Package 'mwana'

Title: An Efficient Workflow for Plausibility Checks and Prevalence Analysis of Wasting in R
Description: A simple and streamlined workflow for plausibility checks and prevalence analysis of wasting based on the Standardized Monitoring and Assessment of Relief and Transition (SMART) Methodology <https://smartmethodology.org/>, with application in R.
Authors: Tomás Zaba [aut, cre, cph] , Ernest Guevarra [aut, cph]
Maintainer: Tomás Zaba <[email protected]>
License: GPL (>= 3)
Version: 0.2.0
Built: 2024-11-19 20:24:04 UTC
Source: https://github.com/nutriverse/mwana

Help Index

A sample data of district level SMART surveys with location anonymised

Description

anthro.01 is a two-stage cluster-based survey with probability of selection of clusters proportional to the size of the population. The survey employed the SMART methodology.

Usage

anthro.01

Format

A tibble of 1,191 rows and 11 columns.

Variable Description
area Location where the survey took place
dos Survey date
cluster Primary sampling unit
team Enumerator IDs
sex Sex, "m" = boys, "f" = girls
dob Date of birth
age Age in months, typically estimated using local event calendars
weight Weight (kg)
height Height (cm)
edema Edema, "n" = no, "y" = yes
muac Mid-upper arm circumference (mm)

Source

Anonymous

Examples

anthro.01

A sample of an already wrangled survey data

Description

A household budget survey data conducted in Mozambique in 2019/2020, known as IOF (Inquérito ao Orçamento Familiar in Portuguese). IOF is a two-stage cluster-based survey, representative at province level (admin 2), with probability of the selection of the clusters proportional to the size of the population. Its data collection spans for a period of 12 months.

Usage

anthro.02

Format

A tibble of 2,267 rows and 14 columns.

Variable Description
province The administrative unit (admin 1) where data was collected.
strata Rural and Urban
cluster Primary sampling unit
sex Sex, "m" = boys, "f" = girls
age calculated age in months with two decimal places
weight Weight (kg)
height Height (cm)
edema Edema, "n" = no, "y" = yes
muac Mid-upper arm circumference (mm)
wtfactor Survey weights
wfhz Weight-for-height z-scores with 3 decimal places
flag_wfhz Flagged observations. 1=flagged, 0=not flagged
mfaz MUAC-for-age z-scores with 3 decimal places
flag_mfaz Flagged observations. 1=flagged, 0=not flagged

Source

Mozambique National Institute of Statistics. The data is publicly available at https://mozdata.ine.gov.mz/index.php/catalog/88#metadata-data_access. Data was wrangled using this package's wranglers. Details about survey design can be gotten from: https://mozdata.ine.gov.mz/index.php/catalog/88#metadata-sampling

Examples

anthro.02

A sample data of district level SMART surveys conducted in Mozambique

Description

anthro.03 contains survey data of four districts. Each district data set presents distinct data quality scenarios that requires tailored prevalence analysis approach: two districts show a problematic WFHZ standard deviation whilst the remaining are all within range.

This sample data is useful to demonstrate the use of the prevalence functions on a multiple-area survey data where there can be variations in the rating of acceptability of the standard deviation, hence require different analyses approaches for each area to ensure accurate estimation.

Usage

anthro.03

Format

A tibble of 943 x 9.

Variable Description
district The location where data was collected
cluster Primary sampling unit
team Survey teams
sex Sex, "m" = boys, "f" = girls
age calculated age in months with two decimal places
weight Weight (kg)
height Height (cm)
edema Edema, "n" = no, "y" = yes
muac Mid-upper arm circumference (mm)

Source

Anonymous

Examples

anthro.03

A sample data of a community-based sentinel site from an anonymized location

Description

Data was generated through a community-based sentinel site conducted across three provinces. Each province's data set presents distinct data quality scenarios, requiring tailored prevalence analysis:

  • "Province 1" has MFAZ's standard deviation and age ratio test rating of acceptability falling within range;

  • "Province 2" has age ratio rated as problematic but with an acceptable standard deviation of MFAZ;

  • "Province 3" has both tests rated as problematic.

This sample data is useful to demonstrate the use of prevalence functions on a multiple-area survey data where variations in the rating of acceptability of the standard deviation exist, hence require different analyses approaches for each area to ensure accurate estimation.

Usage

anthro.04

Format

A tibble of 3,002 x 8.

Variable Description
province location where data was collected
cluster Primary sampling unit
sex Sex, "m" = boys, "f" = girls
age calculated age in months with two decimal places
muac Mid-upper arm circumference (mm)
edema Edema, "n" = no, "y" = yes
mfaz MUAC-for-age z-scores with 3 decimal places
flag_mfaz Flagged observations. 1=flagged, 0=not flagged

Source

Anonymous

Examples

anthro.04

Define wasting

Description

Define if a given observation in the data set is wasted or not, and its respective form of wasting (global, severe or moderate) on the basis of z-scores of weight-for-height (WFHZ), muac-for-age (MFAZ), raw MUAC values and combined case-definition.

