                   AIDS INFORMATION NEWSLETTER
                   Michael Howe, MSLS, Editor
                     AIDS Information Center
                VA Medical Center, San Francisco
                     (415) 221-4810 ext 3305
                        December 16, 1994

                Women and HIV Infection (Part XV)

            Health Care Delivery, Zidovudine Use, and
              Survival of Women and Men with AIDS.
            by BJ Turner, LE Markson, LJ Mckee et al

Abstract:  The aim of this study was to define predictors of
survival for women and men after AIDS diagnosis.  We examined
health care delivery and drug therapy in the year before AIDS
diagnosis for continuously enrolled New York State Medicaid
beneficiaries with AIDS in 1988-1990.  We examined the association
of these factors with survival after AIDS diagnosis.  Of 1,077
women and 1,871 men, 60% of both gender groups were drug users. 
In both risk groups, women had more outpatient visits than men but
were equally likely to visit an AIDS specialist.  In those who were
not drug users, men were twice as likely as women to receive either
zidovudine or Pneumocystis carinii pneumonia prophylaxis.  No
difference appeared among drug users.  Survival after AIDS
diagnosis was similar by gender for those who were not durg users
(RR = 1.09;95% CI = 0.90-1.33).  In drug users, women had a
slightly lower risk of death than men (RR = 0.84;95% CI = 0.72-
0.98).  Risk of death after AIDS diagnosis was higher for persons
starting zidovudine earlier in both risk gorups.  Among drug users,
women received more ambulatory care and survived slighlty longer
than men.  Among those who were not drug users, survival was
similar by gender even after adjusting for differences in care.
(Journal of Acquired Immune Deficinecy Syndromes. 1994;7:1250-62.)

       Survival For Women and Men With AIDS, San Francisco
                            1981-1990

INTRODUCTION:  Previous  studies have suggested that survival
following diagnosis of acquired immunodeficiency syndrome (AIDS)
may be shorter for women than men [1-4].  However, no studies to
date have examined the impact of therapy on temporal trends in
survival by gender.  To answer this questions, we compared survival
trends for women and men diagnosed with AIDS in San Francisco, and
examined the relative effect of antiviral therapy and other
factors, on differences in survival by gender.

METHODS:  Survival following diagnosis of AIDS was calculated based
upon data from 139 women and 7045 men diagnosed with AIDS in San
Francisco between January 1, 1981 and December 21, 1990.  Patients
were followed up prospectively at six-month intervals through May
15, 1991.  Patients initially diagnosed with Kaposi's sarcoma only
were excluded since the disease was rarely diagnosed among women
and its inclusion would bias the results towards longer survival
for men relative to women.

Viral status, subsequent diagnoses, and history of antiviral
therapy or Pneumocystis carinii pneumonia prophylaxis was
ascertained for each patient through prospective follow-up, which
included reviewing medical records and/or contacting physicians at
approximately 6-month intervals.

All analyses assessing the effect of antiretroviral therapy on
survival were based on data from patients who were diagnosed
between January 1, 1987 and December 31, 1990.  This time period
was chosen because antiretroviral therapy first became widely
available in San Francisco in 1987.  Information on therapy was
available for 2,4999 (49.7%) of 5,033 patients with AIDS diagnosed
between 1987 and 1990, including 54.4% of women and 49.5% of men.

Median survival and the cumulative probability of survival at one
year after diagnosis were calculated using the Kaplan-Meier
product-limit method (P1L procedure within the BMDP statistical
software package) [5].

Cox proportional hazards regression analysis (P2L procedure [BMDP])
[6] was used to construct a proportional hazards model for
predicting length of survival and to identify the independent
predictors of survival among those variables studied.  The model
was constructed using a stepwise procedure (maximum partial
likelihood ratio method with a P value of .10 as the entry
criterion and of .15 as the removal criterion at each step.

RESULTS:  Survival following diagnosis for women with AIDS was
significantly shorter than that for men (see Table 1 in print
copy).  When stratified by year of diagnosis, survival improved
over time for both women and men, although survival for women
remained significantly shorter than that for men.  The observed
differences in survival by gender were unaffected when persons
diagnosed at death were excluded from the analysis.

Among women, older age (median survival of 3.2 months for those
aged 60 and older versus 20 months for those aged 20 to 29) and
infection via transfusion (median survival of 3.6 months versus
12.8 and 17.1 months for patients infected via injection drug use
and heterosexual contact, respectively ) were associated with
decreased survival.  Survival did not differ significantly by race
or ethnicity.

Use of either zidovudine (ACT) or dideoxyinosine (ddI) was
associated with increased survival for both women (P < ,005) and
men (P < .001).  Among those who used either zidovudine or ddI,
survival did not differ for men and women.  However, men who did
not use zidovudine or ddI survived longer than did women in the
same category.

