Major Depressive Disorder (MDD) is a common and debilitating mental illness primarily characterized by depressed mood and anhedonia, as well as a heterogenous mixture of other somatic symptoms. Existing therapies for MDD act primarily on monoamine transporters and receptors, with only partial success. A more recent advancement in depression treatment is the identification of the N-methyl-d-aspartate receptor (NMDAR) antagonist (R,S)-ketamine, which has shown efficacy in individuals with treatment resistant depression. Studies have raised the possibility of a mu opioid receptor (MOR)-dependent component to the actions of (R,S)- ketamine, through direct activation of MOR or indirect effects of NMDAR antagonism on endogenous opioids. Considering the ongoing global opioid epidemic, in which over prescribing of opioid painkillers and greater availability of illicit opioids has caused significant morbidity and mortality, studies suggesting an opioid component to ketamine’s actions have caused concern. We undertook the current experiments to better understand 1) the contribution of MORs to the behavioral effects of ketamine, and 2) how very low efficacy agonism of MOR may lead to a better separation between the desirable and undesirable effects of opioid drugs. Previous work from collaborators had established that a single dose of (R,S)-ketamine, specific hydroxynorketamine (HNK) metabolites, and the memantine derivative fluoroethylnormemantine (FENM), given one week before a contextual fear conditioning stressor could prevent the development of learned fear and behavioral despair. Moreover, some of these drugs also prevented the development of learned fear and behavioral despair when given shortly after the contextual fear conditioning stressor. We were interested in how (R,S)- ketamine’s reported MOR activity may contribute to these behavioral effects. We began by testing these compounds in cell-based signaling assays to determine their ability to directly activate MORs. We found that the parent (R)- and (S)-ketamine enantiomers could activate MORs with low potency in a high amplification G protein activation assay. In contrast, in low amplification miniG-based assays, the compounds tested showed little to no activation of MOR, suggesting that (R,S)-ketamine activates MOR with low potency and low efficacy. We moved to experiments in mice with the pseudo-irreversible MOR antagonist methocinnamox (MCAM) and attempted to block various behavioral effects of (R,S)-ketamine and the more selective NMDAR antagonist (FENM). We found that MCAM pretreatment blocked effects of (R,S)-ketamine on antinociception, behavioral despair, and learned fear, while other effects such as anesthesia and changes in hippocampal electrophysiology were not prevented by MCAM. With FENM the antinociceptive effects were less potent and less impacted by MCAM; however, the effects on behavioral despair and learned fear were still MORdependent. These results suggest that the protective effects of (R,S)-ketamine and FENM against stress may be an indirect effect of NMDAR antagonism on endogenous opioids. In a separate line of experiments, we used a range of doses of MCAM to antagonize the behavioral effects of morphine, the MOR agonist antidepressant tianeptine, and the MOR partial agonist 7-OH mitragynine. 7-OH mitragynine is a metabolite of mitragynine, the major alkaloid in kratom, which some reports suggest may have antidepressant and anxiolytic properties. Based on fundamentals of pharmacology, we hypothesized that inherent differences in the presence of “spare receptors” or receptor reserve between brain circuits could be revealed by differential inhibition by MCAM across behaviors and agonists. We assessed the inhibitory potency of MCAM against these drugs in tests of antinociception, hyperlocomotion, behavioral despair, respiration, and gastrointestinal motility. We found that MCAM pretreatment more potently inhibited the low efficacy agonist 7-OH mitragynine in the tests of antinociception, behavioral despair, and constipation. These data suggest that in circuits modulating antinociception, behavioral despair, and constipation, differences in receptor reserve likely facilitate the response to low efficacy agonists. However, our data also argue that the wider therapeutic window of G protein biased, low intrinsic efficacy MOR agonists is not solely a result of differences in the number of “spare” MORs regulating the effects of opioids in different circuits.

