Metabolic syndrome and its associated factors in Ethiopia: A systematic review and meta-analysis

Abstract

Introduction

Metabolic syndrome (MetS) is a common complex entity that has emerged as a worldwide epidemic and major public health concern. The incidence of MetS often parallels the incidence of obesity and it is even worst among people living with comorbidities like; HIV/AIDS, hypertension, and mental illness. Therefore, there was an urgent need to summarize the extent and risk factors of MetS in Ethiopia.

Methodology

This systematic review was conducted according to the PRISMA guideline to investigate the prevalence of MetS and contributing factors. English language-based databases (PubMed, Cumulative Index to Nursing and Allied Health Literature, EMBASE, and Cochrane library) were exhaustively searched to identify studies related to the prevalence of MetS. A random-effects model was employed to estimate the pooled prevalence of MetS, and it was computed using STATA 16.0 software. Heterogeneity analysis was reported using I².

Result

A total of 25 studies with 21,431 study participants were included for this systematic review and meta-analysis. The pooled prevalence of MetS was 30.0% (95% CI: 24.0–36.0%, I2 = 99.19%, p < 0.001) with a high degree of heterogeneity across studies. Subgroup analysis with the target population showed that metabolic syndrome was most prevalent among type II diabetic 56% (95% CI: 47 – 64) and hypertensive patients 44% (95% CI: 35 – 53). Increased age, female gender, being overweight and obese, having a high educational level and income, physical inactivity, and being on treatment of chronic diseases like, diabetes mellitus, hypertension and HIV/AIDS were the most frequently reported risk factors of MetS regardless of the study population.

Conclusion

The prevalence of the MetS is high and rising in Ethiopia. Therefore, the preventative strategy should be considered to reduce the risk of morbidity or mortality related to metabolic syndrome.

Introduction

Non-communicable diseases (NCDs) are the leading causes of death globally, killing more people each year than all other causes combined [1]. The prevalence of NCDs is increasing worldwide in relation to the high population growth rate among developing countries. Nearly 85% of the global premature deaths associated with NCDs occur in low- and middle-income countries [1, 2]. In Ethiopia, the prevalence of diabetes mellitus, hypertension and obesity was 5.2% [3], 19.6% [4] and 5.4% [5], respectively. MetS is a cluster of health problems that set the stage for serious health conditions and places individuals at higher risk of cardiovascular disease, diabetes, and stroke [6]. It is defined as having three or more of the following components: central obesity, hypertension, increased triglyceride level, low level of high-density lipoprotein, and impaired fasting blood glucose level [2, 7].

The MetS is a popular dynamic entity that has become a global epidemic and a major public health problem [6]. In Ethiopia, the overall prevalence of obesity and overweight, which are risk factors of MetS were found to be 5.4% and 20.4% respectively [5]. People with MetS are two times more likely to die, three times as likely to have a heart attack or stroke, and five times more likely to develop type 2 diabetes mellitus compared to individuals without the syndrome [8, 9]. This situation is exacerbated for people living with comorbidities such as HIV/AIDS, hypertension, and mental illness [10–13].

MetS in the hypertensive population varies from 13 to 48% according to the clinical criteria used for its definition [2, 8]. The result of different studies shows an increased prevalence of MetS (17% to 42.5%) among HIV patients due to the long-term use of antiretroviral therapy (ART) [11, 14]. On the other hand, up to 73% of diabetic patients [10] and up to 32% of psychiatric patients [13, 15] can have MetS.

The result of different studies shows that advanced age, female sex, physical inactivity, BMI, and family history of chronic disease are predictor variables of MetS regardless of the study population [16–20]. Several MetS prevalence studies are conducted in Ethiopia; however, the scope of these problems has not been summarized, and their magnitude remains unclear. Therefore, this systematic review and meta-analysis aimed to estimate the pooled prevalence of MetS and its associated factor in Ethiopia.

Methods

Using the PRISMA flow diagram, a systematic review with meta-analysis was performed. While conducting this systematic review, the PRISMA checklist was also strictly followed [21]. An ethical statement was not required.

