| Title: | Blood Pressure Analysis in R |
|---|---|
| Description: | A comprehensive package to aid in the analysis of blood pressure data of all forms by providing both descriptive and visualization tools for researchers. |
| Authors: | John Schwenck [aut, cre],
Irina Gaynanova [aut] |
| Maintainer: | John Schwenck <[email protected]> |
| License: | GPL-3 |
| Version: | 2.1.0 |
| Built: | 2024-11-21 03:50:07 UTC |
| Source: | https://github.com/johnschwenck/bp |
THIS IS A DEPRECATED FUNCTION. USE bp_arv INSTEAD.
arv( data, inc_date = FALSE, subj = NULL, bp_type = 0, add_groups = NULL, inc_wake = TRUE )arv( data, inc_date = FALSE, subj = NULL, bp_type = 0, add_groups = NULL, inc_wake = TRUE )
data |
Required argument. Pre-processed dataframe containing SBP and
DBP with optional ID, VISIT, WAKE, and DATE columns if available.
Use |
inc_date |
Optional argument. Default is FALSE. As ABPM data typically
overlaps due to falling asleep on one date and waking up on another, the |
subj |
Optional argument. Allows the user to specify and subset specific subjects
from the |
bp_type |
Optional argument. Determines whether to calculate ARV for SBP values or DBP values. Default is 0 corresponding to output for both SBP & DBP. For both SBP and DBP ARV values use bp_type = 0, for SBP-only use bp_type = 1, and for DBP-only use bp_type = 2 |
add_groups |
Optional argument. Allows the user to aggregate the data by an
additional "group" to further refine the output. The supplied input must be a
character vector with the strings corresponding to existing column names of the
processed |
inc_wake |
Optional argument corresponding to whether or not to include |
Calculate the Average Real Variability (ARV) at various levels of granularity based on what is supplied (ID, VISIT, WAKE, and / or DATE). ARV is a measure of dispersion that takes into account the temporal structure of the data and relies on the sum of absolute differences in successive observations, unlike the successive variation (SV) which relies on the sum of squared differences.
A tibble object with a row corresponding to each subject, or alternatively a row corresponding to each date, if inc_date = TRUE. The resulting tibble consists of:
ID: The unique identifier of the subject. For single-subject datasets, ID = 1
VISIT: (If applicable) Corresponds to the visit # of the subject, if more than 1
WAKE: (If applicable) Corresponds to the awake status of the subject (0 = asleep |
1 = awake)
ARV_SBP / ARV_DBP: Calculates the average of the absolute differences between
successive measurements. The resulting value averages across the granularity grouping
for however many observations are present.
N: The number of observations for that particular grouping. If inc_date = TRUE,
N corresponds to the number of observations for that date. If inc_date = FALSE, N
corresponds to the number of observations for the most granular grouping available (i.e.
a combination of ID, VISIT, and WAKE)
Any add_groups variables supplied to function argument will be present as a column in the resulting tibble.
Mena et al. (2005) A reliable index for the prognostic significance of blood pressure variability Journal of Hypertension 23(5).505-11, doi:10.1097/01.hjh.0000160205.81652.5a.
# Load data data(bp_hypnos) data(bp_jhs) # Process bp_hypnos hypnos_proc <- process_data(bp_hypnos, sbp = "SYST", dbp = "DIAST", date_time = "date.time", id = "id", wake = "wake", visit = "visit", hr = "hr", pp ="pp", map = "map", rpp = "rpp") # Process bp_jhs data jhs_proc <- process_data(bp_jhs, sbp = "Sys.mmHg.", dbp = "Dias.mmHg.", date_time = "DateTime", hr = "Pulse.bpm.") ## Not run: # ARV Calculation bp_arv(hypnos_proc, add_groups = c("SBP_Category")) bp_arv(jhs_proc, inc_date = TRUE) ## End(Not run)# Load data data(bp_hypnos) data(bp_jhs) # Process bp_hypnos hypnos_proc <- process_data(bp_hypnos, sbp = "SYST", dbp = "DIAST", date_time = "date.time", id = "id", wake = "wake", visit = "visit", hr = "hr", pp ="pp", map = "map", rpp = "rpp") # Process bp_jhs data jhs_proc <- process_data(bp_jhs, sbp = "Sys.mmHg.", dbp = "Dias.mmHg.", date_time = "DateTime", hr = "Pulse.bpm.") ## Not run: # ARV Calculation bp_arv(hypnos_proc, add_groups = c("SBP_Category")) bp_arv(jhs_proc, inc_date = TRUE) ## End(Not run)
Calculate the Average Real Variability (ARV) at various levels of granularity based on what is supplied (ID, VISIT, WAKE, and / or DATE). ARV is a measure of dispersion that takes into account the temporal structure of the data and relies on the sum of absolute differences in successive observations, unlike the successive variation (SV) which relies on the sum of squared differences.
bp_arv( data, inc_date = FALSE, subj = NULL, bp_type = c("both", "sbp", "dbp"), add_groups = NULL, inc_wake = TRUE )bp_arv( data, inc_date = FALSE, subj = NULL, bp_type = c("both", "sbp", "dbp"), add_groups = NULL, inc_wake = TRUE )
data |
Required argument. Pre-processed dataframe containing SBP and
DBP with optional ID, VISIT, WAKE, and DATE columns if available.
Use |
inc_date |
Optional argument. Default is FALSE. As ABPM data typically
overlaps due to falling asleep on one date and waking up on another, the |
subj |
Optional argument. Allows the user to specify and subset specific subjects
from the |
bp_type |
Optional argument. Determines whether to calculate ARV for SBP values or DBP values, or both. For both SBP and DBP ARV values use bp_type = 'both', for SBP-only use bp_type = 'sbp, and for DBP-only use bp_type = 'dbp'. If no type specified, default will be set to 'both' |
add_groups |
Optional argument. Allows the user to aggregate the data by an
additional "group" to further refine the output. The supplied input must be a
character vector with the strings corresponding to existing column names of the
processed |
inc_wake |
Optional argument corresponding to whether or not to include |
A tibble object with a row corresponding to each subject, or alternatively a row corresponding to each date, if inc_date = TRUE. The resulting tibble consists of:
ID: The unique identifier of the subject. For single-subject datasets, ID = 1
VISIT: (If applicable) Corresponds to the visit # of the subject, if more than 1
WAKE: (If applicable) Corresponds to the awake status of the subject (0 = asleep |
1 = awake)
ARV_SBP / ARV_DBP: Calculates the average of the absolute differences between
successive measurements. The resulting value averages across the granularity grouping
for however many observations are present.
N: The number of observations for that particular grouping. If inc_date = TRUE,
N corresponds to the number of observations for that date. If inc_date = FALSE, N
corresponds to the number of observations for the most granular grouping available (i.e.
a combination of ID, VISIT, and WAKE)
Any add_groups variables supplied to function argument will be present as a column in the resulting tibble.
Mena et al. (2005) A reliable index for the prognostic significance of blood pressure variability Journal of Hypertension 23(5).505-11, doi:10.1097/01.hjh.0000160205.81652.5a.
# Load data data(bp_hypnos) data(bp_jhs) # Process bp_hypnos hyp_proc <- process_data(bp_hypnos, sbp = "SYST", dbp = "DIAST", date_time = "date.time", id = "id", wake = "wake", visit = "visit", hr = "hr", pp ="pp", map = "map", rpp = "rpp") # Process bp_jhs data jhs_proc <- process_data(bp_jhs, sbp = "Sys.mmHg.", dbp = "Dias.mmHg.", date_time = "DateTime", hr = "Pulse.bpm.") # ARV Calculation bp_arv(hyp_proc, add_groups = c("SBP_Category"), bp_type = 'sbp') bp_arv(jhs_proc, inc_date = TRUE)# Load data data(bp_hypnos) data(bp_jhs) # Process bp_hypnos hyp_proc <- process_data(bp_hypnos, sbp = "SYST", dbp = "DIAST", date_time = "date.time", id = "id", wake = "wake", visit = "visit", hr = "hr", pp ="pp", map = "map", rpp = "rpp") # Process bp_jhs data jhs_proc <- process_data(bp_jhs, sbp = "Sys.mmHg.", dbp = "Dias.mmHg.", date_time = "DateTime", hr = "Pulse.bpm.") # ARV Calculation bp_arv(hyp_proc, add_groups = c("SBP_Category"), bp_type = 'sbp') bp_arv(jhs_proc, inc_date = TRUE)
Calculate the mean and median at various levels of granularity based on what is supplied (ID, VISIT, WAKE, and / or DATE) for either SBP, DBP, or both.
bp_center( data, inc_date = FALSE, subj = NULL, bp_type = c("both", "sbp", "dbp"), add_groups = NULL, inc_wake = TRUE )bp_center( data, inc_date = FALSE, subj = NULL, bp_type = c("both", "sbp", "dbp"), add_groups = NULL, inc_wake = TRUE )
data |
Required argument. Pre-processed dataframe with SBP and DBP columns
with optional ID, VISIT, WAKE, and DATE columns if available.
Use |
inc_date |
Optional argument. Default is FALSE. As ABPM data typically
overlaps due to falling asleep on one date and waking up on another, the |
subj |
Optional argument. Allows the user to specify and subset specific subjects
from the |
bp_type |
Optional argument. Determines whether to calculate center for SBP values or DBP values, or both. For both SBP and DBP ARV values use bp_type = 'both', for SBP-only use bp_type = 'sbp, and for DBP-only use bp_type = 'dbp'. If no type specified, default will be set to 'both' |
add_groups |
Optional argument. Allows the user to aggregate the data by an
additional "group" to further refine the output. The supplied input must be a
character vector with the strings corresponding to existing column names of the
processed |
inc_wake |
Optional argument corresponding to whether or not to include |
A tibble object with a row corresponding to each subject, or alternatively a row corresponding to each date if inc_date = TRUE. The resulting tibble consists of:
ID: The unique identifier of the subject. For single-subject datasets, ID = 1
VISIT: (If applicable) Corresponds to the visit # of the subject, if more than 1
WAKE: (If applicable) Corresponds to the awake status of the subject (0 = asleep |
1 = awake)
SBP_mean / DBP_mean: Calculates the mean of systolic blood pressure readings
for the specified time granularity.
SBP_med / DBP_med: Calculates the median of systolic blood pressure readings
for the specified time granularity.
N: The number of observations for that particular grouping. If inc_date = TRUE,
N corresponds to the number of observations for that date. If inc_date = FALSE, N
corresponds to the number of observations for the most granular grouping available (i.e.
a combination of ID, VISIT, and WAKE)
Any add_groups variables supplied to function argument will be present as a column in the resulting tibble.
# Load data data(bp_hypnos) data(bp_jhs) # Process bp_hypnos hyp_proc <- process_data(bp_hypnos, sbp = "SYST", dbp = "DIAST", date_time = "date.time", id = "id", wake = "wake", visit = "visit", hr = "hr", pp ="pp", map = "map", rpp = "rpp") # Process bp_jhs data jhs_proc <- process_data(bp_jhs, sbp = "Sys.mmHg.", dbp = "Dias.mmHg.", date_time = "DateTime", hr = "Pulse.bpm.") # BP Center Calculation bp_center(hyp_proc, subj = c(70417, 70435))# Load data data(bp_hypnos) data(bp_jhs) # Process bp_hypnos hyp_proc <- process_data(bp_hypnos, sbp = "SYST", dbp = "DIAST", date_time = "date.time", id = "id", wake = "wake", visit = "visit", hr = "hr", pp ="pp", map = "map", rpp = "rpp") # Process bp_jhs data jhs_proc <- process_data(bp_jhs, sbp = "Sys.mmHg.", dbp = "Dias.mmHg.", date_time = "DateTime", hr = "Pulse.bpm.") # BP Center Calculation bp_center(hyp_proc, subj = c(70417, 70435))
The bp_children data set contains information on 1,283 children from Bristol, UK,
who each took three blood pressure readings per visit over the course of two observation
periods (at ages 9 and 11) and had their body mass index (BMI), physical activity, and
sedentary time information recorded. The study examined how sedentary behavior and physical
activity affected children progressing through primary school to better understand the
relationship between elevated blood pressure in children and its impact on the development
of cardiovascular disease into adulthood
bp_childrenbp_children
A data frame containing 15 variables pertaining to blood pressure and physical activity as follows:
Child ID
Index of the blood pressure reading
Systolic Blood Pressure (mmHg)
Diastolic Blood Pressure (mmHg)
1 = Male
2 = Female
Household Highest Education:
1 = Up to GCSE/O level or equiv
2 = A level/NVQ or equiv
3 = Degree/HND or equiv
4 = Higher degree (MSc/PhD) or equiv
Visit #
Age in Years
Height (ft)
Weight (lbs)
Body Mass Index
Number of days of child accelerometer data overall
Average minutes per day over all valid days
Average sedentary minutes per day over all valid days
Average minutes of Moderate to Vigorous Physical Activity (MVPA) per day over all valid days
(Solomon-Moore E, Salway R, Emm-Collison L, Thompson JL, Sebire SJ, Lawlor DA, Jago R (PI). 2020).
