Risk factors for cardiovascular disease (CVD) are negatively correlated with socio-economic status (SES) in high-income countries (HIC) but there has been little research on their distribution by household SES within low-and middle-income countries (LMICs). Considering the limited data from LMICs, this paper examines the association between behavioural and cardiovascular risk factors and household SES in Papua New Guinea (PNG).
Reported here are results of 671 participants from the 900 randomly selected adults aged 15–65 years. These adults were recruited from three socioeconomically and geographically diverse surveillance sites (peri-urban community, rural Highland and an Island community) in PNG in 2013–2014. We measured their CVD risk factors (behavioural and metabolic) using a modified WHO STEPS risk factor survey and analysis of blood samples. We assessed SES by education, occupation and creating a household wealth index based on household assets. We calculated risk ratios (RR) and their 95% confidence intervals (CI) using a generalized linear model to assess the associations between risks and SES.
Elevated CVD risk factors were common in all SES groups but the CVD metabolic risk factors were most prevalent among homemakers, peri-urban and rural highlands, and the highest (4th and 5th) wealth quintile population. Adults in the highest wealth quintile had high risks of obesity, elevated HbA1c and metabolic syndrome (MetS) that were greater than those in the lowest quintile although those in the highest wealth quintiles were less likely to smoke tobacco. Compared to people from the Island community, peri-urban residents had increased risks of increased waist circumference (WC) (RR: 1.67, 95%CI: 1.21–2.31), hypertension (RR: 2∙29, 95%CI: 1∙89–4.56), high cholesterol (RR: 2∙22, 95%CI: 1∙20–4∙10), high triglycerides (RR: 1∙49, 95%CI: 1∙17–1∙91), elevated HbA1c (RR: 5∙54, 95%CI: 1∙36–21∙56), and Metabolic syndrome (MetS) (RR: 2∙04, 95%CI: 1∙25–3∙32). Similarly, Rural Highland residents had increased risk of obesity (Waist Circumference RR: 1∙70, 95%CI: 1∙21–3∙38, Waist-Hip-Ratio RR:1∙48, 95%CI: 1∙28–1∙70), hypertension (RR: 2∙60, 95%CI: 1∙71–3∙95), high triglycerides (RR: 1∙34, 95%CI: 1∙06–1∙70) and MetS (RR: 1∙88, 95%CI: 1∙12–3∙16) compared to those in the rural Island site.
CVD risk factors are common in PNG adults but their association with SES varies markedly and by location. Our findings show that all community members are at risk of CVD weather they are part of high or low SES groups. These results support the notion that the association between CVD risk factors and SES differ greatly accordingly to the type of SES measure used, risk factors and the population studied. In addition, our findings contribute further to the limited literature in LMIC. Longitudinal studies are needed to monitor changes in rapidly changing societies such as PNG to inform public health policy for control and prevention of NCDs in the country.
Citation: Rarau P, Pulford J, Gouda H, Phuanukoonon S, Bullen C, Scragg R, et al. (2019) Socio-economic status and behavioural and cardiovascular risk factors in Papua New Guinea: A cross-sectional survey. PLoS ONE 14(1): e0211068. https://doi.org/10.1371/journal.pone.0211068
Editor: Rasheed Ahmad, Dasman Diabetes Institute, KUWAIT
Received: September 24, 2018; Accepted: January 7, 2019; Published: January 23, 2019
Copyright: © 2019 Rarau et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The minimal database that supports the findings of this study is deposited in Figshare and will be made available to all researchers upon request to the project ethics committee, PNG Institute of Medical Research Institutional Ethical Review Board (IRB). Request should be sent to Mrs Norries Pomat email: email@example.com and Dr William Pomat email: William.firstname.lastname@example.org OR author Dr Patricia Rarau email: email@example.com
Funding: Unrestricted grant from PNG ExxonMobil Partnerships 2013-2014. Melbourne Research, University of Melbourne, PhD Nossal Scholarship for Global Health, Dr. Patricia Rarau. The funder of the study had no role in study design, data collection, data analysis, data interpretation, or writing of this manuscript. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication.
Competing interests: The commercial funding PNG ExxonMobil Partnerships for this study does not alter our adherence to PLOS ONE Editorial policies on sharing data and materials.
