Package 'PhysioClinical' reference manual

Title:	Clinical Outcome Measures and Responder Analysis
Description:	Validated clinical outcome-measure scoring (FMA, Berg, ARAT, TUG, 6MWT, Barthel, Katz, Lawton-Brody IADL, ...), MDC/MCID responder classification, ICF tagging, interval (Rasch) analysis of ordinal ADL/IADL instruments, and normative deviation scoring for the x-biosignal rehabilitation ecosystem.
Authors:	Yusuke Matsui [aut, cre]
Maintainer:	Yusuke Matsui <[email protected]>
License:	MIT + file LICENSE
Version:	0.7.0
Built:	2026-10-03 13:42:13 UTC
Source:	https://github.com/x-biosignal/PhysioClinical

Apply a score crosswalk to new values

Description

Apply a score crosswalk to new values

Usage

applyEquating(equating, values)
applyEquating(equating, values)

Arguments

equating

a score_equating from equateScores().

values

numeric source-scale values to convert.

Value

numeric target-scale equivalents (linearly interpolated within the crosswalk, clamped to its range).

Examples

set.seed(1)
ability <- rnorm(200)
barthel <- pmin(100, pmax(0, round((ability + 3) * 16 / 5) * 5))
fim <- pmin(126, pmax(18, round(18 + (ability + 3) * 18)))
applyEquating(equateScores(barthel, fim), c(40, 60, 80))
set.seed(1)
ability <- rnorm(200)
barthel <- pmin(100, pmax(0, round((ability + 3) * 16 / 5) * 5))
fim <- pmin(126, pmax(18, round(18 + (ability + 3) * 18)))
applyEquating(equateScores(barthel, fim), c(40, 60, 80))

Convert a ClinicalScore to a one-row data.frame

Description

Convert a ClinicalScore to a one-row data.frame

Usage

## S3 method for class 'ClinicalScore'
as.data.frame(x, ..., stringsAsFactors = FALSE)
## S3 method for class 'ClinicalScore'
as.data.frame(x, ..., stringsAsFactors = FALSE)

Arguments

x

A ClinicalScore.

...

Ignored.

stringsAsFactors

Ignored (kept for the S3 generic signature).

Value

A one-row data.frame with the total, each subscale, the stratum and the metadata columns.

Dual MDC-vs-MCID responder classification

Description

Classifies each subject's pre-to-post change with the two-criterion responder framework (Beaton 2001; de Vet 2006): a change is cross-tabulated on whether it exceeds the Minimal Detectable Change (is it real, above measurement error?) and the Minimal Clinically Important Difference (is it clinically meaningful?), giving four categories — true_responder (exceeds both), subclinical_change (real but below MCID), measurement_error (claims MCID but within noise, only when MCID < MDC) and non_responder. Vectorized over subjects and direction-aware.

Usage

classifyResponder(
  baseline,
  followup,
  instrument = NULL,
  population = NULL,
  mdc = NULL,
  mcid = NULL,
  direction = c("increase", "decrease")
)

## S3 method for class 'responder_classification'
summary(object, ...)
classifyResponder(
  baseline,
  followup,
  instrument = NULL,
  population = NULL,
  mdc = NULL,
  mcid = NULL,
  direction = c("increase", "decrease")
)

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

Arguments

baseline, followup

Numeric vectors of pre / post scores (one per subject).

instrument, population

Optional instrument / population used to look up mdc/mcid from the clinimetric store when they are not given.

mdc, mcid

Optional explicit MDC / MCID thresholds (positive); override the store lookup.

direction

"increase" (default) if higher is better, or "decrease" if lower is better.

object

A "responder_classification".

...

Unused.

Value

A "responder_classification" data.frame with the change, the direction-aware improvement, the MDC/MCID crossing flags and the four-level classification factor. A subject with a missing baseline/followup gets NA flags and classification (missing, not assumed a non-responder); summary() counts it.

summary returns the MDC x MCID contingency table.

References

Beaton DE et al. (2001); de Vet HCW et al. (2006).

Examples

classifyResponder(c(20, 25, 30), c(31, 26, 45), mdc = 5.2, mcid = 9)
classifyResponder(c(20, 25, 30), c(31, 26, 45), mdc = 5.2, mcid = 9)

Construct a ClinicalInstrument

Description

Construct a ClinicalInstrument

Usage

ClinicalInstrument(
  id,
  name = NA_character_,
  version = NA_character_,
  items,
  item_ranges,
  item_type = "interval",
  subscales = list(),
  aggregation = "sum",
  direction = "higher_better",
  strata = list(),
  item_values = list(),
  item_recode = list(),
  source_ref = NA_character_
)
ClinicalInstrument(
  id,
  name = NA_character_,
  version = NA_character_,
  items,
  item_ranges,
  item_type = "interval",
  subscales = list(),
  aggregation = "sum",
  direction = "higher_better",
  strata = list(),
  item_values = list(),
  item_recode = list(),
  source_ref = NA_character_
)

Arguments

id, name, version, items, item_ranges, item_type, subscales, aggregation, direction, strata, item_values, item_recode, source_ref

Instrument specification fields (see the class slots).

Value

A ClinicalInstrument.

Examples

ClinicalInstrument(
  id = "toy", name = "Toy scale", items = c("q1", "q2"),
  item_ranges = list(q1 = c(0, 4), q2 = c(0, 4)))
ClinicalInstrument(
  id = "toy", name = "Toy scale", items = c("q1", "q2"),
  item_ranges = list(q1 = c(0, 4), q2 = c(0, 4)))

Clinical instrument specification

Description

A declarative specification of a clinical outcome measure: its items, their admissible ranges and measurement level, its subscales, how items aggregate into scores, the direction of clinical benefit, and the interpretation strata.

Usage

## S4 method for signature 'ClinicalInstrument'
show(object)
## S4 method for signature 'ClinicalInstrument'
show(object)

Arguments

object

A ClinicalInstrument.

Slots

id: Short unique identifier (e.g. "berg").
name: Human-readable name.
version: Specification version string.
items: Character vector of item identifiers, in order.
item_ranges: Named list; each entry a length-2 numeric c(min, max) for the corresponding item.
item_type: Character vector (recycled to the item count) of "interval" or "ordinal".
subscales: Named list mapping each subscale name to a character vector of its item identifiers; length 0 for single-scale instruments.
aggregation: "sum" or "mean".
direction: "higher_better" or "higher_worse".
strata: List of interpretation bands, each a list with label, lower and upper.
item_values: Optional named list; for an item with a discrete (possibly non-integer) admissible value set - e.g. the Modified Ashworth ⁠1+⁠ = 1.5 - the numeric vector of allowed values. Items absent from the list are validated only against item_ranges / item_type.
item_recode: Optional named list; for an item whose raw responses are recoded before scoring (e.g. the SF-36 0-100 recode, or a reversed item), a named numeric map from the raw response (a character key) to the scored value. Applied after validation, before aggregation.
source_ref: Citation / provenance string.

Clinical score result

Description

The result of scoring a set of item responses against a ClinicalInstrument.

Usage

## S4 method for signature 'ClinicalScore'
show(object)
## S4 method for signature 'ClinicalScore'
show(object)

Arguments

object

A ClinicalScore.

Slots

instrument_id: Identifier of the scoring instrument.
total: Total score (or NA if unresolved under the missing policy).
subscales: Named numeric vector of subscale scores.
stratum: Interpretation stratum label the total falls in (or NA).
items_used: Character vector of the item identifiers actually scored.
missing_handling: The missing-data policy applied.
timestamp: Optional character timestamp.
subject_id: Optional subject identifier.

Define a Goal Attainment Scaling goal

Description

Describes one GAS goal: its attainment levels (typically -2:2 with 0 = expected outcome), a weight, and an optional ICF linkage. When both importance and difficulty are given the (Turner-Stokes) weight is their product; otherwise weight is used (default 1, i.e. unweighted GAS).

Usage

defineGoal(
  description,
  levels = -2:2,
  weight = 1,
  importance = NULL,
  difficulty = NULL,
  icf_tag = NULL
)
defineGoal(
  description,
  levels = -2:2,
  weight = 1,
  importance = NULL,
  difficulty = NULL,
  icf_tag = NULL
)

Arguments

description

Free-text goal description.

levels

Ordered numeric attainment levels (must include 0, the expected outcome); default -2:2.

weight

Goal weight when importance/difficulty are not supplied (default 1, positive).

importance, difficulty

Optional 0+ ratings; if both given the weight is importance * difficulty.

icf_tag

Optional ICF code (e.g. "d450") linking the goal to the International Classification of Functioning.

Value

A "gas_goal" object.

References

Kiresuk & Sherman (1968); Turner-Stokes (2009).

Examples

defineGoal("Walk 100 m unaided", importance = 3, difficulty = 2,
           icf_tag = "d450")
defineGoal("Walk 100 m unaided", importance = 3, difficulty = 2,
           icf_tag = "d450")

Equate one clinical score scale to another (build a crosswalk)

Description

Constructs a conversion from a "source" scale to a "target" scale using a linking sample measured on both. Equipercentile equating (the default) maps a source score to the target score with the same percentile rank, so it handles the non-linear, bounded relationship typical of ADL scales; linear equating matches only the mean and SD.

Usage

equateScores(
  from,
  to,
  method = c("equipercentile", "linear"),
  from_scores = sort(unique(from[is.finite(from)]))
)
equateScores(
  from,
  to,
  method = c("equipercentile", "linear"),
  from_scores = sort(unique(from[is.finite(from)]))
)

Arguments

from

Numeric source-scale scores from the linking sample.

to

Numeric target-scale scores from the linking sample (equipercentile uses the two marginal distributions; linear uses their means and SDs).

method

"equipercentile" (default) or "linear".

from_scores

Source scores to tabulate (default: the sorted unique observed source scores).

