Prevalent,uncontrolled,and undiagnosed diabetes mellitus among urban adults in Dire Dawa,Eastern Ethiopia: A population-based cross-sectional study

Abstract

Objective:

Globally, 8.8% of adults were estimated to have diabetes mellitus, with the low-and middle-income countries sharing the largest burden. However, the research evidence for targeted interventions is lacking in sub-Saharan Africa, particularly in Ethiopia. Therefore, this study aimed to assess the prevalence of diabetes mellitus, disaggregated by the epidemiology of diabetes mellitus morbidity and associated factors among adults in Dire Dawa town, Eastern Ethiopia.

Methods:

Data from a total of 872 randomly sampled adults aged 25–64 years were obtained for analysis using the World Health Organization STEPwise approach to non-communicable disease risk factors surveillance instruments. We estimated the prevalence of diabetes mellitus disaggregated by the previous diabetes mellitus diagnosis status and by the current blood sugar level control status. The bivariable and multivariable binary logistic regression model was used to identify correlates of diabetes mellitus, along with STATA version 14.2 for data management and analysis. All statistical tests were declared significant at p-value < 0.05.

Results:

14.9% (95% confidence interval: 12.1, 17.4) of adults aged 25–64 years had diabetes mellitus in the study sample with 58.5% (95% confidence interval: 49.7, 66.7) on diabetes mellitus medication. Among adults currently taking diabetes mellitus medications, 30.3% (95% confidence interval: 19.8, 45.6) had uncontrolled diabetes mellitus. The magnitude of previously undiagnosed diabetes mellitus was 6.2% (95% confidence interval: 4.8, 8.0) in the study sample and 41.5% (95% confidence interval: 33.3, 50.3) among the diabetics. The odds of diabetes mellitus were higher among adults over the age of 55 years (adjusted odds ratio = 2.1, 95% confidence interval: 1.2, 3.6), currently married adults (adjusted odds ratio = 2.3, 95% confidence interval: 1.2, 4.4), and overweight adults (adjusted odds ratio = 1.6, 95% confidence interval: 1.1, 2.1). Adults with primary education (adjusted odds ratio = 0.4, 95% confidence interval: 0.2, 0.8) and no formal education (adjusted odds ratio = 0.5, 95% confidence interval: 0.2, 0.9) had lower odds of diabetes mellitus.

Conclusion:

The prevalence of diabetes mellitus among adults was high in Dire Dawa, with a third of the diabetics having poor control of their blood sugar levels and, nearly four in ten were previously undiagnosed. Adults who were overweight, currently married, and those over 55 years need to be targeted for regular diabetes health checkups and community-based screening. Also, a mechanism should be instituted to track a patient’s adherence to medications and promote diabetes self-care management.

Keywords

Adult diabetes mellitus Dire Dawa undiagnosed diabetes STEPS survey

Introduction

Diabetes mellitus (DM) is a chronic disease caused by an inherited and/or acquired deficiency in the production of insulin by the pancreas, or by the ineffectiveness of the insulin produced.^1,2 The global prevalence of DM among adults over 18 years of age rose from 4.7% in 1980 to 8.8% in 2017.^3–5 In 2015, an estimated 4.0 million deaths were directly caused by DM.³ Almost half of all deaths that occur before the age of 70 were attributable to high blood glucose. The World Health Organization (WHO) projected that DM will be the seventh leading cause of death in 2030 and by 2045, it will affect 628.6 million people worldwide.²

Previously thought to be a disease of the affluent, DM is now rapidly increasing in low- and middle-income countries (LMICs). The International Diabetes Federation (IDF) estimated in 2017 that more than 79% of people with diabetes live in LMICs.^3,6 In the African region, the proportion of all deaths due to DM that occurring before 60 years of age was estimated to be 77.0%.^3,7 Specifically in Ethiopia, 3.8% (4.0% males and 3.6% females) of the population had diabetes, accounting for 1% of overall mortality in the country.⁴ The national 2016 STEPS survey report showed that 5.9% (6.0% male and 5.8% females) of Ethiopians had raised blood glucose levels and 98.4% had at least one risk factor for DM.⁸

The burden of DM has an economic implication, affecting the health system, and also adding an extra burden on the affected individuals and their families.⁹ For example, globally an estimated gross domestic product loss due to DM, including both the direct and indirect costs, will total US$ 1.7 trillion, while LMICs have a total loss of US$ 800 billion due to the unacceptably high burden of DM.^10,11 Besides the economic burden on the health-care system and national economy, DM imposes a catastrophic out-of-pocket personal expenditure from loss of income due to disability and premature death.¹¹ DM is also a major cause of blindness, kidney failure, heart attacks, stroke, amputation, and death. The negative impacts of DM are unacceptably high as there are effective public health and clinical interventions including a healthy diet, regular physical activity, maintaining normal body weight, and avoiding tobacco use.¹

Prior community-based studies conducted elsewhere reported the prevalence of DM to range from 1.4% to 19.1%^12–16 with associated factors including age, education status, occupational status, positive family history of DM, fruit and vegetable consumption, physical activity, smoking, and alcohol consumption.^15,17–21 However, to the best of our knowledge, there are no recent population-level studies in Dire Dawa concerning the magnitude of DM and associated factors that help guide informed decisions and the development of strategies. In addition, prior evidence accumulated regarding the prevalence of previously undiagnosed and uncontrolled DM was limited despite being an important public health intervention targeting morbidity, mortality, and improvement in the quality of life.^18,22 Besides, this study used a standardized and validated tool, the World Health Organization STEPwise approach to non-communicable disease risk factors surveillance (WHO NCD STEPS)²³ with a large community-based sample. In summary, this study aimed to assess the prevalent, uncontrolled, and undiagnosed DM and associated factors among urban adults in Dire Dawa, Eastern Ethiopia.

