Skip to content

Issue 14: Use percent ED visits due to COVID data so we have data for whole timerange #61

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
kaitejohnson merged 8 commits into from
Jan 5, 2026
Merged

Issue 14: Use percent ED visits due to COVID data so we have data for whole timerange #61

Show file tree
Hide file tree
Changes from 6 commits
Commits
Show all changes
8 commits
Select commit Hold shift + click to select a range
75c662e
add revised function + data
Jan 5, 2026
2aec003
add link to original nssp data
Jan 5, 2026
3de8f80
add link to original nssp data
Jan 5, 2026
a7080f7
wrong fp
Jan 5, 2026
491dd00
fix linting
Jan 5, 2026
706ec14
Merge branch 'main' into 14-use-pct-ed-visits
kaitejohnson Jan 5, 2026
463e688
fix title
Jan 5, 2026
45101c3
Merge branch '14-use-pct-ed-visits' of https://github.com/epiforecast…
Jan 5, 2026
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
47 changes: 20 additions & 27 deletions R/fig_data.R
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -198,16 +198,17 @@ get_bar_chart_seq_count <- function(obs_data,
return(p)
}

#' Hospital admissions over time
#' Percent of hospital admissions due to covid over time
Copy link

@coderabbitai coderabbitai bot Jan 5, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

⚠️ Potential issue | 🟡 Minor

Correct terminology: ED visits, not hospital admissions.

The title states "Percent of hospital admissions due to covid" but the data source (NSSP) and y-axis label (line 245) correctly reference "ED visits" (emergency department visits). Hospital admissions and ED visits are distinct metrics.

🔎 Proposed fix 
-#' Percent of hospital admissions due to covid over time
+#' Percent of ED visits due to COVID over time
📝 Committable suggestion

‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.

Suggested change
#' Percent of hospital admissions due to covid over time
#' Percent of ED visits due to COVID over time
🤖 Prompt for AI Agents 
In @R/fig_data.R around line 201, Update the roxygen title string that currently
reads "Percent of hospital admissions due to covid over time" to use correct
terminology "Percent of ED visits due to COVID-19 over time" (match
capitalization if desired) and ensure any other metadata/title variables in the
same block are consistent with the y-axis label and NSSP source; specifically,
edit the roxygen comment line starting with "#' Percent of hospital admissions
due to covid over time" and verify there are no other occurrences of "hospital
admissions" in this help/plot block.

#'
#' @param location Location to plot (abbreviation)
#' @param date_range Vector of date range to plot, will use min and max
#' @param temporal_granularity Temporal granularity to plot
#' @param location_data Data.frame of location metadata to translate codes to
#' abbreviations
#' @param plot_name Name of plot
#' @param output_fp File.path to save plot
#' @param url URL for hospital admissions
#' @param output_fp File path to save plot
#' @param data_fp Filepath to NSSP data at the state and national level.
#' Originally obtained from: https://data.cdc.gov/Public-Health-Surveillance/NSSP-Emergency-Department-Visit-Trajectories-by-St/rdmq-nq56/about_data #nolint
#'
#' @returns ggplot object
#' @autoglobal
Expand All @@ -220,37 +221,28 @@ get_plot_hosp_admissions <- function(location_to_plot,
"output", "figs",
"data_figs"
),
url = "https://raw.githubusercontent.com/CDCgov/covid19-forecast-hub/refs/heads/main/target-data/covid-hospital-admissions.csv") { # nolint
if (location_to_plot != "US minus CA") {
raw_data <- read_csv(url) |>
left_join(location_data, by = "location") |>
filter(
abbreviation == !!location_to_plot,
date >= min(date_range),
date <= max(date_range)
)
} else {
raw_data <- read_csv(url) |>
left_join(location_data, by = "location") |>
filter(
!abbreviation %in% c("US", "CA"),
date >= min(date_range),
date <= max(date_range)
) |>
group_by(date) |>
summarise(value = sum(value, na.rm = TRUE))
}

data_fp = file.path("input", "nssp_states_and_national.csv")) { # nolint

nssp_cdcgov <- read_csv(data_fp)
raw_data <- nssp_cdcgov |>
left_join(location_data,
by = c("geography" = "location_name") # nolint
) |>
rename(date = week_end) |>
filter(
abbreviation == !!location_to_plot,
date >= min(date_range),
date <= max(date_range)
)
Comment on lines +226 to +236
Copy link

