ISSN: 2047-2986

Facebook-f Twitter Youtube Linkedin Envelope
ISOGH

IMPACT: Google Scholar: H5-Index = 69; H5-Median = 102 (12th in the World in Public Health); Web of Science/Clarivate: Impact factor = 5.8 (5-year) / 5.5 (2-year); Scopus: Citescore = 6.2

Articles

Country-specific key lifestyle factors and health outcomes for resource allocation in the general population: a network analysis across 29 countries

Jiaying Li1,2, Daniel Yee Tak Fong1, Kris Yuet Wan Lok1, Janet Yuen Ha Wong3, Mandy Man Ho1, Edmond Pui Hang Choi1, Vinciya Pandian2, Patricia M Davidson4, Wenjie Duan5, Marie Tarrant6, Jung Jae Lee1, Chia-Chin Lin1, Oluwadamilare Akingbade7,8, Khalid M Alabdulwahhab9, Mohammad Shakil Ahmad10, Mohamed Alboraie11, Meshari A Alzahrani12, Anil S Bilimale13, Sawitree Boonpatcharanon14, Samuel Byiringiro3, Muhammad Kamil Che Hasan15, Luisa Clausi Schettini16, Walter Corzo17, Josephine M. De Leon18, Anjanette S. De Leon18, Hiba Deek19, Fabio Efficace20, Mayssah A El Nayal21, Fathiya El-Raey22, Eduardo Ensaldo-Carrasco23, Pilar Escotorin24, Oluwadamilola Agnes Fadodun25, Israel Opeyemi Fawole26, Yong-Shian Shawn Goh 27, Devi Irawan 28, Naimah Ebrahim Khan29, Binu Koirala2, Ashish Krishna30, Cannas Kwok31, Tung Thanh Le32, Daniela Giambruno Leal33, Miguel Ángel Lezana-Fernández34, Emery Manirambona35, Leandro Cruz Mantoani36, Fernando Meneses-González34, Iman Elmahdi Mohamed37, Madeleine Mukeshimana38, Chinh Thi Minh Nguyen32 , Huong Thi Thanh Nguyen32, Khanh Thi Nguyen32, Son Truong Nguyen32, Mohd Said Nurumal15, Aimable Nzabonimana39, Nagla Abdelrahim Mohamed Ahmed Omer40, Oluwabunmi Ogungbe2, Angela Chiu Yin Poon41, Areli Reséndiz-Rodriguez42, Busayasachee Puang-Ngern14, Ceryl G Sagun18, Riyaz Ahmed Shaik10, Nikhil Gauri Shankar43, Kathrin Sommer20, Edgardo Toro33, Hanh Thi Hong Tran32, Elvira L Urgel18, Emmanuel Uwiringiyimana35, Tita Vanichbuncha14, Naglaa Youssef44

1School of Nursing, Li Ka Shing Faculty of Medicine, University of Hong Kong, Hong Kong SAR, China
2School of Nursing, Johns Hopkins University, Baltimore, Maryland, USA
3School of Nursing and Health Studies, Hong Kong Metropolitan University, Hong Kong SAR, China
4Vice-Chancellor and Principal, University of Wollongong, Wollongong, Australia
5East China University of Science and Technology, Department of Social Work, Shanghai, China
6School of Nursing, The University of British Columbia, Kelowna, British Columbia, Canada
7The Nethersole School of Nursing, The Chinese University of Hong Kong, Hong Kong, China
8Institute of Nursing Research, Osogbo, Osun State, Nigeria
9College of Medicine, Majmaah University, Al Majmaah, Saudi Arabia
10Department of Family & Community Medicine, College of Medicine, Majmaah University, Majmaah, Saudi Arabia
11Department of Internal Medicine, Al-Azhar University, Cairo, Egypt
12Department of Urology, College of Medicine, Majmaah University, Al Majmaah, Saudi Arabia
13School of Public Health, JSS Medical College, JSS AHER, Mysuru, India
14Department of Statistics, Chulalongkorn Business School, Bangkok, Thailand
15Kulliyyah of Nursing, International Islamic University, Kuantan, Malaysia
16Italian Association against Leukaemia, Lymphoma and Myeloma, Rome, Italy
17Diálogos Guatemala, Guatemala, Guatemala
18School of Nursing, Centro Escolar University, Manila, Philippines
19Nursing Department, Faculty of Health Science, Beirut Arab University, Lebanon
20Data Centre and Health Outcomes Research Unit, Italian Group for Adult Hematologic Disease, Rome, Italy
21Department of Psychology, Beirut Arab University, Beirut, Lebanon
22Department of hepatogastroenterology and infectious diseases, Damietta faculty of medicine, Al-Azher University, Kairo, Egypt
23Ergonomics Research Centre, University of Guadalajara, Jalisco, Mexico
24Laboratory of Applied Prosocial Research, Department of Basic, Developmental and Educational Psychology, Autonomous University of Barcelona, Barcelona, Spain
25Faculty of Health Sciences, University of Lethbridge, Lethbridge, Alberta, Canada
26Faculty of Nursing, Ladoke Akintola University of Technology, Ogbomosho, Nigeria
27Alice Lee Centre for Nursing Studies, National University of Singapore, Singapore, Singapore
28School of Nursing, Wijaya Husada Health Institute, Bogor, Indonesia
29Department of Optometry, University of Kwazulu-Natal, Durban, South Africa
30Ecove, Ghaziabad, India
31School of Nursing, Paramedicine and Health Care Science, Charles Sturt University, New South Wales, Australia
32Nam Dinh University of Nursing, Nam Dinh, Vietnam
33Pontificia Universidad Católica de Valparaíso, School of Social Work, Valparaíso, Chile
34Research Department, National Commission for Medical Arbitration, Mexico, Mexico
35College of Medicine and Health Sciences, University of Rwanda, Kigali, Rwanda
36Department of Physiotherapy, Presidente Prudente, Faculty of Science and Technology, São Paulo State University (UNESP), São Paulo, Brazil
37Pharmacology and Toxicology Department, Faculty of Pharmacy, Benghazi University, Benghazi, Libya
38School of Nursing and Midwifery, College of Medicine and Health Sciences, University of Rwanda, Kigali, Rwanda
39Centre for Language Enhancement, College of Arts and Social Sciences, University of Rwanda, Huye, Rwanda
40Faculty of Medicine, Alzaiem Alazhari University, Khartoum North, Sudan
41Faculty of Health Sciences and Sports, Macao Polytechnic University, Macao, China
42National Autonomous University of Mexico, Mexico City, Mexico
43Mental Health and Learning division, Wrexham Maelor Hospital, Wrexham, UKyy
44Medical-surgical Nursing Department, Faculty of Nursing, Cairo University, Cairo, Egypt

