Inadequate public health resources coupled with a chronically overburdened health system leave a large proportion of South Africans unable to access mental health care. Low-income pregnant women with common mental disorders (CMDs) are arguably more vulnerable to falling through the treatment gap, given the low rates of detection during pregnancy and the numerous additional barriers to care. The direct and indirect financial and personal costs associated with perinatal mental illness are substantial, while the high prevalence rates of perinatal CMDs make this an area in need of urgent attention. Integrating task shifting approaches into perinatal primary health care services is a promising solution. The first chapter introduces the thesis, providing context to the studies that are presented in later chapters and an overview of the research questions that informed them. The second chapter constitutes a systematic review of the literature relevant to the studies. Chapters 3 to 6 report on the findings of the studies, briefly described in the abstract below. The prevalence and risk factors associated with perinatal psychological distress - a plausible precursor for common mental disorders (CMDs) - are not widely understood in under-resourced settings. The first study (Chapter 3) investigates the prevalence and predictors of psychological distress in the antenatal period. Data were collected from 664 pregnant women who reported for antenatal care to any one of 11 Midwife and Obstetric Units (MOU) across the greater Cape Town area. Psychological distress was measured using the Symptom Response Questionnaire (SRQ-20; cut-off value of 7/8), while data pertaining to risk factors were collected via a demographic questionnaire, the Alcohol, Smoking and Substance Involvement Screening Test (ASSIST) and the Multidimensional Scale of Perceived Social Support (MSPSS). The prevalence of antenatal psychological distress was 38.6%. Risk factors included low socio-economic status (SES) (OR = 1.45, 95% CI: 1.24-1.68); recent physical abuse and/or rape (OR = 1.94, 95% CI: 1.57-2.40); complications during a previous delivery (OR = 1.18, 95% CI: 1.01-1.38); having given birth before (OR = 1.61, 95% CI: 1.21-2.14). The high prevalence rate of psychological distress is consistent with those found in other South African studies of perinatal CMDs. Appropriate, context-specific, and effective interventions are better served by investigating a broader range of symptoms associated with perinatal CMDs in these settings. The second study (Chapter 4) examines the mental health literacy (MHL) of pregnant women, including their perceptions of the causes of mental illness during pregnancy and best treatment approaches. Understanding the factors that represent barriers to accessing care is important to the development of accessible interventions. Globally, low levels of mental health literacy have often been identified as one such treatment barrier. However, little is known about how pregnant women perceive and understand mental illness during this time, particularly in South Africa. A convenience sample of 262 pregnant women attending routine antenatal appointments at a Midwife and Obstetrics Unit (MOU) were recruited to participate in the study. Participants were presented with one of five possible vignettes, depicting a woman with perinatal mental illness, as defined by the DSM 5, including ante- and postnatal depression, panic disorder, substance dependence and schizophrenia. Participants were then asked to provide a diagnosis and completed two scales assessing aspects of mental health literacy. The results from this study showed that more than three quarters of respondents (77.4%) did not identify the signs and symptoms described in the vignettes as those consistent with mental illness. More than half of all participants (57.5%) were of the view that all the disorders depicted were "typical of a weak character", while stress was the most widely held explanation for symptoms of all disorders. Participants were most confident in the therapeutic potential of psychological services, especially consulting with a counsellor or social worker. These were closely followed by lifestyle and self-help options as the most endorsed means to addressing psychiatric symptoms during pregnancy. Notably, seeking help from a spiritual or religious advisor was comparably as popular among participants as seeking help from a psychologist or social worker. Given the elevated prevalence of perinatal mental illness, these findings are cause for concern. Developing socio-culturally nuanced understandings of how perinatal mental illness is perceived should be emphasized as central to the development of successful interventions. The third study (Chapters 5 and 6) investigates the feasibility and acceptability of, as well as the preliminary responses to an adapted Problem Solving Therapy (PST) intervention to treat psychological distress. Given the large treatment gap that exists in public mental health, support for task shifting evidence based mental health treatments