Usage

define_wasting(
  df,
  zscores = NULL,
  muac = NULL,
  edema = NULL,
  .by = c("zscores", "muac", "combined")
)

Arguments

df

A data set object of class data.frame to use. It must have been wrangled using this package's wrangling functions for WFHZ or MUAC, or both (for combined) as appropriate.
zscores

A vector of class double of WFHZ or MFAZ values. If the class does not match the expected type, the function will stop execution and return an error message indicating the type of mismatch.
muac

A vector of class integer or numeric of raw MUAC values in millimeters. If the class does not match the expected type, the function will stop execution and return an error message indicating the type of mismatch.
edema

A vector of class character of edema. Default is NULL. If the class does not match the expected type, the function will stop execution and return an error message indicating the type of mismatch. Code values should be "y" for presence and "n" for absence of bilateral edema. If different, the function will stop execution and return an error indicating the issue.
.by

A choice of the criterion by which the case-definition should done. Choose zscores for WFHZ or MFAZ, muac for raw MUAC and combined for combined.

Value

Three vectors named gam, sam and mam, of class numeric, same length as inputs, containing dummy values: 1 for case and 0 for not case. This is added to df. When combined is selected, vector's names become cgam, csam and cmam.

Examples

## Case-definition by z-scores ----
z <- anthro.02 |>
  define_wasting(
    zscores = wfhz,
    muac = NULL,
    edema = edema,
    .by = "zscores"
  )
head(z)

## Case-definition by MUAC ----
m <- anthro.02 |>
  define_wasting(
    zscores = NULL,
    muac = muac,
    edema = edema,
    .by = "muac"
  )
head(m)

## Case-definition by combined ----
c <- anthro.02 |>
  define_wasting(
    zscores = wfhz,
    muac = muac,
    edema = edema,
    .by = "combined"
  )
head(c)

Identify, flag outliers and remove them

Description

Identify outlier z-scores for weight-for-height (WFHZ) and MUAC-for-age (MFAZ) following the SMART methodology. The function can also be used to detect outliers for height-for-age (HFAZ) and weight-for-age (WFAZ) z-scores following the same approach.

For raw MUAC values, outliers constitute values that are less than 100 millimeters or greater than 200 millimeters.

Removing outliers consist in setting the outlier record to NA and not necessarily to delete it from the data set. This is useful in the analysis procedures where outliers must be removed, such as the analysis of the standard deviation.

Usage

flag_outliers(x, .from = c("zscores", "raw_muac"))

remove_flags(x, .from = c("zscores", "raw_muac"))

Arguments

x

A vector of class numeric of WFHZ, MFAZ, HFAZ, WFAZ or raw MUAC values. The latter should be in millimeters. If the class is different than expected, the function will stop execution and return an error message indicating the type of mismatch.
.from

A choice between zscores and raw_muac for where outliers should be detected and flagged from.

Details

For z-score-based detection, flagged records represent outliers that deviate substantially from the sample's z-score mean, making them unlikely to reflect accurate measurements. For raw MUAC values, flagged records are those that fall outside the acceptable fixed range. Including such outliers in the analysis could compromise the accuracy and precision of the resulting estimates.

The flagging criterion used for raw MUAC values is based on a recommendation by Bilukha, O., & Kianian, B. (2023).

Value

A vector of the same length as x for flagged records coded as 1 for is a flag and 0 not a flag.

References

Bilukha, O., & Kianian, B. (2023). Considerations for assessment of measurement quality of mid‐upper arm circumference data in anthropometric surveys and mass nutritional screenings conducted in humanitarian and refugee settings. Maternal & Child Nutrition, 19, e13478. Available at https://doi.org/10.1111/mcn.13478

SMART Initiative (2017). Standardized Monitoring and Assessment for Relief and Transition. Manual 2.0. Available at: https://smartmethodology.org.

Examples

## Sample data of raw MUAC values ----
x <- anthro.01$muac

## Apply the function with `.from` set to "raw_muac" ----
m <- flag_outliers(x, .from = "raw_muac")
head(m)

## Sample data of z-scores (be it WFHZ, MFAZ, HFAZ or WFAZ) ----
x <- anthro.02$mfaz

# Apply the function with `.from` set to "zscores" ----
z <- flag_outliers(x, .from = "zscores")
tail(z)

## With `.from` set to "zscores" ----
z <- remove_flags(
  x = wfhz.01$wfhz,
  .from = "zscores"
)

head(z)

## With `.from` set to "raw_muac" ----
m <- remove_flags(
  x = mfaz.01$muac,
  .from = "raw_muac"
)

tail(m)

Calculate child's age in months

Description

Calculate child's age in months based on the date of birth and the date of data collection.

Usage

get_age_months(dos, dob)

Arguments

dos

A vector of class Date for the date of data collection. If the class is different than expected, the function will stop execution and return an error message indicating the type of mismatch.
dob

A vector of class Date for the child's date of birth. If the class is different than expected, the function will stop execution and return an error message indicating the type of mismatch.

Value

A vector of class numeric for child's age in months. Any value less than 6.0 and greater than or equal to 60.0 months will be set to NA.

Examples

## Take two vectors of class "Date" ----
surv_date <- as.Date(
  c(
    "2024-01-05", "2024-01-05", "2024-01-05", "2024-01-08", "2024-01-08",
    "2024-01-08", "2024-01-10", "2024-01-10", "2024-01-10", "2024-01-11"
  )
)
bir_date <- as.Date(
  c(
    "2022-04-04", "2021-05-01", "2023-05-24", "2017-12-12", NA,
    "2020-12-12", "2022-04-04", "2021-05-01", "2023-05-24", "2020-12-12"
  )
)

## Apply the function ----
get_age_months(
  dos = surv_date,
  dob = bir_date
)

A sample MUAC screening data from an anonymized setting

Description

A sample MUAC screening data from an anonymized setting

Usage

mfaz.01

Format

A tibble with 661 rows and 4 columns.

