Malnutrition among men in Ghana: socio-demographic factors and rural-urban differences

Abstract

Background

Research on malnutrition in sub-Saharan Africa, including Ghana, has primarily focused on women and children, with less attention paid to men. We estimate the prevalence and examine the correlates and place differentials in malnutrition among Ghanaian adult males aged 20 to 59 years.

Methods

We analyzed data from the seventh round of the Ghana Living Standards Survey. The analytical sample includes 8,117 (weighted) adult males who had valid data on anthropometric measures and sociodemographic characteristics of interest. Malnutrition was assessed using respondents’ body mass index (BMI). Respondents were classified as being underweight, normal weight, overweight/obese using the standard World Health Organization BMI cut-off points. Descriptive and multinomial logistic regression analysis techniques were employed in analyzing the data.

Results

One in four (25.3%) adult males aged 20–59 years in Ghana were identified to be overweight/obese. Increasing age, being married, having at least junior high school education, being of middle or high wealth status and being a Christian or Moslem were associated with increased odds of being overweight/obese while living in a rural area and having access to improved water sources were associated with a lower likelihood of being overweight/obese.

Conclusion

Interventions on lifestyle behavior changes aimed at addressing overweight and obesity, including eating healthy diets and being physically active should target married, educated and wealthy men who are most at risk of being overweight/obese. Additionally, men need to be educated on the increased risk of overweight/obesity associated with increasing age and encouraged to adopt healthier lifestyles as they age.

Introduction

Malnutrition, including both undernutrition and overnutrition, is a global public health concern affecting adults - both males and females, and children. The age-standardized global prevalence of underweight was estimated to have declined by about 5 percentage points from 13.8% to 8.8% among men and 14.6% to 9.7% among women, between 1975 and 2014 while the estimates for obesity show an increase from 3·2% to 10·8% among men, and from 6·4% to 14·9% among women [1]. While the global prevalence of underweight has been observed to be declining generally, the prevalence of obesity among adults worldwide has doubled since 1990 [1]. The World Health Organization (WHO) estimates that in 2022, 2.5 billion adults were overweight, and this includes 890 million who were living with obesity, while 390 million were underweight [2].

Globally, the prevalence of overweight/obesity and diet-related non-communicable diseases is on the increase while undernutrition and micronutrient deficiencies continue to persist in southeast Asia and sub-Saharan Africa. And while urban areas in Africa are recognized to be more obesogenic compared to rural areas [3–5], recent trends indicate that unique factors are contributing to a similar increase in obesity rates even among rural populations [3].

Globally, the statistics on malnutrition are often provided for women and children while those for men are either limited or non-existent in some contexts, especially in sub-Saharan Africa. Also, the narrative around malnutrition in sub-Saharan Africa has primarily focused on women and children, with less attention paid to men [6]. However, evidence over the past four decades show that the prevalence of obesity has risen among both women and men across all regions, including sub-Saharan Africa [7]. Also, although the prevalence of obesity is typically higher among females compared to males, recent evidence indicates that the gap in the prevalence of obesity between males and females is closing [8]. Arinda et al. (2021) [6] report that there is no single country in the African region that is on course to meet the targets for over-nutrition among men. In Ghana, previous research indicates that the prevalence of obesity is higher among females compared to males [9–12]. The results of a recent systematic review and meta-analysis estimates the pooled overweight prevalence to be 16.5% among males and 25.9% among females [13].

Overweight and obesity are, however, not the only conditions of malnutrition that affect Ghanaian male adults. Undernutrition also affects Ghanaian males, although previous research shows a persistence of undernutrition among females and children under-five [14, 15]. The 2022 Global Nutrition Report shows that between 2000 and 2016, the prevalence of underweight was higher among males than females and projections between 2016 and 2020 suggested a sustained higher prevalence of underweight among males compared to females [16]. Some localized studies have also found a higher prevalence of underweight among males compared to females. For instance, Nonterah et al. (2018) found that underweight was more prevalent among men compared to women (18.3% versus 13.1%) among an adult population in rural Northern Ghana [17].

