Thesis advisor: Eric Dearing / Despite global efforts toward improving gender equity in education, gender-based disparities in learning opportunities and academic outcomes are still prevalent. Many gender differences in learning and academic outcomes begin emerging early in life, even before children start formal schooling. Both theory and empirical evidence point toward children’s exposure to gender-differentiated treatment from caregivers that arises from gendered stereotypes, expectations, and cultural norms. As young children’s brains have evolved to detect the subtlest nuances in their environments, even small gender differences in learning experiences could have lasting consequences in their developmental trajectories and later-life outcomes. In turn, understanding gender disparities in early learning opportunities is critical to catalyzing young girls’ and boys’ positive learning trajectories from early on. A major challenge for the field, however, has been the very limited attention given to these issues in low- and- middle-income countries (LMICs), where about 90% of the world’s children live. The present dissertation consists of three empirical papers focused on young children in LMICs, providing significant contributions to the cumulative knowledge on gender differences in early learning. Respectively, the three papers: (1) provide a thorough accounting of gender disparities across and within 71 countries, (2) uncover specific caregiving mechanisms that help explain these gender disparities in early learning outcomes across and within countries, and (3) delve into the perceptions and meaning makings of mothers and fathers about gendered parenting practices at the local level. By so doing, this dissertation pushes forward our understanding of gender disparities in early learning opportunities and development across multiple ecological levels. The results may help inform global policy and practice aimed at gender equity, a matter of social justice that has proven to affect the social fabric, economic vitality, and political stability of nations. / Thesis (PhD) — Boston College, 2024. / Submitted to: Boston College. Lynch School of Education. / Discipline: Counseling, Developmental and Educational Psychology.

Smoking cessation can reduce health risk and prevent millions of tobacco-related deaths. However, cessation rates are low in low-and-middle income countries (LMICs), with only a small proportion of smokers intending to quit. Given the paucity of literature to support tobacco cessation programs in LMICs, this study aimed to: 1) identify factors associated with intention to quit smoking, 2) assess the relationship between health care provider quit advice/tobacco screening and utilization of cessation assistance, and 3) examine the relationship between home smoking rule and smoking intensity across three stages of smoking cessation (precontemplation, contemplation and preparation) in LMICs. Data were obtained from the Global Adult Tobacco Survey, 2009-2012, a nationally representative household survey of noninstitutionalized civilians aged 15 years and older. Weighted multivariable regression analyses were conducted using SAS version 9.4. Adjusted odds ratios (OR), percent change in smoking intensity and associated 95% confidence intervals (CI) were estimated. Home smoking rule and exposure to anti-smoking messages were the important factors associated with contemplation and preparation to quit smoking. Approximately 1%, 7%, 9% and 15% used quitline, medical treatment, counseling/cessation clinic and cessation assistance (all three combined), respectively, in the past year. Quit advice was significantly associated with utilization of counseling/cessation clinic (OR=3.89, 95% CI=2.8–5.5), medical treatment (OR=1.71, 95% CI=1.2–2.4) and cessation assistance (OR=2.60, 95% CI=2.0–3.4). Tobacco screening was associated with utilization of counseling/cessation clinic (OR=2.60, 95% CI=1.1–5.9) and medical treatment (OR=1.71, 95% CI=1.2–2.4). Living in a completely smoke-free home was associated with a 22.5% (95% CI=17.1%–28.0%), an 18.6% (95% CI=9.0%–28.2%), and a 19.4% (95% CI=3.9%–34.9%) significant reduction in smoking intensity among smokers in precontemplation, contemplation and preparation, respectively. In conclusion, the results suggest that smoke-free home, anti-smoking campaigns, and health care provider intervention promote smoking cessation in LMICs. Therefore, comprehensive smoke-free policies, anti-smoking media campaigns and integration of tobacco screening and quit advice into the health care system are important for tobacco cessation in LMICs, suggesting the need for full implementation of the World Health Organization Framework Convention for Tobacco Control Articles 8 and 11 – 13.

