3 Using the Package in R

3.1 Installation

If you do not have R installed on your computer, please go to Appendix 1. Once you have R installed, you can install this package in the R console:

install.packages("devtools")
devtools::install_github("resplab/accept")

3.2 ACCEPT Prediction Functions

Sample Data

To get started, there is an R data frame with the package of sample patient data. I have printed columns 1-13 and 14-19 separately because there isn’t enough space:

library(accept)
samplePatients = accept::samplePatients
print(samplePatients[,1:13])
> # A tibble: 2 x 13
>   ID     male   age smoker oxygen statin  LAMA  LABA   ICS  FEV1   BMI
>   <chr> <dbl> <dbl>  <dbl>  <dbl>  <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
> 1 10001     1    70      1      1      1     1     1     1    33    25
> 2 10002     0    42      0      1      0     1     1     0    40    35
> # … with 2 more variables: SGRQ <dbl>, LastYrExacCount <dbl>
print(samplePatients[,14:19])
> # A tibble: 2 x 6
>   LastYrSevExacCo… randomized_azit… randomized_stat… randomized_LAMA
>              <dbl>            <dbl>            <dbl>           <dbl>
> 1                1                0                0               0
> 2                0                0                0               0
> # … with 2 more variables: randomized_LABA <dbl>, randomized_ICS <dbl>

ID male age smoker oxygen statin LAMA LABA ICS FEV1

BMI SGRQ LastYrExacCount LastYrSevExacCount randomized_azithromycin randomized_statin randomized_LAMA randomized_LABA randomized_ICS

Patient ID. This is useful if you want to look at many patients at once. If you don't have an ID, you can just make up a number here.

Gender of the patient. 1 = male, 0 = female

Patient age at the beginning of the study.

Whether or not a patient currently smokes. 1 = smoker, 0 = non-smoker

Domiciliary Oxygen Therapy: At-home oxygen therapy, used to treat COPD. This metric measures whether or not a patient uses oxygen therapy, 1 = Yes, 0 = No.

Statins: generally used to lower cholesteral in blood, but some studies have recently shown they can be effective in COPD treatment. This metric asks whether or not a patient has used statins in the past year.

Long-Acting Muscarinic receptor Antagonists: used for treatment of COPD. This metric asks whether or not a patient has used LAMAs in the past year.

Long-Acting β2 Agonists: used for treatment of COPD. This metric asks whether or not a patient has used LABAs in the past year.

Inhaled Glucocorticoids: used for treatment of COPD. This metric asks whether or not a patient has used inhaled glucocorticoids in the past year.

Prediction % FEV1 Score.

Body Mass Index: ratio of weight (kg) to height squared (m^2)

St. George's Respiratory Questionnaire: This questionnaire is to help doctors/patients determine how much COPD is affecting day-to-day life. The score from the questionnaire is correlated to cough, sputum, and wheezing symptoms. Questionnaire Download
SGRQ Overview

Number of exacerbations (severe and mild/moderate) in the last year.

Number of severe exacerbations in the last year.

Whether the patient was randomized to receive azithromycin. Should be 0 for prediction (0,1)

Whether the patient was randomized to receive statins. Should be 0 for prediction (0,1)

Whether the patient was randomized to receive LAMAs. Should be 0 for prediction (0,1)

Whether the patient was randomized to receive LABAs. Should be 0 for prediction (0,1)

Whether the patient was randomized to receive ICs. Should be 0 for prediction (0,1)

Exacerbation Rate

To get a prediction for exacerbation rate, you will need to pass in a patient vector:

