Abstracts Vol. 1 [International Conference on AIDS (11th: 1996: Vancouver, Canada)]

Monday, July 8, 1996

Mo.C.570 - Mo.D.131

Mo.C.570
CORE GROUPS CAUSE PRIMARY INFECTION TO DOMINATE HIV TRANSMISSION
EVEN WHEN MORE THAN 90% OF VIRUS IS EXCRETED DURING LATER STAGES
OF INFECTION
Koopman ames S, Welch G, Jacquez JA, Simon CP Lange K, Pollock 5, Foxnian B. University
of Michigan, Ann Arbor MlI, USA
Objective: To evaluate how core group structure affects the proportion of viral excretion during primary infection which causes primary infection to dominate HIV transmission dynamics.
Methods:-The role of primary and late stages of HIV infection on transmission dynamics is a
function of both: I) how much virus is made available for transmission at different stages;
and 2) how transmissions during different stages are connected into chains of infection.The
way people enter and leave core groups and the rates at which partnerships are formed
inside and out of core groups markedly affects 2).To assess this effect, HIV transmission in
age structured homosexual populations was modeled with partnership formation rates and
the frequency of core group sexual behaviors rising then falling with age and with preferential mixing within both age and core groups.The comparative effects of eliminating transmission during early and late stages of infection was examined at specified rates of contact
within and outside of core groups, rates of entering and leaving core groups, and transmission probabilities during different stages.
Results: Given active core groups with high turnover rates, eliminating transmission during
primary infection stopped epidemics when less than 5% of virus was excreted during primary infection. As core groups became more like the majority population, primary infection
dominated transmission only at higher levels of viral excretion during primary infection.
Eliminating transmission from late stage infection did not stop epidemics unless core groups
accounted for a very small fraction of all partnerships.
Conclusions: Core group conformation affects the way transmissions during different stages
of infection are connected into chains. Highly likely core group conformations allow transmissions during primary infection to dominate the spread of infection even when the
amount of virus excreted during primary infection is only a small fraction of all virus excreted. Consequently, vaccines which fail to prevent infection but which reduce or eliminate
contagiousness during primary infection will, in most situations, stop the HIV epidemic.
Vaccine trials with inexpensive and feasible modifications can test for this effect.
J. S. Koopman MD MPH, Dept. of Epidemiology SPH-I, 109 Observatory, Ann Arbor, MIl
48 109-2029, (e-mail) [email protected], (tel) 313-763-5629
Mo.C.571
RISK OF HIV TRANSMISSION DURING THE SEROCONVERSION VERSUS THE
POST-SEROCONVERSION PERIOD
Duerr Ann*, Mundee Y**, Flowers L*, Xia J***, Kamtorn N**, Prasertwitayakij W**,
Nantachit N**,Vannareumol P**, Jeeyapant 5**, Nelson KE****. *Centers for Disease
Control and Prevention, Atlanta, GA, USA,** Chiang Mai University Chiang Mai,
Thailand,**Mayo Clinic, Rochester, MN, USA,****Johns Hopkins Univ, Baltimore, MD, USA
Objective: Plasma viremia is very high shortly after infection with HIV. Whether this represents a time of high risk of sexual transmission was investigated in a study of HIV-infected
Thai blood donors and their wives.
Methods: Couples were selected from an ongoing study of HIV transmission from HIVinfected (HIV+) male blood donors to their wives. Premarital sex was very rare; women
with HIV risk other than sex with the index case were excluded. Males in this cohort with
documented seroconversion dates after marriage were identified retrospectively from
blood bank records.The seropositivity rate among the wives of these men (SC men) was
compared to that among a group of women who married men who were HIV-positive at
the time of marriage (PSC men); wives of PSC men were therefore not exposed during the
seroconversion period. HIV status of wives was assessed by ELISA, Western blot and p24
antigen at enrollment; seronegativity was confi rmed at 6 and 12 months.
