Diet and physical activity among adults in Lesotho: a population-based cross-sectional survey

¹Division of Clinical Epidemiology, University Hospital Basel, Basel, Switzerland, ²Department of Clinical Research, University of Basel, Basel, Switzerland ³SolidarMed Lesotho, Maseru, Lesotho, ⁴Ministry of Health Lesotho, Maseru, Lesotho, ⁵SolidarMed Switzerland, Lucerne, Switzerland, ⁶Oslo Center for Biostatistics and Epidemiology, Oslo University Hospital, University of Oslo, Oslo, Norway, ⁷Bristol Medical School, University of Bristol, Bristol, United Kingdom, ⁸Department of Medicine, Division of General Internal and Emergency Medicine, Kantonsspital Aarau, Aarau, Switzerland

^&Corresponding author
Felix Gerber, Division of Clinical Epidemiology, University Hospital Basel, Basel, Switzerland

Introduction    

Non-communicable diseases (NCDs), particularly cardiovascular diseases (CVDs) and diabetes, are the leading cause of global mortality and morbidity. Low- and middle-income countries (LMICs) are bearing a disproportionate CVD burden that is expected to increase substantially over the coming years [1,2]. Insufficient physical activity and dietary risks, including underconsumption of healthy food such as fruits and vegetables, and overconsumption of unhealthy food such as processed meat, excess salt, and sweetened beverages, are the most important behavioural CVD risk factors. Dietary risks alone are associated with almost seven million deaths and over 150 million disability adjusted life years (DALYs) annually [3], while low physical activity contributes several million additional DALYs each year [4]. Lesotho, a small landlocked country in Southern Africa, faces a significant double-burden of both communicable and non-communicable diseases, including a rising prevalence of cardiovascular conditions such as arterial hypertension and diabetes [5,6]. As in most other LMICs, obesity has overtaken underweight as the primary cause of nutrition-associated disease burden [7].

A recent population-based survey revealed an adult prevalence of overweight of 40%, with women being disproportionately affected [8]. Although national strategies and programmes to address NCDs are being implemented, efforts are often focused on managing existing disease rather than preventing its onset [9,10]. Assessing and quantifying exposure and effects of individual diet components is complex [1], and while some recent data on CVD-related dietary risks and physical activity exist for Lesotho, they remain limited [11]. Understanding current dietary patterns and activity levels across different demographic groups is essential for designing effective, context-specific interventions. To address this gap, we conducted a representative population-based cross-sectional survey to provide population-level data on dietary habits and physical activity levels in rural and urban communities in northeastern Lesotho. The primary objectives of this study were to: i) describe the consumption patterns of fruits, vegetables, salt, and unhealthy foods among adults in Lesotho; ii) assess physical activity levels using standardized instruments; iii) describe patterns in dietary habits and physical activity across socioeconomic groups and rural and urban settings.

Methods     

Study design and setting: this study is part of the Community-Based Chronic Care Lesotho (ComBaCaL) project that aims at improving chronic disease care in Lesotho [12]. We conducted a population-based cross-sectional survey investigating the prevalence and risk factors of chronic diseases in the districts Butha-Buthe and Mokhotlong in northeastern Lesotho between November 1^st, 2021, and August 31^st, 2022. Both districts are characterized by a central town and large rural areas with challenging transportation infrastructure. The total population is approximately 250,000 inhabitants. This study is reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines for cross-sectional studies [13].

Sampling strategy: we applied a multistage cluster sampling design in which administrative units (towns, peri-urban areas, and villages) served as primary sampling clusters and individuals within households as secondary sampling units. The sampling frame was based on official lists of all administrative areas in Butha-Buthe and Mokhotlong districts obtained from the Ministry of Health and derived from the 2016 national census. To achieve sufficient cluster sizes, villages with fewer than 30 households were merged with adjacent villages where geographically possible. Administrative areas with fewer than 30 households that could not be combined, uninhabited settlements and government-only facilities (e.g. military or police sites) were excluded. After these steps, 785 eligible clusters remained, each comprising at least 30 households. Clusters were stratified by district (Butha-Buthe vs Mokhotlong), settlement type (urban vs rural), and geographic accessibility (hard to reach vs easy to reach). Accessibility was classified as difficult if reaching the nearest health facility required crossing rivers or mountainous terrain or travelling more than 10 km. From this stratified sampling frame, 120 clusters were selected using computer-generated random sampling performed by an independent statistician. A household was defined as one or more persons living in the same living space and sharing resources and daily living arrangements.

