Bridging the Gap: Conference Record [Abstract book, International Conference on AIDS (12th: 1998: Geneva, Switzerland)]

12th World AIDS Conference

Abstracts 42446-42450 863

Results: A total of 349 men and 158 women were selected during this period.
We observed an increasing number of patient consultations over time. Median
age was 36 years for men and 30 years for women. Among the 507 patients, 21%
came for only one consultation, and thus never received their HIV result. The
median number of consultations was three, and the median length of follow-up
was 2 months. Only 10% of the patients have had regular consultations, defined
as having at least one consultation per month. The main reasons for consultations
were: diarrhoea (29%), malaria (15%), dermatitis (14%), pneumonia (14%), oral
candidiasis (8%) and ENT infection (4%). Half of the patients live more than 15
km from our Unit. The average cost of transport to our Unit per patient was 1500
F cfa (US$ 3). Furthermore, the cost of a ten-day treatment for an episode of
diarrhoea (including IV rehydration at our day care unit) was 6000 F cfa (US$ 12)
on average, not affordable by all patients.
Conclusion: In our experience, only a small proportion of patients are regularly
followed up, making difficult a personalised care approach. We envision promoting home care services to reduce the global cost of health care and to reduce the
number of those patients lost to follow-up.
42446 Costs of care for HIV-positive and HIV-negative adult
patients in Kenyatta National Hospital (KNH), Nairobi,
Kenya in 1997
Charles Gilks', L. Guinness', G. Arthur2, S.M. Bhatt2. 1Liverpool School Tropical
Medicine Liverpool, UK; 2Medicine, University of Nairobi Nairobi, Kenya
Objective: To evaluate the costs and cost-effectiveness of hospital care for adult
medical patients admitted to KNH, Nairobi in 1997 as part of a longitudinal study
of the impact of HIV/AIDS on hospital services.
Methods: Consecutive patients, acutely admitted to the medical wards were
enrolled. Full clinical information and data on in-patient costs [drugs, ancillary
services, ward staff, non-curative/hotel services] were collected and analysed
by diagnosis, HIV status and outcome [improved; transferred to TB services;
worse/died] to assess the cost-effectiveness of care. Care bottlenecks and financing mechanisms were also reviewed.
Results: 518 patients in 15 24-hour take periods over 4 months were enrolled.
207 (40%) were HIV-1 positive and bed occupancy was 187%. Data comparing
1997 with studies in 1988/89 & 1992 are presented elsewhere. Length of stay
and costs per admission were similar for HIV positive and negative patients. The
mean length of stay was 9.7 days, and the mean cost per admission was Ksh
8253 ($150.88). Hotel and ward staff costs were the main items with hospital drug
costs (excluding TB drugs purchased by TB programme) representing 10-12%.
There was no significant difference between the mean outcome costs for HIV
positive and negative patients who improved or died; transferred patients incurred
significantly more costs (150%). Delayed departure was noted in 115 (22%) of
admissions: patients were not collected by relatives (52%) or experienced difficulties in settling bills (20%). Such bottlenecks constituted about 20% of total costs.
Overall 62% of costs were born by the hospital, 33% by the patient but only 5%
by NHIF a national health insurance scheme.
Conclusions: As the HIV/AIDS epidemic has matured in Kenya, the costs and
effectiveness of caring for patients with HIV-related diseases are now similar to
HIV-negative patients in KNH a large urban acute care facility. Important care
bottlenecks are identified which if rectified will increase care capacity. The rate
of cost recovery is low, especially through NHIF. The improvement of existing
or alternative financing mechanisms should be explored. Wider use of user fees
may increase bottlenecks with little impact on finances.
42447 An HIV/AIDS ambulatory care model for optimal patient
care in Bangkok, Thailand
Achara Chaovavanich1, Somsit Tansuphaswadikul1, S. Siripataradural,
P. Ruansawang', Y. Inthong', V. Furner2, J. Gold2. 1Bamrasnaradura Hospital,
Tiwanont Road, Nontaburi, Bangkok, Thailand; 2Albion Street Centre, Sydney,
Australia
Background: Thailand has been at the forefront of the HIV/AIDS epidemic in
south east Asia, and the focus on the many successful prevention strategies
has now shifted to addressing the burden of care for people with HIV/AIDS,
within hospital and community based health service settings. Successful models
of patient care are now an important focus.
Methods: Prospective data over 18 months, was collected on all patients presenting to the ambulatory care service of the Bamrasnaradura Hospital, Bangkok,
a public infectious diseases hospital with a significant burden of care within an
urban setting. Data included demography, patient visits, admission rate, procedures performed and diagnoses. Care was provided by nurse practitioners and
doctors within an ambulatory care service.
Results: A total of 2,138 patients (87% male) attended for care, with an average
of 1.7 visits per person. The admission rate was 16% (617 patients). A number
of procedures were carried out including lumbar punctures (1,666), infusions
(802) and blood transfusions (66) by nurse practitioners. The major presenting
diagnoses were Cryptococcosis (1,404 patients), and TB (758), oral candidiasis

(386) and PCP (36).
Conclusions: This model of care enabled a large number of patients to receive appropriate medical and nursing care with minimal hospitalisations. The
multidisciplinary approach to care in this setting will be outlined emphasising the
advantages of this model in a resource constrained country.