Data sources and search strategy

By visiting legitimate databases and indexing services, a systematic search was conducted. Therefore, Cumulative Index to Nursing and Allied Health Literature (CINAHL), Scopus, PubMed, EMBASE (Ovid), and other supplementary sources, Google Scholar and Cochrane library have been used. Advanced search strategies were applied in major databases to retrieve relevant findings closely related to metabolic syndrome, prevalence, risk factors and Ethiopia. The search was conducted with the aid of carefully selected keywords and indexing terms. These include “Metabolic syndrome,” “prevalence,” and “Ethiopia”–related keywords. Boolean operators (“OR,” “AND”) and truncation were used to identify relevant articles and records that meet our research question. The search was carried out from the outset on 20 September 2020. Until the end of the data extraction period, all published and unpublished articles on the prevalence of MetS among study participants in Ethiopia were included. Bibliography lists from all eligible articles were also hand-searched to identify additional papers potentially relevant for inclusion.

Inclusion and exclusion criteria

This systematic review included all observational studies conducted in the English language addressing the prevalence of metabolic syndrome in Ethiopia. Nonetheless, articles with insufficient information, records with missing outcomes of interest, findings from personal opinion observations, case reports and series, systematic reviews, and qualitative studies were excluded.

Article screening process

Records identified from various electronic databases, indexing services, and directories were saved and exported to Covidence software. Using Covidence software, duplicate records were removed. Because of differences in reference styles across sources, some duplicates were addressed manually. The two authors (BK and TJ) did the initial title and abstract screening. During the selection process, three categories (yes, no, maybe) were used. The full text of studies considered ‘yes’ or ‘maybe’ during the screening was assessed based on two authors' eligibility criteria (TJ and BK). Then, full-text screening was conducted by two authors (BK and TJ). In each case, if any discrepancies arose, it is resolved by discussion.

Data extraction and synthesis

Two authors (BK and TJ) extracted relevant data independently using a Microsoft Excel-prepared structured data abstraction format. Study characteristics (study setting, first author, publication year, study design, population characteristics, and sample size) and study outcomes (the prevalence of metabolic syndrome and associated factors) are included in the extracted data. Any disagreements were resolved by discussion and by crosschecking the papers.

Methodological quality assessment of studies

Two authors (TJ and BK) using the Newcastle–Ottawa scale [22] independently evaluated the methodological quality and risk of bias of the included studies, rating study quality out of 10 points (stars). The instrument contained significant indicators categorized into three major domains for ease evaluation. The first section assesses the methodological quality of a study, which has a maximum of five stars. The second section considers the comparability of the study and takes two stars, and the remaining section assesses the outcomes of studies related to statistical analysis. This critical assessment was conducted to assess the internal and external validity of the included studies. For the final decision, two authors (TJ and BK) 's mean score was taken, and studies were included with a score of five and above points/stars.

Outcome measurements and data analysis

The outcome measurements of this systematic review in terms of the prevalence of MetS among the participants of the study were reported in the included study. STATA/MP 16.0 software was used to compute the pooled prevalence of MetS. Heterogeneity analysis was reported using I2. In addition, the sub-group analysis of the target population and study settings was conducted. Furthermore, contributing factors were also summarized in a table as clinical factors and other characteristics.

Result

Study selection

The literature search identified a total of 387 records from several sources. After removing duplicate records with Covidence software, the remaining 286 records were screened using their titles and abstracts, and 253 of them were excluded. The full-text of 33 papers was then evaluated as per the predetermined eligibility criteria for inclusion. Eight articles were also excluded as the outcome of interest was missing, insufficient, and unacceptable methodological quality. Finally, 25 articles met the eligibility criteria and quality assessment in the present systematic review (Fig. 1).

Fig. 1.