Participant recruitment and eligibility: clusters were included if the respective village chief granted verbal permission to conduct the survey. Within selected clusters, households were eligible when the household head or another adult household representative provided verbal consent. Household members were then randomly selected using a pre-programmed algorithm implemented in the Open Data Kit (ODK) data collection system [14], incorporating age, sex, and settlement type. For the overall survey, participants aged 10 years or older were eligible. For this specific analysis, only adult participants aged 18 years or older were considered. Selected individuals were enrolled after providing written informed consent.

Sample size: the required sample size was estimated using the World Health Organization (WHO) stepwise approach to NCD risk factor surveillance (STEPS) sample size calculator based on expected prevalence of key cardiovascular risk factors, including elevated blood pressure and diabetes. The target sample for adults 18 years or older was 6,000.

Data collection procedures: a study team of typically eight members, including nurses, nursing assistants, and lay health workers, visited households to collect data on tablets using ODK collect [14]. Upon reaching a village, verbal consent from the village chief was obtained before contacting household heads. With the consent of the household head, the team documented the age and sex of each household member. Comprehensive study data, including demographic details, anthropometric measurements, and laboratory results, were collected from participants randomly selected by an algorithm after written individual informed consent was obtained.

Bias minimization: several measures were taken to minimize potential sources of bias. Selection bias was reduced through multistage random sampling at the cluster, household, and individual levels. Information bias was addressed by using standardized and validated instruments for dietary and physical activity assessment. Data were collected electronically using tablet-based forms with built-in range and consistency checks to reduce data entry errors. Nevertheless, dietary intake and physical activity were self-reported and may be subject to recall bias and social desirability bias.

Collected data

Sociodemographic and socioeconomic characteristics: we collected demographic data on gender, age, marital status, educational attainment, and employment status. Household wealth was assessed using the Demographic and Health Survey (DHS) questionnaire for Lesotho [15], which assesses various socioeconomic indicators, including water sources, utilities, household assets, food security, housing materials, and cooking fuel types. Household wealth scores were calculated following the methodology outlined by the DHS and categorized into quintiles [16].

Medical history and anthropometric measurements: information on prior diabetes or hypertension diagnosis was collected, either self-reported or as documented in the participant's health booklet. Standing height was measured to the nearest millimeter using a portable stadiometer (Seca 213, Seca, Hamburg, Germany), with participants barefoot or wearing thin socks. Body weight was measured using a digital scale (Beurer, Ulm, Germany). Body mass index (BMI) was calculated as weight in kilograms divided by height in meters squared and categorized according to WHO criteria as underweight (18.5 kg/m²), normal weight (18.5-24.9 kg/m²), overweight (25.0-29.9 kg/m²), and obesity (≥30.0 kg/m²).

Dietary assessment: consumption of fruits, vegetables and salt was assessed using questions from the WHO STEPS manual [17]. Participants were asked about the number of days on which they consume fruits or vegetables per week and subsequently how many servings of fruits or vegetables they typically eat on one of those days. A standard serving of fruit or vegetables is defined as approximately 80 g. For fruits, this corresponds to one medium-sized piece (e.g., one apple, banana, or orange). For vegetables, one serving equals about half a cup of cooked or chopped vegetables, or one full cup of raw leafy vegetables. Starchy tubers such as potatoes and cassava are not counted as vegetables in this definition [17]. The number of days per week fruit or vegetables were consumed was then multiplied by the number of portions consumed and divided by seven to obtain the mean number of fruit and vegetable portions consumed per day. To assess salt consumption, four questions of the WHO STEPS manual were asked, including the frequency of salt being added to a meal during cooking and just before or during eating, the perceived importance of reducing salt intake and whether there was a perceived health risk associated with high salt consumption.