is growing. However, the gaps in our knowledge are threefold. First, most research has used lay counsellors to deliver interventions. No research has used Registered Counsellors (RC) to conduct interventions. Second, very little is known about the potential outcomes of task shifting an adapted PST intervention to reduce symptoms of psychological distress. Third, data regarding the feasibility and acceptability of such interventions in South African Midwife and Obstetric Units (MOUs) is very limited. Results from the study are presented in two chapters. Chapter 5 focuses on the intervention participants. Thirty-eight women who screened positive for high CMD symptoms on the Edinburgh Postnatal Depression Scale (EPDS) at their first antenatal visit were recruited to participate in the intervention. Of these, 22 completed the preand post-intervention interviews. Using mixed methods, preliminary responses to the threesession PST intervention, as well as participants' perceptions of the intervention's feasibility and acceptability were explored. Primary outcomes included psychological distress as measured by the Symptom Response Questionnaire (SRQ-20) and CMD symptoms, as measured by the EPDS. A short semi-structured post-intervention interview was also conducted approximately three months after each participant's last session. On the primary outcome measures, significant reductions were seen on EPDS scores (z = -3.0, p < 0.01) as well as the SRQ-20 scores (z = -3.5, p = <0.01). Several significant reductions were also seen on secondary outcomes. Reductions in impairment to functioning were also noted, with all three Sheehan disability scales reflecting less disruption to work (z = -2.3, p = 0.02), social life (z = -3.3, p < 0.01), as well as family and home responsibilities (z = -2.5, p = 0.01). Perceived Stress Scale scores were also significantly reduced (z = -3.4, p < 0.01). Significant changes were seen on two problem-solving styles, with reduced 'negative problem orientation' scores (z = -3.1, p < 0.01) and 'avoidant style' scores (z = -3.0, p < 0.01) Participants felt that the intervention was feasible and acceptable. The intervention's acceptability lay primarily in the opportunity for participants to talk confidentially to a non-judgmental and empathic person about their problems. The intervention materials seemed to serve as an extension of the therapeutic process. Factors that were identified by participants as representing potential barriers to the intervention included lack of transport or money, work commitments and stigma. Chapter 6 explores the intervention's feasibility and acceptability from the perspectives of 6 stakeholders who were involved with the project. Semi-structured interviews were conducted with each stakeholder. Data from the interviews showed that the stakeholders felt that the intervention was helpful to patients and a valuable resource for the facility to have. Some expressed concern about how stigma associated with mental illness might be a barrier to patients who need mental health care. To the staff, the project's value seemed to lie primarily in the support it provided in managing emotionally distressed patients. Having a resource to refer patients to appeared to provide overburdened staff with some relief. None of the stakeholders reported that the screening and referral procedures added to their workload. Some stakeholders felt detection of psychological problems among patients was compromised without mental health screening. Staff felt that a walk-in counselling service would serve to improve future interventions. Limitations to the first study included its cross-sectional design and use of a screening tool to measure the prevalence of psychological distress. The second study was limited by the employment of vignettes to collect data. While they are useful tools to elicit population-specific responses, their adaption for those purposes means that they are not standardised. The study was further limited by the use of a convenience sample. Finally, the third study was limited by a small sample size owing to low retention rates. However, low retention rates are not uncommon among antenatal and low-income populations, where structural barriers to accessing care are often more pronounced. The lack of a comparison group was an additional limitation. In conclusion, data from this study support task shifting evidence based treatments to Registered Counsellors to treat the highly prevalent antenatal psychological distress. Difficulties distinguishing CMD symptoms from normal pregnancy experiences may influence the uptake of counselling services and represent a barrier to care. Improving mental health literacy may be a necessary supplement to future interventions. Future research should focus on evaluating real-world models of integrated mental health in primary care settings. How psycho-education programmes might impact upon the uptake of services at antenatal care facilities will also contribute to broadening our knowledge of developing effective and appropriate interventions.