Licensed under a CC-BY Creative Commons Attribution 4.0 International
Original Paper: doi:10.1371/journal.pone.0232333
Data: doi:10.5281/zenodo.1049587
Principal Investigator (PI): Russ Jago (University of Bristol)
Calculate the coefficient of variation at various levels of granularity based on what is supplied (ID, VISIT, WAKE, and / or DATE) for either SBP, DBP, or both. CV is a measure of dispersion
bp_cv( data, inc_date = FALSE, subj = NULL, bp_type = c("both", "sbp", "dbp"), add_groups = NULL, inc_wake = TRUE )bp_cv( data, inc_date = FALSE, subj = NULL, bp_type = c("both", "sbp", "dbp"), add_groups = NULL, inc_wake = TRUE )
data |
Required argument. Pre-processed dataframe with SBP and DBP columns
with optional ID, VISIT, WAKE, and DATE columns if available.
Use |
inc_date |
Optional argument. Default is FALSE. As ABPM data typically
overlaps due to falling asleep on one date and waking up on another, the |
subj |
Optional argument. Allows the user to specify and subset specific subjects
from the |
bp_type |
Optional argument. Determines whether to calculate CV for SBP values or DBP values, or both. For both SBP and DBP ARV values use bp_type = 'both', for SBP-only use bp_type = 'sbp, and for DBP-only use bp_type = 'dbp'. If no type specified, default will be set to 'both' |
add_groups |
Optional argument. Allows the user to aggregate the data by an
additional "group" to further refine the output. The supplied input must be a
character vector with the strings corresponding to existing column names of the
processed |
inc_wake |
Optional argument corresponding to whether or not to include |
A tibble object with a row corresponding to each subject, or alternatively a row corresponding to each date if inc_date = TRUE. The resulting tibble consists of:
ID: The unique identifier of the subject. For single-subject datasets, ID = 1
VISIT: (If applicable) Corresponds to the visit # of the subject, if more than 1
WAKE: (If applicable) Corresponds to the awake status of the subject (0 = asleep |
1 = awake)
CV_SBP / CV_DBP: Calculates the ratio of standard deviation to the mean. CV_SBP
or CV_DBP is useful for comparing the degree of variation from one data series
to another.
SD_SBP / SD_DBP: For completeness, the cv function also includes the
standard deviation as a comparison metric to measure spread around the average.
N: The number of observations for that particular grouping. If inc_date = TRUE,
N corresponds to the number of observations for that date. If inc_date = FALSE, N
corresponds to the number of observations for the most granular grouping available (i.e.
a combination of ID, VISIT, and WAKE)
Any add_groups variables supplied to function argument will be present as a column in the resulting tibble.
# Load data data(bp_hypnos) data(bp_jhs) # Process bp_hypnos hyp_proc <- process_data(bp_hypnos, sbp = "SYST", dbp = "DIAST", date_time = "date.time", id = "id", wake = "wake", visit = "visit", hr = "hr", pp ="pp", map = "map", rpp = "rpp") # Process bp_jhs data jhs_proc <- process_data(bp_jhs, sbp = "Sys.mmHg.", dbp = "Dias.mmHg.", date_time = "DateTime", hr = "Pulse.bpm.") # CV Calculation bp_cv(hyp_proc, inc_date = TRUE, add_groups = "SBP_Category", bp_type = 'sbp') bp_cv(jhs_proc, add_groups = c("meal_time")) # Notice that meal_time is not a column from process_data, but it still works# Load data data(bp_hypnos) data(bp_jhs) # Process bp_hypnos hyp_proc <- process_data(bp_hypnos, sbp = "SYST", dbp = "DIAST", date_time = "date.time", id = "id", wake = "wake", visit = "visit", hr = "hr", pp ="pp", map = "map", rpp = "rpp") # Process bp_jhs data jhs_proc <- process_data(bp_jhs, sbp = "Sys.mmHg.", dbp = "Dias.mmHg.", date_time = "DateTime", hr = "Pulse.bpm.") # CV Calculation bp_cv(hyp_proc, inc_date = TRUE, add_groups = "SBP_Category", bp_type = 'sbp') bp_cv(jhs_proc, add_groups = c("meal_time")) # Notice that meal_time is not a column from process_data, but it still works
The bp_ghana data set includes 757 subjects across 32 community health centers who
were partitioned into intervention groups according to a pragmatic cluster-randomized trial
with 389 in the health insurance coverage (HIC) group and 368 in another group consisting
of a combination of HIC with a nurse-led task-shifting strategy for hypertension control
(TASSH) (this group is denoted TASSH + HIC). This study was an effort to assess the
comparative effectiveness of HIC alone versus the combination of TASSH + HIC on reducing
systolic blood pressure among patients with uncontrolled hypertension in Ghana. Baseline
blood pressure measurements and 12 month follow-up results were collected among subjects,
85% of whom had 12 month data present
bp_ghanabp_ghana
A data frame containing 21 variables pertaining to blood pressure as follows:
Subject ID
How much time elapsed between readings: Baseline measurement, 6 month follow-up, 12 month follow-up
Systolic Blood Pressure
Diastolic Blood Pressure
Age in Years
0 = Male
1 = Female
Highest Degree Earned:
1 = No Schooling
2 = Primary Schooling (Grades 1 to 6)
3 = Junior Secondary Schooling (JSS) (Grades 7-8)
4 = Secondary School (Grades 9-11)
5 = Completed Secondary School
6 = Technical school certificate
7 = Some college but no degree
8 = Associate degree
9 = Graduate or Professional school (MD, JD, etc.)
999 = Not specified
0 = Unemployed
1 = Employed
0 = Illiterate
1 = Literate
1 = Smoker
2 = Occasional
3 = Ex-Smoker
4 = Non-Smoker
Income level expressed in home currency - Ghanaian cedi (GhC)
Site Location in Ghana:
1 = Suntreso
2 = Ananekrom
3 = Manhiya
4 = Apatrapa
5 = Nkawie
6 = Afrancho
7 = Kumsai South
8 = Mampongteng
9 = Mampong
10 = Bomfa
11 = Ejura
12 = Kofiase
13 = Kokofu
14 = Asuofia
15 = Konongo
16 = Subirisu
17 = Effiduase
18 = Abuakwa
19 = Tafo
20 = KMA
21 = Methodist
22 = Kenyasi
23 = Asonamaso
24 = Juansa
25 = Juaso
26 = Sekyredumase
27 = Nkenkaasu
28 = Berekese
29 = Bekwai
30 = Dwease
31 = Kuntanese
32 = Foase
Cardiovascular Risk Assessment (%) at Baseline:
1) <10%
2) 10-20%
3) 20-30%
4) 30 to 40%
999= Missing Data
Body Mass Index
Classification of BMI:
1 = Underweight (<18.5 kg/m^2)
2 = Normal (18.5 to 24.9 kg/m^2)
3 = Overweight (25.0 to 29.9 kg/m^2)
4 = Obese (>30 kg/m^2)
All activity in weighted MET minutes (per week) at Baseline
0 = Control
1 = Treatment
0 = Rural
1 = Urban
# of doctors on staff
# of nurses on staff
# of patients seen annually
(Ogedegbe G, Plange-Rhule J, Gyamfi J, Chaplin W, Ntim M, Apusiga K, Iwelunmor J, Awudzi KY, Quakyi KN, Mogavero JN, Khurshid K, Tayo B, Cooper R. 2019).
Licensed under a CC-0 1.0 (Creative Commons) Universal Public Domain Dedication License
Original Paper: doi:10.1371/journal.pmed.1002561
Data: doi:10.5061/dryad.16c9m51
The bp_hist function serves to display the frequencies of the SBP
and DBP readings. These histograms are formatted to complement the bp_scatter
function.
bp_hist(data, subj = NULL, bins = 30, na.rm = TRUE)bp_hist(data, subj = NULL, bins = 30, na.rm = TRUE)
data |
A processed dataframe resulting from the |
subj |
Optional argument. Allows the user to specify and subset specific subjects
from the |
bins |
An integer specifying how many bins to use for histogram plots. This is a ggplot parameter; default value set to 30 |
na.rm |
An logical value specifying whether or not to remove empty values from data frame. This is a ggplot parameter; default value set to TRUE. |
A list containing four objects: three histogram visual graphics corresponding to
the SBP / DBP totals, SBP frequency, and DBP frequency, and a fourth list element
corresponding to a plot legend object for use in the bp_report function
data("bp_jhs") data("bp_hypnos") hyp_proc <- process_data(bp_hypnos, bp_type = 'ABPM', sbp = "syst", dbp = "DIAST", date_time = "date.time", id = "id", wake = "wake", visit = "visit", hr = "hr", map = "map", rpp = "rpp", pp = "pp", ToD_int = c(5, 13, 18, 23)) jhs_proc <- process_data(bp_jhs, sbp = "Sys.mmHg.", dbp = "Dias.mmHg.", date_time = "DateTime", hr = "pulse.bpm.") rm(bp_hypnos, bp_jhs) bp_hist(hyp_proc) bp_hist(jhs_proc)data("bp_jhs") data("bp_hypnos") hyp_proc <- process_data(bp_hypnos, bp_type = 'ABPM', sbp = "syst", dbp = "DIAST", date_time = "date.time", id = "id", wake = "wake", visit = "visit", hr = "hr", map = "map", rpp = "rpp", pp = "pp", ToD_int = c(5, 13, 18, 23)) jhs_proc <- process_data(bp_jhs, sbp = "Sys.mmHg.", dbp = "Dias.mmHg.", date_time = "DateTime", hr = "pulse.bpm.") rm(bp_hypnos, bp_jhs) bp_hist(hyp_proc) bp_hist(jhs_proc)
ABPM measurements for 5 subjects with Type II diabetes. These data are part of a larger study sample that consisted of patients with Type 2 diabetes recruited from the general community. To be eligible, patients with Type 2 diabetes, not using insulin therapy and with a glycosylated hemoglobin (HbA_1c) value at least 6.5
bp_hypnosbp_hypnos
A data frame with the following columns:
Integer. The index corresponding to the reading of a particular subject for a given visit.
Character. The date-time value corresponding to the given reading.
Integer. The systolic blood pressure reading.
Integer. The mean arterial pressure value.
Integer. The diastolic blood pressure reading.
Integer. The heart rate value (measured in beats per minute – bpm).
Integer. The pulse pressure value calculated as the systolic value - the diastolic value.
Integer. The rate pressure product calculated as the systolic reading multiplied by the heart rate value.
Integer. A logical indicator value corresponding to whether or not a subject is awake (WAKE = 1) or not.
Integer. A unique identifier for each subject.
Integer. A value associated with the visit number or a particular subject.
Character. A date-valued column indicating the date of the given reading. Dates are specified according to their
actual date (i.e. 01:00 corresponds to the next date, even if the subject is awake. Awake-state is indicated via WAKE).
Single-subject self-monitored blood pressure readings over 108 days (April 15, 2019 - August 01, 2019). This data set has been processed and uploaded to the Harvard Dataverse for public use. It contains variables pertaining to Date/Time, Systolic BP, Diastolic BP, and Heart Rate. The data assumes a threshold blood pressure of 135 / 85 which is used to calculate excess amounts.
bp_jhsbp_jhs
A data frame with the following columns:
A POSIXct-formatted column corresponding to the date and time of the corresponding reading in local time.
Integer. The month corresponding to the Date column.
Integer. The day of the month corresponding to the Date column.
Integer. The year corresponding to the Date column.
Character. The day of the week corresponding to the Date column.
Integer. The hour corresponding to the DateTime column.
Character. The estimated meal time corresponding to the DateTime column.
Integer. The systolic blood pressure reading.
Integer. The diastolic blood pressure reading.
Integer. The difference between the Sys.mmHg. and Dias.mmHg. column. This is also known as the Pulse Pressure.
Integer. The heart rate value (measured in beats per minute – bpm).
Calculate the Peak and Trough defined as the max BP - average BP and average BP - min BP, respectively.
bp_mag( data, inc_date = FALSE, subj = NULL, bp_type = c("both", "sbp", "dbp"), add_groups = NULL, inc_wake = TRUE )bp_mag( data, inc_date = FALSE, subj = NULL, bp_type = c("both", "sbp", "dbp"), add_groups = NULL, inc_wake = TRUE )
data |
Required argument. Pre-processed dataframe with SBP and DBP columns
with optional ID, VISIT, WAKE, and DATE columns if available.
Use |
inc_date |
Optional argument. Default is FALSE. As ABPM data typically
overlaps due to falling asleep on one date and waking up on another, the |
subj |
Optional argument. Allows the user to specify and subset specific subjects
from the |
bp_type |
Optional argument. Determines whether to calculate magnitude for SBP values or DBP values, or both. For both SBP and DBP ARV values use bp_type = 'both', for SBP-only use bp_type = 'sbp, and for DBP-only use bp_type = 'dbp'. If no type specified, default will be set to 'both' |
add_groups |
Optional argument. Allows the user to aggregate the data by an
additional "group" to further refine the output. The supplied input must be a
character vector with the strings corresponding to existing column names of the
processed |
inc_wake |
Optional argument corresponding to whether or not to include |
A tibble object with a row corresponding to each subject, or alternatively a row corresponding to each date if inc_date = TRUE. The resulting tibble consists of:
ID: The unique identifier of the subject. For single-subject datasets, ID = 1
VISIT: (If applicable) Corresponds to the visit # of the subject, if more than 1
WAKE: (If applicable) Corresponds to the awake status of the subject (0 = asleep |
1 = awake)
Peak_SBP / Peak_DBP: Measures the difference between the max value and the average
Trough_SBP / Trough_DBP: Measures the difference between the average and the
min value
N: The number of observations for that particular grouping. If inc_date = TRUE,
N corresponds to the number of observations for that date. If inc_date = FALSE, N
corresponds to the number of observations for the most granular grouping available (i.e.
a combination of ID, VISIT, and WAKE)
Any add_groups variables supplied to function argument will be present as a column in the resulting tibble.
