Pub Date : 2025-10-08DOI: 10.1016/j.ssmph.2025.101866
Amanda Sursely , John R. Pamplin II , Megan E. Gilster , Carol Coohey , Katherine M. Keyes , Jonathan M. Platt
Objectives
To identify spatiotemporal trends in suicide clusters across multiple intersecting sociodemographic groups of US adults from 2005 to 2020.
Methods
US Vital Statistics data were used to calculate county-level suicide rates in adults from 2005 to 2020. Rates were calculated for 16 intersecting sociodemographic groups, defined by mutually exclusive combinations of race (American Indian/Alaskan Native (AIAN), Asian/Pacific Islander (API), Black, White), ethnicity (Hispanic, non-Hispanic), and sex (male, female). For each group, spatiotemporal suicide clusters were identified using the SaTScan space-time statistic, based on counties where suicide rates were higher (hot clusters) or lower (cold clusters) than expected. We also calculated the average proportion of deaths contained in clusters, to quantify the importance of clusters relative to total suicide rates.
Results
Overall, 275 clusters were identified, comprising 11.4 % of total suicides. Clusters ranged from 1 among non-Hispanic Black women to 135 among non-Hispanic White men. The number and magnitude of hot clusters (RR > 1) increased over time among Hispanic White men and women, and the highest magnitude clusters were observed in non-Hispanic Native American women. No clusters were identified for Hispanic AIAN women or Hispanic API men. Hot clusters were most prevalent in Western counties across all groups. Cold clusters were most prevalent in urban counties.
Conclusions
Suicide rates and spatial clusters have increased since 2005, with distinct patterning between sociodemographic groups. The identification of specific communities that may be at disproportional risk for suicide highlight opportunities to tailor individual- and population-level prevention efforts in emerging high-risk groups.
{"title":"Differences across intersecting racial, ethnic, and gender identity in the spatiotemporal patterning of suicide clusters in the US from 2005-2020","authors":"Amanda Sursely , John R. Pamplin II , Megan E. Gilster , Carol Coohey , Katherine M. Keyes , Jonathan M. Platt","doi":"10.1016/j.ssmph.2025.101866","DOIUrl":"10.1016/j.ssmph.2025.101866","url":null,"abstract":"<div><h3>Objectives</h3><div>To identify spatiotemporal trends in suicide clusters across multiple intersecting sociodemographic groups of US adults from 2005 to 2020.</div></div><div><h3>Methods</h3><div>US Vital Statistics data were used to calculate county-level suicide rates in adults from 2005 to 2020. Rates were calculated for 16 intersecting sociodemographic groups, defined by mutually exclusive combinations of race (American Indian/Alaskan Native (AIAN), Asian/Pacific Islander (API), Black, White), ethnicity (Hispanic, non-Hispanic), and sex (male, female). For each group, spatiotemporal suicide clusters were identified using the SaTScan space-time statistic, based on counties where suicide rates were higher (hot clusters) or lower (cold clusters) than expected. We also calculated the average proportion of deaths contained in clusters, to quantify the importance of clusters relative to total suicide rates.</div></div><div><h3>Results</h3><div>Overall, 275 clusters were identified, comprising 11.4 % of total suicides. Clusters ranged from 1 among non-Hispanic Black women to 135 among non-Hispanic White men. The number and magnitude of hot clusters (RR > 1) increased over time among Hispanic White men and women, and the highest magnitude clusters were observed in non-Hispanic Native American women. No clusters were identified for Hispanic AIAN women or Hispanic API men. Hot clusters were most prevalent in Western counties across all groups. Cold clusters were most prevalent in urban counties.</div></div><div><h3>Conclusions</h3><div>Suicide rates and spatial clusters have increased since 2005, with distinct patterning between sociodemographic groups. The identification of specific communities that may be at disproportional risk for suicide highlight opportunities to tailor individual- and population-level prevention efforts in emerging high-risk groups.</div></div>","PeriodicalId":47780,"journal":{"name":"Ssm-Population Health","volume":"32 ","pages":"Article 101866"},"PeriodicalIF":3.1,"publicationDate":"2025-10-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145267635","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-06DOI: 10.1016/j.ssmph.2025.101869
Daniel Nettle , Coralie Chevallier , Kate E. Pickett , Matthew T. Johnson , Elliott A. Johnson , Melissa Bateson
There is some evidence that income volatility (fluctuations in income over time) negatively affects mental and physical health, independently of the level of income. Evidence to date has examined fluctuations from year to year or from day to day, whereas a more relevant timescale might be month to month. Here, we use data from the Changing Cost of Living Study, a longitudinal panel from the UK and France with monthly data (n = 484). We examine the association between month-to-month income volatility and two outcomes, self-rated general health and anxiety-depression (a composite measure derived from GAD-7 and PHQ-8 scores). Higher volatility was associated with worse health on both measures, with volatility accounting for similar amounts of variation as the level of income. Some association between income volatility and health is to be expected as a consequence of the concavity of the income-health relationship: because of concavity, a downward fluctuation damages health more than the equivalent upward fluctuation improves it. We show that the observed associations are 3 and 4 times too strong to be explained by this mechanism alone. We suggest that volatility, because it introduces uncertainty and stress, has substantial direct health effects. This claim is important for public policy: it means that policies and institutions that smooth people's income streams can have beneficial health effects even if they don't raise anyone's income.