@coderabbitai coderabbitai bot Jan 5, 2026

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

⚠️ Potential issue | 🔴 Critical

🧩 Analysis chain

🏁 Script executed:

#!/bin/bash
# Verify the NSSP CSV schema and check for a sample of join key values

if [ -f "input/nssp_states_and_national.csv" ]; then
  echo "=== CSV column names ==="
  head -1 input/nssp_states_and_national.csv
  
  echo ""
  echo "=== Sample geography values (first 10 unique) ==="
  tail -n +2 input/nssp_states_and_national.csv | cut -d',' -f1 | sort -u | head -10
else
  echo "NSSP CSV file not found at input/nssp_states_and_national.csv"
  echo ""
  echo "Searching for NSSP-related CSV files in the repository..."
  find . -name "*nssp*.csv" -o -name "*states*.csv" | head -20
fi

Repository: epiforecasts/evalvariantnowcasthub

Length of output: 608

🏁 Script executed:

#!/bin/bash
# Find where location_data is created/loaded in the R file
grep -n "location_data" R/fig_data.R | head -20

Repository: epiforecasts/evalvariantnowcasthub

Length of output: 248

🏁 Script executed:

#!/bin/bash
# View the function signature and parameter documentation
sed -n '200,240p' R/fig_data.R

Repository: epiforecasts/evalvariantnowcasthub

Length of output: 1772

🏁 Script executed:

#!/bin/bash
# Find where get_plot_hosp_admissions is called
echo "=== Function calls to get_plot_hosp_admissions ==="
rg "get_plot_hosp_admissions" --type r -B 2 -A 5

echo ""
echo "=== Search for location_data creation/loading ==="
rg "location_data\s*<-|location_data\s*=" --type r -B 2 -A 2 | head -50

Repository: epiforecasts/evalvariantnowcasthub

Length of output: 5289

🏁 Script executed:

#!/bin/bash
# Find the tar_target definition for location_data
rg "tar_target.*location_data|name\s*=\s*location_data" --type r -B 5 -A 10

Repository: epiforecasts/evalvariantnowcasthub

Length of output: 928

🏁 Script executed:

#!/bin/bash
# Find location_fp definition
echo "=== Finding location_fp ==="
rg "location_fp\s*=" --type r -B 2 -A 2

echo ""
echo "=== Search for location CSV files ==="
find . -name "*location*" -type f | grep -E "\.(csv|tsv)$" | head -20

Repository: epiforecasts/evalvariantnowcasthub

Length of output: 454

🏁 Script executed:

#!/bin/bash
# Search for location_fp assignment
echo "=== Finding location_fp assignment ==="
rg "location_fp.*<-|location_fp\s*=" --type r -B 1 -A 1

echo ""
echo "=== Check targets config file for location_fp ==="
grep -n "location_fp" targets/config_targets.R | head -20

Repository: epiforecasts/evalvariantnowcasthub

Length of output: 430

🏁 Script executed:

#!/bin/bash
# View the tar_target definition for location_fp
sed -n '30,45p' targets/config_targets.R

Repository: epiforecasts/evalvariantnowcasthub

Length of output: 849

🏁 Script executed:

#!/bin/bash
# Fetch and inspect the FluSight locations CSV
echo "=== Fetching FluSight locations CSV ==="
curl -s "https://raw.githubusercontent.com/cdcepi/FluSight-forecast-hub/refs/heads/main/auxiliary-data/locations.csv" | head -5 > /tmp/locations.csv

if [ -s /tmp/locations.csv ]; then
  echo "=== Column names ==="
  head -1 /tmp/locations.csv
  
  echo ""
  echo "=== First few rows ==="
  head -6 /tmp/locations.csv
else
  echo "Failed to fetch locations CSV"
fi

Repository: epiforecasts/evalvariantnowcasthub

Length of output: 742

Correct the join key: use location instead of location_name.