# Load data data(bp_hypnos) data(bp_jhs) # Process bp_hypnos hyp_proc <- process_data(bp_hypnos, sbp = "SYST", dbp = "DIAST", date_time = "date.time", id = "id", wake = "wake", visit = "visit", hr = "hr", pp ="pp", map = "map", rpp = "rpp") # Process bp_jhs data jhs_proc <- process_data(bp_jhs, sbp = "Sys.mmHg.", dbp = "Dias.mmHg.", date_time = "DateTime", hr = "Pulse.bpm.") # BP Magnitude Calculation bp_mag(hyp_proc) bp_mag(jhs_proc, inc_date = TRUE)# Load data data(bp_hypnos) data(bp_jhs) # Process bp_hypnos hyp_proc <- process_data(bp_hypnos, sbp = "SYST", dbp = "DIAST", date_time = "date.time", id = "id", wake = "wake", visit = "visit", hr = "hr", pp ="pp", map = "map", rpp = "rpp") # Process bp_jhs data jhs_proc <- process_data(bp_jhs, sbp = "Sys.mmHg.", dbp = "Dias.mmHg.", date_time = "DateTime", hr = "Pulse.bpm.") # BP Magnitude Calculation bp_mag(hyp_proc) bp_mag(jhs_proc, inc_date = TRUE)
The bp_preg data set includes 209 women each with 30 minute recordings during the Pregnancy Day
Assessment Clinic (PDAC) observation window for up to a maximum of 240 minutes (i.e. a maximum
of 8 total readings per subject per observation window in addition to an initial "booking" reading
before the PDAC assessment). This recent study in obstetrics and gynecology investigated
pregnancy-induced hypertension (PIH) and pre-eclampsia (PE) prediction by analyzing cardiac and
physiological information to determine whether the blood pressure assessment of the first observation
window of 1 hour (60 minutes) was sufficient relative to the standard 4 hour (240 minute) window.
bp_pregbp_preg
A data frame containing 55 variables related to physiological assessment during pregnancy. The data variable dictionary is as follows:
Binary indicator for whether an individual exhibits abdominal pain (1 = yes) or not (0 = no)
Fetal heart rate accelerations
Admission from PDAC
Maternal age at estimated date of confinement
Alkaline phosphatase u/L
Alanaine transaminase u/L
Binary indicator for Antenatal diagnosis of pre-eclampsia 1 = yes, 0 = no
Current antihypertensive medication
Binary indicator for whether an individual is asymptomatic (1 = yes) or not (0 = no)
Baseline fetal heart rate (normal, bradycardia, tachycardia)
Body Mass Index
micromol/L
Cardiotocogram
Date of attendance modified by adding a random number of days between -7 and +7
Diastolic Blood Pressure
Fetal heart rate decelerations
Maternal date of birth modified by adding a random number of days between -7 and +7
Medical review during PDAC
Estimated date of confinement modified by adding a random number of days between -7 and +7
Emergency Caesarean section
Final diagnosis: 0 = nil, 1 = pregnancy induced hypertension, 2 = pre-eclampsia
Gestational age (weeks) at attendance
Gamma glutamyl transaminase u/L
Gravidity: number of pregnancies including the current one
Binary indicator for whether an individual has a headache (1 = yes) or not (0 = no)
Maternal Height (cm)
Binary indicator for whether an individual exhibits Hyperreflexia (1 = yes) or not (0 = no)
Induction of labour
Subject ID #
Intrapartum diagnosis of pre-eclampsia 1 = yes, 0 = no
Binary indicator for whether an individual exhibits Nausea (1 = yes) or not (0 = no)
Binary indicator for whether an individual exhibits Oedema (1 = yes) or not (0 = no)
Parity: number of pregnancies proceding to 20 or more weeks, not including the current one
Prescribed aspirin in this pregnancy
Past history of eclampsia
Past history of essential hypertension
Past history of pre-eclampsia
Smoking history
Pre-eclampsia Integrated Estimate of Risk (%)
Postnatal diagnosis of pre-eclampsia 1 = yes, 0 = no
10^9 per mL
Number of previous PDAC assessments this pregnancy
Generic privacy consent form permits participation in audit
Systolic Blood Pressure
Binary indicator for whether an individual exhibits Shortness of Breath (SOB) (1 = yes) or not (0 = no)
Binary indicator for whether an individual is safe for discharge at 1 hour (1 = yes) or not (0 = no)
Binary indicator for whether an individual is safe for discharge at 1 hour - other - comments
Self-discharge from PDAC
30 minute recordings during the Pregnancy Day Assessment Clinic (PDAC) observation window for up to a maximum of 240 minutes including the "Booking" recording
mmol/L
micromol/L
Spot urine protein:creatinine ratio mg/mmol
Fetal heart rate variability as judged visually
Binary indicator for whether an individual exhibits Visual Disturbances (1 = yes) or not (0 = no)
Pre-pregnancy or early pregnancy weight (kg)
(McCarthy EA, Carins TA, Hannigan Y, Bardien N, Shub A, Walker S. 2015)
Licensed under a CC-0 1.0 (Creative Commons) Universal Public Domain Dedication License
Calculates the range (max - min) of both SBP and DBP values in addition to max and min values for reference with the option to specify date as an additional level of granularity
bp_range( data, inc_date = FALSE, subj = NULL, add_groups = NULL, inc_wake = TRUE )bp_range( data, inc_date = FALSE, subj = NULL, add_groups = NULL, inc_wake = TRUE )
data |
Required dataframe with SBP and DBP columns corresponding to
Systolic and Diastolic BP. This dataframe should come from |
inc_date |
Optional argument. Default is FALSE. As ABPM data typically
overlaps due to falling asleep on one date and waking up on another, the |
subj |
Optional argument. Allows the user to specify and subset specific subjects
from the |
add_groups |
Optional argument. Allows the user to aggregate the data by an
additional "group" to further refine the output. The supplied input must be a
character vector with the strings corresponding to existing column names of the
processed |
inc_wake |
Optional argument corresponding to whether or not to include |
A tibble with SBP_max, SBP_min, SBP_range, DBP_max, DBP_min, DBP_range
and any additional optional columns included in data such as ID, VISIT,
WAKE, and DATE. If inc_date = TRUE, each row will correspond to a date.
The resulting tibble consists of:
ID: The unique identifier of the subject. For single-subject datasets, ID = 1
VISIT: (If applicable) Corresponds to the visit # of the subject, if more than 1
WAKE: (If applicable) Corresponds to the awake status of the subject (0 = asleep |
1 = awake)
SBP_max / DBP_max: Finds the maximum value for the given grouping granularity
SBP_min / DBP_min: Finds the minimum value for the given grouping granularity
SBP_range / DBP_range: Calculates the range between the max and min values
N: The number of observations for that particular grouping. If inc_date = TRUE,
N corresponds to the number of observations for that date. If inc_date = FALSE,
N corresponds to the number of observations for the most granular grouping available
(i.e. a combination of ID, VISIT, and WAKE)
Any add_groups variables supplied to function argument will be present as a column in the resulting tibble.
# Load bp_hypnos data(bp_hypnos) data(bp_jhs) # Process bp_hypnos hypnos_proc <- process_data(bp_hypnos, sbp = "SYST", dbp = "DIAST", date_time = "date.time", id = "id", wake = "wake", visit = "visit", hr = "hr", pp ="pp", map = "map", rpp = "rpp") # Process bp_jhs data jhs_proc <- process_data(bp_jhs, sbp = "Sys.mmHg.", dbp = "Dias.mmHg.", date_time = "DateTime", hr = "Pulse.bpm.") # Calculate BP range bp_range(hypnos_proc) bp_range(jhs_proc, inc_date = TRUE, add_groups = c("meal_time")) # Notice that meal_time is not a column from process_data, but it still works# Load bp_hypnos data(bp_hypnos) data(bp_jhs) # Process bp_hypnos hypnos_proc <- process_data(bp_hypnos, sbp = "SYST", dbp = "DIAST", date_time = "date.time", id = "id", wake = "wake", visit = "visit", hr = "hr", pp ="pp", map = "map", rpp = "rpp") # Process bp_jhs data jhs_proc <- process_data(bp_jhs, sbp = "Sys.mmHg.", dbp = "Dias.mmHg.", date_time = "DateTime", hr = "Pulse.bpm.") # Calculate BP range bp_range(hypnos_proc) bp_range(jhs_proc, inc_date = TRUE, add_groups = c("meal_time")) # Notice that meal_time is not a column from process_data, but it still works
The bp_rats data set includes arterial blood pressure waveform time series data
of the SS(n = 9) and SS.13 (n = 6) genetic strains of Dahl rats sampled at 100 Hz who
were each administered a low and high salt diet. It is taken from the study of Bugenhagen
et al.(2010) which sought to investigate the origins of the baroreflex dysfunction in
salt-sensitive Dahl rats on Hypertension.
bp_ratsbp_rats
A data frame containing 5 variables pertaining to arterial continuous waveform blood pressure as follows:
The type of a particular Dahl Rat (either SS or SSBN13) corresponding to the sodium intake administered:
ss_hs = SS rat given the high sodium diet
ss_ls = SS rat given the low sodium diet
ssbn13_hs = SSBN13 rat given the high sodium diet
ssbn13_ls = SSBN13 rat given the low sodium diet
ID # corresponding to a particular rat. There are 9 SS rats and 6 SSBN13 rats.
Continuous waveform data of each rat's arterial (blood) pressure sampled at 100 Hz.
Amount of time elapsed, expressed in seconds
Amount of time elapsed, expressed in minutes
(Goldberger A., Amaral L., Glass L., Hausdorff J., Ivanov P. C., Mark R., Bugenhagen S.M., Cowley A.W. Jr, Beard D.A., ... \& Stanley H. E. 2000).
Licensed under a ODC-BY (Creative Commons) Open Data Commons Attribution License 1.0
Original Paper: doi:10.1152/physiolgenomics.00027.2010
Data: doi:10.13026/C20597
The bp_report function serves to aggregate various data visuals and metrics
pertaining to the supplied data set into a clean formatted report.
bp_report( data, subj = NULL, inc_low = TRUE, inc_crisis = TRUE, group_var = NULL, save_report = TRUE, path = NULL, filename = "bp_report", width = 11, height = 8.5, filetype = "pdf", units = "in", scale = 1.25, plot = TRUE, hist_bins = 30 )bp_report( data, subj = NULL, inc_low = TRUE, inc_crisis = TRUE, group_var = NULL, save_report = TRUE, path = NULL, filename = "bp_report", width = 11, height = 8.5, filetype = "pdf", units = "in", scale = 1.25, plot = TRUE, hist_bins = 30 )
data |
Required argument. A processed dataframe resulting from the
|
subj |
Optional argument. Allows the user to specify and subset specific subjects
from the |
inc_low |
Optional logical argument dictating whether or not to include the "Low" category for BP classification column (and the supplementary SBP/DBP Category columns). Default set to TRUE. |
inc_crisis |
Optional logical argument dictating whether or not to include the "Crisis" category for BP classification column (and the supplementary SBP/DBP Category columns). Default set to TRUE. |
group_var |
A categorical column of the input data set that the individual points are to
be grouped / separated by for a given plot. Cannot contain more than 10 levels (to avoid
overcrowding the plot). This is different from the |
save_report |
A logical value indicating whether to save the BP report output as a separate file.
The default is |
path |
Optional argument. A string corresponding to the respective file path by which the
report is to be saved. Do not include trailing slashes (i.e. ~/loc/) or the file name (i.e. ~/loc/testfile.pdf).