Value

an object of class score_equating: a list with table (data.frame from, to), method, and the linking-sample size. Convert new values with applyEquating().

References

Kolen MJ, Brennan RL (2004) Test Equating, Scaling, and Linking; Houlden H et al. (2006) Clin Rehabil 20:153-159 (Barthel<->FIM).

Examples

set.seed(1)
ability <- rnorm(200)
barthel <- pmin(100, pmax(0, round((ability + 3) * 16 / 5) * 5))
fim <- pmin(126, pmax(18, round(18 + (ability + 3) * 18)))
eq <- equateScores(barthel, fim)          # Barthel -> FIM crosswalk
head(eq$table)
set.seed(1)
ability <- rnorm(200)
barthel <- pmin(100, pmax(0, round((ability + 3) * 16 / 5) * 5))
fim <- pmin(126, pmax(18, round(18 + (ability + 3) * 18)))
eq <- equateScores(barthel, fim)          # Barthel -> FIM crosswalk
head(eq$table)

Anchor-based MCID

Description

Estimates the MCID from change scores anchored to an external improvement indicator. "roc" returns the change cut-off that maximizes the Youden index for classifying the anchor; "mean_change" the mean change among the (minimally) improved; "predictive" the change at which a logistic model of the anchor crosses probability 0.5.

Usage

estimateMCID_anchor(
  change,
  anchor,
  method = c("roc", "mean_change", "predictive"),
  direction = c("increase", "decrease")
)
estimateMCID_anchor(
  change,
  anchor,
  method = c("roc", "mean_change", "predictive"),
  direction = c("increase", "decrease")
)

Arguments

change

Numeric change scores (follow-up minus baseline).

anchor

Improvement indicator aligned to change: logical, or 0/1, where TRUE/1 marks an (minimally) improved case.

method

"roc" (default), "mean_change" or "predictive".

direction

"increase" (default) if a higher change means improvement, or "decrease" if a lower change does.

Value

The MCID estimate (on the change scale).

References

Jaeschke et al. (1989); Copay et al. (2007).

Examples

set.seed(1)
change <- c(rnorm(50, 1, 2), rnorm(50, 8, 2))
anchor <- rep(c(0, 1), each = 50)
estimateMCID_anchor(change, anchor)
set.seed(1)
change <- c(rnorm(50, 1, 2), rnorm(50, 8, 2))
anchor <- rep(c(0, 1), each = 50)
estimateMCID_anchor(change, anchor)

Distribution-based MCID (fraction-of-SD rule)

Description

The distribution-based Minimal Clinically Important Difference as a fraction of the baseline standard deviation (Norman et al. 2003: the 0.5 SD rule).

Usage

estimateMCID_distribution(baseline_sd, fraction = 0.5)
estimateMCID_distribution(baseline_sd, fraction = 0.5)

Arguments

baseline_sd

Baseline standard deviation of the outcome.

fraction

Fraction of the SD (default 0.5).

Value

The MCID estimate (fraction * baseline_sd).

References

Norman GR, Sloan JA, Wyrwich KW (2003). Med Care 41(5).

Examples

estimateMCID_distribution(baseline_sd = 12)
estimateMCID_distribution(baseline_sd = 12)

Distribution-based Minimal Detectable Change from test-retest data

Description

Estimates the MDC from a test-retest reliability study: the intraclass correlation gives the reliability, the standard error of measurement is $SD\sqrt{1 - ICC}$ , and $MDC = SEM \times z \times \sqrt{2}$ (delegating to PhysioCore::icc/sem/mdc).

Usage

estimateMDC(
  test_retest,
  method = "distribution",
  confidence = 0.95,
  model = c("twoway", "oneway")
)
estimateMDC(
  test_retest,
  method = "distribution",
  confidence = 0.95,
  model = c("twoway", "oneway")
)

Arguments

test_retest

A subjects-by-occasions numeric matrix (or data.frame) of repeated measurements.

method

Estimation method; currently "distribution".

confidence

Confidence level for the MDC (default 0.95).

model

ICC model passed to PhysioCore::icc (default "twoway").

Value

A list with mdc, sem, icc and confidence.

References

Shrout & Fleiss (1979); de Vet et al. (2006).

Examples

set.seed(1)
base <- rnorm(40, 50, 10)
retest <- base + rnorm(40, 0, 3)
estimateMDC(cbind(base, retest))
set.seed(1)
base <- rnorm(40, 50, 10)
retest <- base + rnorm(40, 0, 3)
estimateMDC(cbind(base, retest))

Import a FHIR R4 Observation as a clinical score

Description

Parses an HL7 FHIR R4 Observation (as produced by toFHIRObservation) back into a ClinicalScore, the import counterpart of the to*/write* exporters. The instrument_id and each subscale name are recovered from the bundled LOINC map (reverse code lookup, then display, then the verbatim text); the total and subscale values come from valueQuantity (a dataAbsentReason becomes NA); the subject id strips a leading "Patient/"; and the timestamp comes from effectiveDateTime.

Usage

fromFHIR(observation)
fromFHIR(observation)

Arguments

observation

A FHIR Observation as a list (a toFHIRObservation result or a parsed jsonlite object), a JSON string, or a path to a .json file.

Details

The round-trip is lossy for three slots that toFHIRObservation does not serialize: items_used and missing_handling cannot be recovered, and stratum is re-derived from the total only when the instrument is registered (via getInstrument), otherwise NA.

Value

A ClinicalScore.

Examples

sc <- scoreInstrument("katz_adl",
  stats::setNames(rep(1, 6), getInstrument("katz_adl")@items),
  subject_id = "P01")
fromFHIR(toFHIRObservation(sc))@total
sc <- scoreInstrument("katz_adl",
  stats::setNames(rep(1, 6), getInstrument("katz_adl")@items),
  subject_id = "P01")
fromFHIR(toFHIRObservation(sc))@total

Tabulate a GAS result

Description

Tabulate a GAS result

Usage

gasSummary(object, ...)
gasSummary(object, ...)

Arguments

object

A "gas_result" from scoreGAS.

...

Unused.

Value

A data.frame with one row per goal (description, ICF tag, weight, attained level and its weighted contribution) plus the T-score as an attribute.

Look up a per-instrument clinimetric constant

Description

Retrieves a published clinimetric statistic (MDC, MCID, MCII, SEM, ICC, ...) for a measurement instrument from the packaged, provenanced constants store. The return value carries the source reference (DOI, population n, provenance note) so a looked-up threshold is always attributable.

Usage

getClinimetric(instrument, statistic, population = NULL)
getClinimetric(instrument, statistic, population = NULL)

Arguments

instrument

Instrument name (e.g. "FMA-UE", "10mWT"); case insensitive.

statistic

One of "MDC", "MCID_anchor", "MCID_dist", "MCII", "SEM", "ICC" (case insensitive).

population

Optional population/context key to disambiguate multiple rows (e.g. "chronic_stroke_minimal").

Value

A "clinimetric" data.frame of the matching row(s) with value, CI, method, reference_doi, population_n and provenance_note; or NA with a warning when nothing (or no such population) matches.

Examples

getClinimetric("10mWT", "MCII")
getClinimetric("FMA-UE", "MCID_anchor", population = "chronic_stroke_minimal")
getClinimetric("10mWT", "MCII")
getClinimetric("FMA-UE", "MCID_anchor", population = "chronic_stroke_minimal")

Retrieve a registered clinical instrument

Description

Retrieve a registered clinical instrument

Usage

getInstrument(id)
getInstrument(id)

Arguments

id

Instrument identifier.

Value

The ClinicalInstrument.

Retrieve a normative reference from the registry

Description

Retrieve a normative reference from the registry

Usage

getNormative(id, version = "latest", root = NULL)
getNormative(id, version = "latest", root = NULL)

Arguments

id

Artifact identifier.

version

Semantic version, or "latest" (default) for the highest registered version.

root

Registry root (see registerNormative()).

Value

The stored GovernedNormativeReference.

Construct a governed normative reference

Description

Construct a governed normative reference

Usage

GovernedNormativeReference(
  id,
  modality,
  metric,
  version = "1.0.0",
  provenance = list(),
  consent = list(),
  license = list(),
  governance = list(),
  strata_vars = character(0),
  model = list(),
  n = NA_real_
)
GovernedNormativeReference(
  id,
  modality,
  metric,
  version = "1.0.0",
  provenance = list(),
  consent = list(),
  license = list(),
  governance = list(),
  strata_vars = character(0),
  model = list(),
  n = NA_real_
)

Arguments

id, modality, metric

Identifying strings.

version

Semantic version (default "1.0.0").

provenance, , license, governance

Governance lists (see GovernedNormativeReference); provenance$source, consent$status and license$spdx are required.

strata_vars

Character vector of stratification variables.

model

List holding the normative model (stratified mean/sd table or LMS coefficients).

n

Reference sample size.

Value

A validated GovernedNormativeReference.