We controlled for the potential effects of confounders on survival
in a proportional hazards model.  The patients' initial diagnoses
were the best independent predictors of survival, followed by the
patients' age, whether they received therapy with zidovudine or
ddI, version of the case definition, and acquiring HIV infection
through a blood transfusion.  Gender, race or ethnicity, and year
of diagnosis were not significant predictors and were excluded by
the model.  This model suggests that the shorter survival observed
for women is due to factors other than gender.

The proportional hazards model also suggests that improved survival
for both women and men was associated with younger age, use of
zidovudine or ddI, an initial diagnosis of Pneumocystis carinii
pneumonia only, and an initial diagnosis of one opportunistic
infection only.  Survival was shorter for transfusion recipients
and for patients diagnosed under the 1987 revision of the case
definition.  After controlling for therapy, year of diagnosis was
excluded from the final model due to a lack of significance.

EDITORIAL NOTE:  This study indicates that survival following
diagnosis of AIDS has improved for both men and women in recent
years, although survival for women remains significantly shorter
than that for men.  The data also suggest that therapy may be
partially responsible for these recent improvements in survival.

Recent studies have suggested that women with AIDS may have a lower
use of health care resources [7] and less use of zidovudine therapy
[4] when compared with men.  In our study, persons using zidovudine
or ddI were more likely to be white, homosexual or bisexual men,
and were less likely to be women.  Therefore, decreased access to
(and/or lack of utilization of) antiretroviral therapy may partly
explain the shorter survival observed for women.

Other factors such as older age, other competing health risks and
diagnosis later in the course of HIV infection may also partly
explain the shorter survival observed for women.  Because of the
limited clinical and laboratory information available, it is
uncertain whether the shorter survival in women could also be
attributed to diagnosis later in the course of HIV infection.  A
later diagnosis among women could reflect poorer access to care or
a lack of awareness regarding HIV status.  In addition, the
observed temporal improvements in survival for women and men could
reflect diagnosis earlier in the course of HIV infection.  However,
other studies which controlled for CD4 cell counts at the time of
diagnosis demonstrated benefits from therapy [8] and recent
improvements in survival [9].

Because these data are observational in nature, they should not be
used to evaluate the clinical safety and efficacy of drug therapy,
which is more properly evaluated in carefully controlled clinical
trials.  Instead, this study may be used as confirmatory evidence
of longer survival for persons utilizing antiviral therapy.

By 1987, HIV/AIDS had become the eighth leading cause of death
among women or reproductive age in the United States.  By 1991,
HIV/AIDS was expected to become one of the five leading causes of
death among women of reproductive age [10].  Despite these trends,
the natural history of HIV infection in women has not been well
characterized.  Relatively few women have been included in clinical
drug trials or in epidemiologic cohort studies.  Recent reports
have suggested that these concerns are part of a broader problem
of inadequate research, diagnosis and treatment for women's health
care problems [11].

Recent advances in therapy provide hope that survival will continue
to improve for both men and women with AIDS.  Providing access for
both women and men to HIV testing and counseling, and early
monitoring and medical intervention is needed to help bridge the
gender gap in survival following a diagnosis of AIDS.

References

1.  Rothenberg R. Woelfel M. Stoneburner R. et al. Survival with
the acquired immunodeficiency syndrome: experience with 5833 cases
in New York City. NEJM. 1987;317:1297-1302.

2.  Whyte BM. Swanson CE. Cooper DA. Survival of patients with the
acquired immunodeficiency syndrome in Australia. Med J Aust.
1989;150:358-62.

3.  Friedland GH. Saltzman B. Vileno J. et al. Survival differences
in patients with AIDS. J Acq Immune Defic Syndr. 1991;4:144-53.

4.  Moore RD. Hidalgo J. Sugland BW. Zidovudine and the natural
history of the acquired immunodeficiency syndrome. NEJM.
1991;324:1412-16.

5.  Benedetti J. Yuen K. Young L. Life tables and survival
functions. In: Dixon WJ, ed. BMDP Statistical Software. Berkeley,
CA: University of California Press, 1985:557-75.

6.  Hopkins A. Survival analysis with covariates:  Cox models. In:
Dixon WJ, ed. BMDP Statistical Software. Berkeley, CA: University
of California Press, 1985:576-94.

7.  Turner BJ. Markson LE. McKee L. et al. Survival patterns of
women and men with AIDS: Impact of health care use prior to AIDS
[abstract TuD112].  In: Program and abstracts of the Seventh
International Conference on AIDS (Florence, Italy). Rome, Italy:
Istituto Superiore di Sanita, 1991.

8.  Horsburgh RC. Handon D. Fann SA. et al. Predictors of survival
in HIV infection include CD4+ cell count, AIDS defining condition
and therapy but not sex, age, race, or risk activity [abstract
MD3175].  In: Program and abstracts of the Seventh International
Conference on AIDS (Florence, Italy). Rome, Italy: Istituto
Superiore di Sanita, 1991. 

