for age, sex, education level, marital status, current feeling about household income, employment status, presence of heart/circulation problems or high blood pressure, body mass index, smoking status, alcohol consumption and physical activity Slovenia. These countries generally had quality of care rankings in activity was highest in the first quartile and lowest in the third. the second or third quartiles. Belgium, France, the Czech Republic, BMI did not differ between quartiles. Patterns of heart or circulation Hungary, the Netherlands and Poland have a mix of state, social and problems, diabetes and mean number of depressive symptoms private healthcare regulation, financing and provision. Countries varied by quartile with no clear trend. Results from the negative binomial regression model are shown in with this system were found in all quartiles of diabetes care. Country-specific scores are presented in Supplementary table S1. table 2 as the ratio of depressive symptoms between people with Table 1 presents differences in demographic, lifestyle and clinical diabetes compared with those without diabetes by quartile of factors between quartiles. The mean age of participants varied from diabetes care quality. Having diabetes was associated with 44.2 years (quartile 4)–49.2 years (quartile 2). There were approxi- increased depressive symptoms across all quartiles. Adjustment mately equal proportions of men and women in each quartile. Over for demographic, clinical and lifestyle characteristics attenuated half of participants in all quartiles were married, in a partnerships or these associations, though they were still statistically significant. cohabitating. Under 10% of the population in quartiles 1, 2 and 4 The association between diabetes and depressive symptoms were unemployed or permanently sick/disabled, while nearly 13% of increased as quality of diabetes care decreased. In the fully participants in the 3rd quartile were unemployed. Education levels adjusted model, statistically significant differences in depressive were highest in the first and second quartiles and lowest in the third. symptoms associated with diabetes are observed between the first Satisfaction with income followed a decreasing trend by quartile of (highest) quartile of diabetes care and all other quartiles and between care. Smoking trends were similar between quartiles, with 22 to 28% the second and fourth quartiles (figure 1). Figure 2 presents the main analysis stratified by sex. Men with of the population as current smokers. Frequency of alcohol con- sumption decreased from higher to lower quartiles, and physical diabetes in countries in the highest quartile of diabetes care showed Downloaded from https://academic.oup.com/eurpub/article/28/5/872/4959186 by DeepDyve user on 13 July 2022 876 European Journal of Public Health Figure 2 Percentage (%) increase in number of depressive symptoms in people with diabetes versus without diabetes by sex. Estimates adjusted for age, education level, marital status, current feeling about household income, employment status, presence of heart/circulation problems or high blood pressure, body mass index, smoking status, alcohol consumption and physical activity 6,14 no increased depressive symptoms, but higher depressive symptom associations. This may be partially attributable to women’s scores were observed in men with diabetes in the second, third and greater use of diabetes-related healthcare services, higher fourth quartiles. In women, diabetes was associated with increased adherence to glucose monitoring and selfcare, or increased social 30–33 depressive symptoms even in countries with the highest quality of support. However, in countries with the highest quality of diabetes care, but differences between quartiles were not statistically care, diabetes was not associated with increased depressive significant. A similar pattern of increasing depressive symptoms symptoms in men but there was an association in women. This among people with diabetes in lower quartiles of diabetes care was may allude to an added burden of managing diabetes in addition observed in each level of education (Supplementary figure S1). to childcare and family responsibilities, which are largely the role of women in all European countries. Women also report more health conditions and diabetes complications than men, such as poorer Discussion physical functioning, higher cholesterol levels, and more diabetes- The results of this analysis suggest that diabetes is more strongly related complications, which may contribute to increased depressive 30,35 related to depressive symptoms in European countries with poorer symptoms. Similar to the main results, analyses stratified by quality of diabetes care. While diabetes was associated with an educational status suggest that lower quality of diabetes care is increased rate of depressive symptoms across all levels of diabetes associated with higher depressive symptoms across all educational care quality, this association was weakest in countries with the levels. As well, higher educational levels appear to be associated with highest quality of care (RR = 1.03, 95% CI 1.00–1.05) and fewer depressive symptoms in diabetes in countries with higher strongest is countries with the poorest quality of diabetes care levels of quality of care, consistent with prior research. This may (RR = 1.22, 95% CI 1.14–1.31). These results suggest that be due to more positive attitudes towards diabetes and better coping improved quality of diabetes care may reduce some of the psycho- strategies among those with more education. logical burden associated with living with diabetes. Of course, quality of care is only possible pathway linking diabetes These results are consistent with interventions that have reported with an increased risk of depression. Our findings might therefore improved depression scores in participants who received increased also reflect wider issues within the welfare or health care systems or diabetes monitoring, education or prevention. One trial showed that indeed other individual factors such as the social stigma related to people with Type 2 diabetes who used a glucose-monitoring device, diabetes. However, by controlling for lifestyle factors and socio- kept a diary of glucose levels and intake, and received counselling demographics, we hope to have taken into account some of these experienced an improvement in depressive symptoms. Those who wider issues and isolated the role of quality of care. only received general counselling about their diet and lifestyle did not. Another trial showed that the implementation of a diabetes prevention program was associated with a decrease in depressive Strengths symptoms in the