Academic Work | Allison Londerée

RSV Vaccine Attitudes in a Digital Health Intervention

Wed, 15 Apr 2026 12:00:00 -0400

This case study is based on published work from my time at Lirio on a digital health intervention designed to increase RSV vaccination among adults over 60. Rather than treating outreach as a one-way campaign, the project looked closely at what people actually texted back and used those replies to better understand attitudes, friction points, and behavior-change opportunities.

Highlights

Analyzed unsolicited SMS replies to understand RSV vaccine attitudes at scale.
Used structural topic modeling plus expert thematic coding for interpretable insights.
Findings informed how live intervention messaging could be improved over time.

Problem

When the RSV vaccine was introduced for older adults in the United States in 2023, many of the drivers of uptake were still unknown.

For teams designing outreach interventions, this creates a familiar challenge: how do you improve messaging when beliefs, concerns, and barriers are still emerging in real time?

Rather than relying solely on structured surveys or downstream outcomes, we turned to a more immediate signal: what people actually texted back.

Approach

The intervention was launched with a large community pharmacy chain using SMS precision nudging to encourage vaccination. As patients replied, we accumulated a large, messy, and highly valuable dataset of unsolicited responses.

However, the scale of the data introduced a new constraint: there were far too many replies for behavioral experts to review manually, and insights were needed quickly to inform an active intervention.

My role as the data scientist was to bridge that gap—developing a scalable way to extract structure and meaning from free-text responses, while preserving the nuance required for behavioral interpretation.

Methods

The study population included 2,481,987 eligible adults aged 60 and older, of whom 35,716 people replied, generating 105,848 text responses. After removing operational texts such as STOP and HELP, the analytic dataset included 46,964 unsolicited replies.

We used a mixed-methods workflow that combined computational modeling with expert interpretation:

Structural topic modeling (STM) to identify latent themes across the full dataset
Thematic analysis on a hand-coded subset, conducted by behavioral health experts
Alignment of model-derived topics with expert-reviewed themes to ensure interpretability
Characterization of topics along dimensions such as sentiment and practical vs. emotional function
Comparisons across time, message condition, and available covariates (e.g., flu vaccination history, insurance type)

This approach allowed us to scale insight generation without losing the behavioral context needed for intervention design.

Topic Map of 30 topics in text message replies to a precision nudging digital health intervention for RSV vaccination. Shape and colour indicate the mapping of structural topic model topics onto the 10 thematic analysis topics. Black star = Wrong recipient, teal circle = Help, red cross = Stop, purple plus = Already vaccinated, brown square = Will not get vaccinated, yellow triangle = Benign, and grey diamond = Nonsense. Size indicates topic proportions. Topics 1-10 are largest, topics 11-20 are of middle size, and topics 21-30 are smallest.

Outcome / Impact

The mixed-method approach of modeling and expert review converged to produce a grounded, interpretable map of how adults responded to a newly introduced vaccine in a real-world setting.

Key findings included:

expressed attitudes became less negative later in the intervention
individuals without prior flu vaccination and those with commercial insurance were more likely to express refusal
messages framed around emotional consequences generated the highest engagement
messages framed around anticipated regret generated the lowest engagement

Beyond the publication, this project demonstrated a practical pattern: behavioral data generated during live interventions can be rapidly structured and fed back into design decisions.

Instead of treating outreach as static, this approach enables a tighter loop between what people say, how we interpret it, and how interventions evolve.

Tools

Behavioral science, thematic analysis, structural topic modelling, text analysis, digital health experimentation, intervention evaluation.

Reward Representations and Mindsets

Thu, 01 Oct 2020 18:05:20 -0400

One line of research examines a feature that has been shown to be important for self-control; mindset. The mindset that people adopt when making decisions about food is associated with more or less self-control success. For example, when people evaluate foods with a taste mindset (i.e., focusing on the taste of the food), they desire to consume more unhealthy foods than when they adopt a health mindset (i.e., focusing on the health benefits of the food). Functional neuroimaging studies examining neural responses to food have shown that appetitive food images are associated with activity in the brain’s reward system. Understanding how different mindsets may alter neural responses to appetitive food cues may offer insight into why taste mindsets are associated with poorer self-control than health mindsets. A first study ( ) investigated the neural representations of food items varying in taste and health using representational similarity analysis (RSA).Neural activity patterns in the orbital frontal cortex (OFC) spontaneously encodes food health, whereas tastiness was associated with greater neural dissimilarity. Subsequent analyses using model dissimilarity matrices that encode overall tastiness magnitude demonstrated that the neural representation of foods grows more distinct with increasing tastiness but not with increasing health. This suggests neural and cognitive representations of food categories that are the highest in tastiness are more refined than those lower in tastiness.