Variable Description
sex Sex, "m" = boys, "f" = girls
months calculated age in months with two decimal places
edema Edema, "n" = no, "y" = yes
muac Mid-upper arm circumference (mm)

Source

Anonymous

Examples

mfaz.01

A sample SMART survey data with MUAC

Description

A sample SMART survey data with MUAC

Usage

mfaz.02

Format

A tibble with 303 rows and 7 columns.

Variable Description
cluster Primary sampling unit
sex Sex, "m" = boys, "f" = girls
age calculated age in months with two decimal places
edema Edema, "n" = no, "y" = yes
mfaz MUAC-for-age z-scores with 3 decimal places
flag_mfaz Flagged observations. 1=flagged, 0=not flagged

Source

Anonymous

Examples

mfaz.02

Check whether IPC Acute Malnutrition (IPC AMN) sample size requirements were met

Description

Evidence on the prevalence of acute malnutrition used in the IPC AMN can come from different sources: surveys, screenings or community-based surveillance system. The IPC set minimum sample size requirements for each source. This function helps in verifying whether those requirements were met or not depending on the source.

Usage

mw_check_ipcamn_ssreq(df, cluster, .source = c("survey", "screening", "ssite"))

Arguments

df

A data set object of class data.frame to check.
cluster

A vector of class integer or character of unique cluster or screening or sentinel site IDs. If a character vector, ensure that names are correct and each name represents one location for accurate counts. If the class does not match the above expected type, the function will stop execution and return an error message indicating the type of mismatch.
.source

The source of evidence. A choice between "survey" for representative survey data at the area of analysis; "screening" for screening data; "ssite" for community-based sentinel site data.

Value

A summary table of class data.frame, of length 3 and width 1, for the check results. n_clusters is for the total number of unique clusters or screening or site IDs; n_obs for the correspondent total number of children in the data set; and meet_ipc for whether the IPC AMN requirements were met.

References

IPC Global Partners. 2021. Integrated Food Security Phase Classification Technical Manual Version 3.1.Evidence and Standards for Better Food Security and Nutrition Decisions. Rome. Available at: https://www.ipcinfo.org/ipcinfo-website/resources/ipc-manual/en/.

Examples

mw_check_ipcamn_ssreq(
  df = anthro.01,
  cluster = cluster,
  .source = "survey"
)

Estimate the prevalence of combined wasting

Description

Estimate the prevalence of wasting based on the combined case-definition of weight-for-height z-scores (WFHZ), MUAC and/or edema. The function allows users to get the prevalence estimates in accordance with the complex sample design properties; this includes applying survey weights when needed or applicable. Before estimating, the function evaluates the quality of data by calculating and rating the standard deviation of WFHZ and MFAZ, as well as the p-value of the age ratio test. Prevalence will be calculated only when the rating of all test is as not problematic concurrently. If either of them is problematic, it cancels out the analysis and NAs get thrown.

Outliers are detected in both WFHZ and in MUAC data set (through z-scores) based on SMART flags get excluded prior being piped into the actual prevalence analysis workflow.

Usage

mw_estimate_prevalence_combined(df, wt = NULL, edema = NULL, .by = NULL)

Arguments

df

A data set object of class data.frame to use. This must have been wrangled using this package's wrangling functions for both WFHZ and MUAC data sequentially. The order does not matter. Note that MUAC values should be converted to millimeters after using the MUAC wrangler. If this is not done, the function will stop execution and return an error message. Moreover, the function uses a variable called cluster where the primary sampling unit IDs are stored. Make sure to rename your cluster ID variable to cluster, otherwise the function will error and terminate the execution.
wt

A vector of class double of the final survey weights. Default is NULL assuming a self-weighted survey, as in the ENA for SMART software; otherwise a weighted analysis is computed.
edema

A vector of class character of edema. Code will be "y" for presence and "n" for absence of bilateral edema. Default is NULL.
.by

A vector of class character or numeric of the geographical areas or respective IDs for where the data was collected and for which the analysis should be summarised at.

Details

A concept of "combined flags" is introduced in this function. It consists of defining as flag any observation that is flagged in either flag_wfhz or flag_mfaz vectors. A new column cflags for combined flags is created and added to df. This ensures that all flagged observations from both WFHZ and MFAZ data are excluded from the prevalence analysis.

A glimpse on how cflags are defined:

flag_wfhz flag_mfaz cflags
1 0 1
0 1 1
0 0 0

Value

A summarised table of class data.frame for the descriptive statistics about combined wasting.

Examples

## When .by and wt are set to NULL ----
mw_estimate_prevalence_combined(
  df = anthro.02,
  wt = NULL,
  edema = edema,
  .by = NULL
)

## When wt is not set to NULL ----
mw_estimate_prevalence_combined(
  df = anthro.02,
  wt = wtfactor,
  edema = edema,
  .by = NULL
)

Estimate the prevalence of wasting based on z-scores of muac-for-age (MFAZ)

Description

Calculate the prevalence estimates of wasting based on z-scores of muac-for-age and/or bilateral edema. The function allows users to get the prevalence estimates calculated in accordance with the complex sample design properties; this includes applying survey weights when needed or applicable.

Before estimating, the function evaluates the quality of data by calculating and rating the standard deviation of z-scores of MFAZ. If rated as problematic, the prevalence is estimated based on the PROBIT method.

Outliers are detected based on SMART flags and get excluded prior prevalence analysis.