intervention group compared with the control This is the first nationally comparative study of depressive symptoms group. The authors of both trials suggest that decreased and diabetes across Europe and provides novel evidence on quality depressive symptoms may be related to a heightened sense of 27,28 of diabetes care and psychological symptoms associated with control over diabetes. However, a meta-analysis of web-based diabetes. This analysis used a large, representative dataset of partici- interventions that promoted better diabetes care did not show that pants in 19 European countries. The ESS7 ensured representative- such interventions were associated with decreased depression. ness though methods such as strict probability sampling and re- Further analyses in this study examined associations between contacting initial non-respondents. As well, survey weighting and diabetes and depressive symptoms comparing men and women. In countries with poorer quality of care, men showed a stronger asso- multiple imputation of missing variables were used in this analysis to minimize potential bias from selection of participants and non- ciation between diabetes and depression than women. This is consistent with previous meta-analyses that report sex-specific response. Due to the number of countries included, comparisons Downloaded from https://academic.oup.com/eurpub/article/28/5/872/4959186 by DeepDyve user on 13 July 2022 Diabetes and depressive symptoms across Europe 877 were possible between different levels of quality of diabetes care in this paper do not necessarily represent those of the funders or using an objective, international measure. UKCRC. The funders had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript. Limitations Conflicts of interest: None declared. Limitations of this study include uncertainty in the temporal order of diabetes and depressive symptoms, as the dataset was cross- sectional. Consequently, causality cannot be inferred. Unmeasured Key points confounding may be present due to differences in mental health care across Europe. However, there are currently no tools available that  Diabetes is associated with increased depressive symptoms allow a detailed comparative assessment of mental health care across across Europe Europe. Unmeasured confounding may also be present on an  An increased association between diabetes and depressive individual level from diabetes complications that were not symptoms was observed in countries with lower quality of measured the European Social Survey. This analysis was also diabetes care compared with those with higher quality of unable to differentiate between Type 1 and Type 2 diabetes, diabetes care Higher quality of diabetes care may reduce the burden or although elevated depressive symptoms are associated with both types of diabetes in European populations. stress of disease and is associated with fewer depressive symptoms among adults with diabetes Diagnosed diabetes was assessed using self-reported measures and may have been underestimated. Yet validation studies suggest that self-reported diagnosed diabetes has high concordance with 39,40 physician records or antidiabetic drug use. Furthermore, mis- References classification of diabetes would bias the association between diabetes and depressive symptoms towards the null. 1 WHO Regional Office for Europe. Diabetes: Data and statstics. World Health Finally, quality of diabetes care may differ substantially within a Organization, Regional Office for Europe. Available at: http://www.euro.who.int/en/ country, and the measure used in this study does not capture health-topics/noncommunicable-diseases/diabetes/data-and-statistics (10 regional or local granularity. The authors of the Euro Diabetes November 2016, date last accessed). Index acknowledge this limitation, but suggest that this index may 2 Lustman PJ, Clouse RE. Depression in diabetic patients: the relationship between nonetheless be meaningful when interpreted with caution. Further mood and glycemic control. J Diabetes Complications 2005; 19:113–22. research should consider more local measures of quality of diabetes 3 de Groot M, Anderson R, Freedland KE, et al. Association of depression and care. diabetes complications: a meta-analysis. Psychosom Med 2001; 63:619–30. 4 Park M, Katon WJ, Wolf FM. Depression and risk of mortality in individuals with Conclusions diabetes: a meta-analysis and systematic review. Gen Hosp Psychiatry 2013; 35:217– Results suggest an inverse association between national quality of 5 Nouwen A, Winkley K, Twisk J, et al. Type 2 diabetes mellitus as a risk factor for the diabetes care and depressive symptoms among people with diabetes. onset of depression: a systematic review and meta-analysis. Diabetologia 2010; In countries with high quality of diabetes care, diabetes was 53:2480–6. associated with a small increase in depressive symptoms while 6 Ali S, Stone MA, Peters JL, et al. The prevalence of co-morbid depression in adults countries with poorer quality of care showed a larger increase in with Type 2 diabetes: a systematic review and meta-analysis. Diabet Med 2006; depressive symptoms among people with diabetes. This association 23:1165–73. persisted when adjusted for individual demographic, clinical and 7 Holt RI, Phillips DI, Jameson KA, et al. The relationship between depression and behavioural characteristics. Quality of diabetes care may therefore diabetes mellitus: findings from the Hertfordshire Cohort Study. Diabetic Med J Br be important not only for physical complications of diabetes but also Diabetic Assoc 2009; 26:641–8. for depressive symptoms and psychosocial outcomes. 8 Knol MJ, Heerdink ER, Egberts AC, et al. Depressive symptoms in subjects with diagnosed and undiagnosed type 2 diabetes. Psychosom Med 2007; 69:300–5. Supplementary data 9 Hasan SS, Mamun AA, Clavarino AM, Kairuz T. Incidence and risk of depression associated with diabetes in adults: evidence from longitudinal studies. Community Supplementary data are available at EURPUB online. Ment Health J 2015; 51:204–10. 10 Schabert J, Browne JL, Mosely K, Speight J. Social stigma in diabetes: a framework to Funding understand a growing problem for an increasing epidemic. The Patient 2013; 6:1–10. 