Abstract
Objectives. We used statewide data to assess HIV disease stage at initial diagnosis and laboratory indications for initiating antiretroviral therapy among South Carolina residents with newly diagnosed HIV infection.
Methods. Initial CD4+ counts and viral loads among individuals diagnosed with HIV between May 2004 and April 2005 were categorized according to current staging and treatment guidelines.
Results. Of 759 individuals who had a CD4+ count reported, 34% and 56% had counts of 200 cells/mm3 or below and 350 cells/mm3 or below, respectively. CD4+ counts of 200 cells/mm3 or below were significantly associated with male gender (adjusted odds ratio [AOR] = 2.07; 95% confidence interval [CI] = 1.36, 3.16), age above 29 years (AOR = 2.45; 95% CI = 1.51, 3.96), and hospital-reported patients (AOR = 2.17; 95% CI = 1.41, 3.36). The same characteristics were significant risk factors for elevated viral loads.
Conclusions. At least in South Carolina, HIV diagnoses are delayed in a significant percentage of patients. New testing strategies need to be implemented to encourage earlier HIV diagnoses, and future studies should evaluate the effects of expanded routine testing on earlier detection.
Approximately 800 people are newly diagnosed with HIV infection in South Carolina every year.1 For the past several years, the state has consistently ranked among the top 10 in terms of annual AIDS case rates in the United States. For example, in 2004, South Carolina was ranked 26th among the 50 states and the District of Columbia in overall population but 10th in number of AIDS cases per 100 000 population.2
In combination with a patient's clinical state, CD4+ T-cell counts and HIV-1 RNA levels are used in determining disease stage and initiation of therapy. CD4+ T-cell counts are used in laboratory staging of HIV (levels below 200 cells/mm3 are classified as laboratory AIDS), and both CD4+ counts (below 350 cells/mm3) and HIV-1 RNA levels (above 100 000 copies/mL) are used to determine need for commencement of antiretroviral therapy.
Anecdotal evidence suggests that many HIV-infected patients are diagnosed at an advanced disease stage. Several studies have demonstrated that patients who initiate highly active antiretroviral therapy (HAART) with CD4+ T-cell counts above 200 cells/mm3, HIV-1 RNA levels below 50 000 copies/mL, and no clinical AIDS-defining events are less likely to experience adverse clinical outcomes.3–5 By contrast, individuals who begin HAART when their CD4+ T-cell counts are below 200 cells/mm3 do not respond as well to therapy and have worse prognoses than those who initiate therapy at higher CD4+ T-cell counts.6
The lack of complete reporting of HIV laboratory markers (CD4+ T-cell counts and HIV-1 RNA levels) in many states has not allowed for quantification of disease stage at initial diagnosis. Only a few small studies have been able to link newly diagnosed cases to laboratory markers of immune deficiency and stage of HIV disease. Capitalizing on the unique opportunity provided by the robust and complete electronic HIV/AIDS reporting system in South Carolina, we used initial CD4+ T-cell counts and plasma viral RNA (viral loads) to characterize newly diagnosed HIV patients at the statewide (population) level.
METHODS
Study Design
South Carolina instituted confidential name-based reporting of HIV/AIDS on February 1, 1986. Since January 2004, state legislation has mandated that all laboratories electronically report all results on viral loads and CD4+ T-cell counts to the South Carolina Department of Health and Environmental Control (SC-DHEC). These data are stored in the HIV/AIDS Reporting System (HARS) database. South Carolina HARS data quality ratings exceed Centers for Disease Control and Prevention (CDC) minimum standards in terms of reporting timeliness (95% of cases are reported within 6 months of a diagnosis) and completeness of reporting (98% complete reporting based on a comparison with other data sources; SC-DHEC, unpublished data, December 2005).
In this study, we conducted a population-based, statewide cross-sectional analysis of HARS data using initial CD4+ T-cell counts and viral loads among individuals newly diagnosed with HIV between May 1, 2004, and April 30, 2005. To be included in this study, patients were required to have a definitive diagnosis of HIV infection and a laboratory test reported to HARS. They also had to be South Carolina residents and to have been 13 years or older at the time of their first positive HIV test.
We used the first CD4+ T-cell counts and viral load test results reported after initial diagnosis (and before initiation of antiretroviral therapy) to (1) assess patients' disease stage prior to their initiation of antiretroviral therapy and (2) determine the percentages of newly diagnosed individuals requiring HAART or opportunistic infection prophylaxis (or both) at the time of their initial diagnosis. (Opportunistic infection prophylaxis is recommended at CD4+ T-cell counts below 200 cells/mm3 or in the presence of AIDS-related opportunistic infections such as Kaposi's sarcoma, Pneumocystis pneumonia, Molluscum contagiosum, and Mycobacterium avium complex infections.) Our findings should provide a baseline for comparisons with future studies.
