Github repository at https://github.com/DavidPatShuiFong/framinghamRiskEquation

library(framinghamRiskEquation)
packageVersion("framinghamRiskEquation")
#> [1] '0.1.0'

Documentation

?framinghamRiskEquation::framingham_riskequation
framingham_riskequation R Documentation

Framingham Risk Equation

Description

Calculate cardiovascular disease risk according to Framingham Risk Equation

Usage 

framingham_riskequation(df, years = 5, outcome = "CVD")

Arguments

df

a dataframe, which must include the following columns/fields:

  • InternalID (numeric) case identifier

  • CardiovascularDisease (logical - TRUE or FALSE)

  • Diabetes (logical)

  • SmokingStatus (character) - “Smoker” if a smoker. All other values ignored

  • UrineAlbuminDate (character), UrineAlbuminValue (double), UrineAlbuminUnit (character)

  • PersistentProteinuria (logical)

  • eGFRDate (date), eGFRValue (double), eGFRUnits (character)

  • FamilialHypercholesterolaemia (logical)

  • LVH (logical) = left ventricular hyp0ertrophy

  • Cholesterol (double) in millimoles per litre (mmol/L or mM)

  • Triglycerides (double) in millimoles per litre (mmol/L or mM)

  • CholHDLRatio (double)

  • BP (character, two numbers separated by “/”) in mmHg

  • Sex (character) “Female” or not

  • Ethnicity (character) “Aboriginal”, “Torres Strait Islander”, “Aboriginal/Torres Strait Islander” or not

  • Age (double)

years

number of years to predict (from 4 to 12). default is 5 years.

outcome

(default is “CVD”)

“CHD” - coronary heart disease. includes myocardial infarction, death from coronary heart disease plus angina pectoris and coronary insufficiency.

“CVD” cardiovascular disease (the default) includes coronary heart disease, stroke (including transient ischaemia), congestive heart failure and peripheral vascular disease.

Details

Equations and examples sourced from:

  • “Guidelines for the management of Absolute cardiovascular disease risk. 2012” by the National Vascular Disease Prevention Alliance of Australia. http://www.cvdcheck.org.au/pdf/Absolute_CVD_Risk_Full_Guidelines.pdf

  • “Cardiovascular Disease Risk Profiles” by Keaven M. Anderson, Patricia M. Odell, Peter W.F. Wilson, William B. Kannel in ‘American Heart Journal’ 1991; 121:293-298

  • “An Updated Coronary Risk Profile - A Statement for Health Professionals” by Keaven M. Anderson, Peter W.F. Wilson, Patricia M. Odell, William B. Kannel in AHA (American Heart Association) Medical/Scientific Statement, sourced from http://ahajournals.org

Value

a dataframe

  • InternalID - the internalID of the person’s row in the df (dataframe) parameter

  • frisk - numeric (a number, or ‘NA’ if not enough information to computer),

  • friskHI - either ‘NA’ or ‘>15%’. >15%’ are groups considered equivalent risk to already having ischaemic heart disease.

Examples 

framingham_riskequation(data.frame(
  InternalID = 1, BP = "135/80", Sex = "Female",
  Age = 55,
  SmokingStatus = "Smoker", CholHDLRatio = 230 / 48, Diabetes = TRUE, LVH = FALSE,
  CardiovascularDisease = FALSE, PersistentProteinuria = FALSE, eGFRValue = NA,
  eGFRUnits = NA, UrineAlbuminValue = NA, UrineAlbuminUnits = NA,
  FamilialHypercholesterolaemia = NA, Cholesterol = 5.96, Ethnicity = NA
),
outcome = "CHD", years = 10
)
# this comes from "Cardiovascular disease risk profiles" (Anderson 1991)
# the worked answer in the paper is 0.22. this function returns 0.2189125
# the same risk-factors with outcome = "CVD" and years = 5 returns 0.180
#  (cvdcheck.org.au reports 18%)
# the same risk-factors except Sex = "Male" with outcome = "CVD" and years = 5 returns 0.202
#  (cvdcheck.org.au reports 20%)