Examples

GovernedNormativeReference(
  id = "gait_speed_adult", modality = "gait", metric = "gait_speed",
  version = "1.0.0",
  provenance = list(source = "Bohannon 1997", doi = "10.1093/ageing/26.1.15"),
  consent = list(status = "public_aggregate"),
  license = list(spdx = "CC-BY-4.0", redistribution_ok = TRUE),
  governance = list(custodian = "Matsui Lab", access_level = "open"),
  strata_vars = c("age", "sex"),
  model = list(type = "strata",
               table = data.frame(age = 70, sex = "M", mean = 1.3, sd = 0.2)),
  n = 230)
GovernedNormativeReference(
  id = "gait_speed_adult", modality = "gait", metric = "gait_speed",
  version = "1.0.0",
  provenance = list(source = "Bohannon 1997", doi = "10.1093/ageing/26.1.15"),
  consent = list(status = "public_aggregate"),
  license = list(spdx = "CC-BY-4.0", redistribution_ok = TRUE),
  governance = list(custodian = "Matsui Lab", access_level = "open"),
  strata_vars = c("age", "sex"),
  model = list(type = "strata",
               table = data.frame(age = 70, sex = "M", mean = 1.3, sd = 0.2)),
  n = 230)

Governed normative reference

Description

An S4 container for a normative reference distribution together with the governance metadata a clinical artifact must carry: provenance, consent, license and custodianship. Validity requires the provenance source, consent status and license identifier to be present, so an ungoverned reference cannot be constructed.

Slots

id: Stable artifact identifier.
modality: Signal modality (e.g. "gait", "hrv").
metric: The normed quantity (e.g. "gait_speed").
version: Semantic version string "x.y.z".
provenance: List with at least source; optionally doi, collection_date.
consent: List with at least status; optionally ethics_id.
license: List with at least spdx; optionally redistribution_ok.
governance: List with custodian and access_level.
strata_vars: Character vector of stratification variables (e.g. c("age", "sex")).
model: List holding the normative model — either a stratified mean/sd table or LMS (lambda/mu/sigma) coefficients.
n: Reference sample size.

List available clinimetric constants

Description

List available clinimetric constants

Usage

listClinimetrics(instrument = NULL)
listClinimetrics(instrument = NULL)

Arguments

instrument

Optional instrument to filter by (case insensitive).

Value

A data.frame of the stored clinimetric constants.

Examples

listClinimetrics("6MWT")
listClinimetrics("6MWT")

List available clinical instruments

Description

List available clinical instruments

Usage

listInstruments()
listInstruments()

Value

A character vector of registered instrument ids.

List registered normative references

Description

List registered normative references

Usage

listNormative(root = NULL)
listNormative(root = NULL)

Arguments

root

Registry root (see registerNormative()).

Value

A data.frame of id, version, modality, metric, access_level and n, newest version last.

Modified Ashworth Scale level label for a value

Description

The inverse of the MAS arithmetic mapping: returns the ordinal label for a numeric MAS value (so 1.5 round-trips to "1+").

Usage

masLevelLabel(value)
masLevelLabel(value)

Arguments

value

Numeric MAS value(s).

Value

Character MAS level label(s).

Examples

masLevelLabel(1.5)
masLevelLabel(1.5)

Match the normative stratum for a set of covariates

Description

Selects the row of a governed normative reference's model table that matches the supplied covariates: categorical stratification variables (e.g. sex) are matched exactly, numeric ones (e.g. age, speed) by nearest value. A covariate falling outside the tabulated range is flagged as an extrapolation.

Usage

matchStratum(ref, covariates)
matchStratum(ref, covariates)

Arguments

ref

A GovernedNormativeReference.

covariates

Named list of covariate values covering the reference's strata_vars.

Value

A one-row data.frame (the matched stratum) with an "extrapolation" attribute (logical).

Examples

ref <- GovernedNormativeReference("gs", "gait", "gait_speed",
  provenance = list(source = "x"), consent = list(status = "public"),
  license = list(spdx = "CC0-1.0"),
  governance = list(custodian = "lab", access_level = "open"),
  strata_vars = c("age", "sex"),
  model = list(type = "strata", table = data.frame(
    age = c(60, 70), sex = "M", mean = c(1.4, 1.3), sd = 0.2)))
matchStratum(ref, list(age = 68, sex = "M"))
ref <- GovernedNormativeReference("gs", "gait", "gait_speed",
  provenance = list(source = "x"), consent = list(status = "public"),
  license = list(spdx = "CC0-1.0"),
  governance = list(custodian = "lab", access_level = "open"),
  strata_vars = c("age", "sex"),
  model = list(type = "strata", table = data.frame(
    age = c(60, 70), sex = "M", mean = c(1.4, 1.3), sd = 0.2)))
matchStratum(ref, list(age = 68, sex = "M"))

Classify change scores against the minimal detectable change (MDC)

Description

A first responder-analysis primitive: a change is a real improvement/decline only if it exceeds the MDC (measurement noise). Uses PhysioCore::mdc.

Usage

mdcResponder(change, sem_value, confidence = 0.95)
mdcResponder(change, sem_value, confidence = 0.95)

Arguments

change

Numeric change score(s) (follow-up minus baseline).

sem_value

The standard error of measurement (see PhysioCore::sem).

confidence

Confidence level for the MDC (default 0.95).

Value

Character vector: "improved", "stable", or "declined".

Examples

mdcResponder(c(5, 0.5, -6), sem_value = 1.5)
mdcResponder(c(5, 0.5, -6), sem_value = 1.5)

Batch normative deviation over a metric table

Description

Applies normativeZScore to every row of a metric table: each row supplies a value plus the covariate columns named by the reference's strata_vars.

Usage

normativeDeviation(metric_table, ref, deviation_z = 2)
normativeDeviation(metric_table, ref, deviation_z = 2)

Arguments

metric_table

A data.frame with a value column and one column per ref stratification variable.

ref

A GovernedNormativeReference.

deviation_z

Deviation-flag threshold (see normativeZScore()).

Value

metric_table with added z, percentile, deviation_flag and extrapolation columns.

Normative z-score / percentile for an observation

Description

Standardizes an observed value against a governed normative reference, selecting the covariate-matched stratum and applying either a Gaussian $(value - \mu)/\sigma$ (stratified mean/sd model) or the Cole (1990) LMS transform (LMS model). An unmatched or unsupported stratum raises an error rather than returning a silent NA.

Usage

normativeZScore(value, ref, covariates = list(), deviation_z = 2)
normativeZScore(value, ref, covariates = list(), deviation_z = 2)

Arguments

value

Numeric observed value.

ref

A GovernedNormativeReference.

covariates

Named list of covariates (e.g. list(age = 68, sex = "M")) covering the reference's strata_vars.

deviation_z

Absolute z above which deviation_flag is set (default 2).

Value

A list with z, percentile (0-100), deviation_flag and extrapolation.

References

Cole TJ (1990). The LMS method for constructing normalized growth standards. Eur J Clin Nutr 44(1), 45-60.

Examples

ref <- GovernedNormativeReference("gs", "gait", "gait_speed",
  provenance = list(source = "x"), consent = list(status = "public"),
  license = list(spdx = "CC0-1.0"),
  governance = list(custodian = "lab", access_level = "open"),
  strata_vars = "sex",
  model = list(type = "strata",
               table = data.frame(sex = "M", mean = 1.3, sd = 0.2)))
normativeZScore(1.0, ref, list(sex = "M"))
ref <- GovernedNormativeReference("gs", "gait", "gait_speed",
  provenance = list(source = "x"), consent = list(status = "public"),
  license = list(spdx = "CC0-1.0"),
  governance = list(custodian = "lab", access_level = "open"),
  strata_vars = "sex",
  model = list(type = "strata",
               table = data.frame(sex = "M", mean = 1.3, sd = 0.2)))
normativeZScore(1.0, ref, list(sex = "M"))

Convert a PROMIS summed score to a T-score via a lookup table

Description

The other official PROMIS scoring path for a full short form: map the summed raw score to a T-score with the instrument's published summed-score-to-T-score table. The table is not bundled - obtain it from the HealthMeasures scoring manual for the specific short form.

Usage

promisRawToT(raw, conversion_table)
promisRawToT(raw, conversion_table)

Arguments

raw

Summed raw score.

conversion_table

A data.frame with a raw column, a tscore column and optionally an se column (the published conversion table).

Value

A list with tscore and se (NA if the table has no se column); NA with a warning if raw is not in the table.

Examples

tab <- data.frame(raw = 4:8, tscore = c(21.5, 30.1, 35.7, 40.2, 44.0))
promisRawToT(6, tab)
tab <- data.frame(raw = 4:8, tscore = c(21.5, 30.1, 35.7, 40.2, 44.0))
promisRawToT(6, tab)

Rasch (interval) analysis of a clinical ADL/IADL instrument

Description

Fits a polytomous Rasch model to a cohort's responses on a registered instrument and returns the psychometric summary that published ADL/IADL analyses report: the item difficulty hierarchy, category threshold ordering (with disordered-threshold flags), person and item measures on an interval (logit) scale, item fit (infit/outfit), person/item separation reliability, and the raw-score to interval-measure conversion. The Partial Credit Model is the default (it accommodates the Barthel Index's mixed item lengths); pass model = "RSM" for a uniform-format scale.

Usage

raschAnalyze(instrument, responses, model = c("PCM", "RSM"), ...)
raschAnalyze(instrument, responses, model = c("PCM", "RSM"), ...)

Arguments

instrument

A ClinicalInstrument or an instrument id (e.g. "barthel", "lawton_iadl").

responses

A persons x items matrix or data.frame of responses on the instrument's native scale (see raschRecode()).

model

"PCM" (default) or "RSM".

...

Passed to PhysioAppKit::pcm_measure().

Details

Requires the suggested package PhysioAppKit (the domain-neutral engine).