The majority of persons within South Africa (of whom 16% have claimed to have suffered from common mental disorders such as depression) that use medical treatment do so through primary care (Patel et al., 2007; Williams et al., 2007). However, studies have shown that general practitioners (GPs) often overlook, ignore, misdiagnose and even offer inappropriate treatment for mental illness (c.f. Lotrakul & Saipanish, 2009; Qwabe, 2009). Since South Africa is comprised of a multitude of diverse peoples from varying culture backgrounds, cultural diversity needs to be considered within every interaction, especially when GPs consult with individuals with depression. This study, thus, aims to explore GPs’ management of depression by investigating diagnosis or detection, treatment and referral patterns of GPs where their considerations of patient’s cultural worldviews are also investigated. This study utilised a semi-structured interview schedule on a convenient sample of six GP’s. Thematic content analysis was used to analyse salient themes from the data. Eight themes were found, namely: diagnosing, treating and referring patients with depression, cultural implications in general practice, training of GPs, the evolution of the medical field and its practices, disadvantaged communities and access to health care resources as well as the ethics of practice. These results are discussed in relation to local and international literature in the field.

This dissertation is an exploration of the putative mediators of complicated grief treatment (CGT) in an effort to gain a better understanding of the mechanisms by which the treatment exerts its effects. This three-paper dissertation utilizes data from an NIMH-funded randomized controlled trial of CGT (Shear et al., 2005), which showed that CGT is more effective than Interpersonal Psychotherapy (IPT) in reducing symptoms of complicated grief (CG). The first paper examines a broad range of ancillary outcomes including symptoms of anxiety, depression, complicated grief and sleep disturbance due to bad dreams. Antidepressant use is examined as a possible moderator since half the sample was taking antidepressants and those taking antidepressants had a marginally better response rate in CGT than those not taking them (59% vs. 42% in CGT and 40% vs. 19% in IPT). CGT was more effective than IPT in reducing cognitive symptoms of anxiety, depression as measured by the Hamilton Rating Scale for Depression (HRSD), somatic symptoms of depression, guilt/self-blame, negative thoughts about the future, avoidance and poor sleep due to bad dreams. The difference in treatment effect on the HRSD for CGT over IPT was more pronounced for participants not taking antidepressants where CGT reduced depression but IPT did not. Paper two examines possible mediators specific to the model of CGT including: guilt/self-blame specific to the death or deceased; negative thoughts about the future; avoidance of reminders of the loss; anxiety and depression (intense negative emotions). Antidepressants are also examined as a potential moderator to explore whether their use affects the mediating role of the identified variables. All of these variables emerged as either full or partial mediators of CGT. Antidepressant use had no effect on the mediating role of these variables. Paper three examines whether alliance (measured at week 4) predicts subsequent change in grief symptoms (controlling for early symptom change) and if so, whether it accounts for the difference in treatment effect between CGT and IPT (mediation). Working alliance emerged as a mediator of CGT, accounting for 28% of the treatment effect found between CGT/IPT and grief symptoms. Discussion sections for each paper summarize study findings, limitations and implications for future research.

Bridging the fields of engineering and psychiatry, this dissertation proposes a novel framework for the rational dosing of electric and magnetic seizure therapy, including electroconvulsive therapy (ECT) and magnetic seizure therapy (MST), for the treatment of psychiatric disorders such as medication resistant major depression and schizophrenia. The objective of this dissertation is to develop computational modeling tools that allow ECT and MST stimulation paradigms to be biophysically optimized ex vivo, prior to testing safety and efficacy in preclinical and clinical trials. Despite therapeutic advances, treatment resistant depression (TRD) remains a largely unmet clinical need. ECT is highly effective for TRD, but its side effects limit its real-world clinical utility. Modifications of treatment technique (e.g., electrode placement, stimulus parameters, novel paradigms such as MST) significantly improve the tolerability of convulsive therapy. However, we know relatively little about the distribution of the electric field (E-field) induced in the brain to inform spatial targeting of ECT and MST. Lacking an understanding of biophysical and physiological mechanisms, refinements in ECT/MST technique rely exclusively on time-consuming and costly clinical trials. Consequently, key questions remain unanswered about how to position the ECT electrodes or MST coil for targeted brain stimulation. Addressing this knowledge gap, this dissertation proposes a new platform that will inform an improved spatial targeting of ECT and MST through state-of-the-art computer simulations of the E-field distribution in human and