Variables
The demographic variables analyzed included gender, race/ethnicity (Black, White, and Other, including Hispanic, Asian, and Pacific Islander), age at HIV diagnosis, mode of exposure category (men having sex with men [MSM], injection drug use, heterosexual, unspecified), area of residence at time of HIV diagnosis (rural or urban), and source of report (county health department, hospital, private physician/group practice, state corrections facility or mental health department, or other [including laboratories, blood banks, private businesses, and federal institutions]).
Initial CD4+ T-cell counts (with a cutoff of 200 cells/mm3) were used to classify stage of disease. CD4+ counts (350 or below vs more than 350 cells/mm3) in combination with HIV-1 RNA levels (cutoff of 100 000 copies/mL) were used to determine need for commencement of antiretroviral therapy. The CD4+ cutoff value of 200 cells/mm3 reflects the current laboratory criterion for an AIDS diagnosis, below which prophylactic medications for opportunistic infections are recommended in addition to HAART7; the 350 cells/mm3 cutoff is the value below which the October 2005 Department of Health and Human Services guidelines recommend that antiretroviral therapy be offered to patients.7 According to these guidelines, patients with viral loads above 100 000 copies/mL should also be offered antiretroviral therapy.7
Statistical Analysis
We used SAS version 9.1.2 (SAS Institute Inc, Cary, NC) and Epi Info 6.04d (CDC, Atlanta, Ga) statistical software in conducting descriptive analyses. In addition, we conducted univariate and multivariable logistic regression analyses to determine whether there were associations (crude and adjusted odds ratios) between the independent variables and CD4+ T-cell count and viral load categories. The reference categories in the multiple logistic regression models were female gender, White race, the 20- to 29-year age group, heterosexual mode of exposure, rural residence, and private physician or group medical practice as source of HIV diagnosis.
RESULTS
Study Population
Between May 1, 2004, and April 30, 2005, 973 patients with HIV infection were reported to SC-DHEC. Of these patients, 102 were not residents of South Carolina, 5 were younger than 13 years, and 14 did not have a definitive HIV diagnosis, leaving 852 patients who met our inclusion criteria. Among this group of 852 individuals, 759 had initial CD4+ counts and 673 patients had initial plasma viral loads reported to and documented by HARS during the study period. Demographic and laboratory characteristics of the study population are shown in Tables 1, 2, and 3.
TABLE 1.
Results of Univariate Analysis of Initial CD4+ T-Cell Counts and HIV Disease Stage (Relative Odds of Initial Count of 200 cells/mm3 or Below) Among South Carolina Residents 13 Years or Older Diagnosed Between May 1, 2004, and April 30, 2005
| Initial CD4+ T-Cell Count |
||||
| No. (%) | ≤ 200, No. (%) | > 200, No. (%) | Crude OR (95% CI) | |
| Total | 759 (100) | 259 (34) | 500 (66) | … |
| Gender | ||||
| Male | 510 (67) | 195 (38) | 315 (62) | 1.79 (1.28, 2.50) |
| Female (Ref) | 249 (33) | 64 (26) | 185 (74) | 1.00 |
| Race/ethnicity | ||||
| Black | 551 (73) | 185 (34) | 366 (66) | 1.00 (0.70, 1.44) |
| White (Ref) | 173 (23) | 58 (34) | 115 (66) | 1.00 |
| Othera | 33 ( 4) | 16 (48) | 17 (52) | 1.87 (0.88, 3.96) |
| Age at diagnosis, y | ||||
| 13–19 | 32 (4) | 1 (3) | 31 (97) | 0.12 (0.02, 0.87) |
| 20–29 (Ref) | 174 (23) | 38 (22) | 136 (78) | 1.00 |
| 30–39 | 236 (31) | 93 (39) | 143 (61) | 2.33 (1.49, 3.63) |
| 40–49 | 207 (27) | 76 (37) | 131 (63) | 2.08 (1.31, 3.28) |
| > 49 | 110 (15) | 51 (46) | 59 (54) | 3.09 (1.84, 5.20) |
| Exposure category | ||||
| MSM | 259 (34) | 91 (35) | 168 (65) | 1.19 (0.83, 1.71) |
| IDU | 40 (5) | 18 (45) | 22 (55) | 1.79 (0.91, 3.52) |
| MSM and IDU combined | 10 (1) | … | … | … |
| Heterosexualb (Ref) | 255 (34) | 80 (31) | 175 (69) | 1.00 |
| Unspecified | 195 (26) | 67 (34) | 128 (66) | 1.15 (0.77, 1.70) |
| Residence at diagnosis | ||||
| Rural (Ref) | 242 (32) | 81 (33) | 161 (67) | 1.00 |
| Urban | 517 (68) | 178 (34) | 339 (66) | 1.04 (0.76, 1.44) |
| Source of report | ||||
| County health department | 246 (32) | 51 (21) | 195 (79) | 0.49 (0.31, 0.76) |
| Hospital | 223 (29) | 120 (54) | 103 (46) | 2.17 (1.44, 3.28) |
| Private/group practice (Ref) | 169 (22) | 59 (35) | 110 (65) | 1.00 |
| State facilityc | 57 (8) | 7 (12) | 50 (88) | 0.26 (0.11, 0.61) |
| Otherd | 64 (8) | 22 (34) | 42 (66) | 0.98 (0.53, 1.79) |
Note. OR = odds ratio; CI = confidence interval; MSM = men having sex with men; IDU = injection drug use. Ellipses indicate variable was not analyzed or was too small for further analysis.
