Microbicides Development Program, Tanzania???Baseline Characteristics of an Occupational Cohort and Reattendance at 3 Months

of 6

Please download to get full document.

View again

All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
PDF
6 pages
0 downs
2 views
Share
Description
Microbicides Development Program, Tanzania???Baseline Characteristics of an Occupational Cohort and Reattendance at 3 Months
Tags
Transcript
  Microbicides Development Program, Tanzania—BaselineCharacteristics of an Occupational Cohort  and Reattendance at 3 Months  ANDREW VALLELY, MBBS, DTMH, MRCP, MS C ,*† STELLA KASINDI, MD, MPH,†IAN R. HAMBLETON, BA, MS C , P H D,*‡ LOUISE KNIGHT, BS C , MS C ,*‡ TOBIAS CHIRWA, P H D,*‡REBECCA BALIRA, BS C ,*‡ JOHN CHANGALUCHA, BS C , MS C ,‡DEBORAH WATSON-JONES, BA, BM, BC H , MS C , P H D,*† DEAN EVERETT, BS C , MS C ,*‡ AWENE GAVYOLE, MD, MPH,† JOCELYN MOYES, MD,*† MAR PUJADES-RODRı´GUEZ, MD,*†DAVID A. ROSS, MA, BM, BC H , MS C , P H D,* AND RICHARD J. HAYES, BS C , MS C , DS C * Objectives:  To determine baseline characteristics of an occupationalcohort of women in Mwanza City, Tanzania, and factors associated withreattendance at 3 months, in preparation for a microbicide trial. Study Design:  One thousand five hundred seventy-three womenaged 16–54 years working in food outlets and recreational facilitieswere enrolled, interviewed, and examined at community-based repro-ductive health clinics, provided specimens for HIV/STI and pregnancytesting, and attended 3 monthly clinical follow-up.  Results:  Baseline prevalence of HIV was 25.5%; pregnancy 9.7%;herpes simplex virus type-2 74.6%; active syphilis 10.2%, bacterialvaginosis 52.6%; gonorrhea 5.5%; chlamydia 5.9%; and trichomoni-asis 12.3%. Reattendance at 3 months was 74.1% and was higher inolder women, less mobile women, and in those who received an HIV-negative result at enrollment. Conclusions:  Baseline characteristics of this occupational groupsuggest their suitability for microbicide trials. A screening round,locally appropriate informed consent procedures, and effective com-munity tracing may help reduce losses to follow-up in such settings. MWANZA IS 1 OF 6 CENTERS in sub-Saharan Africa partici-pating in the Microbicides Development Program (MDP), aninternational partnership for the development of vaginal microbi-cides for HIV prevention. 1 In preparation for phase III trials of candidate microbicides,investigators in a variety of settings have found preliminary fea-sibility and acceptability studies helpful. 2–7 In Tanzania, high ratesof HIV and sexually transmitted infections (STIs) have beenreported among women working in bars, restaurants, and guest-houses in areas situated along major transit routes 8–11 and incommercial centers adjacent to newly established gold mines. 12,13 Female workers in such facilities are reported to periodicallysupplement their income through transactional sex 13,14 and, al-though not necessarily perceived as commercial sex workerswithin the wider community, are nonetheless at increased risk of STIs and HIV infection. 