Usage

mw_estimate_prevalence_mfaz(df, wt = NULL, edema = NULL, .by = NULL)

Arguments

df

A data set object of class data.frame to use. This must have been wrangled using this package's wrangling function for MUAC data. The function uses a variable name called cluster where the primary sampling unit IDs are stored. Make sure to rename your cluster ID variable to cluster, otherwise the function will error and terminate the execution.
wt

A vector of class double of the final survey weights. Default is NULL assuming a self weighted survey, as in the ENA for SMART software; otherwise, when a vector of weights if supplied, weighted analysis is done.
edema

A vector of class character of edema. Code should be "y" for presence and "n" for absence of bilateral edema. Default is NULL.
.by

A vector of class character or numeric of the geographical areas or respective IDs for where the data was collected and for which the analysis should be summarized at.

Value

A summarized table of class data.frame of the descriptive statistics about wasting.

Examples

## When .by = NULL ----
mw_estimate_prevalence_mfaz(
  df = anthro.04,
  wt = NULL,
  edema = edema,
  .by = NULL
)

## When .by is not set to NULL ----
mw_estimate_prevalence_mfaz(
  df = anthro.04,
  wt = NULL,
  edema = edema,
  .by = province
)

Estimate the prevalence of wasting based on MUAC for survey data

Description

Calculate the prevalence estimates of wasting based on MUAC and/or bilateral edema. Before estimating, the function evaluates the quality of data by calculating and rating the standard deviation of z-scores of muac-for-age (MFAZ) and the p-value of the age ratio test; then it sets the analysis path that best fits the data:

  • If all tests are rated as not problematic, a normal analysis is done.

  • If standard deviation is not problematic and age ratio test is problematic, prevalence is age-weighted. This is to fix the likely overestimation of wasting when there are excess of younger children in the data set.

  • If standard deviation is problematic and age ratio test is not, or both are problematic, analysis gets cancelled out and NAs get thrown.

Outliers are detected based on SMART flags on the MFAZ values and then get excluded prior being piped into the actual prevalence analysis workflow.

Usage

mw_estimate_prevalence_muac(df, wt = NULL, edema = NULL, .by = NULL)

mw_estimate_smart_age_wt(df, edema = NULL, .by = NULL)

Arguments

df

A data set object of class data.frame to use. This must have been wrangled using this package's wrangling function for MUAC data. Make sure MUAC values are converted to millimeters after using the wrangler. If this is not done, the function will stop execution and return an error message. The function uses a variable name called cluster where the primary sampling unit IDs are stored. Make sure the data set has this variable and its name has been renamed to cluster, otherwise the function will error and terminate the execution.
wt

A vector of class double of the final survey weights. Default is NULL assuming a self weighted survey, as in the ENA for SMART software; otherwise, when a vector of weights if supplied, weighted analysis is done.
edema

A vector of class character of edema. Code should be "y" for presence and "n" for absence of bilateral edema. Default is NULL.
.by

A vector of class character or numeric of the geographical areas or respective IDs for where the data was collected and for which the analysis should be summarized at.

Value

A summarized table of class data.frame of the descriptive statistics about wasting.

References

SMART Initiative (no date). Updated MUAC data collection tool. Available at: https://smartmethodology.org/survey-planning-tools/updated-muac-tool/

See Also

mw_estimate_smart_age_wt() mw_estimate_prevalence_mfaz() mw_estimate_prevalence_screening()

Examples

## When .by = NULL ----
mw_estimate_prevalence_muac(
  df = anthro.04,
  wt = NULL,
  edema = edema,
  .by = NULL
)

## When .by is not set to NULL ----
mw_estimate_prevalence_muac(
  df = anthro.04,
  wt = NULL,
  edema = edema,
  .by = province
)

## An application of `mw_estimate_smart_age_wt()` ----
.data <- anthro.04 |>
  subset(province == "Province 2")

mw_estimate_smart_age_wt(
  df = .data,
  edema = edema,
  .by = NULL
)

Estimate the prevalence of wasting based on MUAC for non survey data

Description

It is common to estimate prevalence of wasting from non survey data, such as screenings or any other community-based surveillance systems. In such situations, the analysis usually consists only in estimating the point prevalence and the counts of positive cases, without necessarily estimating the uncertainty. This is the job of this function.

Before estimating, it evaluates the quality of data by calculating and rating the standard deviation of z-scores of muac-for-age (MFAZ) and the p-value of the age ratio test; then it sets the analysis path that best fits the data.

  • If all tests are rated as not problematic, a normal analysis is done.

  • If standard deviation is not problematic and age ratio test is problematic, prevalence is age-weighted. This is to fix the likely overestimation of wasting when there are excess of younger children in the data set.

  • If standard deviation is problematic and age ratio test is not, or both are problematic, analysis gets cancelled out and NAs get thrown.

Outliers are detected based on SMART flags on the MFAZ values and then get excluded prior being piped into the actual prevalence analysis workflow.

Usage

mw_estimate_prevalence_screening(df, muac, edema = NULL, .by = NULL)

Arguments

df

A data set object of class data.frame to use. This must have been wrangled using this package's wrangling function for MUAC data. Make sure MUAC values are converted to millimeters after using the wrangler. If this is not done, the function will stop execution and return an error message with the issue.
muac

A vector of raw MUAC values of class numeric or integer. The measurement unit of the values should be millimeters. If any or all values are in a different unit than the expected, the function will stop execution and return an error message indicating the issue.
edema

A vector of class character of edema. Code should be "y" for presence and "n" for absence of bilateral edema. Default is NULL. If class, as well as, code values are different than expected, the function will stop the execution and return an error message indicating the issue.
.by

A vector of class character or numeric of the geographical areas or respective IDs for where the data was collected and for which the analysis should be summarized at.