framingham_riskequation(data.frame(
  InternalID = 2, BP = "130/80", Sex = "Male",
  Age = 55,
  SmokingStatus = "Smoker", CholHDLRatio = 240 / 45, Diabetes = FALSE, LVH = FALSE,
  CardiovascularDisease = FALSE, PersistentProteinuria = FALSE, eGFRValue = NA,
  eGFRUnits = NA, UrineAlbuminValue = NA, UrineAlbuminUnits = NA,
  FamilialHypercholesterolaemia = NA, Cholesterol = 6.22, Ethnicity = NA
),
outcome = "CHD", years = 10
)
# this comes from "An Updated Coronary Risk Profile" (Anderson 1991)
# the worked answer in the paper is 0.192, this function returns 0.1919
# the same risk-factors with outcome = "CVD" and years = 5 returns 0.133
#  (cvdcheck.org.au reports 13%)
# the same risk-factors except LVH = TRUE (outcome = CVD, years = 5) returns 0.211
#  (cvdcheck.org.au reports 21%)

framingham_riskequation(data.frame(
  InternalID = 3, BP = "130/80", Sex = "Female",
  Age = 55,
  SmokingStatus = "Smoker", CholHDLRatio = 240 / 45, Diabetes = FALSE, LVH = FALSE,
  CardiovascularDisease = FALSE, PersistentProteinuria = FALSE, eGFRValue = NA,
  eGFRUnits = NA, UrineAlbuminValue = NA, UrineAlbuminUnits = NA,
  FamilialHypercholesterolaemia = NA, Cholesterol = 6.22, Ethnicity = NA
),
outcome = "CHD", years = 10
)
# this comes from "An Updated Coronary Risk Profile" (Anderson 1991)
# the worked answer in the paper is 0.135, this function returns 0.1349
# the same risk-factors with outcome = "CVD" and years = 5 returns 0.088
#  (cvdcheck.org.au reports 9%)
# the same risk-factors except LVH = TRUE (outcome = CVD, years = 5) returns 0.150
#  (cvdcheck.org.au reports 15%)

Examples

Example 1

This example comes from “Cardiovascular disease risk profiles” (Anderson 1991).

The worked answer in the paper is \(0.22\).

framingham_riskequation(
  data.frame(InternalID = 1, BP = "135/80", Sex = "Female",
             Age = 55, SmokingStatus = "Smoker",
             CholHDLRatio = 230/48, Diabetes = TRUE, LVH = FALSE,
             CardiovascularDisease = FALSE, PersistentProteinuria = FALSE,
             eGFRValue = NA, eGFRUnits = NA,
             UrineAlbuminValue = NA, UrineAlbuminUnits = NA,
             FamilialHypercholesterolaemia = NA, Cholesterol = 5.96,
             Ethnicity = NA),
  outcome = "CHD", years = 10
)
InternalID frisk friskHI
1 0.2189125 NA

The same risk-factors with outcome = “CVD” and years = 5.

The calculator on cvdcheck.org.au reports \(18\%\).

framingham_riskequation(
  data.frame(InternalID = 1, BP = "135/80", Sex = "Female",
             Age = 55, SmokingStatus = "Smoker",
             CholHDLRatio = 230/48, Diabetes = TRUE, LVH = FALSE,
             CardiovascularDisease = FALSE, PersistentProteinuria = FALSE,
             eGFRValue = NA, eGFRUnits = NA,
             UrineAlbuminValue = NA, UrineAlbuminUnits = NA,
             FamilialHypercholesterolaemia = NA, Cholesterol = 5.96,
             Ethnicity = NA),
  outcome = "CVD", years = 5
)
InternalID frisk friskHI
1 0.1803611 NA

The same risk-factors except Sex = “Male” with outcome = “CVD” and years = 5.