Increasing rates of adult-onset diabetes in low- to middle-income countries (LMICs) present a growing healthcare concern that requires a regulatory approach and local community engagement. Cross-sector collaboration (CSC) is an effective strategy for improving community health. CSC is not effectively used to manage adult-onset diabetes in LMICs, nor is there public policy advocating CSC in the management of diabetes. This qualitative exploratory case study explored how leadership and resources affect the use of CSC to assist in managing the adult-onset of diabetes in a CSC in Cameroon. A purposeful sampling strategy was used to identify participants for semistructured interviews of 10 executives involved in the CSC. CSC theory provided a theoretical framework. Collected data were organized by audio recording, transcribing, translating, member checking, and thematically coding data using NVivo 12 software. Data from a review of documents and researcher observations were triangulated with interview data. Findings revealed that poverty, access to health care, and rising diabetes prevalence contributed to environmental turbulence. Goal-setting and objectives, intersectoral-communication, fostering trust and respect, and sharing data and motivating people presented leadership effectiveness challenges. The following resource challenges emerged in the CSC: public policy, money, skills and expertise, equipment and supplies, and connection to the target population. This increased scientific understanding could help government policy-makers and nongovernmental organizations expand public policy leading to a decrease in the prevalence and burden of adult-onset diabetes.

INTRODUCTION: Epidemiological transition theory, first postulated by Omran in 1971, provides a useful framework for understanding cause-specific mortality changes and may contribute usefully to predictions about cause-specific mortality. However, understandings of mortality transitions and associated epidemiological changes remain poorly defined for public health practitioners due to lack of evidence from low- and middle-income countries. Therefore, understanding of the concept and development of epidemiological transition theory as well as population burden of premature mortality attributable to risk factors is needed. OBJECTIVES: This thesis aims to understand how epidemiological transition theory has been applied in different contexts, using available evidence on mortality transitions from high, middle- and low- income countries, as well as the contribution of risk factors to mortality transitions, particularly for premature mortality. METHODS: A Medline literature search from 1971 to 2013 was conducted to synthesise published evidence on mortality transition (paper I). A descriptive analysis of trends in cause of death using INDEPTH data was conducted, focusing on specific causes of death in 12 INDEPTH sites in Africa and Asia, using the INDEPTH 2013 standard population structure for appropriate comparisons across sites (paper II). A retrospective dynamic cohort database was constructed from Swedish population registers for the age range 30-69 years during 1991-2006, to measure reductions in premature non-communicable disease mortality using a life table method (paper III). Prospective cohort data from Västerbotten Intervention Programme from 1990 to 2006 were used to measure the magnitude of premature non-communicable disease mortality reductions associated with risk factor changes for each period of time (paper IV). FINDINGS: There were changes in emphasis in research on epidemiological transition over the four decades from 1971 to 2013, from cause of death to wide-ranging aspects of the determinants of mortality with increasing research interests in low-and middle-income countries, with some unconsidered aspects of social determinants contributing to deviations from classic theoretical pathways. Mortality rates declined in most sites, with the annual reductions in premature adult mortality varied across INDEPTH sites, Sweden, which now is at late stage of epidemiological transition stage, achieved a 25% reduction in premature mortality during 1991-2006. Overall downward trends in risk factors have helped to reduce premature mortality in the population of Västerbotten County, but some benefits were offset by other increasing risks. The largest mortality changes accrued from reductions in smoking, hypertension and hypercholesterolaemia. CONCLUSIONS: This thesis established patterns of current epidemiological transition in high, middle-and low-income countries (Asia and Africa), where the theory fits the transition patterns in some countries, but with some needs for further adjustments in other settings, as well as deviations from the classical ET theory in the last four decades. It highlights the need to identify the burden of mortality and morbidity, particularly for reducing mortality occurring before the age of 70 years and its attribution to risk factors, which are a major public health challenge. This informs shifting of public health priorities and resources towards prevention and control of chronic non-communicable disease risk factors.

