High dimensional predictions of suicide risk in 4.2 million US Veterans using ensemble transfer learning.

We present an ensemble transfer learning method to predict suicide from Veterans Affairs (VA) electronic medical records (EMR). A diverse set of base models was trained to predict a binary outcome constructed from reported suicide, suicide attempt, and overdose diagnoses with varying choices of study design and prediction methodology. Each model used twenty cross-sectional and 190 longitudinal variables observed in eight time intervals covering 7.5 years prior to the time of prediction. Ensembles of seven base models were created and fine-tuned with ten variables expected to change with study design and outcome definition in order to predict suicide and combined outcome in a prospective cohort. The ensemble models achieved c-statistics of 0.73 on 2-year suicide risk and 0.83 on the combined outcome when predicting on a prospective cohort of [Formula: see text] 4.2 M veterans. The ensembles rely on nonlinear base models trained using a matched retrospective nested case-control (Rcc) study cohort and show good calibration across a diversity of subgroups, including risk strata, age, sex, race, and level of healthcare utilization. In addition, a linear Rcc base model provided a rich set of biological predictors, including indicators of suicide, substance use disorder, mental health diagnoses and treatments, hypoxia and vascular damage, and demographics.

Deep sequential neural network models improve stratification of suicide attempt risk among US veterans.

OBJECTIVE: To apply deep neural networks (DNNs) to longitudinal EHR data in order to predict suicide attempt risk among veterans. Local explainability techniques were used to provide explanations for each prediction with the goal of ultimately improving outreach and intervention efforts. MATERIALS AND METHODS: The DNNs fused demographic information with diagnostic, prescription, and procedure codes. Models were trained and tested on EHR data of approximately 500 000 US veterans: all veterans with recorded suicide attempts from April 1, 2005, through January 1, 2016, each paired with 5 veterans of the same age who did not attempt suicide. Shapley Additive Explanation (SHAP) values were calculated to provide explanations of DNN predictions. RESULTS: The DNNs outperformed logistic and linear regression models in predicting suicide attempts. After adjusting for the sampling technique, the convolutional neural network (CNN) model achieved a positive predictive value (PPV) of 0.54 for suicide attempts within 12 months by veterans in the top 0.1% risk tier. Explainability methods identified meaningful subgroups of high-risk veterans as well as key determinants of suicide attempt risk at both the group and individual level. DISCUSSION AND CONCLUSION: The deep learning methods employed in the present study have the potential to significantly enhance existing suicide risk models for veterans. These methods can also provide important clues to explore the relative value of long-term and short-term intervention strategies. Furthermore, the explainability methods utilized here could also be used to communicate to clinicians the key features which increase specific veterans' risk for attempting suicide.

Prioritizing vaccination based on analysis of community networks.

Many countries that had early access to COVID-19 vaccines implemented vaccination strategies that prioritized health care workers and the elderly. As barriers to access eased, vaccine prioritization strategies have been relaxed. However, these strategies are still an important tool for decision makers to manage new variants, plan for future booster shots, or stage mass vaccinations. This paper explores the impact of vaccine prioritization strategies using networks that represent communities with different demographics and connectivity. The impact of vaccination is compared to non-medical intervention to reduce transmission. Several sources of uncertainty are considered, including vaccine willingness and mask effectiveness. This paper finds that while prioritization strategies can have a large impact on reducing deaths and peak hospitalization, selecting the best strategy depends on community characteristics and the desired objective. Additionally, in some cases random vaccination performs as well as more targeted prioritization strategies. Understanding these trade-offs is important when planning vaccine distribution.

Impact of an auditory hallucination simulation coupled with a speaker diagnosed with schizophrenia on mental illness stigma in pharmacy students.