Consumption of unhealthy food items was evaluated using an abbreviated food frequency questionnaire (FFQ) from an obesity assessment tool previously used in South Africa [18] adapted to reflect local dietary habits. The FFQ assessed the consumption of eight unhealthy food and beverage items that are widely available and frequently consumed in Lesotho. The selection of items was based on the inputs from the local study team and community members. The eight unhealthy food and beverage items were grouped in the following categories: i)sweet beverages: a) fruit juices made from concentrate and b) sugar-sweetened, carbonated drinks; ii) sweet snacks: a) sweets or candy, b) biscuits or muffins, and c) chocolates; iii)fried snacks: a) crisps, b) fries or chips, and c) makoenya (a local type of fried bread bun). For each food item, participants were asked about the frequency of consumption during the past week with the possible responses being: 'None per week,' 'Once to twice a week,' 'Three to four times a week,' or 'Five or more times a week.'

Physical activity: physical activity was assessed using the International Physical Activity Questionnaire Short Form (IPAQ-SF). Participants reported the frequency (days per week) and duration (time per day) of vigorous-intensity activity, moderate-intensity activity, and walking during the previous 7 days. Total physical activity levels were classified as low, moderate, or high according to the standard IPAQ scoring protocol [19].

Analyses and statistical methods: data quality was ensured through careful data cleaning. The analyses were primarily descriptive. Implausible values and outliers were coded as missing. For fruit and vegetable intake, reported consumption of ≥20 servings per day was considered implausible and treated as an outlier, consistent with approaches used in previous studies [20]. For physical activity variables, outliers were identified and handled according to the IPAQ-SF data processing guidelines [21]. Variables assessing salt use and consumption of unhealthy foods and beverages did not show implausible extreme values; however, occasional missing data occurred due to non-response. Participant characteristics and study outcomes were summarised using medians with interquartile ranges, counts with percentages, and means with 95% Wald confidence intervals. Analyses were adjusted for clustering at the village level, and no sampling weights were applied. All analyses were performed using Stata version 16.1 (StataCorp, College Station, TX, USA). Graphs were generated using Excel (Microsoft, Redmond, WA, USA).

Ethical approval: all study procedures were conducted in accordance with the Declaration of Helsinki and applicable ethical standards. Participants received detailed information about the study procedures in their local setting and provided written informed consent prior to enrolment. For participants who were illiterate, consent was documented by thumbprint in the presence of an independent witness, who provided a signature. After completion of the consent process, one signed copy of the consent form was retained by the study team, and one copy was provided to the participant. The study protocol was reviewed and approved by the Ethics Committee Northwest and Central Switzerland (EKNZ; ID AO_2021-00056) and by the National Health Research Ethics Committee of Lesotho (ID 139-2021).

Results     

Participant characteristics: of 785 clusters in the two districts, 120 clusters (60 urban and 60 rural) were randomly selected. Within these clusters, 7,471 individuals from 3,498 households were screened for participation. Of these, 42 individuals declined participation or did not provide consent, and 17 were excluded due to age below 10 years or missing birthdate information. A total of 7,412 participants from 3,485 households were enrolled in the survey (response rate 99.2%). For the present analysis, 6,061 adults aged ≥18 years were included, while 1,351 adolescents aged 10-17 years were excluded.

Demographic, socioeconomic, medical history and body mass index: a total of 6,061 participants aged ≥18 years were enrolled in the survey; 52.2% were female and 47.8% male. The median age of participants was 39 years [IQR: 27-58]. 52.2% of participants were female. A similar proportion of participants from the two districts (52.5% Butha-Buthe, 47.5% Mokhotlong) and from rural (47.8%) and (peri)urban (52.2%) settlements were included, 58.5% of all participants were married or in a committed relationship, 56.7% had no or only primary education, with more men than women having no formal education (15.9% versus 4.8%). About half (50.9%) of the study population had a regular income, with a discrepancy between men and women (71.3% vs 32.3%). 815 (13.5%) and 125 (2.1%) participants had a history of arterial hypertension and type 2 diabetes, respectively, with higher rates among women (19.0% and 3.0%) than men (7.5% and 1.1%). Overall, 40.0% of the adult participants were either overweight (BMI 25 - 29.9 kg/m²) or obese (BMI ≥30 kg/m²), with 21.7% being overweight and 18.3% obese. Both overweight and obesity were more common in women (28% vs 14.9% and 30.7% vs 4.9%), while underweight (BMI <18.5 kg/m²) was more common in men (13.4% vs 3.8%) (Table 1).