This thesis studies the determinants and outcomes of the skill systems in Mexico and Turkey, two crucial cases of middle-income countries (MICs). Despite the similarities in their previous institutional environments and links to the global economy, Turkey has experienced higher economic growth and better social development compared to Mexico. This PhD project focuses on the Mexican and Turkish skill systems, which have been significant institutions that affect the (different) economic and social characteristics of advanced industrialised countries in the literature on comparative capitalisms (CCs). This study builds its theoretical approach on the arguments of CCs literature and the claims of globalisation, where it draws from the studies on global value chains (GVCs). It scrutinises the patterns of convergence due to globalisation and ongoing divergence because of the variation of institutional structures. It examines the complementarities and outcomes of the skill systems, which will then help to understand the divergence of development experiences between Mexico and Turkey. This PhD project carries out a multi-level research and focuses the empirical study on auto parts-automotive value chains (AACs), which is an important industry for many MICs including Mexico and Turkey. The multi-level research first includes a macro-level study of local institutions in Mexico and Turkey, as well as the convergence patterns through the AACs. This is complemented with a micro-level analysis of firms' strategies to find workers with technical skills in production functions. The findings show that while there is some convergence in the Mexican and Turkish auto parts producers' skill needs, firms from these two countries adopt different strategies to address their needs. The distinctive national institutions in Mexico and Turkey, which are shaped by the differences in the state's involvement, are the main reasons for this divergence. The differences between the Mexican and Turkish skill systems have created different outcomes for both the firms and workers in these countries, and hence are expected to have contributed to their diverging development paths.

INTRODUÇÃO: Países de baixa e média renda representam mais de 85% da população mundial de 7 bilhões de pessoas. Nesses países, a população idosa apresentou acelerado crescimento nas últimas décadas. Esse rápido crescimento está sendo acompanhado pelo aumento dos agravos relacionados ao envelhecimento e por mudanças no padrão de morbimortalidade. Entre as condições que têm demonstrado associação com aumento de mortalidade, está a depressão. Porém, grande parte das pesquisas sobre a associação da depressão com mortalidade é de países de alta renda. OBJETIVOS: 1) Investigar a associação entre depressão e mortalidade em idosos de países de baixa e média renda. 2) Investigar a associação de depressão/sintomas depressivos e mortalidade por todas as causas e causas específicas em uma coorte de idosos residentes na zona oeste de São Paulo em um período de 10 anos de seguimento. MÉTODOS: 1) Foi realizada uma revisão sistemática para identificar estudos observacionais que investigaram a associação entre mortalidade e depressão em países de renda média ou baixa. Dois revisores trabalharam independentemente para selecionar artigos, extrair dados e avaliar a qualidade dos estudos. Uma meta-análise foi realizada a partir dos dados extraídos dos estudos incluídos na revisão sistemática. 2) Foi realizado um estudo de coorte utilizando a amostra do \"São Paulo Ageing & Health Study\" (SPAH), investigação que incluiu idosos com 65 anos ou mais, residentes de áreas economicamente menos favorecidas na região Oeste da cidade de São Paulo. Os participantes foram identificados por arrolamento domiciliar e entrevistados em seus domicílios seguindo protocolo padronizado de pesquisa. A definição de depressão e sintomas depressivos foi realizada a partir dos instrumentos Geriatric Mental State (GMS) e Inventário Neuropsiquiátrico (NPI). Os dados de seguimento foram extraídos a partir das declarações de óbitos dos residentes do estado de São Paulo através de procedimento de linkage de dados com os bancos de declarações de óbitos da fundação SEADE. Foram examinadas as associações independentes entre depressão e sintomas depressivos e mortalidade através de modelos de riscos proporcionais de Cox e também foi elaborada curva de sobrevida de Kaplan Meier. RESULTADOS: 1) Foram incluídos 10 estudos, com um total de 13.828 participantes (2.402 deprimidos e 11.426 não-deprimidos) de 6 países (Brasil, 4 artigos; China, 2 artigos; Botswana, Índia, África do Sul e Coréia do Sul, 1 artigo). O risco relativo total não ajustado (RR) de mortalidade em idosos com depressão em relação aos não deprimidos foi de 1,62 (IC95% 1,39-1,88; p < 0,001), com alta heterogeneidade (I2, 66%; IC95% 33-83; p < 0,005). Após o ajuste para o viés de publicação, o RR geral diminuiu para 1,60 (IC95%; 1,37-1,86). Não foram observadas diferenças significativas entre os subgrupos, exceto aqueles definidos pela qualidade do estudo. Os estudos de alta qualidade tiveram um RR agrupado de 1,48 (IC95% 1,32-1,67), enquanto os estudos de baixa qualidade resultaram em um RR agrupado de 1,82 (IC95% 1,25-2,65). 