Methods and materials

Study setting and design

A population-based cross-sectional study was conducted from 1 June to 21 June 2017 in Dire Dawa City which is the capital of Dire Dawa Administration and located 515 km east of Addis Ababa, the capital of Ethiopia. The Dire Dawa administration consists of the city of Dire Dawa and the surrounding rural areas. The city has nine urban kebeles (equivalent to a sub-district, the lowest administrative unit) and 28 rural sub-districts. According to the Central Statistical Authority report of 2013, the total population of the administration was 405,444 with a population density of 4530/km² (11,700/sq mi); among the total population, 196,777 were male and 208,666 were female.²⁴

Populations

This study obtained data from a sample calculation for a study on metabolic syndrome where 872 participants provided valid observations from a total of 903. A sample size of 903 was calculated using the Open Epi version 3.01,²⁵ considering a 95% confidence level (CI), 3% degree of precision, 1.5 design effect, and 17% prevalence of metabolic syndrome (MetSyndr) based on the study conducted in Addis Ababa.²⁶ Eligible participants were adults aged 25–64 years who lived in Dire Dawa for at least 6 months before the survey. Pregnant women, physically and/or mentally disabled individuals, and participants who refused to consent for the study were excluded. A multi-stage random sampling technique was used to select sub-districts, households within the selected sub-districts, and eligible adults in the selected households. From a total of nine urban sub-districts in Dire Dawa City, five were randomly selected. The sample size was proportionally distributed to each of the sub-districts based on the number of households in that administrative unit. A systematic random sampling technique was employed to select households to be visited. During a household visit, the number of adults aged between 25 and 64 years was identified, and if two or more eligible adults were found in a household, a lottery method was used to select one adult for the interview.

Instruments and data collection

This study used the WHO NCD STEPS instrument which consisted of three steps for measuring the risk of NCD risk factors.²³ The first step consists of a collection of core and expanded socio-demographic and behavioral characteristics of the study population. The second step involves core and expanded physical measurement, and the third step consists of biochemical measurement.²³ The data collection tool was primarily developed in English and then translated to Amharic and Afan Oromo, the two widely spoken languages in the study setting by language experts. To check for consistency, back translation to the English language was made before data collection. The training was performed on the study objectives and the STEPS survey procedure. A pre-test was conducted in a sub-district that was not included in the study to check that the data collectors and responders understood the questions translated to the local languages and also measures consistency in measurement among the data collectors. Accordingly, necessary adjustments were made before the actual data collection according to the feedback from the pre-test.

Data were collected using a face-to-face interview technique by health-care professionals who had a Bachelor of Science (BSc) in nursing. Before the interview, study participants were informed about the purpose and procedure of the study, and written informed voluntary consent was obtained. Data were collected on weekends and in the afternoon on workdays during which time eligible adults were expected to be at home in the study setting.

Variable measurements

A digital glucometer (Accu-Chek^® blood glucometer; material no. 06453970018; batch/lot no. 496915) was used to measure capillary blood sugar after participants were asked about the time lapsed from the last meal. Capillary blood samples were collected three times on different occasions (for three consecutive days) from a single study participant, and the average of the three measurements was used for analysis to minimize errors. We used the International Diabetes Association’s (IDA) definition to define DM: fasting blood glucose (FBG) ⩾126 mg/dL or random blood glucose (RBG) >200 mg/dL.³ In addition, subjects who were taking DM medications during the survey were identified as previously diagnosed with DM.²⁷

Previously undiagnosed DM was defined as participants who had not had their blood sugar tested before and not taking DM medications during the survey and had an FBG ⩾126 mg/dL or RBG >200 mg/dL.³ Similarly, uncontrolled DM (or poor glycemic control) was defined when FBG ⩾126 mg/dL or RBG >200 mg/dL among those who were taking DM medication during the survey.¹

Blood pressure (BP) was measured using a semi-automatic BP monitor (Microlife BP A50; Microlife AG, Widnaiu, Switzerland) three times, each 3–5 min apart, from the left arm while the participant was in a sitting position, and the average value of the last two measurements was used to define hypertension; persistent systolic BP ⩾140 mm Hg or diastolic BP ⩾90 mm Hg or reported use of antihypertensive medication.²⁸

Anthropometric measurement was carried out following standard procedures and using calibrated instruments.²⁹ Weight was measured using a standard digital scale and a stadiometer to measure height, and the results were recorded to the nearest 0.5 cm. Body mass index (BMI) was defined as a person’s weight in kilograms divided by the square of the person’s height in meters (kg/m²) and categorized as follows: BMI <18.5, underweight; BMI 18.5–24.9, normal; and BMI ⩾25, overweight and obese.³⁰ The digital BP apparatus was used to measure BP. Abdominal obesity was defined as waist circumference: in males >94 cm and females >80 cm.²³

To measure physical activity, the WHO global recommendation on physical activity was used. Accordingly, the combination of moderate- and vigorous-intensity activity accumulating at least 600 metabolic equivalent (MET)-minutes per week was used as a cut-off to define physically active (achieved 600 or more MET-minutes per week) versus inactive adults (achieving less than 600 MET-minutes per week).^23,31

Current tobacco smoke was defined when a participant reported that he or she smoked tobacco products daily during the data collection period. Similarly, participants who drank alcohol in the past 30 days of the data collection period were considered as current alcohol users.²³

Quality control

The data were collected by trained and experienced field staff. The overall data collection process was supervised by research assistants who had a health background and at least had a BSc degree. A 3-day intensive training was given to the data collectors and the supervisors. The data collection instruments were pre-tested on 5% of sample households outside the selected sub-districts in the city to correct possible problems that may arise during actual data collection. The blood glucose test was conducted according to the standard operating procedures. Supervisors and the research team checked the data for completeness on an ongoing basis.

Data processing and analysis

The data were entered into Epi Data version 3.1 and exported to STATA version 14.2 statistical software. Both descriptive and analytic analyses were performed. The study findings were presented as mean values, percentages with their corresponding 95% CIs, tables, figure, and chi-square tests. Bivariable and multivariable binary logistic regression models were run to identify factors associated with DM. Variables with a p-value of less than 0.25 were considered for the multivariable binary logistic regression. Odds ratio with 95% CI was presented, and the Hosmer–Lemeshow goodness-of-fit test was used to assess model adequacy. All statistical tests were declared significant at p-value < 0.05. Multicollinearity was checked, and no significant correlation was found among variables included in the multivariable analysis.