Despite the high prevalence of underweight among Ghanaian male adults, the diminishing gap in the prevalence of overweight/obesity between females and males and the rising prevalence of overweight/obesity among males [18], only a few studies have examined the prevalence and predictors of malnutrition among adult males in Ghana. Furthermore, there remains a paucity of research investigating the socio-demographic correlates of malnutrition (undernutrition and overnutrition) and place differentials in malnutrition among males. This study uses nationally representative data to estimate the prevalence, socio-demographic predictors and urban rural differentials in malnutrition among adult males in Ghana.

This study conceptualizes men’s nutritional status as an outcome shaped by structured and interdependent determinants operating across individual, intermediate, and proximal levels of influence. Thus, the study conceptualizes malnutrition among males not only as an individual-level outcome, but also shaped by structural, household, and environmental pathways [19, 20].

At the individual level, socioeconomic factors such as age, education and occupation, shape access to resources and exposure to nutrition-related risks. Evidence shows strong socioeconomic and age-related gradients in body mass index and nutritional outcomes, reflecting both biological processes and structural inequalities [21, 22]. At the intermediate level, structural determinants are expressed through household conditions, including household wealth, place of residence, and access to water and sanitation. Household wealth plays a central role in shaping diet quality, food access, and energy expenditure patterns, while marital status influences dietary practices, social support, and health-related behaviors [23, 24]. Place of residence further mediates nutritional outcomes by structuring access to food environments, healthcare, and lifestyle opportunities within the context of ongoing nutrition transitions [20].

Environmental conditions, particularly access to safe water and adequate sanitation are critical pathways linking household context to nutritional status, primarily through their effects on infectious disease exposure, gut health, and nutrient absorption [19, 25].

Understanding the predictors of malnutrition among men and the place-based variations is crucial due to the differing socio-environmental, economic, and cultural factors that shape health behaviors and access to resources. These disparities may allow the identification of unique risk profiles for malnutrition and inform tailored interventions and public health strategies for addressing malnutrition among Ghanaian adult males.

Methodology

Study area and design

The study was conducted in Ghana using nationally representative data from the Ghana Living Standard Survey (GLSS 7), which covers both rural and urban areas across all regions of the country. Data collection for GLSS 7 was carried out between October 2016 and October 2017. The survey employed a two-stage stratified sampling technique. In the first stage, enumeration areas (EAs) were selected based on the 2010 Population and Housing Census, with 1,000 EAs serving as the primary sampling units (PSUs). In the second stage, 15 households were systematically selected from each PSU as the secondary sampling units (SSUs), based on household listings in both rural and urban areas.

Out of approximately 15,000 selected households across the 1,000 EAs, 14,009 households responded to the survey, yielding a response rate of 93.3%. Members of selected households were interviewed, and data on socio-demographic characteristics as well as anthropometric measures (weight and height) were collected. Anthropometric measurements were obtained through direct measurement using calibrated scales and stadiometers to ensure accuracy and consistency.

Study population

For the present study, males aged 20 to 59 years with complete socio-demographic information and valid anthropometric measurements were included to facilitate accurate computation of Body Mass Index (BMI). A final analytic sample of 8,117 men aged 20 to 59 years was obtained after data cleaning (see Fig. 1). Eligibility criteria were sequentially applied as follows:

Fig. 1.

Flow diagram showing the sample selection process

• The sample was first restricted to individuals aged 20–59 years (n = 11,189), reflecting standard practice in adult population studies to minimize age-related heterogeneity in anthropometric outcomes.

• Respondents without anthropometric measurements (n = 2,223) were excluded, as missing height and weight data preclude the construction of the key outcome variable, BMI.