INTRODUCTION: Hallucination simulations improve student empathy but increase desired social distance from individuals with schizophrenia, while direct contact reduces social distance. This study describes the implementation of combining an auditory hallucination simulation with a speaker diagnosed with schizophrenia and its impact on mental illness stigma. METHODS: Pharmacy students in their last year of didactic instruction (N = 346) attended a presentation by a speaker diagnosed with schizophrenia then participated in a hallucination simulation. Mental illness stigma was measured before and after the intervention using the Opening Minds Survey for Health Care Professionals (OMS-HC). Related-samples Wilcoxon-signed rank tests were used to evaluate changes in OMS-HC scores. An inductive qualitative analysis was conducted on student perceptions of patients with psychosis. RESULTS: OMS-HC total scores were reduced by an average of 2.0 ± 5.6 (P = .005) for the first-year pilot and 2.3 ± 7.0 (P < .001) for the subsequent year. OMS-HC attitudes (P = .005) and disclosure/help-seeking (P < .005) subscales decreased both years. There was no significant change in the social distance subscale (P = .205) the first year and a significant decrease (P = .015) the second year. The themes identified from the open-ended comments were increased awareness, sympathy, empathy, inspiration/admiration, discomfort, and change to patient care. CONCLUSIONS: The combination of a speaker with schizophrenia and auditory hallucination simulation effectively reduced mental illness stigma. The combination is likely effective for reducing stigmatizing attitudes and willingness to disclose mental illness with variable reduction in desired social distance.

Analysis of mobility data to build contact networks for COVID-19.

As social distancing policies and recommendations went into effect in response to COVID-19, people made rapid changes to the places they visit. These changes are clearly seen in mobility data, which records foot traffic using location trackers in cell phones. While mobility data is often used to extract the number of customers that visit a particular business or business type, it is the frequency and duration of concurrent occupancy at those sites that governs transmission. Understanding the way people interact at different locations can help target policies and inform contact tracing and prevention strategies. This paper outlines methods to extract interactions from mobility data and build networks that can be used in epidemiological models. Several measures of interaction are extracted: interactions between people, the cumulative interactions for a single person, and cumulative interactions that occur at particular businesses. Network metrics are computed to identify structural trends which show clear changes based on the timing of stay-at-home orders. Measures of interaction and structural trends in the resulting networks can be used to better understand potential spreading events, the percent of interactions that can be classified as close contacts, and the impact of policy choices to control transmission.

Combining Antidepressants with ß-Blockers: Evidence of a Clinically Significant CYP2D6 Drug Interaction.

BACKGROUND: The ß-blockers and antidepressants are two of the most commonly prescribed drug classes in the United States. Several antidepressants are potent inhibitors of cytochrome P450 2D6 liver enzymes (CYP2D6) and can increase the plasma concentrations of certain ß-blockers when administered concomitantly, potentially leading to serious medical consequences such as hypotension, bradycardia, and falls. OBJECTIVE: The primary objective of this investigation was to determine whether initiating an antidepressant in patients receiving ß-blockers increased the risk of hemodynamic adverse events. Our primary outcome was time to hospital admissions or emergency department (ED) visits for an International Classification of Diseases-9 diagnosis suggestive of excessive ß-blockade. METHODS: We conducted a survival analysis for adults continuously enrolled in the California Medicaid system (Medi-Cal) between 2004 and 2012. Eligible patients were required to be receiving ß-blocker medications that are primarily CYP2D6 substrates (e.g., metoprolol, propranolol, or carvedilol). Univariate and multivariable analyses were performed for patients who concurrently received antidepressants with ß-blockers. An additional multivariable analysis analyzed the association of this combination upon hospitalizations or ED visits for all causes. RESULTS: A total of 21,292 beneficiaries met the inclusion criteria, and 4.3% of patients required hospitalization or ED visits within 30 days of co-medication. In multivariable analysis, patients receiving antidepressants with moderate to strong CYP2D6 inhibitory potential (fluoxetine, paroxetine, duloxetine, or bupropion) had a greater risk for hospitalization or ED visits for hemodynamic events than those initiated on antidepressants with weak CYP2D6 inhibition for 30 days or less when each was compared with patients receiving no antidepressants (hazard ratio [HR] 1.53, 95% confidence interval [CI] 1.03-2.81; p=0.04 vs HR 1.24; 95% CI 0.82-1.88; p=0.30). Other demographic variables associated with increased morbidity included advanced age, male sex, higher ß-blocker doses, and African American race or Hispanic ethnicity. CONCLUSIONS: Results of this analysis suggest that initiation of certain antidepressants was associated with an increased risk for serious medical sequelae among patients concurrently receiving ß-blockers. Greater risk was observed with antidepressants that potently inhibit the CYP2D6 enzyme, implying that increased morbidity may be mediated by a metabolic drug interaction.