This dissertation seeks to increase understanding about why older African Americans have equal or lower rates of mental illness than whites; despite the general tendency that disadvantaged individuals have worse mental health than advantaged individuals. To explain the unexpected finding from a methodological standpoint, this study uses the 2008 and 2012 Waves of the Health and Retirement Study and examines racial differences in the factor structures of mental illness focusing on symptoms included in the CES-D (Center for Epidemiologic Studies Depression) and the BAI (Beck Anxiety Inventory). To understand the implications of these racial differences in the stress process, this study also examines how discrimination, as a type of stressor, manifests into different mental health outcomes (somatic or mood-based symptoms) between older African Americans and whites. The analysis employs structural equation modeling and finds that modeling CES-D and BAI items as mood-based and somatic constructs of general distress has better model fit than modeling these items than as separate constructs of depression and anxiety (Chapter 2). In regard to racial differences in the factor structure of these models, the analysis finds that the factor structure of the somatic model differs between the racial groups whereas the factor structure of the mood-based model does not differ. When examining how discrimination manifests into different mental health outcomes, the analysis shows that discrimination is related to increased mood-based symptoms among whites and unrelated to mood-based symptoms among African Americans (Chapter 3). Additionally, everyday discrimination and lifetime discrimination are related to increased somatic symptoms among whites. Among African Americans, physical disability and financial status based discrimination are related to increased somatic symptoms, and lifetime discrimination and ancestry-based discrimination are related to decreased somatic symptoms. The findings from this dissertation suggest that current measures of mental illness (e.g., CES-D, BAI, etc.) may not measure the same constructs across racial groups and that modeling mental illness as mood-based and somatic constructs may give better insights into the prevalence of mental illness among older adults. Further, this study sheds light on the potential effects of discrimination on an alternative and more culturally specific expression of mental illness symptoms (somatic symptoms). This study also extends the literature on discrimination and mental health by illustrating how different forms of discrimination impact different mental health outcomes. / A Dissertation submitted to the Department of Sociology in partial fulfillment of the requirements for the degree of Doctor of Philosophy. / Spring Semester, 2016. / April 7, 2016. / African American, anxiety, depression, older adults / Includes bibliographical references. / Koji Ueno, Professor Directing Dissertation; Ming Cui, Committee Member; Amy Burdette, Committee Member; Miles Taylor, Committee Member; , .

OBJECTIVE: To evaluate the feasibility and acceptability, and preliminary efficacy of Modified Comprehensive Behavioral Intervention for Tics (MCBIT) therapy for patients with persistent chronic tic disorders (CTD) and co-occurring attention deficit hyperactivity disorder (ADHD). METHOD: Seventeen child and adolescent patients aged 10-17 with CTD and co-occurring ADHD were randomly assigned to the MCBIT group (n=9) or to a control group, where they received traditional Comprehensive Behavioral Intervention for Tics (CBIT) therapy (n=8). Both groups received ten fifty-five-minute weekly treatment sessions, and two fifty-five-minute biweekly relapse prevention sessions. RESULTS: Sixteen of the seventeen participants completed the study, and acceptability ratings in both treatment groups were high with no significant differences in expectation of improvement. The MCBIT and CBIT groups in combination showed significant improvement in measures of tic severity, ADHD symptom severity, and tic impairment and group differences were not statistically significant. CONCLUSION: The results indicate that MCBIT treatment is feasible and acceptable for youth with CTD and ADHD, and the findings demonstrate preliminary support for the modified CBIT treatment’s efficacy in reducing tic and ADHD symptoms and improving tic-related quality of life. Additional studies with more participants are warranted to further examine the role of a modular behavioral treatment approach in targeting commonly co-occurring disorders simultaneously and successfully.

Methamphetamine users may suffer from a range of co-morbid psychiatric disorders. Predictors of treatment outcome in substance dependence may include both such co-morbidity and readiness for change. The nature of the relationship between psychiatric co-morbidity and readiness for change has not been systematically studied. Therefore, this study aimed to assess the prevalence and patterns of psychiatric disorders in individuals dependent on methamphetamine; determine whether there is a relationship between such co-morbidity and readiness for change; and identify factors associated with readiness for change in this group.