• A total of 784 observations with implausible anthropometric values were excluded using established outlier detection criteria. The removal of biologically implausible height and weight measurements is widely recommended in epidemiological research to address potential measurement and data entry errors that may bias statistical estimates (Boone-heinonen et al., 2019) [26]. Specifically, observations with heights < 136 cm or > 220 cm and weights < 50 kg or > 200 kg were classified as implausible and excluded prior to analysis [27, 28].

• Finally, 65 respondents with missing socio-demographic covariates were excluded to ensure consistency in multivariable analyses. Given the small proportion of such cases, complete-case analysis is unlikely to introduce significant bias while maintaining interpretability.

Study variables

Dependent variable

The outcome variable of interest was the BMI of males aged 20 to 59 years calculated as weight in kilograms divided by height in meters squared (kg/m²). The study participants were classified into the respective BMI categories based on the standard World Health Organization (WHO) cut-off values as underweight = BMI < 18.5, normal weight = BMI 18.5–24.9, overweight = BMI 25.0–29.9.0.9, and obese = BMI ≥ 30.0 [29].

Independent variables

Building on the UNICEF framework for malnutrition [30], we incorporate diverse perspectives from Development Studies to conceptualize the basic determinants of malnutrition among men, recognizing that several demographic and socioeconomic factors could contribute to insufficient dietary intake and, consequently, malnutrition [31, 32]. The covariates explored in this study included the respondents age, level of education, marital status and occupation as well as the wealth status of the household they belong to. Other household variables controlled for include source of drinking water and type of toilet facility.

In the GLSS 7, household wealth was constructed using an approach similar to that used in the Demographic and Health Surveys [33], which uses a set of household assets which typically include the type of flooring material, source of water supply, type of sanitation facilities, availability of electricity, and ownership of key household assets such as a radio, television, telephone, and refrigerator, and type of vehicle. A wealth index was generated using these household assets and the households were categorized as being of low, middle or high status based on the computed wealth index. Access to an improved water source is defined as access to any of the following water sources: piped water, public taps, standpipes, tube wells, boreholes, protected dug wells, protected springs, or rainwater [34]. Households with improved sanitation are those that use a pour-flush toilet, ventilated improved latrine, composting toilet, or pit latrine with a slab. Households lacking these facilities are deemed as using unimproved sanitation facilities.

Estimation techniques

In examining the determinants of malnutrition, we employed a multinomial logistic regression model (mlogit in STATA version 18), which is suitable for analyzing categorical dependent variables with more than two categories. In specifying the model, the dependent variable, BMI was categorized into three groups: normal weight, underweight and overweight/obese with normal weight being the reference category. The mlogit model estimates the probability/relative risk of an individual being in each category relative to the reference group, based on a set of independent variables. The model was specified for the total sample and by place (urban versus rural). Given that the GLSS 7 employs a nationally representative, stratified two-stage sampling design, all analyses incorporated sampling weights and adjusted for clustering and stratification using the survey (‘svy’) procedures to ensure unbiased estimates and correct standard errors.

Results

Background characteristics of study respondents

All reported estimates account for the complex survey design of GLSS 7, including sampling weights, clustering, and stratification. Table 1 presents a description of the sociodemographic and anthropometric characteristics of the study sample. The mean age of participants was 36.0 years (95% CI: 35.6–36.3), with slightly higher values observed among rural residents (36.4 years; 95% CI: 36.0–36.9) compared to their urban counterparts (35.6 years; 95% CI: 35.1–36.0). Approximately 88% of the men in the study had some form of formal education. Majority of the respondents were in the high wealth category (52.9%), followed by those in the low (29.7%) and medium (17.4%) wealth categories. About 52% of the households had access to improved water sources, while another 52% used improved toilet facilities. The distribution by employment status shows that 12.3% were not employed, 10.2% were employed in highly skilled jobs, 4.7% were in medium-skilled jobs, 68.3% were in service sector jobs, and 4.4% were in elementary occupations. In terms of marital status, 34.6% of the sample were never married. The anthropometric data showed that 4.5% of the men were underweight, 70.2% had normal weight, 21.2% were overweight, and 4.1% were obese.

