Mental Distress and Subjective Quality of Life Among Adolescents and Young Adults Living in Deprived Urban Areas in South America – A Cross-sectional Study

Abstract

Background

There has been extensive research on mental distress and quality of life, and various studies have shown an association between high distress and poorer subjective quality of life (SQoL). However, hardly any research has addressed these issues in young people in deprived urban areas who are exposed to a range of risk factors for experiencing mental distress.

Objective

The study aimed to assess the SQoL and symptoms of anxiety and depression among adolescents and young adults living in deprived urban areas in South America and explore the link between SQoL and symptoms.

Methods

In a cross-sectional study, we recruited participants aged 15/16 and 20–24 years old in deprived areas in Bogotá, Buenos Aires and Lima. Participants rated symptoms of depression and anxiety on the Patient Health Questionnaire-8 and General Anxiety Disorder-7 respectively and SQoL on the Manchester Short Assessment of Quality of Life.

Results

We recruited 2396 participants (1077 adolescents and 1319 young adults), with on average moderate symptom levels of depression and anxiety. Variations of satisfaction with different life domains were largely similar in the two age groups. Overall, participants were explicitly dissatisfied with their personal safety and financial situation, and most satisfied with the people they lived with, their sex life, family relations and friendships. Higher scores on both depression and anxiety scales showed moderate associations (from r = −0.31 to r = −0.52) with poorer SQoL in both age groups.

Conclusion

The SQoL of young people with moderate symptom levels of depression and anxiety and living in deprived urban areas in South America reflects the adverse environment they live in and points to positive resources in social relationships reducing mental distress and improving social conditions may lead to an improvement of quality of life in this population.

Resumen

Antecedentes

Se han realizado extensas investigaciones sobre la angustia mental y la calidad de vida, y varios estudios han demostrado una asociación entre una gran angustia y una peor calidad de vida subjetiva (SQoL). Sin embargo, casi ninguna investigación ha abordado estos problemas en jóvenes de zonas urbanas desfavorecidas que están expuestos a una variedad de factores de riesgo, de sufrir angustia mental.

Objetivo

El estudio tuvo como objetivo evaluar la SQoL y los síntomas de ansiedad y depresión entre los adolescentes y los adultos jóvenes que viven en áreas urbanas desfavorecidas en América del Sur, y explorar el vínculo entre la SQoL y los síntomas.

Métodos

En un estudio transversal, reclutamos participantes de 15/16 y 20/24 años en zonas desfavorecidas de Bogotá, Buenos Aires y Lima. Los participantes calificaron los síntomas de depresión y ansiedad en el Cuestionario de salud del paciente-8 y el Trastorno de ansiedad general-7, respectivamente y la SQoL en la evaluación corta de calidad de vida de Manchester.

Resultados

Reclutamos a 2.396 participantes (1.077 adolescentes y 1.319 adultos jóvenes), con niveles de síntomas de depresión y ansiedad moderados en promedio. Las variaciones de satisfacción con diferentes ámbitos de la vida fueron muy similares en los 2 grupos de edad. En general, los participantes estaban explícitamente insatisfechos con su seguridad personal y su situación financiera, y más satisfechos con las personas con las que vivían, su vida sexual, sus relaciones familiares y sus amistades. Las puntuaciones más altas en las escalas de depresión y ansiedad mostraron asociaciones moderadas (r = −0,31 a r = −0,52) con una peor SQoL en ambos grupos de edad.

Conclusión

La SQoL de jóvenes con niveles moderados de síntomas de depresión y ansiedad, y que viven en áreas urbanas desfavorecidas en América del Sur, refleja el entorno adverso en el que viven, y apunta a recursos positivos en las relaciones sociales. Reducir la angustia mental y mejorar las condiciones sociales puede conducir a una mejora de la calidad de vida en esta población.

