Marcelo Cássio Mendes Lúcio1; Janice Sepúlveda Reis1; Alexandra Dias Moreira2; Tatiane Géa Horta Murta1; Pedro Wesley Rosário1
BACKGROUND: Occupational aspects have been described as additional risk factors for type 2 diabetes mellitus (DM2). Workers whose job interferes with healthy eating and physical activity might be more susceptible to disease.
OBJECTIVE: To investigate sociodemographic, clinical, occupational and lifestyle factors associated with DM2 among employees of a public hospital in Belo Horizonte, Minas Gerais, Brazil.
METHODOLOGY: Cross-sectional study with 443 employees of a public hospital in Belo Horizonte. We administered a sociodemographic and occupational questionnaire and the Finnish Diabetes Risk Score (FINDRISC) and collected biochemical data. We performed multivariate logistic regression analysis to investigate factors associated with diabetes. The significance level was set to 5%.
RESULTS: 6.3% of the participants had a diagnosis of DM2 and 13% were found to be at high or very high risk to develop disease within 10 years. Participants aged above 54, with abnormal waist circumference, who took antihypertensive drugs or had family history of diabetes exhibited higher odds of developing disease.
CONCLUSION: Older age, abdominal obesity, hypertension and family history of diabetes were associated with diagnosis of DM2.
Keywords: diabetes mellitus, type 2; risk factors; occupational health; epidemiology.
INTRODUÇÃO: Características ocupacionais têm sido apontadas como fatores de risco adicionais para o desenvolvimento do diabetes mellitus tipo 2 (DM2). Trabalhadores cujas rotinas de trabalho dificultam a adoção de hábitos saudáveis relacionados à alimentação e atividade física podem estar mais vulneráveis a desenvolver esse agravo.
OBJETIVO: Estimar os fatores sociodemográficos, clínicos, ocupacionais e de hábitos de vida associados ao DM2 entre trabalhadores de uma empresa pública de Belo Horizonte.
METODOLOGIA: Trata-se de um estudo transversal com 443 trabalhadores de uma empresa hospitalar pública de Belo Horizonte (MG). Foram aplicados questionários envolvendo características sociodemográficas, ocupacionais, juntamente com o Finnish Diabetes Risk Score (FINDRISC) e coletados dados bioquímicos. Para análise dos fatores associados ao diabetes, utilizou-se a regressão logística multivariada, considerando nível de significância de 5%.
RESULTADOS: Entre os trabalhadores, 6,3% tinham diagnóstico de diabetes tipo 2 e 13% encontravam-se na faixa de risco alto/muito alto de desenvolver a doença nos próximos 10 anos. Observou-se que as pessoas com idade acima de 54 anos, circunferência da cintura alterada, que usavam medicamentos para hipertensão arterial e que tinham história de diabetes na família apresentaram maior chance de desenvolver a doença em comparação a pessoas com menos de 45 anos, cintura normal, que não usavam medicamento para hipertensão e sem história familiar.
CONCLUSÃO: A idade avançada, a obesidade abdominal, a hipertensão arterial e a história familiar foram fatores associados ao diagnóstico de diabetes mellitus tipo 2.
Palavras-chave: diabetes mellitus tipo 2; fatores de risco; saúde ocupacional; epidemiologia.
Diabetes mellitus (DM) is associated with countless acute and chronic complications which have considerably negative impact on the quality of life of patients and their families1. The prevalence of DM was 7.6% among adults in Brazil at the end of the 1980s and increased to about 9% in 2016 in the country as a whole and to 10% in São Paulo and Belo Horizonte1,2.
Most cases of DM correspond to two etiopathogenic categories known as type 1 (DM1) and 2 (DM2). The latter accounts for 90 to 95% of DM cases worldwide1. Different from DM1, DM2 is preventable. While countless factors were found to be associated with this condition, the most significant ones derive from lifestyle changes which took place in the past decades, including increase of dietary fats and rapidly absorbed carbohydrates, as well as in the consumption of sugary drinks. When associated with physical inactivity these factors increase the risk of overweight, obesity and DM23.
The pathophysiology of DM2 involves complex interactions between genetic predisposition and environmental risk factors. As is known, there is a close relationship between obesity/overweight and risk of DM2. Also the distribution of the body fat influences the risk of disease, being abdominal obesity the variety associated with the highest risk. Other known risk factors include sedentary lifestyle, hypertension, dyslipidemia, history of gestational diabetes and age. The prevalence of disease also varies as a function of ethnicity. Genetic predisposition is considerable, whence the relevance of the family history1.