We used multiple imputations by chained equations (MICEs)³² to handle missing values for five variables which altogether accounted for 14.6% of the data. Before the imputation, we checked for the pattern of these missing values using the Little’s test in STATA to check whether missing was completely at random (MCAR) which was defined when the test was not significant.³³

Results

Socio-demographic characteristics

A total of 872 adults participated in the survey with 822 providing complete responses to all the survey items. We used the MICE method to complete the incomplete responses. In this survey, females constituted 67.2% of the study sample and the mean age was 40.4 years. Five hundred and eighty two (66.7%) of the participants were married, and 85.5% had achieved some level of formal education (Table 1).

Table 1.

Socio-demographic characteristics of adults aged 25–65 years in Dire Dawa City, 2017 (n = 872).

Variables	Imputed analysis data		Complete case analysis data
	Frequency	Percentage	Frequency	Percentage
Sex
Male	286	32.8	286	32.8
Female	586	67.2	586	67.2
Age in years^a, mean (SD) = 40.4 (13.0)
25–34	371	42.6	366	42.6
35–44	187	21.4	184	21.4
45–54	121	13.9	120	13.9
55–64	193	22.1	190	22.1
Educational status^a
No formal education	126	14.4	119	14.2
Primary (1–8 grade)	333	38.2	315	37.7
Secondary (9–12 grade)	312	35.8	303	36.3
College/University	101	11.6	99	11.8
Marital status
Married	582	66.8	582	66.8
Not married	173	19.8	173	19.8
Divorced/widowed	117	13.4	117	13.4
Occupational status
Employed (office work)	232	26.6	232	26.6
Merchant	119	13.7	119	13.7
Unemployed	375	43.0	375	43.0
Others^b	146	16.7	146	16.7

SD: standard deviation.

Variables for which missing responses were imputed.

Housewives, pensioners, daily laborer, and janitor.

Behavioral, diet, and physical measurements

Of the total participants, 8.5% were current cigarette smokers and current smokers were 7.7% among the diabetics. The prevalence of alcohol consumption in the past 30 days before the survey was 13.6% among all the participants, and it was 13.9% among the diabetics. Two hundred and ninety five (33.8%) and 27.6% ate five or more servings of fruits and vegetables per week, respectively. Regarding physical activity, 44.7% of the total participants did not achieve the recommended 600 MET tasks minutes per week and this was even higher among the diabetics, 51.5%. Among the diabetics, 49% were overweight/or obese and 57% had abdominal obesity. This was high when compared to the prevalence of the same in the total participants: 39% were overweight/or obese versus 46% had abdominal obesity (Table 2).

Table 2.

Behavioral, diet, and physical characteristics of adults age 25–65 years in Dire Dawa City, 2017 (n = 872).

Tobacco status	Total (n = 872 (%))	Diabetes mellitus (DM)^a (n = 130 (%))	DM^b (n (%))
Currently smoke cigarette
Yes	74 (8.5)	10 (7.7)	10 (7.7)
No	798 (91.5)	120 (92.3)	120 (92.3)
Current use of alcohol
Yes	119 (13.6)	18 (13.9)	18 (13.9)
No	753 (86.4)	112 (86.1)	112 (86.1)
Fruit consumption per week
Two or fewer servings	467 (53.6)	54 (41.5)	54 (41.5)
Three to four servings	110 (12.6)	13 (10.0)	13 (10.0)
Five or more servings	295 (33.8)	63 (48.5)	63 (48.5)
Vegetable consumption per week
Two or fewer serving	481 (55.2)	88 (67.7)	88 (67.7)
Three to four servings	150 (17.2)	11 (8.5)	11 (8.5)
Five or more servings	241 (27.6)	31 (23.8)	31 (23.8)
Physical activity (MET-minutes)
<600	390 (44.7)	67 (51.5)	67 (51.5)
⩾600	482 (55.3)	63 (48.5)	63 (48.5)
Body mass index (BMI)^c (kg/m²)
Underweight (<18.5)	65 (7.5)	7 (5.4)	7 (6.6)
Normal (18.5–24.9)	467 (53.5)	60 (46.1)	43 (40.6)
Overweight (⩾25)	340 (39.0)	63 (48.5)	56 (52.8)
Abdominal obesity
Yes	405 (46.4)	74 (56.9)	74 (56.9)
No	467 (53.6)	56 (43.1)	56 (43.1)

Imputed analysis data.

Complete case analysis data.

Variables for which missing responses were imputed.

Prevalence of DM

The overall prevalence of DM among adults was 14.9% (95% CI: 12.1, 17.4) with 58.5% (95% CI: 49.7, 66.7) taking DM medication during the survey. The prevalence of DM was slightly higher among males than females, 15.0% (95% CI: 11.3, 19.7) versus 14.8% (95% CI: 12.2, 18.0). However, the difference was not statistically significant (χ²(degree of freedom (df)) = 0.0054 (1), p-value = 0.941).

The prevalence of previously undiagnosed DM was 6.2% (95% CI: 4.8, 8.0) in the study sample, and this accounted for 41.5% (95% CI: 33.3, 50.3) of the diabetics. The prevalence of previously undiagnosed DM was significantly higher among participants with a formal education than with no formal education (χ²(df) = 4.3 (1), p-value = 0.038). In addition, previously undiagnosed DM was significantly higher among participants with abdominal obesity (χ²(df) = 8.82 (1), p-value = 0.003) (Table 3).

Table 3.

The prevalence of undiagnosed and uncontrolled DM with participant characteristics, 2017.

