Get active facilities from a dataset

This helper function identifies and returns only active facilities from a dataset based on their reporting patterns. It's a convenience wrapper around classify_facility_activity that filters the data to include only facilities classified as "Active".

Usage

get_active_facilities(
  data,
  hf_col,
  date_col = "date",
  key_indicators = c("allout", "conf", "test", "treat", "pres"),
  method = 3,
  nonreport_window = 6,
  reporting_rule = "any_non_na",
  return_summary = FALSE
)

Arguments

data

A data frame containing health facility data

hf_col

Character. Name of the column containing unique facility IDs

date_col

Character. Name of the date column. Default is "date"

key_indicators

Character vector of indicator columns used to determine reporting activity. Default is c("allout", "conf", "test", "treat", "pres")

method

Numeric or character. Classification method (1, 2, 3, "method1", "method2", "method3"). Default is 3 (dynamic activation/inactivation)

nonreport_window

Integer. Number of consecutive non-reporting periods before a facility is considered inactive (for method 3). Default is 6

reporting_rule

Character. What counts as reporting: "any_non_na" (default, 0 counts as reported) or "positive_only" (requires >0)

return_summary

Logical. If TRUE, returns a summary tibble instead of filtered data. Default is FALSE

Value

If return_summary = FALSE (default): Returns the input data frame filtered to include only rows for active facilities, with an added activity_status column.

If return_summary = TRUE: Returns a summary tibble with columns:

• n_total: Total number of unique facilities

• n_active: Number of active facilities

• n_inactive: Number of inactive facilities

• pct_active: Percentage of facilities that are active

• pct_inactive: Percentage of facilities that are inactive

Details

This function is particularly useful for:

• Pre-filtering data before calculating reporting rates

• Understanding the proportion of operational facilities

• Ensuring denominators only include facilities that should be reporting

The function uses classify_facility_activity internally to determine facility status based on reporting patterns in the key indicators.

See also

classify_facility_activity, calculate_reporting_metrics

Examples

# Create sample data
sample_data <- data.frame(
  date = rep(as.Date(c("2024-01-01", "2024-02-01", "2024-03-01")), each = 4),
  facility_id = rep(c("HF001", "HF002", "HF003", "HF004"), 3),
  district = rep(c("North", "North", "South", "South"), 3),
  allout = c(
    100, 150, 0, NA,    # Month 1: HF001 and HF002 active,
                        #  HF003 reports 0, HF004 never reports
    110, 160, 0, NA,    # Month 2
    105, 155, NA, NA    # Month 3: HF003 stops reporting
  ),
  conf = c(
    10, 15, 0, NA,
    11, 16, 0, NA,
    10, 15, NA, NA
  )
)

# Get only active facilities
active_only <- get_active_facilities(
  data = sample_data,
  hf_col = "facility_id",
  date_col = "date",
  key_indicators = c("allout", "conf")
)
#> ℹ Facility activity classification:
#>   3 of 4 facilities are active (75%)
#>   1 facilities are inactive (25%)
#> ✔ Returning data filtered to 3 active facilities (of 4 total)

# Get summary of facility activity
activity_summary <- get_active_facilities(
  data = sample_data,
  hf_col = "facility_id",
  date_col = "date",
  key_indicators = c("allout", "conf"),
  return_summary = TRUE
)
#> ℹ Facility activity classification:
#>   3 of 4 facilities are active (75%)
#>   1 facilities are inactive (25%)