8,10,12 Previous studies in Tanzania havefocussed on women working in bars (modern and traditional),guesthouses, hotels, and restaurants. Traditional bars are known as vilabu  or pombe shops in Tanzania and sell locally brewed beer,typically made from fermented honey or bananas. Pombe shopstend to cluster in economically deprived areas compared to modernbars selling commercially brewed beer, a fact which is reflected intheir client base. 14 In preparation for the MDP301 randomized, placebo-controlledclinical trial of the candidate vaginal microbicide PRO2000/5, wedecided to explore the feasibility of recruiting women from abroader occupational group and, specifically, to include womenknown locally as  mamalishe  who serve food in makeshift facilitiesalong roadsides and in market areas, typically adjacent to  vilabu .Although  mamalishe  are numerous in urban centers in Tanzania,the prevalence and incidence of HIV and STIs, risk factors forinfection, and their sexual behaviors have not previously beendescribed. Women working in a broad range of food and recre-ational facilities in 10 administrative wards in Mwanza City,northern Tanzania, were therefore enrolled into a feasibility studyconducted from July 2002 to December 2004. In this paper wedescribe the baseline sociodemographic and behavioral character-istics of this cohort, baseline prevalence of HIV, other STIs andpregnancy, and factors associated with HIV infection and reatten-dance at 3 months. The authors thank the staff of the National Institute for Medical Re-search, Tanzania, and the African Medical and Research Foundation,Tanzania, for their support and assistance in carrying out this study. Theauthors also acknowledge the contribution of Sheena McCormack and staff of the Clinical Trials Unit, Medical Research Council, London, UK (tech-nical support); Tania Cruitti, STD/HIV Research and Intervention Unit,Institute of Tropical Medicine, Antwerp, Belgium ( T. vaginalis  PCR test-ing); and Eloise Turner, London School of Hygiene and Tropical Medicine,UK (logistics support). The Authors also thank the women of Mwanza whoparticipated in the study.Correspondence: Andrew Vallely, MBBS, DTMH, MRCP, MSc, NIMR/ AMREF/LSHTM Collaborative Research Projects Mwanza, PO Box 1482,Mwanza, Tanzania. E-mail: andrew.vallely@lshtm.ac.uk.Received for publication August 7, 2006, and accepted December 20,2006. From the *London School of Hygiene and Tropical Medicine,Keppel St., London, United Kingdom; †African Medical and Research Foundation, Lake Zone Program, Mwanza, Tanzania; and ‡National Institute for Medical Research, Mwanza Center,Mwanza, Tanzania Sexually Transmitted Diseases,  September 2007, Vol. 34, No. 9, p.638–643DOI: 10.1097/01.olq.0000258431.18986.f3Copyright © 2007, American Sexually Transmitted Diseases Association All rights reserved. 638  Methods Study Population In March 2002 fieldworkers identified and visited all food andrecreational facilities in 10 administrative wards in Mwanza City,classified facilities according to predetermined criteria, and re-corded the number of women working at each facility. An esti-mated 2494 women were working in 953 facilities.By October 2002, weekly community-based reproductive healthclinics had been established within selected guesthouses in eachward. Fieldworkers visited each facility to inform women aboutthe study. Free reproductive health services, including syndromicmanagement of STIs, family planning, health education, and vol-untary HIV counseling and testing (VCT) were provided.Ethical clearance was obtained from the National Medical Re-search Coordinating Committee in Tanzania and the LondonSchool of Hygiene and Tropical Medicine. A trained nurse inter-viewer provided detailed information about the study to women attheir first clinic visit. Written informed consent was obtained fromall participants before enrollment.Participants were asked to attend a study clinic every 3 months,when a study nurse collected demographic and sexual behaviorinformation using a pretested questionnaire in Swahili. Gyneco-logical examination by a trained clinical officer was offered atentry and at 6- and 12-month follow-up visits. Vaginal swabs (forbacterial vaginosis and  Trichomonas vaginalis ) and a cervicalswab (for  Neisseria gonorrheae  and  Chlamydia trachomatis ),urine for pregnancy testing and venous blood for HIV, syphilis,and HSV-2 serology were collected at each visit. STIs weremanaged according to Tanzanian syndromic management guide-lines. 