	Undiagnosed DM (n = 130)		χ²(df)	Uncontrolled DM (n = 76)		χ²(df)
	Yes (n = 54)	No (n = 76)		Yes (n = 23)	No (n = 53)
Sex
Male	17 (31.5)	26 (34.2)	0.11 (1)	12 (52.2)	14 (26.4)	4.73 (1)*
Female	37 (68.5)	50 (65.8)		11 (47.8)	39 (73.6)
Age in years
25–54	31 (57.4)	59 (77.6)	6.06 (1)*	14 (60.9)	45 (84.9)	5.34 (1)*
55+	23 (42.6)	17 (22.4)		9 (39.1)	8 (15.1)
Educational status
No formal education	14 (25.9)	9 (11.8)	4.30 (1)*	7 (30.4)	2 (3.8)	10.92 (1)**
Formal education	40 (74.1)	67 (88.2)		16 (69.6)	51 (96.2)
Marital status
Currently married	43 (79.6)	59 (77.6)	0.07 (1)	16 (69.6)	43 (81.1)	1.24 (1)
Currently unmarried	11 (20.4)	17 (22.4)		7 (30.4)	10 (18.9)
Fruit consumption per week
<Five servings	20 (37.0)	47 (61.8)	7.78 (1)*	16 (69.6)	31 (58.5)	0.83 (1)
Five or more servings	34 (63.0)	29 (38.2)		7 (30.4)	22 (41.5)
Vegetable consumption per week
Fewer than five servings	47 (87.0)	52 (68.4)	6.02 (1)*	17 (73.9)	35 (66.0)	0.46 (1)
Five or more servings	7 (13.0)	24 (31.6)		6 (26.1)	18 (34.0)
Body mass index (BMI)
<25 kg/m²	25 (46.3)	42 (55.3)	1.02 (1)	7 (30.4)	35 (66.0)	8.22 (1)*
⩾25 kg/m²	29 (53.7)	34 (44.7)		16 (69.6)	18 (34.0)
Abdominal obesity
Yes	39 (72.2)	35 (46.1)	8.82 (1)*	13 (56.5)	22 (41.5)	1.45550 (1)
No	15 (27.8)	41 (53.9)		10 (43.5)	31 (58.5)

DM: diabetes mellitus; df: degree of freedom.

*p < 0.05; **p < 0.01.

The prevalence of uncontrolled DM among those taking DM medications during the survey was 30.3% (95% CI: 19.8, 45.6) and was significantly higher among females than males (χ²(df) = 4.73 (1), p-value = 0.030). Similarly, the prevalence of uncontrolled DM was significantly higher among participants with formal education compared to those with no formal education (Table 3).

Factors associated with DM

In this study, we identified factors associated with the overall DM including the following: sex, age, educational status, marital status, occupation, current alcohol use, smoking, fruit consumption, vegetable consumption, physical activity, BMI, abdominal obesity, and hypertension. All were considered using bivariate binary logistic regression and four variables with p-value ⩾ 2.5 (occupational status, alcohol consumption, fruit consumption, and abdominal obesity) were removed leaving 9 to proceed with the multivariable binary logistic regression model. In the bivariate model, age, marital status, BMI, and hypertension were significantly associated with DM. Most of the variables significant in the crude analysis also maintained the same in the multivariable model. After controlling for other variables in the multivariable model, age (55–64 years, adjusted odds ratio (AOR) = 2.1, 95% CI: 1.2, 3.6), marital status (currently married, AOR = 2.3, 95% CI: 1.2, 4.4), and BMI (BMI ⩾25 kg/m², AOR = 1.6, 95% CI: 1.1, 2.1) had a higher odds of developing DM compared to the reference groups. However, no formal educational status (AOR = 0.5, 95% CI: 0.2, 0.9) and primary school attendance (AOR = 0.4, 95% CI: 0.2, 0.8) had lower odds of developing DM (for details, see Table 4).

Table 4.

Multi-level correlates of diabetes mellitus among adults in Dire Dawa, Eastern Ethiopia, 2017.

Characteristics	Imputed data analysis		Complete case data analysis
	COR (95% CI)	AOR (95% CI)	COR (95% CI)	AOR (95% CI)
Sex
Male	1.1 (0.7, 1.5)	1.1 (0.7, 1.6)	1.1 (0.7, 1.6)	1.1 (0.7, 1.7)
Female	Ref.	Ref.	Ref.	Ref.
Age in years^a
25–34	Ref.	Ref.	Ref.	Ref.
35–44	1.4 (0.8, 2.4)	1.2 (0.7, 2.1)	1.3 (0.8, 2.4)	1.5 (0.8, 2.9)
45–54	1.9 (1.1, 3.4)*	1.5 (0.8, 2.8)	1.8 (1.0, 3.3)*	1.8 (1.1, 3.8)*
55–64	2.2 (1.3, 3.5)*	2.1 (1.2, 3.6)*	2.3 (1.4, 3.8)*	2.8 (1.4, 5.7)*
Educational status^a
No formal education	1.1 (0.5, 2.0)	0.5 (0.2, 0.9)*	0.8 (0.4, 1.6)	0.3 (0.1, 0.8)*
Primary (1–8 grade)	0.6 (0.4, 1.2)	0.4 (0.2, 0.8)*	0.7 (0.4, 1.3)	0.4 (0.2, 0.9)*
Secondary (9–12 grade)	0.8 (0.5, 1.5)	0.7 (0.4, 1.2)	0.8 (0.4, 1.5)	0.6 (0.3, 1.2)
College/university	Ref.	Ref.	Ref.	Ref.
Marital status
Never married	Ref.	Ref.	Ref.	Ref.
Currently married	2.6 (1.4, 4.8)*	2.3 (1.2, 4.4)*	2.6 (1.3, 4.9)*	2.0 (1.0, 4.1)*
Widowed/divorced	1.8 (0.8, 4.0)	1.5 (0.7, 3.6)	1.8 (0.8, 4.3)	1.3 (0.5, 3.3)
Currently smoke tobacco products
Yes	0.9 (0.4, 1.8)	0.9 (0.4, 1.9)	0.9 (0.4, 1.8)	0.9 (0.4, 1.9)
No	Ref.	Ref.	Ref.	Ref.
Vegetable consumption per week
Two or fewer servings	1.5 (0.97, 2.4)	1.5 (0.96, 2.5)	1.4 (0.9, 2.3)	1.2 (0.7, 2.1)
Three to four servings	0.5 (0.3, 1.1)	0.5 (0.3, 1.1)	0.4 (0.2, 1.1)	0.4 (0.2, 1.1)
Five or more servings	Ref.	Ref.	Ref.	Ref.
Physical activity
⩽600 MET-minutes	1.4 (0.9. 2.0)	1.1 (0.7, 1.6)	1.4 (0.9, 2.1)	1.2 (0.8, 1.8)
⩾600 MET-minutes	Ref.	Ref.	Ref.	Ref.
Body mass index (BMI)^a
Underweight	0.8 (0.4, 1.9)	0.8 (0.3, 1.8)	1.1 (0.5, 2.6)	1.1 (0.5, 2.7)
Overweight	1.5 (1.1, 2.3)*	1.6 (1.1, 2.1)*	2.0 (1.3, 3.0)*	1.7 (1.1, 2.6)*
Normal	Ref	Ref.	Ref.	Ref.
Hypertension (HTN)
Yes	1.7 (1.2, 2.6)*	1.3 (0.8, 2.0)	2.1 (1.4, 3.2)*	1.5 (0.9, 2.4)
No	Ref	Ref.	Ref.	Ref.