{"title":"Why is income volatility associated with poor health? Longitudinal evidence from the UK and France","authors":"Daniel Nettle , Coralie Chevallier , Kate E. Pickett , Matthew T. Johnson , Elliott A. Johnson , Melissa Bateson","doi":"10.1016/j.ssmph.2025.101869","DOIUrl":"10.1016/j.ssmph.2025.101869","url":null,"abstract":"<div><div>There is some evidence that income volatility (fluctuations in income over time) negatively affects mental and physical health, independently of the level of income. Evidence to date has examined fluctuations from year to year or from day to day, whereas a more relevant timescale might be month to month. Here, we use data from the Changing Cost of Living Study, a longitudinal panel from the UK and France with monthly data (n = 484). We examine the association between month-to-month income volatility and two outcomes, self-rated general health and anxiety-depression (a composite measure derived from GAD-7 and PHQ-8 scores). Higher volatility was associated with worse health on both measures, with volatility accounting for similar amounts of variation as the level of income. Some association between income volatility and health is to be expected as a consequence of the concavity of the income-health relationship: because of concavity, a downward fluctuation damages health more than the equivalent upward fluctuation improves it. We show that the observed associations are 3 and 4 times too strong to be explained by this mechanism alone. We suggest that volatility, because it introduces uncertainty and stress, has substantial direct health effects. This claim is important for public policy: it means that policies and institutions that smooth people's income streams can have beneficial health effects even if they don't raise anyone's income.</div></div>","PeriodicalId":47780,"journal":{"name":"Ssm-Population Health","volume":"32 ","pages":"Article 101869"},"PeriodicalIF":3.1,"publicationDate":"2025-10-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145267634","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-03DOI: 10.1016/j.ssmph.2025.101867
Xuemei Zhang , Tianxin Pan , Barbara McPake
Background
The term “reversal hypothesis” has been attached to the observation that the socio-economic status (SES) gradient of non-communicable diseases (NCDs) tends to reverse over time as country's economic and social development progresses. We examined this hypothesis in China between 1991 and 2020, to test whether a divergence in the experiences of men and women can be identified and to explore the possible contributors to any disparity.
Methods
Using nine waves of China Health and Nutrition Survey (CHNS, 1991–2015) and six waves of the China Family Panel Studies (CFPS, 2010–2020), we investigated the SES-NCD gradient among participants aged 18 years and older. We examined the relationship between self-reported doctor diagnosed NCDs and SES proxied by formal educational years by gender using multivariable logistic regression and Blinder-Oaxaca decomposition models.
Findings
Prevalence of a list of NCDs in the overall population increased from 3.52 % (1991) to 13.72 % (2020). The association between SES and NCDs has shifted from positive to negative over time, shifting earlier among women compared to men. For both men and women, the difference in prevalence between high and low SES groups was mostly explained by the socio-demographic and risk behaviour variables included in the models: age, being overweight/obesity, employment status, residency, and region. However, the contribution size of these factors and consistency was different by gender and changed over time.
Conclusions
China has experienced a reversal in the SES-NCD gradient which occurred first among women. The explanations of SES differences are complex and differ by gender.
{"title":"Examining the reversal hypothesis on socioeconomic status and noncommunicable diseases, and gender disparities in China: a nationwide longitudinal analysis between 1991 and 2020","authors":"Xuemei Zhang , Tianxin Pan , Barbara McPake","doi":"10.1016/j.ssmph.2025.101867","DOIUrl":"10.1016/j.ssmph.2025.101867","url":null,"abstract":"<div><h3>Background</h3><div>The term “reversal hypothesis” has been attached to the observation that the socio-economic status (SES) gradient of non-communicable diseases (NCDs) tends to reverse over time as country's economic and social development progresses. We examined this hypothesis in China between 1991 and 2020, to test whether a divergence in the experiences of men and women can be identified and to explore the possible contributors to any disparity.</div></div><div><h3>Methods</h3><div>Using nine waves of China Health and Nutrition Survey (CHNS, 1991–2015) and six waves of the China Family Panel Studies (CFPS, 2010–2020), we investigated the SES-NCD gradient among participants aged 18 years and older. We examined the relationship between self-reported doctor diagnosed NCDs and SES proxied by formal educational years by gender using multivariable logistic regression and Blinder-Oaxaca decomposition models.</div></div><div><h3>Findings</h3><div>Prevalence of a list of NCDs in the overall population increased from 3.52 % (1991) to 13.72 % (2020). The association between SES and NCDs has shifted from positive to negative over time, shifting earlier among women compared to men. For both men and women, the difference in prevalence between high and low SES groups was mostly explained by the socio-demographic and risk behaviour variables included in the models: age, being overweight/obesity, employment status, residency, and region. However, the contribution size of these factors and consistency was different by gender and changed over time.</div></div><div><h3>Conclusions</h3><div>China has experienced a reversal in the SES-NCD gradient which occurred first among women. The explanations of SES differences are complex and differ by gender.</div></div>","PeriodicalId":47780,"journal":{"name":"Ssm-Population Health","volume":"32 ","pages":"Article 101867"},"PeriodicalIF":3.1,"publicationDate":"2025-10-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145416903","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-02DOI: 10.1016/j.ssmph.2025.101868
Najma Abdi , Adam Szpiro , Stephen J. Mooney , Quynh Nguyen , Jana A. Hirsch , Brian E. Saelens
Introduction
Gentrification may displace lower-income populations, particularly in neighborhoods where the majority of residents identify as Black, Indigenous, and/or People of Color (BIPOC). While the health impacts of gentrification and displacement have been explored, little is known about how pedestrian infrastructure (e.g. sidewalks) may be associated with gentrification. This study assesses whether sidewalk availability is associated with gentrification and changes in BIPOC population, with attention to differences between urban and suburban/small-town contexts where patterns of demographic change may diverge, including potential change in BIPOC population by context.