The online calculator on cvdcheck.org.au reports \(20\%\).

framingham_riskequation(
  data.frame(InternalID = 1, BP = "135/80", Sex = "Male",
             Age = 55, SmokingStatus = "Smoker",
             CholHDLRatio = 230/48, Diabetes = TRUE, LVH = FALSE,
             CardiovascularDisease = FALSE, PersistentProteinuria = FALSE,
             eGFRValue = NA, eGFRUnits = NA,
             UrineAlbuminValue = NA, UrineAlbuminUnits = NA,
             FamilialHypercholesterolaemia = NA, Cholesterol = 5.96,
             Ethnicity = NA),
  outcome = "CVD", years = 5
)
InternalID frisk friskHI
1 0.2016813 NA

Example 2

This comes from “An Updated Coronary Risk Profile” (Anderson 1991).

The worked answer in the paper is \(0.192\).

framingham_riskequation(
  data.frame(InternalID = 2, BP = "130/80", Sex = "Male",
             Age = 55, SmokingStatus = "Smoker",
             CholHDLRatio = 240/45, Diabetes = FALSE, LVH = FALSE,
             CardiovascularDisease = FALSE, PersistentProteinuria = FALSE,
             eGFRValue = NA, eGFRUnits = NA,
             UrineAlbuminValue = NA, UrineAlbuminUnits = NA,
             FamilialHypercholesterolaemia = NA, Cholesterol = 6.22,
             Ethnicity = NA),
  outcome = "CHD", years = 10
)
InternalID frisk friskHI
2 0.1919419 NA

The same risk-factors with outcome = “CVD” and years = 5.

The online calculator at cvdcheck.org.au reports \(13\%\).

framingham_riskequation(
  data.frame(InternalID = 2, BP = "130/80", Sex = "Male",
             Age = 55, SmokingStatus = "Smoker",
             CholHDLRatio = 240/45, Diabetes = FALSE, LVH = FALSE,
             CardiovascularDisease = FALSE, PersistentProteinuria = FALSE,
             eGFRValue = NA, eGFRUnits = NA,
             UrineAlbuminValue = NA, UrineAlbuminUnits = NA,
             FamilialHypercholesterolaemia = NA, Cholesterol = 6.22,
             Ethnicity = NA),
  outcome = "CVD", years = 5
)
InternalID frisk friskHI
2 0.1332337 NA

The same risk-factors except LVH = TRUE (outcome = CVD, years = 5).

The online calculator at cvdcheck.org.au reports \(21\%\).

framingham_riskequation(
  data.frame(InternalID = 2, BP = "130/80", Sex = "Male",
             Age = 55, SmokingStatus = "Smoker",
             CholHDLRatio = 240/45, Diabetes = FALSE, LVH = TRUE,
             CardiovascularDisease = FALSE, PersistentProteinuria = FALSE,
             eGFRValue = NA, eGFRUnits = NA,
             UrineAlbuminValue = NA, UrineAlbuminUnits = NA,
             FamilialHypercholesterolaemia = NA, Cholesterol = 6.22,
             Ethnicity = NA),
  outcome = "CVD", years = 5
)
InternalID frisk friskHI
2 0.2113987 NA

Example 3

This comes from “An Updated Coronary Risk Profile” (Anderson 1991)

The worked answer in the paper is \(0.135\).

framingham_riskequation(
  data.frame(InternalID = 3, BP = "130/80", Sex = "Female",
             Age = 55, SmokingStatus = "Smoker",
             CholHDLRatio = 240/45, Diabetes = FALSE, LVH = FALSE,
             CardiovascularDisease = FALSE, PersistentProteinuria = FALSE,
             eGFRValue = NA, eGFRUnits = NA,
             UrineAlbuminValue = NA, UrineAlbuminUnits = NA,
             FamilialHypercholesterolaemia = NA, Cholesterol = 6.22,
             Ethnicity = NA),
  outcome = "CHD", years = 10
)
InternalID frisk friskHI
3 0.1348653 NA

The same risk-factors with outcome = “CVD” and years = 5.