Association between Use of Methadone, Other Central Nervous System Depressants, and QTc Interval-Prolonging Medications and Risk of Mortality in a Large Cohort of Women Living with or at Risk for Human Immunodeficiency Virus Infection.

STUDY OBJECTIVE: To evaluate the association between use of methadone, other central nervous system (CNS) depressants, and QTc interval-prolonging medications and risk of mortality among human immunodeficiency virus (HIV)-infected and at-risk HIV-uninfected women. DESIGN: Multicenter, prospective, observational cohort study (Women's Interagency HIV Study [WIHS]). PARTICIPANTS: A total of 4150 women enrolled in the WIHS study between 1994 and 2014 who were infected (3119 women) or not infected (1031 women) with HIV. MEASUREMENTS AND MAIN RESULTS: Data on medication utilization were collected from all study participants via interviewer-administered surveys at 6-month intervals (1994-2014). Mortality was confirmed by National Death Index data. With age defining the time scale for the analysis, Cox proportional hazards models were used to estimate hazard ratios (HRs) for all-cause mortality in HIV-infected and -uninfected women and non-acquired immunodeficiency syndrome (AIDS) deaths in HIV-infected women. A total of 1046 deaths were identified, of which 429 were considered non-AIDS deaths. Use of benzodiazepines, CNS depressants (excluding methadone), and number of medications with conditional QTc interval-prolonging effects were each associated with all-cause mortality in multivariate models of HIV-infected women: hazard ratio (HR) 1.28, 95% confidence interval (CI) 1.01-1.60, p=0.037; HR 1.61, 95% CI 1.35-1.92, p<0.0001; and HR 1.15 per drug, 95% CI 1.00-1.33, p=0.047, respectively. Other explanatory variables for all-cause mortality in this model included HIV viral load, CD4+  cell count, renal function, hemoglobin and albumin levels, HIV treatment era, employment status, existence of depressive symptoms, ever use of injection drugs, and tobacco smoking. Of interest, use of CNS depressants (excluding methadone) was also associated with non-AIDS deaths (HR 1.49, 95% CI 1.49-2.2, p<0.0001). Although use of benzodiazepines and conditional QT interval-prolonging medications were associated with increased risk of non-AIDS mortality (HR 1.32 and 1.25, respectively), the effect was not statistically significant (p>0.05). CONCLUSION: In this cohort of HIV-infected and at-risk HIV-uninfected women, use of benzodiazepines, CNS depressants, and conditional QTc interval-prolonging medications were associated with a higher risk of mortality independent of methadone and other well-recognized mortality risk factors. Care must be taken to assess risk when prescribing these medications in this underserved and at-risk patient population.

Chief complaint classification with recurrent neural networks.