Value

The poly_rasch fit (see PhysioAppKit::pcm_measure()) with an added class clin_rasch and extra fields: instrument (id), item_hierarchy (items ordered hardest-to-easiest by calibrated location) and recoded (the category matrix analysed).

Examples


set.seed(1)
items <- getInstrument("barthel")@items
# a small synthetic cohort of weighted Barthel responses
resp <- t(replicate(50, {
  ability <- stats::rnorm(1)
  vapply(c(2, 1, 1, 2, 2, 2, 2, 3, 3, 2), function(m)
    c(0, 5, 10, 15)[min(m, max(0, round(ability + m / 2))) + 1], numeric(1))
}))
colnames(resp) <- items
if (requireNamespace("PhysioAppKit", quietly = TRUE)) {
  fit <- raschAnalyze("barthel", resp)
  fit$item_hierarchy
}

set.seed(1)
items <- getInstrument("barthel")@items
# a small synthetic cohort of weighted Barthel responses
resp <- t(replicate(50, {
  ability <- stats::rnorm(1)
  vapply(c(2, 1, 1, 2, 2, 2, 2, 3, 3, 2), function(m)
    c(0, 5, 10, 15)[min(m, max(0, round(ability + m / 2))) + 1], numeric(1))
}))
colnames(resp) <- items
if (requireNamespace("PhysioAppKit", quietly = TRUE)) {
  fit <- raschAnalyze("barthel", resp)
  fit$item_hierarchy
}

Recode clinical instrument responses to consecutive Rasch categories

Description

Turns a persons x items table of responses on an instrument's native scale (e.g. the weighted Barthel levels 0/5/10/15) into the consecutive integer categories 0..m that a Rasch model expects, using the instrument's declared admissible item_values as the category order.

Usage

raschRecode(instrument, responses)
raschRecode(instrument, responses)

Arguments

instrument

A ClinicalInstrument or an instrument id.

responses

A persons x items matrix or data.frame; columns are matched to the instrument's items by name, or taken in the instrument's item order if unnamed.

Value

An integer matrix of consecutive category codes (0..m per item), with the instrument's item names as columns.

Examples

resp <- rbind(c(10, 5, 5, 10, 10, 10, 10, 15, 15, 10),
              c(5, 0, 0, 5, 5, 5, 5, 10, 5, 0))
colnames(resp) <- getInstrument("barthel")@items
raschRecode("barthel", resp)
resp <- rbind(c(10, 5, 5, 10, 10, 10, 10, 15, 15, 10),
              c(5, 0, 0, 5, 5, 5, 5, 10, 5, 0))
colnames(resp) <- getInstrument("barthel")@items
raschRecode("barthel", resp)

Responder analysis of pre-to-post change on the Rasch interval scale

Description

Classifies each person's change between two occasions using interval (logit) measures from raschAnalyze() / PhysioAppKit::pcm_measure(). The Rasch model's own standard errors give a distribution-free standard error of measurement and hence the Minimal Detectable Change; the pre/post change is then run through mdcResponder() (reliable change) and, when an MCID on the logit scale is supplied, classifyResponder() (the MDC x MCID framework).

Usage

raschResponder(
  pre,
  post,
  mcid = NULL,
  confidence = 0.95,
  direction = c("increase", "decrease")
)
raschResponder(
  pre,
  post,
  mcid = NULL,
  confidence = 0.95,
  direction = c("increase", "decrease")
)

Arguments

pre, post

Baseline and follow-up measures: a poly_rasch fit, a list with theta/theta_se, or a numeric measure vector. se is required (from the fit) to derive the MDC.

mcid

Optional MCID on the logit scale; enables the four-level classifyResponder() output.

confidence

Confidence level for the MDC (default 0.95).

direction

"increase" (default; higher measure = better) or "decrease".

Details

The two occasions must be on a common metric – calibrate the items once and anchor them, or co-calibrate – and the persons must align by row.

Value

a data.frame, one row per person: pre, post, change, improvement (direction-aware), sem, mdc, reliable_change (from mdcResponder()) and, if mcid given, classification (from classifyResponder()). Attributes sem, mdc, mcid record the values used.

Examples

pre  <- list(theta = c(0, 0, 0, 0), theta_se = rep(0.3, 4))
post <- list(theta = c(1.5, 0.1, -1.2, 0.05), theta_se = rep(0.3, 4))
raschResponder(pre, post, mcid = 1.0)
pre  <- list(theta = c(0, 0, 0, 0), theta_se = rep(0.3, 4))
post <- list(theta = c(1.5, 0.1, -1.2, 0.05), theta_se = rep(0.3, 4))
raschResponder(pre, post, mcid = 1.0)

Register a clinical instrument

Description

Adds an instrument to the registry, either a ClinicalInstrument object, a parsed-spec list, or the path to a YAML spec file.

Usage

registerInstrument(spec, overwrite = FALSE)
registerInstrument(spec, overwrite = FALSE)

Arguments

spec

A ClinicalInstrument, a spec list, or a path to a .yaml file.

overwrite

Replace an existing instrument with the same id? Default FALSE (an existing id is an error).

Value

Invisibly, the registered ClinicalInstrument.

Examples

inst <- ClinicalInstrument(id = "toy2", items = c("a", "b"),
  item_ranges = list(a = c(0, 1), b = c(0, 1)))
registerInstrument(inst)
inst <- ClinicalInstrument(id = "toy2", items = c("a", "b"),
  item_ranges = list(a = c(0, 1), b = c(0, 1)))
registerInstrument(inst)

Register a normative reference in the versioned registry

Description

Persists a validated GovernedNormativeReference to the registry at <root>/<id>/<version>.rds, writing a sibling governance manifest.json. The artifact must pass both object validity and validateNormativeManifest.

Usage

registerNormative(ref, root = NULL, overwrite = FALSE)
registerNormative(ref, root = NULL, overwrite = FALSE)

Arguments

ref

A GovernedNormativeReference.

root

Registry root directory; defaults to the option PhysioClinical.normative_root or the package's user data dir.

overwrite

Overwrite an already-registered version (default FALSE).

Value

The written .rds path, invisibly.

Examples

ref <- GovernedNormativeReference("gs", "gait", "gait_speed",
  provenance = list(source = "x"), consent = list(status = "public"),
  license = list(spdx = "CC0-1.0"),
  governance = list(custodian = "lab", access_level = "open"))
registerNormative(ref, root = tempfile("norm"))
ref <- GovernedNormativeReference("gs", "gait", "gait_speed",
  provenance = list(source = "x"), consent = list(status = "public"),
  license = list(spdx = "CC0-1.0"),
  governance = list(custodian = "lab", access_level = "open"))
registerNormative(ref, root = tempfile("norm"))

Compare a measured joint ROM to the reference (and the other side)

Description

Compares a measured joint range of motion against the packaged reference angle (see romReference()) and, when supplied, the contralateral (usually unaffected) side - the two comparisons clinicians read a goniometry measurement against.

Usage

romNormalcy(measured, joint, motion, contralateral = NULL)
romNormalcy(measured, joint, motion, contralateral = NULL)

Arguments

measured

Measured range of motion, in degrees.

joint, motion

The joint and motion (case insensitive); must identify a single reference row.

contralateral

Optional measured ROM of the other side, in degrees.

Value

A one-row data.frame (class "rom_normalcy") with reference_deg, percent_of_normal (NA for an extension-to-neutral motion whose reference is 0 deg), deficit_vs_reference (reference minus measured) and limited (measured below reference); plus contralateral_deg, deficit_vs_contralateral and percent_of_contralateral when a contralateral value is given.

Examples

romNormalcy(100, "knee", "flexion")
romNormalcy(100, "knee", "flexion", contralateral = 130)
romNormalcy(100, "knee", "flexion")
romNormalcy(100, "knee", "flexion", contralateral = 130)

Joint range-of-motion reference values (goniometry)

Description

Returns the packaged normal active range-of-motion reference values for the major peripheral joints (shoulder, elbow, forearm, wrist, hip, knee, ankle). These are the consensus clinical goniometry reference angles (in degrees) of the American Academy of Orthopaedic Surgeons, as tabulated by Norkin & White, Measurement of Joint Motion. They are single reference angles (not a distribution), so use romNormalcy() for a reference / side-to-side comparison; for a stratified z-score build a GovernedNormativeReference() from a cohort dataset and use normativeZScore().

Usage

romReference(joint = NULL, motion = NULL)
romReference(joint = NULL, motion = NULL)

Arguments

joint

Optional joint to filter by (e.g. "knee"; case insensitive).

motion

Optional motion to filter by (e.g. "flexion"; case insensitive).

Value

A data.frame with joint, motion, plane, reference_deg and source.

References

American Academy of Orthopaedic Surgeons (1965). Joint Motion: Method of Measuring and Recording. Norkin CC, White DJ. Measurement of Joint Motion: A Guide to Goniometry. F.A. Davis.

Examples

romReference("knee")
romReference(motion = "flexion")
romReference("knee")
romReference(motion = "flexion")

Score a 10-metre walk test (10mWT)

Description

Computes comfortable/fast gait speed and its ambulation category, with an optional normative z-score. Accepts a gait speed directly, a walk time, or a PhysioMoCap signal-derived result (walk_test_report from PhysioMoCap::instrumented10mWT or gait_parameters from PhysioMoCap::calculateGaitParameters).