DOI: 10.7189/jogh.15.04011
Cite as: Li J, Fong DYT, Lok KYW, Wong JYH, Man Ho M, Choi EPH, Pandian V, Davidson PM, Duan W, Tarrant M, Lee JJ, Lin CC, Akingbade O, Alabdulwahhab KM, Ahmad MS, Alboraie M, Alzahrani MA, Bilimale AS, Boonpatcharanon S, Byiringiro S, Hasan MKC, Schettini LC, Corzo W, De Leon JM, De Leon AS, Deek H, Efficace F, El Nayal MA, El-Raey F, Ensaldo-Carrasco E, Escotorin P, Fadodun OA, Fawole IO, Goh YS, Irawan D, Khan NE, Koirala B, Krishna A, Kwok C, Le TT, Leal DG, Lezana-Fernández MÁ, Manirambona E, Mantoani LC, Meneses-González F, Mohamed IE, Mukeshimana M, Nguyen CTM, Nguyen HTT, Nguyen KT, Nguyen ST, Nurumal MS, Nzabonimana A, Omer NAMA, Ogungbe O, Poon ACY, Reséndiz-Rodriguez A, Puang-Ngern B, Sagun CG, Shaik RA, Shankar NG, Sommer K, Toro E, Tran HTH, Urgel EL, Uwiringiyimana E, Vanichbuncha T, Youssef N. Country-specific key lifestyle factors and health outcomes for resource allocation in the general population: a network analysis across 29 countries. J Glob Health. 2025;15:04011

(18 pages)

Download PDF (120)

Download PDF

(18 pages)

Share:

Abstract

Background

We aimed to identify the central lifestyle, the most impactful among lifestyle factor clusters; the central health outcome, the most impactful among health outcome clusters; and the bridge lifestyle, the most strongly connected to health outcome clusters, across 29 countries to optimise resource allocation for local holistic health improvements.

Methods

From July 2020 to August 2021, we surveyed 16 461 adults across 29 countries who self-reported changes in 18 lifestyle factors and 13 health outcomes due to the pandemic. Three networks were generated by network analysis for each country: lifestyle, health outcome, and bridge networks. We identified the variables with the highest bridge expected influence as central or bridge variables. Network validation included nonparametric and case-dropping subset bootstrapping, and centrality difference tests confirmed that the central or bridge variables had significantly higher expected influence than other variables within the same network.

Results

Among 87 networks, 75 were validated with correlation-stability coefficients above 0.25. Nine central lifestyle types were identified in 28 countries: cooking at home (in 11 countries), food types in daily meals (in one country), less smoking tobacco (in two countries), less alcohol consumption (in two countries), less duration of sitting (in three countries), less consumption of snacks (in five countries), less sugary drinks (in five countries), having a meal at home (in two countries), taking alternative medicine or natural health products (in one country). Six central health outcomes were noted among 28 countries: social support received (in three countries), physical health (in one country), sleep quality (in four countries), quality of life (in seven countries), less mental burden (in three countries), less emotional distress (in 13 countries). Three bridge lifestyles were identified in 19 countries: food types in daily meals (in one country), cooking at home (in one country), overall amount of exercise (in 17 countries). The centrality difference test showed the central and bridge variables had significantly higher centrality indices than others in their networks (P < 0.05).

Conclusions

In 29 countries, cooking at home, less emotional distress, and overall amount of exercise emerged as common central lifestyle, health outcome, and bridge lifestyle factors, respectively. However, notable regional variations necessitate tailored interventions and resource allocations to effectively address unique local key variables and promote holistic health in each locale. The study’s cross-sectional design and self-reported data may limit generalisability, emphasising the need for cautious interpretation and further longitudinal research.

Keywords

global; across-country comparisons; lifestyle; health outcomes; network analysis

Lifestyle factors – including dietary risks, alcohol and tobacco consumption, sedentary behaviours, and physical inactivity – significantly influence health outcomes both immediately and over the long term. Short-term impacts manifest in changes to physical and psychological quality of life [1], sleep [2], and social well-being [3]. Conversely, persistent unhealthy lifestyle behaviours substantially elevate the risk for non-communicable diseases such as cardiovascular disease [4], diabetes, cancer, and chronic respiratory diseases [5], ultimately affecting life expectancy and mortality [6]. The global prevalence of non-communicable diseases underscores the critical importance of lifestyle in health promotion. However, implementing comprehensive lifestyle interventions that cover all lifestyle aspects at the population level is complex and resource-intensive, necessitating the identification of the most impactful lifestyle for targeted interventions and effective resource allocation for effective holistic health promotion, especially for those low- and middle-income countries with heavier burden and less resources.

Network analysis is a technique that examines the effects of one variable within its network cluster or its connections with another network cluster while considering all other within-network variables as confounders [7]. In network analysis theory, the central lifestyle and health outcome variables are the most impactful in their respective networks, offering the highest predictive value and potential for substantial improvements when modified. The bridge lifestyle, which is the lifestyle factor most strongly connected to the health outcome network, serves a similar function but impacts the entire health network [7]. The presence of most impactful lifestyle is supported by the interconnected relationships identified by previous studies within lifestyles, health outcomes, and between them. For instance, physical inactivity is associated with other lifestyle behaviours, such as smoking, alcohol consumption and an unhealthy diet [8,9] Additionally, physical inactivity affects various interim health outcomes, including mental health and sleep quality [10]. Moreover, behaviours like smoking, drinking, and unhealthy eating also adversely impact other health outcomes, such as mental health and sleep quality [11]. Building on these intertwined relationships, a study utilised network analysis to identify central lifestyle factors, health outcomes, and bridge lifestyles using a sample drawn from various countries [12]. However, global approaches may overlook crucial cross-country variations in lifestyle patterns and cultural norms, limiting the effectiveness of interventions in specific regions. Therefore, identifying central lifestyle factors at the country level is essential to tailor interventions to local contexts.

