Spur Afrika Mobile Health Clinic 2020 Report

# required libraries

library(airtabler) # database access
library(dplyr)     # data manipulation
library(tidyr)     # 'tidy' data
library(magrittr)  # piping
library(lubridate) # time functions
library(finalfit)  # summary tables
library(lme4)      # regressions
library(jtools)    # pretty regression tables
library(summarytools)# pretty regression tables
library(kableExtra)  # pretty tables
library(highcharter) # charts
library(viridisLite) # colour palettes
library(sjPlot)      # regression table display, requires recent version

# read database

airtable_tables <- list(
  twenty = c("Children"),
  eighteen = c("Children", "Schools", "FindingCodes")
)

airtable <- mapply(function(x, y) airtabler::airtable(x, y), base_key, airtable_tables)
rawdata <- lapply(airtable, function(x) x$Children$select_all())

schools_2018 <- airtable[[2]]$Schools$select_all() %>% select("id", "SchoolName")
findingcodes_2018 <- airtable[[2]]$FindingCodes$select_all() %>% select("id", "Description")

rawdata$eighteen$School <- lapply(
  rawdata[["eighteen"]]$School,
  function(x) {
    unlist(lapply(
      x, function(x) schools_2018[schools_2018$id == x, "SchoolName"]),
      use.names = FALSE)
  }
)
rawdata$eighteen$School[sapply(rawdata$eighteen$School, is.null)] <- NA
rawdata$eighteen$School <- unlist(rawdata$eighteen$School)
rawdata$eighteen$Diagnosis <- lapply(
  rawdata[["eighteen"]]$Diagnosis,
  function(x) {
    unlist(
      lapply(
        x,
        function(x) findingcodes_2018[findingcodes_2018$id == x, "Description"]
      ), use.names = FALSE
    )
  }
)
rawdata$eighteen$Diagnosis[sapply(rawdata$eighteen$Diagnosis, is.null)] <- NA

# choose the required data

# some columns not present in 2018 data
addcols <- function(data, cname) {
  add <-cname[!cname %in% names(data)]
  
  if (length(add) != 0) data[add] <- as.character(NA)
  data
}

data <- lapply(rawdata, function(x) 
  x %>%
    addcols(c("Clinician", "Oral hygiene (daily)")) %>%
    select(
      c("id", "Gender", "Date of birth", "Date seen",
        "School", "Diagnosis", "Clinician", "Sponsored",
        "Oral hygiene (daily)")
    ) %>%
    replace_na(list(Clinician = "Other")) %>%
    mutate(
      Gender = factor(Gender, c("Male", "Female")),
      School = as.factor(School),
      Clinician = as.factor(Clinician),
      Sponsored = replace_na(Sponsored, FALSE),
      OralHygiene = as.numeric(`Oral hygiene (daily)`)
    )
)

# * **Gender** Male, Female and NA. 'NA' is 'not available'
# * **School** School attended (if recorded)
# * **Clinician** The clinician/'surveyor' who saw the child. 
# * **Sponsored** children are supported by the Spur Afrika child development program.

# Some data is 'Factorized' : **Gender**, **School**, **Sponsored** and **Clinician**

# calculate ages
data <- lapply(data, function(x)
  x %>%
    mutate(
      `Date seen` = as.Date(`Date seen`),
      `Date of birth` = as.Date(`Date of birth`)
    ) %>%
    mutate(AgeDays = `Date seen` - `Date of birth`) %>%
    mutate(AgeYears = floor(time_length(AgeDays, "years"))) %>%
    mutate(AgeGroup = as.factor(
      paste0(
        as.character(floor((AgeYears+1)/2)*2-1),
        "-",
        as.character(floor((AgeYears+1)/2)*2)
      ) # two year age groups
    ))
)
# * **AgeDays** Age in days
# * **AgeYears** Age in years
# * **AgeGroup** Grouped into 'two years' groups

# add categories to data

find_string <- function(x, y) {
  # find y string in x. returns vector of TRUE/FALSE
  unlist(
    lapply(
      x,
      function(z) any(toupper(y) %in% toupper(z))
    )
  )
}

