Join tracts to buffers

Author

Affiliation

Vagish Hemmige

Montefiore Medical Center/ Albert Einstein College of Medicine

This section identifies which census tracts fall within transplant center catchment areas.

Earlier steps created:

• Census tract–level HIV population estimates (Merged_tracts)
• Distance-based or travel-time buffers around transplant centers

Here, we determine which tracts intersect each center’s catchment area using spatial joins.

Source code

The full R script is available at:

• R/join_tracts_to_buffers.R

This R script file is itself reliant on the following helper files:

• R/setup.R
• R/functions.R

---

Linking Census Tracts to Transplant Center Catchment Areas

This code performs spatial intersection joins between:

• Census tract–level data (Merged_tracts)
• Transplant center buffer polygons

The goal is to determine which tracts are geographically reachable from transplant centers under each scenario.

---

Objects Created

Three nested list objects are generated:

• tracts_in_buffers_active
• tracts_in_buffers_HIV
• tracts_in_buffers_HOPE

Each is indexed by:

1. Organ
   
2. Distance
   
3. Year

---

What the Spatial Join Does

For each:

• Organ
  
• Distance
  
• Year

The script performs:

st_join(…, join = st_intersects)

This means:

A census tract is included if its geometry intersects the transplant center buffer polygon.

In practical terms:

• If any part of a tract lies within a center’s catchment area,
• That tract is considered geographically accessible to that center.

---

Important Note on Duplication

Because this join is performed against individual center buffers, a tract can appear multiple times if:

• It falls within the catchment area of more than one transplant center.

This duplication is intentional at this stage because:

• It preserves center-level relationships.
• It allows downstream analyses of overlap and multi-center accessibility.

Later steps that use united buffers eliminate this duplication when calculating total population coverage.

---

Why This Step Matters

This spatial join enables calculation of:

• Number of people living with HIV within reach of transplant centers
• Differences in geographic access between:
  • Active centers
  • HIV R+ centers
  • HOPE centers
• Changes over time (e.g., 2017 vs 2022)

By linking tract-level population data to center catchment areas, the analysis translates geography into measurable access.

---

#ST join to individual tracts and transplant centers


tracts_in_buffers_active<-list()
tracts_in_buffers_HIV<-list()
tracts_in_buffers_HOPE<-list()




for (organ_loop in organ_list){
  for (distance_loop in distance_list){
    for (year_loop in year_list){

      message(paste("Currently on:", organ_loop, "", distance_loop, "for year:", year_loop))
      message("Calculating tracts for active...")
      
      #This leads to duplication of tracts if they fall in more than one transplant center's catchment area)      
      tracts_in_buffers_active[[organ_loop]][[distance_loop]][[year_loop]] <- 
        st_join(Merged_tracts[[year_loop]], 
                Transplant_centers_active_buffer[[organ_loop]][[distance_loop]][[year_loop]], 
                join = st_intersects)
      
      
      message("Calculating tracts for HIV...")
      
      tracts_in_buffers_HIV[[organ_loop]][[distance_loop]][[year_loop]] <- 
        st_join(Merged_tracts[[year_loop]], 
                Transplant_centers_HIV_buffer[[organ_loop]][[distance_loop]][[year_loop]], 
                join = st_intersects)
      
      message("Calculating tracts for HOPE...")
      
      tracts_in_buffers_HOPE[[organ_loop]][[distance_loop]][[year_loop]] <- 
        st_join(Merged_tracts[[year_loop]], 
                Transplant_centers_HOPE_buffer[[organ_loop]][[distance_loop]][[year_loop]], 
                join = st_intersects)

      
            
    }
  }
}

Other portions of the analysis

• Setup: Defines global paths, data sources, cohort inclusion criteria, and analysis-wide constants.
• Functions: Reusable helper functions for cohort construction, matching, costing, and modeling.
• Tables: Summary tables and regression outputs generated from the final models.
• Figures:Visualizations of costs, risks, and model-based estimates.
• About: methods, assumptions, and disclosures