11 Rumball-Smith J, Barthold D, Nandi A, Heymann J. Diabetes associated with early EG received a scholarship to complete this research from the Queen labor-force exit: a comparison of sixteen high-income countries. Health Aff Elizabeth Scholars Program, funded by Universities Canada, (Millwood) 2014; 33:110–5. Community Foundations of Canada and the Rideau Hall 12 Jeon YH, Essue B, Jan S, et al. Economic hardship associated with managing chronic Foundation, in collaboration with the Institute for Health and illness: a qualitative inquiry. BMC Health Serv Res 2009; 9:182. Social Policy, McGill University. CB and KT are part of the 13 van Dooren FE, Denollet J, Verhey FR, et al. Psychological and personality factors in HiNEWS project—Health Inequalities in European Welfare type 2 diabetes mellitus, presenting the rationale and exploratory results from The States—funded by NORFACE (New Opportunities for Research Maastricht Study, a population-based cohort study. BMC Psychiatry 2016; 16:17. Funding Agency Cooperation in Europe) Welfare State Futures 14 Chen S, Zhang Q, Dai G, et al. Association of depression with pre-diabetes, un- programme (grant reference: 462-14-110). For more details on diagnosed diabetes, and previously diagnosed diabetes: a meta-analysis. Endocrine NORFACE, see http://www.norface.net/11. CB is a member of 2016; 53:35–46. Fuse, the Centre for Translational Research in Public Health. 15 Peyrot M, Egede LE, Campos C, et al. Ethnic differences in psychological outcomes Funding for Fuse comes from the British Heart Foundation, among people with diabetes: USA results from the second Diabetes Attitudes, Cancer Research UK, Economic and Social Research Council, Wishes, and Needs (DAWN2) study. Curr Med Res Opin 2014; 30:2241–54. Medical Research Council, the National Institute for Health Research, under the auspices of the UK Clinical Research 16 Mier N, Bocanegra-Alonso A, Zhan D, et al. Clinical depressive symptoms and Collaboration, and is gratefully acknowledged. The views expressed diabetes in a binational border population. J Am Board Fam Med 2008; 21:223–33. Downloaded from https://academic.oup.com/eurpub/article/28/5/872/4959186 by DeepDyve user on 13 July 2022 878 European Journal of Public Health 17 Bambra C. Cash Versus Services: ‘Worlds of Welfare’ and the Decommodification of 29 Pal K, Eastwood SV, Michie S, et al. Computer-based diabetes self-management Cash Benefits and Health Care Services. J Soc Policy 2005; 34:195–213. interventions for adults with type 2 diabetes mellitus. Cochrane Datab Syst Rev 2013;Cd008776. 18 The ESS Data Archive. ESS7-2014 Documentation Report. Bergen: Norwegian Centre for Reserach Data, 2015. Available at: http://www.europeansocialsurvey.org/ 30 Rossi MC, Lucisano G, Pintaudi B, et al. The complex interplay between clinical and docs/round7/survey/ESS7_data_documentation_report_e03_0.pdf (11 November person-centered diabetes outcomes in the two genders. Health Qual Life Outcomes 2016, date last accessed). 2017; 15:41. 19 Eikemo TA, Bambra C, Huijts T, Fitzgerald R. The first pan-European sociological 31 Yu MK, Lyles CR, Bent-Shaw LA, Young BA. Sex disparities in diabetes process health inequalities survey of the general population: the European Social Survey of care measures and self-care in high-risk patients. J Diabetes Res 2013; rotating module on the social determinants of health. Eur Sociol Rev 2017; 2013:575814. 33:137–53. 32 Kramer HU, Ruter G, Schottker B, et al. Gender differences in healthcare utilization 20 ESS7 – 2014 Fieldwork Summary and Deviations. Available from: http://www. of patients with diabetes. Am J Manag Care 2012; 18:362–9. europeansocialsurvey.org/data/deviations_7.html (1 Februray 2018, last accessed 33 Mathew R, Gucciardi E, De Melo M, Barata P. Self-management experiences among date). men and women with type 2 diabetes mellitus: a qualitative analysis. BMC Fam 21 Karim J, Weisz R, Bibi Z, Rehman S. Validation of the eight-item Center for Pract 2012; 13:122. Epidemiologic Studies Depression Scale (CES-D) among older adults. Curr Psychol 34 Aliaga C. How is the time of women and men distributed in Europe? Eurostat, 2006 Res Rev 2015; 34:681–92. Contract No.: KS-NK-06-004. Available at: http://ec.europa.eu/eurostat/documents/ 22 Sluik D, Boeing H, Montonen J, et al. Associations between general and abdominal 3433488/5438881/KS-NK-06-004-EN.PDF/7af4b8e8-1d40-478c-8410-e41204e0b29c adiposity and mortality in individuals with diabetes mellitus. Am J Epidemiol 2011; (15 May 2017, date last accessed). 174:22–34. 35 Kautzky-Willer A, Harreiter J, Pacini G. Sex and gender differences in risk, patho- 23 Cebolla Garrofe ´ B, Bjo ¨ rnberg A, Yung Phang A. Euro Diabetes Index. Health physiology and complications of type 2 diabetes mellitus. Endocr Rev 2016; Consumer Powerhouse, 2014. Available at: http://www.healthpowerhouse.com/files/ 37:278–316. EDI-2014/EDI-2014-report.pdf (8 November 2016, date last accessed). 36 Gois CJ, Ferro AC, Santos AL, et al. Psychological adjustment to diabetes mellitus: 24 European Social Survey. Weighting European Social Survey Data. April 24 2014. highlighting self-integration and self-regulation. Acta Diabetol 2012; 49:S33–40. Report No. Available at: https://www.europeansocialsurvey.org/docs/methodology/ 37 Katon W, von Korff M, Ciechanowski P, et al. Behavioral and clinical factors ESS_weighting_data_1.pdf (16 November 2016, date last accessed). associated with depression among individuals with diabetes. Diabetes Care 2004; 25 Bjo ¨ rnberg A. Euro Health Consumer Index. Health Consumer Powerhouse, 2014. 27:914–20. Available at: http://www.healthpowerhouse.com/files/EHCI_2014/EHCI_2014_ 38 Pouwer F, Geelhoed-Duijvestijn PH, Tack CJ, et al. Prevalence of comorbid report.pdf (2 May 2017, date last accessed). depression is high in out-patients with Type 1 or Type 2 diabetes mellitus. Results 26 Bohm K, Schmid A, Gotze R, et al. Classifying OECD Healthcare Systems: A from three out-patient clinics in the Netherlands. Diabetic Med J Br Diabetic Assoc Deductive Approach. 2012. No. 165. Available at: https://www.econstor.eu/ 2010;27:217–24. bitstream/10419/64809/1/726547131.pdf (15 May 2017, date last accessed). 39 Pastorino S, Richards M, Hardy R, et al. Validation of self-reported diagnosis 27 Siebolds M, Gaedeke O, Schwedes U. Self-monitoring of blood glucose—psycho- of diabetes in the 1946 British birth cohort. Prim Care Diabetes 2015; 9: logical aspects relevant to changes in HbA1c in type 2 diabetic patients treated 397–400. with diet or diet plus oral antidiabetic medication. Patient Educ Couns 2006; 40 Bringer J, Fontaine P, Detournay B, et al. Prevalence of diagnosed type 2 diabetes 62:104–10. mellitus in the French general population: the INSTANT study. Diabetes Metab 28 Giel KE, Enck P, Zipfel S, et al. Psychological effects of prevention: do participants 2009; 35:25–31. of a type 2 diabetes prevention program experience increased mental distress? Diabetes Metab Res Rev 2009; 25:83–8. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png European Journal of Public Health Oxford University Press

The association between diabetes and depressive symptoms varies by quality of diabetes care across Europe

Loading next page...
