Statistics 227

Final Assignment Problem 2
Due noon Wednesday, 11 December 1996


Data set corrected as of 4-Dec-96 5:15 pm CST


Duchenne Muscular Dystrophy, DMD, is a genetically transmitted disease, passed from a mother to her children. Affected female offspring usually suffer no apparent symptoms and may unknowingly carry the disease. Male offspring with the disease die at a young age. Not all cases of the disease come from an affected mother. A fraction, perhaps one third, of the cases arise spontaneously, to be genetically-transmitted by an affected female. This is the most widely held view at present. The incidence of DMD is about 1 in 10,000 male births. The population risk that a woman is a DMD carrier is about 1 in 3,300.

The detection problem arises when, for example, a woman suspects that she may be a carrier. This suspicion may start with the detection of an affected male child in the family. Although carriers of DMD usually have no physical symptoms, they tend to exhibit elevated levels of certain serum enzymes or proteins, such as creatine kinase (CK), hemopexin (H), lactate dehydroginase (LD) and pyruvate kinase (PK). The levels of these enzymes may also depend on age and season.

A program to develop an effective method for screening female relatives of boys with DMD began under the direction of Dr. M. Thompson, the Hospital for Sick Children, Toronto. The purpose of the program was to develop a procedure for informing a woman of the likelihood that she is a carrier, based on measurements of serum markers and her family pedigree.

Levels of the enzymes were measured in known carriers and in a group of non-carriers using standard laboratory procedures. The determinations are made by rather delicate, indirect measurement. The data are presented in Table 38.1. In this table, the measurements from separate blood samples are identified by a hospital ID number. In some cases, more than one sample was obtained from an individual. This observation number is given as well as the date the sample was taken.

[For this examination, only the first sample from each individual is included--RT. A Stata dictionary that gives suggested names and a bit of information on this data set can also be retrieved from this Web site; the name of the dictionary page is T38.dct.]

The first two serum markers (CK and H) may be measured rather inexpensively from frozen serum. The second two (PK and LD) require fresh serum. Do the second two increase the detection rate in an important way? Should age and season be taken into account?

This data set and documentation were obtained from Chapter 38 of Data, by David F. Andrews and Agnes M. Herzberg, Springer-Verlag, 1985. The data and the description above were originally provided by M. Percy, Mount Sinai Hospital, Toronto.