Cardiovascular diseases (CVDs) are leading causes of mortality throughout the world.[1–3] More than three quarters of all CVD-related deaths occur in low-and middle-income countries (LMICs), most commonly due to coronary heart disease or stroke.[1, 2]
The major risk factors for CVD are well described: tobacco smoking, high blood pressure, high total and low-density lipoprotein (LDL) cholesterol, type 2 diabetes mellitus (T2DM) and physical inactivity.[1, 3] The co-occurrence of cardiovascular risk factors and metabolic syndrome (MetS), a cluster of at least three of five CVD risk factors, that is associated with the development of both CVDs and T2DM, further increases lifetime risk of a CVD event.[3, 5]
An inverse association between socio-economic status (SES) with CVDs and their risk factors has been reported in many high income countries (HIC).[6–8] A review of studies conducted in HICs reported higher risk of incidence of stroke among those with lower SES compared to those with higher SES. Although complex, this association appears to be related to behavioural, environmental and early life factors that are more common among individuals of lower than higher SES, such as poorer access to health care, use of alcohol, tobacco and drugs, low birth weight and high rates of childhood illnesses.[9, 10]
The evidence for a similar association between CVD risk and SES in low- and middle-income countries (LMICs) is less clear. A recent review reported that high SES in LMICs was associated with a higher prevalence of CVD risk factors such as low physical activity and excessive consumption of salt and fat while low SES was associated with other forms of CVD risk such as tobacco and alcohol use and limited fruit and vegetable consumption. Furthermore, these relationships varied for the different CVD risk factors. An earlier review also reported variable prevalence of CVD risk factors among different socioeconomic groups in LMICs. These variations have been attributed to the complex interplay between the underlying cultural, social, economic, environmental and political determinants of health [9, 13, 14], and more specifically, the stage and type of economic development  and the associated impact on dietary and behavioural patterns.
The prevalence of CVD risk factors in the Western Pacific is substantial and growing.[17–19] Papua New Guinea (PNG) is no exception.[20, 21] A recent review article of general population prevalence studies indicated a substantial increase in CVD risk has occurred between 1960s and 2007 (submitted for publication). In addition, findings of a recent cross-sectional study showed a high prevalence of CVD risk factors among peri-urban residents compared to rural adults.  However, the pace and extent to which PNG populations are exposed to CVD risk factors differs greatly across ethnic and regional groups.
Little is known about the distribution of CVD risk factors according to SES in PNG. Given the differences in sub-national economic development profiles, PNG offers a unique opportunity to study the relationship between SES and CVD risk in divergent economic contexts within a LMIC. Therefore, we aimed to measure the distribution of these risk factors in three sites that are at varying levels of socioeconomic development
Materials and methods
PNG is a lower-middle income country of 7.9 million people. The country’s economy is heavily reliant on extractive industries, particularly the mining of minerals. More than 80% of the population still live in rural areas, the majority of whom are engaged in subsistence farming or cash cropping. Populations within PNG differ by genetics, duration and exposure to globalization and development.[24–26]
Economic development (measured by fluctuations in GDP growth (annual %), which peaked at 18% in 1993 and again at 15% in 2014),  has led to rapid social and demographic changes that are unevenly distributed across the country, with provinces at various stages of development.[24, 28]
We collected data from people living in the three sites of the PNG Institute for Medical Research’s integrated Health and Demography Surveillance System (iHDSS): West Hiri (Central Province), Asaro (Eastern Highlands Province) and Karkar Island (Madang Province).The West Hiri (peri-urban) site comprises villages of Austronesian ancestry, distributed along the coastline north-west of and in close proximity to PNG’s capital city, Port Moresby.[29, 30] The villages with a population estimated at around 11531, surround a large gas processing plant and are located within an hour’s drive of Port Moresby, a rapidly growing city established over a century ago. It is the most developed of the three sites, with good access to government services and infrastructure, such as education and health care, electricity and water supply. In contrast, the Asaro is a highlands rural site comprised of people who are of Non-Austronesian ancestry.. The population live in hamlets around 40–45 kilometres north-east of the Eastern highlands town of Goroka. The Asaro people are primarily subsistence farmers, but many earn cash through smallholder production of coffee, employment on plantations and selling garden products.[29, 31] The people of Karkar Island are of both Austronesian and Non-Austronesian ancestry  living on a volcanic island 30 kilometres off the northern coastline of the provincial capital, Madang. Karkar is a rural site and the least developed site of the three; most adult residents are subsistence farmers or unskilled labourers and access to Madang is by boat to the mainland and then by road, a journey of at least an hour.
