preview

Repeat Gonorrhea Infection

Decent Essays

The lack of difference in the rates of repeat gonorrhea infection by region of birth might be attributable to access of care. In the U.S. Army, all active duty personnel have access to free medical care. In this regard, African-American personnel, who are from the south and are low-income before military service, would then have higher health care access. This may result in rates of repeat gonorrhea infection being similar by region of birth after joining the U.S. Army. Another possibility is that region of birth is a key determinant of gonorrhea incidence, but not for repeat infections, and other behavioral, biological, and socio-cultural factors may influence the risk of repeat infection among women. Newmand and Berman (2008), report …show more content…

Army personnel, the crude hazard ratios for the risk of repeat gonorrhea infection by the Andersen-Gill model were similar to those produced by the Prentice-Williams-Peterson model. Koren et al. (2012) reported a similar finding after analyzing recurrent coronary events. One would expect that because these approaches differ in assumptions and answer different research questions, they would yield different results (Villegas et al., 2013; Amorim and Cai, 2014). The Andersen-Gill model assumes a constant hazard of events, whereas the Prentice-Williams-Peterson model assumes that the occurrence of the first event increases the likelihood of the occurrence of the second event. Perhaps this contradictory finding is due to the small frequency of the repeated events per subject. The percentage of women soldiers with a third, fourth, fifth, and seventh episode of gonorrhea infection was only 1.87%, 0.36%, 0.16%, and 0.06%, respectively. There is no evidence in the literature to corroborate this argument. Additionally, the Prentice-Williams-Peterson model produced more robust estimates (i.e., narrower confidence intervals) than the Andersen-Gill model. Kelly and Lim (2000) argue that when the correlation between multiple events within-subject is low, the standard errors are higher for the Andersen-Gill model than the Prentice-Williams-Peterson model. This suggests that the occurrence of the first gonorrhea infection may not cause an increase in

Get Access