PRISMA flow chart of the study selection process

Characteristics of the included studies

The characteristics of the articles included in the present systematic review and meta-analysis were summarized in Table 1. A total of 25 studies with 21,431 study participants were included for this systematic review and meta-analysis. From whom, 3520 participants had metabolic syndrome. All the included studies employed a cross-sectional study design in common and the year of publication of included studies ranges from 2011 to 2020. The number of study participants ranged from 159 in UOGCSH [23] to 9141 in Nation based survey [24] and targeted a wide range of population characteristics. The prevalence of metabolic syndrome ranged from 4.8% in Nation based survey [24] to 70.1% in HUCSH [25]. Regarding geographic distribution, the 25 studies were obtained from five regions, multinational and two city administration (Addis Ababa and Dire Dawa): nine studies were conducted in Southern Nations Nationalities and Peoples of Ethiopia [2, 25–32], four studies in Oromia region [33–36], four studies at the capital city of Ethiopia, Addis Ababa [1, 37–39], three studies in Amhara region [23, 40, 41], two studies in Harare region [42, 43], one study in Tigray region [44], in Dire Dawa [45] and at the national level [24]. No study was included from Somali, Gambela, Benishangul-Gumuz, and Afar regions. The average quality scores of studies range from 5 to 9.5, as per the Newcastle–Ottawa scale.

Table 1.

Characteristics of the included studies in the systematic review

Authors	Publication year	Region	Study setting	Study design	Study population	Sample size	Prevalence of MetS, N (%)	Event rate	SE
Tadewos et al. [2]	2017	SNNP	HUCSH	CS	Hypertensive patients	238	48.7	0.487	0.032
Teshome et al. [31]	2020	SNNP	HUCSH	CS	Psychiatric patients	245	26.9	0.269	0.028
Woyesa et al. [25]	2017	SNNP	HUCSH	CS	type-II DM patients	319	70.1	0.701	0.026
Tadewos et al. [29]	2017	SNNP	HUCSH	CS	type-II DM patients	270	45.9	0.459	0.030
Bune et al. [30]	2019	SNNP	Gedeo zone	CS	HIV patients	633	22	0.220	0.016
Tesfaye et al. [32]	2014	SNNP	HUCSH	CS	HIV patients	374	23.8	0.238	0.022
Kerie et al. [28]	2019	SNNP	Mizan-Aman town	CS	Participants in age >  = 18 years	534	9.6	0.096	0.013
Hirigo et al. [26]	2016	SNNP	HUCSH	CS	HIV patients	185	24.3	0.243	0.032
Bune et al. [27]	2020	SNNP	Gedeo zone	CS	HIV patients	633	42.5	0.425	0.020
Solomon et al. [1]	2019	AA	SPHMMC	CS	Adult outpatients	325	20.3	0.203	0.022
Bekele et al. [37]	2020	AA	Addis Ababa city high schools	CS	Adolescents	824	12.4	0.124	0.011
Tran et al. [38]	2011	AA	Addis Ababa	CS	Working adults	1,935	17.9	0.179	0.009
Workalemahu et al. [39]	2013	AA	Addis Ababa	CS	Working adults	1,843	16.98	0.1698	0.009
Asaye et al. [35]	2018	Oromia	JUSH	CS	Psychiatric patients	360	28.9	0.289	0.024
Abda et al. [36]	2016	Oromia	JUSH	CS	OPD adult patients	225	26.2	0.262	0.029
Berhane et al. [33]	2012	Oromia	JUSH	CS	HIV patients	313	21.1	0.211	0.023
Bosho et al. [34]	2018	Oromia	Jimma health centre	CS	HIV patients	268	20.5	0.205	0.025
Belete et al. [23]	2018	Amhara	UOGCSH	CS	type-II DM patients	159	53.5	0.535	0.040
Birarra et al. [41]	2018	Amhara	UOGCSH	CS	type-II DM patients	256	57	0.570	0.031
Tachebele et al. [40]	2017	Amhara	UOGCSH	CS	Hypertensive patients	300	39.3	0.393	0.028
Ataro et al. [43]	2020	Harare	Jugal Hospital	CS	HIV patients	375	26.7	0.267	0.023
Cheneke et al. [42]	2019	Harare	Harar town	CS	Hormonal contraceptive users	365	27.7	0.277	0.023
Gebremeskel et al. [44]	2019	Tigray	ACSH	CS	type-II DM patients	419	51.1	0.511	0.024
Mengesha et al. [45]	2020	Dire Dawa	Dire Dawa	CS	Adults aged 25 to 64	872	9.5	0.095	0.010
Gebreyes et al. [24]	2018	9 regions	Multi-center	CS	Participants in the age range of 15–69 years	9141	4.8	0.048	0.002