By default, if not |
filename |
Optional argument. A string corresponding to the name of the report. The default is
"bp_report". The string cannot begin with a number or non-alphabetical character.
|
width |
Optional argument. An numeric value corresponding to the width of the output document. The default is set to 12 inches. |
height |
Optional argument. An numeric value corresponding to the height of the output document. The default is set to 8.53 inches. |
filetype |
A string corresponding to he particular type of file that the report is to be saved as. Although PDF is the default possible options include:
|
units |
A character string corresponding to the unit of measurement that the width and height correspond to in the exported output. The default is inches ("in"), but centimeters ("cm") and millimeters ("mm") are also available. |
scale |
A multiplicative scaling factor for the report output. |
plot |
A logical value indicating whether to automatically produce the plot of bp_report, or suppress the output. The default value is TRUE. If false, the returned object is a grob that can be plotted using |
hist_bins |
An integer specifying how many bins to use for histogram plots. This is a ggplot parameter; default value set to 30 |
If plot = TRUE, the function produces a plot of BP report that contains scatterplot of BP values by stages (see bp_scatter), histograms of BP values by stages (see bp_hist) and frequency tables of BP values by stages and day of the week/time of the day (see dow_tod_plots). If plot = FALSE, the function returns the grob object that can be plotted later using grid.arrange. If save_report = TRUE, the report will be automatically saved at the current working directory (can be checked using getwd()) or at specified file path.
data("bp_jhs") data("bp_hypnos") hyp_proc <- process_data(bp_hypnos, sbp = "syst", dbp = "DIAST", date_time = "date.time", id = "id", wake = "wake", visit = "visit", hr = "hr", map = "map", rpp = "rpp", pp = "pp", ToD_int = c(5, 13, 18, 23)) jhs_proc <- process_data(bp_jhs, sbp = "Sys.mmHg.", dbp = "Dias.mmHg.", date_time = "DateTime", hr = "pulse.bpm.") rm(bp_hypnos, bp_jhs) ## Not run: # Single-subject Report # save_report = FALSE for illustrative purposes # plot = TRUE will automatically generate a plot bp_report(jhs_proc, save_report = FALSE, plot = TRUE) # Multi-subject Report # save_report = FALSE for illustrative purposes # plot = FALSE will suppress the plot output and return a grob object out = bp_report(hyp_proc, group_var = 'VISIT', save_report = FALSE, plot = FALSE) gridExtra::grid.arrange(out) ## End(Not run)data("bp_jhs") data("bp_hypnos") hyp_proc <- process_data(bp_hypnos, sbp = "syst", dbp = "DIAST", date_time = "date.time", id = "id", wake = "wake", visit = "visit", hr = "hr", map = "map", rpp = "rpp", pp = "pp", ToD_int = c(5, 13, 18, 23)) jhs_proc <- process_data(bp_jhs, sbp = "Sys.mmHg.", dbp = "Dias.mmHg.", date_time = "DateTime", hr = "pulse.bpm.") rm(bp_hypnos, bp_jhs) ## Not run: # Single-subject Report # save_report = FALSE for illustrative purposes # plot = TRUE will automatically generate a plot bp_report(jhs_proc, save_report = FALSE, plot = TRUE) # Multi-subject Report # save_report = FALSE for illustrative purposes # plot = FALSE will suppress the plot output and return a grob object out = bp_report(hyp_proc, group_var = 'VISIT', save_report = FALSE, plot = FALSE) gridExtra::grid.arrange(out) ## End(Not run)
Display all SBP and DBP
readings on a scatterplot with deliniation of BP according to the 8
mutually exclusive levels of Hypertension as in Lee et al (2020) (the default), or the
levels set by the American Heart Association (AHA).
bp_scatter( data, plot_type = c("stages2020", "AHA"), subj = NULL, group_var = NULL, wrap_var = NULL, inc_crisis = TRUE, inc_low = TRUE, bp_cutoffs = list(c(100, 120, 130, 140, 180), c(60, 80, 80, 90, 120)), bp_type = c("hbpm", "abpm", "ap") )bp_scatter( data, plot_type = c("stages2020", "AHA"), subj = NULL, group_var = NULL, wrap_var = NULL, inc_crisis = TRUE, inc_low = TRUE, bp_cutoffs = list(c(100, 120, 130, 140, 180), c(60, 80, 80, 90, 120)), bp_type = c("hbpm", "abpm", "ap") )
data |
A processed dataframe resulting from the |
plot_type |
String corresponding to the particular type of plot to be displayed. Default
plot ( |
subj |
Optional argument. Allows the user to specify and subset specific subjects
from the |
group_var |
A categorical column of the input data set that the individual points are to
be grouped / separated by for a given plot. Cannot contain more than 10 levels (to avoid
overcrowding the plot). This is different from the |
wrap_var |
A categorical column of the input data set that the plots are to be segmented
by. If there are multiple levels such as time of day, or visit number, the output will include a
matrix with each plot corresponding to an individual level. This differs from the |
inc_crisis |
A TRUE / FALSE indicator of whether or not to include the "Crisis" (Hypertensive)
category to the scatter plot. This is only activated in conjunction with |
inc_low |
A TRUE / FALSE indicator of whether or not to include the "Low" (Hypotension)
category to the scatter plot. This is only activated in conjunction with |
bp_cutoffs |
A list containing two vectors corresponding to SBP and DBP cutoffs, respectively. Each vector contains 5 values. The SBP vector (100, 120, 130, 140, 180) corresponds to the upper limits for the following stages: Low (0-100), Normal (100-120), Elevated (120-130), Stage 1 Hypertension (130-140), Stage 2 Hypertension (140-180). When utilizing Lee et al (2020) guidelines, additional stages are included: Isolated Systolic Hypertension for Stage 1 (ISH - S1) (130-140), Isolated Diastolic Hypertension for Stage 1 (IDH - S1) (0-130), ISH - S2 (140-180), and IDH - S2 (0-140). The DBP vector (60, 80, 80, 90, 120) corresponds to the upper limits for the following stages: Low (0-60), Normal (60-80), Elevated (0-80), Stage 1 Hypertension (80-90), Stage 2 Hypertension (90-120). The upper limit of the "Elevated" category repeats in the DBP vector and matches that of Normal. This because according to most guidelines, there is no distinction between DBP cutoffs for Normal and Elevated - these stages are discerned by SBP, not DBP. When utilizing Lee et al (2020) guidelines, additional stages are included: Isolated Diastolic Hypertension for Stage 1 (ISH - S1) (0-80), Isolated Diastolic Hypertension for Stage 1 (IDH - S1) (80-90), ISH - S2 (0-90), and IDH - S2 (90-120). Any SBP reading below 100 or DBP reading below 60 is considered Hypotension ("Low"). Any SBP reading above 180 or DBP reading above 120 is considered a Crisis. When If |
bp_type |
Required argument specifying which of the three BP data types
("HBPM", "ABPM", or "AP") the input data is. Default HBPM - Home Blood Pressure Monitor | ABPM - Ambulatory Blood Pressure | AP - Arterial Pressure NOTE: |
There are eight total stages according to Lee et al (2020) with the options to include two additional categories for "Low" (Hypotension) and Hypertensive "Crisis". The categories are as follows:
Low - (Optional) Legacy category for consistency with AHA stages. According to the AHA, low blood pressure is any reading with SBP < 100 and DBP < 60, and is depicted in light blue in the scatter plot. This is always displayed in "AHA" plot, and can be displayed in "stages2020" plot by setting inc_low = TRUE.
Normal - SBP readings less than 120 and DBP readings
less than 80. Reading within this range that either have SBP > 100 or DBP > 60 are also considered Normal by AHA. Normal BP is depicted in green in the scatter plot.
Elevated - SBP readings between 120 - 129 and DBP
readings less than 80. Coincides with Elevated stage as defined by AHA. Without intervention to control the condition, individuals are likely to develop Hypertension. Elevated BP is depicted in yellow in the scatter plot.
Stage 1 - All (SDH) - SBP readings between 130 - 139 and DBP
readings between 80 - 89. Stage 1 Hypertension will typically result in doctors prescribing
medication or lifestyle changes. Stage 1 BP is depicted in dark orange in the scatter plot. These readings correspond to Stage 1 as defined by AHA.
Stage 1 - Isolated Diastolic Hypertension (IDH) - SBP readings
less than 130, but DBP readings between 80 - 89. This alternative stage 1 level accounts
for unusually high diastolic readings, but fairly normal systolic readings and is depicted in
orange in the plot. These readings correspond to Stage 1 as defined by AHA.
Stage 1 - Isolated Systolic Hypertension (ISH) - SBP readings
between 130 - 139, but DBP readings less than 80. This alternative stage 1 level accounts
for unusually high systolic readings, but fairly normal diastolic readings and is depicted in
orange in the plot. These readings correspond to Stage 1 as defined by AHA.
Stage 2 - All (SDH) - SBP readings between 140 - 180 and DBP readings
between 90 - 120. Stage 2 Hypertension will typically result in doctors prescribing both
medication and lifestyle changes. Stage 2 BP is depicted in bright red in the scatter plot. These readings correspond to Stage 2 as defined by AHA.
Stage 2 - Isolated Diastolic Hypertension (IDH) - SBP readings
less than or equal to 140, but DBP readings greater than or equal to 90. This alternative
stage 2 level accounts for unusually high diastolic readings, but fairly normal systolic readings
and is depicted in red. These readings correspond to Stage 2 as defined by AHA.
Stage 2 - Isolated Systolic Hypertension (IDH) - SBP readings
greater than or equal to 140, but DBP readings less or equal to 90. This alternative
stage 2 level accounts for unusually high systolic readings, but fairly normal diastolic readings
and is depicted in red. These readings correspond to Stage 2 as defined by AHA.
Crisis - (Optional) Legacy category for consistency with AHA stages. According to the AHA, hypertensive crisis is defined as a SBP reading exceeding 180 or a
DBP reading exceeding 120. This stage requires medical attention immediately.
Crisis is depicted in red in the scatter plot. This is always displayed in "AHA" plot, and can be displayed in "stages2020" plot by setting inc_crisis = TRUE.
A scatter plot graphic using the ggplot2 package overlaying each reading (represented as points) onto a background that contains each stage
Lee H, Yano Y, Cho SMJ, Park JH, Park S, Lloyd-Jones DM, Kim HC. Cardiovascular risk of isolated systolic or diastolic hypertension in young adults. Circulation. 2020; 141:1778–1786. doi:10.1161/CIRCULATIONAHA.119.044838
Unger, T., Borghi, C., Charchar, F., Khan, N. A., Poulter, N. R., Prabhakaran, D., ... & Schutte, A. E. (2020). 2020 International Society of Hypertension global hypertension practice guidelines. Hypertension, 75(6), 1334-1357. doi:10.1161/HYPERTENSIONAHA.120.15026
data("bp_jhs") data("bp_hypnos") data("bp_ghana") hypnos_proc <- process_data(bp_hypnos, bp_type = 'abpm', sbp = "syst", dbp = "DIAST", date_time = "date.time", id = "id", wake = "wake", visit = "visit", hr = "hr", map = "map", rpp = "rpp", pp = "pp", ToD_int = c(5, 13, 18, 23)) jhs_proc <- process_data(bp_jhs, sbp = "Sys.mmHg.", dbp = "Dias.mmHg.", date_time = "DateTime", hr = "pulse.bpm.") rm(bp_hypnos, bp_jhs) # HYPNOS Data bp_scatter(hypnos_proc, inc_crisis = TRUE, inc_low = TRUE, group_var = "wake", wrap_var = "day_of_week") # JHS Data bp_scatter(jhs_proc, plot_type = "AHA", group_var = "time_of_day") # Ghana Data Set #(Note that column names are of proper naming convention so no processing needed) bp_scatter(bp::bp_ghana, inc_crisis = TRUE, inc_low = FALSE, group_var = "TIME_ELAPSED")data("bp_jhs") data("bp_hypnos") data("bp_ghana") hypnos_proc <- process_data(bp_hypnos, bp_type = 'abpm', sbp = "syst", dbp = "DIAST", date_time = "date.time", id = "id", wake = "wake", visit = "visit", hr = "hr", map = "map", rpp = "rpp", pp = "pp", ToD_int = c(5, 13, 18, 23)) jhs_proc <- process_data(bp_jhs, sbp = "Sys.mmHg.", dbp = "Dias.mmHg.", date_time = "DateTime", hr = "pulse.bpm.") rm(bp_hypnos, bp_jhs) # HYPNOS Data bp_scatter(hypnos_proc, inc_crisis = TRUE, inc_low = TRUE, group_var = "wake", wrap_var = "day_of_week") # JHS Data bp_scatter(jhs_proc, plot_type = "AHA", group_var = "time_of_day") # Ghana Data Set #(Note that column names are of proper naming convention so no processing needed) bp_scatter(bp::bp_ghana, inc_crisis = TRUE, inc_low = FALSE, group_var = "TIME_ELAPSED")
The bp_sleep_metrics function serves to calculate nocturnal metrics
from relevant medical literature.
bp_sleep_metrics(data, subj = NULL)bp_sleep_metrics(data, subj = NULL)
data |
User-supplied data set containing blood pressure data. Must
contain a Systolic blood pressure ( |
subj |
Optional argument. Allows the user to specify and subset specific subjects
from the |
NOTE: Any reference to "sleep" in the bp package refers to an individual's nocturnal period; "sleep" is used in an informal sense for intuitive purposes. Technically, from a clinical perspective, indication of sleep is currently obtained through means of EEG or other highly specialized equipment. For all intents and purposes, sleep in the context of this package refers to actigraphy-inferred nocturnal periods of rest.
The calculation of BP metrics related to sleep is based on averages of BP readings from four periods as identified in Kairo et al. (2003):
presleep or evening (2 hours before sleep start),
prewake (2 hours before wake),
postwake or morning (2 hours after wake),
lowest (3 measurements centered at the minimal BP reading over sleep).