2) Dos 1.967 participantes do SPAH, 443 (22,52%) apresentaram sintomas depressivos e 90 (4,58%) preenchiam critérios para diagnóstico de depressão pela CID-10 no início do seguimento. Foram observados 688 (34,98%) óbitos no seguimento. Presença de sintomas depressivos (HR=1,23; IC95% 1,03-1,46; p=0,022) e depressão (HR=1,75; IC 95% 1,29-2,37; p < 0,001) no início do seguimento estiveram positivamente associadas à mortalidade por todas as causas. Sintomas depressivos (HR=1,26; IC 95% 1,04-1,53; p=0,018) persistiram associados a mortalidade mesmo após controle por variáveis confundidoras, enquanto depressão (HR=1,20; IC95% 0,85-1,67; p=0,296) perdeu significância estatística ao se adicionar ao modelo variáveis relacionadas a comorbidades. Considerando causas especificas de óbitos, sintomas depressivos apresentaram associação significativa com óbitos por doenças cardiovasculares (HR=1,48; IC95% 1,12-1,95; p=0,006) e depressão apresentou associação com óbitos por câncer (HR=2,26; IC95% 1,17-4,36; p=0,015). CONCLUSÃO: A depressão está associada ao excesso de mortalidade em idosos que vivem em países de baixa e média renda. Além disso, esse excesso de mortalidade não difere substancialmente do encontrado em países de alta renda. Já sintomas depressivos e depressão estiveram associados ao aumento de mortalidade em idosos entre os participantes do SPAH. Dado que sintomas depressivos e depressão aumentam o risco de óbito, um melhor entendimento sobre essa relação e estratégias de provisão de serviços de saúde mental, em especial em países de média e baixa renda, são necessários já que essas condições clínicas modificáveis muitas vezes não são reconhecidas ou tratadas adequadamente / INTRODUCTION: Low and middle income countries represent more than 85% of the world population of 7 billion people. In these countries, the elderly population has grown rapidly in recent decades. This rapid growth is being accompanied by the increase in aggravations related to aging and changes in the morbidity and mortality pattern. Among the risk factors that have been shown to be associated with excess mortality are depression. However, much of the research on the association of depression with mortality is from high-income countries. OBJECTIVES: 1) To investigate the association between depression and mortality in the elderly living in low- and middle-income countries 2) To investigate the association of depressive / depressive symptoms and all-cause and specific-cause mortality in a cohort of elderly living in the western zone of São Paulo over a period of 10 years of follow-up. METHODS: 1) A systematic review was conducted to identify observational studies that investigated the association between mortality and depression in middle- and low-income countries. Two reviewers independently worked to select articles, extract data, and evaluate the quality of the studies. A meta-analysis was performed from the data extracted from the studies included in the systematic review. 2) A cohort study was carried out using a sample from the São Paulo Aging and Health Study (SPAH), an investigation that included elderly people aged 65 years and over, from economically disadvantaged areas in the western region of the city of São Paulo. Participants were identified by household survey and interviewed in their homes following a standardized research protocol. The definition of depression and depressive symptoms was performed using the Geriatric Mental State (GMS) and Neuropsychiatric Inventory (NPI) instruments. The follow-up data were extracted from the statements of deaths of the residents of the state of São Paulo through a data linkage procedure with the database of death declarations of the SEADE foundation. We examined the independent associations between depression and depressive symptoms and mortality through Cox proportional hazards models and also elaborated the Kaplan Meier survival curve. RESULTS: 1) Ten studies were included, with a total of 13,828 participants (2,402 depressed and 11,426 non-depressed) from 6 countries (Brazil, 4 articles; China, 2 articles; Botswana, India, South Africa and South Korea, 1 article) were included. The overall unadjusted relative risk (RR) of mortality in depressed relative to non-depressed participants was 1.62 (95% CI 1.39-1.88; p < 0.001), with high heterogeneity (I2=66%; 95% CI 33-83; p < 0.005). After adjustment for publication bias, the RR decreased to 1.60 (95% CI 1.37-1.86; p < 0.001). No significant differences were observed between subgroups except those defined by study quality. The high quality studies had a pooled RR of 1.48 (95% CI 1.32-1.67; p<0.001), while the low quality studies resulted had a pooled RR of 1.82 (95% CI 1.25-2.65; p < 0.005). 