15 Women were encouraged to use the free drop-in reproduc-tive health service at any time between booked appointments.Fieldworkers visited women in their workplace and/or at home1–2 weeks before each scheduled appointment to remind them toattend clinic and conducted up to 3 subsequent weekly visits forwomen who missed an appointment.A community liaison system based on geographical clusters of facilities in each ward was established to promote dialogue be-tween study participants and researchers.  Laboratory Methods All laboratory investigations were conducted at the NationalInstitute for Medical Research (NIMR) Mwanza Center STI Lab-oratory except polymerase chain reaction (PCR) for  T. vaginalis , TABLE 1. Baseline Prevalence of Pregnancy, HIV, and STIs by AgePathogen NPrevalence by Age P  20(n  114)20–24(n  355)25–34(n  710)35  (n  393) All(n  1,572)*Pregnancy 1,560 20 (17.7%) 44 (12.4%) 60 (8.5%) 27 (6.9%) 151 (9.7%) 0.001SerologyHIV 1,568 8 (7.0%) 68 (19.2%) 214 (30.3%) 109 (27.7%) 399 (25.5%)   0.001HSV-2 1,563 40 (35.4%) 213 (60.2%) 571 (81.0%) 342 (87.5%) 1,166 (74.6%)   0.001TPPA    /RPR   1,560 6 (5.3%) 33 (9.3%) 85 (12.1%) 35 (9.0%) 159 (10.2%) 0.08Genital swabs N. gonorrheae  (NG) 1,515 9 (8.3%) 27 (8.0%) 37 (5.4%) 10 (2.6%) 83 (5.5%) 0.01 C. trachomatis  (CT) 1,515 8 (7.4%) 28 (8.3%) 41 (6.0%) 13 (3.4%) 90 (5.9%) 0.04 T. vaginalis † 219 3/16 (18.8%) 7/49 (14.3%) 8/84 (9.5%) 9/70 (12.9%) 27/219 (12.3%) 0.66Bacterial vaginosis ‡ 1,518 49 (45.8%) 196 (58.2%) 377 (54.5%) 176 (46.1%) 798 (52.6%) 0.003Genital symptomsNonmenstrual bleeding 1,566 25 (21.9%) 88 (24.9%) 181 (25.6%) 93 (23.7%) 387 (24.7%) 0.80Dysuria 1,572 24 (21.1%) 80 (22.6%) 150 (21.1%) 81 (20.6%) 335 (21.3%) 0.92Genital itching or burning 1,572 63 (55.3%) 176 (49.7%) 371 (52.3%) 179 (45.4%) 789 (50.2%) 0.11Pain during sex 1,567 31 (27.4%) 109 (30.8%) 231 (32.5%) 102 (26.2%) 473 (30.2%) 0.15 Abnormal genital discharge 1,570 46 (40.4%) 109 (30.9%) 231 (32.5%) 120 (30.5%) 506 (32.2%) 0.23Self-reported genital ulcer/sore 1,566 5 (4.4%) 10 (2.9%) 23 (3.3%) 18 (4.6%) 56 (3.6%) 0.55Other genital symptoms 1,570 21 (18.4%) 65 (18.4%) 179 (25.3%) 111 (28.2%) 376 (24.0%) 0.01Clinical signs Vaginal discharge 1,537 13 (12.2%) 34 (10.0%) 46 (6.6%) 26 (6.7%) 119 (7.7%) 0.06Inguinal lymphadenopathy 1,551 0 (—) 2 (0.6%) 13 (1.9%) 10 (2.6%) 25 (1.6%) 0.09Genital ulcers/blisters 1,535 2 (1.9%) 13 (3.8%) 31 (4.4%) 14 (3.6%) 60 (3.9%) 0.61Genital warts 1,538 2 (1.9%) 10 (3.0%) 16 (2.3%) 2 (0.5%) 30 (2.0%) 0.09Cervical mucosal ulceration 1,523 0 (—) 3 (0.9%) 0 (—) 1 (0.3%) 4 (0.3%) 0.07Cervicitis 1,425 6 (5.9%) 10 (3.2%) 19 (2.9%) 12 (3.3%) 47 (3.3%) 0.47Pain during bimanual exam 1,538 20 (18.9%) 84 (24.8%) 244 (34.7%) 158 (40.5%) 506 (32.9%)   0.001Syndromic diagnosisPelvic inflammatory disease (PID) 1,553 19 (17.3%) 83 (24.1%) 248 (35.2%) 157 (40.0%) 507 (32.7%)   0.001Genital ulcer syndrome (GUS) 1,553 1 (0.9%) 9 (2.6%) 9 (1.3%) 6 (1.5%) 25 (1.6%) 0.39 Vaginal discharge syndrome (VDS) 1,551 61 (55.5%) 160 (46.5%) 281 (39.9%) 148 (37.7%) 650 (41.9%) 0.002*One missing value for age. †  A random sample of 220 posterior fornix vaginal swabs collected at baseline were sent to an external