Introduction

Dyslipidaemia, a major modifiable risk factor for cardiovascular diseases (CVD), is characterised by abnormal levels of lipids and lipoproteins in the blood, including elevated total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), triglycerides (TGs), and/or reduced high-density lipoprotein cholesterol (HDL-C) [1]. CVD remains the leading cause of morbidity and mortality globally, accounting for an increasing proportion of deaths in low- and middle-income countries (LMICs) [2,3].

In sub-Saharan Africa (SSA), the burden of dyslipidaemia has risen substantially in recent times due to rapid urbanisation, dietary transitions, and increasingly sedentary lifestyles [4,5]. A study by Noubiap et al. reported continental prevalence estimates of 25% for elevated TC, 37% for low HDL-C, 28.6% for elevated LDL-C, and 17% for elevated TG, with considerable inter-country variation [6]. Despite these figures, national-level data for several African countries, such as Ghana, remain limited or fragmented.

Ghana, like several countries in the region, is becoming a population in health transition, experiencing a significant epidemiological shift marked by a growing burden of non-communicable diseases (NCDs), including cardiovascular disorders [7]. Lifestyle and nutritional changes associated with urbanisation have contributed to the increased prevalence of cardiometabolic risk factors, including obesity, hypertension, diabetes, and dyslipidaemia [5,8,9]. Several studies conducted across different populations have reported varying dyslipidaemia frequencies, reflecting differences in study design, diagnostic criteria, and population characteristics. However, despite these individual studies, a consolidated national synthesis of dyslipidaemia prevalence in Ghana is still lacking.

Therefore, this systematic review aimed to synthesise the available evidence on the burden and patterns of dyslipidaemia among adult Ghanaians, focusing on the reported prevalence, lipid profile subtypes, and variations in demographic and clinical factors. By collating and analysing existing studies, this review seeks to provide a clearer picture of the national situation and identify critical knowledge gaps that can inform policy decisions, health interventions, and future research on cardiovascular disease prevention in Ghana.

Materials and methods

Study selection

All records retrieved from the database searches were imported into Covidence (Veritas Health Innovation, Melbourne, Australia) for systematic review management. Duplicate records were automatically identified and removed from the platform before screening. Two reviewers (AD and ABA) independently screened the titles and abstracts of all the retrieved articles to identify studies that met the inclusion criteria. Full-text versions of the eligible papers were then retrieved and assessed in detail by two additional reviewers (SMA and ENYN) against the predefined inclusion and exclusion criteria. Any disagreements were resolved through discussion, and unresolved conflicts were adjudicated by a third reviewer (ROA).

A total of 247 records were identified through database searches, and no additional records were obtained from other sources. After removing 145 duplicates, 102 records were retained for further screening. After title and abstract review, 40 irrelevant records were excluded. The full texts of the 62 articles were retrieved and assessed for eligibility. Ultimately, 24 studies fulfilled all inclusion criteria and were included in the final synthesis.

The selection process is shown in Fig 1.

Download:

• PNG
  larger image
• TIFF
  original image

Fig 1. PRISMA flow diagram of the study selection process.

The diagram summarises the number of studies identified, screened, excluded, and included in accordance with PRISMA 2020 guidelines.

https://doi.org/10.1371/journal.pone.0350185.g001

Results

Study characteristics

A total of 24 studies published between 2003 and 2025 met the inclusion criteria and were included in this review. The characteristics of the included studies are summarised in Fig 1 and S2 Table, which outline the study locations, populations examined, and lipid parameters measured. Most studies were conducted in urban hospital settings, whereas a few included community-based or mixed (urban and rural) populations (Fig 2B and 2C). These studies represented all major geographical regions of Ghana, with the Ashanti, Greater Accra, and Central Regions being the most frequently studied (Fig 2D).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 2. Characteristics of studies included in the systematic review.

Panel A shows the publication year of the included studies. Panels B and C illustrate the study locations and settings. Panel D depicts the number of studies per region. Panels E and F show the sex and age distributions of study participants, and Panel G summarises the population categories recruited across studies (Numbers represent absolute counts).

https://doi.org/10.1371/journal.pone.0350185.g002

The sample size ranged from 70 to 3,317 participants, and the mean or median age ranged from 20 to 70 years. More females than males were recruited in all studies (Fig 2E). Both apparently healthy individuals and patient populations were represented, with no significant difference in the mean age (Fig 2F). A higher proportion of the studies focused on healthy controls, while others investigated patients with diabetes mellitus, HIV/AIDS, individuals with hypertension, or mixed clinical populations (Fig 2G). Across the included studies, 11.7% of the total study population were diabetics, while 5.8% had HIV/AIDS, and 1.0% were reported to have both HIV/AIDS and dyslipidaemia. Regarding the diagnostic criteria for dyslipidaemia, approximately two-thirds of the studies used the NCEP ATP III criteria. In contrast, the remaining studies defined dyslipidaemia according to ESC, ADA, or local institutional guidelines.

