Background and Objectives: Female inmates have high rates of sexually transmitted diseases (STDs), and many incarcerated women and jail providers believe STDs are acquired within the jail. We investigated a suspected outbreak of trichomoniasis among female inmates and described the epidemiology of trichomonas infection.
Goals of this Study: To determine the likelihood of within-jail acquisition of trichomoniasis.
Study Design: Retrospective chart review of gynecologic visits to the jail medical clinic and comparison of trichomoniasis surveillance data over a 6-year time period.
Results: The minimum prevalence of trichomoniasis infection among 450 female inmates presenting to the medical clinic for gynecologic evaluation was 37%. Most infections were diagnosed early after incarceration, no woman developed a new infection after adequate treatment, and there was no clustering of cases by time or location.
Conclusion: There was no evidence to support within-jail acquisition of trichomoniasis. The high rate of trichomoniasis and other STDs among incarcerated women warrant more comprehensive jail-based STD screening programs.
TRICHOMONAS VAGINALIS is one of the most common sexually transmitted diseases (STDs) in the United States.1 Trichomoniasis has been reported in 5% of women attending family planning clinics, in 13% to 25% of women attending gynecology clinics, in 47% of pregnant inmates, and in 50% to 75% of female sex workers.2-4 Consequences of trichomoniasis include vaginitis, emotional anguish, medical costs, and complications during pregnancy.5
Trichomoniasis may increase the sexual transmission of HIV.6-8 The long-term adverse effects of T. vaginalis infection, if any, are unknown. There has been a recent report that trichomoniasis is associated with pelvic inflammatory disease in women coinfected with Chlamydia trachomatis, suggesting that T. vaginalis infection could be a risk factor for upper tract disease.9
There has been evidence that T. vaginalis is transmitted predominantly through penile-vaginal sexual intercourse since 1947,10 but transmission by means of fomites and other forms of sexual contact, including female-to-female sexual contact,11 continue to concern patients and clinicians. Trichomonads have been recovered from towels, sponges, and toilet seats; however, transmission of infection from environmental sources has not been documented.12 The incubation period for trichomoniasis ranges from 4 to 20 days, average 7 days, but many women may be asymptomatic carriers for years.13
Incarcerated women are at increased risk for STDs for a variety of social and behavioral reasons such as poverty, low education, poor access to health care, and the exchange of money or drugs for sex. In this report, we examined cases of trichomoniasis at a women's correctional facility, the evidence for acquisition of T. vaginalis infection within the facility, and the prevalence of other STDs among the women with trichomoniasis.
In September 1995, a nurse practitioner at a county correctional center for women noted that since the end of March 1995, there had been a large increase in the number of patients diagnosed with trichomoniasis at the facility. Staff and inmates at the facility were concerned about the possibility of environmental transmission and person-to-person transmission of trichomonads. Inmates were lining their undergarments with absorbent pads, hoarding toilet paper rolls to keep for individual use, self-washing the one pair of underwear issued, and blaming the facility's lack of cleanliness for causing their infection. There were reports of patients being diagnosed with trichomoniasis up to 1 year after incarceration and of patients being diagnosed without any recent history of sexual contact. In addition, there were reports of several patients who developed second infections while in jail after appropriate therapy for the first infection.
The jail houses approximately 650 inmates and books approximately 11,500 inmates each year for an average daily booking rate of 34 women per day. The housing areas are separated into minimum, medium, and maximum security, divided among 17 housing areas. Inmates in minimum or medium security housing areas share shower, toilet, and dining facilities, whereas those in maximum security are housed in solitary cells with their own toilets, but with shared showers. Inmates frequently are moved among security housing areas and housing units. There is an on-site medical clinic where staff members perform about 10 pelvic examinations a week. Trichomoniasis is diagnosed by wet-mount preparation of vaginal specimens collected during pelvic examinations and viewing motile trichomonads with light microscopy. Treatment for T. vaginalis followed standard recommendations.