Table 1.

Sociodemographic and anthropometric characteristics of study participants

	Total (%) (N = 8,117)	Rural (%) (n = 4,731)	Urban (%) (n = 3,386)
Socio-demographics
Formal Education
None	12.2	19.8	6.0
Primary	11.1	14.6	8.3
JHS	37.1	40.4	34.5
SHS+	39.6	25.2	51.2
Household wealth
Low	29.7	49.1	14.1
Medium	17.4	18.7	16.3
High	52.9	32.2	69.6
Water source
Unimproved	47.5	36.5	56.3
Improved	52.5	63.5	43.7
Type of toilet
Unimproved	51.9	44.8	57.7
Improved	48.1	55.2	42.3
Occupation
Not employed	12.4	6.7	16.9
Highly skilled	10.1	6.0	13.4
Medium skilled	4.7	1.8	7.0
Service	68.4	81.5	57.8
Elementary	4.4	3.9	4.9
Marital status
Never married	34.8	27.1	40.9
Currently married	60.0	67.4	54.1
Ever married	5.3	5.5	5.2
Anthropometry
Body Mass Index
Underweight	4.5	5.8	3.5
Normal	70.2	79.2	63.0
Overweight	21.2	13.4	27.4
Obese	4.1	1.7	6.0
Age in years (mean, 95% CI)	36.0 (35.6, 36.3)	36.4 (36.0, 36.9)	35.6 (35.1, 36.0)

All values represent weighted percentages (frequencies)

Source: Computed from Ghana Living Standards Survey, 2016/2017

The urban rural disaggregation shows marked variations. For example, men in rural areas exhibit lower educational attainment, with 19.8% having no formal education compared to 6.0% in urban areas. Conversely, urban areas had a higher proportion of participants with secondary education or higher education compared to rural areas (51.2% vs. 25.2%). Approximately half of the participants who reside in rural areas (49.1%), were in low wealth households compared to just 14.1% in urban areas. Infrastructure differences are also evident as a higher proportion of men in rural households have improved water sources (63.5% vs. 43.7%) and toilet facilities (55.2% vs. 42.3%) compared to their urban counterparts. Employment trends also varied, with men in rural areas being dominant in service occupations (81.5%), while urban residents showed higher unemployment on the one hand (16.9%) and higher skilled jobs (13.4%) on the other hand. The anthropometric data reveal a greater prevalence of underweight among men in rural areas compared to urban areas (5.8% versus 3.5%) and higher obesity prevalence among men in urban areas compared to their counterparts in rural areas (6.0% versus 1.7%).

Predictors of malnutrition status

Table 2 presents multinomial logistic regression findings for socio-demographic and place-based correlates of malnutrition among Ghanaian adult males aged 20–59 years (N = 8,117), with normal BMI as the reference category. The results are reported as relative risk ratios (RRR, 95% CI), representing the relative likelihood of underweight or overweight/obesity compared with normal BMI, and should be interpreted as associations rather than causal estimates.

Table 2.

Multinomial logistic regression results for socio-demographic correlates and place differentials in malnutrition among Ghanaian adult males, 20–59 years