Usage

score10MWT(
  x = NULL,
  time_s = NULL,
  distance = 10,
  pace = c("comfortable", "fast"),
  ref = NULL,
  covariates = list()
)
score10MWT(
  x = NULL,
  time_s = NULL,
  distance = 10,
  pace = c("comfortable", "fast"),
  ref = NULL,
  covariates = list()
)

Arguments

x

A gait speed (m/s), a walk_test_report, or a gait_parameters; or NULL to use time_s.

time_s

Walk time in seconds (manual path; speed = distance/time).

distance

Walk distance in metres (default 10).

pace

"comfortable" (default) or "fast".

ref

Optional GovernedNormativeReference for the normative z-score.

covariates

Covariates (e.g. list(age, sex)) for ref.

Value

A "performance_test_score" with gait_speed, ambulation and (if ref) zscore.

References

Perry et al. (1995); Fritz & Lusardi (2009); Perera et al. (2006).

Examples

score10MWT(time_s = 8)         # 10 m / 8 s = 1.25 m/s -> community ambulator
score10MWT(time_s = 8)         # 10 m / 8 s = 1.25 m/s -> community ambulator

Score a 6-minute walk test (6MWT)

Description

Reports the walked distance with, when the demographics are supplied, the Enright & Sherrill (1998) predicted distance, percent-predicted and the lower limit of normal (LLN), plus an optional normative z-score.

Usage

score6MWT(
  distance_m,
  age = NULL,
  sex = NULL,
  height_cm = NULL,
  weight_kg = NULL,
  ref = NULL,
  covariates = list()
)
score6MWT(
  distance_m,
  age = NULL,
  sex = NULL,
  height_cm = NULL,
  weight_kg = NULL,
  ref = NULL,
  covariates = list()
)

Arguments

distance_m

Distance walked in 6 minutes (metres).

age, sex, height_cm, weight_kg

Demographics for the Enright reference equation (all four required for the predicted value).

ref, covariates

Optional normative reference / covariates for a z-score.

Value

A "performance_test_score" with distance, and (when demographics are given) predicted, percent_predicted, lln and below_lln.

References

Enright PL, Sherrill DL (1998). Am J Respir Crit Care Med 158(5), 1384-1387. doi:10.1164/ajrccm.158.5.9710086

Examples

score6MWT(450, age = 60, sex = "male", height_cm = 175, weight_kg = 75)
score6MWT(450, age = 60, sex = "male", height_cm = 175, weight_kg = 75)

Score the Action Research Arm Test (ARAT)

Description

Score the Action Research Arm Test (ARAT)

Usage

scoreARAT(items, ...)
scoreARAT(items, ...)

Arguments

items

Named numeric responses (0-3 per item) for the 19 ARAT items.

...

Passed to scoreInstrument().

Value

A ClinicalScore with the total (0-57) and the grasp/grip/pinch/ gross-movement subscale scores.

Score the Barthel Index (basic activities of daily living)

Description

The Barthel Index rates ten basic ADL domains on their conventional weighted scale (steps of 5; transfers and mobility to 15) for a total of 0-100, higher being more independent.

Usage

scoreBarthel(items, ...)
scoreBarthel(items, ...)

Arguments

items

Named numeric responses on the weighted per-item scale (e.g. feeding = 10, bathing = 5, transfers = 15), or an unnamed vector in the instrument's item order. Allowed values are the conventional weights (0/5/10, and 0/5/10/15 for transfers and mobility).

...

Passed to scoreInstrument() (e.g. missing, subject_id).

Value

A ClinicalScore with the total (0-100) and the dependence stratum (total / severe / moderate / slight dependence / independent).

Examples

scoreBarthel(stats::setNames(
  c(10, 5, 5, 10, 10, 10, 10, 15, 15, 10), getInstrument("barthel")@items))
scoreBarthel(stats::setNames(
  c(10, 5, 5, 10, 10, 10, 10, 15, 15, 10), getInstrument("barthel")@items))

Score the Berg Balance Scale (BBS)

Description

Score the Berg Balance Scale (BBS)

Usage

scoreBerg(items, ...)
scoreBerg(items, ...)

Arguments

items

Named numeric responses (0-4 per item) for the 14 Berg items.

...

Passed to scoreInstrument().

Value

A ClinicalScore with the total (0-56) and the fall-risk stratum.

Score the Brief Pain Inventory (short form)

Description

Scores the BPI severity (worst / least / average / current pain) and interference (general activity, mood, walking, work, relations, sleep, enjoyment of life) items, each 0-10. Severity and interference are reported as their mean subscale scores (each 0-10, the conventional BPI summaries); the total is the grand mean over all items.

Usage

scoreBPI(items, ...)
scoreBPI(items, ...)

Arguments

items

Named numeric responses (e.g. worst_pain = 8, interference_sleep = 5) or an unnamed vector in the instrument's item order.

...

Passed to scoreInstrument() (e.g. missing = "prorate").

Value

A ClinicalScore whose subscales hold the severity and interference means.

Examples

scoreBPI(stats::setNames(c(8, 2, 5, 4, 5, 6, 5, 7, 3, 5, 6),
                         getInstrument("bpi")@items))
scoreBPI(stats::setNames(c(8, 2, 5, 4, 5, 6, 5, 7, 3, 5, 6),
                         getInstrument("bpi")@items))

Score a cohort's instrument responses into interop-ready ClinicalScores

Description

Scores a per-subject table of instrument responses into a list of ClinicalScore objects, each tagged with its subject_id, ready to drive the FHIR / OMOP / CDISC exporters (writeFHIRBundle(), toOMOP(), toCDISC_QS()) directly. The subject_col and item columns can come straight from a cohort's data (e.g. a PhysioCohort subject table joined to captured responses).

Usage

scoreCohort(
  responses,
  instrument,
  subject_col = "subject_id",
  timestamp_col = NULL,
  missing = c("error", "prorate", "na")
)
scoreCohort(
  responses,
  instrument,
  subject_col = "subject_id",
  timestamp_col = NULL,
  missing = c("error", "prorate", "na")
)

Arguments

responses

A data.frame: one row per subject, with a subject-id column and one column per instrument item (columns not matching an item are ignored).

instrument

A ClinicalInstrument or an instrument id (resolved via getInstrument()).

subject_col

Name of the subject-id column (default "subject_id").

timestamp_col

Optional name of a timestamp column stored on each score.

missing

Missing-data policy passed to scoreInstrument() (default "error"; use "prorate" / "na" when only some items are present).

Value

A named list (by subject id) of ClinicalScore objects.

Examples

items <- getInstrument("katz_adl")@items
resp <- data.frame(subject_id = c("P01", "P02"),
  stats::setNames(as.data.frame(rbind(rep(1, 6), c(1, 0, 1, 1, 0, 1))), items))
scores <- scoreCohort(resp, "katz_adl")
vapply(scores, function(s) s@total, numeric(1))
items <- getInstrument("katz_adl")@items
resp <- data.frame(subject_id = c("P01", "P02"),
  stats::setNames(as.data.frame(rbind(rep(1, 6), c(1, 0, 1, 1, 0, 1))), items))
scores <- scoreCohort(resp, "katz_adl")
vapply(scores, function(s) s@total, numeric(1))

Score the EQ-5D-5L descriptive system (level-sum score)

Description

Sums the five EQ-5D-5L dimension levels (1 = no problems to 5 = extreme problems / unable) to a level-sum score of 5-25, higher indicating worse health. The utility index requires a country-specific value set and is not computed here; the EQ-VAS (0-100) is a separate rating.

Usage

scoreEQ5D5L(items, ...)
scoreEQ5D5L(items, ...)

Arguments

items

Named numeric levels (e.g. mobility = 2, pain_discomfort = 3) or an unnamed vector in the instrument's item order (mobility, self-care, usual activities, pain/discomfort, anxiety/depression).

...

Passed to scoreInstrument().

Value

A ClinicalScore with the level-sum score (5-25).

Examples

scoreEQ5D5L(c(mobility = 2, self_care = 1, usual_activities = 2,
              pain_discomfort = 3, anxiety_depression = 2))
scoreEQ5D5L(c(mobility = 2, self_care = 1, usual_activities = 2,
              pain_discomfort = 3, anxiety_depression = 2))

Score the Functional Assessment Measure (FIM+FAM)

Description

Score the Functional Assessment Measure (FIM+FAM)

Usage

scoreFAM(items, ...)
scoreFAM(items, ...)

Arguments

items

Named numeric responses (1-7 per item) for the 30 FIM+FAM items.

...

Passed to scoreInstrument().

Value

A ClinicalScore with the total (30-210) and the combined motor (16 items) and cognitive (14 items) subscale scores.

Score the Functional Independence Measure (FIM)

Description

Score the Functional Independence Measure (FIM)

Usage

scoreFIM(items, ...)
scoreFIM(items, ...)

Arguments

items

Named numeric responses (1-7 per item) for the 18 FIM items.

...

Passed to scoreInstrument().

Value

A ClinicalScore with the total (18-126) and subscale scores for the six FIM domains (self-care, sphincter, transfers, locomotion, communication, social cognition) plus the combined motor (13 items) and cognitive (5 items) scores.

Examples

scoreFIM(stats::setNames(rep(4, 18), getInstrument("fim")@items))
scoreFIM(stats::setNames(rep(4, 18), getInstrument("fim")@items))

Score the Fugl-Meyer Assessment, Lower Extremity (FMA-LE)

Description

Score the Fugl-Meyer Assessment, Lower Extremity (FMA-LE)

Usage

scoreFMALE(items, ...)
scoreFMALE(items, ...)

Arguments

items

Named numeric responses (0-2 per item) for the 17 FMA-LE motor items, or an unnamed vector in the instrument's item order.