Results: Matching of study data to computerized blood bank records since 1988 identified
15 men who seroconverted after marriage.The date of male seroconversion was estimated
as the midpoint between the last HIV-negative donation and the first HIV-positive donation;
median exposure of wives of SC men from this point until enrollment was 7 months (range
3 - 42). 3/I 5 (20%) of wives were HIV+ at enrollment. Among the wives of PSC men the
median exposure period was 1 6 months (range 2 - 63). 31/85 were HIV+; the odds ratio
(OR) for HIV-positivity in wives of SC men as compared to wives of PSC men was 0.44
(.09, 1.85: P0.25). Multiple logistic regression was used to adjust for husbands' CD4I count
at enrollment and number of sexual contacts with husband during his period of seropositivity. Adjusted OR for wives of SC men compared to PSC men =0.31(0.06, 1.49; P=D0.14).
Conclusions:This study ofThai blood donors and their regular sexual partners failed to
show an increased risk of HIV infection among women exposed during the soneroconversion
period. A high rate of infection was seen among women whose only exposure occurred in
the post-seroconversion period.
Ann Duerr; 4770 Buford Hwy, NE, MS:K-34, Atlanta, GA 3034 I, U.S.A.
Telephone:[phone]; Fax:[phone] E-mail: axd9@ccddrh I.em.cdc.gov
Mo.C.572
HIV INFECTIVITY BY CONTRACEPTIVE METHOD FROM A PARTNER STUDY IN
NORTHERN THAILAND
Nagachinta, Tippasvan*, DuerrA A*, Gargiullo PM" ',YutabutraY Y-**Wannarats A* *Tovanabucra 5S**, Sennan S**m*, Suriyanon V*.*,de Boer M'**, Celentano DO**
Nelson KE.***. *CONRAD, Atlanta, GA, USA,*'*CDC, Atlanta, GA, USA,'***johes Hopkins
Univ, Baltimore, MD, USA,**s**Chiang Mai Univ, Chiang MaiThailand
Objective: To measure HIV male to female infectivity (probability of transmission per intercourse) associated with use of different contraceptive methods or pi-egnancy in an ongoing
partner study in Chiang MaiThailand.
Methods: The study population includes HIV-positive blood dono-s and their regular fensale
sexual partners. Female partners had no risk factors for HIV other than sexual exposure to
the index male. 92 female partners (30 HIV+, 62 HIV-) with defined exposure periods
were included in this analysis, specifically those with partners who tested HIV-positive prior
to marriage or had known seroconversion dates. Data collected includes sexual history, S
year contraceptive history, physical exam including pelvic, laboratoiy testing for STO, HIV
(ELISA with Western blot ronfirmation) and imnsunophenotyping. Condom and other barni
er users (who were rare) were excluded from this analysis.The probability model used in
analysis assumed an independent chance of infection per intercourse, and infectivities were
estimated by maximum likelihood.
Results: HIV-positive women did not differ from HIV-negative women in terms of age, education, duration of relationship, pregnancy type and frequency of sexual contact with male
index case. 78% used oral contraceptives (OC) some time during the study period, 26%
used DMPA (not mutually exclusive). Infectivity did not differ significantly between pregnant
women and non-pregnant women using either OCs, DMPA or non-hormonal/no method.
Estimated infectivity for the OC group was.0019 ~+.00I2 per intercourse (95%CI).
Estimated infectivity for the DMPA exposures was.002 I ~.0038. Infectivity for the nonhormonal/no method exposures was.0023 +.0022; that for exposures among pregnant
women was.00 17 +.0033. Estimated infectivity did not differ significantly by male
CD4counts within any group. However the infectivity was slightly higher when male
CD4counts were <200 cells compared to when CD4 counts were 200 cells (P=.77).
Conclusions: Analysis of this well-defined, albeit small, data set did not demonstrate statistically significant increases in infectivity associated with hormonal contraceptives.This suggests
that increased HIV seropositivity among contraceptors found in studies with undefined
exposure characteristics be interpreted cautiously
Tippavan Nagachinta, 4770 Buford Hwy NE, MS:K 34, Atlanta, GA 3034 I, U.S.A.
Telephone:[phone]; Fax:[phone]: E-mail: txn3@ccddrh I.em.cdc.gov
Mo.C.573
HIV INFECTIVITY: INFORMATION FROM EPIDEMIOLOGICAL STUDIES OF
HETEROSEXUAL TRANSMISSION
Shiboski, Stephen, Padian, Nancy University of California San Francisco, San Francisco, USA
Objective: To investigate the utility of epidemiological studies of heterosexual HIV transmission in providing quantitative information about properties of the infectivity, or per contact
infection risk. Also, to present methods for analysis of data from studies of HIV transmission
which make appropriate use of exposure information.