Syndromic surveillance detects and monitors individual and population health indicators through sources such as emergency department records. Automated classification of these records can improve outbreak detection speed and diagnosis accuracy. Current syndromic systems rely on hand-coded keyword-based methods to parse written fields and may benefit from the use of modern supervised-learning classifier models. In this paper, we implement two recurrent neural network models based on long short-term memory (LSTM) and gated recurrent unit (GRU) cells and compare them to two traditional bag-of-words classifiers: multinomial naïve Bayes (MNB) and a support vector machine (SVM). The MNB classifier is one of only two machine learning algorithms currently being used for syndromic surveillance. All four models are trained to predict diagnostic code groups as defined by Clinical Classification Software, first to predict from discharge diagnosis, and then from chief complaint fields. The classifiers are trained on 3.6 million de-identified emergency department records from a single United States jurisdiction. We compare performance of these models primarily using the F1 score, and we measure absolute model performance to determine which conditions are the most amenable to surveillance based on chief complaint alone. Using discharge diagnoses, the LSTM classifier performs best, though all models exhibit an F1 score above 96.00. Using chief complaints, the GRU performs best (F1 = 47.38), and MNB with bigrams performs worst (F1 = 39.40). We also note that certain syndrome types are easier to detect than others. For example, chief complaints using the GRU model predicts alcohol-related disorders well (F1 = 78.91) but predicts influenza poorly (F1 = 14.80). In all instances, the RNN models outperformed the bag-of-words classifiers suggesting deep learning models could substantially improve the automatic classification of unstructured text for syndromic surveillance.

Evaluation of the Impact of a Multimodel Intervention on Prescribing Patterns of Sedative-Hypnotics in a Behavioral Health System.

OBJECTIVE: To describe the effect of a multimodal intervention targeting chronic benzodiazepine and sedative-hypnotic prescriptions in a large behavioral health system. METHODS: This retrospective study analyzed chronic sedative-hypnotic prescription rates in patients seeking mental health services in a large behavioral health system. The multimodal intervention consisted of provider education, coordination of care with all providers involved in patient care, and guideline development and implementation for safe prescribing of sedative-hypnotics. Three time periods were analyzed: preintervention (October 2013-December 2013), 12-month assessment (October 2014-December 2014), and 24-month assessment (October 2015-December 2015). The primary outcome of the study was the change in frequency of chronic (≥ 60 days) sedative-hypnotic prescriptions received before and after the multimodal intervention. The secondary outcome included the change in prescription rates in a priori-defined cohorts: patients on methadone maintenance therapy and patients ≥ 60 years of age. RESULTS: There were 32,944 prescriptions during the study period. The rate of chronic sedative-hypnotic prescriptions decreased from 1,764 (15.3%) to 1,634 (14.9%) to 1,018 (9.8%) between the 3 assessment periods, respectively. A significant decrease occurred between the preintervention period and 24-month assessment (5.5%, P < .0001) and between the 12-month assessment and the 24-month assessment (5.1%, P < .0001). In the elderly, prescription rates decreased significantly between the preintervention period and the 24-month assessment (3.6%, P < .0001) and the 12-month and 24-month assessments (3.2%, P = .0001). In patients in methadone maintenance programs, rates of concomitant chronic sedative-hypnotic prescriptions decreased significantly between all assessment periods: preintervention and 12-month assessment (8.2%, P = .001), 12-month and 24-month assessment (6.3%, P = .002), and preintervention and 24-month assessment (14.5%, P < .0001). Additionally, prescription rates increased significantly for antidepressants (4.1%, P < .05), hydroxyzine (1.1%, P = .01), buspirone (2.1%, P < .05), gabapentin (6.3%, P < .05), and melatonin agonists (0.3%, P < .05) between the preintervention and 24-month assessment periods. CONCLUSIONS: Implementation of a multimodal intervention led to a significant decrease in rates of chronic sedative-hypnotic prescriptions in a large behavioral health system.

A Practitioner-Driven Research Agenda for Syndromic Surveillance.