Diet and physical activity

Fruit and vegetable consumption: the average number of days per week that fruit was consumed was 2.2 [95% CI: 1.83 - 2.57], with a mean daily intake of 0.82 servings [95% CI: 0.62 - 1.03]. For vegetables, the mean consumption frequency was 4.91 days per week [95% CI: 4.79 - 5.04], with an average daily intake of 1.61 servings [95% CI: 1.55 - 1.67]. Overall, the mean daily consumption of fruits and/or vegetables was 2.44 servings [95% CI: 2.22 - 2.65]. Participants in rural environments had a higher average daily intake of fruits and/or vegetables, consuming 2.77 servings [95% CI: 2.38 - 3.17] compared to 2.13 servings [95% CI: 1.98 - 2.27] in urban areas. Only 8.9% [95% CI: 6.2 - 11.7] of participants met the World Health Organisation recommended minimum intake of five or more servings of fruits or vegetables per day [22]. The proportion of participants meeting the criteria was higher in rural than in urban settlements (14.0% [95% CI: 9.0-19.0] vs 4.4% [95% CI: 2.6 - 6.2) (Table 2).

Salt consumption: salt use during food preparation was widespread, with 5,744 participants (94.9%) reporting that they always or often add salt or salty sauce when cooking. Additionally, 952 participants (15.8%) stated that they frequently add salt right before or while eating. The majority of participants (77.9%) considered reducing salt intake to be very important, while 921 participants (15.2%) considered reducing salt consumption as not important or were unsure about its importance. Similarly, 4,316 participants (71.4%) believed that excessive salt can lead to health problems, while 1,731 participant (28.6%) did not believe that too much salt could be harmful or did not know. There were no differences between gender or settlement type regarding self-reported salt consumption during cooking or while eating, while the proportion considering salt reduction as important and excessive salt consumption as a health risk was higher among women than men (Table 2).

Unhealthy food and beverage consumption: a total of 3,219 participants (53.2%) reported no consumption of sweet beverages, and 1,599 (26.5%) consumed sweet beverages one to two times a week. Overall, 5,928 participants (98.0%) reported less than daily consumption of sweet beverages in a typical week. Among the participants, 3,289 (54.4%) reported consuming none of the sweet foods asked about, while 1,228 participants (20.3%) consumed any of the food items asked about one to two times a week. Overall, 5,829 participants (96.4%) reported less than daily consumption of any sweet food in a typical week. The most consumed sweet food was candy, with 2,164 participants (35.8%) reporting consumption at least once a week. 2,173 participants (36.0%) reported no consumption of fried food items, and 1,409 participants (23.3%) consumed any of the food items asked about one to two times a week. In total, 5,628 participants (93.1%) reported less than daily consumption of a fried food item in a typical week. Crisps were the most consumed fried food item, with 2,969 participants (49.1%) eating them at least once a week (Table 3).

Physical activity: physical activity data was available for 6,061 participants. However, 728 participants (12.0%) were excluded from the analysis due to incomplete or inconsistent responses, in line with the IPAQ-SF data processing and analysis guidelines. Among the remaining 5,333 participants, 3,232 (60.6%) reported high and 1,287 (24.1%) moderate physical activity. Only 814 participants (15.3%) reported low physical activity. More men than women reported higher physical activity (67.5% vs. 54.4%), while more women reported moderate and lower physical activity (27.4% vs. 20.5% and 18.2% vs. 12%, respectively). The proportion of participants reporting high physical activity levels was higher in rural compared to (peri)urban areas (63.9% vs 57.2%) (Table 4).

Patterns of diet and physical activity by socioeconomic status: in our sample, a descriptive pattern suggested that individuals in the richest quintile reported consuming more unhealthy food items of all types, including sweetened beverages, snacks and fast food. Moreover, they consumed less fruit and vegetables, and reported lower levels of physical activity, than those in poorer quintiles (Figure 1, Figure 2).

Discussion     

This population-based cross-sectional survey conducted in two districts of Lesotho indicates a low consumption of unhealthy food items, a low intake of fruits and vegetables, and high levels of physical activity. At the same time, overweight and obesity are highly prevalent in this population. Our findings reveal a high prevalence of overweight (28.0%) and obesity (30.7%) among women. In contrast, most men (66.8%) are normal weight, although overweight (14.9%) and obesity (4.9%) remain notable concerns. These results are consistent with national overweight and obesity statistics for Lesotho [23]. The substantial burden of overweight and obesity observed indicates that the growing obesity epidemic is not limited to urban centres but is also affecting rural communities. This trend aligns with patterns observed across many LMICs, where obesity rates have been rising steadily and are projected to continue increasing [7]. The elevated rates of overweight and obesity, particularly among women, cannot be fully explained by the lifestyle behaviours assessed. Reported levels of high physical activity were high for both men (67.5%) and women (54.4%), and intake of unhealthy foods was relatively low. Our results suggest that physical inactivity and the consumption of commonly recognized unhealthy foods are not the primary contributors to overweight in the study population.