The function uses WAKE column to automatically allocate BP measurements to various periods. The following metrics are defined as a
function of the period averages (separately for SBP and DBP)
dip_calc = 1 - mean_sleep_BP/mean_wake_BP (dip proportion)
noct_fall = mean_presleep_BP - mean_lowest_BP (nocturnal fall)
ST_mbps = mean_postwake_BP - mean_lowest_BP (sleep through morning blood pressure surge)
PW_mbps = mean_postwake_BP - mean_prewake_BP (prewake morning blood pressure surge)
ME_avg = (mean_presleep_BP + mean_postwake_BP)/2 (morning-evening average)
ME_diff = mean_postwake_BP - mean_presleep_BP (morning-evening difference)
wSD = ( (wake_SD x HRS_wake) + (sleep_SD x HRS_sleep) ) / (HRS_wake + HRS_sleep) (weighted standard deviation)
The function outputs a list containing 4 tibble objects corresponding to the following tables:
[[1]] |
Counts of how many BP measurements were observed overall ( |
[[2]] |
Summary statistics for systolic BP measurements (SBP): mean SBP value over Sleep and Wake, sd SBP value over Sleep and Wake, mean SBP value over presleep period (evening in Kario et al. (2003)), mean SBP value over prewake period, mean SBP value over postwake period (morning in Kario et al. (2003)), mean SBP value over 3 reading centered at the lowest SBP value during sleep |
[[3]] |
Summary statistics for diastolic BP measurements (DBP), same as for SBP |
[[4]] |
BP metrics associated with sleep as defined above, separately for SBP and DBP |
Kario, K., Pickering, T. G., Umeda, Y., Hoshide, S., Hoshide, Y., Morinari, M., ... & Shimada, K. (2003). Morning surge in blood pressure as a predictor of silent and clinical cerebrovascular disease in elderly hypertensives: a prospective study. Circulation, 107(10), 1401-1406.
hypnos_proc <- process_data(bp_hypnos, sbp = "syst", dbp = "DIAST", date_time = "date.time", id = "id", wake = "wake", visit = "visit", hr = "hr", map = "map", rpp = "rpp", pp = "pp", bp_type = "ABPM") bp_sleep_metrics(hypnos_proc)hypnos_proc <- process_data(bp_hypnos, sbp = "syst", dbp = "DIAST", date_time = "date.time", id = "id", wake = "wake", visit = "visit", hr = "hr", map = "map", rpp = "rpp", pp = "pp", bp_type = "ABPM") bp_sleep_metrics(hypnos_proc)
Adds BP_CLASS, SBP_Category, and DBP_Category columns to supplied dataframe.
bp_stages( data, sbp, dbp, bp_type = c("hbpm", "abpm", "ap"), inc_low = TRUE, inc_crisis = TRUE, data_screen = TRUE, SUL = 240, SLL = 50, DUL = 140, DLL = 40, adj_sbp_dbp = TRUE, guidelines = c("Lee_2020", "AHA", "Custom"), bp_cutoffs = list(c(100, 120, 130, 140, 180), c(60, 80, 80, 90, 120)) )bp_stages( data, sbp, dbp, bp_type = c("hbpm", "abpm", "ap"), inc_low = TRUE, inc_crisis = TRUE, data_screen = TRUE, SUL = 240, SLL = 50, DUL = 140, DLL = 40, adj_sbp_dbp = TRUE, guidelines = c("Lee_2020", "AHA", "Custom"), bp_cutoffs = list(c(100, 120, 130, 140, 180), c(60, 80, 80, 90, 120)) )
data |
User-supplied dataset containing blood pressure data. Must contain data for Systolic blood pressure and Diastolic blood pressure at a minimum. |
sbp |
Required column name (character string) corresponding to Systolic Blood Pressure (mmHg) |
dbp |
Required column name (character string) corresponding to Diastolic Blood Pressure (mmHg) |
bp_type |
Required argument specifying which of the three BP data types
("HBPM", "ABPM", or "AP") the input data is. Default HBPM - Home Blood Pressure Monitor | ABPM - Ambulatory Blood Pressure | AP - Arterial Pressure NOTE: |
inc_low |
Optional logical argument dictating whether or not to include the "Low" category for BP classification column (and the supplementary SBP/DBP Category columns). Default set to TRUE. |
inc_crisis |
Optional logical argument dictating whether or not to include the "Crisis" category for BP classification column (and the supplementary SBP/DBP Category columns). Default set to TRUE. |
data_screen |
Optional logical argument; default set to TRUE. Screens for extreme values in the data
for both |
SUL |
Systolic Upper Limit (SUL). If |
SLL |
Systolic Lower Limit (SLL). If |
DUL |
Diastolic Upper Limit (DUL). If |
DLL |
Diastolic Lower Limit (DLL). If |
adj_sbp_dbp |
Logical indicator to dictate whether or not to run helper functions that adjust / process
SBP & DBP columns in supplied data set. Default set to: |
guidelines |
A string designation for the guidelines to follow when mapping BP
readings to a respective BP stage. |
bp_cutoffs |
A list containing two vectors corresponding to SBP and DBP cutoffs, respectively. Each vector contains 5 values. The SBP vector (100, 120, 130, 140, 180) corresponds to the upper limits for the following stages: Low (0-100), Normal (100-120), Elevated (120-130), Stage 1 Hypertension (130-140), Stage 2 Hypertension (140-180). When utilizing Lee et al (2020) guidelines, additional stages are included: Isolated Systolic Hypertension for Stage 1 (ISH - S1) (130-140), Isolated Diastolic Hypertension for Stage 1 (IDH - S1) (0-130), ISH - S2 (140-180), and IDH - S2 (0-140). The DBP vector (60, 80, 80, 90, 120) corresponds to the upper limits for the following stages: Low (0-60), Normal (60-80), Elevated (0-80), Stage 1 Hypertension (80-90), Stage 2 Hypertension (90-120). The upper limit of the "Elevated" category repeats in the DBP vector and matches that of Normal. This because according to most guidelines, there is no distinction between DBP cutoffs for Normal and Elevated - these stages are discerned by SBP, not DBP. When utilizing Lee et al (2020) guidelines, additional stages are included: Isolated Diastolic Hypertension for Stage 1 (ISH - S1) (0-80), Isolated Diastolic Hypertension for Stage 1 (IDH - S1) (80-90), ISH - S2 (0-90), and IDH - S2 (90-120). Any SBP reading below 100 or DBP reading below 60 is considered Hypotension ("Low"). Any SBP reading above 180 or DBP reading above 120 is considered a Crisis. When If |
Supplied dataframe must adhere to the unified format using the process_data function.
A dataframe with additional columns corresponding to the stages of blood pressure and the supplementary SBP / DBP categories
Lee H, Yano Y, Cho SMJ, Park JH, Park S, Lloyd-Jones DM, et al. Cardiovascular risk of isolated systolic or diastolic hypertension in young adults. Circulation. 2020;141(22):1778–1786. doi:0.1161/CIRCULATIONAHA.119.044838
Muntner, P., Carey, R. M., Jamerson, K., Wright Jr, J. T., & Whelton, P. K. (2019). Rationale for ambulatory and home blood pressure monitoring thresholds in the 2017 American College of Cardiology/American Heart Association Guideline. Hypertension, 73(1), 33-38. doi:10.1161/HYPERTENSIONAHA.118.11946
# Load bp_hypnos data(bp_hypnos) bp_stages(bp_hypnos, sbp = "syst", dbp = "diast") # Load bp_jhs data data(bp_jhs) bp_stages(bp_jhs, sbp = "sys.mmhg.", dbp = "dias.mmhg.")# Load bp_hypnos data(bp_hypnos) bp_stages(bp_hypnos, sbp = "syst", dbp = "diast") # Load bp_jhs data data(bp_jhs) bp_stages(bp_jhs, sbp = "sys.mmhg.", dbp = "dias.mmhg.")
Combines the output from the following functions:
bp_center
bp_cv
bp_arv
bp_sv
bp_mag
bp_range
bp_stats( data, inc_date = FALSE, subj = NULL, bp_type = c("both", "sbp", "dbp"), add_groups = NULL, inc_wake = TRUE )bp_stats( data, inc_date = FALSE, subj = NULL, bp_type = c("both", "sbp", "dbp"), add_groups = NULL, inc_wake = TRUE )
data |
Required argument. Pre-processed dataframe containing SBP and
DBP with optional ID, VISIT, WAKE, and DATE columns if available.
Use |
inc_date |
Optional argument. Default is FALSE. As ABPM data typically
overlaps due to falling asleep on one date and waking up on another, the |
subj |
Optional argument. Allows the user to specify and subset specific subjects
from the |
bp_type |
Optional argument. Determines whether to calculate magnitude for SBP values or DBP values, or both. For both SBP and DBP ARV values use bp_type = 'both', for SBP-only use bp_type = 'sbp, and for DBP-only use bp_type = 'dbp'. If no type specified, default will be set to 'both' |
add_groups |
Optional argument. Allows the user to aggregate the data by an
additional "group" to further refine the output. The supplied input must be a
character vector with the strings corresponding to existing column names of the
processed |
inc_wake |
Optional argument corresponding to whether or not to include |
A tibble object with a row corresponding to each subject, or alternatively a row corresponding to each date, if inc_date = TRUE. The resulting tibble consists of:
ID: The unique identifier of the subject. For single-subject datasets, ID = 1
N: The number of observations for that particular grouping. If inc_date = TRUE,
N corresponds to the number of observations for that date. If inc_date = FALSE, N
corresponds to the number of observations for the most granular grouping available (i.e.
a combination of ID, VISIT, and WAKE)
VISIT: (If applicable) Corresponds to the visit # of the subject, if more than 1
WAKE: (If applicable) Corresponds to the awake status of the subject (0 = asleep |
1 = awake)
Output from the following functions: bp_center, bp_cv, bp_arv, bp_sv,
bp_mag, bp_range
Any add_groups variables supplied to function argument will be present as a column in the resulting tibble.
# Load data data(bp_hypnos) data(bp_jhs) # Process bp_hypnos hyp_proc <- process_data(bp_hypnos, sbp = "SYST", dbp = "DIAST", date_time = "date.time", id = "id", wake = "wake", visit = "visit", hr = "hr", pp ="pp", map = "map", rpp = "rpp") # Process bp_jhs data jhs_proc <- process_data(bp_jhs, sbp = "Sys.mmHg.", dbp = "Dias.mmHg.", date_time = "DateTime", hr = "Pulse.bpm.") # BP Stats Aggregated Table bp_stats(hyp_proc, subj = c(70417, 70435), add_groups = c("SBP_Category"), bp_type = 'sbp') bp_stats(jhs_proc, add_groups = c("SBP_Category"))# Load data data(bp_hypnos) data(bp_jhs) # Process bp_hypnos hyp_proc <- process_data(bp_hypnos, sbp = "SYST", dbp = "DIAST", date_time = "date.time", id = "id", wake = "wake", visit = "visit", hr = "hr", pp ="pp", map = "map", rpp = "rpp") # Process bp_jhs data jhs_proc <- process_data(bp_jhs, sbp = "Sys.mmHg.", dbp = "Dias.mmHg.", date_time = "DateTime", hr = "Pulse.bpm.") # BP Stats Aggregated Table bp_stats(hyp_proc, subj = c(70417, 70435), add_groups = c("SBP_Category"), bp_type = 'sbp') bp_stats(jhs_proc, add_groups = c("SBP_Category"))
Calculate the successive variation (SV) at various levels of granularity based on what is supplied (ID, VISIT, WAKE, and / or DATE)for either SBP, DBP, or both. SV is a measure of dispersion that takes into account the temporal structure of the data and relies on the sum of squared differences in successive observations, unlike the average real variability (ARV) which relies on the sum of absolute differences. $$SV = sqrt(sum(x_i+1 - x_i)^2/n-1)$$
bp_sv( data, inc_date = FALSE, subj = NULL, bp_type = c("both", "sbp", "dbp"), add_groups = NULL, inc_wake = TRUE )bp_sv( data, inc_date = FALSE, subj = NULL, bp_type = c("both", "sbp", "dbp"), add_groups = NULL, inc_wake = TRUE )
data |
Required argument. Pre-processed dataframe with SBP and DBP columns
with optional ID, VISIT, WAKE, and DATE columns if available.
Use |
inc_date |
Optional argument. Default is FALSE. As ABPM data typically
overlaps due to falling asleep on one date and waking up on another, the |
subj |
Optional argument. Allows the user to specify and subset specific subjects
from the |
bp_type |
Optional argument. Determines whether to calculate SV for SBP values or DBP values, or both. For both SBP and DBP ARV values use bp_type = 'both', for SBP-only use bp_type = 'sbp, and for DBP-only use bp_type = 'dbp'. If no type specified, default will be set to 'both' |
add_groups |
Optional argument. Allows the user to aggregate the data by an
additional "group" to further refine the output. The supplied input must be a
character vector with the strings corresponding to existing column names of the
processed |
inc_wake |
Optional argument corresponding to whether or not to include |
NOTE: The canonical standard deviation, independent of the temporal structure using the sample average, is added for comparison: $$SD = sqrt(sum(x_i+1 - xbar)^2/n-1)$$
A tibble object with a row corresponding to each subject, or alternatively a row corresponding to each date if inc_date = TRUE. The resulting tibble consists of:
ID: The unique identifier of the subject. For single-subject datasets, ID = 1
VISIT: (If applicable) Corresponds to the visit # of the subject, if more than 1
WAKE: (If applicable) Corresponds to the awake status of the subject (0 = asleep |
1 = awake)
SV_SBP / SV_DBP: Calculates the square root of the average squared differences
between successive measurements. The resulting value averages across the granularity
grouping for however many observations are present.