Several additional risk factors were investigated in recent years including occupational aspects. Thus, for instance, workers whose job routine hinders them from adopting healthy eating habits and performing physical activity might be more susceptible to disease. This is the case of workers with long working hours, who work night shifts, have several jobs, inadequate meal and rest times and are frequently exposed to considerable stress and anxiety. Recent studies evidenced that circadian rhythm disorders and occupational stress are associated with higher risk of DM2 independently from traditional risk factors such as body weight, diet, physical activity and family history4,5.
Health care workers deserve special attention within this context, inasmuch as they are frequently work night shifts, have long working hours and more than one job, which factors hinder their attempts at adopting a healthy lifestyle. Stress and anxiety further contribute to increase risk among this population of workers6,7.
By comparison to other conditions with a well-established relationship to work, that of DM2 might be less evident within the scope of chronic degenerative disorders, even though it is not less relevant for some occupational groups. Nevertheless, few studies analyzed the association between occupational characteristics and risk of DM2, and even less specifically addressed this relationship among health care workers8,9.
As a function of the aforementioned considerations, the aim of the present study was to investigate sociodemographic, clinical, occupational and lifestyle factors related with DM2 among employees of a public hospital in Belo Horizonte, Minas Gerais, Brazil.
The present cross-sectional study was conducted from May through October 2016 in a public hospital in Belo Horizonte that provides medical, pharmaceutical, dental and social care under an ad hoc social security regime. The hospital employees were invited to participate in the study on the occasion of periodic medical examinations at the Occupational Health and Safety Department (OHSD). Those who met the inclusion and exclusion criteria were considered eligible. We excluded workers with DM diagnosed before age 25, having been hospitalized or receiving insulin since the time of diagnosis, pregnant women and interns (since they are not subjected to periodic examinations, therefore the necessary laboratory data were lacking).
For sample size calculation we considered a 2-point difference on the Finnish Diabetes Risk Score - FINDRISC - for a similar population of workers, significance level of 5% and statistical power of 90%. As a result 212 participants were needed at least.
The sample was characterized based on a questionnaire, anthropometric measurements and laboratory tests. The data were entered on ad hoc form that included the following variables: self-reported DM diagnosis (yes/no), sex (female/male), marital status (with/without partner), job at the hospital, second job, weekly working hours (up to 30, 31-50, >50), night shift (yes/no), smoking (yes/no) and routine laboratory tests defined as mandatory for all employees - except for interns - in the Occupational Health Medical Control Program (fasting glycemia, high-density lipoprotein - HDL, triglycerides).
Occupations were divided in two categories, namely, health care workers and all others, according to the Brazilian Classification of Occupations10.
We also administered FINDRISC, to wit, a questionnaire validated by the Department of Public Health, University of Helsinki, which is simple and quick to administer, inexpensive, noninvasive and reliable to determine risk of DM2 within the next 10 years. FINDRISC analyzes eight variables: age (<45, 45-54, 55-64, >64 years old); waist circumference - WC (normal, increased, abnormal - men: <94, 94-102, >102 cm; women: <80, 80-88, >88 cm, respectively); body mass index - BMI (<25 kg/m2, 25-30 kg/m2 or >30 kg/m2); physical activity (30 minutes/day at least: yes/no); dietary habits (regular fruit and vegetable intake: every day/sometimes); use of antihypertensive drugs (yes/no); family history of DM (grandparent/uncle/aunt/first cousin: yes/no; parent/sibling/child: yes/no); and previous history of high blood sugar (yes/no)11,12.
The final score is obtained by adding the scores on the individual items and ranges from 0 to 24. Individual risk is expressed as odds of developing DM2 within the next 10 years categorized as: low, score <7, estimated 1 in 100 people will develop disease; slightly elevated, score 7-11, estimated 1 in 25 will develop disease; moderate, score 12-14, estimated 1 in 6 will develop diseases; high, score 15-20, 1 in 3 will develop disease; and very high, score >20, estimated 1 in 2 will develop disease.
Although FINDRISC has not yet been validated for use in Brazil, it has already been used in several studies conducted in Portugal and Brazil and also at the State Referral Center for Diabetes Care and Endocrinology of Bahia, Brazil13.