Ref.: reference category; COR: crude odds ratio; CI: confidence interval; AOR: adjusted odds ratio.

Variables for which missing responses were imputed.

p < 0.05.

Discussion

In this study, the overall prevalence of DM was 14.9% and the prevalence of undiagnosed DM was 6.2% among urban adults aged 25–64 years. Among participants taking DM medications during the survey, 30.3% had uncontrolled DM. The factors that were significantly associated with overall DM in this study included age, educational status, marital status, and BMI. In the bivariable χ² analysis, age, educational status, fruit and vegetable consumption, and abdominal obesity had a significant association with previously undiagnosed DM; in addition, age, sex, educational status, and BMI were significantly associated with uncontrolled DM.

The 14.9% prevalence of DM among adults in Dire Dawa identified in our study was higher than the estimates reported in sub-national studies conducted in Ethiopia ranging from 5.1% in Northwest Ethiopia to 11.5% in East Gojjam.^18,20,27 The current finding is also higher compared to different sub-national studies in Africa which reported the prevalence ranging from 1.4% in Uganda to 7.2% in Western Cape.^16,17,19,34 Our finding, however, is comparable with a 12.1% prevalence of DM by Bahijiri et al.³⁵ in Saudi Arabia and a 13% prevalence of DM by Idris et al.³⁶ in Indonesia. The observed higher prevalence of DM in our study compared to others could be due to variations in the level of difference in the family history of diabetes,³⁷ which we did not measure in our study, and may be due to an increased prevalence of higher BMI and abdominal obesity in the study setting. Besides, a higher unemployment rate and poor consumption of fruits and vegetables in the study setting may also have contributed to the observed differences. Previous observational studies conducted elsewhere reported that high exposure to unemployment was related to prediabetes and type 2 diabetes.^38,39 The protective effect of fruit and vegetable consumption against the risk of type 2 DM was established in previous studies reported in a systematic review and meta-analysis.⁴⁰

The magnitude of undiagnosed DM in this study was 6.2% (95% CI: 4.8, 8.0) with a prevalence of 41.5% among the diabetics. The 6.2% prevalence of undiagnosed DM in our study was comparable with a 5.3% pooled prevalence of undiagnosed DM by Asmelash and Asmelash⁴¹ in African adults although the same study reported an 8.7% prevalence among urban adults. The prevalence of previously undiagnosed diabetes among adults with diabetes in our study was also comparable to the 2014 global estimate of 45.8% although regional variations range from 24.1% to 75.1%.⁴² Our finding suggests that the magnitude of previously undiagnosed DM needs the attention of public health agencies and efforts should be directed to design a feasible means of establishing a system that enables community screening, which is useful for early diagnosis and prevention of premature death from complications due to undiagnosed DM.⁴³

In this study, 30.3% of participants taking DM medication did not achieve the target blood glucose levels. Our study finding indicates that diabetics in the study setting had better glycemic control compared to studies reporting a higher rate of poor glycemic control ranging from 53.3% in Saudi Arabia to 80% in Addis Ababa.^44–46 Although we found a lower proportion of adults with uncontrolled diabetes compared with studies in other settings, it is still too high as the main aim of DM treatment is to achieve glycemic control at an individual level as this is an important predictor for diabetic complication and death.^47,48 Thus, integrated clinical and community health intervention should be planned to achieve glycemic control and reduce complications and premature death.

Regarding the correlates of DM, consistent with our findings, several studies reported that the prevalence of overall DM was higher among adults aged over 55 years.^35,49,50 As people age, the tendency to exercise less, lose muscle mass, and weight gain increases, and this, in turn, may increase the risk of DM.⁵¹ Besides, as an individual gets older, the dynamic regeneration capacity of the pancreatic β-cell to maintain euglycemia reduces. The effect of age on the β-cell proliferation and function indirectly contributes to an impaired insulin sensitivity mediated by lifestyle and comorbidity-related risk factors.⁵² Consistent with the previous report by Selvin and Parrinello⁵³ and Fiagbe et al.,⁵⁴ age was also significantly associated with uncontrolled DM. Therefore, with an increase in age, adults should take advantage of adhering to modifiable risk factors including a healthy diet, regular physical activity, maintaining normal body weight, and avoiding alcohol and tobacco use⁵ to reduce the deteriorating effect to that of age.

Corroborating with the findings of previous studies,^13,35,49,55 DM is more likely to occur among overweight and obese adults. Biologically explained, the body of people who are overweight or obese has a limited capacity to use insulin for optimal blood glucose control due to the release of essential factors that lead to the development of insulin resistance.^56,57 Moreover, the presence of ectopic fat and brown adipose tissues among overweight and obese also expose individuals to the DM.⁵⁸ Therefore, routine public health DM–related education campaigns may help to create awareness on lifestyle modifications and the necessity of weight reduction. As the obesity epidemic drives the prevalence of DM, well expressed with the term “diabesity” in Bhupathiraju et al.,⁵⁹ focusing on measures to reduce overweight and obesity may have a positive effect in reducing the prevalence of DM.