N: The number of observations for that particular grouping. If inc_date = TRUE,
N corresponds to the number of observations for that date. If inc_date = FALSE, N
corresponds to the number of observations for the most granular grouping available (i.e.
a combination of ID, VISIT, and WAKE)
Any add_groups variables supplied to function argument will be present as a column in the resulting tibble.
# Load data data(bp_hypnos) data(bp_jhs) # Process bp_hypnos hypnos_proc <- process_data(bp_hypnos, sbp = "SYST", dbp = "DIAST", date_time = "date.time", id = "id", wake = "wake", visit = "visit", hr = "hr", pp ="pp", map = "map", rpp = "rpp") # Process bp_jhs data jhs_proc <- process_data(bp_jhs, sbp = "Sys.mmHg.", dbp = "Dias.mmHg.", date_time = "DateTime", hr = "Pulse.bpm.") # SV Calculation bp_sv(hypnos_proc) bp_sv(jhs_proc, add_groups = c("meal_time")) # Notice that meal_time is not a column from process_data, but it still works# Load data data(bp_hypnos) data(bp_jhs) # Process bp_hypnos hypnos_proc <- process_data(bp_hypnos, sbp = "SYST", dbp = "DIAST", date_time = "date.time", id = "id", wake = "wake", visit = "visit", hr = "hr", pp ="pp", map = "map", rpp = "rpp") # Process bp_jhs data jhs_proc <- process_data(bp_jhs, sbp = "Sys.mmHg.", dbp = "Dias.mmHg.", date_time = "DateTime", hr = "Pulse.bpm.") # SV Calculation bp_sv(hypnos_proc) bp_sv(jhs_proc, add_groups = c("meal_time")) # Notice that meal_time is not a column from process_data, but it still works
Generate a list of pertinent table outputs that detail various information specific to blood pressure.
bp_tables( data, bp_type = c("both", "sbp", "dbp"), bp_perc_margin = NULL, wake_perc_margin = 2, subj = NULL )bp_tables( data, bp_type = c("both", "sbp", "dbp"), bp_perc_margin = NULL, wake_perc_margin = 2, subj = NULL )
data |
A processed dataframe resulting from the |
bp_type |
Optional argument. Determines whether to calculate tables for SBP values or DBP values, or both. For both SBP and DBP ARV values use bp_type = 'both', for SBP-only use bp_type = 'sbp, and for DBP-only use bp_type = 'dbp'. If no type specified, default will be set to 'both' |
bp_perc_margin |
An optional argument that determines which of the marginal totals to include in the raw count tables expressed as percentages. The argument can take on values either NULL (default, both SBP and DBP), 1 (SBP only), or 2 (DBP only). |
wake_perc_margin |
An optional argument that determines which of the marginal totals to include in the tables pertaining to the percentages of awake / asleep readings if applicable (i.e. if the WAKE column is present). The argument can take on values either NULL (both SBP and DBP), 1 (SBP only), or 2 (DBP only). |
subj |
Optional argument. Allows the user to specify and subset specific subjects
from the |
A list of table outputs for various subsets of the data based on which bp_type is selected
data("bp_jhs") data("bp_hypnos") hyp_proc <- process_data(bp_hypnos, bp_type = 'ABPM', sbp = "syst", dbp = "DIAST", date_time = "date.time", id = "id", wake = "wake", visit = "visit", hr = "hr", map = "map", rpp = "rpp", pp = "pp", ToD_int = c(5, 13, 18, 23)) jhs_proc <- process_data(bp_jhs, sbp = "Sys.mmHg.", dbp = "Dias.mmHg.", date_time = "DateTime", hr = "pulse.bpm.") rm(bp_hypnos, bp_jhs) bp_tables(jhs_proc) bp_tables(hyp_proc)data("bp_jhs") data("bp_hypnos") hyp_proc <- process_data(bp_hypnos, bp_type = 'ABPM', sbp = "syst", dbp = "DIAST", date_time = "date.time", id = "id", wake = "wake", visit = "visit", hr = "hr", map = "map", rpp = "rpp", pp = "pp", ToD_int = c(5, 13, 18, 23)) jhs_proc <- process_data(bp_jhs, sbp = "Sys.mmHg.", dbp = "Dias.mmHg.", date_time = "DateTime", hr = "pulse.bpm.") rm(bp_hypnos, bp_jhs) bp_tables(jhs_proc) bp_tables(hyp_proc)
Blood Pressure Time Series Plots
bp_ts_plots( data, index = NULL, subj = NULL, first_hour = 0, rotate_xlab = FALSE, wrap_var = NULL, wrap_row = NULL, wrap_col = NULL, method = NULL, formula = NULL )bp_ts_plots( data, index = NULL, subj = NULL, first_hour = 0, rotate_xlab = FALSE, wrap_var = NULL, wrap_row = NULL, wrap_col = NULL, method = NULL, formula = NULL )
data |
User-supplied data set containing blood pressure data. Must contain a Systolic blood pressure (SBP), Diastolic blood pressure (DBP) and an ID column. Data must also have either a DATE_TIME or DATE column, unless an index column is specified for the x axis. An index column trumps DATE_TIME and DATE if specified. |
index |
An optional user-specified column denoting x-axis values (other
than DATE_TIME or DATE columns). |
subj |
Optional argument. Allows the user to specify and subset specific subjects
from the |
first_hour |
Optional argument denoting a value corresponding to the first hour of the
x-axis for the hour plots. Only applicable to data sets that contain a DATE_TIME column.
It is often easier to visualize a BP time series not from 0 - 23 hours but rather an order
that begins or ends with waking up such as |
rotate_xlab |
An optional logical argument to rotate the x axis labels 90 degrees. The default value is set to FALSE. |
wrap_var |
An optional character argument indicating a column by which to "wrap" the data.
This function utilizes ggplot2's |
wrap_row |
An optional argument specifying how many rows to wrap the plots if |
wrap_col |
An optional argument specifying how many columnss to wrap the plots if |
method |
(ggplot2 plotting arguments) Smoothing method (function) to use. Default is NULL, but also accepts a character vector "lm", "glm", "gam", "loess". NULL implies that the smoothing method will be chosen automatically based on the size of the largest group. See https://ggplot2.tidyverse.org/reference/geom_smooth.html for more details. |
formula |
(ggplot2 plotting arguments) Formula to use in smoothing function. Default is NULL implying y ~ x for fewer than 1,000 observations and y ~ x(x, bs = "cs") otherwise. See https://ggplot2.tidyverse.org/reference/geom_smooth.html for more details. |
If the data does not contain a DATE_TIME column, a single list will be returned with the time-dependent plots for each subject ID. If the data does contain a DATE_TIME column (and index is not specified), a list of two lists will be returned for each subject ID: one corresponding to the time-dependent plots (according to the DATE_TIME values), and another plot corresponding to the HOUR plots which show repeated measurements of BP values throughout a 24-hour period. The index of the output therefore corresponds to whether there is only the time-dependent plot type (the former situation) or there are both time-dependent and hourly plot types (the latter situation).
# Pregnancy Data Set # bp_preg requires the use of the index argument since there are no DATE or # DATE_TIME columns available data_preg <- bp::bp_preg data_preg$Time_Elapsed <- factor(data_preg$Time_Elapsed, levels = c("Booking", "0", "30", "60", "90", "120", "150", "180", "210", "240")) bp::bp_ts_plots(data_preg, index = 'time_elapsed', subj = 1:3) # JHS Data Set # bp_jhs returns two lists since there is a DATE_TIME column: one for # DATE_TIME and one for HOUR data_jhs <- bp::process_data(bp::bp_jhs, sbp = 'sys.mmhg.', dbp = 'dias.mmhg.', hr = 'pulse.bpm.', date_time = 'datetime') bp::bp_ts_plots(data_jhs) # HYPNOS Data Set # bp_hypnos wraps the plots by the visit # since each subject was recorded over # the course of two office visits data_hypnos <- bp::process_data(bp::bp_hypnos, sbp = 'syst', dbp = 'diast', date_time = 'date.time') bp::bp_ts_plots(data_hypnos, wrap_var = 'visit', subj = '70435')# Pregnancy Data Set # bp_preg requires the use of the index argument since there are no DATE or # DATE_TIME columns available data_preg <- bp::bp_preg data_preg$Time_Elapsed <- factor(data_preg$Time_Elapsed, levels = c("Booking", "0", "30", "60", "90", "120", "150", "180", "210", "240")) bp::bp_ts_plots(data_preg, index = 'time_elapsed', subj = 1:3) # JHS Data Set # bp_jhs returns two lists since there is a DATE_TIME column: one for # DATE_TIME and one for HOUR data_jhs <- bp::process_data(bp::bp_jhs, sbp = 'sys.mmhg.', dbp = 'dias.mmhg.', hr = 'pulse.bpm.', date_time = 'datetime') bp::bp_ts_plots(data_jhs) # HYPNOS Data Set # bp_hypnos wraps the plots by the visit # since each subject was recorded over # the course of two office visits data_hypnos <- bp::process_data(bp::bp_hypnos, sbp = 'syst', dbp = 'diast', date_time = 'date.time') bp::bp_ts_plots(data_hypnos, wrap_var = 'visit', subj = '70435')
THIS IS A DEPRECATED FUNCTION. USE bp_cv INSTEAD.
cv( data, inc_date = FALSE, subj = NULL, bp_type = 0, add_groups = NULL, inc_wake = TRUE )cv( data, inc_date = FALSE, subj = NULL, bp_type = 0, add_groups = NULL, inc_wake = TRUE )
data |
Required argument. Pre-processed dataframe with SBP and DBP columns
with optional ID, VISIT, WAKE, and DATE columns if available.
Use |
inc_date |
Optional argument. Default is FALSE. As ABPM data typically
overlaps due to falling asleep on one date and waking up on another, the |
subj |
Optional argument. Allows the user to specify and subset specific subjects
from the |
bp_type |
Optional argument. Determines whether to calculate ARV for SBP values or DBP values. Default is 0 corresponding to output for both SBP & DBP. For both SBP and DBP ARV values use bp_type = 0, for SBP-only use bp_type = 1, and for DBP-only use bp_type = 2 |
add_groups |
Optional argument. Allows the user to aggregate the data by an
additional "group" to further refine the output. The supplied input must be a
character vector with the strings corresponding to existing column names of the
processed |
inc_wake |
Optional argument corresponding to whether or not to include |
Calculate the coefficient of variation at various levels of granularity based on what is supplied (ID, VISIT, WAKE, and / or DATE) for either SBP, DBP, or both. CV is a measure of dispersion
A tibble object with a row corresponding to each subject, or alternatively a row corresponding to each date if inc_date = TRUE. The resulting tibble consists of:
ID: The unique identifier of the subject. For single-subject datasets, ID = 1
VISIT: (If applicable) Corresponds to the visit # of the subject, if more than 1
WAKE: (If applicable) Corresponds to the awake status of the subject (0 = asleep |
1 = awake)
CV_SBP / CV_DBP: Calculates the ratio of standard deviation to the mean. CV_SBP
or CV_DBP is useful for comparing the degree of variation from one data series
to another.
SD_SBP / SD_DBP: For completeness, the cv function also includes the
standard deviation as a comparison metric to measure spread around the average.
N: The number of observations for that particular grouping. If inc_date = TRUE,
N corresponds to the number of observations for that date. If inc_date = FALSE, N
corresponds to the number of observations for the most granular grouping available (i.e.
a combination of ID, VISIT, and WAKE)
Any add_groups variables supplied to function argument will be present as a column in the resulting tibble.
# Load data data(bp_hypnos) data(bp_jhs) # Process bp_hypnos hypnos_proc <- process_data(bp_hypnos, sbp = "SYST", dbp = "DIAST", date_time = "date.time", id = "id", wake = "wake", visit = "visit", hr = "hr", pp ="pp", map = "map", rpp = "rpp") # Process bp_jhs data jhs_proc <- process_data(bp_jhs, sbp = "Sys.mmHg.", dbp = "Dias.mmHg.", date_time = "DateTime", hr = "Pulse.bpm.") ## Not run: # CV Calculation bp_cv(hypnos_proc, inc_date = TRUE, add_groups = "SBP_Category") bp_cv(jhs_proc, add_groups = c("meal_time")) # Notice that meal_time is not a column from process_data, but it still works ## End(Not run)# Load data data(bp_hypnos) data(bp_jhs) # Process bp_hypnos hypnos_proc <- process_data(bp_hypnos, sbp = "SYST", dbp = "DIAST", date_time = "date.time", id = "id", wake = "wake", visit = "visit", hr = "hr", pp ="pp", map = "map", rpp = "rpp") # Process bp_jhs data jhs_proc <- process_data(bp_jhs, sbp = "Sys.mmHg.", dbp = "Dias.mmHg.", date_time = "DateTime", hr = "Pulse.bpm.") ## Not run: # CV Calculation bp_cv(hypnos_proc, inc_date = TRUE, add_groups = "SBP_Category") bp_cv(jhs_proc, add_groups = c("meal_time")) # Notice that meal_time is not a column from process_data, but it still works ## End(Not run)
Calculate the percent and average decline (or potentially reverse) in nocturnal blood pressure.
This function is typically used with ABPM data, or at minimum, data with a corresponding a
WAKE column available to indicate awake vs asleep.