...

Passed to scoreInstrument().

Value

A ClinicalScore with the total (0-34) and the motor and coordination/speed subscale scores.

Examples

scoreFMALE(stats::setNames(rep(2, 17), getInstrument("fma_le")@items))
scoreFMALE(stats::setNames(rep(2, 17), getInstrument("fma_le")@items))

Score the Fugl-Meyer Assessment, Upper Extremity (FMA-UE)

Description

Score the Fugl-Meyer Assessment, Upper Extremity (FMA-UE)

Usage

scoreFMAUE(items, ...)
scoreFMAUE(items, ...)

Arguments

items

Named numeric responses (0-2 per item) for the 33 FMA-UE motor items, or an unnamed vector in the instrument's item order.

...

Passed to scoreInstrument() (e.g. missing, subject_id).

Value

A ClinicalScore with the total (0-66), the four subscale scores (shoulder/elbow/forearm, wrist, hand, coordination/speed) and the severe/moderate/mild stratum (Woodbury 2013).

Examples

scoreFMAUE(stats::setNames(rep(2, 33), getInstrument("fma_ue")@items))
scoreFMAUE(stats::setNames(rep(2, 33), getInstrument("fma_ue")@items))

Score the Fatigue Severity Scale (FSS)

Description

Averages the nine FSS statements (each rated 1 = strongly disagree to 7 = strongly agree) to a mean of 1-7. A mean of 4 or more indicates clinically significant fatigue.

Usage

scoreFSS(items, ...)
scoreFSS(items, ...)

Arguments

items

Named numeric responses (item1 .. item9) or an unnamed vector in the instrument's item order.

...

Passed to scoreInstrument() (e.g. missing = "prorate").

Value

A ClinicalScore with the mean (1-7) and the fatigue stratum.

Examples

scoreFSS(stats::setNames(c(6, 6, 5, 6, 5, 6, 6, 5, 6),
                         getInstrument("fss")@items))
scoreFSS(stats::setNames(c(6, 6, 5, 6, 5, 6, 6, 5, 6),
                         getInstrument("fss")@items))

Score Goal Attainment Scaling (GAS T-score)

Description

Computes the standardized GAS T-score across a set of goals given their attained levels:

$T = 50 + \frac{10 \sum_i w_i x_i}{\sqrt{(1-\rho)\sum_i w_i^2 + \rho (\sum_i w_i)^2}}$

where $x_i$ is the attained level, $w_i$ the weight and $\rho$ the assumed inter-goal correlation. All goals at their expected level ( $x_i = 0$ ) give $T = 50$ .

Usage

scoreGAS(goals, attained_levels, rho = 0.3)
scoreGAS(goals, attained_levels, rho = 0.3)

Arguments

goals

A defineGoal object or a list of them.

attained_levels

Numeric attained level for each goal (each within that goal's levels).

rho

Assumed inter-goal correlation in [0, 1); default 0.3.

Value

A "gas_result" with the T-score, per-goal weights/levels and rho.

References

Kiresuk & Sherman (1968); Turner-Stokes (2009).

Examples

g1 <- defineGoal("Transfers", importance = 3, difficulty = 2)
g2 <- defineGoal("Stairs", importance = 2, difficulty = 3)
scoreGAS(list(g1, g2), attained_levels = c(1, 0))
g1 <- defineGoal("Transfers", importance = 3, difficulty = 2)
g2 <- defineGoal("Stairs", importance = 2, difficulty = 3)
scoreGAS(list(g1, g2), attained_levels = c(1, 0))

Score item responses against a clinical instrument

Description

Validates a set of item responses against a ClinicalInstrument, aggregates them into the total and subscale scores under the chosen missing-data policy, and assigns the interpretation stratum.

Usage

scoreInstrument(
  instrument,
  items,
  missing = c("error", "prorate", "na"),
  subject_id = NA_character_,
  timestamp = NA_character_
)
scoreInstrument(
  instrument,
  items,
  missing = c("error", "prorate", "na"),
  subject_id = NA_character_,
  timestamp = NA_character_
)

Arguments

instrument

A ClinicalInstrument, or an instrument id resolved via getInstrument().

items

A named numeric vector (item id -> response), a one-row data.frame, or an unnamed vector in the instrument's item order.

missing

Missing-data policy: "error" (default; any missing item is an error), "prorate" (scale the observed items up to the full count) or "na" (return NA for any score with a missing item).

subject_id

Optional subject identifier stored on the result.

timestamp

Optional timestamp stored on the result.

Value

A ClinicalScore.

Examples

inst <- ClinicalInstrument(
  id = "toy", items = c("q1", "q2", "q3"),
  item_ranges = list(q1 = c(0, 4), q2 = c(0, 4), q3 = c(0, 4)))
scoreInstrument(inst, c(q1 = 2, q2 = 3, q3 = 4))
inst <- ClinicalInstrument(
  id = "toy", items = c("q1", "q2", "q3"),
  item_ranges = list(q1 = c(0, 4), q2 = c(0, 4), q3 = c(0, 4)))
scoreInstrument(inst, c(q1 = 2, q2 = 3, q3 = 4))

Score the Katz Index of Independence in ADL

Description

Six basic ADL activities (bathing, dressing, toileting, transferring, continence, feeding), each scored independent (1) or dependent (0), for a total of 0-6.

Usage

scoreKatz(items, ...)
scoreKatz(items, ...)

Arguments

items

Named numeric responses (0/1 per activity) or an unnamed vector in the instrument's item order.

...

Passed to scoreInstrument().

Value

A ClinicalScore with the total (0-6) and the impairment stratum.

Examples

scoreKatz(c(bathing = 1, dressing = 1, toileting = 1,
            transferring = 1, continence = 0, feeding = 1))
scoreKatz(c(bathing = 1, dressing = 1, toileting = 1,
            transferring = 1, continence = 0, feeding = 1))

Score the Lawton-Brody IADL scale (instrumental ADL)

Description

Eight instrumental-ADL domains of independent community living (telephone, shopping, food preparation, housekeeping, laundry, transportation, medication, finances), each dichotomised independent (1) / dependent (0), summed to a 0-8 count. Lawton & Brody define no total-score severity bands, so the returned stratum is NA.

Usage

scoreLawton(items, ...)
scoreLawton(items, ...)

Arguments

items

Named numeric responses (0/1 per domain) or an unnamed vector in the instrument's item order. For the abbreviated male version, omit the food-preparation, housekeeping and laundry items and use missing = "na".

...

Passed to scoreInstrument().

Value

A ClinicalScore with the total (0-8).

Examples

scoreLawton(stats::setNames(rep(1, 8), getInstrument("lawton_iadl")@items))
scoreLawton(stats::setNames(rep(1, 8), getInstrument("lawton_iadl")@items))

Score a Modified Ashworth Scale (MAS) rating

Description

Accepts a MAS rating as its ordinal level label ("0", "1", "1+", "2", "3", "4") or the equivalent number, mapping "1+" to 1.5 for arithmetic while preserving the label. Rate one muscle group per call.

Usage

scoreMAS(rating, ...)
scoreMAS(rating, ...)

Arguments

rating

A single MAS level label or number.

...

Passed to scoreInstrument().

Value

A ClinicalScore whose total is the numeric MAS value; the ordinal label is recovered with masLevelLabel().

Examples

scoreMAS("1+")
scoreMAS("1+")

Score the Modified Fatigue Impact Scale (MFIS)

Description

Sums the twenty-one MFIS items (each 0-4) to a total of 0-84 with physical (9-item), cognitive (10-item) and psychosocial (2-item) subscales. A total of 38 or more indicates fatigue.

Usage

scoreMFIS(items, ...)
scoreMFIS(items, ...)

Arguments

items

Named numeric responses (item01 .. item21) or an unnamed vector in the instrument's item order.

...

Passed to scoreInstrument() (e.g. missing = "prorate").

Value

A ClinicalScore with the total (0-84), the three subscale scores and the fatigue stratum.

Examples

scoreMFIS(stats::setNames(rep(2, 21), getInstrument("mfis")@items))
scoreMFIS(stats::setNames(rep(2, 21), getInstrument("mfis")@items))

Score the Mini-Mental State Examination (MMSE)

Description

Sums the eleven MMSE components (orientation to time and place, registration, attention/calculation, recall, naming, repetition, a three-stage command, reading, writing and figure copying) to a total of 0-30, higher being better cognition.

Usage

scoreMMSE(items, ...)
scoreMMSE(items, ...)

Arguments

items

Named numeric responses giving the points earned on each component (e.g. orientation_time = 5, recall = 2), or an unnamed vector in the instrument's item order.

...

Passed to scoreInstrument() (e.g. missing, subject_id).

Value

A ClinicalScore with the total (0-30), the six-domain subscale scores and the impairment stratum (no / mild / severe impairment).

Examples

scoreMMSE(stats::setNames(c(5, 5, 3, 5, 3, 2, 1, 3, 1, 1, 1),
                          getInstrument("mmse")@items))
scoreMMSE(stats::setNames(c(5, 5, 3, 5, 3, 2, 1, 3, 1, 1, 1),
                          getInstrument("mmse")@items))

Score the Montreal Cognitive Assessment (MoCA)

Description

Sums the seven MoCA domains (visuospatial/executive, naming, attention, language, abstraction, delayed recall and orientation) to a total of 0-30. A total below 26 indicates cognitive impairment. One point is conventionally added for participants with 12 or fewer years of formal education (total capped at 30); apply that adjustment to the orientation item (or the total) before or after scoring as appropriate for your protocol.

