DATA SET EXAMPLE TWO:  Substance Use and Menstrual Cycle Function

This dataset is a subset of data from a large cohort study of women and HIV serostatus. Although HIV serostatus had little effect on menstrual function, substance use may be an important determinant of menstrual cycle length and the probability of having a cycle longer than 40 days. Women from the cohort study were enrolled in the menstrual diary substudy and maintained menstrual diaries for 6 months. Any woman who returned at least three consecutive diaries was considered a participant. Information on age, bodymass index  (BMI) (weight/height squared), and depressed mood were obtained at the enrollment visit. Use of crack or intravenous drug use was obtained at enrollment and at the 6 month folllow-up visit. If use was reproted at either visit, the woman was coded as a user.

The data set is structured as follows:

Column		Variable		Coding

   One		ID			8 digit unique ID for each woman

   Two		Age			Years, 18 to 44	missing=99

   Three	BMI			continuous	missing=99

    Four	Depressed Mood		0=no, 1=yes	missing=9

    Five	Ever used Crack		0=no, 1=yes	missing=9

    Six		Cycle length		if cycle is  18-40 days, number of days; else missing (999)

   Seven	Cycle >40 days		if cycle length is <=40 days=0
					if cycle is >40 days=1
					missing =9

   Eight	Used Injection Drugs	0=no, 1=yes	missing=9