We conducted a cross-sectional, community-based survey on randomly selected adults in the three iHDSS sites between April 2013 and October 2014. Our methods have been published elsewhere  but in brief, the sampling frame was a full population census of the adult population of each site. Using a simple randomization procedure, 900 individuals were randomly selected to participate in the survey, of which we aimed to recruit 300 adult participants from each site, stratified by sex and three age groups (15–29, 30–44, 45–65 years). The sample size was calculated to confer a 80% power to detect 10% absolute difference in the proportion of most risk factors from all ages combined between sites.
The eligibility criteria for study participants were: 15–65 years of age, residing in the iHDSS site and not pregnant (for females). Prior to participation, written informed consent was obtained from participants (18+ years) and the parents or guardians of those aged less than 18 years, We interviewed selected participants at their homes or community health facilities, using the modified WHO STEPS NCD Risk Factor Survey Questionnaire. Domains included: demographics, self-reported health status, dietary/nutrition history, tobacco, betelnut and alcohol use, physical activity, participant history of NCD and/or associated treatments. The survey questionnaire was available in English and Tok Pisin (the language used by the study populations at all three sites) and was piloted extensively prior to survey commencement. All survey forms and procedures were completed at a single time point by a trained survey team. Furthermore, The NCD data were linked with household SES data collected in 2015, extracted from the iHDSS database.
Measurement of cardiovascular risk factors
Metabolic risk factors.
According to the International Diabetes Federation and WHO criteria, central obesity was defined as a waist circumference ≥94 cm (males) and ≥80cm (females) and/or by waist to hip ratio ≥0·90 and ≥0·85 for males and females, respectively.[4, 34] Hypertension was defined as the average of the three systolic and diastolic BP of ≥130/85 mmHg and/or those on treatment for hypertension.[4, 35] Non-fasting serum samples were analysed for lipids (cholesterol, triglyceride and HDL-Cholesterol) using Vitros 250/350 Biochemistry System from Ortho Clinical Diagnostics, in batches within a month of collection. Elevated total cholesterol (TC) and triglyceride (TG) levels were defined as levels of >6·2mmol/L and ≥2·3 mmol/L respectively. Low HDL-c levels was defined as <1mmol/L (males) and <1·3mmol/L (females).[4, 36] T2DM was diagnosed if the participant was on anti-diabetic drugs or if their haemoglobin A1c (HbA1c) was ≥6·5%.[4, 37]. We used the IDF definition for metabolic syndrome (MetS), characterised by having large waist circumference (males ≥94cm and females ≥80cm) as a marker of central obesity plus any two of the following: hypertension (≥130/85mmHg), HbA1c levels (≥6·5%), elevated triglycerides (≥1·7mmol/L) and low HDL-c (<1·0/mmol/L M & <1·3mmol/L F).
Behavioural risk factors.
We included behavioural risk factors associated with an increased risk of CVDs and diabetes.[38, 39] Definitions included: smoking, daily use of any form of tobacco product; alcohol use, any amount of alcohol consumption within the last 30 days; chewing betelnut (areca nut, betel leaf/bean and slaked lime), any amount of betelnut chewing within the last 30 days; low fruit and vegetable intake, consumption on five or fewer days per week; high sugar intake, adding at least 6 teaspoons of sugar to hot drink(s) daily or drinking at least 3 cans of soft drink per week; high fried food intake, consuming fried food on at least five days per week; and daily salt intake, adding salt and/or stock cubes directly to food each day.
We used education, occupation and household wealth indices as markers for SES. Though interrelated, each of these indicators influences health status and outcome of individuals differently. We therefore investigated the association between each individual SES indicator (i.e. education, occupation and wealth) and cardiovascular risk factors. Typically, household income comes from many sources that vary by season, making it difficult to collect income data. Therefore, we used a wealth index, an approach used by the World Bank and Demographic and Health Surveys (DHS) as a proxy measure of income/economic status.[41–43]
We used STATA/SE version 14.0 for all data analyses. We constructed the household wealth index from household data including household durables or assets and characteristics of participants’ houses (type, floor materials used and electricity, number of bedrooms, toilet and water quality) and ownership of a bank account. Using principal component analysis (PCA) we constructed wealth quintiles, based on the following variables: water quality (low, medium, high); toilet quality (low, medium, high); type of house (low, medium, high); material used for the floor (low, medium, high); electricity (yes/no); number of bedrooms in the house; ownership of a bank account (yes/no) and number of household assets owned. The list of household assets included: air conditioning; fan; tractor; truck; car; boat or canoe; fridge; freezer; generator; washing machine; computer; landline or mobile phone; cupboard; bedframes; watch/clock, radio; gas or electric cooker; table and chairs. For each site, household quintiles were generated using the “xtile” command which constructed a new variable with five categories; lowest, second, third, fourth and highest quintiles, representing poorest to richest groups, respectively.
