OHDSI · ginberg · Jun 25, 2026 · Jun 19, 2026 · Jun 24, 2026
diff --git a/R/GetCovariates.R b/R/GetCovariates.R
@@ -66,6 +66,9 @@
 #' @param minCharacterizationMean The minimum mean value for characterization output. Values below this will be cut off from output. This
 #'                                will help reduce the file size of the characterization output, but will remove information
 #'                                on covariates that have very low values. The default is 0.
+#' @param minCharacterizationCount The minimum count value for characterization output. Values below this will be cut off from output. This
+#'                                 will help reduce the file size of the characterization output, but will remove information
+#'                                 on covariates that occur in very few cohort entries. The default is 0.
 #' @param tempEmulationSchema    Some database platforms like Oracle and Impala do not truly support
 #'                               temp tables. To emulate temp tables, provide a schema with write
 #'                               privileges where temp tables can be created.
@@ -147,6 +150,7 @@ getDbCovariateData <- function(connectionDetails = NULL,
                                targetTimeRefTable = NULL,
                                aggregated = FALSE,
                                minCharacterizationMean = 0,
+                               minCharacterizationCount = 0,
                                tempEmulationSchema = getOption("sqlRenderTempEmulationSchema"),
                                covariateCohortDatabaseSchema = NULL,
                                covariateCohortTable = NULL) {
@@ -172,7 +176,9 @@ getDbCovariateData <- function(connectionDetails = NULL,
   }
   errorMessages <- checkmate::makeAssertCollection()
   minCharacterizationMean <- utils::type.convert(minCharacterizationMean, as.is = TRUE)
+  minCharacterizationCount <- utils::type.convert(minCharacterizationCount, as.is = TRUE)
   checkmate::assertNumeric(x = minCharacterizationMean, lower = 0, upper = 1, add = errorMessages)
+  checkmate::assertIntegerish(x = minCharacterizationCount, lower = 0, len = 1, add = errorMessages)
   checkmate::reportAssertions(collection = errorMessages)
   if (!is.null(connectionDetails)) {
     connection <- DatabaseConnector::connect(connectionDetails)
@@ -351,7 +357,8 @@ getDbCovariateData <- function(connectionDetails = NULL,
           targetAnalysisRefTable = targetAnalysisRefTable,
           targetTimeRefTable = targetTimeRefTable,
           aggregated = aggregated,
-          minCharacterizationMean = minCharacterizationMean
+          minCharacterizationMean = minCharacterizationMean,
+          minCharacterizationCount = minCharacterizationCount
         )
         tempCovariateData <- do.call(eval(parse(text = fun)), args)
         if (is.null(covariateData)) {

diff --git a/R/GetCovariatesFromCohortAttributes.R b/R/GetCovariatesFromCohortAttributes.R
@@ -21,6 +21,14 @@
 #'
 #' @param covariateSettings   An object of type \code{covariateSettings} as created using the
 #'                            \code{\link{createCohortAttrCovariateSettings}} function.
+#' @param targetDatabaseSchema Optional target database schema. Ignored because cohort attribute covariates do not support exporting directly to target tables.
+#' @param targetCovariateTable Optional target covariate table. Ignored because cohort attribute covariates do not support exporting directly to target tables.
+#' @param targetCovariateContinuousTable Optional target continuous covariate table. Ignored because cohort attribute covariates do not support exporting directly to target tables.
+#' @param targetCovariateRefTable Optional target covariate reference table. Ignored because cohort attribute covariates do not support exporting directly to target tables.
+#' @param targetAnalysisRefTable Optional target analysis reference table. Ignored because cohort attribute covariates do not support exporting directly to target tables.
+#' @param targetTimeRefTable Optional target time reference table. Ignored because cohort attribute covariates do not support exporting directly to target tables.
+#' @param minCharacterizationMean The minimum mean value for characterization output. Ignored because aggregation is not supported for cohort attribute covariates.
+#' @param minCharacterizationCount The minimum count value for characterization output. Ignored because aggregation is not supported for cohort attribute covariates.
 #'
 #' @template GetCovarParams
 #'
@@ -66,7 +74,15 @@ getDbCohortAttrCovariatesData <- function(connection,
                                           cdmVersion = "5",
                                           rowIdField = "subject_id",
                                           covariateSettings,
+                                          targetDatabaseSchema = NULL,
+                                          targetCovariateTable = NULL,
+                                          targetCovariateContinuousTable = NULL,
+                                          targetCovariateRefTable = NULL,
+                                          targetAnalysisRefTable = NULL,
+                                          targetTimeRefTable = NULL,
                                           aggregated = FALSE,
+                                          minCharacterizationMean = 0,
+                                          minCharacterizationCount = 0,
                                           tempEmulationSchema = getOption("sqlRenderTempEmulationSchema")) {
   if (aggregated) {
     stop("Aggregation not implemented for covariates from cohort attributes.")