nonhuman primate (NHP) brain. Part I of this dissertation aims to develop anatomically realistic finite element models of transcranial electric and magnetic stimulation in human and NHPs incorporating tissue heterogeneity and anisotropy derived from structural magnetic resonance imaging (MRI) and diffusion tensor imaging (DTI) data. The NHP models of ECT and MST are created alongside the human model since NHPs are used in preclinical studies on the mechanisms of seizure therapy. Part II of this dissertation aims to apply the model developed in Part I to electric and magnetic seizure therapy. We compute the strength and spatial distributions of the E-field induced in the brain by various ECT and MST paradigms. The relative E-field strength among various regions of interest (ROIs) is examined to select electrode/coil configurations that produce most focal stimulation of target ROIs that are considered to mediate the therapeutic action of ECT and MST. Since E-field alone is insufficient to account for individual differences in neurophysiological response, we calibrate the E-field maps relative to the neural activation threshold via in vivo measurements of the corticospinal tract response to single pulses (motor threshold, MT). We derive an empirical estimate of the neural activation threshold by coupling simulated E-field strength with individually measured MT. The E-field strength relative to an empirical neural activation threshold and corresponding volume of suprathreshold stimulation (focality) is examined to inform the selection of ECT and MST stimulus pulse amplitude that will result in focal ROI stimulation. We contrast the ECT/MST stimulation strength and focality with conventional fixed and individually titrated pulse amplitude necessary to induce a seizure (seizure threshold, ST) to study pulse amplitude adjustment as a novel means of controlling stimulation strength and focality. This work provides a basis for rational dosing of seizure therapies that could help improve their risk/benefit ratio and guide the development of safer alternatives for patients with severe psychiatric disorders.

Introduction: In the US, Black/African Americans comprise the largest proportion of People Living with HIV (PLWH). Depressive symptoms and fatigue are highly prevalent among PLWH. Depressive symptoms are linked to progression of HIV disease, and fatigue is linked to severity of depressive symptoms. Resistance exercise is known to have psychological benefits in non-HIV depressed populations, and these benefits are hypothesized to be intensity-dependent. That is, moderate intensity exercise seems to stimulate better psychological outcomes compared to other exercise intensities. To date, no study has examined the acute psychological effects of resistance exercise intensity with depressed PLWH. Purpose: To test the acute effects of resistance exercise intensity on affect, perceived activation, and perceived distress among sedentary Black/African American PLWH who experience depressive symptoms. Methods: Twenty-five men and 17 women ages 24-66 (47.5±11.2) with a Center for Epidemiologic Studies Depression Scale score of ≥10 completed a battery of questionnaires and 10-repetition maximum (10-RM) muscular strength tests. Participants were randomized into a moderate intensity resistance exercise group (i.e., 70% of 10-RM), n=21, or vigorous intensity resistance exercise group (i.e., 100% of 10-RM), n=21. Both groups completed 3 sets of 10 repetitions for 5 exercises at the assigned intensity. Exercises include squat, chest press, lat pull-down, dumbbell shoulder press, and dumbbell biceps curl. Affect, perceived activation, and perceived distress were measured with the Feeling Scale, Felt Arousal Scale, and Subjective Units of Distress Scale, respectively. Measures were administered at PRE, MID, POST, at 15-minute DELAY, and at 30-minute DELAY. Changes were analyzed using repeated measures ANOVA, with Bonferroni adjustments for post-hoc analyses. Results: There were significant Time x Group interactions for affect (F=2.6, p=0.03, η2=0.1), and perceived distress (F=5.5, p<0.01, η2=0.1), and a main effect of Time for perceived activation (F=16.2, p<0.001, η2=0.3). In the moderate intensity group, affect improved PRE to POST (t=3.2, p<0.01, d=0.7), PRE to DELAY 15 (t=4.1, p<0.01, d=0.9), and PRE to DELAY 30 (t=4.1, p<0.001, d=0.7). In addition, perceived distress was reduced from PRE to MID (t=4.2, p<0.001, d=0.9), PRE to POST (t=5.5, p<0.001, d=1.2), PRE to DELAY 15 (t=5.6, p<0.001, d=1.2), and PRE to DELAY 30 (t=6.7, p<0.001, d=1.5). In the vigorous intensity group, affect declined PRE to MID (t=2.9, p<0.01, d=-0.6), while perceived distress improved PRE to DELAY 15 (t=4.8, p<0.001, d=1.0) and PRE to DELAY 30 (t=3.5, p<0.01, d=0.7). Perceived activation increased in both groups similarly PRE to MID (t=5.1, p<0.01, d=1.5), and PRE to POST (t=6.1, p<0.001, d=1.8). Conclusions: Results suggest that an acute bout of moderate intensity resistance exercise is more effective than vigorous intensity resistance exercise in improving affect, increasing energy, and reducing distress in depressed Black/African American PLWH. However, vigorous intensity resistance exercise also appears to have distress-reducing benefits, but this appears to only occur after exercise. These findings should be considered when prescribing exercise for depressive symptom management in this population.

