death_injury_matrix_count: Generate a death injury matrix from line-level death data

In PHSKC-APDE/rads: Assisted computation of King County public health data

Generate a death injury matrix from line-level death data

Description

Generate death counts for a death injury matrix specifying the intent and mechanism of injury. Needs line-level death data with a properly formatted ICD10 column.

Usage

death_injury_matrix_count(
  ph.data,
  intent = "*",
  mechanism = "*",
  icdcol = "underlying_cod_code",
  kingco = TRUE,
  group_by = NULL,
  ypll_age = NULL,
  death_age_col = NULL
)

Arguments

ph.data	a data.table or data.frame. Must contain death data structured with one person per row and with at least one column of ICD10 death codes.
intent	a character vector of length 1 to 5. It specifies the intent of death that you want returned ("Unintentional", "Suicide", "Homicide", "Undetermined", or "Legal intervention/war"). "none" will ignore the intent and only return the mechanism of death. NOTE You do not have to type the entire keyword for the intent, a partial string match is sufficient and is case insensitive. E.g., intent = c("cide") would return both "Suicide" and "Homicide" and intent = c("un") would return both "Unintentional" and "Undetermined". The default is '*', which selects all possible intents.
mechanism	a character vector of length 1 to 28. It specifies the mechanism of death that you want returned (E.g., "Cut/pierce", "Drowning", "Fall", "Firearm", etc.). "none" will ignore the mechanism and only return the intent of death. To see the complete list of mechanisms, type unique(rads.data::icd10_death_injury_matrix$mechanism) in your R console. NOTE You do not have to type the entire keyword for the mechanism, a partial string match is sufficient and is case insensitive. E.g., mechanism = c("cycl") would return both "Pedal cyclist" and "Motorcyclist". The default is '*', which selects all possible mechanisms
icdcol	a character vector of length one that specifies the name of the column in ph.data that contains the ICD10 death codes of interest. The default is underlying_cod_code, which is found in the properly formatted death data obtained using the get_data_death() function.
kingco	a logical vector of length one. It specifies whether you want to limit the analysis to King County. Note that this only works with data imported from the get_data_death() function. The default is kingco = TRUE.
group_by	a character vector of indeterminate length. This is used to specify all the variables by which you want to group (a.k.a. stratify) the results. For example, if you specified group_by = c('chi_sex', 'chi_race_6'), the results would be stratified by each combination of sex and race. The default is group_by = NULL
ypll_age	an optional numeric vector of length 1. When specified, it should be the age (an integer) used for Years of Potential Life Lost (YPLL) calculations. Valid values are between 1 & 99 (inclusive), though 65 and 85 are the most common. For example, ypll_age = 65 would sum the total number of years that could have been lived had everyone in the data lived to at least 65. Note that this function returns the total number of YPLL. Additional processing is necessary to calculate rates per 100,000. The default is ypll_age = NULL, which will skip YPLL calculations.
death_age_col	an optional character vector of length one that specifies the name of the column in ph.data with the decedents' age at death in years. It is only needed if ypll_age is specified AND if ph.data lacks a column named chi_age. The default is death_age_col = NULL.

Details

The matrix coding is based on the ICE (International Collaborative Effort on Injury Statistics) standard.

Value

The function returns a data.table with a minimum of three columns: mechanism, intent, & deaths. ypll_## and group_by columns will also be returned if specified in the arguments.

Note

The function default is to return the matrix of all intents and mechanisms of death. You can choose to only return the intent or only return the mechanism. If you set both to "none", you will receive a summary of all injury deaths without regard to the intent.

Also note that terrorism codes (U01.#, U02.#, & U03.#) are not included because they are not included in the coding used by WA DOH. If they are needed, they can be obtained from the CDC link below.

Source

rads.data::icd10_death_injury_matrix

References

WA DOH CHAT: https://secureaccess.wa.gov/doh/chat/Content/FilesForDownload/CodeSetDefinitions/CHATInjury(ICE)codes.pdf

CDC: https://www.cdc.gov/nchs/data/ice/icd10_transcode.pdf

Examples

# create synthetic line level data
set.seed(98104)
injurydata <- data.table::data.table(
  cod.icd10 = c(
    # Cut/pierce, Homicide
    rep("X99", round(runif(1, 30, 10000), 0)),
    # Drowning, Unintentional
    rep("W65", round(runif(1, 30, 10000), 0)),
    # Fall, Suicide
    rep("X80", round(runif(1, 30, 10000), 0)),
    # Fire/flame, Undetermined
    rep("Y26", round(runif(1, 30, 10000), 0)),
    # Firearm, Legal intervention/war
    rep("Y350", round(runif(1, 30, 10000), 0)),
    # Poisoning, Unintentional
    rep("X40", round(runif(1, 30, 10000), 0)),
    # Overexertion, Unintentional
    rep("X50", round(runif(1, 30, 10000), 0)),
    # Other land transport, Homicide
    rep("Y03", round(runif(1, 30, 10000), 0)),
    # Pedal cyclist, other, Unintentional
    rep("V10", round(runif(1, 30, 10000), 0)))
)

injurydata[, year := sample(2015:2020, nrow(injurydata), replace = TRUE)]

# example 1: every available combination of mechanism and intent
eg1 <- death_injury_matrix_count(ph.data = injurydata,
                            intent = "*",
                            mechanism = "*",
                            icdcol = "cod.icd10",
                            kingco = FALSE,
                            ypll_age = NULL,
                            death_age_col = NULL)
head(eg1) # note the data are stratified by year because year was in ph.data

# example 2: falls designated as homicides and or suicides
eg2 <- death_injury_matrix_count(ph.data = injurydata,
                            intent = "icide",
                            mechanism = "fall",
                            icdcol = "cod.icd10",
                            kingco = FALSE,
                            ypll_age = NULL,
                            death_age_col = NULL)
head(eg2)

# example 3: summary of all injury deaths regardless of intent and mechanism
eg3 <- death_injury_matrix_count(ph.data = injurydata,
                            intent = "none",
                            mechanism = "none",
                            icdcol = "cod.icd10",
                            kingco = FALSE,
                            ypll_age = NULL,
                            death_age_col = NULL)
eg3[]

# example 4: any intent and mechanism with YPLL_65 given death_age_col
injurydata4 <- data.table::copy(injurydata)
set.seed(98104)
injurydata4[, ageofdeath := rads::round2(rnorm(1, mean = 70, sd = 5 ), 0),
1:nrow(injurydata4)] # synthetic age of death
eg4 <- death_injury_matrix_count(ph.data = injurydata4,
                            intent = "none",
                            mechanism = "none",
                            icdcol = "cod.icd10",
                            kingco = FALSE,
                            ypll_age = 65,
                            death_age_col = "ageofdeath")
eg4[]

# example 5: all suicides, regardless of mechanism, stratified by age

injurydata5 <- data.table::copy(injurydata4)

eg5 <- death_injury_matrix_count(ph.data = injurydata5,
                            intent = "suicide",
                            mechanism = "none",
                            icdcol = "cod.icd10",
                            kingco = FALSE,
                            group_by = 'ageofdeath',
                            ypll_age = NULL,
                            death_age_col = NULL)
eg5[]

PHSKC-APDE/rads documentation built on April 14, 2025, 10:47 a.m.