9.  Pedersen C. Gerstoft J. Tauris P. et al. Trends in survival of
Danish AIDS patients from 1981 to 1989. AIDS. 1990;4:1111-16.

10. Chu SY. Buehler JW. Berkelman RL. Impact of the human
immunodeficiency virus epidemic on mortality in women of
reproductive age, United States. JAMA. 1990;264:225-9.

11. Council on Ethical and Judicial Affairs, American Medical
Association.  Gender disparities in clinical decision making. JAMA.
1991;266:559-62.

[Note:  Tables included in print copy.]

[Reported by: Surveillance Branch, AIDS Office, San Francisco
Department of Public Health. San Francisco Epidemiologic Bulletin,
Vol. 8, No. 12, December 1992.  PP. 47-49.  For a more detailed
discussion of this study, see:  Lemp GF. Hirozawa AM. Cohen JB. et
al. Survival For Women and Men with AIDS.  Journal of Infectious
Diseases. 1992;166:74-9.  See also: Von Overbeck J. Egger M. Smith
GD. et al. Survival in HIV Infection: Do Sex and Category of
Transmission Matter? AIDS. 1994;8:1307-13.]

                       Disease Progression

(From: Centers for Disease Control and Prevention. Training
Bulletin #37. March 25, 1993.)

     QUESTION:  NIH training materials currently include the
     following statement on disease progression:

     'Studies have now shown that in the United States and Europe
     approximately 50 percent of men with HIV infection develop
     AIDS within 10 years of infection.  CDC currently states that
     over time most HIV-infected persons will eventually develop
     AIDS. 
 
     Unfortunately, such information is primarily limited to
     studies on men ... However, other natural history studies have
     incorporated both women and men.  For example, findings from
     a study of persons infected with HIV by transfusion did not
     indicate any overall difference in disease progression between
     men and women.' 
 
     In published proceedings from the December 1990 U.S. Public
     Health Service conference on Women and HIV Infection, however,
     it was reported that women 'are too often diagnosed in the
     later stages of their infection.'  Present data indicate that 
     women are too frequently diagnosed with AIDS at or near the 
     time of death.  According to Dr. Ruth Berkelman, M.D., [then]
     Chief of the Surveillance Branch, Division of HIV/AIDS,
     Centers for Disease Control, 'In 1990, 11% of women were
     diagnosed with AIDS the same month they died.  This compares
     to approximately 8.5% of men who reported homosexual contact
     who were diagnosed with AIDS the same month they died.'

     Additionally, the issue of shorter survival time for women is
     a subject of ongoing discussion in the medical literature,
     whether this disparity is attributed to biological differences
     or to medical or socioeconomic factors including misdiagnosis, 
     later diagnosis, poverty and subsequent lack of access to
     medical services, inaccessibility of experimental treatments,
     etc. 
 
     Will you please comment on these issues, and on CDC's current
     position regarding relative survival time of women and men
     with HIV disease?"

     Answer:  CDC is continuing to examine the full range of
     conditions found in HIV-infected women.  CDC has supported a
     natural history study of HIV-infected injecting drug users,
     approximately 1/3 of whom are women.  In addition, four sites
     have been funded to conduct natural history studies in HIV-
     infected women.  These studies are being conducted with the
     PHS collaborating group for study of HIV infection in women.
     
     These natural history studies continue to examine gynecologic
     problems in HIV-infected women, including the possible
     increased likelihood of certain gynecologic cancers; the
     frequency and severity of infections (including pelvic
     inflammatory disease, pneumonias, and candidal infections of
     the vagina, oral cavity, and esophagus); and survival after
     infection with HIV and after AIDS diagnosis.

     Additional studies are being carried out to determine biologic
     factors in the heterosexual transmission of HIV, as well as
     to help determine: 1) behavioral risk factors for acquisition
     of HIV; 2) biologic risk factors for HIV infection, such as
     genital ulcer disease and hormonal and other factors; 3) early
     course of HIV after infection; and 4) the impact of counseling
     on behaviors in women. 
 
     The effect of social factors such as access to health care,
     cultural values and attitudes, family structure, and social-
     support systems, on the natural history of HIV disease in
     women is also being examined.

     As mentioned in the first paragraph of this response, some of
     the studies on the natural history of HIV infection/AIDS in
     women involve survival analyses.  Both women and heterosexual
     men with AIDS survive, on average, for a shorter period of
     time than men with AIDS who report sexual contact with other
     men, possibly reflecting the impact of other factors, such as
     injecting drug use or low socioeconomic status.  Survival
     following diagnosis of AIDS has improved in recent years
     partly due to better and available treatments.  The medical
     community needs to be more aware of medical concerns
     associated with HIV infection and AIDS in women so that
     critical early intervention can be instituted. 