We estimated cardiovascular risk factor prevalence with 95% CI and used Pearson’s chi-square and Fisher’s exact test, where appropriate, to assess inter-site and socioeconomic differences in the risk factors. Risk ratios (RR) were determined, using the log binomial regression and a modified Poisson regression with a robust error variance for the univariate analysis and multivariate analysis respectively. We used available case analysis to deal with missing values.
The study was approved by the PNGIMR Institutional Review Board (IRB) and the PNG Medical Research Advisory Committee (MRAC) (IRB No. 1208, 23 March 2012; MRAC No 12.34, November 2012) and registered with the University of Melbourne, School of Population and Global Health Human Ethics Advisory Group (ID: 1647305.1). Written informed consent was obtained from individual participants (aged 18+ years) and from the parents or guardians of participants who were less than 18 years of age prior to study participation.
Table 1 presents the socio-economic and demographic characteristics of the study population (N = 671). The three populations were different in terms of education and employment, but comparable in terms of participants’ age and sex. A higher percentage (34∙8%) of the participants from West Hiri had at least a secondary level education, followed by Karkar participants (15∙1%) then Asaro (11%). Overall, 21% of the participants received some, or completed, secondary education and 4% had completed at least some tertiary education. A higher percentage of Asaro and Karkar Island participants were subsistence farmers (73% and 57% respectively) compared to West Hiri (8%). A higher percentage of participants from West Hiri are engaged in paid employment (34%) compared to Karkar Island and Asaro with only 5% in both sites.
The values are numbers and percentages.
Table 2 shows the prevalence and RR of behavioral CVD risk factors by SES. Females were less likely to smoke tobacco (RR 0.34), less likely to consume alcohol (RR 0.2) and sugar (RR 0.68) than males. Most of the behavioral risk factors were highly prevalent among West Hiri and Asaro participants as compared to Karkar residents, except for tobacco smoking. West Hiri participants were more likely to consume alcohol (RR 6.84), more likely to have high sugar (RR 4.34), salt (RR 3.66) and fried food (RR 19.6) intake than Karkar Island participants, Positive associations were observed between adults with tertiary/vocational education and alcohol use (RR 2.41), less fruits and vegetables (RR 2.76), high salt (RR 1.45) and fried food intake (RR 1.52). Similarly, adults in paid employment were more likely to engage in most of the behavioral risk factors compared to home makers. Positive associations were observed between adults engaged in paid employment with tobacco smoking (RR 2.31), alcohol use (RR 3.49), intake of high salt (RR 1.22) and fried food (RR 1.34) than home makers. Adults in the highest quintile had relatively lower risks of tobacco smoking than those in the lowest wealth quintile group, however this was not significant.
Table 3 shows the prevalence and RR of metabolic CVD risk factors by SES. Females were observed to have higher risks for most metabolic CVD risk factors except for hypertension than males (RR 0∙74). They were more likely to have obesity by both WC (RR 5.78) and WHR (RR 1.18), have low HDLc (RR 1.53) and have MetS (RR 4.7) than males. In comparison to Karkar participants, the metabolic CVD risk factors were more common among West Hiri participants, except for central obesity, by WHR (RR 0∙82) and low hdl-c levels (RR 0∙7). West Hiri participants were more likely to have increased WC (RR 2.25), hypertension (RR 2.50), have elevated TC (RR 2.05) and TG (RR 1.43), elevated HbA1c (RR 8.63) and have MetS (RR 2.63) than adults from Karkar Island. Furthermore, adults engaged in paid employment were more likely to be hypertensive (RR 1∙40) but less likely to have obesity by WC (RR 0.49) and WHR (RR 0.79) and have MetS (RR 0∙50) compared to home makers. Conversely, adults in the highest wealth quintile had higher risk of obesity by WC (RR 1∙99) and MetS (RR 1∙84) compared to those in the lowest quintile.