Nocturnal blood pressure decline is an indicator of an individual's natural circadian rhythm. Studies have shown that individuals with diminished circadian rhythms are more likely to exhibit target organ damage. There is a "U-shaped" relationship that exists among the magnitude of nocturnal blood pressure decline; the extreme dippers and the non-dippers (including reverse dippers) are both more prone to mortality risk than normal dippers.
NOTE: Any reference to "sleep" in the bp package refers to an individual's nocturnal period; "sleep" is used in an informal sense for intuitive purposes. Technically, from a clinical perspective, indication of sleep is currently obtained through means of EEG or other highly specialized equipment. For all intents and purposes, sleep in the context of this package refers to actigraphy-inferred nocturnal periods of rest.
dip_calc( data, sleep_start_end = NULL, dip_thresh = 0.1, extreme_thresh = 0.2, inc_date = FALSE, subj = NULL )dip_calc( data, sleep_start_end = NULL, dip_thresh = 0.1, extreme_thresh = 0.2, inc_date = FALSE, subj = NULL )
data |
User-supplied data set that must contain In the event of non-ABPM data (i.e. a data set without a corresponding |
sleep_start_end |
(optional) User-supplied manual override to adjust sleep interval indicating indicate start and end time corresponding to the sleep interval of interest. Must only contain 2 values and must be 24-hour denoted integers Example: NOTE: If the Furthermore, the |
dip_thresh |
Default threshold for normal "Dipping" set to 0.10 (i.e. 10%). This value represents the maximum percentage that BP can fall during sleep and be characterized as "Normal" nocturnal decline (dipping). Specifically, this category includes all dips between 0% and this value. |
extreme_thresh |
Default threshold for "Extreme Dipping" set to 0.20 (i.e. 20%). This value represents the maximum percentage that BP can fall during sleep and be characterized as "Extreme" nocturnal decline (dipping). Specifically, this category includes all dips between the Normal dipping threshold and this value. NOTE: dip_thresh cannot exceed extreme_thresh. |
inc_date |
Default to FALSE. Indicates whether or not to include the date in the grouping of the final output |
subj |
Optional argument. Allows the user to specify and subset specific subjects
from the |
A list containing 2 tibble objects. The first tibble object lists grouped average values for SBP and DBP
for awake and asleep periods. The second dip_pct tibble object lists the dipping percentage and
classification according to the results from the first dip tibble. If inc_date = TRUE these two
tibbles will be broken down further by date. There are 4 classifications a subject can have (assuming a
default dipping threshold of 10% and extreme dipping threshold of 20% according to the original source):
Reverse Dipper - no nocturnal decline (greater or equal to 0%)
Non-Dipper - a nocturnal decline between 0 - 10%
Dipper - a nocturnal decline between 10% and the extreme dipping % (20%)
Extreme Dipper - a nocturnal decline exceeding 20%
Okhubo, T., Imai, Y., Tsuji, K., Nagai, K., Watanabe, N., Minami, J., Kato, J., Kikuchi, N., Nishiyama, A., Aihara, A., Sekino, M., Satoh, H., and Hisamichi, S. (1997). Relation Between Nocturnal Decline in Blood Pressure and Mortality: The Ohasama Study, American Journal of Hypertension 10(11), 1201–1207, doi:10.1016/S0895-7061(97)00274-4.
## Load bp_hypnos data(bp_hypnos) ## Process bp_hypnos hypnos_proc <- process_data(bp_hypnos, sbp = 'syst', dbp = 'diast', date_time = 'date.time', hr = 'hr', pp = 'PP', map = 'MaP', rpp = 'Rpp', id = 'id', visit = 'Visit', wake = 'wake') dip_calc(hypnos_proc)## Load bp_hypnos data(bp_hypnos) ## Process bp_hypnos hypnos_proc <- process_data(bp_hypnos, sbp = 'syst', dbp = 'diast', date_time = 'date.time', hr = 'hr', pp = 'PP', map = 'MaP', rpp = 'Rpp', id = 'id', visit = 'Visit', wake = 'wake') dip_calc(hypnos_proc)
Graphical visualization of the dip_calc output for categories of dipping percentages.
NOTE: Any reference to "sleep" in the bp package refers to an individual's nocturnal period; "sleep" is used in an informal sense for intuitive purposes. Technically, from a clinical perspective, indication of sleep is currently obtained through means of EEG or other highly specialized equipment. For all intents and purposes, sleep in the context of this package refers to actigraphy-inferred nocturnal periods of rest.
dip_class_plot( data, subj = NULL, dip_thresh = 0.1, extreme_thresh = 0.2, thresh_mult = 2, sleep_start_end = NULL )dip_class_plot( data, subj = NULL, dip_thresh = 0.1, extreme_thresh = 0.2, thresh_mult = 2, sleep_start_end = NULL )
data |
User-supplied data set that must contain In the event of non-ABPM data (i.e. a data set without a corresponding |
subj |
Optional argument. Allows the user to specify and subset specific subjects
from the |
dip_thresh |
Default threshold for normal "Dipping" set to 0.10 (i.e. 10%). This value represents the maximum percentage that BP can fall during sleep and be characterized as "Normal" nocturnal decline (dipping). Specifically, this category includes all dips between 0% and this value. |
extreme_thresh |
Default threshold for "Extreme Dipping" set to 0.20 (i.e. 20%). This value represents the maximum percentage that BP can fall during sleep and be characterized as "Extreme" nocturnal decline (dipping). Specifically, this category includes all dips between the Normal dipping threshold and this value. NOTE: dip_thresh cannot exceed extreme_thresh. |
thresh_mult |
Optional argument that serves as a multiplier by which to expand plot sizing for X and Y axis. Default set to 2. |
sleep_start_end |
Optional User-supplied manual override to adjust sleep interval indicating indicate start and end time corresponding to the sleep interval of interest. Must only contain 2 values and must be 24-hour denoted integers Example: |
A scatter plot of all dipping percentage values layered on top of the category plot outlined in Okhubo et al. (1995). dip_thresh and extreme_thresh denote the cutoffs for the Normal and Extreme dipping categories. Any dips below zero are denoted as Inverted (or Reverse) dipping.
The default plot categories are as follows:
INV: Inverted (Reverse) Dipper - no nocturnal decline (greater or equal to 0%)
ND: Non-Dipper - a nocturnal decline between 0 - 10%
DIP: Dipper - a nocturnal decline between 10% and the extreme dipping % (20%)
ED: Extreme Dipper - a nocturnal decline exceeding 20%
Okhubo, T., Imai, Y., Tsuji, K., Nagai, K., Watanabe, N., Minami, J., Kato, J., Kikuchi, N., Nishiyama, A., Aihara, A., Sekino, M., Satoh, H., and Hisamichi, S. (1997). Relation Between Nocturnal Decline in Blood Pressure and Mortality: The Ohasama Study, American Journal of Hypertension 10(11), 1201–1207, doi:10.1016/S0895-7061(97)00274-4.
hypnos_proc <- process_data(bp::bp_hypnos, sbp = "syst", dbp = "DIAST", date_time = "date.time", id = "id", wake = "wake", visit = "visit") dip_class_plot(hypnos_proc)hypnos_proc <- process_data(bp::bp_hypnos, sbp = "syst", dbp = "DIAST", date_time = "date.time", id = "id", wake = "wake", visit = "visit") dip_class_plot(hypnos_proc)
Blood Pressure Tables in Graphical Format
dow_tod_plots(data, subj = NULL)dow_tod_plots(data, subj = NULL)
data |
A processed dataframe resulting from the |
subj |
Optional argument. Allows the user to specify and subset specific subjects
from the |
A list of four 'gtables' that correspond to the tables for Day of Week and
Time of Day broken down by both SBP and DBP.
data("bp_hypnos") hyp_proc <- process_data(bp_hypnos, bp_type = 'ABPM', sbp = "syst", dbp = "DIAST", date_time = "date.time", id = "id", wake = "wake", visit = "visit", hr = "hr", map = "map", rpp = "rpp", pp = "pp", ToD_int = c(5, 13, 18, 23)) rm(bp_hypnos) dow_tod_plots_out <- dow_tod_plots(hyp_proc) grid::grid.draw( gridExtra::grid.arrange(dow_tod_plots_out[[1]], dow_tod_plots_out[[2]], ncol = 2) )data("bp_hypnos") hyp_proc <- process_data(bp_hypnos, bp_type = 'ABPM', sbp = "syst", dbp = "DIAST", date_time = "date.time", id = "id", wake = "wake", visit = "visit", hr = "hr", map = "map", rpp = "rpp", pp = "pp", ToD_int = c(5, 13, 18, 23)) rm(bp_hypnos) dow_tod_plots_out <- dow_tod_plots(hyp_proc) grid::grid.draw( gridExtra::grid.arrange(dow_tod_plots_out[[1]], dow_tod_plots_out[[2]], ncol = 2) )
A helper function to assist in pre-processing the user-supplied
input data in a standardized format for use with other functions in the bp package.
See Vignette for further details.
process_data( data, bp_type = c("hbpm", "abpm", "ap"), guidelines = c("Lee_2020", "AHA", "Custom"), bp_cutoffs = list(c(100, 120, 130, 140, 180), c(60, 80, 80, 90, 120)), ap = NULL, time_elap = NULL, sbp = NULL, dbp = NULL, date_time = NULL, id = NULL, group = NULL, wake = NULL, visit = NULL, hr = NULL, pp = NULL, map = NULL, rpp = NULL, DoW = NULL, ToD_int = NULL, eod = NULL, data_screen = TRUE, SUL = 240, SLL = 50, DUL = 140, DLL = 40, HRUL = 220, HRLL = 27, inc_low = TRUE, inc_crisis = TRUE, agg = FALSE, agg_thresh = 3, collapse_df = FALSE, dt_fmt = "ymd HMS", chron_order = FALSE, tz = "UTC" )process_data( data, bp_type = c("hbpm", "abpm", "ap"), guidelines = c("Lee_2020", "AHA", "Custom"), bp_cutoffs = list(c(100, 120, 130, 140, 180), c(60, 80, 80, 90, 120)), ap = NULL, time_elap = NULL, sbp = NULL, dbp = NULL, date_time = NULL, id = NULL, group = NULL, wake = NULL, visit = NULL, hr = NULL, pp = NULL, map = NULL, rpp = NULL, DoW = NULL, ToD_int = NULL, eod = NULL, data_screen = TRUE, SUL = 240, SLL = 50, DUL = 140, DLL = 40, HRUL = 220, HRLL = 27, inc_low = TRUE, inc_crisis = TRUE, agg = FALSE, agg_thresh = 3, collapse_df = FALSE, dt_fmt = "ymd HMS", chron_order = FALSE, tz = "UTC" )
data |
User-supplied dataset containing blood pressure data. Must contain data for Systolic blood pressure and Diastolic blood pressure at a minimum. |
bp_type |
Required argument specifying which of the three BP data types
("HBPM", "ABPM", or "AP") the input data is. Default HBPM - Home Blood Pressure Monitor | ABPM - Ambulatory Blood Pressure | AP - Arterial Pressure NOTE: |
guidelines |
A string designation for the guidelines to follow when mapping BP
readings to a respective BP stage. |
bp_cutoffs |
A list containing two vectors corresponding to SBP and DBP cutoffs, respectively. Each vector contains 5 values. The SBP vector (100, 120, 130, 140, 180) corresponds to the upper limits for the following stages: Low (0-100), Normal (100-120), Elevated (120-130), Stage 1 Hypertension (130-140), Stage 2 Hypertension (140-180). When utilizing Lee et al (2020) guidelines, additional stages are included: Isolated Systolic Hypertension for Stage 1 (ISH - S1) (130-140), Isolated Diastolic Hypertension for Stage 1 (IDH - S1) (0-130), ISH - S2 (140-180), and IDH - S2 (0-140). The DBP vector (60, 80, 80, 90, 120) corresponds to the upper limits for the following stages: Low (0-60), Normal (60-80), Elevated (0-80), Stage 1 Hypertension (80-90), Stage 2 Hypertension (90-120). The upper limit of the "Elevated" category repeats in the DBP vector and matches that of Normal. This because according to most guidelines, there is no distinction between DBP cutoffs for Normal and Elevated - these stages are discerned by SBP, not DBP. When utilizing Lee et al (2020) guidelines, additional stages are included: Isolated Diastolic Hypertension for Stage 1 (ISH - S1) (0-80), Isolated Diastolic Hypertension for Stage 1 (IDH - S1) (80-90), ISH - S2 (0-90), and IDH - S2 (90-120). Any SBP reading below 100 or DBP reading below 60 is considered Hypotension ("Low"). Any SBP reading above 180 or DBP reading above 120 is considered a Crisis. When If |
ap |
(For AP data only) Required column name (character string) corresponding to continuous Arterial Pressure (AP) (mmHg). Note that this is a required argument so long as bp_type = "AP". Ensure that bp_type is set accordingly. |
time_elap |
(For AP data only) Column name corresponding to the time elapsed for the given AP waveform data. |
sbp |
Required column name (character string) corresponding to Systolic Blood Pressure (mmHg) |
dbp |
Required column name (character string) corresponding to Diastolic Blood Pressure (mmHg) |
date_time |
Optional column name (character string) corresponding to Date/Time, but HIGHLY recommended to supply if available. For DATE-only columns (with no associated time), leave date_time = NULL. DATE-only adjustments are automatic. Dates can be automatically calculated off DATE_TIME column provided that it is called "DATE_TIME" exactly. |
id |
Optional column name (character string) corresponding to subject ID. Typically needed for data corresponding to more than one subject. For one-subject datasets, ID will default to 1 (if ID column not found in dataset) |
group |
Optional column name (character string) corresponding to an additional grouping variable that can be used to further break down data. NOTE that this simply sets the column as "GROUP" so that other functions recognize which column to use as the grouping variable. |
wake |
Optional column name (character string) corresponding to sleep status. A WAKE value of 1 indicates that the subject is awake and 0 implies asleep. |
visit |
Optional column name (character string) corresponding to Visit number |
hr |
Optional column name (character string) corresponding to Heart Rate (bpm) |
pp |
Optional column name (character string) corresponding to Pulse Pressure (SBP - DBP). If not supplied, it will be calculated automatically. |
map |
Optional column name (character string) corresponding to Mean Arterial Pressure |
rpp |
Optional column name (character string) corresponding to Rate Pulse Pressure (SBP * HR). If not supplied, but HR column available, then RPP will be calculated automatically. |
DoW |
Optional column name (character string) corresponding to the Day of the Week.