Methods
Utilizing the Logitudinal Tract Database and U.S. national sidewalk availability, we conducted a longitudinal observational study of 26,498 gentrifiable census tracts limited to urban/suburban/smal town areas across the US in 2010–2020. Sidewalk availability was defined as the proportion of Google Street images within a given census tract that have at least one sidewalk. We used logistic regression to examine the association between sidewalk availability and gentrification and linear regression to assess changes in BIPOC residential composition between 2010 and 2020 in gentrified neighborhoods. Sensitivity and supplementary analysis was conducted for a stratified models by BIPOC thresholds (40 %, 50 %, 60 %) and urban/suburban outcomes.
Results
Sidewalk availability was significantly associated with higher odds of gentrification in all tracts (OR: 1.119; 95 % CI: 1.032–1.212; p = 0.0065) and in BIPOC-majority tracts (OR: 1.216; 95 % CI: 1.088–1.359; p < 0.001). In gentrified neighborhoods, sidewalk availability was associated with a reduction in the percentage of BIPOC residents (−0.016 per 0.1-unit; 95 % CI: −0.027, −0.006; p < 0.002) and a significant decline in the absolute number of BIPOC individuals (−81.2; 95 % CI: −114.78, −47.69; p < 0.001). Supplementary analyses indicated divergence by context: in urban gentrified tracts, each 0.1-unit higher sidewalk availability was associated with −0.020 percentage points in BIPOC share (95 % CI –126.5, −57.2; p < 0.001) and −91.86 BIPOC residents (95 % CI −126.51, −57.22; p < 0.001); in suburban/small-town gentrified tracts, it was associated with +0.036 percentage points (95 % CI: 0.001, 0.071; p = 0.042) and +165 BIPOC residents (95 % CI 45.1, 285.8; p = 0.007).
Conclusion
Our results demonstrate a consistent relationship between sidewalk availability and gentrification, although different associations with BIPOC population change by context, contributing to the ongoing discourse on gentrification, urban development, and neighborhood change.
中产阶级化可能会取代低收入人口,特别是在大多数居民认为是黑人、土著和/或有色人种(BIPOC)的社区。虽然已经探索了高档化和流离失所对健康的影响,但对行人基础设施(例如人行道)如何与高档化联系在一起知之甚少。本研究评估了人行道的可用性是否与高档化和BIPOC人口的变化有关,并关注了城市和郊区/小城镇背景之间的差异,其中人口变化模式可能存在差异,包括BIPOC人口的潜在变化。方法利用美国全国人行道可用性数据库,对2010-2020年美国城市/郊区/小城镇地区的26,498个可士绅化人口普查区进行了纵向观察研究。人行道可用性定义为在给定的人口普查区内至少有一个人行道的100张街道图像的比例。我们使用逻辑回归来检验人行道可用性与高绅化之间的关系,并使用线性回归来评估2010年至2020年高绅化社区BIPOC住宅组成的变化。通过BIPOC阈值(40%、50%、60%)和城市/郊区结果对分层模型进行敏感性和补充分析。结果:人行道的可用性与高绅化的几率显著相关(OR: 1.119; 95% CI: 1.032-1.212; p = 0.0065),在bipoc占多数的地区(OR: 1.216; 95% CI: 1.088-1.359; p < 0.001)。在高绅化社区,人行道的可用性与BIPOC居民百分比的减少(- 0.016 / 0.1单位;95% CI: - 0.027, - 0.006; p < 0.002)和BIPOC个人绝对数量的显著下降(- 81.2;95% CI: - 114.78, - 47.69; p < 0.001)有关。补充分析显示了环境差异:在城市高绅化地区,每增加0.1个单位的人行道可用性,BIPOC份额就会增加- 0.020个百分点(95% CI -126.5, - 57.2; p < 0.001), BIPOC居民的比例就会增加- 91.86个百分点(95% CI - 126.51, - 57.22; p < 0.001);在郊区/小城镇中产阶级化地区,这与+0.036个百分点(95% CI: 0.001, 0.071; p = 0.042)和+165个BIPOC居民(95% CI 45.1, 285.8; p = 0.007)相关。我们的研究结果表明,人行道可用性与高档化之间存在一致的关系,尽管与BIPOC人口变化之间存在不同的关联,这为高档化、城市发展和社区变化的持续讨论做出了贡献。
{"title":"The role of sidewalk availability in gentrification: A longitudinal study of U.S. neighborhoods and racial/ethnic composition","authors":"Najma Abdi , Adam Szpiro , Stephen J. Mooney , Quynh Nguyen , Jana A. Hirsch , Brian E. Saelens","doi":"10.1016/j.ssmph.2025.101868","DOIUrl":"10.1016/j.ssmph.2025.101868","url":null,"abstract":"<div><h3>Introduction</h3><div>Gentrification may displace lower-income populations, particularly in neighborhoods where the majority of residents identify as Black, Indigenous, and/or People of Color (BIPOC). While the health impacts of gentrification and displacement have been explored, little is known about how pedestrian infrastructure (e.g. sidewalks) may be associated with gentrification. This study assesses whether sidewalk availability is associated with gentrification and changes in BIPOC population, with attention to differences between urban and suburban/small-town contexts where patterns of demographic change may diverge, including potential change in BIPOC population by context.