Highlights

Mindset (taste vs. health focus) changes how reward-relevant food information is represented.
fMRI + RSA show OFC encodes health spontaneously, while tastiness drives sharper neural distinctions.
The long-term goal is to decode self-control mindsets and improve strategy targeting.

A second study in this line of work aims to expand on this finding, and attempts to exploit the ease at which we are able to simulate the mental states of others. Specifically, this study examines patterns of activation in the brain’s reward system that are associated with mentally simulating mindsets of successful or unsuccessful self-regulators. This study will determine whether there are reliable neural differences associated with these mindsets that can be used to decode the default mindset an individual adopts spontaneously during uninstructed viewing of tempting food cues. If successful, this study lays a framework that can serve as a marker for understanding when some individuals may be successful or unsuccessful when making dietary decisions. I plan to expand this project to help understand when and for whom a variety of self-control strategies (construal, reappraisal, inhibition etc.) may be effective.

A third study integrates research from motivation science (construal level theory) with research from neuroscience (reward cue-reactivity) and computational modeling (attentional drift diffusion model; DDM). Using a combination of functional and structural neuroimaging, eye gaze, decision time, and choice data, this study aims to connect neural reward cue-reactivity to attention towards reward cues and ultimately decision-making. Additionally, this study examines the mediating effects of construal level mindset, the focus on high (i.e., relationship to health goals) or low (i.e., sugariness) level features of an item, on the relationship between cue-reactivity and decision choice. The ultimate goal of this project is to advance a model of self-control that provides new insight into the decisions that people make and the underlying processes that led to those decisions. In doing so, new theoretical and applied research that may ultimately help improve people’s self-control can be developed.

Predicting Self Regulation Across Adolescence

Thu, 01 Oct 2020 17:57:04 -0400

This line of work aims to make predictions about when an individual may be particularly susceptible to giving into tempting cues. Adolescents have been characterized as a vulnerable population due to their heightened sensitivity to rewards. Neurobiological models characterize adolescents’ hypersensitivity to rewards as a result of a developmental imbalance between a slower-to-develop prefrontal cortex and a faster-to-develop striatal system. Thus, this line of work focuses on identifying neural markers of self-regulation in adolescents. A first study ( , Londerée et al., in prep) examined how alternative tobacco marketing influences teen non-smokers using a multimodal approach involving fMRI, DTI, and eye-tracking. Adolescents’ neural and behavioral sensitivity to naturalistic flavored alternative tobacco ads were examined. Neural reward response is increased when tobacco is paired with appetitive flavors, even in non-smokers, due to this heightened sensitivity to reward cues in this population. Individuals that showed an increased attention towards ads subsequently reported a stronger willingness to try alternative tobacco. In addition, structural connectivity between regions involved in self-control (IFG) and reward (OFC) was associated with reduced reward reactivity to food cues. These findings suggest a neural mechanism for how marketing with appetitive cues can make inroads into adolescents’ attention and potentially lead to initiation of unhealthy behaviors.

Highlights

This work studies when adolescents are most vulnerable to reward cues and risky initiation.
Multimodal neuroimaging and behavioral data identify potential self-regulation markers.
Findings link cue-reactivity and attention patterns to willingness and later substance-use risk.

A second ongoing study aims to elucidate the relationship between environment (stress, exposure to violence), past experience, and substance use initiation in adolescents. Adolescents are vulnerable to neural changes after exposure to violence. It is hypothesized that these changes can increase risk-taking, reward-seeking, and ultimately substance use initiation. This project contains multimodal measures, including a multi-substance cue-reactivity task, DTI, and resting state functional connectivity, to compare with blood samples measuring stress-induced inflammation (CRP) and substance use reports. This work aims use baseline as well as longitudinal neural changes to predict subsequent substance use and identify structural and functional neural mediators. Future studies in this line of work will attempt to apply these structural, functional and behavioral markers to other populations and other types of tempting cues.