The multivariate analysis showed West Hiri and Asaro adults, respectively had higher risk for low fruit/vegetable consumption (RR 102.9 & RR 13.7), high alcohol use (RR 7.93 & RR 3.20), high sugar (RR 4.63 & RR 2.49), salt (RR 3.50 & RR 3.72) and fried food intake (RR 52.6 & RR 56.2), obesity (WC) (RR 1.67 & RR 1.70), hypertension (RR 2.93 & RR 2.60), high cholesterol (RR 2.22 & RR 1.79), triglyceride (RR 1.49 & RR 1.34), and MetS (RR 2.04 & RR 1.88) compared to Karkar adults. (Table 4) Multivariate analysis showed adults with higher education had less risk of having elevated HbA1c levels (RR 3.04e-08) but 1∙12 times the risk of eating fried food than those with lower than primary education. Adults in paid employment had 43% reduced risk of eating less fruits and vegetables compared to home makers. Adults in the highest wealth quintiles had 1∙55, 1∙98 and 1∙70 times the risk of having high sugar intake, obesity (WC) and MetS, respectively, compared to those in lowest quintile.
We have previously reported on the prevalence of NCD risk factors in three different regions of PNG . We observed high prevalence of low fruit and vegetable intake, high intake of sugar, salt and fried food and alcohol, obesity (WC), hypertension and low HDL-c levels among those with higher education compared to adults in the lowest education group. However, tobacco smoking, central obesity (WHR) and high triglyceride levels were more prevalent among adults with lower education. Furthermore, among those in paid employment, we observed higher prevalence of tobacco smoking, alcohol use, less fruits and vegetables intake, high intake of sugar, salt and fried food, hypertension and elevated HbA1c levels. In addition, we found high sugar intake, obesity (WC), hypertension, elevated HbA1c levels and MetS was more prevalent among the highest wealth quintile group compared to those in the lowest quintile group. However, high levels of tobacco smoking, alcohol use and betelnut chewing and high TC were more prevalent among adults in the lowest wealth quintile group.
The patterning of CVD risk factors in our study supports the notion that risk factors and disease risk are impacted differently according to the influences of socioeconomic transitions experienced by different individuals and their communities. A recent review of the association between CVD and SES in LMICs reported that people in low SES groups generally had higher prevalence of alcohol, tobacco smoking and consumed less vegetables and fruits while adults of high SES groups consume more processed food and were less physically active. Furthermore, reviews of studies which were conducted in other LMICs have also shown large variation in these associations where the direction of the associations differed from one country to another and differed for the different CVD risk factors.[11, 12, 45]
A study in India which collected data from participants aged 26–32 between 1998 to 2002, observed higher prevalence of CVD risk factors such as obesity and hypertension among those with higher SES, but the pattern did not hold for tobacco smoking. Similar findings were observed in a more recent study which looked at the socio-demographic patterning of CVD risk factors among rural populations from major cities in India. This variability in the CVD risk factors is also observed in our findings across the three study samples. For example, adults with higher educational attainment were observed to have high salt and fried food intake, consume less fruits and vegetables, but were less likely to smoke tobacco, have high TC and HbA1c levels.
Some of our findings differ to those from other LMICs such as that observed in a study among adults in Aleppo, Syria in 2006. The Aleppo study observed an inverse association between education and CVD risk factors. Their results showed diabetes, obesity, hypertension and high cholesterol levels were highly prevalent among people with low education levels. An inverse association was also observed between employment status and CVD risk factors, with a higher prevalence of hypertension and diabetes among those with no employment compared to the employed. In contrast, our results showed adults with paid employment had a higher prevalence of tobacco smoking, alcohol use, consumed less fruits and vegetables, high sugar, salt and fried food intake, obesity, hypertension, HbA1c and MetS compared to adults with home duties.
An article published in 2012 which looked at the association between CVD risk factors and education, occupational and SES in an Asian Indian population, reported these risk factors to be highly prevalent among adults with low SES. The study observed a positive association between low SES and truncal obesity, high triglyceride and low HDL-c levels, tobacco smoking and low physical activity. These findings were in contrast to earlier studies from India that reported people in lower SES groups had a lower CVD risks.
Urbanization in PNG appears to be accelerating the CVD risk in rural areas to levels comparable to longer-term urban or peri-urban populations, such as that seen in other LMICs.[51, 52] As economic development progresses, the SES gradient tends to become more regressive.(9, 12) This is clearly illustrated by Kerr GD et al and Allen L et al.[8, 11] These two articles reviewed studies which looked at the association of SES and NCD and risks factors which showed that at advanced stages of socioeconomic transition (HIC), there are clear SES gradients of NCDs and risk factors  while in LMIC, countries are at different stages of these emerging, hence mix trends are observed. It is apparent that different stages of socioeconomic transition produce different distributions of risk factors across different SES groups, thus adds to the understanding of which groups change which behaviors at these stages of transition. In the current study, we observed mixed findings for the association between SES and CVD risk factors.