data <- lapply(data, function(x)
  x %>%
    mutate(
      PollutionDiagnosis = find_string(
        x$Diagnosis,
        c("Adenitis",
          "Asthma", "Bronchiectasis", "Bronchitis",
          "Lower respiratory tract infection", "Respiratory tract infection",
          "Otitis media", "Otitis media - acute", "Siusitis",
          "Tonsilitis", "Allergic Bronchitis", "Allergic rhinitis",
          "Upper respiratory tract infection", "Viral illness",
          "Viral pharyngitis", "Viral upper respiratory tract infection")
      ),
      PollutionFinding = find_string(
        x$Diagnosis, c("Conjunctivitis - allergic", "Allergic conjunctivitis",
                       "Cough", "Cough - cold weather",
                       "Cough - exertional",
                       "Cough - nocturnal", "Dry eyes", "Dyspnoea", "Dyspnoea - exercise",
                       "Environmental irritant", "Eye inflammation", "Eye irritation", "Photophobia",
                       "Eye pain", "Eyes dry", "Itchy eyes", "Lacrimation",
                       "rhonchi", "runny nose", "Short of breath",  "Smoke irritation", "Sore throat",
                       "Viral pharyngitis",
                       "Watery eyes", "Wheezing", "Rhinorrhoea")
      ),
      PollutionDiagnosisOrFinding = PollutionDiagnosis | PollutionFinding,
      Respiratory = find_string(
        x$Diagnosis, c("Viral upper respiratory tract infection", "Wheezing", "rhonchi", "Viral pharyngitis",
          "Tracheomalacia", "Tonsilitis", "Sore throat", "Rhinorrhoea", "Respiratory tract infection",
          "Nasal congestion", "Dyspnoea", "Dyspnoea - exercise", "Dyspnoea - nocturnal",
          "Cough - nocturnal", "Cough - exertional", "Cough - cold weather", "Cough", "Bronchitis",
          "Asthma", "Allergic rhinitis", "Allergic Bronchitis", "Bronchiectasis", "Short of breath",
          "Tuberculosis", "Upper respiratory tract infection")
      ),
      EarsNose = find_string(
        x$Diagnosis, c("difficulty hearing", "Dry nasal mucosa", "Ear pain", "Ear wax",
                       "Epistaxis", "Foreign body in ear", "Hearing loss", "Hunger pains",
                       "Impacted earwax",
                       "Jaw and temporomandibular joint pain", "Jaw pain",
                       "Lymphadenopathy - cervical", "Nose bridge pain",
                       "Otitis media", "Otitis media - acute", "Otitis externa",
                       "Rhinitis", "runny nose", "Sinusitis", "Viral pharyngitis",
                       "Tonsilitis"
                      )
      ),
      Gastrointestinal = find_string(
        x$Diagnosis, c("Abdominal pain", "Constipation", "Diarrhoea", "Dysphagia", "Dyspepsia",
                       "Epigastric pain", "FODMAP intolerance",
                       "Food intolerance", "Gastro-oesophageal reflux", "Gastro-oesophageal reflux disease (GORD)",
                       "Gastroenteritis", "Gastrointestinal infection", "Glossitis", "Indigestion", "Iron deficiency",
                       "Irritable bowel syndrome", "nausea", "Umbilical hernia")
      ),
      Dermatological = find_string(
        x$Diagnosis, c("Acne", "Atopic Urticaria", "Burns", "Cellulitis", "Dermatitis", "dry eyelid",
          "Eczema", "facial blemish", "Infected wound", "Ingrown toenail",
          "Molluscum contagiosum", "Psoriasis", "rash", "Rashes",
          "Scabies", "Scar", "Skin laceration", "Skin lesion", "Sebaceous cyst", "Tinea", "Tinea capitis")
      ),
      Psychological = find_string(x$Diagnosis, c("Anxiety")),
      Opthalmological = find_string(
        x$Diagnosis, c("Astigmatism", "Blindness","Allergic conjunctivitis", "Conjunctivitis - allergic",
                       "Dry eyes", "Episcleritis", "Eye inflammation",
                       "Eye irritation", "Eye lesion", "Eye pain", "Eyesight poor", "Eyes dry",
                       "Hyperopia", "Iritis",
                       "Itchy eyes", "Lacrimation", "Myopia", "Orbital cellulitis",
                       "Photophobia", "Poor vision",
                       "Pterygium", "Refractive error", "Strabismus",
                       "Watery eyes")
      ),
      Musculoskeletal = find_string(
        x$Diagnosis, c("Arm pain", "Back pain", "Cervicogenic headache", "Hand pain",
          "Joint pain", "Knee pain", "Leg pain", "Lumbago", "Muscle spasm", 
          "Musculoskeletal injury",
          "Musculoskeletal pain", "Neck pain", "Prosthetic limb", "Wrist pain")
      ),
      Dental = find_string(
        x$Diagnosis, c("Caries", "Dental calculus", "Dental carie", "Dental Cavity",
                       "Dental decay", "Dental fracture", "Dental overcrowding",
          "Dental pain", "Dental plaque", "Gingivitis", "Gum disease", "Halitosis",
          "Pericoronitis", "Toothache", "Toothache - cold drink or food",
          "Tooth Hypomineralization", "Tooth loose")
      ),
      Gynaecological = find_string(
        x$Diagnosis, c("Dysmenorrhoea", "Menorrhagia",
                       "Mittleschmerz pain", "Pelvic inflammatory disease",
                       "pre-menstrual pain", "Vaginal infection")
      ),
      Urological = find_string(
        x$Diagnosis, c("Dysuria", "Haematuria", "Nephritis",
                       "Non specific urethritis", "Pyelonephritis",
                       "Urinary tract infection")
      ),
      Neurological  = find_string(
        x$Diagnosis, c("Headache", "Migraine", "Seizures", "Tension headache")
      )
    )
)