Introduction

Countries across Latin America are experiencing rapid and marked urbanization. It is estimated that about 55% of the population worldwide reside in urban areas, a figure projected to increase to 68% by 2050.¹ In Latin America, already approximately 80% of the population live in large urban areas, making it the most urbanized region in the world.² In practically all major cities, there are large parts with relative deprivation. In those deprived areas, there are a range of risk factors for mental distress, such as poverty, gang and domestic violence, easy availability of illegal drugs, limited educational options, and shortage of regular employment. The same factors can also be assumed to affect quality of life.^3–5

While the serious risk factors in deprived areas are likely to have an influence on both, mental distress and quality of life, there is the question as to whether the two are correlated as it has been found in other groups and settings. Several studies have shown a link between higher symptom levels and poorer quality of life. These studies however were conducted in different contexts and usually in clinical samples.^6,7

Against this background, we aimed to study the subjective quality of life (SQoL) and levels of mental distress among young people living in deprived neighbourhoods in large cities in South America. We operationalised mental distress as symptoms of anxiety and depression. In our focus on young people, we followed the concept of ‘youth’ by the United Nations that defines an age range of 15–24 years, and studied two distinct groups within that range, i.e. adolescents of 15–16 years of age and young adults of 20–24 years of age.

In these groups, we assessed symptoms of anxiety and depression, investigated SQoL in a range of different life domains and explored how mental distress and SQoL are associated. The study was conducted as multi-site research in deprived areas of three large cities across the continent, i.e. in Bogotá, Buenos Aires and Lima.

Methods

Study design

This is a cross-sectional exploratory study and part of a wider research programme “Building resilience and resources to reduce depression and anxiety in young people from urban neighbourhoods in Latin America (OLA)”. The programme is funded by the Medical Research Council (MR/S03580X/1) in the United Kingdom and carried out in partnership with universities in Argentina, Colombia, Peru and the United Kingdom. The overall aim is to identify the factors, resources, and activities that contribute to the prevention and recovery of depression and anxiety among young people in low-income urban settings in Latin America (for details see the protocol Priebe et al.).⁸

Participants and procedure

In each of the three cities, the 50% most deprived areas were identified using nationally available indices,⁸ i.e. the Human Development Index of the United Nations Development Programme in Bogotá and Lima⁹ and the Unmet Basic Needs Index in Buenos Aires.¹⁰ In the identified deprived areas, we recruited participants who were either 15–16 (adolescents) or 20–24 years old (young adults). All participants, or the parent or legal guardian in the case of adolescents, provided written informed consent. In the latter cases, the approach to the parent or legal guardian followed the assent by the adolescent participant.

Participants were excluded if: they had been diagnosed with a serious mental disorder (psychosis, bipolar disorder, schizophrenia) or a learning disability; they were illiterate; or lacked capacity to provide informed consent for young adults and assent for adolescents.

Participants were recruited and assessed between April 2021 and November 2022. During the first 10–15 months of the 19-month period of data collection some social restrictions were in place because of the COVID-19 pandemic. The type and extent of the restrictions varied substantially both between cities and within each city over time. Therefore, recruitment strategies varied across the three cities, reflecting different regulations and practical options. In Bogotá, adolescents were recruited through 13 schools and young adults through 14 educational centres. In Buenos Aires, recruitment was conducted through collaboration with community organisations, an online initiative using the official communication channels of the University of Buenos Aires. In Lima, 13 schools, five non-governmental organisations, two government programmes, two universities and one technical education centre supported recruitment. In addition, participants – mostly adolescents – were recruited through social media adverts.

The informed assent and consent process was conducted either in person, by telephone with an audio recording, by sending a photo or scan, or via an online form. All participants received a reimbursement for their participation, equivalent to $10 US dollars.

Each participant filled out a battery of questionnaires, either on paper or online. It should be added that the data were managed through Redcap, entered directly by participants or recorded by the research team depending on the social restrictions in place at the given place and time.