	Overall Sample (N = 8,117)				Urban (n = 3,386)				Rural (n = 4,731)
	Underweight		Overweight/obesity		Underweight		Overweight/obesity		Underweight		Overweight/obesity
Marita status [Never married]
Currently married	1.10	(0.69–1.75)	1.82***	(1.47–2.25)	2.23*	(1.01–4.94)	2.00***	(1.54–2.60)	0.71	(0.44–1.17)	1.35	(0.95–1.93)
Formerly married	1.32	(0.70–2.50)	1.10	(0.73–1.68)	3.00	(0.95–9.30)	1.07	(0.62–1.84)	0.94	(0.45–1.99)	1.02	(0.59–1.77)
Formal education [None]
Primary	0.55*	(0.35–0.87)	1.22	(0.90–1.66)	0.37	(0.12–1.17)	0.88	(0.54–1.44)	0.69	(0.42–1.14)	1.58*	(1.10–2.28)
JHS	0.83	(0.54–1.29)	1.35*	(1.03–1.77)	0.60	(0.25–1.40)	0.98	(0.63–1.51)	1.11	(0.66–1.87)	1.66**	(1.21–2.28)
SHS +	1.06	(0.67–1.65)	1.52**	(1.13–2.06)	0.89	(0.37–2.11)	1.14	(0.72–1.80)	1.09	(0.60–1.98)	1.71**	(1.14–2.56)
Age	1.01	(0.99–1.04)	1.03***	(1.02–1.04)	0.95*	(0.91–0.99)	1.03***	(1.02–1.04)	1.05***	(1.03–1.07)	1.02*	(1.00–1.03)
Household wealth [Low]
Middle	1.01	(0.65–1.56)	1.24	(0.99–1.56)	1.12	(0.56–2.22)	1.17	(0.83–1.65)	1.03	(0.60–1.78)	1.17	(0.86–1.59)
High	0.86	(0.63–1.16)	1.45***	(1.19–1.77)	1.17	(0.68–2.02)	1.28	(0.95–1.73)	0.68	(0.45–1.05)	1.62***	(1.25–2.11)
Employment [Not employed]
High-Skilled	0.88	(0.44–1.75)	1.12	(0.82–1.54)	0.89	(0.38–2.06)	0.95	(0.66–1.37)	1.34	(0.48–3.78)	1.62	(0.91–2.91)
Medium-Skilled	0.45	(0.11–1.78)	0.99	(0.67–1.47)	0.33	(0.04–2.60)	0.99	(0.64–1.53)	1.22	(0.23–6.40)	0.92	(0.39–1.18)
Service and Manual	1.03	(0.66–1.59)	0.96	(0.74–1.23)	0.96	(0.53–1.73)	1.03	(0.77–1.37)	1.46	(0.69–3.06)	0.73	(0.46–1.15)
Elementary and other	1.15	(0.48–2.75)	0.96	(0.62–1.50)	1.99	(0.71–5.58)	0.94	(0.54–1.63)	0.63	(0.19–2.11)	0.91	(0.44–1.90)
Religion [None, Traditional, Others]
Christianity	0.92	(0.66–1.28)	1.37*	(1.06–1.77)	2.13	(0.88–5.25)	1.41	(0.96–2.08)	0.75	(0.50–1.13)	1.28	(0.92–1.78)
Islam	0.72	(0.45–1.15)	1.57**	(1.18–2.08)	1.32	(0.49–3.53)	1.59*	(1.05–2.40)	0.78	(0.42–1.45)	1.48*	(1.02–2.17)
Location [Urban]
Rural	1.17	(0.85–1.60)	0.43***	(0.36–0.51)
Water source [Unimproved]
Improved	0.95	(0.70–1.30)	0.70***	(0.60–0.80)	0.97	(0.59–1.57)	0.64***	(0.53–0.78)	1.01	(0.66–1.53)	0.74*	(0.59–0.93)
Type of toilet [Improved]
Unimproved	1.17	(0.84–1.62)	0.95	(0.82–1.11)	1.10	(0.67–1.80)	0.85	(0.69–1.05)	1.25	(0.81–1.93)	1.12	(0.89–1.42)

Values represent Relative Risk Ration (RRR) and confidence intervals in parenthesis

[] Reference category

* p < 0.05; ** p < 0.01; ***p < 0.001

Source: Computed from Ghana Living Standards Survey, 2016/2017

Overall Sample

Marital status was significantly associated with overweight/obesity but not underweight. Currently married men were more likely to be overweight/obese than their never-married counterparts (RRR 1·82, 95% CI 1·47–2·25; p < 0·001). Being formerly married showed no significant association with either outcome. A significant educational gradient was observed for both outcomes. Men with primary-level schooling were less likely to be underweight compared to those with no formal education (0·55, 0·35–0·87; p = 0·010). Conversely, JHS completion (1·35, 1·03–1·77; p = 0·031) and SHS or above (1·52, 1·13–2·06; p = 0·006) were each associated with a higher likelihood of overweight/obesity. Older age was independently associated with a greater likelihood of overweight/obesity (1·03, 1·02–1·04; p < 0·001), with no significant association for underweight.