AA Addis Ababa, HUCSH Hawassa University comprehensive Specialized Hospital, ACSH Ayider comprehensive specialized hospital, UOGCSH University of Gondar comprehensive specialized hospital, JUSH Jimma university specialized hospital, SPHMMC St. Paul’s Hospital Millennium Medical College, SE standard error, CS cross sectional, SNNP Southern Nations Nationalities and Peoples of Ethiopia

Study outcome measures

Prevalence of metabolic syndrome in Ethiopia

From the 25 studies describing metabolic syndrome, the pooled prevalence of metabolic syndrome in Ethiopia was found to be 30.0% (95% confidence interval [CI]: 24.0 -36.0%). As the I² statistic revealed, there is a high degree of heterogeneity across studies (I2 = 99.19%, p < 0.001). Random effects model was assumed for this meta-analysis (Fig. 2).

Fig. 2.

Forestplot illustrating the pooled analysis of 25 studies reporting metabolic syndrome in Ethiopia

Sensitivity and subgroup analyses

There was no significant change in the degree of heterogeneity even if we attempted to exclude the expected outliers and one or more of the studies from the analysis. Therefore, we were subjected to include all the studies for the meta-analysis.

We also conducted a subgroup analysis based on geographical distribution and target population. Subgroup analysis based on region revealed that the highest prevalence of metabolic syndrome was observed in the Amhara region, 50.0% (95% CI: 38—61), followed by the south region with 35.0% (95% CI: 22–47). A lower pooled estimate was observed at Ethiopia's capital city, Addis Ababa, 17.0% (95% CI: 14 – 19), as depicted in Table 2. Another subgroup analysis with the target population showed that metabolic syndrome was most prevalent among type II diabetic 56% (95% CI: 47 – 64) and hypertensive patients 44% (95% CI: 35 – 53) (Fig. 3.).

Table 2.

Subgroup analysis based on the region of studies

Region	Number of publications	Pooled estimate (95% CI)	Heterogeneity (I2)
Harare	2	27.0% (24-30)	0.00%
Amhara	3	50.0% (38 – 61)	89.89%
Oromia	4	24.0% (20 – 28)	64.14%
Addis Ababa	4	17.0% (17 – 19)	83.81%
SNNP	9	35.0% (22 – 47)	98.71%
Others	3	21.0% (7 – 38)	99.47%
Overall	25	30.0% (24 – 36)	99.19%

SNNP Southern Nations Nationalities and Peoples, Others: Tigray, Dire Dawa and multi-national.

Fig. 3.

Subgroup analysis based on the target population of studies

Factors associated with the prevalence of metabolic syndrome in Ethiopia

In this systematic review, various factors contribute to the prevalence of metabolic syndrome. Increased age (1, 2, 23, 28, 30–32, 35, 36, 42–45), female gender (23, 26, 28–32, 35–37, 40–42, 44, 45), overweight, and obesity (1, 2, 23, 26, 29, 31, 32, 34, 36, 37, 40, 44, 45), having high educational level (1, 26, 28, 34, 40, 42), physical inactivity (28, 37, 44, 45), who are on treatment of chronic diseases like, diabetes mellitus, hypertension and HIV/AIDS (32, 33, 36, 43) were the most frequently contributing factors for the development of metabolic syndrome. Furthermore, duration of diagnosis with diabetes (23, 29, 44), having high-income status (1, 2, 42) and who have a chronic disease (36, 44) and hyperlipidemia (32, 36) were associated factors for the prevalence of metabolic syndrome in Ethiopia (Table 3)

Table 3.

