Question

Cohort Study

A local group of obstetricians (OB/GYNs) has hired your team to conduct a cohort study among pregnant women seen at their three satellite offices. The study is intended to find out whether a mother’s body mass index (BMI) at 12 weeks gestation influences her risk for preterm delivery. Over a period of six months, you enrolled 600 pregnant women during their first OB appointment at one of the three satellite offices. Women were excluded from the study if they were determined to be beyond 12 weeks gestation at their first appointment or they had a previous pregnancy that resulted in a preterm delivery (before 36 weeks gestation). The BMIs of all participants were measured at 12 weeks gestation and 98 women were considered to have a high BMI (25.0 or greater). After following all of the participants through delivery without any loss to follow-up, you found that 52 women delivered preterm infants, 20 among those who had a high BMI.

1. What is the relative risk (calculation)

2. Can you calculate a measure to describe the potential impact of the exposure on the disease in the study population (i.e., attributable risk (AR)) and/or in the total population (i.e., population attributable risk (PAR))?

Answer 1

1. To find the relative risk of mothers having high BMI to have preterm delivery,

Find the ratio: P(No. of mothers with high BMI have preterm delivery)/P(no. of mothers have preterm delivery with BMI otherwise)

Relative Risk = 0.625

Indicating that the relative risk of mothers having a preterm delivery and high BMI is 0.625 times that of mothers having a preterm delivery with BMI otherwise.

where P is the probability of occurance.

Delivery Pre-term Otherwise Total 20 32 52 98 BMI High Otherwise Total 600(but exluded certain cases) 20/52 0.385 RR 30/52-0.615 0.385/0.615 0.625

2. Population Attribute Risk (PAR) is a measure of how much portion of the event that occurs can be attributed to a certain exposure/risk.

Here, the risk is High BMI

and the event is preterm delivery.

No. of events in the total population = 52

No. of events without exposure (BMI complications) = 32

Then, PAR = 52-32

= 20

Another measure to describe the potential impact of the exposure on the disease may be PAR percent

Here, PAR % = * 100

= 0.3846 * 100

= 38.46%