The focus of the research study is to explore chronically ill African American, Latino, and White patients' causal attributions of symptoms of depression and factors that predict depression care treatment preferences among these groups. Research has demonstrated that perception of illness impacts what treatments a person will deem appropriate for their mental health problems and from whom they will seek treatment. Research also indicates that certain ethnic groups are more likely to seek treatment for their symptoms of depression in the primary care setting. Yet, it is unclear how they actually perceive their symptoms and what best predicts the treatments that they are likely to consider acceptable. A convenient sample of 109 HIV+ adults, 79 diabetic adults, and 3 adults with both conditions were recruited for this study. Participants had to be receiving services for either HIV, diabetes, or both conditions in one of the three central Austin facilities and be a representative from one of three racial/ethnic groups: African Americans, Latino, and White. Differences were found across ethnicity with regard to causal beliefs and treatment preferences for the symptoms of depression both among the HIV and the diabetic subgroups. Latinos in both groups were more likely than Whites to prefer counseling or a single form of treatment over combined treatment methods. Diabetic Latinos were more likely to prefer counseling for symptoms of depression. HIV seropositive individuals who reported the least number of symptoms of physical illness were more likely to attribute their symptoms of depression to stressful life events, whereas those who reported the greatest number of symptoms of physical illness were more likely to attribute their symptoms of depression to their medical illness. Additionally among the HIV subgroup, individuals who reported high stress tended to predict the preferences for treatment provided by a psychiatrist/psychologist and Whites scored highest on this factor. Finally, differences in depression scores across race/ethnicity were also revealed. The utility of assessing a patient's understanding of symptoms of depression in order to determine how personal illness models impact treatment preferences and knowledge of patient's causal attributions can aid medical social workers and physicians in collaborative management of chronic illness and depression are discussed.

xvii, 93 p. : ill. (some col.) / Since Rosenzweig's "Dodo Bird Verdict" in 1936, the "common" versus "specific" factors debate has continued to polarize the field of psychotherapy. Treatment expectancy is an important but often overlooked common factor. The current study investigated the role of treatment expectancy in the Treatment of Adolescents with Depression Study (TADS). Four-hundred three adolescents ( M age =14.62, SD =1.56) filled out the Treatment Expectancy for Adolescents (TEA) measure prior to treatment randomization to one of four treatments: fluoxetine (FLX), cognitive behavior therapy (CBT), their combination (COMB), and placebo (PBO). Adolescents randomized to CBT or COMB also filled out the CBT Rationale Acceptance and Expectation for Improvement (C-RAEI) form during their second session of CBT. Before finding out their treatment assignments, adolescents endorsed higher treatment expectancies for COMB than CBT and medication only. Family income levels below $75,000 and higher levels of depression severity, hopelessness, and suicidality were associated with lower expectations for improvement with CBT. The presence of a comorbid anxiety disorder diagnosis was associated with lower expectations for medication without CBT. Separate random coefficients and logistic regression models identified treatment expectancy as a predictor of outcome for three primary outcome measures in TADS, irrespective of treatment assignment. Severity of depression moderated this relationship; mild to moderately depressed adolescents appeared to be more sensitive to the effects of treatment expectancy than marked to severely depressed adolescents. The opposite results were found for the self-rated outcome measure in TADS based on the C-RAIE. For marked to severely depressed adolescents assigned to CBT or COMB, acceptance of treatment rationale and expectancy for improvement were associated with treatment response. These results suggest that treatment expectancy is an important common factor of treatment for mild to moderately depressed adolescents prior to treatment initiation, although it may be especially important for initially skeptical, marked to severely depressed adolescents to "buy in" to treatment after treatment initiation. Treatment effects were still found after controlling for the effects of treatment expectancy on outcome. It seems that both the "common" factor of treatment expectancy and the "specific" factor of treatment assignment contributed to outcome in TADS. / Committee in charge: Anne D. Simons, Chair; Gordon Nagayama Hall, Member; Holly Arrow, Member; Jeffrey Todahl, Outside member