Methods: Data from three CDC-sponsored studies of transmission in partners of previously infected individuals are used to estimate infectivity: two based on partners of individuals
infected via transfusion, and one consisting of partners of individuals infected from a variety
of sources. Newly developed statistical methods which account for the incomplete nature
of the exposure data from these studies are used to estimate the infectivity and associated
estimation error to investigate effects of behavioral risk factors and to assess evidence for
heterogeneity of infectivity across partnerships. Infectivity estimates from other studies and
biological data on HIV infectivity are also reviewed.
Results: Estimates of the constant per-contact infectivity from the three studies are similar,
ranging from 0.0009 to 0.002 with overlapping 95% confidence intervals. Additional nonparametric estimates show some evidence for variation with time following infection of the
primary infected partner, and for heterogeneity across partnerships. Due to the incompleteness of the exposure data and to large uncertainty in the estimates, little can be concluded
about how the infectivity changes with time. In particular, no information is provided about
change in the period immediately following infection of the primary infected partner.
Conclusions: Although epidemiological studies provide useful information on risk factors for
transmission, they do not yield reliable estimates of HIV infectivity unless it is assumed to be
constant over time and across partnerships.These limitations are largely due to the inadequacy of available exposure data. More complete exposure information and additional data
on biological properties of infectiousness and susceptibility are needed for better estimates.
Stephen Shiboski, Department of Epidemiology & Biostatistics, Box 0560, UCSF: San
Francisco, CA 94 I143-0560.Tel: (4 I15)476-0533, fax:476-60 I14, email: [email protected]>
Mo.D. 130
ATTITUDES AND TRAINING OF HEALTH CARE WORKERS ON AIDS
IN HONG KONG
Lau, IT F, Lee S S. Scientific Committee on AIDS, Government of Hong Kong
Objectives: To understand AIDS-related knowledge/attitudes among health care workers
(HCW) in Hong Kong; to study factors affecting avoidance of treating AIDS patients.
Methods: The sampling frame consisted of the membership lists of the relevant professional
bodies. Self-administered questionnaires were sent to 20,800 health care workers (HCW)
in 1995.The final sample (n - 4523) composed of 1 174 doctors, 1949 nurses, 689 medical
laboratory technicians, 19I occupational therapists, 247 radiotherapists, 2 16 physiotherapists, 57 optometrists.
Results: (I) The HCW had moderate knowledge level on the more technical aspects of
AIDS-related matters. Less than 50% knew about the risk for a vertical transmission or the
median length for progression to AIDS. (2) Only 27% of the respondents felt that they have
adequate skill/knowledge to take care of HIV patients. 13% believed that they had attained
adequate knowledge on AIDS. (3) 20% stated that they are reluctant to take care of
HIV/AIDS patients. 32% would be bothered by an HIV infected co-worker (4) 35%
believed that they might be infected through caring of patients. Factors associated with low
level of avoidance included low susceptibility, experience with AIDS patients, knowledge
about transmission of AIDS, endorsement of professional duties and confidece in treating
AIDS patients. Certain groups of HCW are also more likely to avoid patients.
Conclusion: In general, the HCW in Hong Kong feel inronfident to manage AIDS patients.
About 25% of the workfoirce showed some tendency to avoid AIDS patients, and it is seen
that knowledge on AIDS related matters is important to remove the fear to take care of
AIDS patients.
Joseph Lau T F, Centie for Clinical Trials and Epidemiological Research, I I/F, Prince of Wales
Hospital, Shatin, Hong KongTel: 852-2637-6606 Fax:[phone] Email: [email protected]
Mo.D.13I
THE EXPERIENCE OF CREATING A WORKING COOPERATIVE SOCIETY
FORMED BY PEOPLE LIVING WITH HIV
Anania, Serio. PROMUSIDA (Municipal Aids Program), Rosari, Argentina
Issue: The Carpe Diem Coopeiative Society is an alternative way of facing limitations and
labor discrimination suffered by people living with HIV
Project: Two years ago we were hired by the PROMUSIDA (Municipal Program against
AIDS), because of our condition of people living with HIV, to take part in prevention work

IL)
0
v
c
nO
c0
r)
V
C
Q)
<)
IO
0
cn_
0
CO
C
r)
c
4
42