This indicates that other dietary factors may be involved. A study from South Africa [24] used an unquantified food frequency questionnaire covering 61 food items and, unlike our survey, included questions on starchy foods, which are considered potential contributors to weight gain. Further research is needed to better assess dietary patterns and to understand what contributes to the high rates of overweight and obesity in this population. Quantitative dietary assessments that include information on food composition and estimated energy intake (in kilocalories per day) could help provide a more accurate picture of the actual diet in the study population [25,26]. Fruit and vegetable intake in our study population was low, with only 8.9% of participants meeting the WHO recommendation of at least five daily servings [27]. Rural participants were more likely to meet this threshold (14.0%) than urban participants (4.4%). These figures are lower than those reported in recent reviews of dietary intake across LMICs [28] and regions with comparable socioeconomic profiles [20]. Fruit intake was particularly limited, averaging just 0.82 servings per day. This is likely due to Lesotho’s scarce domestic availability of fruits and vegetables that averages only 128g per person per day due to its high-altitude terrain, rain fed small-scale farming systems, and limited fruit and vegetable production capacity [29].

Inadequate fruit intake is a leading dietary risk factor for morbidity and mortality in Southern Africa [30]. The very low levels observed in our study suggest that this risk may be even more pronounced in the population we assessed. Addressing barriers to fruit and vegetable consumption will require improvements in food system resilience, agricultural diversity, and affordability to reduce the burden of diet-related diseases [31]. Salt consumption is widespread in the study population, both during cooking and while eating. Although most participants recognised the importance of reducing salt intake, a notable proportion showed limited awareness of the associated health risks. These patterns are similar to findings from other African countries such as South Africa and Ghana, where salt use remains high despite public health campaigns [32]. To better understand the health impact of sodium consumption, future research should include more detailed assessments, including objective methods such as spot or 24-hour urine collection to quantify actual sodium intake. Our findings, which show that people in the richest socioeconomic group consume more unhealthy food items and fewer fruits and vegetables, and are less active, reflect a pattern frequently observed in LMICs undergoing economic and nutritional transition.

In such contexts, it is often the wealthier population segments who first adopt more sedentary lifestyles and diets high in sugar, fat and ultra-processed foods. This results in a higher burden of obesity and NCDs among these groups [33]. This observation aligns with the reversal hypothesis, which posits that in the early stages of economic development, a higher socioeconomic status (SES) is associated with an unhealthy diet, lower levels of physical activity, and a higher risk of NCDs. However, as countries grow wealthier, this association reverses [34]. As development progresses, better health literacy, greater access to healthcare and stronger institutional support enable higher-SES groups to adopt healthier behaviours. Meanwhile, unhealthy dietary habits and sedentary lifestyles are becoming increasingly prevalent among lower-SES groups due to the affordability and availability of ultra-processed foods [35]. Understanding these shifting patterns is essential for designing effective public health strategies. Interventions in LMICs should target unhealthy behaviours among wealthier individuals early in the transition phase, while anticipating the eventual shift of the NCD burden to poorer segments of the population. Therefore, preventive strategies must be flexible and focus on equity, addressing both current and emerging risk profiles across the socioeconomic gradient [36].

