This package is designed for identifying disease cases from admin data for epidemiological studies. The implementation focused on code readability and re-usability. Three types of functions are included:
Interactive functions (e.g.,
identify_row(), exclude(),
fetch_var()) based on filter and joins from dplyr with
tweaks that fix SQL translation or add features that are not natively
supported by SQL. They also work for local data.frames, and some use the
‘data.table’ package
(vignette("datatable-intro", package = "data.table")) to
speed up processing time for large data. These functions are not as
flexible as dplyr::filter(), but they are general enough to
be useful even outside health research.
Call-building functions (e.g., build_def(),
execute_def()) that facilitate batch execution and re-use
of case definitions. In essence, build_def creates codes of
definitions (which is chain of the interactive functions, e.g.,
define_case()) that are not immediately ran.
execute_def runs built definitions with different input
data.
Miscellaneous functions such as computing age
(compute_duration()), collapsing records within a time
range into one episode (collapse_episode()), and more
(on-going effort). Most of these functions have built-in checks
signalling when things might go wrong, e.g., missing values in
calculated ages. These helpers are introduced in the data wrangling
vignette (vignette("wrangling")).
In health research and surveillance, identifying diseases or events from administrative databases is often the initial step. However, crafting case-finding algorithms is a complex task. Existing algorithms, often written in SAS by experienced analysts, can be complex and difficult to decipher for the growing number of analysts trained primarily in R.
These algorithms may also affect performance if they depend on Data Step in SAS, due to a lack of translation between Data Step and SQL. This can result in SAS downloading data from a remote database to a local machine, leading to poor performance when handling large, population-based databases.
The ‘healthdb’ R package was created to address these challenges. It
minimizes the need to download data and offers an easy-to-use interface
for working with healthcare databases. It also includes capabilities not
supported by ‘SQL’, such as matching strings by ‘stringr’ style regular
expressions, and can compute comorbidity scores
(compute_comorbidity()) directly on a database server. This
vignette will present an example of common use cases.
Simply run:
We will need the following packages for this demo.
Consider the case definition of substance use disorder (SUD) from British Columbia Centre for Disease Control’s Chronic Disease Dashboard,
One or more hospitalization with a substance use disorder diagnostic code, OR Two or more physician visits with a substance use disorder diagnostic code within one year.
We are going to implement this definition. First, let’s make a demo data sets for the two sources:
Physician claims with multiple columns of ICD-9 diagnostic codes
# make_test_dat() makes either a toy data.frame or database table in memory with known number of rows that satisfy the query we will show later
claim_db <- make_test_dat(vals_kept = c("303", "304", "305", "291", "292", glue("30{30:59}"), glue("29{10:29}"), noise_val = c("999", "111")), type = "database")
# this is a database table
# note that in-memory SQLite database stores dates as numbers
claim_db %>% head()
#> # Source: SQL [?? x 6]
#> # Database: sqlite 3.53.1 [:memory:]
#> uid clnt_id dates diagx diagx_1 diagx_2
#> <int> <int> <dbl> <chr> <chr> <chr>
#> 1 79 2 17120 999 <NA> 999
#> 2 67 2 17758 999 <NA> 999
#> 3 23 3 16721 3044 3038 <NA>
#> 4 7 4 16499 3055 3045 <NA>
#> 5 71 4 16567 999 999 <NA>
#> 6 98 4 16632 999 <NA> <NA>Hospitalization with ICD-10 codes
hosp_df <- make_test_dat(vals_kept = c(glue("F{10:19}"), glue("F{100:199}"), noise_val = "999"), type = "data.frame")
# this is a local data.frame/tibble
hosp_df %>% head()
#> uid clnt_id dates diagx diagx_1 diagx_2
#> 1 29 1 2019-10-11 F138 F181 999
#> 2 92 2 2015-07-20 999 999 <NA>
#> 3 71 2 2016-08-17 999 <NA> <NA>
#> 4 60 2 2017-05-14 999 999 999
#> 5 42 2 2017-12-12 F196 F17 F121
#> 6 78 3 2015-05-08 999 <NA> <NA>
# convert Date to numeric to be consistent with claim_db
hosp_df <- hosp_df %>%
mutate(dates = julian(dates))Let’s focus on the physician claims. Extracting clients with at least two records within a year is not difficult, and involves only a few steps. The codes could look like the following using dplyr, however, it does not work because: 1. SQL does not support multiple patterns in one LIKE operation, 2. dbply currently have issue with translating n_distinct.