2) Of the 1,967 participants, 443 (22.52%) had depressive symptoms and 90 (4.58%) had a diagnosis of depression by ICD-10 at the beginning of follow-up. There were 688 (34.98%) deaths at follow-up. Presence of depressive symptoms (HR=1.23; 95% CI 1.03-1.46; p=0.022) and depression (HR=1.75; 95% CI 1.29-2.37; p < 0.001) at the beginning of follow-up were positively associated with all-cause mortality. Depressive symptoms (HR=1.26; 95% CI 1.04-1.53; p=0.018) persisted associated with mortality even after controlling for confounding variables, while depression (HR=1.20; 95% CI 0.85-1.67; p=0.296) lost association when variables related to comorbidities were added to the model. Considering specific causes of death, depressive symptoms had a significant association with death due to cardiovascular diseases (HR=1.48; 95% CI 1.12-1.95; p=0.006), and depression was associated with cancer deaths (HR=2.26; 95% CI 1.17-4.36; p=0.015). CONCLUSION: Depression is associated with excess mortality in the elderly living in low- and middle-income countries. In addition, this excess mortality does not differ substantially from that found in high-income countries. Depressive symptoms and depression were associated with increased mortality in the elderly. Given that depressive symptoms and depression increase the risk of death, a better understanding of this relationship and strategies for providing mental health services, especially in middle- and low-income countries, are necessary since these modifiable clinical conditions are often not recognized or properly handled

Background: Parent behaviours have a long-lasting impact on the health, education, and behaviour of the young people in their care. Group-based parenting interventions are a promising approach to improve parenting, as well as to prevent and reduce physical and emotional maltreatment of children. While a billion of adolescents live in low- and middle- income countries, few evaluations have examined parenting approaches for families with adolescents in these countries. Objectives: This study was nested within a randomised controlled trial of Sinovuyo Teen, a parenting programme for families with adolescents. The trial took place in rural South Africa in 2015-2016 with 552 families in 40 study clusters. Primary intervention outcomes included parenting and child maltreatment. This dissertation aims to: (1) describe the implementation (attendance, engagement, and fidelity) of the intervention delivered within the evaluation, (2) examine if participant characteristics affected attendance and engagement in the intervention, (3) examine if implementation characteristics affected programme results, and (4) examine if participant characteristics affected programme results. Methods: Programme implementation was assessed through observations of all programme group sessions and the records of the implementing organisation. The analyses also draw on participant self-report data from three time-points (baseline, immediate post-test, and follow-up). The data were analysed using a series of correlation and multilevel regression analyses, presented in three papers. Findings: Attendance levels in the intervention were somewhat lower than in previous similar studies, perhaps due to the role of home visits. Generally, more disadvantaged families participated at similar rates as families with more material and social resources. However, a number of factors affected attendance on individual and family levels, e.g. an overcrowded household, and at the community level, e.g. funeral and grant receipt days. Intervention fidelity was similar to the levels reported in high-income countries, thus suggesting that high implementation quality is feasible in a low-resource setting. There was no consistent impact of the variation in participation and implementation on participant outcomes. Baseline risks did not consistently affect variation in treatment effects, confirming recent findings that families at-risk can benefit from parenting support as much or more than less at-risk families.