diff --git a/R/GetCovariatesFromOtherCohorts.R b/R/GetCovariatesFromOtherCohorts.R
@@ -37,6 +37,9 @@
 #' @param minCharacterizationMean The minimum mean value for binary characterization output. Values below this will be cut off from output. This
 #'                                will help reduce the file size of the characterization output, but will remove information
 #'                                on covariates that have very low values. The default is 0.
+#' @param minCharacterizationCount The minimum count value for binary characterization output. Values below this will be cut off from output. This
+#'                                 will help reduce the file size of the characterization output, but will remove information
+#'                                 on covariates that occur in very few cohort entries. The default is 0.
 #' @template GetCovarParams
 #'
 #' @export
@@ -57,6 +60,7 @@ getDbCohortBasedCovariatesData <- function(connection,
                                            targetTimeRefTable = NULL,
                                            aggregated = FALSE,
                                            minCharacterizationMean = 0,
+                                           minCharacterizationCount = 0,
                                            tempEmulationSchema = getOption("sqlRenderTempEmulationSchema")) {
   errorMessages <- checkmate::makeAssertCollection()
   checkmate::assertClass(connection, "DatabaseConnectorConnection", add = errorMessages)
@@ -70,7 +74,9 @@ getDbCohortBasedCovariatesData <- function(connection,
   checkmate::assertClass(covariateSettings, "covariateSettings", add = errorMessages)
   checkmate::assertLogical(aggregated, len = 1, add = errorMessages)
   minCharacterizationMean <- utils::type.convert(minCharacterizationMean, as.is = TRUE)
+  minCharacterizationCount <- utils::type.convert(minCharacterizationCount, as.is = TRUE)
   checkmate::assertNumeric(x = minCharacterizationMean, lower = 0, upper = 1, add = errorMessages)
+  checkmate::assertIntegerish(x = minCharacterizationCount, lower = 0, len = 1, add = errorMessages)
   checkmate::reportAssertions(collection = errorMessages)
   if (!missing(cohortId)) {
     warning("cohortId argument has been deprecated, please use cohortIds")
@@ -177,7 +183,8 @@ getDbCohortBasedCovariatesData <- function(connection,
     targetAnalysisRefTable = targetAnalysisRefTable,
     targetTimeRefTable = targetTimeRefTable,
     aggregated = aggregated,
-    minCharacterizationMean = minCharacterizationMean
+    minCharacterizationMean = minCharacterizationMean,
+    minCharacterizationCount = minCharacterizationCount
   )
 
   sql <- "TRUNCATE TABLE #covariate_cohort_ref; DROP TABLE #covariate_cohort_ref;"

diff --git a/R/GetDefaultCovariates.R b/R/GetDefaultCovariates.R
@@ -37,6 +37,9 @@
 #' @param minCharacterizationMean The minimum mean value for binary characterization output. Values below this will be cut off from output. This
 #'                                will help reduce the file size of the characterization output, but will remove information
 #'                                on covariates that have very low values. The default is 0.
+#' @param minCharacterizationCount The minimum count value for binary characterization output. Values below this will be cut off from output. This
+#'                                 will help reduce the file size of the characterization output, but will remove information
+#'                                 on covariates that occur in very few cohort entries. The default is 0.
 #' @template GetCovarParams
 #'
 #' @examples
@@ -75,6 +78,7 @@ getDbDefaultCovariateData <- function(connection,
                                       targetTimeRefTable = NULL,
                                       aggregated = FALSE,
                                       minCharacterizationMean = 0,
+                                      minCharacterizationCount = 0,
                                       tempEmulationSchema = getOption("sqlRenderTempEmulationSchema")) {
   if (!is(covariateSettings, "covariateSettings")) {
     stop("Covariate settings object not of type covariateSettings")
@@ -96,7 +100,9 @@ getDbDefaultCovariateData <- function(connection,
   }
   errorMessages <- checkmate::makeAssertCollection()
   minCharacterizationMean <- utils::type.convert(minCharacterizationMean, as.is = TRUE)
+  minCharacterizationCount <- utils::type.convert(minCharacterizationCount, as.is = TRUE)
   checkmate::assertNumeric(x = minCharacterizationMean, lower = 0, upper = 1, add = errorMessages)
+  checkmate::assertIntegerish(x = minCharacterizationCount, lower = 0, len = 1, add = errorMessages)
   checkmate::reportAssertions(collection = errorMessages)
 