Anthropometric characteristics of study participants

To examine the distributions of anthropometric variables, participants were grouped by diagnosis. ANOVA was used to examine differences between the patient and control groups. The analysis showed that the patients had BMI, waist circumference, and hip circumference values similar to those of controls (Fig 3). Subsequently, no significant difference in waist-to-hip circumference ratio was observed between the groups. However, the diabetes and hypertension groups had significantly higher systolic and diastolic blood pressure readings than the control group (Figs 3F and 3G).

Download:

• PNG
  larger image
• TIFF
  original image

Fig 3. Anthropometric characteristics of study participants.

The figure presents anthropometric variables retrieved from the included records: (A) BMI; (B) waist circumference; (C) hip circumference; (E) waist-to-hip ratio; (F) systolic blood pressure; and (G) diastolic blood pressure.

https://doi.org/10.1371/journal.pone.0350185.g003

Lipid profile of study participants

Most studies (approximately two-thirds) used the National Cholesterol Education Program Adult Treatment Panel III (NCEP ATP III) criteria to define dyslipidaemia, whereas a few applied guidelines from the European Society of Cardiology (ESC), the American Diabetes Association (ADA), or local diagnostic standards such as those used at the Korle-Bu Teaching Hospital (KBTH). Nearly all studies reported data on TC, TG, LDL-C, and HDL-C, although some failed to specify the cut-off thresholds (S2 Table). Overall, the included studies demonstrated substantial heterogeneity in study populations, diagnostic criteria, and settings, reflecting the diversity of dyslipidaemia research in Ghana.

Lipid profile analysis revealed no significant differences in total cholesterol or LDL-C levels between the control and patient groups (Fig 4A, B). However, the HDL-C and TG levels were significantly different between the diabetic and control groups (Fig 4C, D). Although the VLDL-C levels were higher in the diabetes/hypertension and HIV groups than in the control group, these differences were not statistically significant (Fig 4E). The proportions of individuals with high TC, LDL-C, TG, and low HDL-C levels (based on retrieved records) were evaluated to compare the participant groups. Similarly, the frequency of low HDL-C levels was also assessed in these groups. Of all the variables examined, only high TC levels showed a statistically significant difference among the groups (p = 0.021) (Table 1).

Download:

• PNG
  larger image
• TIFF
  original image

Table 1. Distribution of Lipid Profile Abnormalities Across Participant Groups.

https://doi.org/10.1371/journal.pone.0350185.t001

Download:

• PNG
  larger image
• TIFF
  original image

Fig 4. Lipid profiles of the participants.

Panels A–E illustrate the distribution of lipid parameters across participant groups. Total cholesterol (A), LDL-cholesterol (B), HDL-cholesterol (C), triglycerides (D), and VLDL-cholesterol (E) were compared between controls and clinical groups, including diabetes/hypertension and HIV.

https://doi.org/10.1371/journal.pone.0350185.g004

Prevalence of dyslipidaemia

The reported prevalence of dyslipidaemia varies substantially across studies and population groups. Among apparently healthy adults, the overall prevalence ranged from 4.0% to 62.8%, whereas studies involving individuals with chronic conditions, such as diabetes mellitus or HIV infection, tended to show higher rates. Across most studies, low HDL-C was the most frequently reported abnormality, followed by elevated TC and LDL-C, whereas hypertriglyceridaemia was the least common lipid disturbance.

The prevalence of individual lipid abnormalities reported in the included studies is summarised in Table 2. Considerable variation was observed across the regions and study populations. Elevated total cholesterol ranged from approximately 2.3% to 62.8%, while elevated cholesterol ranged from 2.3% to 72.4%. Hypertriglyceridaemia generally showed lower prevalence across studies, ranging from about 2.1% to 48.4%. In contrast, HDL-C emerged as one of the most frequently reported lipid abnormalities, with prevalence estimates ranging from 7.6% to 91.5%. These findings highlight substantial heterogeneity in lipid abnormality patterns across the included studies.

Download:

• PNG
  larger image
• TIFF
  original image

Table 2. Regional distribution of dyslipidaemia and major lipid fractions reported in the included studies.

https://doi.org/10.1371/journal.pone.0350185.t002

Overall, the distribution demonstrated considerable geographical variation in the burden and patterns of dyslipidaemia across Ghana, with a higher prevalence generally observed in the Southern and Middle Belts than in the Northern regions. However, the wide variability in reported lipid parameters across studies makes it challenging to derive a precise national estimate of the prevalence of dyslipidaemia.