If not supplied, but DATE or DATE_TIME columns available, then DoW will be created
automatically. DoW values must be abbreviated as such |
ToD_int |
Optional vector of length 4, acceptable values are from 0 to 23 in a an order corresponding to hour for Morning, Afternoon, Evening, Night). This vector allows to override the default interval for the Time-of-Day periods: if NULL, the Morning, Afternoon, Evening, and Night periods are set at 6, 12, 18, 0 respectively, where 0 corresponds to the 24th hour of the day (i.e. Midnight). For example, ToD_int = c(5, 13, 18, 23) would correspond to a period for Morning starting at 5:00 (until 13:00), Afternoon starting at 13:00 (until 18:00), Evening starting at 18:00 (until 23:00), and Night starting at 23:00 (until 5:00) |
eod |
Optional argument to adjust the delineation for the end of day (eod). The supplied value should be a character string with 4 characters representing the digits of 24-hour time, e.g. "1310" corresponds to 1:10pm. For individuals who
do not go to bed early or work night-shifts, this argument adjusts the |
data_screen |
Optional logical argument; default set to TRUE. Screens for extreme values in the data
for both |
SUL |
Systolic Upper Limit (SUL). If |
SLL |
Systolic Lower Limit (SLL). If |
DUL |
Diastolic Upper Limit (DUL). If |
DLL |
Diastolic Lower Limit (DLL). If |
HRUL |
Heart Rate Upper Limit (HRUL). If see https://www.cdc.gov/physicalactivity/basics/measuring/heartrate.htm |
HRLL |
Heart Rate Upper Limit (HRUL). If |
inc_low |
Optional logical argument dictating whether or not to include the "Low" category for BP classification column (and the supplementary SBP/DBP Category columns). Default set to TRUE. |
inc_crisis |
Optional logical argument dictating whether or not to include the "Crisis" category for BP classification column (and the supplementary SBP/DBP Category columns). Default set to TRUE. |
agg |
Optional argument specifying whether or not to aggregate the data based on the amount of time
between observations. If |
agg_thresh |
Optional argument specifying the threshold of how many minutes can pass between readings (observations) and still be considered part of the same sitting. The default is set to 3 minutes. This implies that if two or more readings are within 3 minutes of each other, they will be averaged together (if agg is set to TRUE). |
collapse_df |
Optional argument that collapses the dataframe to eliminate repeating rows after aggregation. |
dt_fmt |
Optional argument that specifies the input date/time format (dt_fmt). Default set to "ymd HMS" but can take on any format specified by the lubridate package. |
chron_order |
Optional argument that specifies whether to order the data in chronological (Oldest dates & times at the top / first) or reverse chronological order (Most recent dates & times at the top / first). TRUE refers to chronological order; FALSE refers to reverse chronological order. The default is set to FALSE (i.e. most recent observations listed first in the dataframe). See https://lubridate.tidyverse.org/reference/parse_date_time.html for more details. |
tz |
Optional argument denoting the respective time zone. Default time zone set to "UTC". See
Use |
A processed dataframe object with standardized column names and formats to use with the rest of bp package functions. The following standardized column names are used throughout
BP_TYPE |
One of AP, HBPM or ABPM |
ID |
Subject ID |
SBP |
Systolic Blood Pressure |
DBP |
Diastolic Blood Pressure |
SBP_CATEGORY |
Ordinal, SBP characterization into "Low" < "Normal"<"Elevated"<"Stage 1"< "Stage 2" < "Crisis". "Low" is not included if |
DBP_CATEGORY |
Ordinal, DBP characterization into "Low" < "Normal"<"Elevated"<"Stage 1"< "Stage 2" < "Crisis". "Low" is not included if |
BP_CLASS |
Blood pressure categorization based on paired values (SBP, DBP) into one of the 8 stages according to Lee et al. 2020. See |
HR |
Heart Rate |
MAP |
Mean Arterial Pressure |
PP |
Pulse Pressure, SBP-DBP |
DATE_TIME |
Date and time in POSIXct format |
DATE |
Date only in Date format |
MONTH |
Month, integer from 1 to 12 |
DAY |
Day, integer from 1 to 31 |
YEAR |
Four digit year |
DAY_OF_WEEK |
Ordinal, with "Sun"<"Mon"<"Tue"<"Wed"<"Thu"<"Fri"<"Sat" |
TIME |
Time in character format |
HOUR |
Integer, from 0 to 23 |
TIME_OF_DAY |
One of "Morning", "Afternoon", "Evening" or "Night" |
Lee H, Yano Y, Cho SMJ, Park JH, Park S, Lloyd-Jones DM, Kim HC. Cardiovascular risk of isolated systolic or diastolic hypertension in young adults. Circulation. 2020; 141:1778–1786. doi:10.1161/CIRCULATIONAHA.119.044838
Muntner, P., Carey, R. M., Jamerson, K., Wright Jr, J. T., & Whelton, P. K. (2019). Rationale for ambulatory and home blood pressure monitoring thresholds in the 2017 American College of Cardiology/American Heart Association Guideline. Hypertension, 73(1), 33-38. doi:10.1161/HYPERTENSIONAHA.118.11946
Omboni, S., Parati, G*., Zanchetti, A., Mancia, G. Calculation of trough: peak ratio of antihypertensive treatment from ambulatory blood pressure: methodological aspects Journal of Hypertension. October 1995 - Volume 13 - Issue 10 - p 1105-1112 doi:10.1097/00004872-199510000-00005
Unger, T., Borghi, C., Charchar, F., Khan, N. A., Poulter, N. R., Prabhakaran, D., ... & Schutte, A. E. (2020). 2020 International Society of Hypertension global hypertension practice guidelines. Hypertension, 75(6), 1334-1357. doi:10.1161/HYPERTENSIONAHA.120.15026
# Load bp_hypnos data("bp_hypnos") # Process data for bp_hypnos hypnos_proc <- process_data(bp_hypnos, bp_type = 'abpm', sbp = 'syst', dbp = 'diast', date_time = 'date.time', hr = 'hr', pp = 'PP', map = 'MaP', rpp = 'Rpp', id = 'id', visit = 'Visit', wake = 'wake', data_screen = FALSE) hypnos_proc # Load bp_jhs data data("bp_jhs") # Process data for bp_jhs # Note that bp_type defaults to "hbpm" and is therefore not specified jhs_proc <- process_data(bp_jhs, sbp = "Sys.mmHg.", dbp = "Dias.mmHg.", date_time = "DateTime", hr = "Pulse.bpm.") jhs_proc# Load bp_hypnos data("bp_hypnos") # Process data for bp_hypnos hypnos_proc <- process_data(bp_hypnos, bp_type = 'abpm', sbp = 'syst', dbp = 'diast', date_time = 'date.time', hr = 'hr', pp = 'PP', map = 'MaP', rpp = 'Rpp', id = 'id', visit = 'Visit', wake = 'wake', data_screen = FALSE) hypnos_proc # Load bp_jhs data data("bp_jhs") # Process data for bp_jhs # Note that bp_type defaults to "hbpm" and is therefore not specified jhs_proc <- process_data(bp_jhs, sbp = "Sys.mmHg.", dbp = "Dias.mmHg.", date_time = "DateTime", hr = "Pulse.bpm.") jhs_proc
Adjusts WAKE column to reflect start and end of user-defined sleep period. If sleep_start_end is NULL and WAKE column is included in input data, this function will not alter data.
sleep_int(data, sleep_start_end = NULL, adj = TRUE)sleep_int(data, sleep_start_end = NULL, adj = TRUE)
data |
Supplied data from function |
sleep_start_end |
User-supplied sleep interval to indicate start and end time of the sleep interval of interest. Must only contain 2 values and must be 24-hour denoted integers |
adj |
Logical indicating whether or not to perform necessary adjustments / additions to data for WAKE column (i.e. set the default sleep time from 11PM - 6 AM if no interval specified and a DATE_TIME column is available). Default set to TRUE. adj == FALSE essentially acts as a pass statement |
Dataframe with added / adjusted WAKE column moved after DATE_TIME column containing binary 1 (awake) and 0 (asleep)
THIS IS A DEPRECATED FUNCTION. USE bp_sv INSTEAD.
sv( data, inc_date = FALSE, subj = NULL, bp_type = 0, add_groups = NULL, inc_wake = TRUE )sv( data, inc_date = FALSE, subj = NULL, bp_type = 0, add_groups = NULL, inc_wake = TRUE )
data |
Required argument. Pre-processed dataframe with SBP and DBP columns
with optional ID, VISIT, WAKE, and DATE columns if available.
Use |
inc_date |
Optional argument. Default is FALSE. As ABPM data typically
overlaps due to falling asleep on one date and waking up on another, the |
subj |
Optional argument. Allows the user to specify and subset specific subjects
from the |
bp_type |
Optional argument. Determines whether to calculate ARV for SBP values or DBP values. Default is 0 corresponding to output for both SBP & DBP. For both SBP and DBP ARV values use bp_type = 0, for SBP-only use bp_type = 1, and for DBP-only use bp_type = 2 |
add_groups |
Optional argument. Allows the user to aggregate the data by an
additional "group" to further refine the output. The supplied input must be a
character vector with the strings corresponding to existing column names of the
processed |
inc_wake |
Optional argument corresponding to whether or not to include |
Calculate the successive variation (SV) at various levels of granularity based on what is supplied (ID, VISIT, WAKE, and / or DATE)for either SBP, DBP, or both. SV is a measure of dispersion that takes into account the temporal structure of the data and relies on the sum of squared differences in successive observations, unlike the average real variability (ARV) which relies on the sum of absolute differences. $$SV = sqrt(sum(x_i+1 - x_i)^2/n-1)$$
NOTE: The canonical standard deviation, independent of the temporal structure using the sample average, is added for comparison: $$SD = sqrt(sum(x_i+1 - xbar)^2/n-1)$$
A tibble object with a row corresponding to each subject, or alternatively a row corresponding to each date if inc_date = TRUE. The resulting tibble consists of:
ID: The unique identifier of the subject. For single-subject datasets, ID = 1
VISIT: (If applicable) Corresponds to the visit # of the subject, if more than 1
WAKE: (If applicable) Corresponds to the awake status of the subject (0 = asleep |
1 = awake)
SV_SBP / SV_DBP: Calculates the square root of the average squared differences
between successive measurements. The resulting value averages across the granularity
grouping for however many observations are present.
N: The number of observations for that particular grouping. If inc_date = TRUE,
N corresponds to the number of observations for that date. If inc_date = FALSE, N
corresponds to the number of observations for the most granular grouping available (i.e.
a combination of ID, VISIT, and WAKE)
Any add_groups variables supplied to function argument will be present as a column in the resulting tibble.
# Load data data(bp_hypnos) data(bp_jhs) # Process bp_hypnos hypnos_proc <- process_data(bp_hypnos, sbp = "SYST", dbp = "DIAST", date_time = "date.time", id = "id", wake = "wake", visit = "visit", hr = "hr", pp ="pp", map = "map", rpp = "rpp") # Process bp_jhs data jhs_proc <- process_data(bp_jhs, sbp = "Sys.mmHg.", dbp = "Dias.mmHg.", date_time = "DateTime", hr = "Pulse.bpm.") # SV Calculation ## Not run: bp_sv(hypnos_proc) bp_sv(jhs_proc, add_groups = c("meal_time")) # Notice that meal_time is not a column from process_data, but it still works ## End(Not run)# Load data data(bp_hypnos) data(bp_jhs) # Process bp_hypnos hypnos_proc <- process_data(bp_hypnos, sbp = "SYST", dbp = "DIAST", date_time = "date.time", id = "id", wake = "wake", visit = "visit", hr = "hr", pp ="pp", map = "map", rpp = "rpp") # Process bp_jhs data jhs_proc <- process_data(bp_jhs, sbp = "Sys.mmHg.", dbp = "Dias.mmHg.", date_time = "DateTime", hr = "Pulse.bpm.") # SV Calculation ## Not run: bp_sv(hypnos_proc) bp_sv(jhs_proc, add_groups = c("meal_time")) # Notice that meal_time is not a column from process_data, but it still works ## End(Not run)