# make the data wide
wide_data <- lapply(data, function(x)
  x %>%
    unnest(Diagnosis) %>% # each finding/condition/diagnosis has its own row
    mutate(yesno = TRUE) %>% # 'dummy' column
    distinct() %>%           # get rid of duplicates (there shouldn't be any)
    spread(Diagnosis, yesno, fill = FALSE) # go 'wide'
  # each finding/diagnosis will now have its own column
  # if the patient had the 'Diagnosis' in their list, then 
  #  the column entry will have 'yesno' = TRUE
  #  otherwise will equal the 'fill' = FALSE
)

# Diagnoses/conditions e.g. asthma, bronchitis.
# 
# Findings e.g. cough, itchy eyes.
# 
# In the database, both diagnoses and findings information are found in the **Diagnosis** column.
# 
# Created summary columns are **PollutionDiagnosis**, **PollutionFinding** and **PollutionDiagnosisOrFinding**. These variables are set to TRUE if the child has a condition or finding which could be aggravated or caused by air pollution.

# filter by age and clinician

filtered_wide_data <- lapply(wide_data, function(x)
  x %>%
    filter(AgeYears <= 18) # children only
)
# there are some respondents who are in university etc. and 20+ years old

### Data filtering

# *Only* children up to the age of 16 years is included in the analysis.

category_levels <- lapply(percentage_school, function(x) as.character(names(x)))
sponsored_levels <- lapply(percentage_sponsored, function(x) as.character(names(x)))
percentage_school_df <- mapply(function(x, y)
  as.data.frame(x) %>%
    mutate(Category = y),
  percentage_school, category_levels, SIMPLIFY = FALSE
)
percentage_sponsored_df <- mapply(function(x, y)
  as.data.frame(x) %>%
    mutate(Category = y),
  percentage_sponsored, sponsored_levels, SIMPLIFY = FALSE)

percentage_df <- mapply(function(x, y)
  x %>%
    rename(proportion = `x`) %>%
    mutate(group = 'School (not sponsored)') %>%
    rbind(y %>%
            rename(proportion = `x`) %>%
            mutate(group = 'Sponsored')
    ),
  percentage_school_df, percentage_sponsored_df, SIMPLIFY = FALSE
)

percentage_df[["twenty"]] %>%
  hchart('bar', hcaes(x = 'Category', y = 'proportion', group = 'group')) %>%
    hc_title(text = "Finding categories, children, 2020") %>%
  hc_subtitle(text = paste(
    "Up to age 18 years inclusive seen in Spur Afrika clinic, Kibera, Nairobi<br>",
    "General school clinic and sponsored (Spur Afrika)"
  )) %>%
  hc_tooltip(pointFormat = "<span style=\"color:{point.color}\">●</span> {series.name}: <strong>{point.y:.2f}</strong>") %>%
  hc_exporting(enabled = TRUE)

percentage_df[["eighteen"]] %>%
  hchart('bar', hcaes(x = 'Category', y = 'proportion', group = 'group')) %>%
  hc_title(text = "Finding categories, children, 2018") %>%
  hc_subtitle(text = paste(
    "Up to age 18 years inclusive seen in Spur Afrika clinic, Kibera, Nairobi<br>",
    "General school clinic and sponsored (Spur Afrika)"
  )) %>%
  hc_tooltip(pointFormat = "<span style=\"color:{point.color}\">●</span> {series.name}: <strong>{point.y:.2f}</strong>") %>%
  hc_exporting(enabled = TRUE)

pollution_data_2018 <- data$eighteen %>%
  filter(AgeYears <= 16)
pollution_data_2020 <- data$twenty %>%
  filter(AgeYears <= 16)
  
pollution_model_2018 <- glm(
  PollutionDiagnosisOrFinding ~ Gender + Sponsored,
  data = pollution_data_2018,
  family = binomial(link = "logit")
)

pollution_model_2020a <- glm(
  PollutionDiagnosisOrFinding ~ Gender + Sponsored,
  data = pollution_data_2020,
  family = binomial(link = "logit")
)

pollution_model_2020b <- glmer(
  PollutionDiagnosisOrFinding ~ Gender + Sponsored + (1 | Clinician),
  data = pollution_data_2020,
  family = binomial(link = "logit")
)

pollution_model_2020c <- glmer(
  PollutionDiagnosisOrFinding ~ Gender + Sponsored + Gender * Sponsored + (1 | Clinician),
  data = pollution_data_2020,
  family = binomial(link = "logit")
)

tab_model(
  pollution_model_2018, pollution_model_2020a, pollution_model_2020b, pollution_model_2020c,
  dv.labels = c(
    "2018",
    "2020 - simple model",
    "2020 - with clinician random effect",
    "2020 - with gender:sponsored interaction"
  ),
  show.aic = TRUE,
  title = paste(
    "Conditions potentially attributable to air pollution, children up to age of 16 years inclusive.",
    "Spur Afrika clinics"
  )
)