1
 
/lp/ou_press/the-association-between-diabetes-and-depressive-symptoms-varies-by-CVaanhhF0J

References (40)

Publisher
Oxford University Press
Copyright
Copyright © 2022 European Public Health Association
ISSN
1101-1262
eISSN
1464-360X
DOI
10.1093/eurpub/cky050
Publisher site
See Article on Publisher Site

Abstract

Downloaded from https://academic.oup.com/eurpub/article/28/5/872/4959186 by DeepDyve user on 13 July 2022 872 European Journal of Public Health Table A1 Gender inequality index, life expectancy at birth males and females and gender gap in life expectancy 2015 Life expectancy at birth 2015 Country GII 2015 Males Females Austria AT 0.078 78.8 83.7 4.9 Belgium BE 0.073 78.7 83.4 4.7 Bulgaria BG 0.223 71.2 78.2 7.0 Croatia HR 0.141 74.4 80.5 6.1 Cyprus CY 0.116 79.9 83.7 3.8 Czech Republic CZ 0.129 75.7 81.6 5.9 Denmark DK 0.041 78.8 82.7 3.9 Estonia EE 0.131 73.2 82.2 9.0 Finland FI 0.056 78.7 84.4 5.7 France FR 0.102 79.2 85.5 6.3 Germany DE 0.066 78.3 83.1 4.8 Greece GR 0.119 78.5 83.7 5.2 Hungary HU 0.252 72.3 79.0 6.7 Ireland IE 0.127 79.6 83.4 3.8 Italy IT 0.085 80.3 84.9 4.6 Latvia LV 0.191 69.7 79.5 9.8 Lithuania LT 0.121 69.2 79.7 10.5 Luxembourg LU 0.075 80.0 84.7 4.7 Malta MT 0.217 79.7 84.0 4.3 Netherlands NL 0.044 79.9 83.2 3.3 Poland PL 0.137 73.5 81.6 8.1 Portugal PT 0.091 78.1 84.3 6.2 Romania RO 0.339 71.5 78.7 7.2 Slovakia SK 0.179 73.1 80.2 7.1 Slovenia SI 0.053 77.8 83.9 6.1 Spain ES 0.081 80.1 85.8 5.7 Sweden SE 0.048 80.4 84.1 3.7 United Kingdom GB 0.131 79.2 82.8 3.6 GII: Gender inequality index. The European Journal of Public Health, Vol. 28, No. 5, 872–878 The Author(s) 2018. Published by Oxford University Press on behalf of the European Public Health Association. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/ 4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. doi:10.1093/eurpub/cky050 Advance Access published on 3 April 2018 ......................................................................................................... The association between diabetes and depressive symptoms varies by quality of diabetes care across Europe 1,2 3 3 Eva A. Graham , Katie H. Thomson , Clare L. Bambra 1 Department of Epidemiology, Biostatistics, and Occupational Health, Faculty of Medicine, McGill University, Montreal, QC, Canada 2 Douglas Mental Health University Institute, Montreal, QC, Canada 3 Institute of Health and Society, Faculty of Medical Sciences, Newcastle University, Newcastle upon Tyne, UK Correspondence: Eva A. Graham, Douglas University Mental Health Institute, Frank B. Common Pavilion, F-2116, 6875 Boul. LaSalle, Montreal, QC H4H 1R3, Canada, Tel: +1 5147616131 ext. 6140, e-mail: eva.graham@mail.mcgill.ca Background: Depressive symptoms are more common in adults with diabetes and may arise from the physical and psychosocial burden of disease. Better quality of diabetes care may be associated with a reduced disease burden and fewer depressive symptoms. Methods: This cross-sectional study included 34 420 participants from 19 countries in the European Social Survey Round 7 (2014–2015). Countries were grouped into quartiles based on their quality of diabetes care as measured in the Euro Diabetes Index 2014. Individual-level depressive symptoms were measured using the 8-item Center for Epidemiologic Studies—Depression Scale. Negative binomial regression was used to compare the number of depressive symptoms between adults with and without diabetes in each quartile of diabetes care quality. Analyses included adjustment for covariates and survey weights. Results: In countries with the highest quality of diabetes care, having diabetes was associated with only a 3% relative increase in depressive symptoms (95% CI 1.00–1.05). In countries in the second, third and fourth (lowest) quartiles of diabetes care quality, having diabetes was associated with a 13% (95% CI 1.08–1.17), 13% (1.08–1.19) and 22% (1.14–1.31) relative increase in depressive symptoms, respectively. Conclusion: The as- sociation between diabetes and depressive symptoms appears stronger in European countries with lower quality of diabetes care. Potential pathways for this association include the financial aspects of diabetes care, access to services and differential exposure to the social determinants of heath. Further research is needed to unpack these mechanisms and improve the quality of life of people with diabetes across Europe. ......................................................................................................... Downloaded from https://academic.oup.com/eurpub/article/28/5/872/4959186 by DeepDyve user on 13 July 2022 Diabetes and depressive symptoms across Europe 873 non-respondents. The mean response rate was 56% across all ESS7 Introduction countries and ranged from 31% in Germany to 69% in Lithuania. iabetes is highly prevalent in European adults, with a prevalence This study included all adult participants (age 18+) from ESS7 D of approximately 10.3% of men and 9.6% of women. countries except Estonia, as it did not collect information on Depression among people with diabetes is associated with poorer depressive symptoms, and Israel, as it is not located geographically diabetes control and treatment adherence, increased diabetes com- in Europe. All survey data and documentation are freely available at 3 4 plications, and a higher risk of mortality. Although it is generally www.europeansocialsurvey.org. accepted that depression or its symptoms are twice as prevalent 5,6 among people with diabetes, estimates vary across Europe. In a Measures UK study, 5.0% of participants with diabetes had possible cases of Self-reported depressive symptoms were measured using the 8-item depression compared with 3.8% without diabetes. In a large city in Center for Epidemiologic Studies—Depression Scale (CESD-8). the Netherlands, 29.7% of participants with diagnosed diabetes Depressive symptoms included feeling depressed, not feeling reported elevated depressive symptoms, in contrast to 19.4% of par- happy, feeling lonely, not enjoying life, feeling sad, feeling that ticipants with normal glucose levels. everything was an effort, feeling that you could not get going, and The relationship between diabetes and depression may be partially sleeping problems. Participants indicated how much of the time attributable to the physical and psychological burden of managing during the past week they endorsed each symptom on a 4-point diabetes and its complications. This burden may include financial scale (never, some of the time, most of the time or all or almost difficulties due to diabetes expenses, effects of social stigma or un- 10–12 all of the time). The CESD-8 supports a single factor structure, and employment as a result of ill health. People with diabetes may a final depressive symptom score was created by summing partici- also experience stress from the daily psychological burden of pants’ responses for each symptom. Possible scores ranged from 0 monitoring dietary intake, physical activity levels and blood sugar 9,13 (no symptoms) to 32 (all symptoms present nearly all the time). levels. The burden of disease hypothesis is supported by cross- Diabetes was measured in the ESS by asking participants whether sectional evidence that diagnosed diabetes is more strongly 8,14 they had or had experienced diabetes in the past 12 months. Self- associated with depression than undiagnosed diabetes. As well, reported diabetes has a high sensitivity in Europe when compared having diagnosed diabetes is associated with an approximate 27% with medical records, reported medication or repeated self-report of increased risk of new-onset depression compared with not having diabetes (sensitivity = 0.79). diabetes. Quality of diabetes care was measured using