Professional Documents
Culture Documents
ABSTRACT The goal o f this study wlls to examine where people with acquired immune de-
ficiency syndrome (AIDS) in the United States liw and the degree to which AlDS is present
in rural areas. AIDS cases reported to the Centersfir Disease Control and P r m t w n (CDC)
in 1996 were categorized by metropolitan statistical area (MSA) size and compared to the
general population. Data w e analyzed by region, racelethnicity and risk exposure; AIDS in-
cidence rates m e compared m m time by M S A size. Relatim to the US. population, AIDS
cases were disproportionately black (43 percent us. 11 percent), male (80 percent vs. 48 per-
cent), and from the Northeast (32 percent us. 20 percent). In all regions, a greater proportion
of AIDS cases reside in large MSAs compared with the general population. Risk exposures
d i e little by MSA size, except in the Northeast. The PTopOrtion ofpeople with AIDS who
reside in large MSAs exceeds the proportion of the population in those areas, especially when
racelethnicity is consit&red. AIDS rufes hmR increased in non-MSAs relatiw to,large MSAs,
y t do not indicate that the epidpmic is increasing rapidly in rural areas. F w AIDS cases
are reported from smaller communities, y t require medical and social services that may bur-
den the rural health care system.
A
s the focus of the response to the hu- posure Analysis of the geographic distribution of
man immunodeficiency virus (HIV) and AIDS cases can highlight new areas of incidence as
acquired immunodeficiency syndrome well as Aanges in other locations that may require
(m) epidemic increasingly moves to- attention.
ward early testing and treatment, it is Although the majority of people with AIDS have
essential to monitor the magnitude and to dmracterize been reported from urban areas, over the course of
the population of infected people who require and use
prevention and care resources. Prevention and plan-
ning rely on current trends in HIV and AJDS epide- We thank Mitzi h4ays jbr her assistance with data generation and mnage-
miology to maximize access to counseling, testing, ment and John Karon fbr his assistance with statistical analyses. FmQr-
medical and other services among the appropriate ther injbrmation, mtnct: Shari Steinberg, Division of HIVIAIDS P r m -
populations. Accurate surveillance for HIV infection twn-Sumeillance and Epidemiology, Nationnl Centerjbr HlV STD and
and AIDS can idenbfy new populations at increased TB P r m t i o n , Centersfirr Disease Conttol and P r m t i o n , 1600 Clifton
risk for infection and changes in patterns of HIV ex- Road, Mailstop E 4 7 , Atknfa, GA 30333; e-mail s m 2 @ c d c . p
12 Vd.16, No.1
the distribution of the adult population, ATDS cases
were disproportionately reported from the Northeast Table 1. Reported Adolescent and Adult AIDS
(32 percent of cases vs. 20 percent of the general p o p Cases and Percentage Distributions of
ulation); more of the cases were in people who were AIDS Cases and the Adolescent and Adult
Hispanic (16 percent vs. 9 percent) or black, non-His- Population in the United States and
panic (43 percent vs. 11 percent) and male (80percent District of Columbia, 1996.
vs. 48 percent; Table 1).
Of the AIDS cases, 83 percent were in people from
MSAs with populations of more than soO,O00, al- AIDS
though only 62 percent of the general population re- U.S. Rate
sides in those areas. Overall, the annual AIDS rate in Poplation per
these areas was 42 per l00,OOO population. Smaller AIDS (N= 100,OOO
(N=66,158) 212,379,502) Population
MSAs,hawever, had 10 percent of AIDS cases,com-
pared with 18 percent of the population (rate=18 per
lO0,OOO). Seven percent of people with AIDS and 20 (Per- (Per-
(Number) centage) centage)
percent of the population live in non-MSAs (rate=lO
per lO0,OOo). Region'
One-third of AIDS cases from the large MsAs were Northeast 20,831 32 20 49.4
reported from the Northeast (N=19,059); another Midwest 6,772 10 24 13.5
third (N=18,716) were from the South pable 2). This south 25,476 39 35 34.2
is somewhat comparable to the distribution of the West 13,079 20 22 28.6
population in large MSAs; 25 percent are in the Racelethnidty