Factors associated with the prevalence of metabolic syndrome

Authors	Sample size	Prevalence of MetS, n (%)	Factors affecting the prevalence of MetS
Tadewos et al. (2)	238	48.70	Age over 60 years, overweight, obesity, high monthly income, retirement from job, divorced, widowed and married
Solomon et al. (1)	325	20.5	Older age, Amhara ethnicity, overweight, higher income and higher education level
Bekele et al. (37)	824	12.4	Female gender, smoker, alcohol drinker, no vigorous or modern physical activity, overweight, obesity, and more time spent for sedentary activities
Asaye et al. (35)	360	28.9	Age greater than 30 years old, female gender and regularly eating high protein and fat
Abda et al. [36]	225	26.2	Female gender, increased age, on treatment of hypertension, diabetes, hyperlipidemia, BMI > 25 kg/m2, presence of HTN, DM and hyperlipidemia
Teshome et al. [31]	245	26.9	Female gender, age > 40 years, overweight, obesity and duration of mental illness more than 5 years
Ataro et al. [43]	375	26.7	Age over 40 years, lack of formal education and using tenofovir-lamivudine-lopinavir/ritonavir regimens
Belete et al. [23]	159	53.5	Increased age, female gender, high BMI, having diabetes for over 5 years and poor glycemic control
Tadewos et al. [29]	270	45.9	Female gender, duration of DM since its diagnosis, overweight and obesity
Gebremeskel et al. [44]	419	51.1	Female gender, age, physical inactivity, inadequate intake of fruits, family history, overweight, and obesity, duration of DM > 10yrs, presence of family history of DM and presence of chronic disease
Berhane et al. [33]	313	21.1	Taking the antiretroviral therapy for more than 12 months and male gender
Mengesha et al. [45]	872	9.5	Increased age, female gender overweight and obesity, high waist circumference, and physical inactivity
Bune et al. [30]	633	22	Increased age and female gender
Tesfaye et al. [32]	374	23.8	Female gender, increased age, using D4T + 3TC + EFV regimen, overweight and obesity, and having total cholesterol of at least 200 mg/dl
Kerie et al. [28]	534	9.6	Female gender, high educational status, physical inactivity, increased age
Birarra et al. [41]	256	57	Female gender
Tachebele et al. [40]	300	39.3	Female gender, high educational status, abnormal BMI which included both high and low BMI
Bosho et al. [34]	268	20.5	Having formal education and body mass index above 25 kg/m2
Hirigo et al. [26]	185	24.3	Female gender and BMI > 25 kg/m2, ART use more than 48 months, and high educational status
Cheneke et al. [42]	365	27.7	Female gender, women with age ≥ 40 years, having high income status and hormonal contraceptive use for more than 42 weeks

Publication bias

In our estimates, publication bias was assessed using funnel plots under the fixed-effects model, which helped us visualize each funnel plot's symmetry status. In principle, visual assessment of the effect estimates from larger studies that spread narrowing at the top of the plot, with more broadly dispersed estimates at the bottom of the plot among smaller studies, may inform the existence of bias [46]. The exclusion of unpublished research and the inclusion of various individual analyses with a range of sample sizes in our study may have affected and maximized publication bias risk [47] (Fig. 4.).

Fig. 4.

Publication bias using funnel plot of standard error by logit rate

Discussion

A MetS is a group of abnormal laboratory and physical findings, such as dyslipidemia, hypertension, glucose intolerance, and central obesity due to the integration between biochemical, clinical, physiological, and metabolic factors [48]. This systematic review and meta-analysis aimed to determine the prevalence of MetS in Ethiopia. Accordingly, the pooled prevalence of MetS in Ethiopia was found to be 30.0%. This is in agreement with similar studies conducted in the Latin American population (24.9%) [49], Gulf Cooperation Council countries (27.3%) [50], Bangladesh (30%) [6] and USA (33%) [51]. However, the finding of this study is greater than the study conducted in Ghana, 21.2% [52] and India (19.52%) [53] and lower that the study conducted in Saudi Arabia (39.8%) [54]. The possible explanation for this discrepancy might be due to the difference in life style and feeding habit among the populations.

In subgroup analysis, we found a higher prevalence of MetS among type 2 diabetes mellitus patients (56%) when compared to other subgroups. The possible explanation might be due to the fact that, type 2 diabetic patients are insulin resistant and usually obese. The two risk factors: Obesity and insulin resistance, are highly interconnected and are considered as the principal defects underlying the pathophysiology of MetS [10]. Blood lipids also play a vital role in the development of MetS among diabetic patients. Because of the impaired action of insulin, there is an increase in triglyceride hydrolysis and release of non-esterified fatty acids, associated with alteration of blood lipid concentration and metabolism, resulting in dyslipidemia [55].

Other subgroups with a higher prevalence of MetS were hypertension patients (44%). This increased prevalence could be attributed to antihypertensive medications. As evidenced by different kinds of literature, antihypertensive drugs like; thiazide diuretics and beta-blockers increase total cholesterol, low-density lipoprotein cholesterol, and triglyceride levels, particularly at high doses, and contribute to insulin resistance and worsening glycemic control in hypertension patients with diabetes [56, 57].