</div></div><div><h3>Methods</h3><div>Utilizing the Logitudinal Tract Database and U.S. national sidewalk availability, we conducted a longitudinal observational study of 26,498 gentrifiable census tracts limited to urban/suburban/smal town areas across the US in 2010–2020. Sidewalk availability was defined as the proportion of Google Street images within a given census tract that have at least one sidewalk. We used logistic regression to examine the association between sidewalk availability and gentrification and linear regression to assess changes in BIPOC residential composition between 2010 and 2020 in gentrified neighborhoods. Sensitivity and supplementary analysis was conducted for a stratified models by BIPOC thresholds (40 %, 50 %, 60 %) and urban/suburban outcomes.</div></div><div><h3>Results</h3><div>Sidewalk availability was significantly associated with higher odds of gentrification in all tracts (OR: 1.119; 95 % CI: 1.032–1.212; p = 0.0065) and in BIPOC-majority tracts (OR: 1.216; 95 % CI: 1.088–1.359; p < 0.001). In gentrified neighborhoods, sidewalk availability was associated with a reduction in the percentage of BIPOC residents (−0.016 per 0.1-unit; 95 % CI: −0.027, −0.006; p < 0.002) and a significant decline in the absolute number of BIPOC individuals (−81.2; 95 % CI: −114.78, −47.69; p < 0.001). Supplementary analyses indicated divergence by context: in urban gentrified tracts, each 0.1-unit higher sidewalk availability was associated with −0.020 percentage points in BIPOC share (95 % CI –126.5, −57.2; p < 0.001) and −91.86 BIPOC residents (95 % CI −126.51, −57.22; p < 0.001); in suburban/small-town gentrified tracts, it was associated with +0.036 percentage points (95 % CI: 0.001, 0.071; p = 0.042) and +165 BIPOC residents (95 % CI 45.1, 285.8; p = 0.007).</div></div><div><h3>Conclusion</h3><div>Our results demonstrate a consistent relationship between sidewalk availability and gentrification, although different associations with BIPOC population change by context, contributing to the ongoing discourse on gentrification, urban development, and neighborhood change.</div></div>","PeriodicalId":47780,"journal":{"name":"Ssm-Population Health","volume":"32 ","pages":"Article 101868"},"PeriodicalIF":3.1,"publicationDate":"2025-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145220898","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-09-26DOI: 10.1016/j.ssmph.2025.101865
Katherine LeMasters , Sara N. Levintow , Jennifer Lao , Erin McCauley , Craig Waleed , Zaire Cullins , M. Forrest Behne , Lauren Brinkley-Rubinstein
Purpose
Solitary confinement is associated with increased mortality post-release. Little is known about its use following recent reforms and COVID-19 challenges, and its association with in-custody deaths. We investigated patterns of and associations between solitary confinement and mortality in North Carolina (NC) state prisons.
Methods
Using weekly housing data from Disability Rights NC, we created a retrospective cohort of those newly incarcerated in NC prisons between 2021 and 2023. We calculated the weekly proportion of individuals experiencing solitary confinement for 2+ weeks by type (e.g., administrative purposes). We calculated all-cause mortality rates stratified by those who did and did not experience solitary confinement.
Results
Nearly one-quarter of 41,525 individuals in NC state prisons experienced solitary confinement. By the end of follow-up, over 6.0 % of individuals were in solitary confinement per week, and there were 43 in-custody deaths. Those who never experienced solitary confinement had an all-cause mortality rate of 1.96 per 100,000 person-weeks (95 % CI: 1.32, 2.82), and those who experienced solitary confinement had an all-cause mortality rate of 4.23 per 100,000 person-weeks (95 % CI: 2.31, 7.09).