Syndromic surveillance has expanded since 2001 in both scope and geographic reach and has benefited from research studies adapted from numerous disciplines. The practice of syndromic surveillance continues to evolve rapidly. The International Society for Disease Surveillance solicited input from its global surveillance network on key research questions, with the goal of improving syndromic surveillance practice. A workgroup of syndromic surveillance subject matter experts was convened from February to June 2016 to review and categorize the proposed topics. The workgroup identified 12 topic areas in 4 syndromic surveillance categories: informatics, analytics, systems research, and communications. This article details the context of each topic and its implications for public health. This research agenda can help catalyze the research that public health practitioners identified as most important.

Drug Interactions with Lithium: An Update.

Lithium has been used for the management of psychiatric illnesses for over 50 years and it continues to be regarded as a first-line agent for the treatment and prevention of bipolar disorder. Lithium possesses a narrow therapeutic index and comparatively minor alterations in plasma concentrations can have significant clinical sequelae. Several drug classes have been implicated in the development of lithium toxicity over the years, including diuretics and non-steroidal anti-inflammatory compounds, but much of the anecdotal and experimental evidence supporting these interactions is dated, and many newer medications and medication classes have been introduced during the intervening years. This review is intended to provide an update on the accumulated evidence documenting potential interactions with lithium, with a focus on pharmacokinetic insights gained within the last two decades. The clinical relevance and ramifications of these interactions are discussed.

Modeling Evacuation of a Hospital without Electric Power.

Hospital evacuations that occur during, or as a result of, infrastructure outages are complicated and demanding. Loss of infrastructure services can initiate a chain of events with corresponding management challenges. This report describes a modeling case study of the 2001 evacuation of the Memorial Hermann Hospital in Houston, Texas (USA). The study uses a model designed to track such cascading events following loss of infrastructure services and to identify the staff, resources, and operational adaptations required to sustain patient care and/or conduct an evacuation. The model is based on the assumption that a hospital's primary mission is to provide necessary medical care to all of its patients, even when critical infrastructure services to the hospital and surrounding areas are disrupted. Model logic evaluates the hospital's ability to provide an adequate level of care for all of its patients throughout a period of disruption. If hospital resources are insufficient to provide such care, the model recommends an evacuation. Model features also provide information to support evacuation and resource allocation decisions for optimizing care over the entire population of patients. This report documents the application of the model to a scenario designed to resemble the 2001 evacuation of the Memorial Hermann Hospital, demonstrating the model's ability to recreate the timeline of an actual evacuation. The model is also applied to scenarios demonstrating how its output can inform evacuation planning activities and timing.

Modeling hospitals' adaptive capacity during a loss of infrastructure services.

Resilience in hospitals - their ability to withstand, adapt to, and rapidly recover from disruptive events - is vital to their role as part of national critical infrastructure. This paper presents a model to provide planning guidance to decision makers about how to make hospitals more resilient against possible disruption scenarios. This model represents a hospital's adaptive capacities that are leveraged to care for patients during loss of infrastructure services (power, water, etc.). The model is an optimization that reallocates and substitutes resources to keep patients in a high care state or allocates resources to allow evacuation if necessary. An illustrative example demonstrates how the model might be used in practice.

Detection and treatment rates for perinatal depression in a state Medicaid population.