## not run
claim_db %>%
# identify the target codes
filter(if_any(starts_with("diagx"), ~ str_like(., c("291%", "292%", "303%", "304%", "305%")))) %>%
# each clnt has at least 2 records on different dates
group_by(clnt_id) %>%
# the n_distinct step is mainly for reducing computation in the next step
filter(n_distinct(dates) >= 2) %>%
# any two dates within one year?
filter((max(dates) - min(dates)) <= 365)
## endHere’s how you could use healthdb to achieve these
steps:
Identify rows contains the target codes. Use
?identify_row to see a list of supported matching
types.
result1 <- claim_db %>%
identify_row(
vars = starts_with("diagx"),
match = "start",
vals = c(291:292, 303:305)
)
#> ℹ Identify records with condition(s):
#> • where at least one of the diagx, diagx_1, diagx_2 column(s) in each record
#> • contains a value satisfied SQL LIKE pattern: 291% OR 292% OR 303% OR 304% OR 305%
#> ℹ To see the final query generated by 'dbplyr', use dplyr::show_query() on the output.
#> To extract the SQL string, use dbplyr::remote_query().Bonus: remove clients with exclusion codes
This step is not in the substance use disorder definition, but other disease definitions often require exclusion of some ICDs that contradicts the ones of interest. Let’s say we want to remove clients with code “111” here.
We first identify “111” from the source, then exclude clients in the
output from the previous step’s result. exclude() take
either a data set (via the excl argument) or expression (condition
argument) as input. For the former, it performs an anti join matching on
the by argument (see dplyr::join_by()). For the latter, it
is the opposite of filter, i.e.,
filter(!(some_expression)).
result2 <- result1 %>%
exclude(
excl = identify_row(claim_db, starts_with("diagx"), "in", "111"),
by = "clnt_id"
)
#> ℹ Identify records with condition(s):
#> • where at least one of the diagx, diagx_1, diagx_2 column(s) in each record
#> • contains a value exactly matched values in set: "111"
#> ℹ Exclude records in `data` through anti_join with `excl` matching on (by argument): "clnt_id"Restrict the number of records per client
result3 <- result2 %>% restrict_n(
clnt_id = clnt_id,
n_per_clnt = 2,
count_by = dates,
# here we use filter mode to remove records that failed the restriction
mode = "filter"
)
#> ℹ Apply restriction that each client must have at least 2 records with distinct
#> dates. Clients/groups which did not met the condition were excluded.Restrict the temporal pattern of diagnoses
restrict_date() supports more complicated patterns like
having n diagnoses at least i days apart within j years (see
vignette("if_date_logic") for how this works under the
hood). Note that when SQL interpret order of dates, the result could be
not deterministic if there were duplicate dates within client.
Therefore, a unique row id (uid) has to be supplied to get consistent
result.
result4 <- result3 %>% restrict_date(
clnt_id = clnt_id,
date_var = dates,
n = 2,
within = 365,
uid = uid,
# here we use flag mode to flag records that met the restriction instead of removing those
mode = "flag"
)
#> ℹ Apply restriction that each client must have 2 records that were within 365
#> days. Records that met the condition were flagged.Fetch variables from other tables by matching common keys
Up to this point, the result is only a query and have not been downloaded. Hopefully, the data has been shrunken to a manageable size for collection.
# Class of result4
class(result4)
#> [1] "tbl_SQLiteConnection" "tbl_dbi" "tbl_sql"
#> [4] "tbl_lazy" "tbl"
# execute query and download the result
result_df <- result4 %>% collect()
# Number of rows in source
nrow(claim_db %>% collect())
#> [1] 100
# Number of rows in the current result
nrow(result_df)
#> [1] 27Our data now only contains diagnoses which are probably not enough
for further analyses. Let’s say we want to gather client demographics
such as age and sex from other sources. This certainly can be done with
multiple dplyr::left_join() calls. Here we provide the
fetch_var() function to make the codes more concise. Note
that the input must be a named object and not from a pipe (i.e., don’t
do this data %>% some_action %>% fetch_var()).
# make two look up tables
age_tab <- data.frame(
clnt_id = 1:50,
age = sample(1:90, 50),
sex = sample(c("F", "M"), 50, replace = TRUE)
)
address_tab <- data.frame(
clnt_id = rep(1:50, 5), year = rep(2016:2020, each = 50),
area_code = sample(0:200, 50, replace = TRUE)
)
# get year from dates for matching
result_df <- result_df %>% mutate(year = lubridate::year(as.Date(dates, origin = "1970-01-01")))
# note that keys must be present in all tables
fetch_var(result_df,
keys = c(clnt_id, year),
linkage = list(
# the formula means from_table ~ get_variable
# |clnt_id means matching on clnt_id only
age_tab ~ c(age, sex) | clnt_id,
address_tab ~ area_code
)
) %>%
select(uid, clnt_id, dates, age, sex, area_code) %>%
head()
#> # A tibble: 6 × 6
#> uid clnt_id dates age sex area_code
#> <int> <int> <dbl> <int> <chr> <int>
#> 1 8 8 16839 5 F 43
#> 2 19 8 17608 5 F 43
#> 3 33 9 17841 18 F 112
#> 4 6 9 18620 18 F 112
#> 5 10 12 17729 65 F 68
#> 6 14 12 18170 65 F 68To complete the definition, we need to repeat the process shown above
with hospitalization data. Some studies may use more than a handful of
data sources to define their sample. We packed steps 1-4 in one function
define_case(), and provide tools to perform batch execution
with different data and parameters to meet those needs.