Northeast and 30 percent in the South. AIDS cases White 26,199 40 76 16.3
from the large MSAs, however, are disproportionately Black 28,307 43 11 11 7.4
Hispanic 10,726 16 9 54.8
in people of minority race / ethnjaty; 43 percent of the
Other 926 1 4 11.1
cases are in blacks who comprise 13 percent of the
SeX
population. Of the cases, 18 percent were in Hispanics
Male 52,902 80 48 48.1
who comprise 12 percent of the general population. Female 13,256 20 52 12.9
In non-MSAs, the same racial and ethnic disparity
Metropolitan Statistical h a (MSA)
is found. Of the cases,41 percent are in blacks
m O,OO + 54,597 83 62 41.6
(N=1,777) and 8 percent are in Hispanics (N=343). Of 50,OOO to s00,OOO 6,738 10 18 17.7
the general non-WA population, 8 percent is black Non-MSA 4,336 7 20 10.1
and 4 percent is Hispanic. The south has the highest
proportion (58 percent) of nonmetropoolitan AIDS cas-
1. Northeast=Connecticut,Maine, Massachusetts, New
es, althaugh the rate in the non-MSA South (14 per
Hampshue, New Jersey, New York, Pennsylvania, Rhode Island
100,OOO) is similar to that in the non-MsA Northeast and Vermont; Midwest=Illinois, Indiana, Iowa,Kansas,
(13 per lO0,OOo). Michigan, Minnffota, Missouri, Nebraska, North Dakota, Ohio,
AlDs rates differ by race, age and location; however, South Dakota and Wisconsin; South=Alabama, Arkansas,
Delaware, District of Columbia, Florida, Georgia, Kentucky,
some pattems are evident. Rates are generally highest Louisiana, Maryland, Mississippi, North Carolina, Oklahoma,
in the large MsAs within each race, age group and re- South Carolina,-T Texas, V u G a and Wst Virginia;
gion. For whites, rates are lowest in the Midwest; rates West=Alaska, MOM, California, Colorado, Hawaii, Idaho,
for whites in the other three regions are similar in Montana, Nevada, New Mexico, Oregon, Utah,Washington and
each corresponding MSA population category. Rates Wyoming.
for blacks are 3 to 32 times higher than rates for
whites; rates for Hispanics fall between those for
blacks and whites in nearly every location. The North-
east has the highest rates for minority races/ethnia-
ties, regardless of MSA size. Rates within each MSA were lowest for people older than 65 years of age at
and region category were highest for people 30 to 49 the time of AIDS diagnosis.
years of age at diagnosis, followed by people 13 to 29 Despite some of the lower rates in the South, one
years of age and people 50 to 64 years of age, and must consider the number of cases to realize the effect
Men Women
Northeast
MSA soO,OOO+ 12,511 39 45 11 5 5,033 46 51 3
MSA 50,OOO to 500,OOO 772 36 48 8 9 236 44 52 4
Non-MSA 322 43 40 7 9 76 39 57 4
Midwest
MSA 5oo,OOO+ 3,652 69 18 6 8 730 40 57 3
MSA 50,OOO to 5oo,ooO 665 66 18 7 9 91 20 74 6
Non-MSA 485 59 23 8 10 88 28 68 4
south
MSA 5OO,OOO+ 14,087 58 22 13 7 4,181 35 61 3
MSA 50,OOO to 500,OOO 2,767 58 19 13 9 838 26 69 5
Non-MSA 1,902 52 22 15 11 577 25 71 5
West
MSA 5OO,OOO+ 8,619 76 12 3 9 1,069 38 55 7
MSA 50,OOO to 500,000 883 66 17 4 14 99 31 59 10
Non-MSA 3% 62 19 7 13 60 47 50 3
White
MSA 5oo,OOO+ 15,066 76 12 4 8 1,899 45 49 5
MSA 50,OOO to 5OO,O00 2,528 72 13 5 11 387 31 63 5
Non-MSA 1,512 67 14 7 12 290 33 63 4
Black
MSA 5OO,OOO+ 15,926 42 38 14 7 7,056 41 56 3
MSA 50,OOO to 5oo,OOO 1,933 41 33 17 9 767 28 67 5
Non-MSA 1,307 39 32 18 10 455 25 71 4
Hispanic
MSA 5oo,OOO+ 7,247 50 34 10 6 1,921 37 60 4
MSA 50,OOO to 500,OOO 566 48 33 15 4 97 32 64 4
Non-MSA 235 40 42 10 9 40 29 67 4
Note: MSA=metropolitan statistical arra; MSM=men who have sex with men; IDU=inje&on drug use; Het=heterosexual contact. Other
includes MSM-IDU (menonly), adult hemophilia, adult transfusion and adult risk undetermined.
1. Data are adjusted for reporting delays and the redistribution of unreported mode of exposure Totals include all race/ethniaty and risk
groups. Percentages given are row percentages.
of cases among blacks is rather high. Both the absolute groups. Although the rates will likely remain highest
and the proportional burden of the epidemic in any in the aties, there has been a trend of increasing rates
area should be considered in assessing the effect of in the non-MSAs relative to the rates in large MAs.