42448 Surgical complications in HIV infected patients
Sian Jones1, C. Smith', D.N. Rose2. 1Mount Sinai Medical Center, New York.
NY; 2Long ISaInd Jewish Medical Center, New York, NY, USA
Objectives: To examine the variables associated with surgical complications in
HIV infected patients.
Design: Retrospective chart review
Methods: Charts were reviewed on all patients who underwent major surgical
procedures and also carried a diagnosis of HIV infection at Mount Sinai Medical
Center between 1990 and 1995.
Results: Fifty eight patients were evaluated. The median age was 39 and
the median CD4 count was 85. All but 3 procedures were performed under
general anesthesia, and 31 patients underwent surgery on an emergent basis.
Fifteen (26%) patients sustained 20 complications, most (70%) were infectious
in nature. Two deaths occurred from sepsis and/or renal failure. Among the 16
variables studied, only serum albumin - 2.5 mg/dl, creatinine -2.0 mg/dl and CD4
count - 50 were significantly associated with postoperative complications. In a
multiple regression analysis only serum albumin independently predicted risk for
surgical-complication.
Conclusions: Postoperative complications of major surgery in patients with
advanced HIV disease and severe surgical illness are most strongly associated
with a low serum albumin level.
42449 Dual epidemics of HIV and injection drug use lead to
shorter survival for first nations in Vancouver, Canada
Irene L. Goldstone1, L. Demerais2, R.S. Hogg', T. Perry1, R. Albert3,
S. Markowski3, A. McLeod'13. 1BC Centre for Excellence in HIV/AIDS, 608-1081
Burrard St., Vancouver; 2 Vancouver Native Health Society, Vancouver; "St.
Paul's Hospital, Vancouver, BC, Canada
Issues: First Nations persons represent 4% of the population of British Columbia.
However, in 1997 they represented 13.6% (62) of new HIV diagnoses and 19%
(24) of HIV/AIDS deaths reported to St. Paul's Hospital's AIDS Care Team (SPH)
and Vancouver Native Health Society (total deaths = 127).
Project: Data were collected from a chart review of all HIV/AIDS deaths at SPH
in 1997 and the BC Centre's Treatment Program for an analysis and comparison
of demographic characteristics and utilization patterns in First Nations persons
and others.
Results: Of 62 deaths at SPH, 10 (16%) were First Nations (6 M, median age
36; 3 F, median age 45; 1 transgendered, 27) vs. 52 (49 M, median age 43; 6
women, median age 39). Transmission was attributed to injection drug use in 8
(80%) of First Nations vs 16 (35%); AIDS was diagnosed in 8 (80%) vs 37 (71%).

Survival days, median, HIV dx to death
Survival days, median, AIDS to death
Last reported CD4 count, median
Last reported viral load, median
Ever received antiretroviral therapy, median
Hospital admissions, median, HIV dx to death
Hospital admissions, median, AIDS to death
Hospital days, median, HIV dx to death
Hospital days, median, AIDS to death
Palliative care days, median, HIV dx to death

First Nations n = 10
854 (189-4758)
95 (2-259)
50 (20-220)
32,000
(22,000-54.000)
7 (70%1
5 (4-12)
3(1-9)
46 (2-123)
24 (270))
3 (3-3)2/10

Other n = 52
1672 (30-5488)
463 (3-2255)
40 (10-490)
16,000
(20,000-62,000)
41 (79%)
3 (1-12)
3(1-8)
27 (2-130)
27(3-126)
11.5 (1-67) 22/52

Lessons Learned: First Nations persons were sicker earlier, with shorter
survival, more admissions and hospital days and less palliative care than nonaboriginal persons. Improved access to primary care and support for adherence
to care and therapies earlier in the course of HIV is urgently needed to enhance
the quality of living and dying. As the epidemic in First Nations evolves, this profile
will provide a benchmark to measure future outcomes of care.
42450 Changes in EEIV hospitalizations since the introduction
of multidrug therapies at a local tertiary HIV referral
centre - Impact on hospital and community resources
Charles Chan', Meaghen Hyland2, G. Bailey3, R.H. Hyland3, A. Phillips3,
C.K.N. Chan2. 1200 Elizabeth St., Iden 220 Toronto Ontario M5G2C4; 2The
Toronto Hospital Toronto ON; 3 Wellesley-Central Hospital Toronto ON, Canada
Objectives: To review the pattern of HIV hospitalization at a tertiary HIV referral
centre for assessing changes after the implementation of multidrug therapies in
Canada in late 1995.
Design: Retrospective review of all HIV hospitalization at a tertiary care institution.
Methods: The characteristics of HIV admissions were examined between
1990-97 using data from the Toronto Wellesley Central Hospital Health Records
Department. The number of HIV admissions, patients, deaths and average
length of stay (ALOS) were noted in order to examine the burden of HIV admissions on hospital resources. The number of comorbidities, number of discharges/patient/year, and the number of alternate level of care (ALC) cases were
studied in order to estimate the complexity of HIV illness and whether resources
in the community are sufficient. ALC cases are those ready for discharge but
remain in hospital because of inadequate community resources.
Results: Between 1990-96, the number of HIV admissions increased from
102/yr to 397/yr. The actual number of HIV patients admitted also increased