In total, out of 2402 recruited participants, six had more than two missing items on the MANSA and were therefore excluded. No participant had more than one missing item on the GAD-7 or PHQ-8.

Thus, the total sample for this study included 2396 participants (gender: 1556 female; 813 male; 27 missing or other), 1077 adolescents (gender: 701 female; 358 male; 18 missing or other) and 1319 young adults (gender: 855 female; 455 male; 9 missing or other).

Measures

To evaluate the sociodemographic variables, a section was included in the “ad hoc” questionnaire asking about gender and age.

Participants self-rated the levels of depression symptoms on the Patient Health Questionnaire-8 (PHQ-8) and of anxiety symptoms on the General Anxiety Disorder-7 (GAD-7).

The PHQ-8 is an eight-item questionnaire that assesses various symptoms of depression, such as sadness, loss of pleasure or interest, difficulty concentrating, fatigue, and changes in sleep and/or appetite. Participants are asked to indicate, on a self-rating scale, how often they have experienced each symptom in the last fifteen days. The scoring scale includes four options, ranging from zero to three: zero represents “No days,” one “Several days (1–6 days),” two “More than half of the days (7–11 days).” and three “Almost every day (12 or more days).” This instrument is widely used in the scientific field due to its ease of application and its reliability in detecting depression.¹¹

The GAD-7 is a questionnaire consisting of seven questions about common symptoms of anxiety, such as worrying excessively, feeling nervous, having difficulty relaxing, and experiencing restlessness or tension. In a self-administered questionnaire, participants are asked to rate how often they have experienced these symptoms in the past 15 days. The scoring scale offers four options, ranging from zero to three: zero indicates “No days”, one “Several days (1–6 days)”, two “More than half of the days (7–11 days)” and three “Almost every day (12 or more days).” This instrument is widely used in scientific research due to its ease of application and its reliability in detecting anxiety.¹²

SQoL was assessed on the Manchester Short Assessment of Quality of Life (MANSA).¹³ This measure includes 12 items that address satisfaction in various aspects of life, such as life in general, employment or unemployment, financial situation, social relationships, sexual life, leisure activities, accommodation, state of health life, safety, family, physical health and mental health. Participants rated their satisfaction on each item using a 7-point scale, ranging from 1, “couldn’t be worse,” to 7, “couldn’t be better.” with 4 being a “neutral middle point”.

The mean score of the 12 items reflects the level of SQoL. A difference of the mean score of 0.3 scale points can reflect a one point difference in at least three different life domains (out of the 12 domains) or a difference of three points in one domain, e.g. from the ‘neutral middle point’ to ‘couldn’t be better’, or a difference of one point in one domain and of two points in another. Such a difference may be regarded as practically relevant.

We also calculated two dimensions of SQoL as identified by Petkari et al.¹⁴ One dimension is called ‘Life and Health’ and includes satisfaction with life as a whole, employment status (work or unemployment), financial situation, social relationships, friendships and the family, sex life, leisure activities, as well as physical and mental health. The other dimension is ‘Living Environment’ and includes satisfaction with accommodation, living conditions, personal safety in the area where one lives.

All three scales are well established and have been widely used in mental health research.

Data analysis

Descriptive and inferential statistics were used for data analysis. For descriptive analyses, frequencies, means and standard deviations were reported. In the case of the MANSA, an item-by-item description was made with the objective of comparing each quality of life indicator between the different age groups. For the inferential analysis, Pearson's correlation analysis was conducted, Cohen's (1988) criterion is followed for the interpretation of the effect size of the correlations: .10 = Small, .30 = Moderate, and .50 = Large.¹⁵ On PHQ-8 and GAD-7, one missing item was mean substituted. On MANSA up to two missing items were substituted in line with instructions for the scale. Participants with more missing items were excluded.

