TABLE 1. Demographic and Behavioral Characteristics of Maricopa County Male Arrestee Population.

	Age Group, y										Source
	15–17	18–19	20–24	25–29	30–34	35–39	40–44	45–54	55–64	>65
Age distribution of arrestees	4%		22%	19%	16%	11%	10%	14%	4%	1%	CHS
Prevalence of infection
CT	18%		9%	11%	8%	3%	0%	0%	0%	0%	10
GC	7%		7%	7%	2%	3%	2%	3%	0%	0%	10
Sexual behavior
Mean duration of partnership (mo)*,†	16	14	17	27	41	40	57	57	57	57	15
Proportion without partner on day of release, %‡	18	12	7	6	5	6	9	9	9	9	15
No. sex acts per year*,§	20–41	73–127	73–127	62–108	51–93	51–93	48–86	48–86	40–73	32–62	16
Proportion condom protected sex§	0.58	0.39	0.39	0.27	0.18	0.18	0.14	0.14	0.09	0.09	17

^*Number of partners/sex acts estimated as average of reported number of partners/sex acts weighted by proportion reporting under each category of number of partners/sex acts among those sexually active.

^†Assuming a person is in a partnership, its remaining duration, that is, the duration before the partnership ends, is determined as an exponentially distributed random number. The mean duration (in months) was estimated as 12 divided by (number of partners per year − 1).

^‡Estimated using proportion who has had sex but currently does not have a partner. The same proportion is used to determine the duration of no partnership as a geometrically distributed random number. These values are also applied after a partnership ends to determine the proportion that would not immediately have a new partner.

^§Note that in simulating transmissions in a partnership, the behavior of both persons involved, that is, the inmate and his female partner assuming heterosexual, is considered. Because the source data for sex acts and condom use are from the general population, assuming female partners are from the general population and assuming not much age difference between inmates and their partners, we used sex acts and condom use data of women.