laboratory for  T. vaginalis  PCR. It wasnot possible to process 1/220 specimens. ‡ Using Ison-Hay classification criteria. Vol. 34  ●  No. 9  639 MICROBICIDES DEVELOPMENT PROGRAM, TANZANIA  which was conducted by an external laboratory. Venous blood atenrollment was tested for HIV on both the Murex HIV Ag/Abcombination ELISA (Murex Biotech, Dartford, UK) and the Uni-Form II HIV Ag/Ab microELISA system (BioMe´rieux, France),and for herpes simplex virus type-2 (HSV-2) by the Kalon HSV-2ELISA (Kalon Biologicals, Aldershot, UK), which has been shownto be sensitive and specific in a validation study carried out onAfrican sera. 16 Manufacturer’s kit insert instructions were used todefine cutoff values for positive and negative HSV-2 results on theKalon assay. Optical densities greater than 1.1 were consideredpositive. Syphilis serology was by RPR (Omega Diagnostics,Alva, Scotland, UK) and TPPA (Fujirebio Inc, Tokyo, Japan).Active syphilis was defined as RPR positive/TPPA positive. Urinespecimens were tested for pregnancy using the Quickstick    HCGtest (IPAS, San Diego). Bacterial vaginosis (BV) was determinedby microscopic examination of a Gram-stained slide, using theIson-Hay criteria. 17 Vaginal and cervical swabs were tested byPCR for  N. gonorrheae ,  C. trachomatis  (Roche Diagnostics,Mannheim, Germany), and  T. vaginalis  (Shaio and Kenge PCRmethods). Statistical Methods Data were double-entered and verified. The association of STIprevalence with age was examined using the    2 -test for groupdifferences. Logistic regression was used to examine factors asso-ciated with HIV infection and reattendance after 3 months. Foreach outcome, effects were adjusted for age, clinic site, facility type,and other variables that were statistically significant ( P  0.05). Toexplore the relationship between mobility and reattendance, we con-structed 2 composite variables with 3 levels (low, moderate, high):“travel” based on time away from home and “permanence” based onmoving home or job in the previous year. All analyses were per-formed using Stata 8 (Stata Inc, 2002). Results General Characteristics of the Study Population Of 1573 women recruited into the cohort, 61% were  mamalishe (839/1573; 53%) or worked in pombe shops (115/1573; 7%). Theremaining two-fifths worked in restaurants/bars (335/1573; 21%) TABLE 2. Association of Sociodemographic, Behavioral, and Biological Characteristics With HIVInfection Among 1573 Women Who Attended at BaselineCharacteristicTotalNumber% HIVPositiveOR(95% CI)*OR(95% CI) †  Age (yr) 1,572  P  0.001  P  0.001  20 114 (7.3%) 7.0 1 120–24 355 (22.6%) 19.2 3.14 (1.46–6.75) 2.46 (1.12–5.41)25–34 710 (45.2%) 30.1 5.72 (2.74–11.94) 4.91 (2.29–10.53)35   393 (25.0%) 27.7 5.09 (2.40–10.79) 5.17 (2.34–11.41)Facility type 1,572  P  0.001  P  0.25Mamalishe/pombe 954 (60.7%) 22.4 1 1Guesthouse/hotel 284 (18.1%) 31.7 1.73 (1.28–2.33) 1.29 (0.90–1.84)Bar/restaurant/grocery 335 (21.3%) 28.4 1.53 (1.14–2.05) 1.28 (0.90–1.80)Education 1,573  P  0.07  P  0.17Primary incomplete 376 (23.9%) 25.0 1 1Primary complete 1,032 (65.6%) 26.7 1.02 (0.78–1.35) 1.05 (0.77–1.44)Secondary 130 (8.3%) 16.2 0.55 (0.32–0.93) 0.60 (0.34–1.08)Other 35 (2.2%) 22.9 0.95 (0.41–2.19) 0.74 (0.28–1.94)Marital status 1,573  P  0.001  P  0.001Married 586 (37.3%) 18.8 1 1Separated/divorced/widowed 665 (42.3%) 33.4 2.16 (1.66–2.82) 1.76 (1.30–2.38)Single 322 (20.5%) 20.8 1.58 (1.10–2.27) 1.50 (1.00–2.25)Lifetime sexual partners 1,424 ‡ P  0.001  P  0.001  5 892 (62.6%) 19.1 1 15–10 410 (28.8%) 32.2 1.92 (1.47–2.52) 1.74 (1.31–2.30)10–15 65 (4.6%) 36.9 2.22 (1.30–3.79) 1.84 (1.06–3.19)15   57 (4.0%) 38.6 2.43 (1.38–4.28) 2.00 (1.10–3.62)Partners in past 3 mo 1,565  P  0.05  P  0.490 223 (14.3%) 22.9 1 11 1,016 (64.9%) 24.4 1.09 (0.77–1.55) 1.31 (0.88–1.96)2 226 (14.4%) 29.7 1.51 (0.98–2.35) 1.22 (0.74–2.01)3   100 (6.4%) 32.0 1.78 (1.03–3.06) 1.00 (0.52–1.94)Condom usage with main partnerin past 3 mo1,566  P  0.08  P  0.74No partner 236 (15.1%) 22.9 1 1Never 933 (59.6%) 24.7 1.09 (0.77–1.54) 1.22 (0.81–1.83)Sometimes 263 (16.8%) 27.8 1.45 (0.95–2.21) 1.14 (0.71–1.84) Always 134 (8.6%) 30.6 1.62 (0.99–2.65) 1.30 (0.76–2.23)*OR adjusted for age. † OR adjusted for age, clinic site, facility type, marital status, number of lifetime sexual partners. ‡ Participants (149/1573) responded “not known.” 640  Sexually Transmitted Diseases  ●  September 2007 VALLELY ET AL  or in guesthouses/hotels (284/1573; 18%). The mean age was 29.5years. Women working as  mamalishe  or in pombe shops were onaverage 3.4 years older than women working in other types of facilities (95% CI 2.6, 4.2;  P  0.001).Most women had received some formal education, but 90% hadnot progressed beyond primary-level schooling. Guesthouse/hotelworkers were more likely to have attended secondary education(63/284; 22%) when compared with restaurant/bar workers (27/ 335; 8%) or  mamalishe  /pombe shop workers (40/954; 4%).In all, 37% were married or living with a partner, 42% wereseparated, divorced, or widowed, and 21% remained single.  Ma-malishe  and pombe shop workers were more likely to be married(488/954; 51%) than restaurant/bar workers (55/335; 16%) orwomen working in guesthouses/hotels (43/284; 15%).Study participants were highly mobile: 34% (535/1566) of womenhad lived at their current address for less than 1 year, and 55%(857/1571) had been with their current employer for less than 1 year.  Mamalishe  and pombe shop workers were less likely to have traveledin the 3 months preceding the first clinic visit than women working inother types of facilities and were less likely to have changed theirhome address or place of work during the previous year.  Baseline Prevalence of STIs, Symptoms, and Signs High prevalences of STIs were observed at baseline (Table 1).HSV-2 seroprevalence was 74.6% overall and increased with age( P  0.001). Women aged  24 years had the highest prevalence of gonorrhea and chlamydia and prevalence declined with increasingage ( P  0.08 and  P  0.04, respectively).High prevalences of trichomoniasis (12.3%), active syphilis(10.2%), and bacterial vaginosis (52.6%) were also observed.Overall, 925/1573 (58.8%) of women at baseline had gonorrhea,chlamydia, syphilis, or bacterial vaginosis.Clinical features of STIs were common at entry: 1241/1573(78.9%) reported 1 or more STI symptom and 657/1573 (41.8%)had at least 1 clinical sign. Vaginal discharge syndrome (VDS)was diagnosed in 41.9% of women, pelvic inflammatory disease(PID) in 32.7%, and genital ulcer syndrome (GUS) in 1.6% (Table1). PID increased significantly with age ( P  0.001).The association between presenting symptoms and signs waspoor: Among the 1241 women who reported at least 1 genitalsymptom at baseline, 653 (52.6%) had no clinical signs on exam-ination. Of the 657 women with 1 or more clinical signs, 69(10.5%) were asymptomatic.The proportion of women with 1 or more clinical signs whoreported symptoms varied by pathogen and was 83% (29/35) inwomen with  C. trachomatis , 91% (314/347) in women with bac-terial vaginosis, and 97% (37/38) in women with  N. gonorrheae . Pregnancy Overall, 9.7% of women were pregnant at baseline, with signif-icantly higher prevalence in younger women ( P  0.001; Table 1).  