Value

A summarized table of class data.frame of the descriptive statistics about wasting.

References

SMART Initiative (no date). Updated MUAC data collection tool. Available at: https://smartmethodology.org/survey-planning-tools/updated-muac-tool/

See Also

mw_estimate_prevalence_muac() mw_estimate_smart_age_wt()

Examples

mw_estimate_prevalence_screening(
  df = anthro.02,
  muac = muac,
  edema = edema,
  .by = province
)

## With `edema` set to `NULL` ----
mw_estimate_prevalence_screening(
  df = anthro.02,
  muac = muac,
  edema = NULL,
  .by = province
)

## With `.by` set to `NULL` ----
mw_estimate_prevalence_screening(
  df = anthro.02,
  muac = muac,
  edema = NULL,
  .by = NULL
)

Estimate the prevalence of wasting based on z-scores of weight-for-height (WFHZ)

Description

Calculate the prevalence estimates of wasting based on z-scores of weight-for-height and/or bilateral edema. The function allows users to get the prevalence estimates calculated in accordance with the complex sample design properties; this includes applying survey weights when needed or applicable.

Before estimating, the function evaluates the quality of data by calculating and rating the standard deviation of z-scores of WFHZ. If rated as problematic, the prevalence is estimated based on the PROBIT method.

Outliers are detected based on SMART flags and get excluded prior being piped into the actual prevalence analysis workflow.

Usage

mw_estimate_prevalence_wfhz(df, wt = NULL, edema = NULL, .by = NULL)

Arguments

df

A data set object of class data.frame to use. This must have been wrangled using this package's wrangling function for WFHZ data. The function uses a variable name called cluster where the primary sampling unit IDs are stored. Make sure to rename your cluster ID variable to cluster, otherwise the function will error and terminate the execution.
wt

A vector of class double of the final survey weights. Default is NULL assuming a self weighted survey, as in the ENA for SMART software; otherwise, when a vector of weights if supplied, weighted analysis is done.
edema

A vector of class character of edema. Code should be "y" for presence and "n" for absence of bilateral edema. Default is NULL.
.by

A vector of class character or numeric of the geographical areas or respective IDs for where the data was collected and for which the analysis should be summarised at.

Value

A summarised table of class data.frame of the descriptive statistics about wasting.

Examples

## When .by = NULL ----
### Start off by wrangling the data ----
data <- mw_wrangle_wfhz(
  df = anthro.03,
  sex = sex,
  weight = weight,
  height = height,
  .recode_sex = TRUE
)

### Now run the prevalence function ----
mw_estimate_prevalence_wfhz(
  df = data,
  wt = NULL,
  edema = edema,
  .by = NULL
)

## Now when .by is not set to NULL ----
mw_estimate_prevalence_wfhz(
  df = data,
  wt = NULL,
  edema = edema,
  .by = district
)

## When a weighted analysis is needed ----
mw_estimate_prevalence_wfhz(
  df = anthro.02,
  wt = wtfactor,
  edema = edema,
  .by = province
)

Clean and format the output table returned from the MFAZ plausibility check for improved clarity and readability

Description

Clean and format the output table returned from the MFAZ plausibility check for improved clarity and readability. It converts scientific notations to standard notations, round values and rename columns to meaningful names.

Usage

mw_neat_output_mfaz(df)

Arguments

df

An object of class data.frame returned by this package's plausibility checker for MFAZ data, containing the summarized results to be formatted.

Value

A data.frame object of the same length and width as df, with column names and values formatted for clarity and readability.

Examples

## First wrangle age data ----
data <- mw_wrangle_age(
  df = anthro.01,
  dos = dos,
  dob = dob,
  age = age,
  .decimals = 2
)

## Then wrangle MUAC data ----
data_mfaz <- mw_wrangle_muac(
  df = data,
  sex = sex,
  age = age,
  muac = muac,
  .recode_sex = TRUE,
  .recode_muac = TRUE,
  .to = "cm"
)

## Then run plausibility check ----
pl <- mw_plausibility_check_mfaz(
  df = data_mfaz,
  flags = flag_mfaz,
  sex = sex,
  muac = muac,
  age = age
)

## Now neat the output table ----
mw_neat_output_mfaz(df = pl)

Clean and format the output table returned from the MUAC plausibility check for improved clarity and readability.

Description

Clean and format the output table returned from the plausibility check of raw MUAC data for improved clarity and readability. It converts scientific notations to standard notations, round values and rename columns to meaningful names.

Usage

mw_neat_output_muac(df)

Arguments

df

An object of class data.frame returned by this package's plausibility checker for raw MUAC data, containing the summarized results to be formatted.

Value

A data.frame object of the same length and width as df, with column names and values formatted for clarity and readability.

Examples

## First wranlge MUAC data ----
df_muac <- mw_wrangle_muac(
  df = anthro.01,
  sex = sex,
  muac = muac,
  age = NULL,
  .recode_sex = TRUE,
  .recode_muac = FALSE,
  .to = "none"
)

## Then run the plausibility check ----
pl_muac <- mw_plausibility_check_muac(
  df = df_muac,
  flags = flag_muac,
  sex = sex,
  muac = muac
)

## Neat the output table ----

mw_neat_output_muac(df = pl_muac)

Clean and format the output table returned from the WFHZ plausibility check for improved clarity and readability

Description

Clean and format the output table returned from the WFHZ plausibility check for improved clarity and readability. It converts scientific notations to standard notations, round values and rename columns to meaningful names.