A recent study by Gouda et al (submitted and will be published in due course) of verbal autopsy studies in PNG, from 1970–2001 and 2009–2014, showed that an epidemiological transition in mortality is underway in PNG. Deaths from NCDs, particularly diabetes, stroke and ischaemic heart disease, are on the rise, more so in peri-urban West Hiri and Asaro than Karkar Island sites. This mortality data is evidence of the existing and high prevalence of NCD risk factors among populations who have had a greater and increasing exposure to modernization and earning an income through employment or cash crops as observed in our study.(20)
Strengths and limitations
The strengths of this study include our use of robust objective measures, standard methods and instruments to collect data as well as blood sample analysis. The limitations include the cross-sectional design, limiting our ability to draw causal inferences, and therefore, only providing a snapshot in a rapidly-changing society. In addition, the target sample of 900 adults was not achieved. It proved difficult to have study participants fasting for the blood sample collection, therefore non-fasting blood samples were used for the analysis. The behavioral risk factors were self-reported using standardized questions. Finally, we created a wealth index from household data, an approach that is commonly used, but was challenging to construct and interpret, and the possibility of misclassification across the wealth quintiles could be a threat to validity.
The findings from this study highlight variability in CVD risk factor prevalence that is not fully explained by SES, therefore, further studies in such settings would enable a better understanding of these associations. Our findings suggest that all community members are at risk of CVD whether they are part of high or low SES groups. Such changes should be of major concern for PNG policy makers as it suggests the rise in CVD is not confined to urban areas. To conclude, this study provides baseline from which trends can be observed. If the adverse health impacts from consumption of alcohol, tobacco and processed foods are to be mitigated equitably, whole of population strategies, such as those identified by WHO as “Best Buys”, need to be implemented throughout PNG, but with consideration of adapting them where needed to fit diverse local contexts.
The authors thank all research participants, government, church health authorities and community leaders in the respective sites. The authors are grateful to the NCD Study field, laboratory and the iHDSS staff. The authors also acknowledge the PNGIMR for providing the administrative and technical support to conduct this study and the Melbourne University Statistical Group for statistical advice provided
- 1. WHO. Cardiovascular Diseases (CVDs) 2017 [updated May 2017; cited 2017 December 2017]. Available from: http://www.who.int/mediacentre/factsheets/fs317/en/.
- 2. Roth GA, Johnson C, Abajobir A, Abd-Allah F, Abera SF, Abyu G, et al. Global, Regional, and National Burden of Cardiovascular Diseases for 10 Causes, 1990 to 2015. Journal of the American College of Cardiology. 2017.
- 3. Shah AD, Langenberg C, Rapsomaniki E, Denaxas S, Pujades-Rodriguez M, Gale CP, et al. Type 2 diabetes and incidence of cardiovascular diseases: a cohort study in 1·9 million people. The Lancet Diabetes & Endocrinology. 2015;3(2):105–13. https://doi.org/10.1016/S2213-8587(14)70219-0.
- 4. Federation ID. The IDF consensus worldwide definition of the Metabolic Syndrome. 2006 2006. Report No.
- 5. Wilson P, Meigs J. Cardiometabolic risk: a Framingham perspective. International journal of obesity. 2008;32:S17–S20. pmid:18469835
- 6. Kaplan GA, Keil JE. Socioeconomic factors and cardiovascular disease: a review of the literature. Circulation. 1993;88(4):1973–98.
- 7. Kuper H, Adami H-O, Theorell T, Weiderpass E. The Socioeconomic Gradient in the Incidence of Stroke. A Prospective Study in Middle-Aged Women in Sweden. 2007;38(1):27–33. pmid:17138948
- 8. Kerr GD, Slavin H, Clark D, Coupar F, Langhorne P, Stott DJ. Do vascular risk factors explain the association between socioeconomic status and stroke incidence: a meta-analysis. Cerebrovasc Dis. 2011;31. pmid:20980755
- 9. Wilkinson RG, Marmot M. Social determinants of health: the solid facts: World Health Organization; 2003.
- 10. Liu RS, Burgner DP, Sabin MA, Magnussen CG, Cheung M, Hutri-Kähönen N, et al. Childhood Infections, Socioeconomic Status, and Adult Cardiometabolic Risk. Pediatrics. 2016;137(6). pmid:27235447
- 11. Allen L, Williams J, Townsend N, Mikkelsen B, Roberts N, Foster C, et al. Socioeconomic status and non-communicable disease behavioural risk factors in low-income and lower-middle-income countries: a systematic review. The Lancet Global health. 2017;5(3):e277–e89. Epub 2017/02/15. pmid:28193397; PubMed Central PMCID: PMCPMC5673683.