BACKGROUND: The purpose of this investigation was to assess detection and treatment rates for perinatal depression among women enrolled in the California State Medicaid (Medi-Cal) program in comparison to female beneficiaries of reproductive age who did not give birth during the same study period. METHODS: Investigators conducted a retrospective longitudinal cohort analysis of women between the ages of 18 and 39 years old who were continuously enrolled in the Medi-Cal fee-for-service program between January 2006 and December 2009. The perinatal cohort consisted of women with evidence of a live birth occurring between October 2007 and March 2009. The control cohort consisted of women in the same age group and health plan without evidence of pregnancy during this time frame. The primary outcome of this investigation was diagnosis of depression during 3 contiguous 9-month time frames: immediately prior to presumed conception, during pregnancy, and throughout the postpartum period. Secondary outcomes included within-group and cohort comparisons of treatment patterns (antidepressant or psychotherapy). A multivariable analysis of demographic factors predicting depression diagnosis or treatment was conducted as well. RESULTS: A total of 6030 women was identified in the perinatal cohort, and 56,709 women were included in the control group. The perinatal cohort was significantly less likely than nonpregnant controls to receive a diagnosis of depression both during pregnancy (prevalence=1.6% vs 3.5%; OR=0.45; 95% CI=0.35-0.55) and postpartum (2.2% vs 3.6%; OR=0.59; 95% CI=0.50-0.71). Similar differences were noted in antidepressant prescribing patterns apparent during these 2 time frames. A subgroup analysis of women who received a depression diagnosis revealed that only 48% of the perinatal cohort was provided any treatment during pregnancy (vs 72% of the control group; p<0.0001) or postpartum (57% vs 73%; p<0.0001). Specific demographic factors predicting a lower prevalence of depression detection or treatment included Hispanic descent, age <25 years, or primary residence in an rural setting. CONCLUSIONS: Depression was often overlooked and undertreated among women who are pregnant or postpartum in comparison to services delivered to similar nonpregnant controls. Significant disparities in the healthcare received by certain subpopulations of perinatal women suggest that research into barriers to care and subsequent interventions are warranted.

A pilot study evaluating genetic and environmental factors for postpartum depression.

OBJECTIVE: To assess the influence of genetic and environmental risk factors upon postpartum depression. DESIGN: Case-control, prospective study. SETTING: The University of California at San Francisco Obstetric and Gynecology Clinic. PARTICIPANTS: Mothers screened for postpartum depression six weeks after delivery with the Edinburgh Postnatal Depression Scale and recruited as cases and controls. MEASUREMENTS: Eligible subjects completed a series of assessments and a structured clinical interview to confirm diagnosis of depression. Deoxyribonucleic acid was obtained for genotyping of 81 single nucleotide polymorphisms in 12 genes hypothesized to be postpartum depression-related. RESULTS: Twenty-four cases and 24 controls were eligible for analysis. Three single necleotide polymorphisms in the serotonin 2A receptor (HTR2A) gene were associated with postpartum depression. The strongest association at a functional promoter polymorphism (rs6311), a functional promoter single nucleotide polymorphisms (p=0.002, odds ratio 0.25, 95% confidence interval:0.10-0.63), was a finding robust to population stratification. Gene-wide association was significant for HTR2A (permuted p=0.008), but not when corrected for all 12 genes. Analysis of demographic and psychosocial risk factors identified distressed relationship, unplanned pregnancy, and a previous history of depression as significant predictive variables (p≤0.05). CONCLUSIONS: This pilot data suggests deoxyribonucleic acid variations in HTR2A may be associated with postpartum depression. Psychosocial variables were also identified as risk factors. The relative influence of these variables on the manifestation of postpartum depression is yet to be determined.

Clinical and economic outcomes of a pilot project examining pharmacist-focused collaborative care treatment for depression.

OBJECTIVE: To assess the clinical and economic impact of a pharmacist-focused health management program for patients with depression. DESIGN: Prospective, nonrandomized, proof-of-concept investigation. SETTING: Asheville, NC, from July 2006 through December 2007. PARTICIPANTS: Employees or adult dependents with depressive symptoms who agreed to enroll in an employer-sponsored treatment program conducted at two ambulatory clinics where consultative services were provided. Participants were included in the analysis if they participated in the program for at least 1 year and had two or more documented visits with a pharmacist. INTERVENTION: Outpatient-based pharmacists provided assessment, self-management services follow-up, and treatment recommendations to primary care providers within a collaborative care management model. MAIN OUTCOME MEASURES: Changes in severity of depressive symptoms and impact on overall health care costs for employers and beneficiaries. RESULTS: Of the 151 beneficiaries referred to the program, 130 (82%) remained under pharmacist care for a minimum of 1 year and were included in the aggregate analysis. Statistically significant improvements were observed for Patient Health Questionnaire (PHQ)-9 scores from baseline to endpoint (11.5 ± 6.6 to 5.3 ± 4.7 [mean ± SD], P < 0.0001). The clinical response rate was 68% with a 56% remission rate. In economic subgroup analysis (n = 48), annual medical costs decreased from an average of $6,351 per enrollee to $5,876, which was lower than the projected value ($7,195). Total health care costs to the employer increased from $7,935 per enrollee to $8,040, which was lower than the projected value ($9,023). CONCLUSION: Patients in the first year of the program had significant improvement in the PHQ-9 clinical indicator of depression severity. Total health care costs per patient per year were reduced compared with projected costs without the program. Employers expressed their appreciation for this collaborative care program and continued to offer this voluntary health benefit after the study's conclusion.

