Abstract
Technical Project: Optimizing Patient Flow in Post-Surgical Observation
My capstone team partnered with a Mid-Atlantic teaching hospital that had identified overcrowding and suspected inefficiencies in its post-surgical observation care. Working with their internal teams, we used a mixed-methods approach to analyze patient flow. Specifically focusing on intake, bed placement, and discharge workflows. The quantitative analysis covered 593 encounters from the Surgical Observation Unit and a separate dataset from the unified pre- and post-operative unit. We supplemented this with in-person observations, four semi-structured interviews with bed center staff, a surgeon, and a post-operative unit director, and finally a follow-up survey designed to evaluate our proposed solutions. Three themes surfaced across the data. First, patient classification and placement decisions were made case-by-case with no standardized criteria or clear ownership. One stakeholder told us they had asked who owns the decision and never gotten an answer. Second, data collection was unstructured to the point that staff relied on experience-based judgment rather than evidence, with one interviewee summarizing the situation as “it’s all anecdotal.” Third, even when relevant data existed in the EHR, it was difficult to access in a form usable for operational decisions. The data itself contained quality issues such as including encounter durations approaching 687 hours and records with negative lengths of stay. We proposed six solutions across classification, staff knowledge, and data infrastructure, and found, consistent with existing literature on emergency department interventions, that combined approaches outperform single interventions. The project’s contribution is showing that the systemic issues well-documented in ED settings, including fragmented decision-making, communication breakdowns, and data gaps, also exist in post-surgical observation, where they had not previously been systematically examined.
STS Research: How Fitness Tracking Apps Shape Gen-Z’s Relationship with Health
My STS thesis investigates how calorie and fitness tracking apps shape behavior and health-related self-perception among Gen-Z users. I surveyed 51 UVA undergraduates: 22 current users, 18 former users, and 11 non-users. The central finding is a dissonance pattern. Sixty-eight percent of users described their app’s overall impact as positive, but 74% of those same users also reported at least one harmful behavior linked to their tracking, including exercising while sick or injured, working out when exhausted, restricting food to meet calorie targets, or compulsively checking the app. Current users reported harmful behaviors at nearly twice the rate of former users (77% vs. 39%), and they remained significantly more likely to recommend their app to a friend (8.0 vs. 5.3 out of 10, p < .001), even while reporting higher levels of stress and social comparison. I read these findings through three STS frameworks: Latour’s theory of nonhuman delegation, Sadowski’s analysis of data as capital, and Morozov’s critique of techno solutionism. Together, they suggest that these apps function as behavioral enforcement so effectively that users no longer distinguish the app’s goals from their own. The dashboard overrides felt experience, the metric replaces bodily authority, and engagement gets mistaken for health. The recommendation is not to abandon tracking. It is to redesign it: replacing binary streak counters with rolling averages, social leaderboards with private progress dashboards, and daily targets with prompts that ask users to check in with how they actually feel. The current generation of fitness apps is optimized for retention; the next should be optimized for well-being.