Discussion

This systematic review sought to determine the available evidence on the burden and patterns of dyslipidaemia among adult Ghanaians. The findings revealed wide variation in reported dyslipidaemia prevalence across studies, ranging from 3.0% to 72.4%, reflecting differences in study design, diagnostic criteria, population characteristics, and geographic coverage. Low HDL-C consistently emerged as the most common lipid abnormality, followed by elevated total cholesterol and LDL-C, whereas hypertriglyceridaemia was the least frequent. Overall, the available evidence indicates a substantial burden of dyslipidaemia in Ghana, particularly in urban and clinical populations, consistent with the broader epidemiological transition occurring in the country.

The prevalence patterns observed in this review broadly align with the regional trends reported across sub-Saharan Africa. In a meta-analysis by Noubiap et al. (2018), the pooled prevalence of low HDL-C was 37%, elevated total cholesterol was 25%, and elevated LDL-C was 28.6%, indicating that low HDL-C remains the dominant lipid abnormality across the continent [6]. However, this also raises the question of whether a comprehensive study should be conducted to categorise the SSA ranges. More importantly, this finding reveals a population-level disparity in lipid metabolism and environmental risk profiles that differs from Western trends, in which elevated TC is the most prevalent form of dyslipidaemia in Europe (54%) and the United States (48%) [12]. The implications for health outcomes are significant in this regard. HDL-C plays a crucial role in transporting cholesterol away from the arterial walls, thereby limiting the progression of atherosclerosis through anti-inflammatory and antioxidant pathways. The observed high prevalence of low HDL-C levels supports previous studies that reported a high burden and prevalence of cardiovascular diseases in the Ghanaian population [8,13]. Similarly, the significantly higher systolic and diastolic blood pressures in the diabetes and hypertension groups relative to the controls underscores the shared pathophysiology that links metabolic dysregulation, dyslipidaemia, and hypertension. The dominant prevalence of low HDL-C over elevated LDL-C levels also highlights the importance of validating Western-developed therapies in non-Western populations. In particular, statins, which are used to reduce cardiovascular risk and lower LDL-C levels, have limited effects on HDL-C levels [14]. Therefore, cardiovascular prevention strategies targeting HDL-C levels, such as lifestyle modifications, may require a greater emphasis in the Ghanaian context.

Studies that explored dyslipidaemia in people living with HIV/AIDS found elevated levels of VLDL-C relative to controls; however, this difference did not reach statistical significance. The impact of the virus on patients’ immunity and ongoing antiretroviral therapy may have introduced uncontrolled variables, contributing to the lack of statistical significance. Notably, Ghana shifted away from older Non-Nucleoside Reverse Transcriptase Inhibitor-based treatments towards Dolutegravir-based regimens as a first-line regimen in 2019 [15], whereas many of the studies sampled in this review were published well before that year. Additionally, the included studies reported specific antiretroviral regimens, treatment duration, and viral suppression status with varying levels of detail. Therefore, given that earlier Nucleoside Reverse Transcriptase Inhibitors (NRTIs), Non-Nucleoside Reverse Transcriptase Inhibitors (NNRTIs), and Protease Inhibitors (PIs) are known causes of altered lipid metabolism leading to dyslipidaemia [16], variation in antiretroviral treatment may have contributed to the observed lack of significance. Another complication in this patient population is that HIV infection, as a result of chronic inflammation, independently affects lipid metabolism and leads to dyslipidaemia [17]. Unfortunately, the included studies did not clearly distinguish between HIV-related and antiretroviral-related dyslipidaemia.

The current study also found a stark regional disparity in dyslipidaemia prevalence between the Northern Regions of Ghana and the Southern and Middle Belts. This observation parallels Ghana’s economic development and urbanisation trends. Indeed, studies in LMICs have identified a two-fold risk of developing cardiometabolic conditions as a result of residing in urban regions [18,19], which is associated with physical inactivity, smoking and urban dietary patterns. However, this observation must be interpreted in light of significant sampling bias, as 67% of participants were recruited from the urban south, namely the Ashanti, Greater Accra, and Central Regions of the country. The apparent regional disparities in dyslipidaemia prevalence may therefore point to differences in research infrastructure, access to healthcare, and health-seeking behaviour rather than population-level differences in disease burden. Another limitation arises from the overrepresentation of clinical populations compared with participants from community settings. Among the 24 included studies, 67% of the study population was hospital-based, and 33% were recruited from community settings. This likely inflated the prevalence estimates by oversampling individuals already seeking healthcare for metabolic, cardiovascular, and other chronic disorders. Without data on the true population-level burden in rural communities and the northern regions of Ghana, it is difficult to determine whether the results reflect genuine geographical variation in dyslipidaemia prevalence.