Usage

mw_neat_output_wfhz(df)

Arguments

df

An object of class data.frame returned by this package's plausibility checker for WFHZ data, containing the summarized results to be formatted.

Value

A data.frame object of the same length and width as df, with column names and values formatted for clarity and readability.

Examples

## First wrangle age data ----
data <- mw_wrangle_age(
  df = anthro.01,
  dos = dos,
  dob = dob,
  age = age,
  .decimals = 2
)

## Then wrangle WFHZ data ----
data_wfhz <- mw_wrangle_wfhz(
  df = data,
  sex = sex,
  weight = weight,
  height = height,
  .recode_sex = TRUE
)

## Now run the plausibility check ----
pl <- mw_plausibility_check_wfhz(
  df = data_wfhz,
  sex = sex,
  age = age,
  weight = weight,
  height = height,
  flags = flag_wfhz
)

## Now neat the output table ----
mw_neat_output_wfhz(df = pl)

Check the plausibility and acceptability of muac-for-age z-score (MFAZ) data

Description

Check the overall plausibility and acceptability of MFAZ data through a structured test suite encompassing sampling and measurement-related biases checks in the data set. The test suite in this function follows the recommendation made by Bilukha, O., & Kianian, B. (2023) on the plausibility of constructing a comprehensive plausibility check for MUAC data similar to WFHZ to evaluate its acceptability when the variable age exists in the data set.

The function works on a data frame returned from this package's wrangling function for age and for MFAZ data.

Usage

mw_plausibility_check_mfaz(df, sex, muac, age, flags)

Arguments

df

A data set object of class data.frame to check.
sex

A vector of class numeric of child's sex.
muac

A vector of class numeric of child's MUAC in centimeters.
age

A vector of class double of child's age in months.
flags

A vector of class numeric of flagged records.

Details

Whilst the function uses the same test checks and criteria as that of WFHZ in the SMART plausibility check, the percent of flagged data is evaluated using a different cut-off points, with a maximum acceptability of 2.0%, as shown below:

Excellent Good Acceptable Problematic
0.0 - 1.0 >1.0 - 1.5 >1.5 - 2.0 >2.0

Value

A summarized table of class data.frame, of length 17 and width 1, for the plausibility test results and their respective acceptability ratings.

References

Bilukha, O., & Kianian, B. (2023). Considerations for assessment of measurement quality of mid‐upper arm circumference data in anthropometric surveys and mass nutritional screenings conducted in humanitarian and refugee settings. Maternal & Child Nutrition, 19, e13478. https://doi.org/10.1111/mcn.13478

SMART Initiative (2017). Standardized Monitoring and Assessment for Relief and Transition. Manual 2.0. Available at: https://smartmethodology.org.

See Also

mw_wrangle_age() mw_wrangle_muac() mw_stattest_ageratio() flag_outliers()

Examples

## First wrangle age data ----
data <- mw_wrangle_age(
  df = anthro.01,
  dos = dos,
  dob = dob,
  age = age,
  .decimals = 2
)

## Then wrangle MUAC data ----
data_muac <- mw_wrangle_muac(
  df = data,
  sex = sex,
  age = age,
  muac = muac,
  .recode_sex = TRUE,
  .recode_muac = TRUE,
  .to = "cm"
)

## And finally run plausibility check ----
mw_plausibility_check_mfaz(
  df = data_muac,
  flags = flag_mfaz,
  sex = sex,
  muac = muac,
  age = age
)

Check the plausibility and acceptability of raw MUAC data

Description

Check the overall plausibility and acceptability of raw MUAC data through a structured test suite encompassing sampling and measurement-related biases checks in the data set. The test suite in this function follows the recommendation made by Bilukha, O., & Kianian, B. (2023).

Usage

mw_plausibility_check_muac(df, sex, muac, flags)

Arguments

df

An object of class data.frame to check. It must have been wrangled using this package's wrangling function for MUAC.
sex

A vector of class numeric of child's sex.
muac

A vector of class double of child's MUAC in centimeters.
flags

A vector of class numeric of flagged records.

Details

Cut-off points used for the percent of flagged records:

Excellent Good Acceptable Problematic
0.0 - 1.0 >1.0 - 1.5 >1.5 - 2.0 >2.0

Value

A summarized table of class data.frame, of length 9 and width 1, for the plausibility test results and their respective acceptability ratings.

References

Bilukha, O., & Kianian, B. (2023). Considerations for assessment of measurement quality of mid‐upper arm circumference data in anthropometric surveys and mass nutritional screenings conducted in humanitarian and refugee settings. Maternal & Child Nutrition, 19, e13478. https://doi.org/10.1111/mcn.13478

SMART Initiative (2017). Standardized Monitoring and Assessment for Relief and Transition. Manual 2.0. Available at: https://smartmethodology.org.

See Also

mw_wrangle_muac() flag_outliers()

Examples

## First wranlge MUAC data ----
df_muac <- mw_wrangle_muac(
  df = anthro.01,
  sex = sex,
  muac = muac,
  age = NULL,
  .recode_sex = TRUE,
  .recode_muac = FALSE,
  .to = "none"
)

## Then run the plausibility check ----
mw_plausibility_check_muac(
  df = df_muac,
  flags = flag_muac,
  sex = sex,
  muac = muac
)

Check the plausibility and acceptability of weight-for-height z-score (WFHZ) data

Description

Check the overall plausibility and acceptability of WFHZ data through a structured test suite encompassing sampling and measurement-related biases checks in the data set. The test suite, including the criteria and corresponding rating of acceptability, follows the standards in the SMART plausibility check. The only exception is the exclusion of MUAC checks. MUAC is checked separately using more comprehensive test suite as well.