Cases were defined as the occurrence of T. vaginalis, as recorded in the jail medical data base coded with International Classification of Disease (ICD)-9 diagnosis of trichomonas, unspecified (131.9), in inmates of the facility from 1990 to 1995. To evaluate the accuracy of the ICD-9 case definition, a more strict case definition for 1995 was termed as one of the following: a positive wet-mount (the presence of trichomonads in vaginal secretions by light microscopy) recorded in the patient's chart or a positive cervical Pap smear for trichomonads reported in the laboratory records.
Medical records of all inmates who visited the medical unit in 1995 with symptoms of vaginitis or lower abdominal pain and had a gynecologic examination were reviewed. Additional cases were discovered by reviewing laboratory records of Pap smears performed in 1995. All medical records of 1995 case-patients diagnosed more than 60 days after incarceration were rereviewed to identify any history of antitrichomonal therapy or a clinic visit with gynecologic examination while incarcerated.
Census and booking information for the facility was obtained from the weekly population summary prepared by the facility administrative offices. Housing location at the time of diagnosis was obtained from the medical record.
The facility's laundering procedures were evaluated by interviews with management staff and facility workers. Maximum washing and drying temperatures and duration of drying were assessed and compared with industry standards.
Demographic and clinical information were gathered from medical records and recorded on a standardized data collection form. Active syphilis was defined as a positive nontreponemal serologic test for syphilis (Venereal Disease Research Laboratory test), confirmed by an antitreponemal antibody test (microhemagglutination assay for T. pallidum [MHATP]) and inadequate treatment history. Data were entered into a computer with EPI-INFO 6.02 (Centers for Disease Control, Atlanta, GA). The Chi square test with Yate's correction was used to compare differences in rates and proportions of cases by year and housing location.
In 1995, 152 cases of trichomoniasis were recorded at the facility's medical unit by ICD-9 classification for a prevalence rate of 16.5 cases of trichomoniasis per 1,000 inmates booked. This is a 35% increase, compared with 12.2 cases per 1,000 inmates booked in 1994 (p < 0.02). Comparing rates of trichomoniasis by year from 1990 to 1995, however, showed no significant trend (Figure 1), and the case-rate for 1995 was significantly lower than the previous 5-year average of 19.9 cases per 1,000 inmates booked (p < 0.05).
During 1995, 450 women presented to the medical clinic complaining of symptoms of vaginitis or lower abdominal pain and underwent gynecologic examination. Trichomoniasis was diagnosed in 165 of 450 (37%) women either by light microscopy (n = 131) or by Pap smear (n = 34). The number of cases in 1995 ascertained by chart and laboratory review (n = 165) was only a modest 9% increase over the number (n = 152) found by reviewing ICD-9 reporting data.
Figure 2 shows the time interval from date of incarceration to date of diagnosis. Half of the case-patients with vaginal complaints were diagnosed within 20 days of incarceration. Case-patients who were diagnosed more than 60 days after incarceration (n = 19) did not have previous antitrichomonal therapy or a previous clinic visit with gynecologic examination. The median time interval from date of incarceration to date of documented clinic request (n = 73) was 12 days (range, 1 to 101).
There were no significant differences among case rates when compared by housing location or month of diagnosis. An inspection of the laundry facilities showed no breach in sanitization practices. Wash and dry temperatures in the laundry and the duration of drying equaled or surpassed industry standards to sterilize infected clothing, including undergarments.
The mean age (±sd) of case-patients was 34 (±7) years, which was comparable to the mean age of the general facility population of 35 (±4). Of case-patients, 119 (75%) of 158 had been previously booked, 86 (55%) of 157 had a history of STD, 21 (19%) of 108 were intoxicated at the time of booking, and 9 (6%) of 158 were pregnant. Of women with trichomoniasis who were tested, Neisseria gonorrhoeae was detected in 1% (1 of 115), C. trachomatis in 5% (6 of 114), and active syphilis in 8% (5 of 66). Of the women with N. gonorrhoeae at the facility, 20% (1 of 5) had trichomoniasis, whereas 38% (6 of 16) of the women with C. trachomatis had trichomoniasis.