Chronic disease is rising globally, but LMICs may require novel approaches to management suited to a low-resource setting. Private health care providers in LMICs have experimented and developed new models for organizing, financing, and delivering care. This thesis examines some of the innovations that have emerged in LMICs to manage chronic disease. Using a common performance measurement framework and drawing from a database of over 1200 innovative health programs, I identify and evaluate 46 programs addressing chronic disease in LMICs. I then go on to identify, innovative practices used by the subset of 19 diabetes focussed programs and catalogue them according to the Chronic Care Model (CCM). Delivery system design is the most commonly used domain of the CCM, which even earlier has shown to have great potential to impact health outcomes. Few of the identified innovations may also have the potential for ‘Reverse innovation’ in high income countries.

Chronic disease is rising globally, but LMICs may require novel approaches to management suited to a low-resource setting. Private health care providers in LMICs have experimented and developed new models for organizing, financing, and delivering care. This thesis examines some of the innovations that have emerged in LMICs to manage chronic disease. Using a common performance measurement framework and drawing from a database of over 1200 innovative health programs, I identify and evaluate 46 programs addressing chronic disease in LMICs. I then go on to identify, innovative practices used by the subset of 19 diabetes focussed programs and catalogue them according to the Chronic Care Model (CCM). Delivery system design is the most commonly used domain of the CCM, which even earlier has shown to have great potential to impact health outcomes. Few of the identified innovations may also have the potential for ‘Reverse innovation’ in high income countries.

Background: To address the huge human resources for health (HRH) crisis that Zambia and other low and middle income countries (LMICs) are experiencing, most LMICs have engaged the services of small scale community-based health worker (CBHW) programmes. However, several challenges affect the CBHWs’ ability to deliver services. Integration of national CBHW programmes into health systems is an emerging innovative strategy for addressing the challenges. Integration is important because it facilitates recognition of CBHWs in the national primary health care system. However, the integration process has not been optimal, and a more comprehensive understanding of the factors that shape the integration process is lacking. This study aimed at addressing this gap by analysing the integration process of national CBHW programmes in health systems in LMICs, with a special emphasis on Zambia. Methodology: This was a qualitative study that used case study and systematic review study designs. The case study focused on Zambia and analysed the integration processes of Community Health Assistants (CHAs) into the health system at district level (Papers I-III). Data collected using key informant interviews, participant observation, in-depth interviews and focus group discussions were analysed using thematic analysis. The systematic review analysed, using thematic and pathways analysis, the integration process of national CBHWs into health systems in LMICs (Brazil, Ethiopia, India and Pakistan)-(Paper IV). The framework on the integration of health innovations into health systems guided the overall analysis. Results: Factors that facilitated the integration of CHAs into the health system in Zambia included the HRH crisis which triggered the willingness by the Ministry of Health to develop and support implementation of the integration strategy-the CHA strategy. In addition, the attributes of the CHA strategy, such as the perceived competence of CHAs compared to other CBHWs, enhanced the community’s confidence in the CHA services. Involvement of the community in selecting CHAs also increased the community’s sense of programme ownership. However, health system characteristics such as limited support by some support staff, supply shortages as well as limited integration of CHAs into the district governance system affected CHAs’ ability to deliver services. In other LMICs, as in Zambia, the HRH problems necessitated the development of integration strategies. In addition, the perceived relative advantage of national CBHWs with regard to delivering health services compared to the other CBHWs also facilitated the integration process. Furthermore, the involvement of community members and some politicians in programme processes enhanced the perceived legitimacy, credibility and relevance of programmes in other LMICs. Finally, the integration process within the existing health systems enhanced programme compatibility with health system elements such as financing. However, a rapid scale-up process, resistance from other health workers, ineffective incentive structures, and discrimination of CBHWs based on social, gender and economic status inhibited the integration process of national CBHWs into the health systems. Conclusion: Strengthening the integration process requires fully integrating the programme into the district health governance system; being aware of the factors that can influence the integration process such as incentives, supplies and communication systems; clear definition of tasks and work relationships; and adopting a stepwise approach to integration process.