Conclusions
Solitary confinement is common in NC prisons and is associated with elevated death rates in custody. There is an urgent need to document the types and duration of solitary stays to ultimately end this punitive practice.
{"title":"The use of solitary confinement and in-custody mortality in North Carolina State Prisons, 2021–2023","authors":"Katherine LeMasters , Sara N. Levintow , Jennifer Lao , Erin McCauley , Craig Waleed , Zaire Cullins , M. Forrest Behne , Lauren Brinkley-Rubinstein","doi":"10.1016/j.ssmph.2025.101865","DOIUrl":"10.1016/j.ssmph.2025.101865","url":null,"abstract":"<div><h3>Purpose</h3><div>Solitary confinement is associated with increased mortality post-release. Little is known about its use following recent reforms and COVID-19 challenges, and its association with in-custody deaths. We investigated patterns of and associations between solitary confinement and mortality in North Carolina (NC) state prisons.</div></div><div><h3>Methods</h3><div>Using weekly housing data from Disability Rights NC, we created a retrospective cohort of those newly incarcerated in NC prisons between 2021 and 2023. We calculated the weekly proportion of individuals experiencing solitary confinement for 2+ weeks by type (e.g., administrative purposes). We calculated all-cause mortality rates stratified by those who did and did not experience solitary confinement.</div></div><div><h3>Results</h3><div>Nearly one-quarter of 41,525 individuals in NC state prisons experienced solitary confinement. By the end of follow-up, over 6.0 % of individuals were in solitary confinement per week, and there were 43 in-custody deaths. Those who never experienced solitary confinement had an all-cause mortality rate of 1.96 per 100,000 person-weeks (95 % CI: 1.32, 2.82), and those who experienced solitary confinement had an all-cause mortality rate of 4.23 per 100,000 person-weeks (95 % CI: 2.31, 7.09).</div></div><div><h3>Conclusions</h3><div>Solitary confinement is common in NC prisons and is associated with elevated death rates in custody. There is an urgent need to document the types and duration of solitary stays to ultimately end this punitive practice.</div></div>","PeriodicalId":47780,"journal":{"name":"Ssm-Population Health","volume":"32 ","pages":"Article 101865"},"PeriodicalIF":3.1,"publicationDate":"2025-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145220897","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-09-18DOI: 10.1016/j.ssmph.2025.101864
Anu Molarius , Fredrik Granström , Jennifer Ervin
Poor mental health is debilitating and of global public health concern. Unpaid labour has been recognised as an important yet neglected, and highly gendered, determinant of mental health. We examined the contribution of unpaid labour to poor mental health, in relation to other known risk factors, in employed adults. Study data is from 5462 employed Swedish women and men aged 30–49 years responding to a population survey in 2022. Independent associations between domestic work (inclusive of unpaid household and care work) and self-reported diagnosed depression and anxiety were examined by multiple logistic regression. The contributions of economic difficulties, loneliness, physical inactivity, risk consumption of alcohol and job dissatisfaction to depression and anxiety were also investigated. Gender-specific population attributable risks (PAR) were calculated to assess the contributions. In total, 10 % of the women and 6 % of the men reported depression while 12 % and 6 %, respectively, reported anxiety. Experiencing domestic work as burdensome explained 47 % (95 % CI: 31–57 %) (PAR) of the prevalence of depression in women, whereas no independent association was found in men. Loneliness, economic difficulties and physical inactivity accounted for 13–28 % of the prevalence of depression in both women and men. Burdensome domestic work was associated with anxiety in both genders and explained 22–25 % of the prevalence. Whilst longitudinal studies are needed, these results imply that unpaid labour is a highly important contributing factor to poor mental health, especially among women. Promoting a more equal division of unpaid labour has the potential to improve mental health across the working population.