results <- predictACCEPT(samplePatients[1,])
print(t(results))
>                                                         [,1]        
> ID                                                      "10001"     
> male                                                    "1"         
> age                                                     "70"        
> smoker                                                  "1"         
> oxygen                                                  "1"         
> statin                                                  "1"         
> LAMA                                                    "1"         
> LABA                                                    "1"         
> ICS                                                     "1"         
> FEV1                                                    "33"        
> BMI                                                     "25"        
> SGRQ                                                    "50"        
> LastYrExacCount                                         "2"         
> LastYrSevExacCount                                      "1"         
> randomized_azithromycin                                 "0"         
> randomized_statin                                       "0"         
> randomized_LAMA                                         "0"         
> randomized_LABA                                         "0"         
> randomized_ICS                                          "0"         
> predicted_exac_probability                              "0.7799"    
> predicted_exac_rate                                     "1.813428"  
> azithromycin_predicted_exac_probability                 "0.703674"  
> azithromycin_predicted_exac_rate                        "1.406447"  
> predicted_severe_exac_probability                       "0.4394301" 
> predicted_severe_exac_rate                              "0.7688124" 
> azithromycin_predicted_severe_exac_probability          "0.425913"  
> azithromycin_predicted_severe_exac_rate                 "0.577342"  
> predicted_exac_probability_lower_PI                     "0.4425521" 
> predicted_exac_probability_upper_PI                     "0.9806575" 
> predicted_exac_rate_lower_PI                            "0.5843924" 
> predicted_exac_rate_upper_PI                            "3.945464"  
> azithromycin_predicted_exac_probability_lower_PI        "0.3644317" 
> azithromycin_predicted_exac_probability_upper_PI        "0.9531118" 
> azithromycin_predicted_exac_rate_lower_PI               "0.4532394" 
> azithromycin_predicted_exac_rate_upper_PI               "3.059998"  
> predicted_severe_exac_probability_lower_PI              "0.06018855"
> predicted_severe_exac_probability_upper_PI              "0.904737"  
> predicted_severe_exac_rate_lower_PI                     "0.07760276"
> predicted_severe_exac_rate_upper_PI                     "2.254686"  
> azithromycin_predicted_severe_exac_probability_lower_PI "0.05621219"
> azithromycin_predicted_severe_exac_probability_upper_PI "0.898296"  
> azithromycin_predicted_severe_exac_rate_lower_PI        "0.05628177"
> azithromycin_predicted_severe_exac_rate_upper_PI        "1.701869"  
> predicted_exac_probability_lower_CI                     "0.7705132" 
> predicted_exac_probability_upper_CI                     "0.7892394" 
> predicted_exac_rate_lower_CI                            "1.760126"  
> predicted_exac_rate_upper_CI                            "1.868583"  
> azithromycin_predicted_exac_probability_lower_CI        "0.6931493" 
> azithromycin_predicted_exac_probability_upper_CI        "0.7142844" 
> azithromycin_predicted_exac_rate_lower_CI               "1.365329"  
> azithromycin_predicted_exac_rate_upper_CI               "1.447842"  
> predicted_severe_exac_probability_lower_CI              "0.4235962" 
> predicted_severe_exac_probability_upper_CI              "0.4542333" 
> predicted_severe_exac_rate_lower_CI                     "0.7323719" 
> predicted_severe_exac_rate_upper_CI                     "0.8092376" 
> azithromycin_predicted_severe_exac_probability_lower_CI "0.4116278" 
> azithromycin_predicted_severe_exac_probability_upper_CI "0.4400719" 
> azithromycin_predicted_severe_exac_rate_lower_CI        "0.5519708" 
> azithromycin_predicted_severe_exac_rate_upper_CI        "0.6030546"

The predictACCEPT() function returns a data frame with the original patient data, along with the predictions for different treatment options. Here is a summary of the results:

Azithromycin No Treatment

Overall Severe

Probability Rate

Confidence Interval Prediction Interval

azithromycin_predicted_

Prediction if the patient is treated with 250 mg / day of azithromycin

predicted_

Prediction if patient is not treated with azithromycin

exac_

All exacerbations (mild, moderate, severe)

severe_exac_

Severe exacerbations only

probability

Probability of at least one exacerbation next year

rate

Number of exacerbations / year

_upper_CI
_lower_CI

95 % confidence interval

_upper_PI
_lower_PI

Prediction Interval

To visualize the data, there is a graphing function called plotExacerbations(), which creates a Plotly bar graph. You have the option of selecting probability or rate for which prediction you want to see, and either CI or PI to select the confidence interval or prediction interval respectively.

plotExacerbations(results, type="probability", interval = "CI")

plotExacerbations(results, type="probability", interval = "PI")

plotExacerbations(results, type="rate", interval = "CI")