If your case definition also needs an age restriction (e.g., only
count records made when the client was within a certain age range), use
define_case_with_age(). It takes all the same arguments as
define_case() plus birth_date/age
and age_range, applying the age filter before both the
count (n_per_clnt) and temporal
(apart/within) restrictions, so those are
evaluated only on age-eligible records (e.g., “two or more visits while
aged 18-65”). It can be supplied to build_def(def_fn = ...)
just like define_case().
# build the full definition of SUD
sud_def <- build_def(
# name of definition
def_lab = "SUD",
# place holder names for sources
src_labs = c("claim", "hosp"),
def_fn = define_case, # you could alter it and supply your own function
# below are argumets of define_case
fn_args = list(
# if length = 1, the single element will be use for every source
vars = list(starts_with("diagx")),
match = "start", # match ICD starts with vals
vals = list(c(291:292, 303:305), glue("F{10:19}")),
clnt_id = clnt_id,
n_per_clnt = c(2, 1),
date_var = dates,
within = c(365, NULL),
uid = uid,
mode = "flag"
)
)
sud_def
#> # A tibble: 2 × 5
#> def_lab src_labs def_fn fn_args fn_call
#> <chr> <chr> <chr> <list> <list>
#> 1 SUD claim define_case <named list [9]> <language>
#> 2 SUD hosp define_case <named list [9]> <language>Let’s look inside the fn_call list column. Two calls of
define_case() have been made with different parameters. The
data arguments are left empty on purpose for re-usability. For example,
you may want to repeat the analysis with data from different regions or
study periods.
sud_def$fn_call
#> [[1]]
#> define_case(data = , vars = starts_with("diagx"), match = "start",
#> vals = c(291:292, 303:305), clnt_id = clnt_id, n_per_clnt = 2,
#> date_var = dates, within = 365, uid = uid, mode = "flag")
#>
#> [[2]]
#> define_case(data = , vars = starts_with("diagx"), match = "start",
#> vals = glue("F{10:19}"), clnt_id = clnt_id, n_per_clnt = 1,
#> date_var = dates, within = NULL, uid = uid, mode = "flag")Executing the definition is simply a call of
execute_def(). If verbose option is not turned off by
options(healthdb.verbose = FALSE), the output message will
explain what has been done. You could append multiple
build_def() outputs together and execute them all at once.
Definition and source labels will be added to the result to identify
outputs from different calls.
# execute the definition
result_list <- sud_def %>%
execute_def(with_data = list(
claim = claim_db,
hosp = hosp_df
))
#>
#> Actions for definition SUD using source claim_db:
#> → --------------Inclusion step--------------
#> ℹ Identify records with condition(s):
#> • where at least one of the diagx, diagx_1, diagx_2 column(s) in each record
#> • contains a value satisfied SQL LIKE pattern: 291% OR 292% OR 303% OR 304% OR 305%
#> → --------------No. rows restriction--------------
#>
#> ℹ Apply restriction that each client must have at least 2 records with distinct dates. Records that met the condition were flagged.
#> → --------------Time span restriction--------------
#>
#> ℹ Apply restriction that each client must have 2 records that were within 365 days. Records that met the condition were flagged.
#> → -------------- Output all records--------------
#>
#>
#> Actions for definition SUD using source hosp_df:
#>
#> → --------------Inclusion step--------------
#>
#> ℹ Identify records with condition(s):
#> • where at least one of the diagx, diagx_1, diagx_2 column(s) in each record
#> • contains a value satisfied regular expression: ^F10|^F11|^F12|^F13|^F14|^F15|^F16|^F17|^F18|^F19
#>
#> All unique value(s) and frequency in the result (as the conditions require just one of the columns containing target values; irrelevant values may come from other vars columns):
#> 999 F100 F102 F103 F104 F105 F106 F107 F109 F11 F113 F114 F115 F117 F118 F121
#> 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
#> F122 F124 F127 F13 F130 F132 F133 F135 F136 F137 F138 F139 F14 F140 F144 F146
#> 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
#> F147 F148 F149 F15 F153 F154 F156 F157 F158 F159 F16 F160 F163 F165 F166 F167
#> 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
#> F169 F17 F171 F173 F174 F175 F176 F177 F18 F181 F183 F187 F188 F189 F190 F192
#> 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1
#> F194 F196 F197 NAs
#> 1 1 1 1
#> → -------------- Output all records--------------Let’s check the results!