the Euro Diabetes Notably, the stress associated with diabetes may differ based on Index (EDI) 2014 published by Healthcare Consumer Powerhouse. quality of diabetes care. Better quality of care may reduce the The EDI was developed by an expert panel of clinicians and financial, social or psychological burden of diabetes and result in academics to compare European healthcare systems and services. reduced depressive symptoms. However, very little evidence is Indicators of quality of care included diabetes prevention initiatives, available to evaluate this hypothesis. One study conducted in the case finding and screening, range and reach of services (e.g. eye care, USA found that less availability or affordability of diabetes pharmaceuticals availability), access to treatment or care (e.g. patient medication and supplies was associated with increased diabetes- education, coverage of diabetes supplies), procedures available related psychological stress. Another study looked at differences (e.g. annual blood tests, eye exams) and diabetes outcomes. in depressive symptoms between people with diabetes living on Information for each indicator was obtained from national and either side of the USA/Mexico border, which have differing levels international surveys (e.g. WHO World Health Statistics), of quality of health care. No differences in depressive symptoms interviews with health care workers and public health experts, and were observed between groups. However, participants in this study national and regional diabetes registries, programs and reports. were selected from a convenience sample and diabetes care quality A total score for each country was calculated using a weighted was not measured in detail. Further cross-national research is warranted to better elucidate sum of these indicators. Scores ranged from 509 in Lithuania to 946 in Sweden, with a maximum possible score of 1000. associations between diabetes, depression and diabetes care quality in non-USA settings using large samples and robust measures of Importantly, a country’s relative position is more indicative of their quality of care than to their absolute score or ranking and quality of diabetes care. A large proportion of the research on depression and diabetes has been conducted in the USA and may therefore countries were divided into quartiles of quality of care for this analysis. not apply to European populations where healthcare systems generally provide more universal access, also termed higher levels Covariates included age in years, sex, education, marital status, current feeling about household income and employment (see of healthcare decommodification. This study aims to assess whether the association between diabetes and depressive symptoms table 1). Clinical covariates included the presence of heart/circula- tion problems (e.g. stroke) or high blood pressure in the past differs by quality of diabetes care across 19 European countries. It is hypothesized that this relationship will be weaker in countries with 12 months as well as body mass index (BMI) calculated from self- reported height and weight. The continuous covariates age and BMI higher standards of diabetes care. were modelled using fractional polynomial models to account for non-linearly, which select the best higher-order terms for each Methods covariate (e.g. age squared). Self-reported lifestyle covariates including smoking status, alcohol consumption and engagement in Data physical activity. Data come from the European Social Survey (ESS), a repeated cross- sectional survey of social attitudes and values in 21 countries across 18,19 Sample size Europe. This analysis used Round 7 of the ESS (ESS7), as this was the first time detailed information was collected on health A total of 34 467 adults participated in ESS7 from the 19 studies conditions, lifestyle behaviours and the social determinants of included in this analysis. Those with a self-reported height of 110 cm health. Each country in ESS7 used random probability sampling or below were excluded, as this may be due to measurement error to select a representative sample of individuals aged 15 and over in and obesity measures for these participants were extremely high private households, though specific sampling methods varied by (n = 17). Of the remaining sample, 91% had complete information country. Sampling was completed between August 2014 and on all variables. Most participants with incomplete information were December 2015 and several attempts were made to contact initial missing BMI values (n = 1113), presence of heart/circulation Downloaded from https://academic.oup.com/eurpub/article/28/5/872/4959186 by DeepDyve user on 13 July 2022 874 European Journal of Public Health Table 1 Demographic, clinical and lifestyle characteristics of participants in the European social survey 7 by quartile of diabetes care quality Quartiles of quality of Quartile 1 Quartile 2 Quartile 3 Quartile 4 diabetes care (highest) (lowest) N = 8719 N = 9362 N = 8211 N = 8128 Countries in order of quality of diabetes care Sweden, Slovenia, Austria, Ireland, Netherlands, Norway, Belgium, Czech Denmark, France, Portugal, Republic, Switzerland, Germany, Hungary, Poland, United Kingdom Finland Spain Lithuania Age mean (SD) 47.4 (0.30) 49.2 (0.32) 47.5 (0.29) 44.4 (0.32) % Female 51.6 51.9 51.9 52.4 Marital status Married/civil partnership/cohabitating 53.0 56.0 49.6 56.9 Divorced/separated/widowed 15.3 15.6 16.6 13.7 Never Married 31.7 28.4 33.8 29.4 Main activity Employed, retired, housework, other 91.3 91.1 86.9 92.3 Unemployed 4.9 6.3 10.0 6.7 Sick/disabled 3.7 2.6 3.0 0.9 Education Primary or lower secondary 30.3 30.4 55.2 35.6 Secondary 45.5 56.8 30.3 46.2 Tertiary 24.1 12.8 14.4 18.1 Income feeling Living comfortably 45.1 32.4 23.8 12.3 Coping on present income 39.6 50.9 46.7 61.5 Difficult on present income 11.8 13.5 21.7 22.3 Very difficult on present income 3.5 3.1 7.8 3.9 BMI mean(SD) 25.8 (0.08) 25.8 (0.09) 25.7 (0.07) 25.7 (1.00) % Heart/circulation problems or high blood pressure in past 12 months 3.9 6.1 5.0 7.1 % Diabetes in past 12 months 5.0 6.4 5.3 4.2 Total depressive symptoms mean(SD) 4.9 (0.06) 5.4 (0.07) 6.1 (0.07) 5.4 (0.09) Smoking Current 22.5 29.3 28.6 28.3 Former 25.4 23.6 19.9 18.9 Never 52.1 47.1 51.4 52.7 Alcohol use in past 12 months Never 17.3 16.1 27.6 21.7 Once a month or less 20.4 22.5 19.7 34.5 Once a week or less 31.2 34.1 26.3 31.9 Every day or several times a week 31.0 27.3 26.4 11.8 Number of past 7 days did sports or other physical activity mean(SD) 3.4 (0.04) 3.1 (0.05) 2.8 (0.04) 3.0 (0.05) problems (n = 594) or physical activity levels (n = 383). Missing incorporated clustering by country as well as variance from demographic, clinical and lifestyle variables were imputed using multiply imputed datasets. Sampling weights were used in all ordered logistic regression (education level, satisfaction with analyses to account for (i) selection probabilities of each participant current income, number of days of physical activity), multinomial based on each country’s sampling design and (ii) sampling error and logistic regression (sex, presence of heart problems or diabetes, non-response based on age, gender, education and region. marital status, smoking status, alcohol consumption) or linear Population weights were applied to account for differences in the regression (age, BMI) regression with all other variables as population size of each country. All analyses were conducted using predictors. As well, approximately 2.2% of the sample had missing Stata version 12 and used a significance level of 0.05. information on at least one depressive symptom (n = 752). Depressive symptoms were imputed using ordered logistic Results regression on all other depressive