Probability of N Exacerbations (Poisson)

We can also calculate the predicted number of exacerbations in a year:

patientResults = predictACCEPT(samplePatients[1,])
exacerbationsMatrix = predictCountProb(patientResults, n = 10, shortened = TRUE)
print(exacerbationsMatrix)
>                         none severe   1 severe   2 severe 3 or more severe
> no exacerbations         0.16312494 0.00000000 0.00000000       0.00000000
> 1 exacerbation           0.16160772 0.13417676 0.00000000       0.00000000
> 2 exacerbations          0.08005231 0.13292879 0.05518286       0.00000000
> 3 or more exacerbations  0.03452944 0.09424771 0.09342959       0.04074374

The shortened parameter groups the probabilities from 3-10 exacerbations into one category, “3 or more exacerbations.” To see all n exacerbation probabilities:

exacerbationsMatrix = predictCountProb(patientResults, n = 10, shortened = FALSE)
print(exacerbationsMatrix)
>                          0 severe       1 severe      2 severe
> 0 exacerbation(s) 0.1631249390163 0.000000000000 0.00000000000
> 1 exacerbation(s) 0.1616077243627 0.134176756503 0.00000000000
> 2 exacerbation(s) 0.0800523106129 0.132928787049 0.05518286197
> 3 exacerbation(s) 0.0264359163925 0.065846212439 0.05466960969
> 4 exacerbation(s) 0.0065475094051 0.021744593672 0.02708056557
> 5 exacerbation(s) 0.0012973222875 0.005385587148 0.00894289698
> 6 exacerbation(s) 0.0002142093293 0.001067099229 0.00221492992
> 7 exacerbation(s) 0.0000303167117 0.000176195701 0.00043886580
> 8 exacerbation(s) 0.0000037543422 0.000024936702 0.00007246399
> 9 exacerbation(s) 0.0000004132693 0.000003088096 0.00001025572
>                        3 severe      4 severe      5 severe      6 severe
> 0 exacerbation(s) 0.00000000000 0.00000000000 0.00000000000 0.00000000000
> 1 exacerbation(s) 0.00000000000 0.00000000000 0.00000000000 0.00000000000
> 2 exacerbation(s) 0.00000000000 0.00000000000 0.00000000000 0.00000000000
> 3 exacerbation(s) 0.01513003362 0.00000000000 0.00000000000 0.00000000000
> 4 exacerbation(s) 0.01498931014 0.00311126375 0.00000000000 0.00000000000
> 5 exacerbation(s) 0.00742494776 0.00308232608 0.00051182766 0.00000000000
> 6 exacerbation(s) 0.00245196293 0.00152682878 0.00050706718 0.00007016644
> 7 exacerbation(s) 0.00060728935 0.00050420928 0.00025117549 0.00006951382
> 8 exacerbation(s) 0.00012032820 0.00012487992 0.00008294644 0.00003443364
> 9 exacerbation(s) 0.00001986817 0.00002474368 0.00002054374 0.00001137113
>                         7 severe        8 severe     9 severe
> 0 exacerbation(s) 0.000000000000 0.0000000000000 0.000000e+00
> 1 exacerbation(s) 0.000000000000 0.0000000000000 0.000000e+00
> 2 exacerbation(s) 0.000000000000 0.0000000000000 0.000000e+00
> 3 exacerbation(s) 0.000000000000 0.0000000000000 0.000000e+00
> 4 exacerbation(s) 0.000000000000 0.0000000000000 0.000000e+00
> 5 exacerbation(s) 0.000000000000 0.0000000000000 0.000000e+00
> 6 exacerbation(s) 0.000000000000 0.0000000000000 0.000000e+00
> 7 exacerbation(s) 0.000008244955 0.0000000000000 0.000000e+00
> 8 exacerbation(s) 0.000008168270 0.0000008477255 0.000000e+00
> 9 exacerbation(s) 0.000004046149 0.0000008398409 7.747644e-08

To visualize the matrix as a heatmap, we can use the function plotHeatMap:

plotHeatMap(patientResults, shortened = FALSE)