# view the results
purrr::walk(result_list, ~ head(.) %>% print())
#> # Source: SQL [?? x 10]
#> # Database: sqlite 3.53.1 [:memory:]
#> def src uid clnt_id dates diagx diagx_1 diagx_2 flag_restrict_n
#> <chr> <chr> <int> <int> <dbl> <chr> <chr> <chr> <int>
#> 1 SUD claim 23 3 16721 3044 3038 <NA> 0
#> 2 SUD claim 7 4 16499 3055 3045 <NA> 0
#> 3 SUD claim 5 5 17926 3059 3038 <NA> 0
#> 4 SUD claim 34 6 16719 111 3035 <NA> 1
#> 5 SUD claim 36 6 16865 2911 2911 999 1
#> 6 SUD claim 47 6 17213 2923 303 3056 1
#> # ℹ 1 more variable: flag_restrict_date <int>
#> def src uid clnt_id dates diagx diagx_1 diagx_2
#> 1 SUD hosp 29 1 18180 F138 F181 999
#> 2 SUD hosp 42 2 17512 F196 F17 F121
#> 3 SUD hosp 9 3 16991 F194 F175 999
#> 4 SUD hosp 48 8 17908 F109 F177 F146
#> 5 SUD hosp 38 10 16701 F137 F137 999
#> 6 SUD hosp 27 11 16610 F122 F197 999At this point, the result from the claim database
(result[[1]]) has not been collected locally. You could
collect it manually, do further filtering, and then combine with the
result from hospitalization data in any way you want. If you just need a
simple row bind, we have bind_source() with convenient
naming feature.
bind_source(result_list,
# output_name = c(names in the list elements)
src = "src",
uid = "uid",
clnt_id = "clnt_id",
flag = c("flag_restrict_date", NA),
# force_proceed is needed to collect remote tables to local memory
force_proceed = TRUE
)
#> # A tibble: 100 × 5
#> src_No src uid clnt_id flag
#> <int> <chr> <int> <int> <int>
#> 1 1 claim 23 3 0
#> 2 1 claim 7 4 0
#> 3 1 claim 5 5 0
#> 4 1 claim 34 6 1
#> 5 1 claim 36 6 1
#> 6 1 claim 47 6 0
#> 7 1 claim 17 7 0
#> 8 1 claim 8 8 0
#> 9 1 claim 19 8 0
#> 10 1 claim 33 9 0
#> # ℹ 90 more rowspool_case() goes a few steps further than row bind. It
also filters records with valid flags and can summarize by client/group.
Since we had to decide which variables to be summarized in advance, the
output may not be flexible enough to meet your needs.
pool_case(result_list,
def = sud_def,
# your could skip summary with output_lvl = "raw"
output_lvl = "clnt",
# include records only from sources having valid records, see function documentation for more detail and other options
include_src = "has_valid",
force_proceed = TRUE
)
#> # A tibble: 35 × 10
#> def clnt_id first_valid_date first_valid_src last_entry_date last_entry_src
#> <chr> <int> <dbl> <chr> <dbl> <chr>
#> 1 SUD 1 18180 hosp 18180 hosp
#> 2 SUD 2 17512 hosp 17512 hosp
#> 3 SUD 3 16991 hosp 16991 hosp
#> 4 SUD 6 16719 claim 17213 claim
#> 5 SUD 8 17908 hosp 17908 hosp
#> 6 SUD 10 16701 hosp 16701 hosp
#> 7 SUD 11 16610 hosp 16610 hosp
#> 8 SUD 12 16765 claim 18501 hosp
#> 9 SUD 13 16585 hosp 17665 hosp
#> 10 SUD 15 18000 hosp 18000 hosp
#> # ℹ 25 more rows
#> # ℹ 4 more variables: raw_in_claim <dbl>, raw_in_hosp <dbl>,
#> # valid_in_claim <int>, valid_in_hosp <int>Identifying the cases is usually just the beginning. The data
wrangling vignette (vignette("wrangling")) introduces the
helpers for what comes next: reporting sample sizes for your flowchart,
computing age, decoding codes, scoring comorbidity, and reshaping
records into episodes and periods.