Relative to men from low-wealth households, those from high-wealth households were more likely to be overweight/obese (1·45, 1·19–1·77; p < 0·001). Employment category was not significantly associated with either nutritional status outcome, with all confidence intervals crossing unity. Both Christian (1·37, 1·06–1·77; p = 0·017) and Muslim (1·57, 1·18–2·08; p = 0·002) affiliation were associated with a greater likelihood of overweight/obesity relative to men with no religion, traditional beliefs, or other affiliation.

Rural residence was strongly and inversely associated with overweight/obesity compared with urban residence (0·43, 0·36–0·51; p < 0·001), with no significant association for underweight. Access to an improved water source was associated with a lower likelihood of overweight/obesity relative to unimproved sources (0·70, 0·60–0·80; p < 0·001). Toilet facility type was not significantly associated with either outcome.

Urban sub-sample

Among urban men, current marriage was associated with a higher likelihood of both overweight/obesity (2·00, 1·54–2·60; p < 0·001) and underweight (2·23, 1·01–4·94; p = 0·047) relative to never-married men, although the latter estimate was imprecise. Formerly married men showed no significant associations. No educational level was significantly associated with underweight in urban areas. Similarly, no formal education category was significantly associated with overweight/obesity among urban men. Each additional year of age was associated with a lower likelihood of underweight (0·95, 0·91–0·99; p = 0·020) and a higher likelihood of overweight/obesity (1·03, 1·02–1·04; p < 0·001). Neither middle nor high household wealth was significantly associated with either nutritional outcome in urban areas. Employment type was not significantly associated with nutritional status among urban men. Muslim affiliation was associated with a greater likelihood of overweight/obesity relative to the reference religious group (1·59, 1·05–2·40; p = 0·029), whereas Christian affiliation did not reach statistical ssignificance. Neither affiliation was significantly associated with underweight. Access to an improved water source was associated with a lower likelihood of overweight/obesity (0·64, 0·53–0·78; p < 0·001), with no significant association for underweight or toilet facility type.

Rural sub-sample

Marital status was not significantly associated with either underweight or overweight/obesity among rural men. A clear educational gradient in overweight/obesity was however, observed among rural men. JHS completion (1·66, 1·21–2·28; p = 0·002) and SHS or above (1·71, 1·14–2·56; p = 0·009) were each associated with a greater likelihood of overweight/obesity relative to no formal education. Primary schooling showed no significant association with underweight. Older age was associated with a significantly greater likelihood of both underweight (1·05, 1·03–1·07; p < 0·001) and overweight/obesity (1·02, 1·00–1·03; p = 0·042). High household wealth was associated with a greater likelihood of overweight/obesity relative to low wealth (1·62, 1·25–2·11; p < 0·001) while middle wealth and underweight showed no significant associations.

Employment type was not significantly associated with either nutritional outcome among rural men. Muslim affiliation was associated with a greater likelihood of overweight/obesity relative to the reference religious group (1·48, 1·02–2·17; p = 0·042). Christian affiliation and underweight showed no significant associations. Access to an improved water source was associated with a lower likelihood of overweight/obesity (0·74, 0·59–0·93; p = 0·010), with no significant association for underweight. Toilet facility type was not a significant predictor in the rural sub-sample.