The assumption that central lifestyle factors vary across countries is supported by multiple drivers, including socioeconomic and cultural influences, which affect the coexistence and strength of associations among lifestyle behaviours. For instance, limited access to healthy foods is more prevalent in low-income countries [13], which may result in insufficient food availability and reduced variety of food options. In contrast, high-income countries often experience higher rates of sedentary behaviours, such as prolonged sitting and increased screen time [14], due to factors like widespread use of digital technology, the prevalence of desk-based and office jobs, and greater reliance on automobiles for transportation. Additionally, cultural factors, such as the American preference for sweet flavours [15,16] and alcohol prohibition in some areas of Saudi Arabia due to religious and legal restrictions [15,16], can increase or decrease frequency of certain lifestyles. Hong Kong’s high-density urban living may limit physical activity and increase stress levels [17]. In contrast, Rwanda’s predominantly rural environment promotes higher levels of physical activity [18]. The connection between these differences and country-specific key variables can be explained by the pronounced presence of specific lifestyle factors and health outcomes. These factors may strengthen their associations with other coexisting lifestyles or health outcomes, thereby increasing their centrality within a network and making each country’s key variables distinct. Identifying these diverse key lifestyles is crucial for developing effective, tailored local health interventions, yet remains poorly understood.

Therefore, this study aims to assess the interplay between lifestyle factors and health outcomes across 29 countries, and to identify the key variables (central lifestyle, central health outcome, and bridge lifestyle). We hypothesise that each country has distinct central lifestyle factors, health outcomes, and bridge lifestyles. The findings are expected to provide country-specific guidance for local resource allocation for effective holistic health improvements.

METHODS

Study design

This was an international cross-sectional study.

Study settings

This study spanned six regions defined by the World Health Organization, covering 29 countries after excluding Spain due to a limited sample size of 51. Recruitment was conducted primarily through online platforms to optimise reach and facilitate voluntary participation in preferred languages. The published protocol provides detailed information on the study design [19].

Participants and sample size

We recruited participants aged 18 and older using convenience and snowball sampling techniques. While these methods enabled efficient participant recruitment across diverse geographical locations, they also introduce potential selection biases, particularly affecting the representativeness of the socioeconomic and rural/urban diversity within the sample populations. Sample sizes for each network – lifestyle (18 nodes), health outcomes (13 nodes), and bridge (31 nodes) – were determined based on the maximum number of edges: 153, 78, and 465, respectively, adhering to the guideline of at least three participants per parameter [20]. Consequently, the required sample sizes for each network were 459, 234, and 1134 participants, respectively. Sample sizes across countries ranged 150–2238, with most meeting these criteria. Notably, centrality measures in network analysis can be considered reliable even if the sample size falls short of initial estimates, provided they pass the stability test via case-dropping subset bootstrap [20]. To ensure reliability, stability tests were conducted for all 87 networks, with only those networks that passed these tests being reported. We recognise the limitations of these sampling methods and the sample size in some countries, and their potential impact on the generalisability of our findings. To mitigate these effects, stability tests via case-dropping subset bootstrap were conducted for all 87 networks, and only those networks that passed these tests were reported.

Measures

Sociodemographics

The sociodemographic variables included in this study were gender, age, country of residence, marital status, highest educational level attained, and employment status.

Measurement of lifestyle factors and health outcomes

Item generation for the study questionnaire was developed for assessing a thorough literature review and collaborative discussions with public health professionals, nurses, and nutritionists. The chosen variables for lifestyle and health outcomes reflect dimensions commonly acknowledged in public health research, focusing on key lifestyle dimensions, including diet, physical activity, and substance use, as well as health outcomes across physical, psychological, social, and financial well-being. Specifically, 18 lifestyle factors include:

  • dietary (types of food in daily meals, consumption of fruits and vegetables, frozen food/food products, snacks, soft drinks, juices, or other sugary drinks, meals at home, cooking at home, takeout food, traditional Chinese medicine or natural health products, and oral supplements/vitamins)
  • exercise (frequency, duration, type, and overall amount)
  • sedentary behaviours (sitting and screen time duration)
  • addiction behaviours (smoking tobacco and alcohol consumption).
  • The 13 health-related outcomes include:
  • physical well-being (weight, appetite, sleep quality, perceived physical health, and quality of life)
  • psychological well-being (mental burden, emotional distress)
  • social well-being (family disputes, social support provided, social support received, and social activities)
  • financial well-being (income and economic burden).

The initial English version was created and face validity ensured through rigorous reviews that refined the questionnaire by enhancing clarity, eliminating redundancies, and structuring the flow of questions to facilitate ease of completion. To ensure cultural relevance, consultations with domain experts from various countries were conducted, followed by translations into multiple languages using a standard forward-backwards method. Lifestyle factors and health outcomes were assessed using single items, making formal psychometric evaluation inapplicable [19].

Participants rated the impact of COVID19 on 18 lifestyle factors and 13 general health outcomes using a 5-point Likert scale, with 1 indicating ‘substantially reduced’, 3 ‘no change’, and 5 ‘substantially increased’. To ensure consistent item alignment reflecting healthier lifestyles and improved health outcomes, certain items were reverse coded by appending a ‘less’ prefix.

Data collection

Data were collected via online survey platforms and supplemented with offline electronic PDF forms in areas with limited internet access. Both methods were standardised in content and structure to ensure comparability and maintain data quality across different collection environments. Rigorous pre-testing confirmed the equivalence of the two formats, allowing for consistent data collection across all regions. To encourage participation, one Hong Kong dollar was donated to the Red Cross for each completed questionnaire. The response rate was recorded at 75.2%.

Validation and rigor

To enhance internal validity, validation questions were incorporated into the questionnaire. Participants were asked ‘Where does the sun rise every day?’ Additionally, to ensure cultural relevance in Nigeria, the question ‘Where is your STATE capital?’ was used. Additionally, each country’s questionnaire was pilot-tested with a minimum of 10 participants prior to administration. The results from these pilot tests indicated that participant responses were within expected parameters and no confusion regarding the questions was reported; therefore, no modifications to the questionnaire were deemed necessary. Moreover, to mitigate social desirability bias, the survey was conducted anonymously. Furthermore, we performed data quality checks before analysis to remove any responses indicating misreporting or misunderstanding. These measures collectively aim to reduce potential biases and enhance the accuracy and reliability of our data set.

Statistical analysis

Data collected for our study were transferred to a Microsoft Excel database to facilitate stringent quality control. This included removing incomplete or duplicate responses to ensure data integrity. We also addressed data inconsistencies by discarding responses that did not align with validation questions, such as incorrect details regarding sunset times or capital cities, thereby enhancing the accuracy of our data set and the validity of our analysis. Subsequent analyses were performed using R Statistical Software (v4.1.1; R Core Team, Vienna, Austria). The network analyses encompassed four domains: topological overlap assessment, network estimation, stability testing, and the computation of centrality and bridge centrality indices.

Checking topological overlap

We used the ‘networktools’ R package’s ‘goldbricker’ function to identify unique variables to prevent artificial relationships among similar variables in a network. A significance threshold of 0.25 for inclusion and a significance level of 0.01 were applied [21].