symptoms as well as the demo- graphic and clinical characteristics listed above. Twenty complete The highest quality of diabetes care was observed in Nordic imputed datasets were created. The final sample included34 420 countries, while lower quality of care was more prevalent in participants who had complete information following multiple Eastern Europe (table 1). This is consistent with trends observed imputation. in overall healthcare in European countries in the same year, albeit with some discrepancies. Country rankings are also Statistical analysis somewhat consistent with Bohm et al.’s classification of health care systems based on health care regulation, financing and service The outcome of total number of depressive symptoms was modelled provision. Countries with National Health Service or National using negative binomial regression in order to reflect the positively skewed distribution of depressive symptoms in the European Health Insurance systems (healthcare regulation and financing by the state) include Denmark, Finland, Norway, Sweden, Portugal, population. Presence of diabetes, diabetes quality of care quartile and an interaction term were included as predictors. A non-zero Spain, the UK and Ireland. While the majority of these countries interaction term would indicate that the association between rank at top of quality of diabetes care, Spain, Portugal and Ireland diabetes and depression varies depending on the quartile of have lower quality of diabetes care. Countries with Social Health diabetes care quality. Sensitivity analyses examined the same associ- Insurance systems, or regulation and financing by non-governmen- ations by sex and education level. Standard error estimates tal organizations, include Austria, Germany, Switzerland and Downloaded from https://academic.oup.com/eurpub/article/28/5/872/4959186 by DeepDyve user on 13 July 2022 Diabetes and depressive symptoms across Europe 875 Table 2 Rate ratio of diabetes on depressive symptoms by diabetes quartile Quality of diabetes care Model 1 Model 2 Model 3 Quartile 1 1.19 1.16–1.22 1.05 1.03–1.09 1.03 1.00–1.05 a a a Quartile 2 1.37 1.33–1.41 1.14 1.11–1.17 1.13 1.08–1.17 a a a Quartile 3 1.33 1.25–1.42 1.16 1.10–1.21 1.13 1.08–1.19 a,c a,b a,b Quartile 4 1.53 1.36–1.71 1.26 1.15–1.39 1.22 1.14–1.31 Rate ratios estimated from a negative binomial regression model of diabetes, quartile of diabetes care and their interaction on number of depressive symptoms. Model 1: unadjusted. Model 2: adjusted for individual differences in age, education, education  diabetes care quartile, satisfaction with income, gender, marital status, employment status, BMI and heart/circulation problems or high blood pressure. Model 3: additionally adjusted for smoking, alcohol use, alcohol use  diabetes care quartile and number of days doing physical activity. a: Significantly different from Quartile 1. b: Significantly different from Quartile 2. c: Significantly different from Quartile 3. Figure 1 Percentage (%) increase in number of depressive symptoms in people with diabetes versus without diabetes. Estimates adjusted for age, sex, education level, marital status, current feeling about household income, employment status, presence of heart/circulation problems or high blood pressure, body mass index, smoking status, alcohol consumption and physical activity Slovenia. These countries generally had quality of care rankings in activity was highest in the first quartile and lowest in the third. the second or third quartiles. Belgium, France, the Czech Republic, BMI did not differ between quartiles. Patterns of heart or circulation Hungary, the Netherlands and Poland have a mix of state, social and problems, diabetes and mean number of depressive symptoms private healthcare regulation, financing and provision. Countries varied by quartile with no clear trend. Results from the negative binomial regression model are shown in with this system were found in all quartiles of diabetes care. Country-specific scores are presented in Supplementary table S1. table 2 as the ratio of depressive symptoms between people with Table 1 presents differences in demographic, lifestyle and clinical diabetes compared with those without diabetes by quartile of factors between quartiles. The mean age of participants varied from diabetes care quality. Having diabetes was associated with 44.2 years (quartile 4)–49.2 years (quartile 2). There were approxi- increased depressive symptoms across all quartiles. Adjustment mately equal proportions of men and women in each quartile. Over for demographic, clinical and lifestyle characteristics attenuated half of participants in all quartiles were married, in a partnerships or these associations, though they were still statistically significant. cohabitating. Under 10% of the population in quartiles 1, 2 and 4 The association between diabetes and depressive symptoms were unemployed or permanently sick/disabled, while nearly 13% of increased as quality of diabetes care decreased. In the fully participants in the 3rd quartile were unemployed. Education levels adjusted model, statistically significant differences in depressive were highest in the first and second quartiles and lowest in the third. symptoms associated with diabetes are observed between the first Satisfaction with income followed a decreasing trend by quartile of (highest) quartile of diabetes care and all other quartiles and between care. Smoking trends were similar between quartiles, with 22 to 28% the second and fourth quartiles (figure 1). Figure 2 presents the main analysis stratified by sex. Men with of the population as current smokers. Frequency of alcohol con- sumption decreased from higher to lower quartiles, and physical diabetes in countries in the highest quartile of diabetes care showed Downloaded from https://academic.oup.com/eurpub/article/28/5/872/4959186 by DeepDyve user on 13 July 2022 876 European Journal of Public Health Figure 2 Percentage (%) increase in number of depressive symptoms in people with diabetes versus without diabetes by sex. Estimates adjusted for age, education level, marital status, current feeling about household income, employment status, presence of heart/circulation problems or high blood pressure, body mass index, smoking status, alcohol consumption and physical activity 6,14 no increased depressive symptoms, but higher depressive symptom associations. This may be partially attributable to women’s scores were observed in men with diabetes in the second, third and greater use of diabetes-related healthcare services, higher fourth quartiles. In women, diabetes was associated with increased adherence to glucose monitoring and selfcare, or increased social 30–33 depressive symptoms even in countries with the highest quality of support. However, in countries with the highest quality of diabetes care, but differences between quartiles were not statistically care, diabetes was not associated with increased depressive significant. A similar pattern of increasing depressive symptoms symptoms in men but there was an association in women. This among people with diabetes in lower quartiles of diabetes care was may allude to an added burden of managing diabetes in addition observed in each level of education (Supplementary figure S1). to childcare and family responsibilities, which are largely the role of women in all European countries. Women also report more health conditions and diabetes complications than men, such as poorer Discussion physical functioning, higher cholesterol levels, and more diabetes- The results of this analysis suggest that diabetes is more strongly related complications, which may contribute to increased depressive 30,35 related to depressive symptoms in European countries with