### Description

The risk of developing heart failure within 4 years can be predicted using regression equations based on data in the Framingham Study.

Patients:

(1) both men and women

(2) adults aged 45 to 94

X =

= SUM((variable value) * (regression coefficient))

probability of developing heart failure within 4 years =

= 1 / (1 = EXP((-1) * (X)))

Variable for Males

Table 1

Units

Regression Coefficient

intercept

-7.3611

age in years

10 years

0.0313

left ventricular hypertrophy

yes/no

0.8428

vital capacity

100 cL

-0.0030

heart rate

10 bpm

0.0144

systolic blood pressure

20 mm Hg

0.0067

CHD

yes/no

1.5333

valvular disease

yes/no

0.8868

diabetes

yes/no

0.2383

cardiomegaly

yes/no

0.7968

from Table 2, page 1199

where:

• CHD is expanded as congenital heart disease in the legend to the tables, but coronary heart disease seems to be more consistent with the article content.

• cL for vital capacity is interpreted as centiliters

• Left ventricular hypertrophy is based on ECG changes, while a chest radiograph can be used to report cardiomegaly.

• I am a bit confused by the units for age, vital capacity, heart rate and systolic blood pressure. The example calculation given on page 1199 indicates that the patient values for age (in years), vital capacity (in centiliters), blood pressure (in mm Hg) and heart rate (bpm) are used in the calculation, without adjustment to the units listed.

• The presence of a murmur may be used as evidence of valvular disease.

• Table 1 for males and females include vital capacity and chest X-ray (for cardiomegaly), which may not be available in all ambulatory care encounters.

Variable for Males

Table 2

Units

Regression Coefficient

intercept

-9.2087

age in years

10 years

0.0412

left ventricular hypertrophy

yes/no

0.9026

heart rate

10 bpm

0.0166

systolic blood pressure

20 mm Hg

0.00804

CHD

yes/no

1.6079

valvular disease

yes/no

0.9714

diabetes

yes/no

0.2244

from Table 3, page 1200

Variable for Females

Table 1

Units

Regression Coefficient

intercept

-5.4997

age in years

10 years

0.0216

left ventricular hypertrophy

yes/no

1.0072

vital capacity

100 cL

-0.0087

heart rate

10 bpm

0.0092

systolic blood pressure

20 mm Hg

0.0032

CHD

yes/no

1.5358

valvular disease

yes/no

1.2454

diabetes

yes/no

1.4275

cardiomegaly

yes/no

0.4792

valve disease and diabetes

yes/no

-0.9292

from Table 2, page 1199

Variable for Females

Table 2

Units

Regression Coefficient

intercept

-10.7988

age in years

10 years

0.0503

left ventricular hypertrophy

yes/no

1.3402

heart rate

10 bpm

0.0105

systolic blood pressure

20 mm Hg

0.00337

CHD

yes/no

1.5549

valvular disease

yes/no

1.3929

diabetes

yes/no

1.3857

body mass index (BMI)

yes/no (see note)

0.0578

valve disease and diabetes

yes/no

-0.9860

from Table 3, page 1200

where:

• I am not sure how the body mass index can be reported as yes/no. I have used the calculated BMI value in the spreadsheet.