The treatment and analysis of the data was computed using R (2020). The gtsummary package¹⁶ was used for the presentation of the descriptive results. The correlation matrix was calculated with the corrr packet.¹⁷

Opinion of the ethics committees

The study was prospectively registered in the ISRCTN Registry (ISRCTN99961401) and positive ethics opinions were obtained from the Biomedical Research Ethics Committee of the Faculty of Medicine of the University of Buenos Aires; The Faculty of Medicine-Committee of Research and Institutional Ethics and Research of the Pontificia Universidad Javeriana, Bogotá; Institutional Research Ethics Committee of the Universidad Peruana Cayetano Heredia; and Queen Mary Research Ethics Committee.

Results

Mental distress

There were two participants with one missing item on the GAD-7, none with missing items on the PHQ-8, and 48 with one or two missing items on the MANSA so that their missing items were substituted. The mean score on the GAD-7 was 8.51 (SD = 4.92) and on the PHQ-8 10.00 (SD = 5.97). The respective scores were 8.45 (SD = 4.92) and 10.37 (SD = 5.92) for the adolescents, and 8.55 (SD = 4.92) and 9.70 (SD = 6.00) for the young adults. These average symptom levels are regarded as being of borderline clinical significance.

Quality of life

The mean scores of the MANSA overall and of the two dimensions of SQoL are shown for the total sample and the two age groups in Table 1.

Table 1.

Mean scores and standard deviation of MANSA and its two dimensions for the total sample and the two age groups.

	Total samplen = 2396	15–16 years oldn = 1077	20–24 years oldn = 1319
MANSA	4.65 (1.08)	4.77 (1.10)	4.56 (1.05)
Dimension: Life and Health	4.58 (1.16)	4.71 (1.16)	4.47 (1.14)
Dimension: Living Environment	4.81 (1.18)	4.89 (1.20)	4.74 (1.16)

Table 2 shows the satisfaction scores with each life domain addressed in the MANSA for the total sample and the two age groups.

Table 2.

Mean satisfaction scores and standard deviation (SD) with each life domain addressed in the MANSA for the total sample and the two age groups.

	Total samplen = 2396		15-16 years oldn = 1077		20-24 years oldn = 1319
	M	SD	M	SD	M	SD
Satisfaction
Life as a whole	4.57	1.58	4.59	1.59	4.55	1.57
Work/study	4.57	1.69	4.71	1.58	4.45	1.77
Financial situation	3.88	1.71	4.40	1.65	3.45	1.65
Friendships	5.09	1.71	5.15	1.68	5.05	1.73
Leisure activities	4.62	1.66	4.73	1.73	4.53	1.60
Sex life	5.24	1.59	5.35	1.57	5.16	1.60
Physical health	4.44	1.79	4.55	1.81	4.35	1.77
Mental health	4.22	1.88	4.22	1.92	4.21	1.84
Accommodation	5.04	1.67	5.22	1.60	4.89	1.72
Personal safety	3.74	1.73	4.00	1.69	3.53	1.73
People one lives with	5.35	1.52	5.33	1.52	5.37	1.52
Family relationships	5.09	1.57	5.00	1.62	5.17	1.54

Note: Participants rated their satisfaction on each item using a 7-point scale, ranging from 1, “couldn’t be worse,” to 7, “couldn’t be better.” with 4 being a “neutral middle point”.

For the total sample, two scores are below the neutral middle point of 4 indicating explicit dissatisfaction. Participants are dissatisfied with their financial situation and even more so with their personal safety. On the other end of the spectrum, they are particularly satisfied with the people they live with and their sex life. Positive mean scores above 5 are also found for satisfaction with friendships, family relationships and their accommodation.

The age groups differ in their satisfaction with some life domains with a mean difference of more than 0.3 scale points in three life domains: young adults are less satisfied with their financial situation, their accommodation and their personal safety. Across the other life domains the satisfaction scores in the two age groups show a similar pattern.