poorer symptoms. Similar to the main results, analyses stratified by quality of diabetes care. While diabetes was associated with an educational status suggest that lower quality of diabetes care is increased rate of depressive symptoms across all levels of diabetes associated with higher depressive symptoms across all educational care quality, this association was weakest in countries with the levels. As well, higher educational levels appear to be associated with highest quality of care (RR = 1.03, 95% CI 1.00–1.05) and fewer depressive symptoms in diabetes in countries with higher strongest is countries with the poorest quality of diabetes care levels of quality of care, consistent with prior research. This may (RR = 1.22, 95% CI 1.14–1.31). These results suggest that be due to more positive attitudes towards diabetes and better coping improved quality of diabetes care may reduce some of the psycho- strategies among those with more education. logical burden associated with living with diabetes. Of course, quality of care is only possible pathway linking diabetes These results are consistent with interventions that have reported with an increased risk of depression. Our findings might therefore improved depression scores in participants who received increased also reflect wider issues within the welfare or health care systems or diabetes monitoring, education or prevention. One trial showed that indeed other individual factors such as the social stigma related to people with Type 2 diabetes who used a glucose-monitoring device, diabetes. However, by controlling for lifestyle factors and socio- kept a diary of glucose levels and intake, and received counselling demographics, we hope to have taken into account some of these experienced an improvement in depressive symptoms. Those who wider issues and isolated the role of quality of care. only received general counselling about their diet and lifestyle did not. Another trial showed that the implementation of a diabetes prevention program was associated with a decrease in depressive Strengths symptoms in the intervention group compared with the control This is the first nationally comparative study of depressive symptoms group. The authors of both trials suggest that decreased and diabetes across Europe and provides novel evidence on quality depressive symptoms may be related to a heightened sense of 27,28 of diabetes care and psychological symptoms associated with control over diabetes. However, a meta-analysis of web-based diabetes. This analysis used a large, representative dataset of partici- interventions that promoted better diabetes care did not show that pants in 19 European countries. The ESS7 ensured representative- such interventions were associated with decreased depression. ness though methods such as strict probability sampling and re- Further analyses in this study examined associations between contacting initial non-respondents. As well, survey weighting and diabetes and depressive symptoms comparing men and women. In countries with poorer quality of care, men showed a stronger asso- multiple imputation of missing variables were used in this analysis to minimize potential bias from selection of participants and non- ciation between diabetes and depression than women. This is consistent with previous meta-analyses that report sex-specific response. Due to the number of countries included, comparisons Downloaded from https://academic.oup.com/eurpub/article/28/5/872/4959186 by DeepDyve user on 13 July 2022 Diabetes and depressive symptoms across Europe 877 were possible between different levels of quality of diabetes care in this paper do not necessarily represent those of the funders or using an objective, international measure. UKCRC. The funders had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript. Limitations Conflicts of interest: None declared. Limitations of this study include uncertainty in the temporal order of diabetes and depressive symptoms, as the dataset was cross- sectional. Consequently, causality cannot be inferred. Unmeasured Key points confounding may be present due to differences in mental health care across Europe. However, there are currently no tools available that  Diabetes is associated with increased depressive symptoms allow a detailed comparative assessment of mental health care across across Europe Europe. Unmeasured confounding may also be present on an  An increased association between diabetes and depressive individual level from diabetes complications that were not symptoms was observed in countries with lower quality of measured the European Social Survey. This analysis was also diabetes care compared with those with higher quality of unable to differentiate between Type 1 and Type 2 diabetes, diabetes care Higher quality of diabetes care may reduce the burden or although elevated depressive symptoms are associated with both types of diabetes in European populations. stress of disease and is associated with fewer depressive symptoms among adults with diabetes Diagnosed diabetes was assessed using self-reported measures and may have been underestimated. Yet validation studies suggest that self-reported diagnosed diabetes has high concordance with 39,40 physician records or antidiabetic drug use. Furthermore, mis- References classification of diabetes would bias the association between diabetes and depressive symptoms towards the null. 1 WHO Regional Office for Europe. Diabetes: Data and statstics. World Health Finally, quality of diabetes care may differ substantially within a Organization, Regional Office for Europe. Available at: http://www.euro.who.int/en/ country, and the measure used in this study does not capture health-topics/noncommunicable-diseases/diabetes/data-and-statistics (10 regional or local granularity. The authors of the Euro Diabetes November 2016, date last accessed). Index acknowledge this limitation, but suggest that this index may 2 Lustman PJ, Clouse RE. Depression in diabetic patients: the relationship between nonetheless be meaningful when interpreted with caution. Further mood and glycemic control. J Diabetes Complications 2005; 19:113–22. research should consider more local measures of quality of diabetes 3 de Groot M, Anderson R, Freedland KE, et al. Association of depression and care. diabetes complications: a meta-analysis. Psychosom Med 2001; 63:619–30. 4 Park M, Katon WJ, Wolf FM. Depression and risk of mortality in individuals with Conclusions diabetes: a meta-analysis and systematic review. Gen Hosp Psychiatry 2013; 35:217– Results suggest an inverse association between national quality of 5 Nouwen A, Winkley K, Twisk J, et al. Type 2 diabetes mellitus as a risk factor for the diabetes care and depressive symptoms among people with diabetes. onset of depression: a systematic review and meta-analysis. Diabetologia 2010; In countries with high quality of diabetes care, diabetes was 53:2480–6. associated with a small increase in depressive symptoms while 6 Ali S, Stone MA, Peters JL, et al. The prevalence of co-morbid depression in adults countries with poorer quality of care showed a larger increase in with Type 2 diabetes: a systematic review and meta-analysis. Diabet Med 2006; depressive symptoms among people with diabetes. This association 23:1165–73. persisted when adjusted for individual demographic, clinical and 7 Holt RI, Phillips DI, Jameson KA, et al. The relationship between depression and behavioural characteristics. Quality of diabetes care may therefore diabetes mellitus: findings from the Hertfordshire Cohort Study. Diabetic Med J Br be important not only for physical complications of diabetes but also Diabetic Assoc 2009; 26:641–8. for depressive symptoms and psychosocial outcomes. 