Introduction Continuously rising health care and workplace costs associated with mental illness is demanding attention from health care funders in South Africa’s private health care sector. The majority of mental health care costs are generated by in-hospital care, whilst funded access to ambulatory care is limited in this sector. The Medscheme Mental Health Programme (MMHP) is a collaborative care project which aims to promote the integration of good quality mental health care into the primary care setting. In a “treatment-to-target” approach, symptom score trackers are used to systematically monitor response to treatment in order to help identify and modify suboptimal treatment plans timeously (Hattingh 2017b). Aims This study describes the MMHP participants and pathways into and through the MMHP, and its initial clinical outcomes. Methods Principal members and dependant beneficiaries of two participating medical schemes screened for enrolment on the MMHP between 1 August 2016 and 28 February 2018 were included in the study. Persons younger than 18 years were excluded. Symptoms of major depressive disorder (MDD), generalised anxiety disorder (GAD), posttraumatic stress disorder (PTSD) and alcohol abuse were screened for by using the Patient Health Questionnaire-9 (PHQ-9) (Spitzer, Williams, and Kroenke 2002-2015; Kroenke and Spitzer 2002), the Generalised Anxiety Disorder Questionnaire-7 (GAD-7) (Spitzer, Williams, and Kroenke 2002-2015; Spitzer and Kroenke 2006), the Primary Care Post-Traumatic Stress Disorder Screen (PC-PTSD) US Department of Veteran Affairs (2015); (Prins, Ouimette, and Kimerling 2003) and the Alcohol Use Disorders Identification Test (AUDIT) (Babor et al. 2001). The Medscheme Care Manager administered these questionnaires telephonically to screen candidates for enrolment on the Programme and communicated regularly with the associated clinical practitioner regarding treatment response. A specialist psychiatrist reviewed and provided recommendations on problematic cases at set intervals. Using logistic regression, the association between demographic characteristics and scheme type and the presence of moderate or severe symptoms of 1) depression, 2) generalised anxiety disorder, and 3) post-traumatic stress disorder, was assessed. Percentages of the sample with a single condition, one, two and three comorbidities were also analysed, as well as the proportions of co-occurrence per various combinations of conditions. Wilcoxon signed rank tests were used to determine the change in symptom severity between baseline and 10 weeks in those receiving intervention through the MMHP. Linear regression models were created to analyse the predictors of change in clinical scores. Results In the screened group, 48.6% were found to have moderate to severe symptoms of anxiety on the GAD-7, 53.2% of depression on the PHQ-9, and 33.2% of PTSD on the PC-PTSD. Relatively high rates of possible comorbidity were found in this study, especially between depression and anxiety: of those screening positive for any one condition, 73.8% screened positive on the combination of PHQ-9 and GAD-7. Screening positive on the PHQ-9 was found to be a very strong predictor of concomitant positive screening on the GAD-7 (OR = 36.4, CI = 25.3 - 52.2), and vice versa - screening positively on the GAD-7 strongly predicted positive screening on the PHQ-9 (OR = 36.6, CI = 25.4 - 52.6). Strong associations were demonstrated with females and potential depression (OR = 1.51, CI = 1.03 - 2.21) and/or PTSD (OR = 1.65, CI = 1.18 - 2.31), while younger age was significantly associated with higher likelihood of screening positive for potential depression (OR: 0.99, CI= 0.98 - 1.00), PTSD (OR = 0.97, CI 0.96 - 0.98) and/or generalised anxiety disorder (OR = 0.97, CI = 0.96 - 0.98). There were statistically and clinically significant improvements in clinical scores for all four conditions at Week 10 after enrolment on the MMHP, compared to baseline: 21% reduction in mean scores in the AUDIT, 43% in the GAD-7, 45% in the PHQ-9, and 36% in the PC-PTSD. Conclusion In its current form, the MMHP appears to be successful in reaching significantly symptomatic medical scheme beneficiaries, with possible scope to expand its reach. Certain key design elements such as using clinical data to determine risk and need for intervention, treatment target calculation adjusted for baseline, screening for comorbidity, and current referral sources, appear to be appropriate. Given the absence of a control group, however, further research is required to confirm the outcomes of the intervention.