Network estimation

For each country, three networks were analysed: one comprising all remaining lifestyle factors, another encompassing all remaining health outcomes, and a bridge network connecting the two. Pairwise associations were estimated using partial correlation analysis to control for confounding effects from other nodes. The least absolute shrinkage and selection operator was applied to minimise edges and nullify small correlations, while the extended Bayesian information criterion helped select the tuning parameter, resulting in a more interpretable and sparse network [22]. Network estimation and visualisation were performed using the ‘bootnet’ and ‘qgraph’ R packages, respectively [22]. In visual representations, nodes symbolised network items and edges their relationships, with edge thickness indicating the strength of association – blue for positive and red for negative associations.

Network stability

Network stability was evaluated using the ‘bootnet’ package [22]. Edge weight stability was assessed with 95% confidence intervals (CIs) derived from nonparametric bootstrapping; narrower CIs indicated greater network credibility [22]. Centrality stability was measured by the correlation stability coefficient (CS-C), calculated via case-dropping subset bootstrap. A CS-C value above 0.25 was considered acceptable, with values above 0.5 being preferable [22].

Central node, centrality, bridge node, and bridge centrality

A central node significantly influences a network through its connections with other nodes, while bridge nodes connect closely to different clusters within another network [23,24]. In network analysis, betweenness centrality measures how frequently a node occurs on the shortest paths between other pairs, underscoring its intermediary role. Closeness centrality calculates a node’s average distance to all others, indicating network accessibility. Strength centrality totals the edge weights connected to a node, showing its connectivity. Differently, expected influence centrality sums edge weights without absolute conversions, capturing dominant influences in networks with both positive and negative edges. Given our objective to identify nodes that exert significant cumulative effects, particularly in networks with negative interactions, we focus on expected influence centrality to pinpoint central nodes and to identify bridge nodes effectively [25]. The ‘qgraph’ package in R calculated the expected influence index, which accounts for both positive and negative edges to identify central nodes with the highest values. Bridge nodes were assessed using the ‘networktools’ package in R, which utilises the bridge expected influence (one-step) index that aggregates a node’s edges to nodes in other networks [23]. Centrality differences between two nodes were evaluated using Wilcoxon tests with 1000 bootstrapped indices, generated through the ‘bootnet’ package in R. Multiple comparisons were adjusted for using Holm-Bonferroni corrections.

RESULTS

Demographics and item description

After screening 19 094 responses, 16 461 were eligible for analysis. Exclusions included blank or incomplete responses (1940), duplicates (116), inconsistent entries (450), responses from non-participating countries (126), and records missing age or gender information (1). Table 1 outlines the sociodemographic details, means, standard deviations, and abbreviations of lifestyle factors and health outcomes, while Table S1 in the Online Supplementary Document displays the breakdown by country.

Table 1.  Sociodemographic of respondents from 29 countries (n = 16 461)

H – health outcome, L – lifestyle, SD – standard deviation

Non-redundant items remained for network analysis

Redundancy analysis via the Goldbricker method excluded L15 (Table 1), L16, and L17 for overlapping with L18, representing exercise habits more accurately. The final network comprised 15 lifestyle factors and 13 health outcomes with no additional redundancies.

Lifestyle networks across each country

Network stability

The bootstrapped 95% CI analysis confirmed the accuracy of edge weights in the lifestyle network, with narrow intervals suggesting reliable estimates (Figure S1 in the Online Supplementary Document). Besides, CS-C values for expected influence, derived from a case-dropping subset bootstrap, ranged 0.21–0.75 (Table 2; Figure S2 in the Online Supplementary Document). Excluding the Unities States, these values surpassed the 0.25 threshold, affirming the interpretability of 28 country-specific lifestyle networks.

Table 2.  Summary of network measures across 29 countries (n = 16 461)

CS-C – correlation stability coefficient, H – health outcome, L – lifestyle, NA – not applicable, PC-C – partial correlation coefficient

Network structure and central lifestyle

Using Mainland China as an example, Figure 1, Panel A shows the lifestyle network with 49.5% of its 150 edges being nonzero, indicating strong connectivity. The strongest connections were between (the edge connected Lifestyle 7 and Lifestyle 6 (L7–L6), partial correlation coefficient = 0.85), L11–L12 (partial correlation coefficient = 0.75), and L13–L14 (partial correlation coefficient = 0.66) (Table 2). The central node, cooking at home (L7), was confirmed by centrality index plots and bootstrapped difference tests (P<0.05) (Figure 1, Panel B). For the other 27 countries, Figure S3 in the Online Supplementary Document presents the network structures, with nonzero edge percentages ranging from 15.2 in Guatemala to 68.6% in Hong Kong (Table 2). Figure S4 in the Online Supplementary Document contains centrality index plots and results of centrality bootstrapped difference tests. Table S2 in the Online Supplementary Document contains the partial correlation coefficients for all edges across all countries.

Figure 1.  Network structure and centrality difference test of lifestyles (Panels A–B), health outcomes (Panels C–D), and combined (Panels E–F) in mainland China. The abbreviations of nodes in Panels A, C, and E can be found in Table 1. In Panels B, D, and F, a grey cell indicates that there is no significant difference between the corresponding two variables. A dark cell indicates that there is a significant difference between the corresponding two variables at 5% level of significance. A white cell displays the value of the expected influence or bridge expected influence.

Health outcome networks across countries

Network stability

Narrow CIs across countries confirmed the precision of estimated edge weights (Figure S5 in the Online Supplementary Document). Besides, CS-C values (Table 2; Figure S6 in the Online Supplementary Document), varied 0.21–0.75. Australia was the only country below the 0.25 threshold, resulting in 28 interpretable health outcome networks.

Network structure and central health outcome

For Mainland China (Figure 1, Panel C) displays a network structure with 51.3% (40 out of 78) of edges being nonzero. The strongest connections were between H7–H6 (Table 1) and H5–H4 (both 0.67), and H2–H1 (0.42) (Table 2). Quality of life (H5) was identified as the most influential health outcome, confirmed by centrality index and bootstrapped difference tests (P<0.05) (Figure 1, Panel D). For the other 27 countries, Figure S3 in the Online Supplementary Document illustrates the network structures, with nonzero edge percentages ranging from 15.4 in the UK to 68.0% in Indonesia (Table 2). Figure S4 in the Online Supplementary Document contains centrality index plots and results of centrality bootstrapped difference tests. Table S3 in the Online Supplementary Document provides the partial correlation matrix for all edges across all networks.