These findings are likely to be applicable to rural and small urban areas in the mountainous north-eastern regions of Lesotho. However, they may not be representative of Maseru (the capital) or the lowland regions, where food access and dietary patterns may differ. Although Lesotho has unique geographical and economic characteristics, the observed patterns of rising obesity alongside low fruit and vegetable consumption and high salt intake are consistent with trends reported in many LMICs undergoing a nutrition transition. This study has several strengths. It was conducted as part of a large, population-based cross-sectional survey in two districts of Lesotho and included both urban and rural communities, including hard-to-reach areas. The use of stratified random sampling enhances the representativeness of the findings. Standardised and widely used tools, such as the WHO STEPS dietary questions and the IPAQ-SF for physical activity, were employed, allowing for comparisons with national and international data. In addition, the combined assessment of dietary behaviour and physical activity provides a comprehensive view of modifiable lifestyle risk factors for NCDs. This study has important limitations. Dietary intake and physical activity were self-reported, which introduces potential social desirability and recall bias. The dietary assessment focused on a limited number of food items, omitting energy-dense staples such as maize-based dishes, which may play a significant role in local diets and contribute to overweight and obesity. Furthermore, salt intake was assessed only qualitatively, which limits the ability to estimate actual sodium intake. Finally, children and adolescents were not included in this analysis, limiting the generalisability of the findings to the broader population.

Conclusion     

This population-based cross-sectional survey provides valuable insights into dietary habits and physical activity levels among adults in Lesotho. While reported physical activity levels were high and consumption of unhealthy foods relatively low, the prevalence of overweight and obesity, particularly among women, remains substantial. Low fruit and vegetable intake and widespread salt use were observed, alongside gaps in awareness of dietary risks. The findings suggest that the assessed behavioural risk factors alone may not fully explain the burden of obesity in this population, highlighting the need for more nuanced dietary assessments and the inclusion of quantitative intake measures. Public health strategies must take into account changing dietary patterns across the socioeconomic spectrum and prioritise preventive approaches that improve access to healthy, diverse, and affordable diets.

Competing interests     

The authors declare no competing interest.

Authors' contributions     

Alain Amstutz and Niklaus Daniel Labhardt were the principal investigators, they acquired the funding, led the project, and conceptualized the study together with Felix Gerber, Lucia González Fernández, Emmanuel Firima and Thesar Tahirsylaj. Ravi Gupta and Pauline Grimm led the local implementation. Thesar Tahirsylaj and Felix Gerber drafted the first manuscript with input from all authors. Giuliana Sanchez-Samaniego and Tristan Lee led the data management. Malebona Mathilise, Thuso Kabi, Mosoetsi Mokaene, Malehloa Maphenchane, Manthabiseng Molulela, Makhebe Khomolishoese, Matumaole Bane, Mamoronts’ane Pauline Sematle: were responsible for the local implementation and data collection. Madavida Mphunyane, Lebohang Sao, and Mosa Tlahali were: responsible for the collaboration with the Lesotho Ministry of Health and gave input on the study design to ensure alignment with local guidelines and practices. Thesar Tahirsylaj: conducted the analysis with input from Felix Gerber, Frédérique Chammartin and Fabian Raeber. Thesar Tahirsylaj, Felix Gerber, Frédérique Chammarin, Giuliana Sanchez-Samaniego, Niklaus Daniel Labhardt and Alain Amstutz accessed and verified the underlying data. All the authors have read and agreed to the final manuscript.

Acknowledgments     

We would like to acknowledge the SolidarMed team in Lesotho and Switzerland, and the participants for their essential contributions.

Tables and figures     

Table 1: sociodemographic and socioeconomic characteristics, medical history, and BMI among participants in a population-based cross-sectional survey conducted in the districts of Butha-Buthe and Mokhotlong (Lesotho), from November 2021 to August 2022 (n = 6,061)

Table 2: fruit, vegetable and salt consumption among participants in a population-based cross-sectional survey conducted in the districts of Butha-Buthe and Mokhotlong (Lesotho), from November 2021 to August 2022 (n = 6,061)

Table 3: unhealthy food and beverage consumption among participants in a population-based cross-sectional survey conducted in the districts of Butha-Buthe and Mokhotlong (Lesotho), from November 2021 to August 2022 (n = 6,061)

Table 4: level of physical activity among participants in a population-based cross-sectional survey conducted in the districts of Butha-Buthe and Mokhotlong (Lesotho), from November 2021 to August 2022 (n = 6,061)

Figure 1: diet patterns by wealth quintile among participants in a population-based cross-sectional survey conducted in the districts of Butha-Buthe and Mokhotlong (Lesotho), from November 2021 to August 2022 (n = 6,061)

Figure 2: high level of physical activity by wealth quintile among participants in a population-based cross-sectional survey conducted in the districts of Butha-Buthe and Mokhotlong (Lesotho), from November 2021 to August 2022 (n = 6,061)

References