Since the early 1980’s, confidential
name-based surveillance has been conducted for AIDS cases; in this
process, patient names associated with cases have been reported to
local and state public health departments and stored in AIDS registries
for ongoing surveillance activities. Since the advent of the epidemic,
AIDS surveillance data have provided information critical to understanding
the epidemiology of HIV and clinical conditions associated with HIV
infection, have served as the backbone to evaluate the effect of treatment
and prevention efforts, and have allowed population-based monitoring
of the epidemic. However, with advances in treatment, such as highly
active antiretroviral therapy, persons with HIV infection are living
longer without progressing to AIDS. As a result, AIDS incidence has
decreased, and no longer provides the most accurate population-level
information on the state of the HIV epidemic. HIV infection surveillance,
however, provides information on more recently infected persons, and
thereby a more accurate representation of the current trends of the
epidemic. Based on these facts, CDC has recommended that all states
conduct HIV case surveillance as an extension of current AIDS surveillance
activities (1).
As of December 2001, confidential
name-based surveillance was being conducted for HIV infection that
had not progressed to AIDS in 33 states, Guam and the U.S. Virgin
Islands (Connecticut conducted name-based HIV surveillance for pediatric
cases only). In response to community concerns regarding the confidentiality
of name-based HIV reports within public health, however, several states
have elected to implement HIV infection surveillance using alternative
methods to confidential name-based reporting. These alternative methods
utilize coded patient identifiers instead of patient names for either
initial reporting to public health or for long term storage in surveillance
registries. Alternative methods to confidential name-based surveillance
that have been implemented and are presented here include code-based
and name-to-code-based systems. In code-based systems, HIV reports
are submitted to public health departments using a coded patient identifier
comprised of different, partial personal identifiers (e.g., date of
birth, initials of patient name, portions of social security number,
etc.) without patient name. In name-to-code-based systems, HIV reports
are initially reported to public health departments using patient
name; after public health follow up has been conducted and patient
referrals have been offered, names are converted into coded patient
identifiers for storage in the surveillance registry.
As a part of the 1999 guidelines
for national HIV case surveillance (1), CDC specified that states
using alternative methods for conducting HIV case surveillance should
evaluate the role of surveillance data in linking reported persons
to prevention and care programs and determine whether alternatives
to reporting of patient names would reduce confidentiality risks while
meeting the needs for high-quality surveillance data. CDC is working
with these areas to evaluate the proficiency and performance of the
coded patient identifiers within an integrated HIV/AIDS surveillance
system. The results of these evaluations will be reviewed as one aspect
of a larger project being conducted by the Institute of Medicine,
which will, in part, review the quality of integrated HIV/AIDS surveillance
data from systems using a variety of patient identifiers (name, name-to-code,
and code). Until the evaluations are complete, HIV surveillance data
from states conducting alternatives to confidential name-based HIV
reporting are not included in nationally accumulated HIV infection
case count totals. Until the technical and logistical data management
issues of data from systems using coded patient identifiers are resolved,
the existing national HIV/AIDS Reporting System is unable to receive
these data in the same manner as reports gathered within name-based
HIV/AIDS surveillance systems. This technical report represents the
first opportunity to examine demographic characteristics of persons
diagnosed with HIV infections who were residents of states that conduct
HIV infection case surveillance using coded patient identifiers.
As of December 2001, 12 states, the
District of Columbia, and Puerto Rico had implemented alternatives
to confidential name-based reporting for cases of HIV infection (without
AIDS). Of these 14 areas, nine conduct code-based HIV reporting (District
of Columbia, Hawaii, Illinois, Kentucky, Maryland, Massachusetts,
Puerto Rico, Rhode Island, and Vermont), and five states conduct name-to-code-based
HIV reporting (Delaware, Maine, Montana, Oregon, and Washington) (Figure
1). Combined, these states and territory reported 8,116 (19%)
of the 42,156 AIDS cases reported to the CDC in 2000; 54,177 (17%)
of the 322,865 persons living with AIDS at the end of December 2000
resided in these areas (2). In this report, we present the HIV surveillance
data from the five states (Illinois, Maine, Maryland, Massachusetts,
and Washington) and one territory (Puerto Rico) that had implemented
alternative methods of HIV case surveillance by January 1, 2000. The
coded patient identifiers implemented in these six areas included
sex and date of birth in all codes, various components of last name
(4 codes), the last 4 digits of the social security number (4 codes),
race (2 codes), various components of first name (2 codes), zip code
of residence (1 code), and health region (1 code). Aggregate data
in tabular form, as reported to the state or territorial health department,
were provided by respective areas for this report.
Combined, these six areas accounted
for 6,327 (15%) of 42,156 AIDS cases reported to CDC in 2000 (2).
In 2000, these six health departments received 8,563 reports of HIV
infection (Table 1). The 35 areas with confidential
name-based reporting reported 21,704 HIV infection cases during the
same time period (2). Five of six areas highlighted in this report
would rank in the top 10 states in the number of HIV cases reported
to CDC in 2000. HIV infection was more commonly reported among males
than among females in all six areas (Table 2).
Of 1,191 total cases reported in Washington, 980 (82%) were male,
while of 1,926 total cases reported in Maryland, 1,170 (61%) were
male; these states had, respectively, the highest and lowest male-to-female
case ratio among the areas. In all areas, a vast majority of all HIV
infection cases (range 92% to 96%) reported in 2000 were among persons
aged 20-64 years (Table 3). Within this age
range, Washington cases tended to be younger, with 58% of cases among
20-34 year olds, and Maryland cases tended to be older, with 34% of
cases in the 20-34 year-old age group. Areas also varied in the racial
and ethnic breakdown of HIV cases (Table 4);
in Maine, Massachusetts, and Washington more cases were reported for
white persons (83%, 49%, and 73%, respectively) than for any other
racial/ethnic group. In Illinois and Maryland, a majority of cases
were among Black, non-Hispanics (53% and 61%, respectively). These
data continue to stress the varied face of the epidemic among different
geographic areas.
References
|