There was an apparent increase in the number of cases of trichomoniasis in 1995 at the jail compared with 1994, but 1994 represented a 6-year low for the time period studied. There was no increase in the case-rate in 1995 when compared with the 5-year average. Any comparison of annual prevalence of T. vaginalis infection in the population is limited, however, by the changing demographics of the underlying population at risk, changes in women's access to medical services, the provision of medical services, and the diagnostic practices of the clinicians.
To assess the reliability of the jail surveillance system used to detect cases, we compared the accuracy of the recorded ICD-9 diagnoses used in the annual summary with the cases ascertained by medical and laboratory review. The similar results attained by these two surveillance methods imply that the ICD-9 diagnosis for trichomoniasis is indeed a fairly accurate measure of the actual number of cases of trichomoniasis diagnosed at the jail and therefore appropriate for assessing trends. These observed prevalence estimates are likely, however, to underestimate the true prevalence of trichomoniasis because of the lack of sensitivity of microscopy and Pap smear for diagnosis.
To investigate the possibility of within-jail acquisition of infection, the time interval from the date of incarceration to the date of clinic visit request (median = 12 days) was compared with the range of the typical incubation period (4-20 days) and found to be within this interval. Because the time interval to clinic visit request was not beyond the average incubation period of trichomoniasis, this information suggests that most of the women diagnosed with trichomoniasis most likely had prevalent infection at the time of incarceration and merely presented for clinical evaluation while in the facility. In addition, the shape of the frequency distribution demonstrating the time interval from date of incarceration to the date of clinic visit request (Figure 2) showed that the majority of cases occurred early on. Although there were several women diagnosed with trichomoniasis after being in the facility for more than 60 days, in these instances there was no history of previous antitrichomonal treatment within the facility, thereby offering no cases of treatment failure or reinfection after treatment within the facility. The long time intervals to clinic visit request in these cases may be more readily explained by the varied incubation period of trichomoniasis, which can extend in some women to longer than 6 months.1
There was no epidemiologic evidence for environmental transmission of trichomonas infection. The lack of clustering of cases within one housing area or by month of diagnosis is evidence for the lack of environmental transmission in this setting, although the mobility of the population and the large variance in the case-rates by housing unit make this difficult to completely exclude. Shared personal toilet articles, exposure to showers or toilet seats, etc., that might increase the risk of infection might have been expected to cluster in one or two geographic areas, or if specific behaviors were implicated, by month of diagnosis. Clustering by place or time would be expected unless transmission was a widespread environmental problem occurring throughout the jail, but the moderate overall prevalence is consistent with previous studies in similar high-risk patients and does not suggest a larger epidemic.2-4
A majority of patients had a history of previous incarceration, indicating high recidivism, but offering an opportunity for patient education and screening regarding STDs. Almost 20% of case-patients were reported as intoxicated when booked, indicating that a proportion of patients may have been unaware of their symptoms of possible infection when entering the facility, because of their intoxicated status. Finally, more than half of the patients had a prior STD history, which is consistent with previous data demonstrating STD history to be a risk factor for current STD infection.3
The proportion of case-patients coinfected with other bacterial STDs is not large enough to warrant empiric cotreatment, but does indicate the necessity in this and similar facilities for screening case-patients with trichomoniasis for C. trachomatis, N. gonorrhoeae, and syphilis. In particular, with the large proportion of asymptomatic C. trachomatis and N. gonorrhoeae infections and long latency period in syphilis, any marker for current infection would be useful in discovering new cases and preventing adverse clinical sequelae, such as pelvic inflammatory disease.
During 1995, there was an increase in the case-rate of trichomoniasis at the jail over 1994, the year with the lowest case-rate over the previous 5 years. When comparing case-rates over the previous 5 years, however, the 1995 case-rate was not increased over the average. There was no evidence for acquisition of trichomoniasis within the facility, as supported by the occurrence of the majority of cases within 2 weeks of incarceration, absence of time or geographic clustering, absence of cases occurring after appropriate treatment, and absence of defects in environmental sanitization. Rates of coinfection in case-patients support comprehensive STD screening for C. trachomatis, N. gonorrhoeae, and syphilis to increase rates of detection and ensure appropriate treatment in these women who are at high risk for STDs.