Against the established evidence that vegetable and fruit consumption is associated with a reduced risk of diabetes,40,60,⁶¹ we did not find a statistically significant association. The lack of association in our study might be primarily due to the very low prevalence of fruit and vegetable consumption among adults in the study setting. In the bivariable analysis, however, fruit and vegetable consumption was significantly associated with undiagnosed DM. This is consistent with the study conducted by Ford and Mokdad.⁶²

In this study, the occurrence of DM was lower among participants who had primary or no formal education. This finding is in line with a sub-national study in Ghana¹⁴ and might be due to an increase in sedentary lifestyle with higher education and lower physical activities as they were more likely to be employed in an office setting.^63,64 Also, people with a lower level of education are more likely to be engaged in physically laborious activities for work compared to those with a higher educational status⁶⁵ which, in turn, will lower their risk of DM. However, the finding is contrary to the study conducted in Jeddah, Saudi Arabia,³⁵ Belgrade, Serbia,⁶⁶ and European countries.⁶⁷ Inconsistent with the previous studies,^68,69 we found that married adults had increased odds of having DM compared with unmarried adults. Previous studies reported that unmarried men had a reduced odds of having type 2 DM,⁶⁹ whereas another study reported that married women had no significant difference in the risk of having DM compared with the unmarried or widow/separated.⁶⁸ Therefore, we hypothesize that the quality of a marriage or home environment may influence health outcomes. For example, Leong et al.⁷⁰ reported that a spousal history of diabetes increased the risk of DM. However, in our study, we did not assess whether individual developed DM before or after marriage making further explanation difficult on the association between DM and marital status.

The strength

We estimated the population-level prevalence of DM disaggregated by previous diagnosis status and reported the level of glycemic control using the standard STEPS survey tool in a large sample of urban adults. To identify previously undiagnosed elevated blood glucose levels, we collected blood sugar measurements instead of relying on self-reports.

Limitations

While sharing the methodological limitation of cross-sectional studies with others, the effects of income or wealth, family history of diabetes, khat chewing, and eating behaviors other than fruits and vegetables were not assessed. We also did not identify the specific types of fruits or vegetables consumed. The study also did not consider acute and chronic illnesses which may affect blood sugar levels.

Conclusion

The overall prevalence of DM among adults in Dire Dawa was high, and the prevalence of previously undiagnosed DM and poor glycemic control were of great concern. Age, educational status, marital status, and BMI were significantly associated with DM. Therefore, creating awareness through activities targeting adults and identified factors should be devised and institutionalized in the health system in the study setting. Campaigns and community mobilizations to educate the community about early screening and preventive packages targeting modifiable risk factors should be employed to reduce the burden of DM.

Supplemental Material

sj-pdf-1-smo-10.1177_2050312120975235 – Supplemental material for Prevalent, uncontrolled, and undiagnosed diabetes mellitus among urban adults in Dire Dawa, Eastern Ethiopia: A population-based cross-sectional study

Supplemental material, sj-pdf-1-smo-10.1177_2050312120975235 for Prevalent, uncontrolled, and undiagnosed diabetes mellitus among urban adults in Dire Dawa, Eastern Ethiopia: A population-based cross-sectional study by Behailu Hawulte Ayele, Hirbo Shore Roba, Addisu Shunu Beyene and Melkamu Merid Mengesha in SAGE Open Medicine

Footnotes

Acknowledgements

The authors thank Haramaya University for providing us with financial support. The authors would like to extend their gratitude to the Dire Dawa City Administration Health Bureau and the community. The authors would also like to thank Dr Tara Wilfong and Ms Rasha Darghawth for their diligent language edition.

Author contributions

B.H.A., H.S.R., A.S.B., and M.M.M. conceived and designed the study. B.H.A., H.S.R., and M.M.M. adopted data collection instrument and acquisition of data. B.H.A., M.M.M., and H.S.R. carried out the data management. B.H.A. and M.M.M. analyzed the data and interpreted findings. B.H.A. and M.M.M. drafted the manuscript. B.H.A., H.S.R., A.S.B, and M.M.M. performed the critical revision and approval of the final manuscript.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was financially supported by the Haramaya University. The funder had no role in study design, data collection, and analysis, interpretation of data, the decision to publish, or preparation of the manuscript.

Ethical approval

Ethical clearance was obtained from the Institutional Health Research Ethics Review Committee of Haramaya University College of Health and Medical Science (HU, IHRERC) with an approval no. IHRERC/003/ 2016. In addition, permission was obtained from the Dire Dawa City Administration. Based on the objective of the study, an official letter was sent to the selected kebeles in Dire Dawa City Administration. Confidentiality was maintained and all respondents’ questionnaire was prepared anonymously.

Informed consent

The objective of the study was explained to each study participants. Verbal and written informed consent was obtained from each respondent before the interview. Those volunteers and signed on the consent form were involved in the study. Participants with newly diagnosed diabetes and hypertension were referred to nearby health facilities for further investigation and prompt management.

ORCID iDs

Behailu Hawulte Ayele

Addisu Shunu Beyene

Melkamu Merid Mengesha

Availability of data and materials

The data sets used and/or analyzed during this study will be available from the corresponding author on reasonable request.

Supplemental material

Supplemental material for this article is available online.

References

IDF. Definition and diagnosis of diabetes mellitus and intermediate hyperglycemia: report of a WHO/IDF consultation. Geneva: World Health Organization, 2006.

WHO. Guidelines on second- and third-line medicines and type of insulin for the control of blood glucose levels in non-pregnant adults with diabetes mellitus. Geneva: World Health Organization, 2018.

IDF. IDF Diabetes Atlas. 8th ed. Geneva: World Health Organization, 2017.

WHO. Global report on diabetes. Geneva: World Health Organization, 2016.

Zhou

Hajifathalian

, et al. Worldwide trends in diabetes since 1980: a pooled analysis of 751 population-based studies with 4·4 million participants. Lancet 2016; 387(10027): 1513–1530.