Usage

scoreMoCA(items, ...)
scoreMoCA(items, ...)

Arguments

items

Named numeric responses giving the points earned on each domain (e.g. delayed_recall = 3, orientation = 6), or an unnamed vector in the instrument's item order.

...

Passed to scoreInstrument().

Value

A ClinicalScore with the total (0-30) and the impairment stratum (normal / mild / moderate-severe impairment).

Examples

scoreMoCA(stats::setNames(c(5, 3, 6, 3, 2, 5, 6),
                          getInstrument("moca")@items))
scoreMoCA(stats::setNames(c(5, 3, 6, 3, 2, 5, 6),
                          getInstrument("moca")@items))

Score the Medical Research Council (MRC) sum score

Description

Sums manual muscle-test grades (0 = no contraction to 5 = normal power) for six movements tested bilaterally - shoulder abduction, elbow flexion, wrist extension, hip flexion, knee extension and ankle dorsiflexion - to a total of 0-60, higher being stronger. A total below 48 defines ICU-acquired weakness.

Usage

scoreMRCSum(items, ...)
scoreMRCSum(items, ...)

Arguments

items

Named numeric grades (e.g. elbow_flexion_r = 4, ankle_dorsiflexion_l = 3) or an unnamed vector in the instrument's item order.

...

Passed to scoreInstrument() (e.g. missing = "prorate").

Value

A ClinicalScore with the total (0-60), the upper-limb / lower-limb / left / right subscale sums and the weakness stratum.

Examples

scoreMRCSum(stats::setNames(rep(5, 12), getInstrument("mrc_sum")@items))
scoreMRCSum(stats::setNames(rep(5, 12), getInstrument("mrc_sum")@items))

Score the modified Rankin Scale (mRS)

Description

Score the modified Rankin Scale (mRS)

Usage

scoreMRS(grade, ...)
scoreMRS(grade, ...)

Arguments

grade

A single global disability grade, 0-6.

...

Passed to scoreInstrument().

Value

A ClinicalScore whose total is the grade and whose stratum is the disability label.

Examples

scoreMRS(3)
scoreMRS(3)

Score the NIH Stroke Scale (NIHSS)

Description

Score the NIH Stroke Scale (NIHSS)

Usage

scoreNIHSS(items, ...)
scoreNIHSS(items, ...)

Arguments

items

Named numeric responses for the 15 NIHSS components (each within its own range).

...

Passed to scoreInstrument().

Value

A ClinicalScore with the total (0-42) and the severity band (no stroke / minor / moderate / moderate-to-severe / severe).

Score a Pain Numeric Rating Scale (0-10)

Description

A single 0-10 pain-intensity rating, banded into the conventional none / mild / moderate / severe strata. An equivalent 0-100 mm Visual Analogue Scale measures the same construct on a wider range.

Usage

scorePainNRS(items, ...)
scorePainNRS(items, ...)

Arguments

items

The pain rating: a length-1 numeric, or a named vector c(pain_intensity = x).

...

Passed to scoreInstrument() (e.g. subject_id).

Value

A ClinicalScore with the rating (0-10) and its severity stratum.

Examples

scorePainNRS(c(pain_intensity = 6))
scorePainNRS(c(pain_intensity = 6))

Score a PROMIS (or other GRM) short form by EAP

Description

Estimates the latent trait for a set of item responses under Samejima's graded response model by expected a posteriori (EAP) scoring over a normal quadrature, and reports it on the PROMIS T-score metric (mean 50, SD 10).

Usage

scorePROMIS(
  responses,
  calibration,
  quad_points = 61,
  quad_range = 4,
  prior_mean = 0,
  prior_sd = 1
)
scorePROMIS(
  responses,
  calibration,
  quad_points = 61,
  quad_range = 4,
  prior_mean = 0,
  prior_sd = 1
)

Arguments

responses

Named numeric vector of item responses (category 1..K per item; NA for a skipped item), or a one-row data.frame. Names must match calibration$item; an unnamed vector is taken in calibration row order.

calibration

A data.frame with columns item, a (slope) and ordered threshold columns b1, b2, ... (a K-category item uses K-1 thresholds; pad shorter items with NA).

quad_points, quad_range

EAP quadrature: number of nodes and half-width (nodes span ⁠[-quad_range, quad_range]⁠; defaults 61 and 4).

prior_mean, prior_sd

Normal prior on the latent trait (default standard normal, the PROMIS calibration metric).

Details

No item parameters are bundled: supply the official PROMIS calibration (slopes and thresholds) from HealthMeasures. Any GRM-scored instrument works with the same function.

Value

A list with theta, se_theta, tscore (50 + 10 * theta), se_tscore and n_items (items actually scored).

References

Samejima F (1969). Estimation of latent ability using a response pattern of graded scores. Psychometrika Monograph 17. PROMIS scoring: HealthMeasures, www.healthmeasures.net.

Examples

# illustrative calibration (NOT official PROMIS parameters)
cal <- data.frame(item = c("i1", "i2", "i3"), a = c(2.4, 1.9, 2.7),
  b1 = c(-2, -1.5, -1.8), b2 = c(-1, -0.5, -0.7),
  b3 = c(0.2, 0.1, 0), b4 = c(1.4, 1.2, 1.1))
scorePROMIS(c(i1 = 4, i2 = 5, i3 = 4), cal)
# illustrative calibration (NOT official PROMIS parameters)
cal <- data.frame(item = c("i1", "i2", "i3"), a = c(2.4, 1.9, 2.7),
  b1 = c(-2, -1.5, -1.8), b2 = c(-1, -0.5, -0.7),
  b3 = c(0.2, 0.1, 0), b4 = c(1.4, 1.2, 1.1))
scorePROMIS(c(i1 = 4, i2 = 5, i3 = 4), cal)

Score the SF-36 / RAND-36 Health Survey

Description

Scores the RAND 36-Item Health Survey 1.0: each item's raw response is recoded to 0-100 (via the instrument's item_recode map) and averaged within its scale, giving the eight subscale scores (0-100, higher = better health) in the returned subscales: physical functioning, role-physical, role-emotional, energy/fatigue, emotional well-being, social functioning, pain and general health. The health-transition item is not part of the eight scales. The norm-based T-scores and the PCS/MCS component summaries use proprietary population weights and are out of scope.

Usage

scoreSF36(items, ...)
scoreSF36(items, ...)

Arguments

items

Named numeric raw responses on the questionnaire's own answer scales (q01, q03..q36, omitting the health-transition item q02), or an unnamed vector in the instrument's item order.

...

Passed to scoreInstrument() (e.g. missing = "prorate" to score a scale from its completed items).

Value

A ClinicalScore whose subscales hold the eight 0-100 scale scores (the total is the grand mean over all items, not a standard SF-36 score).

References

Hays RD, Sherbourne CD, Mazel RM (1993). The RAND 36-Item Health Survey 1.0. Health Econ 2:217-227.

Examples

# all best-health responses -> every subscale scores 100
best <- stats::setNames(
  c(1, rep(3, 10), rep(2, 7), 1, 1, 1, 1, 6, 6, 1, 1, 6, 6, 1, 6, 5, 5, 1, 5, 1),
  getInstrument("sf36")@items)
scoreSF36(best)@subscales
# all best-health responses -> every subscale scores 100
best <- stats::setNames(
  c(1, rep(3, 10), rep(2, 7), 1, 1, 1, 1, 6, 6, 1, 1, 6, 6, 1, 6, 5, 5, 1, 5, 1),
  getInstrument("sf36")@items)
scoreSF36(best)@subscales

Score a Timed Up and Go test (TUG)

Description

Classifies TUG time against the standard fall-risk cut-offs (< 10 s normal; >= 13.5 s elevated community-dwelling fall risk, Shumway-Cook 2000; >= 30 s likely dependent, Podsiadlo & Richardson 1991), with an optional normative z-score.

Usage

scoreTUG(x, ref = NULL, covariates = list())
scoreTUG(x, ref = NULL, covariates = list())

Arguments

x

A TUG time in seconds, or a PhysioMoCap itug_report (its total_duration is used).

ref, covariates

Optional normative reference / covariates for a z-score.

Value

A "performance_test_score" with time, category and fall_risk (logical, >= 13.5 s).

References

Podsiadlo & Richardson (1991); Shumway-Cook et al. (2000).

Examples

scoreTUG(15)   # >= 13.5 s -> elevated fall risk
scoreTUG(15)   # >= 13.5 s -> elevated fall risk

Score the Wolf Motor Function Test functional-ability scale (WMFT-FAS)

Description

Score the Wolf Motor Function Test functional-ability scale (WMFT-FAS)

Usage

scoreWMFT(fas, ...)
scoreWMFT(fas, ...)

Arguments

fas

Named numeric responses (0-5 per task) for the 15 functional tasks.

...

Passed to scoreInstrument().

Value

A ClinicalScore whose total is the mean functional-ability score.

Show a normative reference (surfacing its governance)

Description

Show a normative reference (surfacing its governance)

Usage

## S4 method for signature 'GovernedNormativeReference'
show(object)
## S4 method for signature 'GovernedNormativeReference'
show(object)

Arguments

object

A GovernedNormativeReference.

Value

object, invisibly.

Export clinical scores to an ADaM BDS (ADQS) data frame

Description

Restructures a QS domain (or scores) into an ADaM Basic Data Structure analysis data set (one PARAMCD per test): AVAL/AVALC carry the numeric/character result and PARCAT1 the questionnaire category.