{"title":"The contribution of unpaid labour to poor mental health in the Swedish working population","authors":"Anu Molarius , Fredrik Granström , Jennifer Ervin","doi":"10.1016/j.ssmph.2025.101864","DOIUrl":"10.1016/j.ssmph.2025.101864","url":null,"abstract":"<div><div>Poor mental health is debilitating and of global public health concern. Unpaid labour has been recognised as an important yet neglected, and highly gendered, determinant of mental health. We examined the contribution of unpaid labour to poor mental health, in relation to other known risk factors, in employed adults. Study data is from 5462 employed Swedish women and men aged 30–49 years responding to a population survey in 2022. Independent associations between domestic work (inclusive of unpaid household and care work) and self-reported diagnosed depression and anxiety were examined by multiple logistic regression. The contributions of economic difficulties, loneliness, physical inactivity, risk consumption of alcohol and job dissatisfaction to depression and anxiety were also investigated. Gender-specific population attributable risks (PAR) were calculated to assess the contributions. In total, 10 % of the women and 6 % of the men reported depression while 12 % and 6 %, respectively, reported anxiety. Experiencing domestic work as burdensome explained 47 % (95 % CI: 31–57 %) (PAR) of the prevalence of depression in women, whereas no independent association was found in men. Loneliness, economic difficulties and physical inactivity accounted for 13–28 % of the prevalence of depression in both women and men. Burdensome domestic work was associated with anxiety in both genders and explained 22–25 % of the prevalence. Whilst longitudinal studies are needed, these results imply that unpaid labour is a highly important contributing factor to poor mental health, especially among women. Promoting a more equal division of unpaid labour has the potential to improve mental health across the working population.</div></div>","PeriodicalId":47780,"journal":{"name":"Ssm-Population Health","volume":"32 ","pages":"Article 101864"},"PeriodicalIF":3.1,"publicationDate":"2025-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145097105","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-09-11DOI: 10.1016/j.ssmph.2025.101863
Fru Awah Wanka , Maame Esi Woode , Coretta Maame Panyin Jonah , Julian Douglas May
Stunting, which affects about one-quarter of South African children below five years, has a long-term effect on a child's development and a potential economic cost to mothers. A stunted child necessitates more attention, thereby reducing the time a mother, especially if she is the primary caregiver, has available to seize labour market opportunities. This study investigated the relationship between child stunting and maternal labour force participation (MLFP) in South Africa, using a bivariate probit regression model. Data was sourced from the 2017 South African National Income Dynamics Study (NIDS). Using maternal height and improved water supply at home as instrumental variables to control for potential endogeneity, the average marginal effect model indicates that caring for a stunted child is associated with a 20 % reduction in MLFP, on average, at a 1 % statistical significance level. In addition to mothers' intrinsic merit of protecting their children, measures to prevent childhood stunting, such as nutrition education and promotion of exclusive breastfeeding, can improve MLFP. A more flexible working arrangement could facilitate mothers' engagement in the labour force while caring for their stunted children.
{"title":"To work or not to work? Child stunting and maternal labour force participation in South Africa","authors":"Fru Awah Wanka , Maame Esi Woode , Coretta Maame Panyin Jonah , Julian Douglas May","doi":"10.1016/j.ssmph.2025.101863","DOIUrl":"10.1016/j.ssmph.2025.101863","url":null,"abstract":"<div><div>Stunting, which affects about one-quarter of South African children below five years, has a long-term effect on a child's development and a potential economic cost to mothers. A stunted child necessitates more attention, thereby reducing the time a mother, especially if she is the primary caregiver, has available to seize labour market opportunities. This study investigated the relationship between child stunting and maternal labour force participation (MLFP) in South Africa, using a bivariate probit regression model. Data was sourced from the 2017 South African National Income Dynamics Study (NIDS). Using maternal height and improved water supply at home as instrumental variables to control for potential endogeneity, the average marginal effect model indicates that caring for a stunted child is associated with a 20 % reduction in MLFP, on average, at a 1 % statistical significance level. In addition to mothers' intrinsic merit of protecting their children, measures to prevent childhood stunting, such as nutrition education and promotion of exclusive breastfeeding, can improve MLFP. A more flexible working arrangement could facilitate mothers' engagement in the labour force while caring for their stunted children.</div></div>","PeriodicalId":47780,"journal":{"name":"Ssm-Population Health","volume":"32 ","pages":"Article 101863"},"PeriodicalIF":3.1,"publicationDate":"2025-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145057300","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-09-10DOI: 10.1016/j.ssmph.2025.101861
Woo-Ri Lee , Sungyoun Chun , Kangju Son , Hyun Seung Choi , You Young An
Hearing loss is a sensory deficit prevalent in older adults. In South Korea, over 40 % of individuals aged 60 and older reportedly experience unilateral or bilateral hearing loss. An association between perceived social support and satisfaction with hearing aid use has been reported. Although most previous studies focused on the direct effects of hearing aid use, its impact on social relationships remains poorly explored. We analyzed data from the Korean Longitudinal Study of Aging between 2006 and 2022 for those aged 65 years or above. Outcome measures in this study were depressive symptoms and self-reported quality of life (QoL) scores. The independent variable in this study was hearing aid use. A regression analysis was performed using a generalized estimating equation (GEE) model. Additionally, the interaction effect between hearing aids and social relationships was examined by performing an interaction term analysis. Of the 554 participants, 177 (31.9 %) used hearing aids. GEE analysis revealed that participants who used hearing aids had a lower depressive symptoms score (β = −0.328, 95 % confidence interval (CI) = −0.625 to −0.031) and higher self-reported QoL (β = 2.193, 95 % CI = 0.060 to 4.327). The interaction term analysis demonstrated a significant interaction effect between hearing aid use and social relationships. Hearing aid use was associated with reduced depressive symptoms and enhanced QoL in older adults with hearing loss. These effects were further moderated by the strength of social relationships, underscoring the importance of both auditory and social interventions in promoting mental and emotional well-being in aging societies.
