Rothman: Oral contraceptives and stroke.

Description Usage Format Source Examples

Description

A case-control study of oral contraceptives and stroke in young women with presence or absence of hypertension. Cases represent thrombotic stroke and controls are hospital controls. The group of no hypertension includes normal blood pressure (<140/90 mm Hg) and borderline hypertension (140-159/90-94 mm Hg). Hypertension group includes moderate hypertension (160-179/95-109 mm Hg) and severe hypertension (180+/110+ mm Hg). This data has been used as an example of join exposure by Rothman for measuring interactions (see examples).

Usage

1

Format

A labelled tibble with 477 rows and 3 variables:

stroke

Thrombotic stroke, factor with levels "No" and "Yes".

oc

Current user of oral contraceptives, factor with levels "Non-user" and "User".

ht

Hypertension, factor with levels "No" (<160/95 mm Hg) and "Yes".

Source

Collaborative Group for the Study of Stroke in Young Women (1975) Oral contraceptives and stroke in young women. JAMA 231:718-722.

Rothman, KJ (2002) Epidemiology. An Introduction. Oxford University Press.

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
data(Rothman)

cross_tab(stroke ~ oc + ht, data = Rothman)

mhor(stroke ~ ht/oc, data = Rothman)

## Model with standard interaction term:
model1 <- glm(stroke ~ ht*oc, data = Rothman, family = binomial)
glm_coef(model1)

## Model considering join exposure:
Rothman$join <- 0
Rothman$join[Rothman$oc == "Non-user" & Rothman$ht == "Yes"] <- 1
Rothman$join[Rothman$oc == "User" & Rothman$ht == "No"] <- 2
Rothman$join[Rothman$oc == "User" & Rothman$ht == "Yes"] <- 3
Rothman$join <- factor(Rothman$join, labels=c("Unexposed", "Hypertension", "OC user",
                       "OC and hypertension"))

require(sjlabelled, quietly = TRUE)
Rothman$join <- set_label(Rothman$join, label = "Exposure")

cross_tab(stroke ~ join, data = Rothman)

model2 <- glm(stroke ~ join, data = Rothman, family = binomial)
glm_coef(model2)
odds_trend(stroke ~ join, data = Rothman)$df
odds_trend(stroke ~ join, data = Rothman)$fig

pubh documentation built on Feb. 17, 2021, 1:08 a.m.