Bridge networks across countries

Network stability

Figure S7 in the Online Supplementary Document illustrates the precision of estimated edge weights for each country with narrow CIs, indicating high accuracy. Besides, CS-C values for bridge networks varied between 0.00–0.75 (Table 2; Figure S8 in the Online Supplementary Document). Ten countries, including Australia, Egypt, Italy, Lebanon, Macau, Republic of Sudan, Rwanda, Saudi Arabia, Singapore, and South Africa, fell below the 0.25 threshold, yielding 19 interpretable bridge networks.

Network structure and bridge lifestyle

Figure 1, Panel E shows Mainland China’s network with 144 of 378 edges being nonzero. The overall amount of exercise (L18) emerged as the most influential bridge lifestyle, significantly outperforming other nodes as shown by centrality bootstrapped difference tests (P<0.05) (Figure 1, Panel F). The corresponding bridge edge is L18–H1 (lose weight). For the other 18 countries, Figure S3 in the Online Supplementary Document displays network structures, with nonzero edge percentages ranging from 6.1 in Guatemala to 46.6% in Indonesia (Table 2). Figure S4 in the Online Supplementary Document contains centrality index plots and results of centrality bootstrapped difference tests. Table S4 in the Online Supplementary Document contains the partial correlation matrix for all networks.

Summary of the central and bridge variables in networks across 29 countries

Nine types of central lifestyle were identified among 28 countries:

  • cooking at home (L7) in Guatemala, Lebanon, Macau, Mainland China, Nigeria, Philippines, Republic of Sudan, Rwanda, Saudi Arabia, South Africa, and Egypt
  • food types in daily meals (L1) in Burundi
  • ess smoking tobacco (L11) in Indonesia and Malaysia
  • less alcohol consumption (L12) in Rwanda and Vietnam
  • less duration of sitting (L13) in Canada, Rwanda, and Singapore
  • less consumption of snacks (L4) in Brazil, Chile, Mexico, South Korea, and UK
  • less sugary drinks (L5) in Hong Kong, India, Libya, Mexico, and Thailand
  • having a meal at home (L6) in Egypt and Libya
  • taking alternative medicine or natural health products (L9) in Australia.
  • Six types of central health outcome were identified among 28 countries:
  • social support received (H9) in Indonesia, Malaysia, and Philippines
  • physical health (H3) in Burundi
  • sleep quality (H4) in Egypt, Guatemala, Saudi Arabia, and Thailand
  • quality of life (H5) in Brazil, Chile, India, Italy, Mainland China, Mexico, and USA
  • less mental burden (H6) in Rwanda, Singapore, and South Africa
  • less emotional distress (H7) in Canada, Hong Kong, India, Lebanon, Libya, Macau, Nigeria, Republic of Sudan, Rwanda, Singapore, South Korea, UK, and Vietnam.
  • Three types of bridge lifestyle were identified among 19 countries:
  • food types in daily meals (L1) in Libya
  • cooking at home (L7) in Nigeria
  • overall amount of exercise (L18) in Brazil, Burundi, Canada, Chile, Guatemala, Hong Kong, India, Indonesia, Mainland China, Malaysia, Mexico, Philippines, South Korea, Thailand, United Kingdom, United States, and Vietnam.

Figure 2 visually summarises the distribution of central lifestyles, health outcomes, and bridge lifestyles across all participating countries.

Figure 2.  Central lifestyles, central health outcomes, and bridge lifestyles by country.

DISCUSSION

This study employed network analysis to examine complex relationships between lifestyle factors and health outcomes across 29 countries, pinpointing key variables for use as potential predictive markers or intervention targets. We identified four central lifestyle categories: diet-related behaviours, natural health product usage, substance use, and sedentary habits; and four key health outcomes: physical health, emotional distress, social support, and quality of life. Additionally, three bridge lifestyle factors – food types in daily meals, home cooking, and overall exercise amount – were recognised as focal points for targeted resource allocation to enhance network-wide health outcomes, given the established causal relationship between lifestyle and health. Modifying central lifestyle factors can enhance all within-network lifestyles and improve long-term health outcomes. Central health outcomes highlight opportunities for beyond-lifestyle interventions, such as better health care service utilisation. Bridge lifestyles offer specific pathways for optimising overall health through targeted adjustments. Shared central or bridge variables across regions highlight the potential for collaborative strategies that leverage common strengths and address shared challenges.

In 28 lifestyle networks, four central lifestyle categories emerged: diet-related behaviours, natural health product usage, substance use, and sedentary habits, reflecting complex cultural, economic, and public health interactions. Specifically, Australia’s focus on natural health products as a central lifestyle, likely driven by strong cultural acceptance and a preference for alternative health practices [26], enhances preventive lifestyles such as diet and exercise. In contrast, central alcohol consumption in Indonesia and Malaysia is influenced by significant binge drinking in Malaysia and rising harm from illicit home-produced alcohol in Indonesia [27,28]. Similarly, tobacco consumption was prominent in Rwanda and Vietnam, likely due to cultural norms and high smoking prevalence [29,30]. Reduced sitting time, significant in Rwanda and Singapore, corresponds with high urbanisation and pandemic-related increases in sedentary behaviour [31,32]. Moreover, diet-related behaviours such as specific food choices, reduced consumption of snacks and sugary drinks, home cooking, and having meals at home are pivotal in Rwanda and 23 other countries. Diet’s centrality within the lifestyle network may be explained by its integral role in daily life and its deep connections to cultural and economic conditions [33,34], along with its coexistence with other lifestyle factors through behavioural and psychological mechanisms, such as self-regulation and discipline. These central behaviours not only predict overall lifestyle patterns but also serve as catalysts for broader modifications, suggesting that interventions targeting these areas could significantly influence entire lifestyle networks. Globally, effective strategies to address these dietary challenges could include enhancing nutritional education and awareness, improving food accessibility and affordability, and subsidising healthy foods or taxing unhealthy ones to leverage diet’s centrality for holistic improvements in overall healthy lifestyles.