 
@@ -115,7 +121,7 @@ getDbDefaultCovariateData <- function(connection,
   settings <- .toJson(covariateSettings)
   rJava::J("org.ohdsi.featureExtraction.FeatureExtraction")$init(system.file("", package = "FeatureExtraction"))
   json <- rJava::J("org.ohdsi.featureExtraction.FeatureExtraction")$createSql(
-    settings, aggregated, cohortTable, rowIdField, rJava::.jarray(as.character(cohortIds)), cdmDatabaseSchema, as.character(minCharacterizationMean)
+    settings, aggregated, cohortTable, rowIdField, rJava::.jarray(as.character(cohortIds)), cdmDatabaseSchema, as.character(minCharacterizationMean), as.character(minCharacterizationCount)
   )
   todo <- .fromJson(json)
   if (length(todo$tempTables) != 0) {

diff --git a/R/UnitTestHelperFunctions.R b/R/UnitTestHelperFunctions.R
@@ -59,6 +59,9 @@
 #' @param minCharacterizationMean The minimum mean value for binary characterization output. Values below this will be cut off from output. This
 #'                                will help reduce the file size of the characterization output, but will remove information
 #'                                on covariates that have very low values. The default is 0.
+#' @param minCharacterizationCount The minimum count value for binary characterization output. Values below this will be cut off from output. This
+#'                                 will help reduce the file size of the characterization output, but will remove information
+#'                                 on covariates that occur in very few cohort entries. The default is 0.
 #' @param ...                    Additional arguments, not used.
 #' @return
 #' Returns an object of type \code{covariateData}, containing information on the covariates.
@@ -73,6 +76,7 @@
                                    covariateSettings,
                                    aggregated = FALSE,
                                    minCharacterizationMean = 0,
+                                   minCharacterizationCount = 0,
                                    ...) {
   writeLines("Constructing length of observation covariates")
   if (covariateSettings$useLengthOfObs == FALSE) {

diff --git a/inst/csv/jarChecksum.txt b/inst/csv/jarChecksum.txt
@@ -1 +1 @@
-25224c0381cb8cf2d969e483400ae8b4caaaf3979df00b43c1197ce43c0cedbb
+148c135b5461287933e45215d633318bd0a81778989061930bcde3604096ccd5
diff --git a/inst/java/featureExtraction-3.13.0.jar → inst/java/featureExtraction-3.14.0.jar b/inst/java/featureExtraction-3.13.0.jar → inst/java/featureExtraction-3.14.0.jar
diff --git a/java/org/ohdsi/featureExtraction/FeatureExtraction.java b/java/org/ohdsi/featureExtraction/FeatureExtraction.java
@@ -515,11 +515,23 @@ else if (temporalSequence)
 	 */
 	public static String createSql(String settings, boolean aggregated, String cohortTable, String rowIdField, String[] cohortDefinitionIds,
 			String cdmDatabaseSchema, String minCharacterizationMean) {
+		return createSql(settings, aggregated, cohortTable, rowIdField, cohortDefinitionIds, cdmDatabaseSchema, minCharacterizationMean, "0");
+	}
+
+	/**
+	 * Construct the SQL for creating and retrieving the features.
+	 *
+	 * @param minCharacterizationCount The minimum count value for characterization output. Values below this will be cut off from output.
+	 * @return A JSON object.
+	 */
+	public static String createSql(String settings, boolean aggregated, String cohortTable, String rowIdField, String[] cohortDefinitionIds,
+			String cdmDatabaseSchema, String minCharacterizationMean, String minCharacterizationCount) {
 
 		long[] idsAsLongs = new long[cohortDefinitionIds.length];
 		for (int i = 0; i < cohortDefinitionIds.length; i++)
 			idsAsLongs[i] = Long.valueOf(cohortDefinitionIds[i]);
-		return createSql(settings, aggregated, cohortTable, rowIdField, idsAsLongs, cdmDatabaseSchema, Double.valueOf(minCharacterizationMean));
+		return createSql(settings, aggregated, cohortTable, rowIdField, idsAsLongs, cdmDatabaseSchema, Double.valueOf(minCharacterizationMean),
+				Integer.valueOf(minCharacterizationCount));
 	}
 
