2 Introduction
2.1 Background
In May 2022, queer and trans New Yorkers faced a public-health crisis that demanded a different kind of response. Mpox cases had already been reported across Europe among gay and bisexual men, with several early clusters linked to large parties and festivals. With LGBTQ Pride season approaching, many U.S. activists anticipated a local outbreak as people traveled nationally and internationally for Pride events. We expected that large-scale vaccination would be needed among LGBTQ communities — but the systems designed to detect and manage emerging outbreaks were already strained.
Limited testing and vaccine shortage
The official data used to understand the size and speed of the outbreak came from a diagnostic system that was slow, limited in capacity, and not built for rapid community-level surveillance. Early in the outbreak, testing for mpox could only be performed in specialized public-health laboratories. Clinicians often had to coordinate with the local health department before collecting specimens, and testing generally required visible lesions to be present.
We anticipated that many gay and bi men would struggle to obtain testing, but what unfolded was worse. Some people left clinical encounters not only untested but without adequate treatment for the severe, sometimes debilitating pain that mpox caused. Even when providers suspected mpox, access to testing was inconsistent: in many jurisdictions, clinicians could not submit specimens without public-health approval, and only a small number of laboratories were authorized to run the PCR assay until commercial labs came online later in July.
At the same time, the United States entered the outbreak with a severe vaccine shortage. When the first U.S. cases were detected, the Strategic National Stockpile had thousands of immediately deployable doses of JYNNEOS, the preferred vaccine. By contrast, the estimated size of the U.S. population at elevated risk — men who have sex with men meeting HIV PrEP indications plus MSM living with HIV — was eventually estimated in the millions. It was clear that vaccine rationing would become necessary, and equally clear that the data needed to support those decisions were limited and slow.
MPX NYC
Out of this urgency, RESPND-MI — the Rapid Epidemiologic Study of Sexual Networks, Demographics, and Mpox Infection — came together. Researchers, clinicians, community organizers, mpox patients, influencers, and party promoters collaborated to build a study that could move at the speed of an outbreak while staying accountable to the people it served. The result was the MPX NYC study: an anonymous, web-based survey and network-mapping project documenting how queer and trans people interacted across the city.
To do this, we measured networks directly, through link tracing, and indirectly, through co-attendance of events and spaces. To preserve anonymity, we developed the MPX NYC Person–Place Mapper, a tool that converts participant-reported locations into U.S. census-tract identifiers before data are saved. This allowed us to collect sensitive spatial data safely and at scale.
Data from the Person–Place Mapper proved immediately useful. During data collection, we shared interim analyses with the New York City Department of Health, which used these reports to refine its mobile mpox vaccination efforts and identify new locations to reach. By the time our survey launched, the initial vaccine shortage had been partially alleviated by splitting doses across people. Vaccine was rationed at the dose level, not the person level, as we had initially expected.
As the immediate crisis shifted, we broadened the project’s goal: to document and share the methods behind the study. The challenges we confronted reflected a deeper question in epidemiology — how to reason about causality when exposures and outcomes are intertwined across people and places. We developed SSNAC (Social and Spatial Network Analysis with Causal Interpretation), a framework for understanding causality in connected populations, where risks and interventions spread through relationships, neighborhoods, and shared spaces.
People’s Department of Health
Our findings show how patterns of movement, mixing, and gathering likely shaped mpox transmission — and how targeted, neighborhood-based interventions can help contain future outbreaks. Alongside these results, we share new descriptive statistics and new tools for measuring sensitive spatial data. In the spirit of open science, we have made our analytic code and data available.
When traditional systems lagged, queer and trans New Yorkers built our own data infrastructure. The result is a unique dataset, new epidemiologic theory, and a foundation for rapid, community-led outbreak response. Together, these efforts offer a glimpse of what a People’s Department of Health could look like.
2.2 What this book presents
This book shares what we learned from that effort. It combines findings, methods, and organizing principles into a single resource for anyone building community-led health infrastructure.
At its core are three intertwined strands:
Findings – what the MPX NYC data reveal about how people and places were connected during the 2022 outbreak.
Science – the frameworks we developed for understanding causal processes in networks, detailed in the SSNAC toolkits.
Organizing – the participatory model that powered RESPND-MI, from the LGBTQ+ Community Forum to bilingual communications and distributed leadership.
2.3 The shape of the study
The MPX NYC survey asked participants about health, networks, and places — linking each location to its corresponding U.S. census tract to preserve anonymity while enabling spatial analysis. These data allowed us to explore how patterns of movement, mixing, and gathering shaped exposure risk and community reach.
The results chapters describe what we found:
People (Chapter 6) — who joined the study
Gatherings (Chapter 7) — where people met for sex or social connection
Movement (Chapter 8) — how participants moved across the city
Mixing (Chapter 9) — how likely different groups are to meet
Outbreaks (Chapter 10) — what we learn about vaccination during an outbreak
And the methods chapter explains how we prepared and conducted the survey, and how we analyzed the data.
Data (Chapter 7) — how we implemented the data collection
Measures (Chapter 8) — quantities we used to describe the data
Analysis (Chapter 9) — how we analyzed the data
Additional descriptions of the data appear in Appendix G (Results). Survey questions and the MPX NYC Person–Place Mapper, the tool we developed to safely solicit sensitive locations, are detailed in Appendix F (Measurement).
2.4 From findings to theory
To interpret these data, we developed a new framework for network-based causal inference. Classical epidemiologic models assume independent individuals; our data showed instead that outcomes and exposures were structured by overlapping social and spatial networks.
The SSNAC framework introduces definitions, graphical models, and analytic tools for reasoning about causality under interference. It extends the formal foundations of causal inference into the context of connected populations, where one person’s exposure may influence another’s outcome, and where risk is shaped by social interactions and relationships in space.
The SSNAC framework is laid out in three appendices:
Context (Appendix A) — the central ideas of SSNAC in words
Description (Appendix B) — descriptive statistics for a network with people connected to spatial units
Causality (Appendix C) — toolkits, formal definitions, and graphical models for reasoning under interference
2.5 From theory to practice
The study was not only a scientific project; it was also a collective act of organizing. Through the RESPND-MI LGBTQ+ Community Forum, dozens of community members and leaders met weekly to make shared decisions about study design, translation, and communication.
We reflect on this organizing effort in the appendices:
Organizing (Appendix D) — our principles, structure, and practices
Marketing (Appendix E) — our approach to community-centered communication