The online calculator cvdcheck.org.au reports \(9\%\).

framingham_riskequation(
  data.frame(InternalID = 3, BP = "130/80", Sex = "Female",
             Age = 55, SmokingStatus = "Smoker",
             CholHDLRatio = 240/45, Diabetes = FALSE, LVH = FALSE,
             CardiovascularDisease = FALSE, PersistentProteinuria = FALSE,
             eGFRValue = NA, eGFRUnits = NA,
             UrineAlbuminValue = NA, UrineAlbuminUnits = NA,
             FamilialHypercholesterolaemia = NA, Cholesterol = 6.22,
             Ethnicity = NA),
  outcome = "CVD", years = 5
)
InternalID frisk friskHI
3 0.0883931 NA

The same risk-factors except LVH = TRUE (outcome = CVD, years = 5).

The online calculator cvdcheck.org.au reports \(15\%\).

framingham_riskequation(
  data.frame(InternalID = 3, BP = "130/80", Sex = "Female",
             Age = 55, SmokingStatus = "Smoker",
             CholHDLRatio = 240/45, Diabetes = FALSE, LVH = TRUE,
             CardiovascularDisease = FALSE, PersistentProteinuria = FALSE,
             eGFRValue = NA, eGFRUnits = NA, 
             UrineAlbuminValue = NA, UrineAlbuminUnits = NA,
             FamilialHypercholesterolaemia = NA, Cholesterol = 6.22,
             Ethnicity = NA),
  outcome = "CVD", years = 5
)
InternalID frisk friskHI
3 0.1500842 NA

Example 4

This comes from “Maximising the effectiveness and cost-effectiveness of cardiovascular disease prevention in the general population” by Andrew M Tonkin, Andrew N Boyden and Stephen Colagiuri in ‘Medical Journal of Australia’ (191:6) 21 September 2009.

Using the provided Australian cardiovascular risk charts, the estimated risk is \(5-9\%\).

The online calculator cvdcheck.org.au (with total cholesterol = 7.2, HDL 1.0) reports \(9\%\).

framingham_riskequation(
  data.frame(InternalID = 4, BP = "140/80", Sex = "Female",
             Age = 46, SmokingStatus = "Smoker",
             CholHDLRatio = 7.2, Diabetes = FALSE, LVH = FALSE,
             CardiovascularDisease = FALSE, PersistentProteinuria = FALSE,
             eGFRValue = NA, eGFRUnits = NA, 
             UrineAlbuminValue = NA, UrineAlbuminUnits = NA,
             FamilialHypercholesterolaemia = NA, Cholesterol = 7.2,
             Ethnicity = NA),
  outcome = "CVD", years = 5
)
InternalID frisk friskHI
4 0.0886549 NA

This example also comes from “Maximising the effectiveness and cost-effectiveness of cardiovascular disease prevention in the general population”.

Using the provided Australian cardiovascular risk charts, the estimated risk is \(20-24\%\).

The online calculator cvdcheck.org.au (with total cholesterol = 7.2, HDL 1.0) reports \(20\%\).

framingham_riskequation(
  data.frame(InternalID = 4, BP = "140/80", Sex = "Female",
             Age = 66, SmokingStatus = "Smoker",
             CholHDLRatio = 7.2, Diabetes = FALSE, LVH = FALSE,
             CardiovascularDisease = FALSE, PersistentProteinuria = FALSE,
             eGFRValue = NA, eGFRUnits = NA, 
             UrineAlbuminValue = NA, UrineAlbuminUnits = NA,
             FamilialHypercholesterolaemia = NA, Cholesterol = 7.2,
             Ethnicity = NA),
  outcome = "CVD", years = 5
)
InternalID frisk friskHI
4 0.1968354 NA