Discussion

In this study, we examined the prevalence and correlates of malnutrition among adult males in Ghana and identified urban-rural differences with respect to the correlates. Our findings reveal that approximately 1 in 4 (25.3%) Ghanaian males aged 20 to 59 years are overweight/obese with higher prevalence in urban areas (28.0%) compared to rural areas (15.0%). The higher prevalence of overweight/obesity in urban settings may be associated with factors such as sedentary occupations, greater availability of processed foods, and a shift toward westernized diets [35, 36]. Among the socio-demographic factors examined, only education was associated with both underweight (among the general sample) and overweight/obesity (among the general sample and the urban sub-sample). The observed association between education and BMI status is consistent with previous studies [37–39]. The plausible reasons for the paradoxical association between education and underweight on the one hand and education and overweight/obesity on the other hand may be because of the complex interplay of socioeconomic, behavioral, and cultural factors [40, 41]. Higher education often correlates with better health literacy and access to resources that promote healthy lifestyles, which can help prevent underweight and overweight conditions [40]. But while educated individuals are more likely to understand the importance of balanced nutrition and regular physical activity, leading to healthier body weight management, the demands and stress associated with higher education and professional careers can lead to unhealthy behaviors leading to an increased likelihood of being overweight/obese [42, 43]. For instance, long working hours and high-stress environments may result in poor dietary choices, lack of physical activity, and increased consumption of alcohol, all of which contribute to weight gain [40, 41]. In the present study, the paradoxical effect of education is evident in the differential effect it showed on malnutrition among men in urban and rural areas. Men with higher education in rural areas tend to have sedentary jobs, such as teaching or administration, potentially contributing to overweight/obesity compared to less educated men in rural areas who engage in labor intensive jobs. Furthermore, higher social status, linked to higher education, may lead to affluent lifestyle patterns, such as high-calorie diets and less manual labor, thus contributing to weight gain [39, 41, 43]. Additionally, cultural perceptions in rural areas might associate larger body sizes with prosperity [44, 45], leading educated men to prioritize weight gain, whereas urban areas might value thinness as part of modern lifestyles. On the contrary, educated men in urban areas may have more access to recreational activities and may be more aware of the negative effects of overweight and obesity.

Other socio-demographic factors, including marital status, age, location (i.e., rural versus urban), household wealth status, religion, and household access to improved water sources were also found to be associated with overweight/obesity. The findings revealed that being married is associated with increased odds of overweight/obesity among men with the effect being more pronounced among men in urban areas. Similar findings have been reported in two other studies. One cross-sectional and retrospective cohort analyses conducted in China showed that married individuals were at a higher risk of being overweight than their unmarried counterparts [46]. The two potential reasons for this finding as given by Liu et al. [46] were that individuals who have never been married may prioritize weight management as a way to appeal to future partners, and married individuals might be constrained by additional family responsibilities, leaving them less time for physical activity. This phenomenon is further corroborated by a study conducted in Iran, which found that, compared to their unmarried counterparts, married men are more likely to undergo changes in behavior, leading to less consistent weight control post-marriage [47].

The association between age and the increased risk of overweight/obesity is consistent among similar age brackets with broader epidemiological trends. Analysis of data from the Global Burden of Disease study demonstrates that the risk of overweight/obesity in adults increases and peaks between the ages of 50 and 54 years among men, with a drop afterwards [48]. Generally, physiological changes, decreased physical activity, and altered metabolism play a role in this age-related pattern [49]. While an increase in age is generally associated with a higher likelihood of being overweight or obese, men in urban areas may have better access to exercise facilities, reducing the likelihood of weight gain or even counteracting the effect.

Our findings also indicate that belonging to medium and high wealth households is associated with higher odds of overweight/obesity. This aligns with observed global patterns. For example, research by Aitsi-Selmi et al. [50] across diverse populations revealed that an increase in wealth was associated with increased odds of obesity in all countries studied. The assertion for this finding is that affluent households are more inclined towards sedentary lifestyles and often have greater access to energy-dense foods, contributing to weight gain [50]. This effect is, however, not observed among urban men compared to their rural counterparts as they may be more knowledgeable and educated about the consequences of unhealthy dietary practices and sedentary lifestyles.