Body weight and height were measured by a nursing assistant at OHSD with a Welmy® scale. BMI was calculated by dividing the body weight (in kg) by height (in meters) squared. We considered obesity as BMI≥30 kg/m2 and overweight as BMI≥25 kg/m2. WC was measured with non-elastic tape measure at the umbilical level as per international recommendations14.
Descriptive analysis included calculation of mean, standard deviation, median and quartile for continuous variables and absolute and relative frequencies for categorical variables. Groups were compared by means of the Mann-Whitney and Kruskal-Wallis tests. Associations were investigated with the χ2 and Fisher's exact tests.
We performed multivariate logistic regression analysis to investigate factors associated with of DM2. We first considered all the variables with p<0.20 on univariate analysis and only those with p<0.05 in at least one category remained in the final model. Analysis was performed with software Stata (Stata Corporation, College Station, Texas) version 12.0. The significance level was set to 5%.
The present study was approved by the research ethics committee of Holy House of Mercy, Belo Horizonte, the OHSD chair and the hospital director. All the participants signed an informed consent form.
The sample comprised 443 employees from different hospital departments. Most participants were female (78.6%) and younger than 45 years old (40.9%). The prevalence of DM2 was 6.3% and that of obesity and overweight 19.2 and 34.8%, respectively. As per the FINDRISC score, 26.3% of the participants were categorized as with low risk of DM, 22.5% with slightly elevated score, 38.2% with moderate risk, 12.8% with high risk and 0.2% with very high risk.
As Table 1 shows, the prevalence of high or very high risk was higher among the workers above age 64 (37.5%), women (14.5%), participants with abnormal WC (27.8%), who did not perform physical activity 30 minutes/day (19.5%), ate fruit and vegetables only sometimes (21.3%), used antihypertensive medication (39.8%) and had a parent/sibling/child with DM (27.8%). We did not find any statistically significant difference for the analyzed occupational variables.
On univariate analysis of risk factors, we found statistically significant association between risk of DM2 and age, BMI, WC, use of antihypertensive medication, high blood sugar and family history of DM. Neither in this case there was association between risk of DM2 and occupational variables (Table 2).
Table 3 describes the results of logistic regression analysis with significance level p<0.05. Only high blood sugar was removed from the model because the confidence interval was too wide. In turn, variables age, BMI, WC, use of antihypertensive medication and family history of DM remained significantly associated with the outcome. The participants older than 54 exhibited 4.61 times higher odds of developing DM2 compared to the participants under 45. Men with WC >102 cm and women with WC >88 cm exhibited 4.10 times higher odds of developing DM than the men with WC <94 cm and the women with WC <80 cm. Participants who had already taken antihypertensive drugs exhibited 2.53 times higher odds of DM2 than those who had never used this type of medication. Participants with family history of DM (parent/sibling/child) exhibited 3.27 times higher odds of DM2 compared to those without family history of disease.
Relative to the analyzed occupational characteristics, smoking and modifiable risk factors, we call the attention to the association between night work and daily fruit/vegetable intake (p=0.029). Workers who eat fruit and/or vegetables every day tend not to work at night. We did not find statistically significant association for any other occupational variable (occupational group, second job, working hours).
Rigorous lifestyle change programs with emphasis on healthy eating and regular physical activity are efficacious to reduce the incidence of DM215. Identifying individuals at high risk of disease and the risk factors with the strongest influence on definite populations is necessary to plan screening and preventive actions.
Thirteen percent of the participants in the present study were categorized as with high/very high risk of developing DM2 within the next 10 years (score≥15); the average score on FINDRISC was 10. FINDRISC was administered in studies conducted in Finland and Portugal, and the average score was 9.1 and 9.3, respectively12,16. In the latter, 12.8% of the participants were rated as with high/very high risk of disease. Among studies in Brazil in which FINDRISC was administered, the rates of participants categorized as with high/very high risk varied considerably, from 11.7% in Ceará to 18.6% in the South region17,18.
The prevalence of obesity and overweight in our sample was 19.2 and 54%, respectively, therefore similar to the national average2. The prevalence of these conditions, as well as that of risk categories, is similar to that reported in other studies and represents a cause of considerable concern. Our findings indicate that also the prevalence of DM and associated risk factors is increasing among the analyzed population of workers.
DM is currently the eighth leading global cause of death, the first being cardiovascular disease (CVD). DM is one of the main risk factors for CVD, and several risk factors are associated with both conditions, including obesity, dyslipidemia, sedentary lifestyle and smoking1. Thus, preventive actions against risk factors for DM have impact not only on the eighth, but also on the first leading cause of death worldwide.
