This study is the first of its kind for Indonesian adolescents. We found evidence of a significant relationship between the major NCD risk factors and current indicators of mental wellbeing in our novel sample. Psychological distress was associated with every risk domain, physical function QoL was associated with adiposity, physical inactivity, and excess sedentary time, and general QoL was associated with substance use, physical inactivity, and diet. Those adolescents with risk factors present in more risk domains had higher mean scores for psychological distress, a scale which has good ability to detect depression and anxiety in this population. 18 Moreover, we saw a relationship between increasing co-occurring risk factors and a reduction in quality of life across two differing scales, physical function and general QoL. This evidence of current diminished wellbeing indicates that intervening on NCD risk factors could improve the mental health and wellbeing for young people now, not just for the benefit of future adult health.
The trends described here between common NCD risks and wellbeing suggest an important relationship between an adolescent’s current wellbeing and their experience of NCD risk, and notably, the number of co-occurring risks they are exposed to. Interestingly, we did not see this pattern replicated in our subsample with multiple biomarker risks. This is consistent with another study in young adults which examined several risk biomarkers, including high fasting glucose, triglycerides, low HDL cholesterol, high blood pressure, and high waist circumference and found a low correlation only between waist circumference and psychological distress. 32 These findings could potentially provide some of the context in which to interpret our results. The social and behavioural nature of many of these risks (such as smoking cigarettes, the foods they consume, or participation in physical activity) could be more closely tied to adolescents’ wellbeing than the biological condition indicated by their blood lipid profile. That is not to say the biological impacts measured here are not relevant to wellbeing, there are likely other mediating factors. A recent meta-analysis of cohort studies examining the relationship between clustered metabolic risks and depressive symptoms showed consistently significant associations in western countries, but not in Asian countries. 33 Meanwhile, a population-based study of adults with diabetes found that symptoms of depression had a stronger association with the experience of diabetic symptoms than measures of glycaemic control (HbA(1c) levels). 34 So, while the relationship with co-occurring biomarkers and mental wellbeing may be less clear in our adolescent sample, there is an interesting and novel association between NCD risks and mental wellbeing. Further research is needed to unpack the direction of these associations, but the importance of this finding and the opportunities to prevent ill-health could be substantial.
There have been calls over the past decade to consider the significance of the major NCD risk factors in a public health approach to prevention and control of common mental disorders. 35, 36 These risks are largely modifiable, and while the effect on mental health on the individual level may not be as significant as preventing the major risks for mental disorder, such as adverse childhood experiences or poly-victimisation, the potential impact on wellbeing at the population level (and cumulative impact across disease outcomes) should not be overlooked. 37, 38 Results from previous studies have been mixed; systematic reviews by Wu et al (2017) and Hoare et al (2014) found a relationship between several NCD risks and HRQOL, and depressive symptoms, respectively, however the authors questioned the quality and representativeness of the studies reviewed in both cases. 16, 39 Our results indicate that there is a potential for effective public health policy and interventions which address behavioural NCD risks to improve the current wellbeing of young people, as well as potentially reducing their likelihood of developing an NCD in later life, or reducing the associated disability experienced from potential NCD outcomes if they arise. 40 If there were effective interventions which could be shown to improve both current wellbeing as well as future health outcomes, the immediate impact on wellbeing could be a stronger motivator for adolescents than those actions promising a healthier future. The mixed results from previous studies, in the context of our strong results indicate the need for further research in this field to examine the complex relationships explored here, in this important age group.
Socio-demographic factors at the individual, household, and societal level are often highlighted in the literature as an important part of the underlying mechanisms to explain the close relationship between mental wellbeing, NCDs, and their risk factors. 35, 40 Our large sample enabled us to describe the distribution of key NCD risk factors and outcomes for adolescent wellbeing by relevant socio-demographic factors. Sex differences, while not ubiquitous, were sizeable where present. Males had a notably higher prevalence of substance use, and females had double the risk of inadequate physical activity. This reflects global findings for sex differences in physical inactivity and substance use. 41–43 Adiposity was more prevalent in highly urbanised Jakarta, and likewise amongst those in the sample who self-identified as being of higher-than-average socio-economic status. While this fits with a common global narrative around urbanisation and obesogenic risk factors, recent findings both globally and from Indonesia encourage a more nuanced analysis, with rural BMI increasing faster (despite being lower overall) than in cities in low- and middle-income regions, and people living in Indonesian cities having a dietary pattern which both increases NCD risk (increased consumption of sugary drinks and ultra-processed food) and protects against NCDs (increased consumption of vegetables and fish). 44, 45
There are some limitations to be addressed. Our study focused on only two of Indonesia’s provinces, which limits generalisability. However, the two provinces selected differ substantially and were chosen for their diversity of setting (urban, rural, peri-urban), population density, adolescent population, and Human Development Index. The cross-sectional nature of this study means that we are unable to address causation, however this study adds to the growing body of research that shows an increasing burden of NCD risk in Indonesian adolescents and to the understanding about the relationship between these risk factors and current wellbeing. The association observed between co-occurring NCD risks and wellbeing, and quality of life is likely bi-directional, and further research is needed to better understand the underlying mechanisms. However, this association indicates that addressing NCD risks now is important for young peoples’ current health and wellbeing and not only for future adult health.
We found evidence of a growing number of young people in Indonesia with co-occurring NCD risks. This trend is consistent with earlier studies that have suggested an increasing prevalence of co-occurring risks over time. 46 The prevalence of insufficient physical activity using the WHO guideline for adolescents (60 min of daily physical activity of moderate-to-vigorous intensity) was very high (96.1%) in our sample. However, this was consistent with other studies using the same definition of physical inactivity. A 2019 study reported overall prevalence of insufficient physical activity as 86.4% overall in 11-17-year-olds in Indonesia, and a study from 2015 Indonesian Global School Health Survey in 13-16-year-olds found a prevalence of 87.8%.41, 46 While our estimate was higher by 7–8 percentage points, we had an older age group (a known risk for decreasing activity), and our sample was only drawn from two provinces, which could explain some of the elevated risk. Similarly, we had very high levels of inadequate fruit and vegetable consumption within our sample, 91.1%. However, analysis of Indonesia’s RISKESDAS 2018 survey data shows similar levels (over 95%) in 15 to 19-year-olds and 18 to 24-year-olds. 47, 48 Our findings here, and the similarity with other published estimates, indicates that these risks are almost universal. While these are well established risk definitions their highly prevalent nature calls into question their practical utility in identifying those at greater risk than the general population.
The mixed method data collection across a large, representative sample, is a strength of this study. We report on self-report survey items, anthropometrics, and biomarkers for this important population. Many of the other studies with a similar focus are largely reliant on self-report survey data only, and do not have objective measures of height, weight, waist circumference, blood pressure, or blood lipid profiles. Our outcome measures for mental wellbeing and quality of life are measured with sound, validated scales, and our earlier work in culturally verifying and validating the K10 within a subsample of diagnostic interviews (MINI-KID) enables reliable estimates. 18 Another strength of this study is the grounding in qualitative enquiry. Early qualitative work influenced much of the original survey design, ensuring culturally relevant and informed data collection. 17, 49 The analysis presented here was informed by more recent qualitative interviews with stakeholders across Indonesia, working in NCD prevention and adolescent health, and an in-depth understanding of the data needs of those stakeholders. 50