In 28 health outcome networks, we identified four primary categories of central health outcomes: physical health (including sleep quality), emotional distress, social support, and overall quality of life. Social support received was central in Indonesia, Malaysia, and the Philippines, reflecting their collective cultural values [35,36]. Physical health was prominent in Burundi, highlighting direct concerns with bodily well-being, while sleep quality was central in Egypt, Guatemala, Saudi Arabia, and Thailand, possibly due to hot climates affecting sleep [37]. Emotional distress or mental burden emerged as central health concerns in countries including Canada, Hong Kong, India, Lebanon, Libya, Macau, Nigeria, Republic of Sudan, Rwanda, Singapore, South Africa, South Korea, UK, and Vietnam, driven by diverse factors. Specifically, political turmoil and economic instability significantly impacted Lebanon, Libya, Sudan, and Nigeria [38], while rapid urbanisation and economic stress affected Singapore, Hong Kong, South Korea, Macau, and others [39]. Quality of life, a multidimensional outcome encompassing physical health, mental health, and social relationships [40], was central in Brazil, Chile, India, Italy, Mainland China, Mexico, and the USA. This highlights the varied needs for health outcome interventions in each country-targeting physical, psychological, and social factors beyond lifestyle changes-to effectively improve holistic health.

In 19 countries with an interpretable bridge network, our study identified three key bridge lifestyles significantly affecting overall health outcomes: daily food types in Libya, home cooking in Nigeria, and exercise in other nations. In the context of network analysis, bridge lifestyle variables serve as critical connectors between the lifestyle network and the health outcomes network. These variables exhibit strong influence within the lifestyle network while simultaneously linking strongly to health outcome networks. Targeting these bridges in interventions can produce widespread, synergistic benefits by improving both the variables themselves and associated lifestyle behaviours, leading to enhanced overall health outcomes. Libya’s reliance on specific meal types, influenced by its desert climate, heavy reliance on food imports, and political instability, highlights the critical role of dietary choices in health outcomes [41,42]. In Nigeria, home cooking may influence health through nutritional control over diets and economic relief via cost-effectiveness [43,44]. All remaining 17 countries or regions had exercise as their bridge lifestyle. The well-documented benefits of exercise on mental and physical health may explain this consistency across various nations [45], suggesting that strategic enhancements in exercise accessibility could lead to substantial health improvements. Globally, effective strategies to improve exercise accessibility include enhancing public infrastructure, implementing workplace wellness programmes, promoting community initiatives, integrating physical activity into urban planning, conducting education campaigns, ensuring inclusivity, partnering with health care providers, leveraging digital platforms, and supporting research and monitoring. Given the well-established causal relationship between lifestyle factors and health outcomes, targeted resource allocation and intensified interventions on bridge lifestyles are crucial for optimising these modifications across diverse settings.

This study identifies target points requiring increased resources and targeted interventions but does not specify which kinds of strategies are most effective. It assumes countries will adopt context-proven strategies. While comprehensive interventions for all lifestyle factors are ideal, they may only be feasible in resource-rich settings. Network theory suggests that prioritising central lifestyles, key health outcomes, and bridging lifestyles achieves greater health improvements than addressing other factors alone, which is particularly useful for resource-limited settings. For example, high-income Macau, middle-income mainland China, and low-income Sudan all identify home cooking as a central lifestyle factor. In resource-rich Macau, allocating substantial resources to home cooking can enhance other lifestyles, while additional resources can directly target other factors. In low-resource Sudan, dedicating a higher proportion of resources to home cooking can maximise impact. Similarly, targeting central health outcomes and bridging lifestyles can holistically improve health outcomes. Policymakers should design cost-effective programmes focused on these key variables and adopt context-specific interventions.

Limitations

Several limitations were noted. First, our online recruitment might cause selection bias, underrepresenting individuals with low socio-economic status and limited digital literacy, possibly skewing the relationships between lifestyle and health outcomes. Future studies should use stratified sampling and offline data collection to improve representativeness and validate findings. Second, our reliance on self-reported data subject to recall bias, social desirability bias, and inaccurate self-assessment, potentially distorting the observed relationships between variables in the network. Although mitigated by validation questions and strict data quality controls, these limitations necessitate caution when interpreting the network structures and their implications. Future research should incorporate more accurate assessments, such as wearable technology or 24-hour recall, to enhance validity. Third, methodological limitations inherent in network analysis, such as assumptions of sparsity and potential overfitting, may affect the stability and generalisability of the identified network structures. The network estimation techniques rely on statistical models that may not fully capture the complex interplay between variables, and small changes in data can lead to different network configurations. Therefore, our network analysis results should be interpreted with consideration of these methodological constraints. Fourth, partial correlations within networks were based on individual perceptions of lifestyle and health outcome changes driven by the external factor of the COVID-19 pandemic. Although we anticipate these changes will revert in equal magnitude but opposite direction, preserving the relationships between variables post-pandemic, the findings require future validation during non-pandemic periods. Fifth, the cross-sectional design of our study during the COVID-19 pandemic limits our findings to associative insights and precludes the analysis of temporal network dynamics. By measuring outcomes relative to pre-pandemic conditions rather than their current status, the study’s applicability to the post-pandemic period may be constrained. However, if we posit a rebound effect of equal magnitude and opposite direction, the findings could retain their relevance in a post-pandemic context. To substantiate and expand upon these insights, we recommend conducting longitudinal studies and randomised controlled trials post-pandemic. Lastly, small sample sizes in several countries limited network interpretability and robust comparisons, especially concerning bridge lifestyles. Increasing sample sizes in future research could address this issue.

CONCLUSIONS

This study delineated central lifestyle, health outcome, and bridge lifestyle variables across 29 countries. Modifications in these identified variables hold the potential to significantly improve holistic lifestyle and holistic health outcomes, as well as holistic health outcomes via lifestyle, respectively. Their substantial impact supports the need for targeted interventions and increased resource allocation to achieve effective, comprehensive health improvements. The variation in these variables across different countries-shaped by diverse cultural, socio-demographic, political, and economic factors-emphasises the importance of customised strategies that exploit the unique effects of central or bridge lifestyles. Additionally, the similarities observed among some countries suggest opportunities for collaborative initiatives to promote global health.

Additional material

Online Supplementary Document

Acknowledgements

We would like to thank Miss Bobo Chan for her tremendous administrative support of the project.

Ethics statement: This study was approved by the Institutional Review Board of the University of Hong Kong-the Hospital Authority Hong Kong West Cluster (reference no: UW 20-272). Informed consent was obtained from all participants, ensuring they were aware of the study’s purpose, data collection methods, and their rights. To protect data privacy and handle sensitive health information confidentially, all participants completed the survey anonymously, and data storage and analysis were conducted in strict compliance with the Declaration of Helsinki, upholding the highest standards of data security and confidentiality.

Data availability: Data are available upon reasonable request from the corresponding author.

[1] Funding: This research received no specific grant from any funding agency, commercial or not-for-profit sectors.