We found that higher HDL was associated with lower risk of DM2, and higher body weight, triglycerides and fasting blood sugar with higher risk (data not shown). These findings lend further support to the use of FINDRISC in clinical practice as an accurate tool to estimate risk of DM2. Risk is five time higher among individuals with hypertension, high triglycerides, low HDL, abnormal blood glucose and abdominal obesity19.
Variables age, WC, use of antihypertensive medication and family history of DM remained significantly associated with the outcome in the regression model. This finding indicates that preventive and screening strategies against DM2 should prioritize workers above age 55, with abnormal WC, history of treatment for hypertension or of parents, siblings or children with DM.
In regard to the analyzed continuous variables, we found the same associations as for risk categories. These results reinforce the relevance of variables body weight, HDL, triglycerides and blood sugar as predictors of risk of DM2 (data not shown). We further emphasize the relationship between score on FINDRISC and diagnosis of DM2 (p<0.001) which corroborates the reliability of this scale as a low-cost and easy-to-apply instrument to stratify risk of DM2.
We did not find any statistically significant correlation between occupational variables - occupational group, working hours, more than one job - and risk of DM2, risk categories, modifiable risk factors considered in FINDRISC or smoking. Almeida et al.8 analyzed the relationship between risk factors for DM2 and nursing work among employees of a public hospital in Fortaleza, Ceará, Brazil, from March 2003 through March 2007. These authors analyzed health care workers exclusively and more particularly, they compared the nursing staff to other occupational groups. The results indicated statistically significant association between nursing work and abdominal obesity, abnormal waist-hip ratio, sedentary lifestyle, smoking and HDL <35 mg/dL. Yet, even though also these authors analyzed risk of DM2 among health care workers, methodological differences hinder possible comparisons with our results.
Some limitations might have impaired association analysis between occupational aspects and risk of DM2 in the present study. For having a cross-sectional design and being based on a questionnaire addressing the participants' current status, the assessment of the impact of occupational characteristics on risk of DM2 might have been impaired. We call the attention to the need for prospective studies to investigate causal relationships between the variables of interest.
Among our results, we emphasize the significant association found between working night shifts and dietary habits. In their study with night workers from a Guarulhos metallurgical industry in 2011, Lopes and Simony20 found that most participants exhibited overweight or obesity and reported considerable intake of fat, fried food, sweets and sugar and low intake of vegetables, fruit, milk and dairy products. In addition, a large part of their sample stated they were aware of differences in their diet when off work and during vacation. Similarly, 57% of the participants in our study who worked night shifts exhibited overweight versus 20.5% of those who did not. This difference, however, was not statistically significant (p=0.082).
The scientific evidence for the effect of lifestyle changes, especially physical activity and healthy eating, for prevention of DM2 has grown in recent years, is consistent and fully justifies actions in this regard1. Health care services should be prepared to act as effectively as possible in the prevention of DM2 as a function of the undisputable individual and social benefits of such actions. Accurate understanding of risk factors is essential for this purpose.
Night work has been a target of scientific research in recent years, leading to more and more evidence for its association with DM2. In the present study, we found association between this variable and dietary habits4,5. Further studies are warranted to achieve a more thorough understanding of this relationship.
1. Milech AI, Oliveira JEP, Vencio S. Diretrizes da Sociedade Brasileira de Diabetes (2015-2016). São Paulo: AC Farmacêutica; 2016.
2. Brasil. Ministério da Saúde. Vigitel Brasil 2016. Vigilância de fatores de risco e proteção para doenças crônicas por inquérito telefônico [Internet]. [cited on 17 Apr. 2017]. Available at: http://portalarquivos.saude.gov.br/images/pdf/2017/abril/17/Vigitel_17-4-17-final.pdf
3. International Diabetes Federation. IDF Diabetes Atlas [Internet]. 8ª ed. [cited on 19 Dec. 2018]. Available at: http://www.diabetesatlas.org/
4. Knutsson A, Kempe A. Shift work and diabetes - a systematic review. Chronobiol Int. 2014;31(10):1146-51. https://doi.org/10.3109/07420528.2014.957308
5. Nyberg ST, Fransson EI, Heikkilä K, Ahola K, Alfredsson L, Bjorner JB, et al. Job strain as a risk factor for type 2 diabetes: a pooled analysis of 124,808 men and women and diabetes - a systematic review. Diabetes Care. 2014;37(8):2268-75. https://doi.org/10.2337/dc13-2936
6. Rodrigues LGM, Queiroz JC, Oliveira LC, Menezes RMP, Lima GAF. Ocorrência do estresse em enfermeiros no ambiente hospitalar. Revista de Enfermagem. 2015;9(4):8054-8. https://doi.org/10.5205/reuol.6235-53495-1-RV.0904supl201510
7. Ferreira NN, Lucca SR. Síndrome de burnout em técnicos de enfermagem de um hospital público do Estado de São Paulo. Rev Bras Epidemiol. 2015;18(1):68-79. http://dx.doi.org/10.1590/1980-5497201500010006
8. Almeida VCF, Zanetti ML, Almeida PC, Damasceno MMC. Ocupação e fatores de risco para diabetes tipo 2: estudo com trabalhadores de enfermagem. Rev Latino-Am Enfermagem 2011;19(3):476-84.