Whitinga

Guariguata

Weil

, et al. IDF diabetes atlas: global estimates of the prevalence of diabetes for 2011 and 2030. Diabet Res Clin Pract 2011; 94(3): 311–321.

IDF. Reports on DM. 5th ed. Geneva: World Health Organization, 2012.

EPHI. Ethiopia STEPS report on risk factors for non-communicable disease and prevalence of selected NCDs. Addis Abeba: Ethiopian Public Health Institute, 2016.

Afroz

Alramadan

Hossain

, et al. Cost-of-illness of type 2 diabetes mellitus in low and lower-middle-income countries: a systematic review. BMC Health Serv Res 2018; 18(1): 972.

10.

Seuring

Archangelidi

Suhrcke

The economic costs of type 2 diabetes: a global systematic review. Pharmacoeconomics 2015; 33(8): 811–831.

11.

Roglic

WHO Global report on diabetes: a summary. Int J Non-commun Dis 2016; 1(1): 3–8.

12.

Animaw

Seyoum

Increasing prevalence of diabetes mellitus in a developing country and its related factors. PLoS ONE 2017; 12(11): e0187670.

13.

Ebrahimi

Emamian

Shariati

, et al. Diabetes mellitus and its risk factors among a middle-aged population of Iran, a population-based study. Int J Diabet Develop Count 2014; 36(22): 189–196.

14.

Gatimu

Milimo

Sebastian

MS.

Prevalence and determinants of diabetes among older adults in Ghana. BMC Public Health 2016; 16: 1174.

15.

Elmadhoun

Noor

Ibrahim

AAA

, et al. Prevalence of diabetes mellitus and its risk factors in urban communities of North Sudan: population-based study. J Diabet Banner 2015; 8(6): 839–846.

16.

Ayah

Joshi

Wanjiru

, et al. A population-based survey of the prevalence of diabetes and correlates in an urban slum community in Nairobi, Kenya. BMC Public Health 2013; 13: 371.

17.

Stanifer

Cleland

Makuka

, et al. Prevalence, risk factors, and complications of diabetes in the Kilimanjaro region: a population-based study from Tanzania. PLoS ONE 2016; 11(10): e0164428.

18.

Aynalem

Zeleke

AJ.

Prevalence of diabetes mellitus and its risk factors among individuals aged 15 years and above in Mizan-Aman Town, Southwest Ethiopia, 2016: a cross-sectional study. Int J Endocrinol 2018; 2018: 9317987.

19.

Bailey

Ayles

Beyers

, et al. Diabetes mellitus in Zambia & The Western Cape province of South Africa; prevalence, risk factors, diagnosis, and management. Diabet Res Clin Pract 2016; 118: 1–11.

20.

Abebe

Berhane

Worku

, et al. mellitus in North West Ethiopia: a community based study. BMC Public Health 2014; 14: 97.

21.

Villegas

Shu

Gao

, et al. Vegetable but not fruit consumption reduces the risk of type 2 diabetes in Chinese women. J Nutr 2008; 138(3): 574–580.

22.

Endris

Worede

Asmelash

Prevalence of diabetes mellitus, prediabetes, and its associated factors in Dessie town, Northeast Ethiopia: a community-based study. Diabetes Metab Syndr Obes 2019; 12: 2799–2809.

23.

WHO. The WHO STEPwise approach to non-communicable disease risk factor surveillance: WHO STEPS surveillance manual. Geneva: World Health Organization, 2017.

24.

FDRE. Statistical report on the 2013 national labour force survey. Addis Ababa: Central Statistical Agency, 2013.

25.

Dean

AGSK

Soe

. Open-Epi: Open Source Epidemiologic Statistics for Public Health 2013 (updated 04. Version 3.01), www.OpenEpi.com

26.

Workalemahu

Gelaye

Berhane

, et al. Physical activity and metabolic syndrome among Ethiopian adults. Am J Hypertens 2013; 26(4): 535–540.

27.

Wondemagegn

Bizuayehu

Abie

, et al. Undiagnosed diabetes mellitus and related factors in East Gojjam (NW Ethiopia): a community-based study. J Public Health Res 2016; 6: 834.

28.

WHO. STEPS: A framework for surveillance. Geneva: World Health Organization, 2003.

29.

CDC. National Health and Nutrition examination survey (NHNES): Anthropometry Procedures Manual, 2007, https://www.cdc.gov/nchs/data/nhanes/nhanes_07_08/manual_an.pdf

30.

Jensen

Ryan

Apovian

, et al. 2013 AHA/ACC/TOS guideline for the management of overweight and obesity in adults: a report of the American College of Cardiology/American heart association task force on practice guidelines and the obesity society. Circulation 2014; 129(25 Suppl. 2): S102–S138.

31.

WHO. GPAQ: global physical activity questionnaire (version 2.0). Geneva: World Health Organization, 2010.

32.

Azur

Stuart

Frangakis

, et al. Multiple imputations by chained equations: what is it and how does it work? Int J Methods Psychiatr Res 2011; 20(1): 9.

33.

Kontopantelis

White

Sperrin

, et al. Outcome-sensitive multiple imputations: a simulation study. BMC Med Res Methodol 2017; 17(1): 2.

34.

Bahendeka

Wesonga

Mutungi

, et al. Prevalence and correlates of diabetes mellitus in Uganda: a population-based national survey. Trop Med Int Health 2016; 21(3): 405–416.

35.

Bahijri

Jambi

Al Raddadi

, et al. The prevalence of diabetes and prediabetes in the adult population of Jeddah, Saudi Arabia: a community-based survey. PLoS ONE 2016; 11(4): e0152559.

36.

Idris

Hasyim

Utama

Analysis of diabetes mellitus determinants in Indonesia: a study from the Indonesian basic health research 2013. Acta Med Indones 2017; 49(4): 291–298.

37.

Valdez

Yoon

Liu

, et al. Family history and prevalence of diabetes in the U.S. population. Diabetes Care 2007; 30(10): 2517.

38.