Baseline HIV Prevalence and Association WithSociodemographic and Behavioral Factors HIV prevalence at enrollment was 25.5% and increased from7.0% in women under 20 years to 30.3% in women aged 25–34years ( P   0.001; Table 1). In a regression analysis adjusting forage, no association was found between HIV infection and ethnicgroup, literacy, or use of vaginal inserts to increase or decreasevaginal wetness (data not shown). Age-adjusted associations withHIV infection are presented in Table 2. Women working in guest-houses, hotels, bars, and restaurants were more likely to be HIVsero-positive at baseline compared to women working as  mamal- TABLE 3. Association of HIV Infection With Other STIs Among 1573 Women Who Attended at BaselineCharacteristicTotalNumberPrevalence of STI in:OR(95% CI)*OR(95% CI) † HIVNegative (n  1,174)HIVPositive (n  399)SerologyHSV-2 1,563 794 (68.0%) 372 (93.9%) 6.28 (4.05–9.75) 5.68 (3.63–8.89)TPPA    /RPR   1,560 112 (9.6%) 47 (11.9%) 1.19 (0.83–1.72) 1.02 (0.70–1.50)Genital swabs N. gonorrheae  (NG) 1,515 61 (5.4%) 22 (5.7%) 1.16 (0.69–1.94) 1.04 (0.61–1.75) C. trachomatis  (CT) 1,515 61 (5.4%) 29 (7.5%) 1.53 (0.96–2.44) 1.36 (0.84–2.21)Bacterial vaginosis 1,518 547 (48.4%) 251 (64.7%) 1.99 (1.56–2.54) 1.86 (1.45–2.38)Genital symptomsNonmenstrual bleeding 1,566 277 (23.7%) 110 (27.6%) 1.22 (0.94–1.58) 1.20 (0.92–1.58)Dysuria 1,572 239 (20.4%) 96 (24.1%) 1.25 (0.95–1.65) 1.27 (0.95–1.68)Genital itching or burning 1,572 555 (47.3%) 234 (58.7%) 1.62 (1.28–2.04) 1.61 (1.26–2.04)Dyspareunia 1,567 331 (28.3%) 142 (35.7%) 1.40 (1.10–1.79) 1.52 (1.17–1.98) Abnormal genital discharge 1,570 356 (30.4%) 150 (37.8%) 1.44 (1.13–1.84) 1.43 (1.11–1.84)Self-reported genital ulcer/sore 1,566 26 (2.2%) 30 (7.5%) 3.79 (2.18–6.60) 3.82 (2.16–6.77)Other genital symptoms 1,570 270 (23.0%) 106 (26.6%) 1.15 (0.88–1.49) 1.11 (0.84–1.46)Clinical signs Vaginal discharge 1,537 91 (8.0%) 28 (7.1%) 0.97 (0.62–1.51) 1.07 (0.68–1.69)Inguinal lymphadenopathy 1,551 8 (0.7%) 17 (4.3%) 5.59 (2.39–13.11) 5.50 (2.28–13.25)Genital ulcers/blisters 1,535 29 (2.5%) 31 (7.9%) 3.21 (1.90–5.44) 3.43 (1.97–5.95)Genital warts 1,538 11 (1.0%) 19 (4.8%) 5.82 (2.68–12.64) 5.57 (2.51–12.37)Cervicitis 1,532 34 (3.0%) 13 (3.3%) 1.21 (0.62–2.34) 1.35 (0.68–2.68)Pain during bimanual examination 1,538 358 (31.3%) 148 (37.7%) 1.23 (0.96–1.57) 1.24 (0.96–1.60)*OR adjusted for age. † OR adjusted for age, clinic site, facility type, marital status, and number of lifetime sexual partners. Vol. 34  ●  No. 9  641 MICROBICIDES DEVELOPMENT PROGRAM, TANZANIA  ishe  or in pombe shops (adjusted OR: 1.73; 95% CI 1.28–2.33) asdid those reporting 3 or more sexual partners in the last 3 months(adjusted OR: 1.78; 95% CI 1.03–3.06).These associations became statistically nonsignificant on multivar-iate analysis after which only marital status and lifetime number of sexual partners remained associated with HIV infection (Table 2).  Association of HIV Infection With STI Pathogens and ClinicalFindings Gonorrhea, chlamydia, and syphilis were not significantly asso-ciated with HIV infection at baseline. HSV-2 and bacterial vagi-nosis were associated with HIV, and these associations persistedafter adjusting for age, clinic site, and other factors (Table 3). Factors Associated With Reattendance at 3 Months A total of 1165/1573 (74.1%) women returned for a second visit at3 months. The odds of reattendance were nearly 6 times higher inwomen aged   35 years than those aged   20 years (adjusted OR:5.75; 95% CI 3.37–9.78; Table 