NAME:  The Benign Breast Disease Study
KEYWORDS:  Logistic Regression
SIZE:  200 observations, 14 variables
TYPE:  Matched Case-Control

SOURCE: The data are from Appendix 5 of
  Hosmer, D.W. and Lemeshow, S. (1989) Applied Logistic Regression,
  John Wiley and Sons, New York.
They are copyrighted and reproduced here with the permission of 
the publisher.


DESCRIPTIVE ABSTRACT:

	These data contain information from a 1 - 3 matched design studying
the risk factors associated with benign breast disease.  The data is comprised
of a subset of large study in which the original data are from a hospital
based case-control study designed to examine the epidemiology of fibrocystic
breast disease.  Data are provided on 50 women who were diagnosed as having
benign breast disease and 150 age matched controls, with three controls per
case.  Matching was based on the age of the subject at the time of interview.

	Cases included women with a biopsy-confirmed diagnosis of fibrocystic
breast disease identified through two hospitals in New Haven, Connecticut.
Controls were selected from among patients admitted to the general surgery,
orthopedic, or otolaryngologic services at the same two hospitals. Trained
interviewers administered a standardized structured questionnaire to collect
information from each subject [see Pastides et. al. (1983) and Pastides,
et al. (1985)].


Table:  Code Sheet for the Benign Breast Disease Data.

Columns		Name					Abbreviations	
-----------------------------------------------------------------------
4-5		Stratum (1-50)				STR	

9		Observation within a Stratum 		OBS	
		(1 = Case,2-4 = Control)

14-15		Age of the Subject at the Interview	AGMT

21		Final Diagnosis 			FNDX
		(1 = Case, 0 = Control)

26-27		Highest Grade in School			HIGD

32		Degree (0 = None, 1 = High School,	DEG
			2 = Jr. College, 3 = College,
			4 = Masters,
			5 = Doctoral)

37		Regular Medical Checkups		CHK
			(1-Yes, 2=No)

42-43		Age at First Pregnancy			AGP1

48-49		Age at Menarche				AGMN

54		# of Stillbirths, Miscarriages, etc.	NLV

58-59		Number of Live Births			LIV

63-65		Weight of the Subject			WT

69-70		Age at Last Menstrual Period		AGLP

75		Marital Status (1 = Married,		MST
			2 = Divorced, 3 = Separated,
			4 = Widowed, 5 = Never Married)		
------------------------------------------------------------------------
Missing values are denoted by "." in the listing of the data.	


PEDAGOGICAL NOTES:
	These data have been used as an example of fitting a logistic 
regression model for matched case-control studies.


REFERENCES:

1.  Pastides, H., Kelsey, J.L., Holford, T.R., and LiVolsi, V.A.,(1985).
    The epidemiology of fibrocystic breast disease.  American Journal of 
    Epidemiology, 121, 440-447.

2.  Pastides, H., Kelsey, J.L., LiVolsi, V.A., Holford, T., Fischer, D.,
    and Goldberg, I.(1983).  Oral contraceptive use and fibrocystic breast 
    disease with special reference to its histopathology.  Journal of the 
    National Cancer Institute, 71, 5-9.

3.  Hosmer and Lemeshow, Applied Logistic Regression, Wiley, (1989).