Usage

toADaM_ADQS(x, ...)
toADaM_ADQS(x, ...)

Arguments

x

A ClinicalScore, a list of them, or a QS-domain data frame from toCDISC_QS().

...

Passed to toCDISC_QS() when x is score(s).

Value

An ADaM BDS data.frame (columns STUDYID, USUBJID, PARAMCD, PARAM, PARCAT1, AVAL, AVALC, AVISIT, AVISITN, ADT).

References

CDISC ADaM Basic Data Structure (BDS).

Export clinical scores to a CDISC SDTM QS (Questionnaires) domain

Description

Maps one or more ClinicalScore objects to a CDISC SDTMIG Questionnaires (QS) domain data frame: the total and each subscale becomes a QS record. QSTESTCD/QSTEST/QSCAT come from the packaged controlled-terminology map (cdisc_ct.csv); an unmapped instrument gets a sponsor-defined QSTESTCD (upper-cased, forced to start with a letter, disambiguated against collisions, and truncated to the SDTM 8-character limit) with a warning. QSTESTCD values in the shipped map are sponsor-defined (ct_status = "sponsor"), not asserted to be published CDISC CT.

Usage

toCDISC_QS(
  scores,
  ct_map = NULL,
  studyid = "STUDY",
  usubjid_map = NULL,
  visitnum = 1L,
  visit = NA_character_
)
toCDISC_QS(
  scores,
  ct_map = NULL,
  studyid = "STUDY",
  usubjid_map = NULL,
  visitnum = 1L,
  visit = NA_character_
)

Arguments

scores

A ClinicalScore or a list of them.

ct_map

Optional data frame overriding the packaged CT map (columns instrument, subscale, qstestcd, qstest, qscat).

studyid

STUDYID value (default "STUDY").

usubjid_map

Optional named vector/list mapping subject_id to a USUBJID; otherwise USUBJID = <studyid>-<subject_id>.

visitnum, visit

VISITNUM / VISIT for all records.

Value

A QS-domain data.frame conforming to the SDTMIG column layout.

References

CDISC SDTMIG QS domain.

Examples

sc <- methods::new("ClinicalScore", instrument_id = "fim", total = 90,
                   subscales = c(motor = 60, cognitive = 30),
                   subject_id = "01-001", timestamp = "2026-07-26")
toCDISC_QS(sc)
sc <- methods::new("ClinicalScore", instrument_id = "fim", total = 90,
                   subscales = c(motor = 60, cognitive = 30),
                   subject_id = "01-001", timestamp = "2026-07-26")
toCDISC_QS(sc)

Export a clinical score as a FHIR R4 Observation

Description

Serializes a ClinicalScore into an HL7 FHIR R4 Observation resource (as a jsonlite-ready list): a survey-category Observation whose code uses the LOINC map (falling back to a text-only code), the total as valueQuantity, and each subscale as a component. Provenance can be attached via derivedFrom.

Usage

toFHIRObservation(
  score,
  subject = NULL,
  effectiveDateTime = NULL,
  derivedFrom = NULL
)
toFHIRObservation(
  score,
  subject = NULL,
  effectiveDateTime = NULL,
  derivedFrom = NULL
)

Arguments

score

A ClinicalScore.

subject

Subject reference (e.g. "Patient/123"); used verbatim. Defaults to the score's subject_id: a bare id becomes "Patient/<id>", while an id that is already a reference (contains a "/" or a urn:/URL scheme) is used as-is.

effectiveDateTime

ISO-8601 datetime; defaults to the score's timestamp.

derivedFrom

Optional character vector of source references.

Value

A "fhir_observation" list.

References

HL7 FHIR R4 Observation; LOINC.

Examples

sc <- methods::new("ClinicalScore", instrument_id = "berg", total = 45,
                   subject_id = "P01", timestamp = "2026-07-26T09:00:00Z")
toFHIRObservation(sc)
sc <- methods::new("ClinicalScore", instrument_id = "berg", total = 45,
                   subject_id = "P01", timestamp = "2026-07-26T09:00:00Z")
toFHIRObservation(sc)

Export clinical scores to OMOP CDM v5.4 tables

Description

Maps one or more ClinicalScore objects to OMOP Common Data Model v5.4 MEASUREMENT and OBSERVATION rows. The total and each subscale become a row, routed to MEASUREMENT or OBSERVATION by the concept map's domain_id. Concept ids are looked up in the packaged omop_concept_map.csv; an unmapped instrument yields measurement_concept_id = 0 (the OMOP "No matching concept") with a warning. No Athena concept ids are fabricated — the shipped map leaves them blank; supply a site concept_map to populate them.

Usage

toOMOP(scores, concept_map = NULL, person_map = NULL, type_concept_id = 0L)
toOMOP(scores, concept_map = NULL, person_map = NULL, type_concept_id = 0L)

Arguments

scores

A ClinicalScore or a list of them.

concept_map

Optional data frame overriding the packaged concept map (columns instrument, subscale, domain_id, measurement_concept_id, unit_concept_id).

person_map

Optional named vector/list mapping subject_id to an integer person_id. When absent, a plain positive-integer subject_id (matching "^[0-9]+$") is used directly; any other id is left as NA with a warning.

type_concept_id

Integer *_type_concept_id for provenance (default 0L = "No matching concept"); set to your site's type concept (e.g. EHR).

Value

A list with MEASUREMENT and OBSERVATION data frames conforming to the OMOP CDM v5.4 column layout.

References

OHDSI OMOP CDM v5.4 (MEASUREMENT, OBSERVATION).

Examples

sc <- methods::new("ClinicalScore", instrument_id = "berg", total = 45,
                   subject_id = "1001", timestamp = "2026-07-26T09:00:00Z")
suppressWarnings(toOMOP(sc))
sc <- methods::new("ClinicalScore", instrument_id = "berg", total = 45,
                   subject_id = "1001", timestamp = "2026-07-26T09:00:00Z")
suppressWarnings(toOMOP(sc))

Validate a FHIR Observation against the bundled JSON schema

Description

Validate a FHIR Observation against the bundled JSON schema

Usage

validateFHIRObservation(obs)
validateFHIRObservation(obs)

Arguments

obs

An "fhir_observation" (or a list).

Value

TRUE if valid; otherwise FALSE with the validation errors as an attribute (requires jsonvalidate).

Validate a normative-artifact governance manifest

Description

Checks that a normative reference's manifest carries every required governance field — a semantic version, and non-empty provenance$source, consent$status, license$spdx, governance$custodian and governance$access_level — so an ungoverned artifact cannot be registered or distributed.

Usage

validateNormativeManifest(manifest)
validateNormativeManifest(manifest)

Arguments

manifest

A manifest list, a GovernedNormativeReference, or a path to a manifest.json file.

Value

TRUE invisibly if valid; otherwise an error naming the missing or invalid fields.

Examples

ref <- GovernedNormativeReference("m", "gait", "gait_speed",
  provenance = list(source = "x"), consent = list(status = "public"),
  license = list(spdx = "CC0-1.0"),
  governance = list(custodian = "lab", access_level = "open"))
validateNormativeManifest(ref)
ref <- GovernedNormativeReference("m", "gait", "gait_speed",
  provenance = list(source = "x"), consent = list(status = "public"),
  license = list(spdx = "CC0-1.0"),
  governance = list(custodian = "lab", access_level = "open"))
validateNormativeManifest(ref)

Median WMFT performance time

Description

Summarises the Wolf Motor Function Test timed-task performance as the median time across tasks, the conventional WMFT time summary.

Usage

wmftMedianTime(times)
wmftMedianTime(times)

Arguments

times

Numeric vector of per-task performance times (seconds).

Value

The median time (seconds).

Write a FHIR Bundle of clinical-score Observations

Description

Write a FHIR Bundle of clinical-score Observations

Usage

writeFHIRBundle(scores, path)
writeFHIRBundle(scores, path)

Arguments

scores

A ClinicalScore, an "fhir_observation", or a list of either.

path

Output .json path.

Value

path, invisibly.

Write OMOP CDM tables to CSV files

Description

Writes each non-empty table from toOMOP() to <path>/<table>.csv (lower-cased, e.g. measurement.csv), following the one-file-per-table OMOP convention.

Usage

writeOMOPTables(tables, path)
writeOMOPTables(tables, path)

Arguments

tables

A ClinicalScore, a list of them, or the list returned by toOMOP().

path

Output directory (created if needed).

Value

A character vector of written file paths, invisibly.

Package 'PhysioClinical'

Help Index

Apply a score crosswalk to new values

Description

Usage

Arguments

Value

See Also

Examples

Convert a ClinicalScore to a one-row data.frame

Description

Usage

Arguments

Value

Dual MDC-vs-MCID responder classification

Description

Usage

Arguments

Value

References

See Also

Examples

Construct a ClinicalInstrument

Description

Usage

Arguments

Value

See Also

Examples

Clinical instrument specification

Description

Usage

Arguments

Slots

Clinical score result

Description

Usage

Arguments

Slots

Define a Goal Attainment Scaling goal

Description

Usage

Arguments

Value

References

See Also

Examples

Equate one clinical score scale to another (build a crosswalk)

Description

Usage

Arguments

Value

References

See Also

Examples

Anchor-based MCID

Description

Usage

Arguments

Value

References

See Also

Examples

Distribution-based MCID (fraction-of-SD rule)

Description

Usage

Arguments

Value

References

See Also

Examples

Distribution-based Minimal Detectable Change from test-retest data

Description

Usage

Arguments

Value

References

See Also

Examples

Import a FHIR R4 Observation as a clinical score