{"title":"Impact of hearing aid use on depression and quality of life in older patients with age-related hearing loss: A focus on social relationships","authors":"Woo-Ri Lee , Sungyoun Chun , Kangju Son , Hyun Seung Choi , You Young An","doi":"10.1016/j.ssmph.2025.101861","DOIUrl":"10.1016/j.ssmph.2025.101861","url":null,"abstract":"<div><div>Hearing loss is a sensory deficit prevalent in older adults. In South Korea, over 40 % of individuals aged 60 and older reportedly experience unilateral or bilateral hearing loss. An association between perceived social support and satisfaction with hearing aid use has been reported. Although most previous studies focused on the direct effects of hearing aid use, its impact on social relationships remains poorly explored. We analyzed data from the Korean Longitudinal Study of Aging between 2006 and 2022 for those aged 65 years or above. Outcome measures in this study were depressive symptoms and self-reported quality of life (QoL) scores. The independent variable in this study was hearing aid use. A regression analysis was performed using a generalized estimating equation (GEE) model. Additionally, the interaction effect between hearing aids and social relationships was examined by performing an interaction term analysis. Of the 554 participants, 177 (31.9 %) used hearing aids. GEE analysis revealed that participants who used hearing aids had a lower depressive symptoms score (β = −0.328, 95 % confidence interval (CI) = −0.625 to −0.031) and higher self-reported QoL (β = 2.193, 95 % CI = 0.060 to 4.327). The interaction term analysis demonstrated a significant interaction effect between hearing aid use and social relationships. Hearing aid use was associated with reduced depressive symptoms and enhanced QoL in older adults with hearing loss. These effects were further moderated by the strength of social relationships, underscoring the importance of both auditory and social interventions in promoting mental and emotional well-being in aging societies.</div></div>","PeriodicalId":47780,"journal":{"name":"Ssm-Population Health","volume":"32 ","pages":"Article 101861"},"PeriodicalIF":3.1,"publicationDate":"2025-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145097107","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-09-10DOI: 10.1016/j.ssmph.2025.101862
A. Maldonado , E. Torres , M. Flores , M. Rodriguez , E.A. Villavicencio , R. Torres , I. Castro , F. Torres , J.C. Loya , N. Alkhouri , S. Carvajal , Garcia
Background
Mexican-origin farmworkers experience disproportionately high rates of chronic diseases which increase their risk for metabolic dysfunction-associated steatosis liver disease (MASLD). While it has been found that neighborhood-level factors influence health as much as individual-level characteristics, including allostatic load (ALoad), these factors have been less examined in MASLD research. This study examined the association between perceived neighborhood environment, ALoad, and MASLD. It also examined whether ALoad is a mediator between perceived neighborhood context and MASLD.
Methods
Multivariable binary logistic and linear regressions were fitted to analyze data from a community-based sample of 151 Mexican-origin farmworkers residing in the Southern Arizona border region. Self-reported data on six dimensions of neighborhood context was collected. Allostatic load was calculated as an index of physiological dysregulation. Hepatic steatosis and fibrosis were assessed by liver stiffness measurements (LSM) and controlled attenuation parameter (CAP) through FibroScan®. MASLD were identified as having a CAP score of ≥288 dB/m.
Results
The mean age was 49.7 ± 14.1 years, mean BMI 31.9 ± 6.5 kg/m2, and 9.9 % had type 2 diabetes. The mean CAP score was, M = 265.8 ± 61.0 with 41.1 % of the sample exhibiting MASLD status. Perceived neighborhood violence was not associated with MASLD status; however, it was negatively associated with ALoad, (p = 0.003). ALoad was a negative mediator between perceived violence and MASLD status (p = 0.004).
Conclusion
The results of this study inform the development of culturally relevant strategies to reduce Mexican-origin farmworkers’ risk for MASLD that are highly responsive to the structural and systemic forces that shape their lived experience.