[2] Authorship contributions: JL analysed the data and drafted the manuscript. DYTF is the principal investigator of the study and is responsible for conducting the study overall. DYTF, JYHW, MH, KYWL and EPHC conceived the study. DYTF, MH, KYWL, JYHW, EPHC, WD, JJL and CCL contributed to the design of the study. All authors contributed to data acquisition, critically appraised and approved the manuscript, and assume responsibility for the contents of the manuscript.

[3] Disclosure of interest: The authors completed the ICMJE Disclosure of Interest Form (available upon request from the corresponding author) and disclose no relevant interests.

references

[1] C Pisinger, S Ladelund, C Glümer, U Toft, M Aadahl, and T Jørgensen. Five years of lifestyle intervention improved self-reported mental and physical health in a general population: the Inter99 study. Prev Med. 2009;49:424-8. DOI: 10.1016/j.ypmed.2009.07.020. [PMID:19664653]

[2] VW-H Wong, FY-Y Ho, YS-H Wong, K-F Chung, W-F Yeung, and CH Ng. Efficacy of lifestyle medicine on sleep quality: A meta-analysis of randomized controlled trials. J Affect Disord. 2023;330:125-38. DOI: 10.1016/j.jad.2023.02.111. [PMID:36863476]

[3] M Parsamehr and SP Rasoulinezhad. The study of the relationship between lifestyle and social health among people of Talesh city. Quarterly Journal of Social Development. 2015;10:35-66.

[4] D Neupane, CS McLachlan, SR Mishra, MH Olsen, HB Perry, and A Karki. Effectiveness of a lifestyle intervention led by female community health volunteers versus usual care in blood pressure reduction (COBIN): an open-label, cluster-randomised trial. Lancet Glob Health. 2018;6:e66-73. DOI: 10.1016/S2214-109X(17)30411-4. [PMID:29241617]

[5] ST Nyberg, A Singh-Manoux, J Pentti, IEH Madsen, S Sabia, and L Alfredsson. Association of Healthy Lifestyle With Years Lived Without Major Chronic Diseases. JAMA Intern Med. 2020;180:760-8. DOI: 10.1001/jamainternmed.2020.0618. [PMID:32250383]

[6] Q Sun, D Yu, J Fan, C Yu, Y Guo, and P Pei. Healthy lifestyle and life expectancy at age 30 years in the Chinese population: an observational study. Lancet Public Health. 2022;7:e994-1004. DOI: 10.1016/S2468-2667(22)00110-4. [PMID:35926549]

[7] D Hevey. Network analysis: a brief overview and tutorial. Health Psychol Behav Med. 2018;6:301-28. DOI: 10.1080/21642850.2018.1521283. [PMID:34040834]

[8] N Noble, C Paul, H Turon, and C Oldmeadow. Which modifiable health risk behaviours are related? A systematic review of the clustering of Smoking, Nutrition, Alcohol and Physical activity (‘SNAP’) health risk factors. Prev Med. 2015;81:16-41. DOI: 10.1016/j.ypmed.2015.07.003. [PMID:26190368]

[9] I Corrales-Gutierrez, F Baena-Antequera, D Gomez-Baya, F Leon-Larios, and R Mendoza. Relationship between eating habits, physical activity and tobacco and alcohol use in pregnant women: Sociodemographic inequalities. Nutrients. 2022;14:557 DOI: 10.3390/nu14030557. [PMID:35276912]

[10] A Kramer. An overview of the beneficial effects of exercise on health and performance. Adv Exp Med Biol. 2020;1228:3-22. DOI: 10.1007/978-981-15-1792-1_1. [PMID:32342447]

[11] M Vajdi and MA Farhangi. A systematic review of the association between dietary patterns and health-related quality of life. Health Qual Life Outcomes. 2020;18:337 DOI: 10.1186/s12955-020-01581-z. [PMID:33046091]

[12] J Li, DYT Fong, KYW Lok, JYH Wong, MM Ho, and EPH Choi. Key lifestyles and interim health outcomes for effective interventions in general populations: A network analysis of a large international observational study. J Glob Health. 2023;13:04125 DOI: 10.7189/jogh.13.04125. [PMID:37861130]

[13] J Qiao, X Lin, Y Wu, X Huang, X Pan, and J Xu. Global burden of non-communicable diseases attributable to dietary risks in 1990–2019. J Hum Nutr Diet. 2022;35:202-13. DOI: 10.1111/jhn.12904. [PMID:33834556]

[14] PT Katzmarzyk, C Friedenreich, EJ Shiroma, and I-M Lee. Physical inactivity and non-communicable disease burden in low-income, middle-income and high-income countries. Br J Sports Med. 2022;56:101-6. DOI: 10.1136/bjsports-2020-103640. [PMID:33782046]

[15] R Alexander Bentley, DJ Ruck, and HN Fouts. US obesity as delayed effect of excess sugar. Econ Hum Biol. 2020;36:100818. DOI: 10.1016/j.ehb.2019.100818. [PMID:31540873]

[16] B Al-Ansari, AM Thow, CA Day, and KM Conigrave. Extent of alcohol prohibition in civil policy in Muslim majority countries: the impact of globalization. Addiction. 2016;111:1703-13. DOI: 10.1111/add.13159. [PMID:26508526]

[17] J Li, DYT Fong, MM Ho, EPH Choi, KYW Lok, and JJ Lee. Prevalence and Factors Associated With Willingness to Sustain Pandemic-Induced Digital Work in the General Population and Moderating Effects of Screen Hours: Cross-Sectional Study. J Med Internet Res. 2024;26:e53321. DOI: 10.2196/53321. [PMID:38805704]

[18] J Li, DYT Fong, KYW Lok, JYH Wong, M Man Ho, and EPH Choi. Global impacts of COVID-19 on lifestyles and health and preparation preferences: An international survey of 30 countries. J Glob Health. 2023;13:06031 DOI: 10.7189/jogh.13.06031. [PMID:37565394]

[19] KY-W Lok, RWT Ko, HSL Fan, JY Wong, EP Choi, and NWM Shek. International survey on fear and childbirth experience in pregnancy and the postpartum period during the COVID-19 pandemic: study protocol. BMJ Open. 2021;11:e050132. DOI: 10.1136/bmjopen-2021-050132. [PMID:34344684]

[20] J Li, DYT Fong, KYW Lok, JYH Wong, MM Ho, and EPH Choi. Key lifestyles and health outcomes across 16 prevalent chronic diseases: A network analysis of an international observational study. J Glob Health. 2024;14:04068 DOI: 10.7189/jogh-14-04068. [PMID:38606605]

[21] J Li, DYT Fong, KYW Lok, JYH Wong, MM Ho, and EPH Choi. Key lifestyles and interim health outcomes for effective interventions in general populations: A network analysis of a large international observational study. J Glob Health. 2023;13:04125 DOI: 10.7189/jogh.13.04125. [PMID:37861130]