8 Knol MJ, Heerdink ER, Egberts AC, et al. Depressive symptoms in subjects with diagnosed and undiagnosed type 2 diabetes. Psychosom Med 2007; 69:300–5. Supplementary data 9 Hasan SS, Mamun AA, Clavarino AM, Kairuz T. Incidence and risk of depression associated with diabetes in adults: evidence from longitudinal studies. Community Supplementary data are available at EURPUB online. Ment Health J 2015; 51:204–10. 10 Schabert J, Browne JL, Mosely K, Speight J. Social stigma in diabetes: a framework to Funding understand a growing problem for an increasing epidemic. The Patient 2013; 6:1–10. 11 Rumball-Smith J, Barthold D, Nandi A, Heymann J. Diabetes associated with early EG received a scholarship to complete this research from the Queen labor-force exit: a comparison of sixteen high-income countries. Health Aff Elizabeth Scholars Program, funded by Universities Canada, (Millwood) 2014; 33:110–5. Community Foundations of Canada and the Rideau Hall 12 Jeon YH, Essue B, Jan S, et al. Economic hardship associated with managing chronic Foundation, in collaboration with the Institute for Health and illness: a qualitative inquiry. BMC Health Serv Res 2009; 9:182. Social Policy, McGill University. CB and KT are part of the 13 van Dooren FE, Denollet J, Verhey FR, et al. Psychological and personality factors in HiNEWS project—Health Inequalities in European Welfare type 2 diabetes mellitus, presenting the rationale and exploratory results from The States—funded by NORFACE (New Opportunities for Research Maastricht Study, a population-based cohort study. BMC Psychiatry 2016; 16:17. Funding Agency Cooperation in Europe) Welfare State Futures 14 Chen S, Zhang Q, Dai G, et al. Association of depression with pre-diabetes, un- programme (grant reference: 462-14-110). For more details on diagnosed diabetes, and previously diagnosed diabetes: a meta-analysis. Endocrine NORFACE, see http://www.norface.net/11. CB is a member of 2016; 53:35–46. Fuse, the Centre for Translational Research in Public Health. 15 Peyrot M, Egede LE, Campos C, et al. Ethnic differences in psychological outcomes Funding for Fuse comes from the British Heart Foundation, among people with diabetes: USA results from the second Diabetes Attitudes, Cancer Research UK, Economic and Social Research Council, Wishes, and Needs (DAWN2) study. Curr Med Res Opin 2014; 30:2241–54. Medical Research Council, the National Institute for Health Research, under the auspices of the UK Clinical Research 16 Mier N, Bocanegra-Alonso A, Zhan D, et al. Clinical depressive symptoms and Collaboration, and is gratefully acknowledged. The views expressed diabetes in a binational border population. J Am Board Fam Med 2008; 21:223–33. Downloaded from https://academic.oup.com/eurpub/article/28/5/872/4959186 by DeepDyve user on 13 July 2022 878 European Journal of Public Health 17 Bambra C. Cash Versus Services: ‘Worlds of Welfare’ and the Decommodification of 29 Pal K, Eastwood SV, Michie S, et al. Computer-based diabetes self-management Cash Benefits and Health Care Services. J Soc Policy 2005; 34:195–213. interventions for adults with type 2 diabetes mellitus. Cochrane Datab Syst Rev 2013;Cd008776. 18 The ESS Data Archive. ESS7-2014 Documentation Report. Bergen: Norwegian Centre for Reserach Data, 2015. Available at: http://www.europeansocialsurvey.org/ 30 Rossi MC, Lucisano G, Pintaudi B, et al. The complex interplay between clinical and docs/round7/survey/ESS7_data_documentation_report_e03_0.pdf (11 November person-centered diabetes outcomes in the two genders. Health Qual Life Outcomes 2016, date last accessed). 2017; 15:41. 19 Eikemo TA, Bambra C, Huijts T, Fitzgerald R. The first pan-European sociological 31 Yu MK, Lyles CR, Bent-Shaw LA, Young BA. Sex disparities in diabetes process health inequalities survey of the general population: the European Social Survey of care measures and self-care in high-risk patients. J Diabetes Res 2013; rotating module on the social determinants of health. Eur Sociol Rev 2017; 2013:575814. 33:137–53. 32 Kramer HU, Ruter G, Schottker B, et al. Gender differences in healthcare utilization 20 ESS7 – 2014 Fieldwork Summary and Deviations. Available from: http://www. of patients with diabetes. Am J Manag Care 2012; 18:362–9. europeansocialsurvey.org/data/deviations_7.html (1 Februray 2018, last accessed 33 Mathew R, Gucciardi E, De Melo M, Barata P. Self-management experiences among date). men and women with type 2 diabetes mellitus: a qualitative analysis. BMC Fam 21 Karim J, Weisz R, Bibi Z, Rehman S. Validation of the eight-item Center for Pract 2012; 13:122. Epidemiologic Studies Depression Scale (CES-D) among older adults. Curr Psychol 34 Aliaga C. How is the time of women and men distributed in Europe? Eurostat, 2006 Res Rev 2015; 34:681–92. Contract No.: KS-NK-06-004. Available at: http://ec.europa.eu/eurostat/documents/ 22 Sluik D, Boeing H, Montonen J, et al. Associations between general and abdominal 3433488/5438881/KS-NK-06-004-EN.PDF/7af4b8e8-1d40-478c-8410-e41204e0b29c adiposity and mortality in individuals with diabetes mellitus. Am J Epidemiol 2011; (15 May 2017, date last accessed). 174:22–34. 35 Kautzky-Willer A, Harreiter J, Pacini G. Sex and gender differences in risk, patho- 23 Cebolla Garrofe ´ B, Bjo ¨ rnberg A, Yung Phang A. Euro Diabetes Index. Health physiology and complications of type 2 diabetes mellitus. Endocr Rev 2016; Consumer Powerhouse, 2014. Available at: http://www.healthpowerhouse.com/files/ 37:278–316. EDI-2014/EDI-2014-report.pdf (8 November 2016, date last accessed). 36 Gois CJ, Ferro AC, Santos AL, et al. Psychological adjustment to diabetes mellitus: 24 European Social Survey. Weighting European Social Survey Data. April 24 2014. highlighting self-integration and self-regulation. Acta Diabetol 2012; 49:S33–40. Report No. Available at: https://www.europeansocialsurvey.org/docs/methodology/ 37 Katon W, von Korff M, Ciechanowski P, et al. Behavioral and clinical factors ESS_weighting_data_1.pdf (16 November 2016, date last accessed). associated with depression among individuals with diabetes. Diabetes Care 2004; 25 Bjo ¨ rnberg A. Euro Health Consumer Index. Health Consumer Powerhouse, 2014. 27:914–20. Available at: http://www.healthpowerhouse.com/files/EHCI_2014/EHCI_2014_ 38 Pouwer F, Geelhoed-Duijvestijn PH, Tack CJ, et al. Prevalence of comorbid report.pdf (2 May 2017, date last accessed). depression is high in out-patients with Type 1 or Type 2 diabetes mellitus. Results 26 Bohm K, Schmid A, Gotze R, et al. Classifying OECD Healthcare Systems: A from three out-patient clinics in the Netherlands. Diabetic Med J Br Diabetic Assoc Deductive Approach. 2012. No. 165. Available at: https://www.econstor.eu/ 2010;27:217–24. bitstream/10419/64809/1/726547131.pdf (15 May 2017, date last accessed). 39 Pastorino S, Richards M, Hardy R, et al. Validation of self-reported diagnosis 27 Siebolds M, Gaedeke O, Schwedes U. Self-monitoring of blood glucose—psycho- of diabetes in the 1946 British birth cohort. Prim Care Diabetes 2015; 9: logical aspects relevant to changes in HbA1c in type 2 diabetic patients treated 397–400. with diet or diet plus oral antidiabetic medication. Patient Educ Couns 2006; 40 Bringer J, Fontaine P, Detournay B, et al. Prevalence of diagnosed type 2 diabetes 62:104–10. mellitus in the French general population: the INSTANT study. Diabetes Metab 28 Giel KE, Enck P, Zipfel S, et al. Psychological effects of prevention: do participants 2009; 35:25–31. of a type 2 diabetes prevention program experience increased mental distress? Diabetes Metab Res Rev 2009; 25:83–8.

Journal

European Journal of Public HealthOxford University Press

Published: Oct 1, 2018

Keywords: diabetes mellitus; diabetes mellitus, type 2; depressive disorders

There are no references for this article.