The NSSP CSV's geography column contains numeric FIPS codes (e.g., "1", "10", "100", "1000"), which correspond to the FluSight locations CSV's location column (e.g., "01", "02", "04"), not location_name (which contains place names like "Alabama", "Alaska"). The current join will fail to match any rows.

Change:

by = c("geography" = "location_name")

To:

by = c("geography" = "location")
🤖 Prompt for AI Agents 
In @R/fig_data.R around lines 226-236, The left_join between nssp_cdcgov and
location_data is using the wrong key; change the join in the pipeline that
produces raw_data (the left_join call involving nssp_cdcgov and location_data)
to match geography to the FluSight `location` column instead of `location_name`
so rows with FIPS codes align correctly (i.e., use by = c("geography" =
"location") in the left_join definition).


p <- ggplot(raw_data) +
geom_bar(aes(x = date, y = value),
geom_line(aes(x = date, y = percent_visits_covid),
fill = "black",
stat = "identity", position = "dodge"
) +
get_plot_theme() +
xlab("") +
ylab("Hospital admissions") +
ylab("Percent of ED visits\ndue to COVID") +
# scale_y_continuous(transform = "log10")+
scale_x_date(
limits = c(min(date_range), max(date_range)),
Expand All @@ -263,7 +255,8 @@ get_plot_hosp_admissions <- function(location_to_plot,
size = 10
),
legend.position = "bottom"
)
) +
ggtitle(glue::glue("{location_to_plot}"))
dir_create(output_fp, recurse = TRUE)
ggsave(file.path(output_fp, glue::glue("{plot_name}.png")),
plot = p,
Expand Down
26 changes: 17 additions & 9 deletions R/globals.R
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -69,8 +69,9 @@ utils::globalVariables(c(
"obs_freq", # <get_plot_obs_clade_freq>
"clades_modeled", # <get_bar_chart_seq_count>
"sequences", # <get_bar_chart_seq_count>
"week_end", # <get_plot_hosp_admissions>
"abbreviation", # <get_plot_hosp_admissions>
"value", # <get_plot_hosp_admissions>
"percent_visits_covid", # <get_plot_hosp_admissions>
"location", # <get_plot_seq_counts_by_loc>
"sequences", # <get_plot_seq_counts_by_loc>
"population", # <get_plot_seq_counts_by_loc>
Expand Down Expand Up @@ -105,6 +106,13 @@ utils::globalVariables(c(
"location", # <get_plot_by_location>
"model", # <get_plot_by_nowcast_date>
"nowcast_date", # <get_plot_by_nowcast_date>
"model_id", # <get_plot_coverage_overall>
"interval_range", # <get_plot_coverage_overall>
"interval_coverage", # <get_plot_coverage_overall>
"empirical_coverage", # <get_plot_coverage_overall>
"90", # <get_plot_coverage_overall>
"50", # <get_plot_coverage_overall>
"interval_label", # <get_plot_coverage_overall>
"model", # <get_plot_overall>
"target_date", # <get_plot_horizon>
"nowcast_date", # <get_plot_horizon>
Expand Down Expand Up @@ -138,13 +146,13 @@ utils::globalVariables(c(
"model", # <get_plot_bias_by_date>
"bias", # <get_plot_bias_by_date>
"avg_bias", # <get_plot_bias_by_date>
"model_id", # <get_plot_coverage_overall>
"location", # <get_plot_coverage_overall>
"interval_range", # <get_plot_coverage_overall>
"interval_coverage", # <get_plot_coverage_overall>
"empirical_coverage", # <get_plot_coverage_overall>
"95", # <get_plot_coverage_overall>
"50", # <get_plot_coverage_overall>
"interval_label", # <get_plot_coverage_overall>
"model_id", # <get_plot_coverage_by_loc>
"location", # <get_plot_coverage_by_loc>
"interval_range", # <get_plot_coverage_by_loc>
"interval_coverage", # <get_plot_coverage_by_loc>
"empirical_coverage", # <get_plot_coverage_by_loc>
"90", # <get_plot_coverage_by_loc>
"50", # <get_plot_coverage_by_loc>
"interval_label", # <get_plot_coverage_by_loc>
NULL
))
Loading