AIDS and the needs for treatment, prevention and so- This trend, though statistically sipficant, points not
cial S e M C e s . so much to a burgeoning epidemic of rural AIDS as to
In general, AIDS rates in large MSAs exceed or a p a slow infiltration of the virus into some smaller com-
proximate the rates in smaller MSAs and non-MSAs munities in the United States.
in all regions of the country for all racial/ethnic By definition, non-MSAs have smaller populations
16 Vd.16, No. 1
and may offer a more limited range of medical and tween diagnosis and death. O f these people, the pro-
social services (Smith, 1990). Access to care and re- portion of interstate moves was greatest among those
saurces is critical for HIV-infected people Better ac- initially from large cities, and most of the moves, both
cess to medical services has been associated with few- inter- and intrastate, were to large metropolitan areas,
er hospitalizations (Cunningham, et al., 1996). AIDS regardless of the sue of the initial place of residence
patients of more experienced physicians have longer The net result was a small relative decrease in large
survival times (Kitahata, et al., 1996), and patients ad- metropolitan areas and a larger relative increase in
mitted to hospitals with more AIDS admissions have nonmetropolitan areas, although the absolute changes
lower mortality rates than do patients admitted to were comparable Migration can pose a practical prob-
hospitals with fewer AIDS admissions (Stone, et al., lem as some governmental AIDS funds are allocated
1992; Turner and Ball, 1992). A study of HIV/AIDS on the basis of the number of cases. Migration after
medical care showed that 68 percent of physicians in diagnosis can increase the burden of providing servic-
small and non-MSA counties in California, although es at the destination, which may lack the appropriate
their experience may be limited, had seen an HIV-in- allocation of resources.
fected patient and were providing care (Lewis, 1996). Our data are also limited by their timeliness: AIDS
Rural residents travel to larger places for medical care diagnosis generally lags behind HIV infection diagno-
for reasons of confidentiality in obtaining care, con- sis. Data on recently infected people are not available
cerns about prejudice in rural communities (Helms, through AIDS case surveillance. HN case surveillance
1993), or a lack of primary care physicians or provid- would provide data to better characterize recent ina-
ers with experience in treating HN infection (Berry, et dence trends, as well as where treatment, seMces and
al., 1996; Graham, et al., 1995; Rx for Rural AIDS, prevention efforts are needed. AIDS incidence among
1995). Some communities, however, have created pro- populations infected with HIV will depend on access
grams to address their AIDS problems and have orga- to treatment and the long-term effectiveness of thera-
nized networks of services and HIV care for affected pies. The assessment of changes in rural AIDS inci-
residents (Fiscus, et al., 1996; McKinney, 1993; Rx for dence is complicated by the lack of knowledge of the
Rural AIDS, 1995). Continued increases in the num- balance between changes in HIV incidence and issues
ber of people with HIV/AIDSoutside urban areas, of access to care and treatment. Trends in AIDS inci-
along with increased survival times for those in treat- dence over time were not analyzed because the data
ment, could place a sigruficant burden on the rural cover a time when AIDS incidence was growing ev-
health system and negatively affect those who are erywhere; instead, the relative impact of AIDS in ur-
infected. ban and rural areas was examined.
Information on where someone became infected or Alhmgh the data confirm the presence of HIV and
where that person may have moved after infection is AIDS in rural communities across the United States,
not available in the nationwide surveillance system; they do not suggest a rampant spread of ATDS to ru-
the data are limited to the persons place of residence ral areas. There are some communities with relatively
reported at the time of AIDS diagnosis. The data do high AIDS rates, especially those that have been af-
not directly measure if people with HIV or AIDS in fected by interacting epidemics of sexually transmitted
rural areas are receiving appropriate health services. diseases and drug use. These areas face difficult Cfial-
Investigators of smaller studies have confirmed migra- lenges to provide adequate care, confidentiality and
tion before or after HIV infection or AIDS diagnosis, services to the affected populations. Every communi-
generally in early studies of movement from urban to ty-urban, suburban or nonurban-should be aware
rural areas after diagnosis (Cohn, et al., 1994; Davis of the likelihood that HIV infection is present in the
and Stapleton, 1991; Verghese 1989). others have re- population and plan accordingly to ensure the oppor-
ported increases in the number of AIDS patients in- tunity for confidential testing, treatment and services.
fected in rural areas (Roberts, et al., 1997; Rumley, et However, repeated reports of alarming proportional
al., 1991). Although the effect of potential misclassifi- increases in AIDS cases in rural areas do not allow for
cation due to migration after diagnosis cannot be appropriate planning or resource allocation propor-
quantified, we assume that we slightly overestimate tional to the affected populations.
the number of cases in larger cities and underestimate Knowing where HIV is prevalent allows more ac-
that in smaller cities and rural areas. According to one curate planning of resource needs, including medical
study (Buehler, et al., 1995), at least 10 percent of peo- and social services, specialized training for care pro-
ple reported with AIDS who died had moved be- viders, and prevention services. The use of antiretro-