9. Vilarinho RMF, Lisboa MTL. Diabetes mellitus: fatores de risco em trabalhadores de enfermagem. Acta Paul Enferm. 2010;23(4):557-61. http://dx.doi.org/10.1590/S0103-21002010000400018
10. Brasil. Ministério do Trabalho e Emprego. Classificação Brasileira de Ocupações [Internet]. Brasil: Ministério do Trabalho e Emprego; 2010. [cited on 19 Dec. 2018]. Available at: http://www.mtecbo.gov.br/cbosite/pages/download?tipoDownload=1
11. Lindström J, Tuomilehto J. The Diabetes Risk Score: a practical tool to predict type 2 diabetes risk. Diabetes Care. 2003;26(3):725-31. https://doi.org/10.2337/diacare.26.3.725
12. Saaristo T, Peltonen M, Lindström J, Saarikoski L, Sundvall J, Eriksson JG, et al. Cross-sectional evaluation of the Finnish Diabetes Risk Score: a tool to identify undetected type 2 diabetes, abnormal glucose tolerance and metabolic syndrome. Diabetes Vasc Dis Res. 2005;2(2):67-72. https://doi.org/10.3132/dvdr.2005.011
13. Centro de Diabetes e Endocrinologia do Estado da Bahia. Protocolos clínicos para assistência ao diabetes na atenção básica de saúde [Internet]. Salvador: Centro de Diabetes e Endocrinologia do Estado da Bahia; 2014 [cited on 19 Dec. 2018]. Available at: https://www.worlddiabetesfoundation.org/sites/default/files/Protocolos%20para%20assistencia%20ao%20DM%20-%20Bahia.pdf
14. World Health Organization. Waist Circumference and Waist-Hip Ratio: Report of a WHO Expert Consultation. Geneva: WHO; 2008.
15. American Diabetes Association. 5 Prevention or Delay of Type 2 Diabetes. Diabetes Care. 2015;38(Suppl. 1):S31-S32. https://doi.org/10.2337/dc15-S008
16. Valente T, Azevedo L. Estudo RADAR - Risco Aumentado de Diabetes em Amarante. Rev Port Med Geral Fam. 2012;28(1):18-24.
17. Marinho NBP. Avaliação do Risco para Diabetes Mellitus Tipo 2 entre Adultos de Itapipoca-Ceará [MA dissertation]. Fortaleza: Universidade Federal do Ceará; 2010.
18. Araújo LO, Silva ES, Mariano JO, Moreira RC, Prezotto KH, Fernandes CAM, et al. Risco para desenvolvimento do diabetes mellitus em usuários da atenção primária a saúde: um estudo transversal. Rev Gaúcha Enferm. 2015;36(4):77-83. http://dx.doi.org/10.1590/1983-1447.2015.04.50195
19. Rydén L, Grant PJ, Anker SD, Berne C, Cosentino F, Danchin N, et al. ESC Guidelines on diabetes, pre-diabetes, and cardiovascular diseases developed in collaboration with the EASD - summary. Diab Vasc Dis Res. 2014;11(3):133-73. https://doi.org/10.1177/1479164114525548
20. Lopes JE, Simony RF. Hábitos alimentares e estado nutricional de trabalhadores noturnos de uma indústria metalúrgica da cidade de Guarulhos, São Paulo, Brasil. Rev Simbio-Logias. 2013;6(9):117-29.
21 de Fevereiro de 2019.
Aceito em 2 de Setembro de 2019.
Fonte de financiamento: nenhuma