Package 'openstudyval'

Open validator for FDA study-data submissions

Description

Validates a study folder against publicly described FDA technical-rejection criteria, SAS XPORT constraints, and an extensible SDTM, SEND, and ADaM structural catalog. Independent of Pinnacle 21 and not an FDA certification.

Author(s)

Shekhar Eranna Chalwadi

See Also

validate_submission, rules_catalog

---

Path to a bundled example study

Description

Example studies intentionally contain conformance issues so a first validation run produces a non-empty report.

Usage

example_study(which = c("sdtm", "adam"))

Arguments

which

"sdtm" or "adam".

Value

Character path to the example directory.

Examples

example_study("sdtm")

---

Validation report object

Description

An osv_report contains issue rows, a dataset inventory, the standard, and the engine label. Severity order is Reject, Error, Warning, Note.

Usage

## S3 method for class 'osv_report'
print(x, ...)

## S3 method for class 'osv_report'
summary(object, ...)

## S3 method for class 'osv_report'
as.data.frame(x, row.names = NULL, optional = FALSE, ...)

Arguments

x, object	An osv_report.
row.names	Ignored. Present for the as.data.frame generic.
optional	Ignored. Present for the as.data.frame generic.
...	Passed to further methods.

Value

summary returns a data frame of one row per rule. as.data.frame returns the issue-level data frame.

Examples

report <- validate_submission(example_study("sdtm"), standard = "SDTM")
print(report)
as.data.frame(report)

---

Read a study-data folder

Description

CSV and RDS files are read with base R. SAS transport files (.xpt) are read with haven when that package is installed.

Usage

read_study_data(path)

Arguments

path	Directory containing study datasets, or a character vector of file paths.

Value

A named list of data frames. Each element has a source_file attribute. Names are the upper-case file stem.

Examples

study <- read_study_data(example_study("sdtm"))
names(study)

---

Rule catalog helpers

Description

The catalog is an original implementation of publicly described expectations. Messages are not copied from Pinnacle 21 or from CDISC conformance-rule publications. Supported check types are dataset_required, variable_required, codelist, iso8601, key_unique, one_to_one, nonmissing, and regex.

Usage

rules_catalog(standard = NULL)

explain_rule(rule_id)

add_custom_rules(rules, catalog = rules_catalog())

severity_levels()

Arguments

standard	Optional filter: "SDTM", "SEND", "ADaM", "XPT", or "TRC".
rule_id	Rule identifier, such as "SDTM024".
rules	Data frame of additional rules with the catalog columns.
catalog	Catalog to extend. Defaults to rules_catalog().

Value

rules_catalog and explain_rule return a data frame. add_custom_rules returns the combined catalog. severity_levels returns a character vector.

Examples

head(rules_catalog("SDTM"))
explain_rule("SDTM024")
severity_levels()

---

Validate datasets already in memory

Description

Same engine as validate_submission for data already loaded.

Usage

validate_datasets(
  datasets,
  standard = c("SDTM", "SEND", "ADaM"),
  define = NULL,
  rules = NULL,
  ct = NULL,
  version_label = "openstudyval 0.1.0",
  path = NULL
)

Arguments

datasets	Named list of data frames, as returned by read_study_data.
standard	One of "SDTM", "SEND", or "ADaM".
define	Optional define.xml path.
rules	Optional extra rules.
ct	Optional controlled terminology data frame.
version_label	Report label.
path	Original folder, used only for technical-rejection file checks.

Value

An object of class osv_report.

Examples

study <- read_study_data(example_study("adam"))
validate_datasets(study, standard = "ADaM")

---

Validate a study-data folder

Description

Reads datasets, then runs XPORT constraints, standard-specific structural checks, the built-in rule catalog, technical-rejection criteria, and an optional define.xml cross-check.

Usage

validate_submission(
  path,
  standard = c("SDTM", "SEND", "ADaM"),
  define = NULL,
  rules = NULL,
  ct = NULL,
  version_label = "openstudyval 0.1.0"
)

Arguments

path	Directory of datasets, or a vector of dataset files.
standard	One of "SDTM", "SEND", or "ADaM".
define	Optional path to define.xml. If NULL, a file named define.xml in path is used when present.
rules	Optional data frame of extra rules. See add_custom_rules.
ct	Optional data frame of controlled terminology. Column names are variable names and values are permitted terms.
version_label	Character label stored on the report.

Details

The result is not a regulatory certification. Agency engines remain the reference for a live submission.

Value

An object of class osv_report.

See Also

validate_datasets, write_validation_report

Examples

report <- validate_submission(example_study("sdtm"), standard = "SDTM")
summary(report)

---

Write an HTML or CSV validation report

Description

Writes a submission-style issue summary. The HTML layout is original and is not a Pinnacle 21 report.

Usage

write_validation_report(report, file)

Arguments

report	An osv_report from validate_submission.
file	Path ending in .html or .csv. An HTML report also writes a sibling CSV.

Value

The written path, invisibly.

Examples

report <- validate_submission(example_study("sdtm"), standard = "SDTM")
dest <- file.path(tempdir(), "osv-report.html")
write_validation_report(report, dest)

---