The present study sought to more rigourously explore outcomes of psychological well-being immediately following a psychotherapeutic yoga class. Specifically, the study hypothesized improvements in state anxiety and subjective well-being as well as an observable relationship between state and trait mindfulness following a yoga intervention, all while controlling for differences between yoga instructors, prior yoga experience, and participant endorsements of psychological symptoms. Previous yoga experience was not found to be a significant factor in any of the tested hypotheses. Findings revealed that psychotherapeutic yoga decreased anxiety and increased subjective well-being, even after controlling for therapist variability, prior yoga experience, and client diagnosis. Results also indicate differential impacts on decreased anxiety and increased subjective well-being by class instructor. This is the first study to examine outcomes of an ongoing yoga based-practices in the naturalistic setting of an outpatient counseling center while rigorously controlling for confounding factors (e.g. therapist variability). Methodological and statistical limitations are discussed in depth, and future directions to improve on this study and clarify the present findings are emphasized.

Heterogeneous treatment responses are commonly observed in patients with mental disorders. Thus, a universal treatment strategy may not be adequate, and tailored treatments adapted to individual characteristics could improve treatment responses. The theme of the dissertation is to develop statistical and machine learning methods to address patients heterogeneity and derive robust and generalizable individualized treatment strategies by integrating evidence from multi-domain data and multiple studies to achieve precision medicine. Unique challenges arising from the research of mental disorders need to be addressed in order to facilitate personalized medical decision-making in clinical practice. This dissertation contains four projects to achieve these goals while addressing the challenges: (i) a statistical method to learn dynamic treatment regimes (DTRs) by synthesizing independent trials over different stages when sequential randomization data is not available; (ii) a statistical method to learn optimal individualized treatment rules (ITRs) for mental disorders by modeling patients' latent mental states using probabilistic generative models; (iii) an integrative learning algorithm to incorporate multi-domain and multi-treatment-phase measures for optimizing individualized treatments; (iv) a statistical machine learning method to optimize ITRs that can benefit subjects in a target population for mental disorders with improved learning efficiency and generalizability. DTRs adaptively prescribe treatments based on patients' intermediate responses and evolving health status over multiple treatment stages. Data from sequential multiple assignment randomization trials (SMARTs) are recommended to be used for learning DTRs. However, due to the re-randomization of the same patients over multiple treatment stages and a prolonged follow-up period, SMARTs are often difficult to implement and costly to manage, and patient adherence is always a concern in practice. To lessen such practical challenges, in the first part of the dissertation, we propose an alternative approach to learn optimal DTRs by synthesizing independent trials over different stages without using data from SMARTs. Specifically, at each stage, data from a single randomized trial along with patients' natural medical history and health status in previous stages are used. We use a backward learning method to estimate optimal treatment decisions at a particular stage, where patients' future optimal outcome increment is estimated using data observed from independent trials with future stages' information. Under some conditions, we show that the proposed method yields consistent estimation of the optimal DTRs, and we obtain the same learning rates as those from SMARTs. We conduct simulation studies to demonstrate the advantage of the proposed method. Finally, we learn DTRs for treating major depressive disorder (MDD) by stage-wise synthesis of two randomized trials. We perform a validation study on independent subjects and show that the synthesized DTRs lead to the greatest MDD symptom reduction compared to alternative methods. The second part of the dissertation focuses on optimizing individualized treatments for mental disorders. Due to disease complexity, substantial diversity in patients' symptomatology within the same diagnostic category is widely observed. Leveraging the measurement model theory in psychiatry and psychology, we learn patient's intrinsic latent mental status from psychological or clinical symptoms under a probabilistic generative model, restricted Boltzmann machine (RBM), through which patients' heterogeneous symptoms are represented using an economic number of latent variables and yet remains flexible. These latent mental states serve as a better characterization of the underlying disorder status than a simple summary score of the symptoms. They also serve as more reliable and representative features to differentiate treatment responses. We then optimize a value function defined by the latent states after treatment by exploiting a transformation of the observed symptoms based on the RBM without modeling the relationship between the latent mental states before and after treatment. The optimal treatment rules are derived using a weighted large margin classifier. We derive the convergence rate of the proposed estimator under the latent models. Simulation studies are conducted to test the performance of the proposed method. Finally, we