The function works on a data frame returned from this package's wrangling function for age and for WFHZ data.

Usage

mw_plausibility_check_wfhz(df, sex, age, weight, height, flags)

Arguments

df

A data set object of class data.frame to check.
sex

A vector of class numeric of child's sex.
age

A vector of class double of child's age in months.
weight

A vector of class double of child's weight in kilograms.
height

A vector of class double of child's height in centimeters.
flags

A vector of class numeric of flagged records.

Value

A summarized table of class data.frame, of length 19 and width 1, for the plausibility test results and their respective acceptability rates.

References

SMART Initiative (2017). Standardized Monitoring and Assessment for Relief and Transition. Manual 2.0. Available at: https://smartmethodology.org.

See Also

mw_plausibility_check_mfaz() mw_plausibility_check_muac() mw_wrangle_age()

Examples

## First wrangle age data ----
data <- mw_wrangle_age(
  df = anthro.01,
  dos = dos,
  dob = dob,
  age = age,
  .decimals = 2
)

## Then wrangle WFHZ data ----
data_wfhz <- mw_wrangle_wfhz(
  df = data,
  sex = sex,
  weight = weight,
  height = height,
  .recode_sex = TRUE
)

## Now run the plausibility check ----
mw_plausibility_check_wfhz(
  df = data_wfhz,
  sex = sex,
  age = age,
  weight = weight,
  height = height,
  flags = flag_wfhz
)

Test for statistical difference between the proportion of children aged 24 to 59 months old over those aged 6 to 23 months old

Description

Calculate the observed age ratio of children aged 24 to 59 months old over those aged 6 to 23 months old and test if there is a statistical difference between the observed and the expected.

Usage

mw_stattest_ageratio(age, .expectedP = 0.66)

Arguments

age

A vector of class numeric of child's age in months. If different than expected, the function will stop execution and return an error message indicating the type of mismatch.
.expectedP

The expected proportion of children aged 24 to 59 months old over those aged 6 to 23 months old. This is estimated to be 0.66.

Details

This function should be used specifically when assessing the quality of MUAC data. For age ratio test of children aged 6 to 29 months old over 30 to 59 months old, as performed in the SMART plausibility check, use nipnTK::ageRatioTest() instead.

Value

A vector of class list of three statistics: p for p-value of the statistical difference between the observed and the expected proportion of children aged 24 to 59 months old over those aged 6 to 23 months old; observedR and observedP for the observed ratio and proportion respectively.

References

SMART Initiative. Updated MUAC data collection tool. Available at: https://smartmethodology.org/survey-planning-tools/updated-muac-tool/

Examples

mw_stattest_ageratio(
  age = anthro.02$age,
  .expectedP = 0.66
)

Wrangle child's age

Description

Wrangle child's age for downstream analysis. This includes calculating age in months based on the date of data collection and the child's date of birth, and setting to NA the age values that are less than 6.0 and greater than or equal to 60.0 months old.

Usage

mw_wrangle_age(df, dos = NULL, dob = NULL, age, .decimals = 2)

Arguments

df

A data set of class data.frame to wrangle age from.
dos

A vector of class Date for date of data collection from the df. Default is NULL.
dob

A vector of class Date for child's date of birth from the df. Default is NULL.
age

A vector of class numeric of child's age in months. In most cases this will be estimated using local event calendars; in some other cases it can be a mix of the former and the one based on the child's date of birth and the date of data collection.
.decimals

The number of decimals places to which the age should be rounded. Default is 2.

Value

A data.frame based on df. The variable age will be automatically filled in each row where age value was missing and both the child's date of birth and the date of data collection are available. Rows where age is less than 6.0 and greater than or equal to 60.0 months old will be set to NA. Additionally, a new variable for df named age_days, of class double, will be created.

Examples

## A sample data ----
df <- data.frame(
  surv_date = as.Date(c(
    "2023-01-01", "2023-01-01", "2023-01-01", "2023-01-01", "2023-01-01"
  )),
  birth_date = as.Date(c(
    "2019-01-01", NA, "2018-03-20", "2019-11-05", "2021-04-25"
  )),
  age = c(NA, 36, NA, NA, NA)
)

## Apply the function ----
mw_wrangle_age(
  df = df,
  dos = surv_date,
  dob = birth_date,
  age = age,
  .decimals = 3
)

Wrangle MUAC data

Description

Calculate z-scores for MUAC-for-age (MFAZ) and identify outliers based on the SMART methodology. When age is not supplied, wrangling will consist only in detecting outliers from the raw MUAC values. The function only works after the age has been wrangled.