Correlation between symptom levels and SQoL

The correlations between levels of symptoms of anxiety and depression (GAD-7 and PHQ-8) and the mean SQoL score (MANSA) as well as the MANSA dimensions ‘Life and Health’ and ‘Environment’ are presented in Table 3.

Table 3.

Pearson correlations between symptoms of anxiety, depression and SQoL scores.

	Total sample		15–16 years old		20–24 years old
	GAD-7	PHQ-8	GAD-7	PHQ-8	GAD-7	PHQ-8
Global SQoL score	−.44***	−.50***	−.49***	−.52***	−.41***	−.51***
Dimension ‘Life and Health’	−.45***	−.52***	−.50***	−.53***	−.41***	−.53***
Dimension ‘Living Environment’	−.34***	−.35***	−.39***	−.39***	−.31***	−.34***

***

p < .001.

All correlations indicate associations of higher symptom scores with poorer SQoL. They are all in the range between r = −0.31 and r = −0.50, most of them within the range of moderate correlations. The associations of symptoms with the dimension ‘Living Environment’ are slightly weaker than with the global SQoL score or the dimension ‘Life and Health’. All correlations are statistically significant on a level of p < 0.001.

Discussion

In the participants of our study living in deprived urban areas in South America we found on average moderate levels of anxiety and depression symptoms. Adolescents and young adults show similar satisfaction patterns across most life domains, but young adults are less satisfied with their financial situation, accommodation and personal safety. Overall, young people are explicitly dissatisfied with their personal safety and with their financial situation while they express a higher level of satisfaction with the people they live with, with their sex life, with friendships and with family relationships. Higher symptoms of anxiety and depression are associated with poorer SQoL, and the correlations are mostly moderate for both types of symptoms and both age groups.

Improving the mental health of young people and their quality of life are a major challenge to societies, and this applies in particular to disadvantaged young people living in deprived urban areas.^18,19 Public health initiatives should be informed by research evidence about the levels of mental distress, quality of life and their association in young people living in deprived urban areas on the continent. Nevertheless, such research is widely lacking. This may reflect the relative absence of research funding and infrastructure as well as the practical challenges of conducting field studies in such settings.

Strengths and limitations

The study recruited a large sample of young people in a challenging context. The study was conducted in the capital cities of three countries across South America, following a consistent protocol on all sites. It used standardised and well-established assessment methods and trained researchers. The number of missing data was minimal, and all analyses could be conducted with almost the complete sample and without the need for imputation of much data.

However, the study also has some limitations. Most importantly, convenience sampling was used to recruit participants and so the study sample is not necessarily representative. This makes absolute frequencies and means more difficult to interpret. However, the associations between symptoms and SQoL should be less affected, as such associations are usually more robust against selection biases. Also, we assessed symptoms at one point in time and did not consider the duration and potential variability of such symptoms over time.

Our study addresses a population that differs significantly from the samples used in most psychological studies, which tend to focus on Western, educated, industrialized, wealthy, and democratic populations.²⁰ The sample of this study does not belong to these categories, since, although it corresponds to Western populations, its members are not part of highly educated societies nor do they live in cities of industrialized or rich countries. Furthermore, the countries in question do not stand out for having consolidated democracies. These differences are fundamental when analyzing the effects they may have on the variables evaluated in this study, since the social, economic and cultural context can significantly influence the results obtained.

Comparisons with the literature

Although the literature on quality of life in people with mental distress is extensive, most research has been conducted in Western countries and in clinical samples rather than in population-based studies. The sample here was population based and lived in deprived areas of three South American capitals. Their average symptom levels of anxiety and even more so of depression are of borderline clinical significance,^13,14 and the young people showed relatively low satisfaction with their mental health.