apply the developed method to real-world studies. We demonstrate the utility and advantage of our method in tailoring treatments for patients with major depression and identify patient subgroups informative for treatment recommendations. In the third part of the dissertation, based on the general framework introduced in the previous part, we propose an integrated learning algorithm that can simultaneously learn patients' underlying mental states and recommend optimal treatments for each individual with improved learning efficiency. It allows incorporation of both the pre- and post-treatment outcomes in learning the invariant latent structure and allows integration of outcome measures from different domains to characterize patients' mental health more comprehensively. A multi-layer neural network is used to allow complex treatment effect heterogeneity. Optimal treatment policy can be inferred for future patients by comparing their potential mental states under different treatments given the observed multi-domain pre-treatment measurements. Experiments on simulated data and real-world clinical trial data show that the learned treatment polices compare favorably to alternative methods on heterogeneous treatment effects and have broad utilities which lead to better patient outcomes on multiple domains. The fourth part of the dissertation aims to infer optimal treatments of mental disorders for a target population considering the potential distribution disparities between the patient data in a study we collect and the target population of interest. To achieve that, we propose a learning approach that connects measurement theory, efficient weighting procedure, and flexible neural network architecture through latent variables. In our method, patients' underlying mental states are represented by a reduced number of latent state variables allowing for incorporating domain knowledge, and the invariant latent structure is preserved for interpretability and validity. Subject-specific weights to balance population differences are constructed using these compact latent variables, which capture the major variations and facilitate the weighting procedure due to the reduced dimensionality. Data from multiple studies can be integrated to learn the latent structure to improve learning efficiency and generalizability. Extensive simulation studies demonstrate consistent superiority of the proposed method and the weighting scheme to alternative methods when applying to the target population. Application of our method to real-world studies is conducted to recommend treatments to patients with major depressive disorder and has shown a broader utility of the ITRs learned from the proposed method in improving the mental states of patients in the target population.

Depression is a major public health problem. Decades of research have been conducted to create a classification system aligned with the complex phenomenological features of depression. The dominant classification system for depression is the latent paradigm, which conceptualizes observable symptoms of depression as effects of an underlying disorder. There is increasing evidence, however, that the latent model is inadequate to inform the prognosis and treatment of depression. Specifically, evidence is accumulating that symptoms of depression do not necessarily arise due to an underlying condition, but that symptoms occur as a network in which each one is causally related to a previous symptom. This dissertation critically evaluated the underlying assumptions of this “network paradigm,” one of the frameworks which had been proposed as an alternative to the traditional latent paradigm, as an appropriate model for studying depression. The first chapter systematically evaluated empirical depression network studies regarding whether the study design included an examination of the paradigm’s assumptions. In the second chapter, I investigated the relationships among depressive symptoms and determined whether causal relationships among depressive symptoms, a key assumption underlying this paradigm, could be a plausible explanation. The last chapter investigated a central controversy within the network literature regarding consistent findings and measurement error. The first chapter found that the majority of depression network studies published in the literature were not capable of providing empirical support of symptom causal relationships and often neglected to investigate the impact of measurement error. The second chapter estimated a significant relationship between two depressive symptoms - sadness and anhedonia, using an inverse probability treatment-weighted regression estimation approach in the context of longitudinal data. Causal relationships among symptoms, a key assumption underlying the network paradigm, may be a plausible explanation for the depressive symptom relationships. The third chapter found that statistical network models are not robust to measurement error through a series of simulation studies. Measurement error remained a general threat against the network paradigm, and existing network findings should be interpreted with caution. Overall, the network paradigm may be appropriate for study depression, but existing findings should be interpreted with caution. There is a need to explore the fundamental assumptions of paradigms prior to widespread application.