Usage

mw_wrangle_muac(
  df,
  sex,
  muac,
  age = NULL,
  .recode_sex = TRUE,
  .recode_muac = TRUE,
  .to = c("cm", "mm", "none"),
  .decimals = 3
)

Arguments

df

A data set object of class data.frame to wrangle data from.
sex

A numeric or character vector of child's sex. Code values should only be 1 or "m" for males and 2 or "f" for females. Make sure sex values are coded in either of the aforementioned before calling the function. If input codes are different than expected, the function will stop execution and return an error message with the type of mismatch.
muac

A vector of class numeric of child's age in months. If the class is different than expected, the function will stop execution and return an error message indicating the type of mismatch.
age

A vector of class numeric of child's age in months.
.recode_sex

Logical. Set to TRUE if the values for sex are not coded as 1 (for males) or 2 (for females). Otherwise, set to FALSE (default).
.recode_muac

Logical. Set to TRUE if the values for raw MUAC should be converted to either centimeters or millimeters. Otherwise, set to FALSE (default)
.to

A choice of the measuring unit to which the MUAC values should be converted; "cm" for centimeters, "mm" for millimeters and "none" to leave as it is.
.decimals

The number of decimals places the z-scores should have. Default is 3.

Value

A data frame based on df. New variables named mfaz and flag_mfaz, of child's MFAZ and detected outliers, will be created. When age is not supplied, only flag_muac variable is created. This refers to outliers detected based on the raw MUAC values.

References

Bilukha, O., & Kianian, B. (2023). Considerations for assessment of measurement quality of mid‐upper arm circumference data in anthropometric surveys and mass nutritional screenings conducted in humanitarian and refugee settings. Maternal & Child Nutrition, 19, e13478. https://doi.org/10.1111/mcn.13478

SMART Initiative (2017). Standardized Monitoring and Assessment for Relief and Transition. Manual 2.0. Available at: https://smartmethodology.org.

See Also

flag_outliers() remove_flags() mw_wrangle_age()

Examples

## When age is available, wrangle it first before calling the function ----
w <- mw_wrangle_age(
  df = anthro.02,
  dos = NULL,
  dob = NULL,
  age = age,
  .decimals = 2
)

### Then apply the function to wrangle MUAC data ----
mw_wrangle_muac(
  df = w,
  sex = sex,
  age = age,
  muac = muac,
  .recode_sex = TRUE,
  .recode_muac = TRUE,
  .to = "cm",
  .decimals = 3
)

## When age is not available ----
mw_wrangle_muac(
  df = anthro.02,
  sex = sex,
  age = NULL,
  muac = muac,
  .recode_sex = TRUE,
  .recode_muac = TRUE,
  .to = "cm",
  .decimals = 3
)

Wrangle weight-for-height data

Description

Calculate z-scores for weight-for-height (WFHZ) and identify outliers based on the SMART methodology.

Usage

mw_wrangle_wfhz(df, sex, weight, height, .recode_sex = TRUE, .decimals = 3)

Arguments

df

A data set object of class data.frame to wrangle data from.
sex

A numeric or character vector of child's sex. Code values should only be 1 or "m" for males and 2 or "f" for females. Make sure sex values are coded in either of the aforementioned before to call the function. If input codes are neither of the above, the function will stop execution and return an error message with the type of mismatch.
weight

A vector of class double of child's weight in kilograms. If the input is of a different class, the function will stop execution and return an error message indicating the type of mismatch.
height

A vector of class double of child's height in centimeters. If the input is of a different class, the function will stop execution and return an error message indicating the type of mismatch.
.recode_sex

Logical. Set to TRUE if the values for sex are not coded as 1 (for males) or 2 (for females). Otherwise, set to FALSE (default).
.decimals

The number of decimals places the z-scores should have. Default is 3.

Value

A data frame based on df. New variables named wfhz and flag_wfhz, of child's WFHZ and detected outliers, will be created.

References

SMART Initiative (2017). Standardized Monitoring and Assessment for Relief and Transition. Manual 2.0. Available at: https://smartmethodology.org.

See Also

flag_outliers() remove_flags()

Examples

mw_wrangle_wfhz(
  df = anthro.01,
  sex = sex,
  weight = weight,
  height = height,
  .recode_sex = TRUE,
  .decimals = 2
)

Convert MUAC values to either centimeters or millimeters

Description

Convert MUAC values to either centimeters or millimeters as required. Before to covert, the function checks if the supplied MUAC values are in the opposite unit of the intended conversion. If not, execution stops and an error message is returned.

Usage

recode_muac(x, .to = c("cm", "mm"))

Arguments

x

A vector of raw MUAC values. The class can either be double or numeric or integer. If different than expected, the function will stop execution and return an error message indicating the type of mismatch.
.to

A choice between cm (centimeters) and mm (millimeters) for the measuring unit to convert MUAC values to. Before to execute the conversion, the function checks if values are in the opposite unit; in case not, the execution stops and an error message is returned. Strive to address the error and try again.

Value

A numeric vector of the same length as x, with values converted to the chosen measuring unit.

Examples

## Recode from millimeters to centimeters ----
muac_cm <- recode_muac(
  x = anthro.01$muac,
  .to = "cm"
)
head(muac_cm)

## Using the `muac_cm` object to recode it back to "mm" ----
muac_mm <- recode_muac(
  x = muac_cm,
  .to = "mm"
)
tail(muac_mm)

A sample SMART survey data with WFHZ standard deviation rated as problematic

Description

A sample SMART survey data with WFHZ standard deviation rated as problematic

Usage

wfhz.01

Format

A tibble with 303 rows and 6 columns.

Variable Description
cluster Primary sampling unit
sex Sex, "m" = boys, "f" = girls
age calculated age in months with two decimal places
edema Edema, "n" = no, "y" = yes
wfhz MUAC-for-age z-scores with 3 decimal places
flag_wfhz Flagged observations. 1=flagged, 0=not flagged

Source

Anonymous

Examples

wfhz.01