The pattern of satisfaction with life domains is similar for adolescents and young adults, although young adults are in comparison less satisfied with their finances, accommodation and personal safety. These differences may reflect that young adults either live independently or still have to live with their families because finding or affording their own accommodation is difficult, that they tend to move around more independently in the area and are expected to generate their own income.⁵

However, the sample as a whole are explicitly dissatisfied with their personal safety, i.e. the average score was below the neutral middle point in the range of dissatisfaction. To our knowledge, there are no other larger scale studies in populations that are most dissatisfied with their personal safety. This unusual finding may be a direct reflection of the environment in which the participants of this study live.²¹ Deprived urban areas in South America do often have significant crime rates and gang activities, and are not considered ‘safe’. Since most of the participants are not in a position to leave the area and move to live elsewhere, they may feel relatively helpless to avoid these risks, and concerns about their safety can dominate their experience.

On the other end of the spectrum of life domains, young people feel delighted with the people they live with, their sex life, their friends and their family. The higher satisfaction with sex life contrasts with studies in clinical samples, some of which expressed the highest dissatisfaction exactly with their sex life.²² The high satisfaction with the people the individual lives with however has also been found in other samples in Europe, in particular in people with depression.²³ The high satisfaction with different types of close social relationships point to relevant social resources, which may be utilised in otherwise adverse circumstances.²⁴ Young people seem to feel comfortable and supported in the closer community they live in and appreciate these relationships. One can only speculate as to whether a hostile external environment fosters stronger and more helpful relationships in people who live together. Interventions to help particularly distressed young people may try to tap into and utilise these existing sources of support in close relationships.

Various studies have shown an association of higher levels of symptoms, in particular of depression, with lower SQoL. Usually, such higher symptom levels are considered in the literature as influential factors impacting on people's SQoL.^6,7,25 Yet, the results show only correlations without any evidence of the direction of influence. It appears plausible that lower mood and anxiety can negatively affect quality of life and make people less satisfied with different domains of their life. At the same time, the influence may also be the other way round, and a poorer quality of life may lead to lower mood and more anxiety.

The correlations between symptom levels and SQoL in this study are overall moderate in size. They are somewhat higher than those found in most clinical samples.^6,23 One reason for this may be that the variability of symptoms tends to be higher in a population-based sample than in clinical samples, and the higher variability can facilitate higher correlations. Also, the possibly mutual influence of mental distress and SQoL might be stronger in young people in deprived urban areas than in clinical samples. In any case, both anxiety and depression are similarly associated with SQoL, and they are linked with global SQoL as well as with both dimensions of it. Thus, the findings of this study suggest that any influence is rather general than specific and that it applies similarly to different types of symptoms, different dimensions of SQoL and different age groups.

All correlations indicate associations of higher symptom scores with poorer SQoL indicating that the correlation of SQoL with symptom levels is not specific for single life domains. Such associations of symptoms with rather global tendencies to appraise one's life more or less positively have been described in the literature before.²⁵ The correlations are all in the range between r = −0.31 and r = −0.50, most of them within the range of moderate correlations. The associations of symptoms with the dimension ‘Living Environment’ are slightly weaker than with the global SQoL score and with the dimension ‘Life and Health’. All correlations are statistically significant on a level of p < 0.001.

Conclusions

Young people's mental health and quality of life are closely related, and reducing psychological distress can contribute to improving their SQoL. Future research should explore the extent to which the mental distress and quality of life of young people living in deprived urban neighbourhoods on different continents are similar or different to the findings of this study that was conducted with participants from three cities in South America. Furthermore, it should be explored how social and economic factors such as insecurity, housing conditions and family support can influence these aspects. Finally, social and health policies and specific interventions need to be developed and tested that can improve both the mental health and the quality of life of young people who live in adverse conditions in large cities, where poverty and the lack of safety networks are major challenges.

Funding

This study was funded by the Medical Research Council (MR/S03580X/1).

Conflicts of interest

The authors have no conflicts of interest to declare.

Acknowledgements

The authors would like to thank all participants of the study for sharing their data, experiences and views with us, and all researchers on the OLA programme in Bogotá, Buenos Aires, Lima and London for their collaboration and support.