 
@@ -558,7 +570,7 @@ public static String createSql(String settings, boolean aggregated, String cohor
 	 */
 	public static String createSql(String settings, boolean aggregated, String cohortTable, String rowIdField, int cohortDefinitionId,
 								   String cdmDatabaseSchema) {
-		return createSql(settings, aggregated, cohortTable, rowIdField, new long[]{cohortDefinitionId}, cdmDatabaseSchema, 0.0d);
+		return createSql(settings, aggregated, cohortTable, rowIdField, new long[]{cohortDefinitionId}, cdmDatabaseSchema, 0.0d, 0);
 	}
 
 	/**
@@ -596,6 +608,17 @@ public static String createSql(String settings, boolean aggregated, String cohor
 	 */
 	public static String createSql(String settings, boolean aggregated, String cohortTable, String rowIdField, long[] cohortDefinitionIds,
 			String cdmDatabaseSchema, double minCharacterizationMean) {
+		return createSql(settings, aggregated, cohortTable, rowIdField, cohortDefinitionIds, cdmDatabaseSchema, minCharacterizationMean, 0);
+	}
+
+	/**
+	 * Construct the SQL for creating and retrieving the features.
+	 *
+	 * @param minCharacterizationCount The minimum count value for characterization output. Values below this will be cut off from output.
+	 * @return A JSON object.
+	 */
+	public static String createSql(String settings, boolean aggregated, String cohortTable, String rowIdField, long[] cohortDefinitionIds,
+			String cdmDatabaseSchema, double minCharacterizationMean, int minCharacterizationCount) {
 
 		JSONObject jsonObject = new JSONObject(settings);
 
@@ -666,7 +689,8 @@ public static String createSql(String settings, boolean aggregated, String cohor
 		jsonWriter.key("sqlConstruction");
 		jsonWriter.value(createConstructionSql(jsonObject, idSetToName, temporal, temporalSequence, aggregated, cohortTable, rowIdField, cohortDefinitionIds, cdmDatabaseSchema));
 
-		String sqlQueryFeatures = createQuerySql(jsonObject, cohortTable, cohortDefinitionIds, aggregated, temporal, temporalSequence, minCharacterizationMean);
+		String sqlQueryFeatures = createQuerySql(jsonObject, cohortTable, cohortDefinitionIds, aggregated, temporal, temporalSequence, minCharacterizationMean,
+				minCharacterizationCount);
 		if (sqlQueryFeatures != null) {
 			jsonWriter.key("sqlQueryFeatures");
 			jsonWriter.value(sqlQueryFeatures);
@@ -743,7 +767,7 @@ private static Object createCleanupSql(JSONObject jsonObject, boolean temporal2)
 	}
 
 	private static String createQuerySql(JSONObject jsonObject, String cohortTable, long[] cohortDefinitionIds, boolean aggregated, boolean temporal, boolean temporalSequence,
-	                                     double minCharacterizationMean) {
+	                                     double minCharacterizationMean, int minCharacterizationCount) {
 		boolean temporalAnnual = isTemporalAnnual(jsonObject);
 		Stream<String> fields = Stream.<Stream<String>>of(
 				aggregated ? Stream.of("cohort_definition_id", "covariate_id", "sum_value") : Stream.of("row_id", "covariate_id", "covariate_value"),
@@ -785,8 +809,15 @@ else if (temporalAnnual)
 		if (aggregated) {
 			sql.append("\n) all_covariates\nINNER JOIN (\nSELECT cohort_definition_id, COUNT(*) AS total_count\nFROM @cohort_table {@cohort_definition_id != -1} ? {\nWHERE cohort_definition_id IN (@cohort_definition_id)}");
 			sql.append(" GROUP BY cohort_definition_id\n) total\n  ON all_covariates.cohort_definition_id = total.cohort_definition_id");
+			List<String> filterConditions = new ArrayList<String>();
 			if (minCharacterizationMean != 0) {
-				sql.append(" WHERE all_covariates.sum_value / (1.0 * total.total_count) >= " + minCharacterizationMean);
+				filterConditions.add("all_covariates.sum_value / (1.0 * total.total_count) >= " + minCharacterizationMean);
+			}
+			if (minCharacterizationCount != 0) {
+				filterConditions.add("all_covariates.sum_value >= " + minCharacterizationCount);
+			}
+			if (!filterConditions.isEmpty()) {
+				sql.append(" WHERE " + filterConditions.stream().collect(Collectors.joining(" AND ")));
 			}
 			sql.append(";");
 		} else {

diff --git a/man/dot-getDbLooCovariateData.Rd b/man/dot-getDbLooCovariateData.Rd
Original file line number	Diff line number	Diff line change
		@@ -1 +1 @@
		25224c0381cb8cf2d969e483400ae8b4caaaf3979df00b43c1197ce43c0cedbb
		148c135b5461287933e45215d633318bd0a81778989061930bcde3604096ccd5