- 12. Sommer I, Griebler U, Mahlknecht P, Thaler K, Bouskill K, Gartlehner G, et al. Socioeconomic inequalities in non-communicable diseases and their risk factors: an overview of systematic reviews. BMC Public Health. 2015;15(1):914. pmid:26385563
- 13. Marmot M. Social determinants of health inequalities. The Lancet. 2005;365(9464):1099–104.
- 14. Havranek EP, Mujahid MS, Barr DA, Blair IV, Cohen MS, Cruz-Flores S, et al. Social determinants of risk and outcomes for cardiovascular disease. Circulation. 2015;132(9):873–98. pmid:26240271
- 15. Zimmet P. Globalization, coca-colonization and the chronic disease epidemic: can the Doomsday scenario be averted? Journal of Internal Medicine. 2000;247:301–10. pmid:10762445
- 16. Zimmet PZ, Alberti K. Introduction: Globalization and the non‐communicable disease epidemic. Obesity. 2006;14(1):1–3. pmid:16493116
- 17. WHO. GLOBAL STATUS REPORT on noncommunicable disease 2014. WHO, editor. Geneva, Switzerland: World Health Organization; 2014 2014. 298 p.
- 18. WHO WPR. Western Pacific Regional Action Plan for the Prevention and Control of Noncommunicable Diseases_2014–2020. WHO, 2014 978 92 9061 655 9.
- 19. Kessaram T, McKenzie J, Girin N, Roth A, Vivili P, Williams G, et al. Noncommunicable diseases and risk factors in adult populations of several Pacific Islands: results from the WHO STEPwise approach to surveillance. Australian and New Zealand journal of public health. 2015;39(4):336–43. pmid:26095921
- 20. Rarau P, Vengiau G, Gouda H, Phuanukoonon S, Kevau IH, Bullen C, et al. Prevalence of non-communicable disease risk factors in three sites across Papua New Guinea: a cross-sectional study. BMJ Global Health. 2017;2(2):e000221. pmid:29242751
- 21. PNG NDoH, HOPE Worldwide P, WHO. Papua New Guinea NCD Risk Factor STEPS Report. Port Moresby: National Department of Health, PNG; 2014.
- 22. Saweri W. The rocky road from roots to rice: a review of the changing food and nutrition situation in Papua New Guinea. Papua New Guinea Medical Journal. 2001;44(3/4):151–63.
- 23. Bank TW. The World Bank in Papua New Guinea: The World Bank; 2017 [cited 2017 February 2017]. Available from: http://www.worldbank.org/en/country/png/overview.
- 24. PNG NSO. Papua New Guinea Demography and Health Survey 2006 National Report. Port Moresby: National Statistical Office; 2009 [cited 2016. Available from: https://www.nso.gov.pg/index.php/projects/demographic-health-survey/48-dhs.
- 25. PNG NSO. 2009–2010 Papua New Guinea Household Income and Expenditure Survey: PNG National Statistical Office; 2013.
- 26. Howes S, Mako AA, Swan A, Walton G, Webster T, Wiltshire C. A lost decade? Service delivery and reforms in Papua New Guinea 2002–20012. The National Research Institute and the Development Policy Centre, Canberra, 2014 2014. Report No.
- 27. Group TWB. 2018 [cited 2018 March 2018]. Available from: https://data.worldbank.org/indicator/NY.GDP.MKTP.KD.ZG?locations=PG.
- 28. Cornish M, Fox R, Howes S, Nicholas W, Prabhakar A, Rova A. PNG survey of recent developments, 2014–15. 2015.
- 29. Gouda H, Vengiau G, Phuanukoonnon S, Siba P. Report of Partnership in Health Project (PiHP) 2011–2012. Report. PNG Institute of Medical Research, 2013.
- 30. Wurm S. Linguistic prehistory in the New Guinea area. Journal of Human Evolution. 1983;12(1):25–35.
- 31. Benediktsson K. Harvesting development: The construction of fresh food markets in Papua New Guinea. Denmark: NIAS Press; 2002.
- 32. Boyce A, Harrison G, Platt C, Hornabrook R, Serjeantson S, Kirk R, et al. Migration and genetic diversity in an island population: Karkar, Papua New Guinea. Proceedings of the Royal Society of London B: Biological Sciences. 1978;202(1147):269–95.