Pharmacologic treatment for postpartum depression: a systematic review.

During the past decade, the medical community has expressed a growing concern over the high prevalence of postpartum depression and the tragic repercussions of untreated illness. However, many questions persist about the pathogenesis of postpartum depression, the natural course of the illness, and the safety and effectiveness of available treatments. To summarize the data on pharmacologic treatments for postpartum depression, we performed a systematic review of four major databases to identify original research published from 1960-September 2009 that featured pharmacologic treatments for depression detected in women during the 12 months after delivery. Pharmacologic treatments included prescription drugs (antidepressants and hormones), herbal remedies, and dietary supplements. Case reports, studies examining the prevention of postpartum depression, and those including diagnosed episodes of depression preceding the postpartum period (i.e., antepartum onset) were excluded. Treatment randomization or the presence of a control group was not required for inclusion in this review. Fourteen investigations met inclusion criteria. Nine studies examined the effects of prescription antidepressants, two investigated hormones, and three featured omega-3 fatty acid supplementation. Significant heterogeneity was evident in study design and prevented a pooled quantitative analysis of treatment effects. The power of most investigations was limited, and numerous confounding biases were evident. Therapeutic effects were documented for prescription antidepressants and hormone supplementation (estrogen derivatives). Tolerability of the interventions in depressed mothers and breastfed infants was not well described. The effectiveness of omega-3 fatty acids was not evident in postpartum depression trials, although significant limitations in study methodology were apparent. Postpartum depression is a common and serious medical problem, but most cases go undetected and untreated. The need to identify safe, effective, and convenient treatments for postpartum depression is urgent, but the current state of the medical literature describing pharmacologic interventions is not impressive. Preliminary evidence documenting the effectiveness of serotonergic antidepressants and hormone supplementation should serve as an impetus for rigorous controlled investigations in the future.

Heart of Pharmacy: a course exploring the psychosocial issues of patient care.

OBJECTIVE: To implement and assess the impact of a course utilizing reflective learning to explore the complex, psychosocial human issues encountered in pharmacy practice. DESIGN: A 1-credit-hour elective course, The Heart of Pharmacy, was offered to all pharmacy students. The course utilized both content and reflective techniques to produce a mutual exploratory learning experience for students, staff, and faculty members. Faculty and staff facilitators observed competencies and used a single group posttest design to assess students' attitudes. In year four, students' written reflections for each session were added and reviewed on a continuous basis throughout the course. ASSESSMENT: Faculty and staff observations indicated that educational outcomes were achieved and student perceptions and evaluations of the course were highly positive. Three major themes were identified in the students' qualitative responses: a recognition of communal support among student and faculty colleagues; a grounding for personal growth and professional formation; a deeper insight into and experience with the role of the pharmacist as compassionate listener and caregiver. CONCLUSION: Faculty observations of student competencies and students' perceptions of this course point to the need for pharmacy education to provide organized, structured reflective learning opportunities for students and faculty members to explore the deeper human issues of pharmacy practice and patient care.