There are as yet no ideal biomarkers of HIV-associated neurocognitive disorders. As astrocytosis is a feature of HIV encephalitis, the marker S100β may hold promise as a biomarker of HAND. We explored associations between S100β and neurocognition in individuals with HIV in Cape Town, South Africa, before and after antiretroviral therapy (ART) was initiated. The S100β levels in the cerebrospinal fluid (CSF) of forty-six participants with HIV, but not yet on antiretroviral therapy, was quantified using an enzyme-linked immunoassay (ELISA). A battery of cognitive tests was performed and the global deficit score (GDS) was calculated. In twenty of these patients, the S100β analysis and the cognitive tests were repeated approximately six months after the initiation of ART. There was no significant association between cerebrospinal fluid S100β and GDS at baseline (r= -0.070; p= 0.66) or after six months of ART (r= 0.16; p= 0.52). Cerebrospinal fluid S100β levels at baseline did not predict a change in neurocognition on ART (B(SE) = 0.001, (0.001), β=0.025, p=0.85). S100β in the cerebrospinal fluid may not adequately reflect neurocognitive impairment in individuals with HIV. Our results further demonstrate that CSF S100β levels are not affected by ART, indicating persistent neuroinflammation.

BACKGROUND: Methamphetamine (MA) use disorder in individuals with severe mental disorders (SMDs) has significant impact on clinical presentation and care. Although treatments exist, these are met by significant challenges. Notably, brief treatments for MA use within the general population have been feasible, acceptable and effective. An individualized, integrated treatment for MA use within a psychiatric inpatient setting would allow adjustment of the treatment according to individual patient needs. It is important to understand the patient needs and potential service barriers to care before formulating a treatment. This study begins to address this gap by seeking to understand the views of healthcare providers on a brief treatment to address MA use among patients with a dual diagnosis. METHODS: Thirteen key stakeholders working with patients with mental disorders including severe mental disorders and co morbid MA use were interviewed using an open-ended semi- structured interview schedule designed to explore their views on a brief treatment for MA use among patients with a dual diagnosis. Interviews were transcribed and the framework approach was used to conduct data analysis. RESULTS: Numerous themes emerged from the data. First, there are multiple risk factors for MA use. Second, this use has a significant impact on multiple aspects of patient presentation and care including individual impacts, family impacts, and impact on care. Third, although treatments for MA use disorders exist, these have significant challenges at multiple levels. Lastly, the integration of a modified brief treatment for MA use in patients with dual diagnosis would be possible if it was adjusted to patient-specific needs within the existing system and if the team adapting the treatment were cognizant of existing and potential challenges. CONCLUSIONS: The adaptation and integration of a brief treatment for MA use among patients with severe mental disorders was considered possible and even necessary if existing and potential challenges were carefully addressed.

Researchers have found a connection between mental health diagnoses and poorer health outcomes, causing an excess morbidity and mortality gap in these populations. These mentally ill individuals have the same somatic illnesses that afflict the general population, but they experience them at higher rates. Mentally ill minority populations are at even higher risk because underprivileged status on its own has been found to correlate with poorer health outcomes. Stigma and mental illness are compounding features of poorer health outcomes. The aim of this study was to highlight how addressing stigma in underprivileged populations may result in more health care utilization and treatment and better overall health outcomes for these at-risk patient groups.

The physician assistant (PA) profession growth rate was 31% in 2019 and is projected to continue to increase in the upcoming years. The rates of depression amongst the general population has been rising over the recent years. Currently, 7.1% of Americans are diagnosed with depression, however, approximately 30% of medical students carry this same diagnosis. Unfortunately, there is minimal data on the prevalence of depression amongst PA students. The literature review for this study is composed of past research on the prevalence of depression among medical students at various different programs. Some studies accounted for variable study measures including a previous diagnosis of mental illness, demographics including race, ethnicity, etc. while others did not. One study that was reviewed assessed if there is correlation between depression and work-related personality traits including: commitment, discipline, dominance, stability, cooperation, and social competence. In summary, the comprehensive review showed significant evidence that the prevalence of depression among medical students, and the small sample size of PA students, is greater than the general population. Studies showed that there is a significant association between the prevalence of depression and certain socio-demographics and work-related personality traits. This thesis proposes a multi-program, prospective cohort study to identify the prevalence of depression in PA students and identify potential factors that may be associated with the development of depression in PA school. The results of this study will then be compared to the general population’s rate of depression.This study will include various PA programs throughout the United States and assess the prevalence of depression among PA students while also accounting for potential variable measures including: sex, age, marital status, race, ethnicity, previous or current diagnosis of mental illness, and familial history. This data will be analyzed utilizing the chi-square test enabling for sub-group analysis as needed. The purpose of this research is to determine if the prevalence of depression among PA students is significantly greater than that of the general population. By identifying the prevalence of depression in PA students there is potential to decrease burnout in the PA profession and prevent potential detrimental effects such as suicide.