{"title":"The association between neighborhood context, allostatic load, and metabolic dysfunction-associated steatosis liver disease in Mexican-origin farmworkers along the Southern Arizona US/Mexico border","authors":"A. Maldonado , E. Torres , M. Flores , M. Rodriguez , E.A. Villavicencio , R. Torres , I. Castro , F. Torres , J.C. Loya , N. Alkhouri , S. Carvajal , Garcia","doi":"10.1016/j.ssmph.2025.101862","DOIUrl":"10.1016/j.ssmph.2025.101862","url":null,"abstract":"<div><h3>Background</h3><div>Mexican-origin farmworkers experience disproportionately high rates of chronic diseases which increase their risk for metabolic dysfunction-associated steatosis liver disease (MASLD). While it has been found that neighborhood-level factors influence health as much as individual-level characteristics, including allostatic load (ALoad), these factors have been less examined in MASLD research. This study examined the association between perceived neighborhood environment, ALoad, and MASLD. It also examined whether ALoad is a mediator between perceived neighborhood context and MASLD.</div></div><div><h3>Methods</h3><div>Multivariable binary logistic and linear regressions were fitted to analyze data from a community-based sample of 151 Mexican-origin farmworkers residing in the Southern Arizona border region. Self-reported data on six dimensions of neighborhood context was collected. Allostatic load was calculated as an index of physiological dysregulation. Hepatic steatosis and fibrosis were assessed by liver stiffness measurements (LSM) and controlled attenuation parameter (CAP) through FibroScan®. MASLD were identified as having a CAP score of ≥288 dB/m.</div></div><div><h3>Results</h3><div>The mean age was 49.7 ± 14.1 years, mean BMI 31.9 ± 6.5 kg/m2, and 9.9 % had type 2 diabetes. The mean CAP score was, <em>M</em> = 265.8 ± 61.0 with 41.1 % of the sample exhibiting MASLD status. Perceived neighborhood violence was not associated with MASLD status; however, it was negatively associated with ALoad, (<em>p</em> = 0.003). ALoad was a negative mediator between perceived violence and MASLD status (<em>p</em> = 0.004).</div></div><div><h3>Conclusion</h3><div>The results of this study inform the development of culturally relevant strategies to reduce Mexican-origin farmworkers’ risk for MASLD that are highly responsive to the structural and systemic forces that shape their lived experience.</div></div>","PeriodicalId":47780,"journal":{"name":"Ssm-Population Health","volume":"32 ","pages":"Article 101862"},"PeriodicalIF":3.1,"publicationDate":"2025-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145049192","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-09-01DOI: 10.1016/j.ssmph.2025.101858
Dorly J.H. Deeg , H. Roeline W. Pasman , Martijn Huisman , Bregje D. Onwuteaka-Philipsen
Background
The mortality decline during the past decades has postponed the age at death. Dying at older ages may involve alterations in health trajectories at the end of life. This study examined 10-year period differences in level and changes in health conditions from 3 months to 3 days before death. Distinction was made between cancer and non-cancer decedents, because their trajectories are known to differ.
Methods
Data were provided by proxies of participants in the Dutch population-based Longitudinal Aging Study Amsterdam, who died between 2005–2009 (midyear 2008) and 2017–2019 (midyear 2018), with complete data for 162 and 124 decedents, respectively. Health conditions included functional limitations, symptom severity, and low consciousness.
Results
Average age at death increased from 81.0 (2008) to 83.5 (2018). 48 % of decedents were men. Cancer constituted 33 % of causes of death. Controlling for sex, age at death, and education, non-cancer decedents experienced more often low consciousness at 3 months before death in 2018 than in 2008. At 3 months in 2018, cancer decedents experienced fewer functional limitations than in 2008 and fewer than non-cancer decedents. In both periods, cancer decedents experienced steeper functional declines than non-cancer decedents. Trajectories of symptom severity were similar in cancer and non-cancer decedents in both periods.
Discussion
In non-cancer decedents, but not in cancer decedents, the dying process was more protracted in 2018 than in 2008. Level and increase in symptom severity did not improve over time for both cancer and non-cancer decedents, suggesting that improvement of palliative care is warranted for both groups.
{"title":"Two decades of end-of-life conditions of older adults: older and more protracted?","authors":"Dorly J.H. Deeg , H. Roeline W. Pasman , Martijn Huisman , Bregje D. Onwuteaka-Philipsen","doi":"10.1016/j.ssmph.2025.101858","DOIUrl":"10.1016/j.ssmph.2025.101858","url":null,"abstract":"<div><h3>Background</h3><div>The mortality decline during the past decades has postponed the age at death. Dying at older ages may involve alterations in health trajectories at the end of life. This study examined 10-year period differences in level and changes in health conditions from 3 months to 3 days before death. Distinction was made between cancer and non-cancer decedents, because their trajectories are known to differ.</div></div><div><h3>Methods</h3><div>Data were provided by proxies of participants in the Dutch population-based Longitudinal Aging Study Amsterdam, who died between 2005–2009 (midyear 2008) and 2017–2019 (midyear 2018), with complete data for 162 and 124 decedents, respectively. Health conditions included functional limitations, symptom severity, and low consciousness.</div></div><div><h3>Results</h3><div>Average age at death increased from 81.0 (2008) to 83.5 (2018). 48 % of decedents were men. Cancer constituted 33 % of causes of death. Controlling for sex, age at death, and education, non-cancer decedents experienced more often low consciousness at 3 months before death in 2018 than in 2008. At 3 months in 2018, cancer decedents experienced fewer functional limitations than in 2008 and fewer than non-cancer decedents. In both periods, cancer decedents experienced steeper functional declines than non-cancer decedents. Trajectories of symptom severity were similar in cancer and non-cancer decedents in both periods.</div></div><div><h3>Discussion</h3><div>In non-cancer decedents, but not in cancer decedents, the dying process was more protracted in 2018 than in 2008. Level and increase in symptom severity did not improve over time for both cancer and non-cancer decedents, suggesting that improvement of palliative care is warranted for both groups.</div></div>","PeriodicalId":47780,"journal":{"name":"Ssm-Population Health","volume":"31 ","pages":"Article 101858"},"PeriodicalIF":3.1,"publicationDate":"2025-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144924711","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}