4). After age-adjustment, facility type,marital status, HIV status, VCT, and permanence were significantlyassociated with reattendance. Only VCT and permanence remainedsignificantly associated during multivariate analysis. Women whoreceived a negative HIV result were more likely to return for a secondvisit than women who received an HIV-positive result (adjusted OR:2.08; 95% CI 1.47–2.95). Travel in the 3 months before the first visitwas not associated with reattendance, but changes in home and/orfacility address in the previous year were. Conclusions The prevalences of HIV, HSV-2, gonorrhea, and bacterial vagino-sis among women in our study were considerably higher than thoseobserved in the general population and are comparable to prevalencesseen in similar occupational cohorts in Tanzania. 8–10,12 In 2002, HIV prevalence among adults aged 15–49 years wasestimated at 8.8% in Tanzania. 18 HIV prevalence among womenaged 15–54 years in rural communities in Mwanza Region hasbeen estimated at around 4%–5% 11,19,20 and in urban and periurbanareas, 9%–13%. 11,18 STI surveillance data for Tanzania are incom-plete, 18 but data from research cohorts suggest that STIs arecommon among women in rural and periurban communities inMwanza Region. 11,19,20 In women working in bars and hotels in Moshi, 10 HIV preva-lence was 26.3%; HSV-2 52.3%; syphilis (TPPA   /RPR  ) 2.2%;gonorrhea 1.1%; chlamydia 10.0%; trichomoniasis 20.5%; and TABLE 4. Sociodemographic and Behavioral Factors Associated With Reattendance at 3 MonthsCharacteristic Total Number % Attending OR (95% CI)* OR (95% CI) †  Age (yr) 1,572  P  0.001  P  0.001  20 114 51.8 1 120–24 355 63.9 1.65 (1.08–2.53) 1.60 (1.01–2.51)25–34 710 75.5 2.87 (1.91–4.31) 2.62 (1.69–4.07)35   393 87.0 6.25 (3.90–10.01) 5.75 (3.37–9.78)Facility type 1,573  P  0.02  P  0.19Mamalishe/pombe 954 77.7 1 1Guesthouse/hotel 284 71.1 0.81 (0.60–1.10) 0.89 (0.62–1.26)Bar/restaurant/grocery 335 66.3 0.68 (0.51–0.90) 0.74 (0.54–1.02)Marital status 1,573  P  0.02  P  0.31Married 586 79.9 1 1Separated/divorced/widowed 665 73.8 0.71 (0.54–0.93) 0.81 (0.59–1.10)Single 322 64.0 0.70 (0.50–0.97) 0.79 (0.55–1.14)HIV status 1,568  P  0.001  P  0.70HIV negative 1,169 75.7 1 1HIV positive 399 68.9 0.58 (0.45–0.76) 0.91 (0.56–1.48) VCT 1,573  P  0.001  P  0.001Refused VCT 512 72.7 1 1Known HIV negative 802 77.2 1.46 (1.12–1.90) 1.58 (1.19–2.10)Known HIV positive 259 67.2 0.72 (0.51–1.01) 0.76 (0.53–1.09)Travel ‡ 1,567  P  0.07  P  0.17Low 1,036 74.3 1 1Moderate 428 76.2 1.11 (0.85–1.46) 1.19 (0.88–1.60)High 103 64.1 0.64 (0.41–0.99) 0.73 (0.45–1.20)Permanence § 1,568  P  0.01  P  0.05High 603 81.4 1 1Moderate 417 72.9 0.79 (0.58–1.08) 0.79 (0.56–1.10)Low 548 66.8 0.65 (0.49–0.87) 0.67 (0.49–0.92)*OR adjusted for age. † OR adjusted for age, clinic site, facility type, VCT, and permanence. ‡ Travel in the 3 months preceding the first clinic visit was categorized as: low (0 nights away fromhome), moderate (   1 night but less than 1 continuous week away from home), and high (   1continuous week away from home on one or more occasions). § Permanence of home and workplace location was categorized as: high (lived in the same house andworked at the same facility during the past year); moderate (lived or worked at the same location forless than 1 year and moved house no more than once in the previous year); low (lived or worked at thesame location for less than 1 year and moved house twice or more in the previous year). 642  Sexually Transmitted Diseases  ●  September 2007 VALLELY ET AL
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks