3.4 Results
The ALIVE GWAS subset includes 1,197 participants drawn from the ALIVE cohort in 1998. The median age at first visit was 34. There were 287 (24%) females and 882 (76%) males. They visited semiannually with 19 average visits per person. This translates into about 9 years of information on each subject. The population was predominantly African American (98%). About half of the participants were HIV positive when they entered the ALIVE cohort. 1,029 participants were injection drug users when they entered the study, while only 502 were still using injection drugs as of their last visit. When they first entered the study, the number of participants who were using more than once a day, less than once a day and not using were 449, 415, and 159, respectively. After 13 years in the program, 231 of the 496 injection drug users were still using more than once a day. Table 2 shows the characteristics of the GWAS subset in the ALIVE cohort. In the ALIVE GWAS subset, the mean age of first injection was 21, and the histogram is right-skewed. More than half used for the first time between ages 14 and 26. The mean age of participants entering the cohort was 34. The years between first use and entering the program were regarded as missing data, although one discreet…show more content… This presented a problem since so much data was not available. There were 103,599 visits recorded for all the participants in the ALIVE GWAS subset, but this represented just one-third of all potential data from the age of first use forward: 66 percent was missing. The largest gaps of information occurred in injection status between the ages of 20 and 30: 75 percent of the data was not recordable. In other words, the younger the participant, the greater the likelihood of unretrievable information. Table 3 shows the distribution of missing data in the ALIVE GWAS