- 33. WHO. STEPwise approach to noncommunicable disease risk factor surveillance (STEPS). Available from: http://www.who.int/chp/steps/instrument/en/.
- 34. WHO EC. Waist Circumference and Waist-Hip Ratio Report of a WHO Expert Consultation 2008. Report. Geneva, Switzerland: WHO, 2011 978 92 4 150149 1.
- 35. Chobanian AV, Bakris GL, Black HR, Cushman WC, Green LA, Izzo JL Jr., et al. Seventh report of the Joint National Committee on Prevention, Detection, Evaluation, and Treatment of High Blood Pressure. Hypertension. 2003;42(6):1206–52. Epub 2003/12/06. pmid:14656957.
- 36. Evaluation EPoD. Executive summary of the third report of the National Cholesterol Education Program (NCEP) expert panel on Detection, Evaluation, and Treatment of high blood cholesterol in adults (Adult Treatment Panel III). Jama. 2001;285(19):2486. pmid:11368702
- 37. Association AD. Standards of medical care in diabetes—2010. Diabetes care. 2010;33(Supplement 1):S11–S61.
- 38. WHO. Global status report on noncommunicable diseases 2010. WHO, editor. Geneva, Switzerland: World Health Organization; 2011 2011. 176 p.
- 39. Javed F, Bello Correra FO, Chotai M, Tappuni AR, Almas K. Systemic conditions associated with areca nut usage: a literature review. Scand J Public Health. 2010;38(8):838–44. pmid:20688790.
- 40. Winkleby MA, Jatulis DE, Frank E, Fortmann SP. Socioeconomic status and health: how education, income, and occupation contribute to risk factors for cardiovascular disease. American Journal of Public Health. 1992;82(6):816–20. PMC1694190. pmid:1585961
- 41. Filmer D, Scott K. Assessing asset indices. Demography. 2012;49(1):359–92. Epub 2011/12/03. pmid:22135117.
- 42. Gwatkin DR, Rutstein S, Johnson K, Suliman E, Wagstaff A, Amouzou A. Socio-economic differences in health, nutrition, and population within developing countries: Washington, DC, World Bank; 2007.
- 43. Rutstein SO, Johnson K. The DHS wealth index. DHS comparative reports no. 6. Calverton: ORC Macro. 2004.
- 44. Peugh JL, Enders CK. Missing Data in Educational Research: A Review of Reporting Practices and Suggestions for Improvement. Review of Educational Research. 2004;74(4):525–56.
- 45. Hosseinpoor AR, Bergen N, Kunst A, Harper S, Guthold R, Rekve D, et al. Socioeconomic inequalities in risk factors for non communicable diseases in low-income and middle-income countries: results from the World Health Survey. BMC public Health. 2012;12(1):912.
- 46. Samuel P. Socio-economic status and cardiovascular risk factors in rural and urban areas of Vellore, Tamilnadu, South India. International journal of epidemiology. 41(5):1315–27. pmid:22366083
- 47. Kinra S, Bowen LJ, Lyngdoh T, Prabhakaran D, Reddy KS, Ramakrishnan L, et al. Sociodemographic patterning of non-communicable disease risk factors in rural India: a cross sectional study. Bmj. 2010;341:c4974. pmid:20876148
- 48. Al Ali R, Rastam S, Fouad FM, Mzayek F, Maziak W. Modifiable cardiovascular risk factors among adults in Aleppo, Syria. International journal of public health. 2011;56(6):653–62. pmid:21814848
- 49. Gupta R, Deedwania PC, Sharma K, Gupta A, Guptha S, Achari V, et al. Association of educational, occupational and socioeconomic status with cardiovascular risk factors in Asian Indians: a cross-sectional study. PLOS one. 2012;7(8):e44098. pmid:22952886
- 50. Kar S, Thakur J, Virdi N, Jain S, Kumar R. Risk factors for cardiovascular diseases: Is the social gradient reversing in northern India. 2010.
- 51. Mohan I, Gupta R, Misra A, Sharma KK, Agrawal A, Vikram NK, et al. Disparities in prevalence of cardiometablic risk factors in rural, urban-poor, and urban-middle class women in India. PloS one. 2016;11(2):e0149437. pmid:26881429
- 52. Makoutodé M. Lifestyle and dietary factors associated with the evolution of cardiometabolic risk over four years in West-African adults: the Benin study. Journal of obesity. 2013;2013.