[22] S Epskamp, D Borsboom, and EI Fried. Estimating psychological networks and their accuracy: A tutorial paper. Behav Res Methods. 2018;50:195-212. DOI: 10.3758/s13428-017-0862-1. [PMID:28342071]

[23] E Fonseca-Pedrero. [Network analysis in psychology]. Pap Psicol. 2018;39:1-12. Spanish

[24] PJ Jones, R Ma, and RJ McNally. Bridge centrality: A network approach to understanding comorbidity. Multivariate Behav Res. 2021;56:353-67. DOI: 10.1080/00273171.2019.1614898. [PMID:31179765]

[25] D Hevey. Network analysis: a brief overview and tutorial. Health Psychol Behav Med. 2018;6:301-28. DOI: 10.1080/21642850.2018.1521283. [PMID:34040834]

[26] JE Harnett, E McIntyre, A Steel, H Foley, D Sibbritt, and J Adams. Use of complementary medicine products: a nationally representative cross-sectional survey of 2019 Australian adults. BMJ Open. 2019;9:e024198. DOI: 10.1136/bmjopen-2018-024198. [PMID:31315853]

[27] M Hatta, M Balkish, K Rozanim, A Hamizatul, and A Norsiah. How Severe is Binge Drinking in Malaysia and Who are at Risk. J Alcohol Drug Depend. 2013;1:2

[28] AN Nurprasetia. Alcohol policies, consumption and government revenue: Indonesia. World Customs J. 2022;16:15-34. DOI: 10.55596/001c.116319

[29] TNP Nguyen, M Hunsberger, J Löve, TA Duong, TH Phan, and NK Luong. Patterns and determinants of tobacco purchase behaviors among male cigarette smokers in Vietnam: A latent class analysis. Tob Induc Dis. 2024;22:1-10. DOI: 10.18332/tid/187869. [PMID:38835515]

[30] R Brathwaite, J Addo, L Smeeth, and K Lock. A systematic review of tobacco smoking prevalence and description of tobacco control strategies in sub-Saharan African countries; 2007 to 2014. PLoS One. 2015;10:e0132401. DOI: 10.1371/journal.pone.0132401. [PMID:26162085]

[31] JH Lau, A Nair, E Abdin, R Kumarasan, P Wang, and F Devi. Prevalence and patterns of physical activity, sedentary behaviour, and their association with health-related quality of life within a multi-ethnic Asian population. BMC Public Health. 2021;21:1939 DOI: 10.1186/s12889-021-11902-6. [PMID:34696751]

[32] JP Nganabashaka, JB Niyibizi, G Umwali, SM Rulisa, C Bavuma, and JC Byiringiro. The effects of COVID-19 mitigation measures on physical activity (PA) participation among adults in Rwanda: An online cross-sectional survey. PLoS One. 2023;18:e0293231. DOI: 10.1371/journal.pone.0293231. [PMID:37943889]

[33] L Desbouys, C Méjean, S De Henauw, and K Castetbon. Socio-economic and cultural disparities in diet among adolescents and young adults: a systematic review. Public Health Nutr. 2020;23:843-60. DOI: 10.1017/S1368980019002362. [PMID:31466544]

[34] K Lobczowska, A Banik, S Forberger, K Kaczmarek, T Kubiak, and A Neumann-Podczaska. Social, economic, political, and geographical context that counts: meta-review of implementation determinants for policies promoting healthy diet and physical activity. BMC Public Health. 2022;22:1055 DOI: 10.1186/s12889-022-13340-4. [PMID:35619065]

[35] Quah S. Families in Asia: Home and kin. Milton Park, Abingdon, Oxon: Routledge; 2008.

[36] HOFSTEDE G. The 6-D model of national culture 2024. Available from: https://geerthofstede.com/culture-geert-hofstede-gert-jan-hofstede/6d-model-of-national-culture/. Accessed: 2 January 2025.

[37] B Tawatsupa, V Yiengprugsawan, T Kjellstrom, S-a Seubsman, A Sleigh, and TCS Team. Heat stress, health and well-being: findings from a large national cohort of Thai adults. BMJ Open. 2012;2:e001396. DOI: 10.1136/bmjopen-2012-001396. [PMID:23135538]

[38] MS Mohamed, DS Panda, FA Fadul, AA Saadeldin, MT Idriss, and YH Khan. Assessment of the impact of COVID-19 and political instability on mental health of university students in Sudan. East Mediterr Health J. 2024;30:272-82. DOI: 10.26719/2024.30.4.272. [PMID:38808403]

[39] A Ventriglio, J Torales, JM Castaldelli-Maia, D De Berardis, and D Bhugra. Urbanization and emerging mental health issues. CNS Spectr. 2021;26:43-50. DOI: 10.1017/S1092852920001236. [PMID:32248860]

[40] RL Schalock. The concept of quality of life: what we know and do not know. J Intellect Disabil Res. 2004;48:203-16. DOI: 10.1111/j.1365-2788.2003.00558.x. [PMID:15025663]

[41] F Nouh. Prevalence of food insecurity in eastern part of Libya: a study of associated factors. Scholars Acad J Biosci. 2021;8:192-8. DOI: 10.36347/sajb.2021.v09i08.002

[42] M Hameed, H Moradkhani, A Ahmadalipour, H Moftakhari, P Abbaszadeh, and A Alipour. A review of the 21st century challenges in the food-energy-water security in the Middle East. Water. 2019;11:682 DOI: 10.3390/w11040682

[43] I Petrikova, R Bhattacharjee, and PD Fraser. The ‘Nigerian Diet’and its evolution: review of the existing literature and household survey data. Foods. 2023;12:443 DOI: 10.3390/foods12030443. [PMID:36765972]

[44] P Mbah and B Olabisi. Promoting household food preparation as a strategy to combating nutritional problems in Nigeria: Key to sustainable development. Education and Science Journal of Policy Review and Curriculum Development. 2015;5:82-92.

[45] RS Kelly, MP Kelly, and P Kelly. Metabolomics, physical activity, exercise and health: A review of the current evidence. Biochim Biophys Acta Mol Basis Dis. 2020;1866:165936. DOI: 10.1016/j.bbadis.2020.165936. [PMID:32827647]

Correspondence to:
Daniel Yee Tak Fong
School of Nursing, Li Ka Shing Faculty of Medicine, University of Hong Kong
5/F, Academic Building 3 Sassoon Road, Pokfulam, Hong Kong
China
[email protected]