Probable explanations for why access to improved water was associated with a reduced likelihood of overweight/obesity could be that access to clean and safe drinking water can lead to a reduction in the consumption of sugary beverages [51], which is a major contributor to weight gain and obesity. With clean water readily available, individuals are more likely to choose water over sugary drinks [51], particularly in urban areas. Additionally, given that safe water is crucial for preparing nutritious foods, such as properly washing fruits and vegetables, access to improved water may lower the risk of consuming obesogenic energy dense foods.

The findings of this study make useful contributions to the literature on malnutrition among adult males in Ghana. To the best of our knowledge, this is one of the few studies that uses a nationally representative sample to estimate the prevalence of malnutrition among Ghanaian adult males and further examines urban rural differentials. The findings are generalizable to the adult male population aged 20–59 years in Ghana and other African countries with a similar socio-demographic profile. These findings, however, need to be considered in light of some limitations. Firstly, the data is from 2016/2017, which is about eight years old, the estimated prevalence may thus not reflect the current prevalence. It is possible that the prevalence of overweight and obesity may have increased since 2016/2017. There is therefore the need to estimate the prevalence of the different conditions of malnutrition among men using more recent data. Secondly, some critical variables, particularly dietary practices and physical activity were not available in the GLSS 7 data. These variables were thus not included in the present paper. Including these variables in the analysis may have yielded different results and provided additional insights on the predictors of malnutrition among Ghanaian adult males. These limitations notwithstanding, the findings of the current study hold true and provide empirical evidence on the prevalence, correlates andrural urban differentials in malnutrition among Ghanaian adult males. Future studies should analyze more recent data and unpack the present findings further by investigating the influence of other variables including dietary practices and physical activity. Third, this study analyses cross-sectional data, the results therefore indicate associations and not causality. Causal relationships should therefore not be implied.

Conclusion

The findings of this study indicate that increasing age, marital status, higher education, middle or high wealth status, and religious affiliation are associated with a higher likelihood of overweight/obesity among Ghanaian adult males aged 20 to 59 years. In contrast, access to improved water sources appears to be associated with a reduced risk of overweight/obesity, particularly in rural areas. These findings emphasize the need for targeted health interventions to promote healthy dietary habits and physical activity, focusing on at-risk groups such as married, educated, and affluent men. The results also reveal striking differences in how education impacts overweight/obesity in rural versus urban contexts. These insights underscore the importance of designing interventions tailored to these contextual differences. Interventions aimed at promoting lifestyle behavior changes should target married, affluent, and educated men, encouraging healthier dietary choices and increased physical activity. Additionally, raising awareness about age-related risks of overweight/obesity is essential for fostering long-term health. Addressing rural urban disparities in access to health-promoting resources and countering culturally driven perceptions of body weight are also critical. By adopting context-specific strategies, policymakers can more effectively combat the growing burden of overweight/obesity among adult males in Ghana and similar developingcountry contexts.

Author contributions

FAAD and AKC conceptualized and designed the study. AKC analyzed the data. FAAD and AKC wrote paper. FAAD and AKC hold responsibility for the content of paper. Both authors have read and approved the final manuscript.

Funding

This study did not receive funding from any institution or organization.

Data availability

The data analysed during the current study are publicly available on the website of the Ghana Statistical Service ([https://microdata.statsghana.gov.gh/index.php/catalog/97](https:/microdata.statsghana.gov.gh/index.php/catalog/97)) and can be accessed on written request.

Declarations

Ethics approval and consent to participate

This study analyzed secondary data from the GLSS 7. Voluntary consent to participate in the survey was sought from participants. Trained interviewers obtained informed consent from participants before conducting the interviews. Participants gave voluntary consent to participate in the survey.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.