The pooled prevalence of MetS among HIV patients was 26%, which is in line with a similar meta-analysis conducted in the global HIV population [11]. This might be due to the fact that long-term use of ART drugs and HIV itself can cause lipid abnormalities [14]. The introduction of ART to the body leads to increased levels of TNF-α, which, in turn, impairs the metabolism of fatty acids and lipid oxidation, resulting in suppressed lipolysis. This, in turn, results in altered fat distribution and subsequent lipid profile changes [12]. Moreover, rapid weight gain may occur following the uptake of ART, with the use of protease inhibitors and nucleoside reverse transcriptase inhibitors, leading to body fat depositions in the abdominal and dorsal regions; this may further contribute to the development of MetS among HIV-infected individuals [58, 59]. Furthermore, the prevalence of MetS is lower in studies conducted in Addis Ababa as compared to the regional studies. This might be due to the difference in the level of awareness and life style of the population.

Female gender, increased age, obesity, overweigh, high educational status, income status, taking treatments for chronic diseases like, diabetes mellitus, hypertension and HIV/AIDS, and physical inactivity were the commonly encountered predictors of MetS in this review regardless of the population studied. Females are at higher risk of developing MetS than males. Physiological events such as puberty, pregnancy, and menopause are closely related to alterations in energy homeostasis and gonadal steroid levels during women’s lives. An increase usually follows these events in insulin resistance and body fat, important components of MetS. Besides, the use of hormonal contraceptives and pathological conditions such as polycystic ovary syndrome and gestational diabetes, which also present with alterations in gonadal steroid levels and increased insulin resistance, might contribute to the occurrence of MetS in women [60].

Increased age and physical inactivity are also predictors of MetS. Physical inactivity, a frequent finding in aged and obese individuals, is associated with negative metabolic consequences such as decreased insulin sensitivity and increased abdominal fat [19]. Obesity and overweight, which are associated with insulin resistance and oxidative stress, play crucial roles in the pathophysiology of this syndrome [20].

High-income and educational status are also another predictor variables on the occurrence MetS across all study populations in this review. The possible mechanisms for the impacts of socioeconomic and educational status on MS maybe they lead to unhealthy behaviors such as smoking and alcohol consumption, easier access to high-calorie foods causing metabolic changes, and easy access to transportation leading to sedentary lifestyles [3, 61, 62].

The strengths of our systematic review include a complete literature search in the various relevant database (PubMed, EMBASE, CINAHL, Scopus, Google, and Google scholar) and proper screening of eligible studies by two independent reviewers. Our review has the following limitations; the meta-analysis result should be interpreted cautiously due to heterogeneity among studies. Finally, we acknowledge that we may not have been able to retrieve unpublished data and grey literature.

Conclusion

MetS was reported in about one-third of Ethiopians. Females, old age groups and people with co-morbidities like type 2 diabetes, hypertension, and HIV/AIDS had higher prevalence of MS. Early preventive measures are required to address the risk components of MetS such as obesity and hypertension which are rapidly rising in Ethiopia.

Abbreviations

ACSH

Ayider comprehensive specialized hospital

ART

Antiretroviral therapy

BMI

Body mass index

CI

Confidence interval

CINAHL

Cumulative Index to Nursing and Allied Health Literature

HIV

Human immune virus

HUCSH

Hawassa University comprehensive Specialized Hospital

JUSH

Jimma University specialized hospital

MetS

Metabolic Syndrome

NCD

Non-communicable disease

PRISMA

Preferred Reporting Items for Systematic Reviews and Meta-Analyses

SPHMMC

St. Paul’s Hospital Millennium Medical College

UOGCSH

University of Gondar comprehensive specialized hospital

WHO

World Health Organization

Author’s contribution

BK, and TJ contributed to conceptualization, data curation, investigation, formal analysis, methodology, writing original draft, review, and editing of the manuscript. All authors approved the submitted version of the manuscript critically.

Funding

The author(s) received no financial support for the research, authorship, and publication of this review.

Data availability

The datasets used during the current study are available from the corresponding author on a reasonable request.

Declaration

Conflicting of interests

The Authors declare that there is no conflict of interest.