Rautio

Varanka-Ruuska

Vaaramo

, et al. Accumulated exposure to unemployment is related to impaired glucose metabolism in middle-aged men: a follow-up of the Northern Finland Birth Cohort 1966. Prim Care Diabetes 2017; 11(4): 365–372.

39.

Robinson

Yateman

Protopapa

, et al. Unemployment and diabetes. Diabet Med 1989; 6(9): 797–803.

40.

Fan

Zhang

, et al. Fruit and vegetable intake and risk of type 2 diabetes mellitus: a meta-analysis of prospective cohort studies. BMJ Open 2014; 4(11): e005497.

41.

Asmelash

The burden of undiagnosed diabetes mellitus in adult African population: a systematic review and meta-analysis. J Diabetes Res 2019; 2019: 4134937.

42.

Beagleya

Guariguataa

Weila

, et al. Global estimates of undiagnosed diabetes in adults. Diabet Res Clin Pract 2014; 103: 150–160.

43.

Goyder

Irwig

Screening for diabetes: what are we really doing?

BMJ 1998; 317(7173): 1644–1646.

44.

Tekalegn

Addissie

Kebede

, et al. Magnitude of glycemic control and its associated factors among patients with type 2 diabetes at Tikur Anbessa Specialized Hospital, Addis Ababa, Ethiopia. PLoS ONE 2018; 13(3): e0193442.

45.

Abouammoh

Alshamrani

MA.

Knowledge about diabetes and glycemic control among diabetic patients in Saudi Arabia. J Diabetes Res 2020; 2020: 1239735.

46.

Gebreyohannes

Netere

Belachew

SA.

Glycemic control among diabetic patients in Ethiopia: a systematic review and meta-analysis. PLoS ONE 2019; 14(8): e0221790.

47.

Simó

Hernández

Treatment of diabetes mellitus: general goals and clinical practice management. Revista Española De Cardiología 2002; 55(8): 845–860.

48.

Huang

Liu

Moffet

, et al. Glycemic control, complications, and death in older diabetic patients. Diabetes Care 2011; 34(6): 1329.

49.

Ajay

Prabhakaran

Jeemon

, et al. Prevalence and determinants of diabetes mellitus in the Indian industrial population. Diabet Med 2008; 25(10): 1187–1194.

50.

Miyakawa

Shimizu

Van Dat

, et al. Prevalence, perception, and factors associated with diabetes mellitus among the adult population in central Vietnam: a population-based, cross-sectional seroepidemiological survey. BMC Public Health 2017; 17: 298.

51.

Sue Kirkman

Briscoe

Clark

, et al. Diabetes in older adults. Diabetes Care 2012; 35(12): 2650–2664.

52.

Lee

Halter

JB.

The pathophysiology of hyperglycemia in older adults: clinical considerations. Diabetes Care 2017; 40(4): 444–452.

53.

Selvin

Parrinello

CM.

Age-related differences in glycaemic control in diabetes. Diabetologia 2013; 56(12): 2549–2551.

54.

Fiagbe

Bosoka

Opong

, et al. Prevalence of controlled and uncontrolled diabetes mellitus and associated factors of controlled diabetes among diabetic adults in the Hohoe municipality of Ghana. J Diabet Manag 2017; 7(5): 135.

55.

Asciutto

Lugo

Gallus

, et al. Prevalence and determinants of diabetes mellitus in a representative sample of Italian adults. J Epidemiol Biostat Public Health 2014; 11(4): 1–8.

56.

Kahn

Hull

Utzschneider

KM.

Mechanisms linking obesity to insulin resistance and type 2 diabetes. Nature 2006; 444: 840.

57.

Eckel

Kahn

Ferrannini

, et al. Obesity and type 2 diabetes: what can be unified and what needs to be individualized? J Clin Endocrin Metab 2011; 96(6): 1654–1663.

58.

Sharma

Lau

DCW

. Obesity and type 2 diabetes mellitus. Canad J Diabet 2013; 37(2): 63–64.

59.

Bhupathiraju

FB.

Epidemiology of obesity and diabetes and their cardiovascular complications. Circul Res 2016; 118(11): 1723–1735.

60.

Liu

Serdula

Janket

, et al. A prospective study of fruit and vegetable intake and the risk of type 2 diabetes in women. Diabet Care 2004; 27(12): 2993–2996.

61.

Wang

Fang

Gao

, et al. Higher intake of fruits, vegetables or their fiber reduces the risk of type 2 diabetes: a meta-analysis. J Diabet Invest 2016; 7(1): 56–69.

62.

Ford

Mokdad

AH.

Fruit and vegetable consumption and diabetes mellitus incidence among U.S. Prev Med 2001; 32(1): 33–39.

63.

Parry

Straker

The contribution of office work to sedentary behavior associated risk. BMC Public Health 2013; 13(1): 296.

64.

Waters

Ling

Chu

AHY

, et al. Assessing and understanding sedentary behavior in office-based working adults: a mixed-method approach. BMC Public Health 2016; 16: 360.

65.

Baker

DW.

Differences in leisure-time, household, and work-related physical activity by race, ethnicity, and education. J General Int Med 2005; 20(3): 259–266.

66.

Konevic

Martinovic

Djonovic

Association of socioeconomic factors and sedentary lifestyle in Belgrade’s suburb, working-class community. Iran J Public Health 2015; 44(8): 1053–1060.

67.

Sacerdote

Ricceri

Rolandsson

, et al. Lower educational level is a predictor of incident type 2 diabetes in European countries: the EPIC-InterAct study. Int J Epidemiol 2012; 41(4): 1162–1173.

68.

Cornelis

Chiuve

Glymour

, et al. Bachelors, divorcees, and widowers: does marriage protect men from type 2 diabetes? PLoS ONE 2014; 9(9): e106720.

69.

Ramezankhani

Azizi

Hadaegh

Associations of marital status with diabetes, hypertension, cardiovascular disease, and all-cause mortality: a long term follow-up study. PLoS ONE 2019; 14(4): e0215593.

70.

Leong

Rahme

Dasgupta

Spousal diabetes as a diabetes risk factor: a systematic review and meta-analysis. BMC Med 2014; 12: 12.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.46 MB