Study heterogeneity posed a challenge in synthesising data from the included studies. Differences in the diagnostic criteria used to define dyslipidaemia likely contributed to the wide prevalence range observed (4.0%–62.8%). Similar variability has been reported across sub-Saharan Africa, where studies employ different definitions, most commonly the NCEP ATP III criteria, alongside modified WHO thresholds or study-specific cut-offs. Across the region, dyslipidaemia prevalence varies substantially depending on the population studied, with community-based studies typically reporting prevalence around 20–25%, while much higher rates have been documented in high-risk groups such as individuals with hypertension (48–73%), HIV infection (~69%), or diabetes (up to 73–88%). These regional patterns are broadly consistent with the variability observed in the Ghanaian studies included in this review and likely reflect the ongoing epidemiological transition across many African countries, characterised by urbanisation, changing dietary patterns, and increasing cardiometabolic risk factors. The lack of standardised diagnostic criteria also complicates comparisons across studies and may limit the applicability of international guidelines developed in Western populations to African contexts with different baseline lipid profiles and cardiovascular risk patterns. Consequently, the heterogeneity of the available data precluded a formal meta-analysis.

Across the reviewed studies, sex-specific dyslipidaemia prevalence data were inconsistently reported, and the included studies recruited more females than males. This limits rigorous analysis of sex differences, which is of particular interest because sex differences in lipid profiles do exist. For instance, oestrogen is known to contribute to relatively favourable lipid profiles in premenopausal women, and these protective effects on lipid metabolism diminish after menopause [20]. Sex differences also exist in lifestyle risk factors for dyslipidaemia, such as dietary patterns, physical activity, alcohol consumption, and smoking. Taken together, the underreporting of sex-stratified data in the original studies limits the ability to clarify the true magnitude of sex differences in the dyslipidaemia burden.

In Ghana, dyslipidaemia is most often detected opportunistically during routine clinical care rather than through structured population-based screening programmes. Many adults are first diagnosed when seeking care for related conditions such as hypertension, diabetes, or other cardiometabolic disorders, or during workplace and community health screening events. Although Ghana’s National Health Insurance Scheme (NHIS) covers many essential health services, routine lipid screening for asymptomatic adults is not consistently implemented across healthcare settings. Consequently, dyslipidaemia frequently remains undiagnosed until individuals present with other cardiovascular risk factors or complications. The findings of the present review, therefore, highlight the need for improved surveillance and earlier detection strategies, including integrating lipid screening into existing non-communicable disease prevention programmes and into routine primary healthcare services. Such approaches align with the World Health Organisation’s recommendations for strengthening integrated cardiovascular risk screening and management within primary healthcare systems in low- and middle-income countries.

This systematic review shows that dyslipidaemia appears to be common among adult Ghanaians, with low HDL-C emerging as the dominant form across diverse study populations. However, precise national prevalence estimates are limited by the significant methodological heterogeneity and sampling bias. The evidence reviewed disproportionately represents urban clinical populations, whereas rural and non-clinical populations remain inadequately covered. Another limitation is the limited reporting of treatment and disease control in the included studies. Few studies provided information on whether dyslipidaemia, diabetes, or HIV infection were adequately controlled, and details regarding the class/ types of medication used were rarely reported. This lack of information limited the ability to evaluate treatment patterns. Nevertheless, the available data suggest that dyslipidaemia is common, particularly among Ghanaian adults, and given the established role of poor lipid metabolism in the development of cardiovascular disease, this warrants public health attention. This involves implementing nationally representative surveillance using standard, context-relevant diagnostic criteria for determining clinically significant dyslipidaemia, conducting research that addresses the identified knowledge gaps, and consolidating the lessons learned into comprehensive cardiovascular disease prevention strategies that are informed by locally relevant data rather than the uncritical adoption of approaches developed in other settings.

Conclusion

This systematic review highlights that dyslipidaemia is highly prevalent among adult Ghanaians, with low HDL-C emerging as the most common lipid abnormality across diverse populations. The wide variation in prevalence across studies reflects methodological heterogeneity, differing diagnostic criteria, and predominance of data from urban and clinical populations. Nevertheless, overall evidence points to a growing cardiometabolic burden in Ghana, paralleling ongoing lifestyle and nutritional changes. Addressing this challenge will require harmonised diagnostic standards, population-based surveillance, and the integration of lipid screening into existing non-communicable disease prevention programmes. Future research should prioritise community-level studies that use context-appropriate lipid thresholds to improve the accuracy of national estimates and guide locally relevant cardiovascular risk-reduction strategies.

Supporting information

Acknowledgments

The authors acknowledge institutional support that facilitated access to the electronic databases and full-text articles used in this review. The authors also appreciate the technical and administrative teams within the contributing institutions for providing an enabling environment during the conduct of this work.