Background Methamphetamine induced psychosis (MAP) and schizophrenia present with similar positive symptoms of psychosis, are characterized by evidence of attentional impairment, and show symptomatic response to treatment with dopamine antagonists. At the same time, MAP is considered a transient condition, while schizophrenia can be conceptualized as a neurodevelopmental disorder. Despite advances in the neurobiology of these two conditions, the extent to which their underlying attentional neurocircuitry show overlaps or differences has not often been directly compared. This thesis compared MAP, schizophrenia and healthy controls, in order to examine overlap and differences in 1) subcortical regulation of cortical inhibition and excitability, 2) resting state cortical and subcortical connectivity, and the dynamics of rhythmic neural activity between states, and 3) cortical-cortical connectivity using event related potential (ERP) responses to stimuli with a continuous performance task (CPT). Methods Outpatients treated for MAP and schizophrenia were recruited through hospitals and psychiatric institutions in the Western Cape. A final cohort of 24 MAP and 28 schizophrenia and 32 healthy control participants were included in the analyses for this thesis. For the cortical silent period (CSP) paradigm, the participant was asked to maintain isometric contraction between the thumb and index finger while TMS pulses at 120% and 140% resting motor threshold (RMT) were delivered to the primary motor cortical area corresponding to the abductor pollicis brevis (APB). Parameters extracted from CSP data included the latency to motor evoked potential (MEP), MEP amplitude and CSP duration. Electroencephalographs (EEGs) were performed with bilateral prefrontal, frontal, frontal temporal, central and parietal electrode locations. Relative EEG frequency power data were extracted from 3 stages during the EEG session, including states of eyes open, eyes closed, and during performance of the CPT. The CPT consisted of a series of random consonant letters. Participants were asked to respond to the letter "S" with a finger press only if it was the 3rd consecutive occurrence. ERP data were extracted and averaged from consecutive cues (S1 and S2), target (S3) and distractor (individual "S") stimuli in the CPT task. ERP data were analysed for group differences in N100, P200, N200 and P300 amplitudes and latencies at each electrode location with sufficient signal quality. Results In the CSP protocol, MAP and schizophrenia groups showed smaller MEP amplitudes at both 120% and 140% RMT stimulation levels in comparison to controls. Both MAP and schizophrenia groups had lower alpha and higher delta relative frequency band power, with schizophrenia showing significant differences from controls at more electrode positions than MAP. While controls demonstrated a decrease in alpha power between the eyes closed and eyes open resting states, this did not occur in MAP or schizophrenia. During the CPT, both MAP and schizophrenia achieved fewer correct targets and showed slower reaction times than healthy controls. In addition, MAP responded more often than the other two groups to the S2 stimulus (which required response inhixii Abstract bition). ERP analysis found smaller N100, larger P200, larger N200 and larger P300 amplitudes in MAP in response to stimuli requiring inhibition than in schizophrenia and controls, whereas schizophrenia showed longer P300 latencies in response to the target and distractor stimuli than in MAP and controls. Conclusions MEP results suggest that MAP and schizophrenia may have similar subcortical dysregulation, suggestive of altered dopaminergic regulation in the basal ganglia-thalamus-cortex loop. EEG frequency power results suggest that MAP and schizophrenia both display an inflexibility of subcortical systems involved in adaptation to environmental changes, suggesting deficiencies in the CT-TRN-TC loop in both MAP and schizophrenia. CPT performance and the pattern of ERP alterations in MAP suggests greater cholinergic impairment during attentional performance in MAP than in schizophrenia. Taken together, while there is considerable overlap in cortical-subcortical inhibition and connectivity in MAP and schizophrenia, there are also important differences; findings that emphasize both the similarities and dissimilarities that are seen clinically.