Hispanic, Asian, or Pacific Islander.
Including heterosexuals who had sexual intercourse with a high-risk individual (such as an injection drug user, a male bisexual, a transfused individual, or someone who was HIV positive).
Primarily the state corrections department, along with mental health facilities.
South Carolina residents reported from other states, blood banks or businesses, federal facilities, laboratories, or unknown sources.
TABLE 2.
Results of Univariate Analysis of Initial CD4+ T-Cell Counts and Indications for HAART (Relative Odds of Initial Count of 350 cells/mm3 or Below) Among South Carolina Residents 13 Years or Older Diagnosed Between May 1, 2004, and April 30, 2005
| Initial CD4+ T-Cell Count |
||||
| No. (%) | ≤ 350, No. (%) | > 350, No. (%) | Crude OR (95% CI) | |
| Total | 759 (100) | 424 (56) | 335 (44) | … |
| Gender | ||||
| Male | 510 (67) | 315 (62) | 195 (38) | 2.08 (1.53, 2.82) |
| Female (Ref) | 249 (33) | 109 (44) | 140 (56) | 1.00 |
| Race/ethnicity | ||||
| Black | 551 (73) | 313 (57) | 238 (43) | 1.27 (0.90, 1.79) |
| White (Ref) | 173 (23) | 88 (51) | 85 (49) | 1.00 |
| Othera | 33 ( 4) | 22 (67) | 11 (33) | 1.93 (0.88, 4.23) |
| Age at diagnosis, y | ||||
| 13–19 | 32 (4) | 12 (38) | 20 (62) | 0.74 (0.34, 1.60) |
| 20–29 (Ref) | 174 (23) | 78 (45) | 96 (55) | 1.00 |
| 30–39 | 236 (31) | 139 (59) | 97 (41) | 1.76 (1.19, 2.62) |
| 40–49 | 207 (27) | 120 (58) | 87 (42) | 1.70 (1.13, 2.55) |
| > 49 | 110 (15) | 75 (68) | 35 (32) | 2.64 (1.60, 4.35) |
| Exposure category | ||||
| MSM | 259 (34) | 147 (57) | 112 (43) | 1.10 (0.78, 1.56) |
| IDU | 40 (5) | 27 (68) | 13 (32) | 1.75 (0.86, 3.53) |
| MSM and IDU combined | 10 (1) | … | … | … |
| Heterosexualb (Ref) | 255 (34) | 138 (54) | 117 (46) | 1.00 |
| Unspecified | 195 (26) | 106 (54) | 89 (46) | 1.00 (0.69, 1.45) |
| Residence at diagnosis | ||||
| Rural (Ref) | 242 (32) | 135 (56) | 107 (44) | 1.00 |
| Urban | 517 (68) | 289 (56) | 228 (44) | 1.00 (0.74, 1.37) |
| Source of report | ||||
| County health department | 246 (32) | 114 (46) | 132 (54) | 0.69 (0.47, 1.02) |
| Hospital | 223 (29) | 159 (71) | 64 (29) | 1.99 (1.30, 3.02) |
| Private/group practice (Ref) | 169 (22) | 94 (56) | 75 (44) | 1.00 |
| State facilityc | 57 (8) | 22 (39) | 35 (61) | 0.50 (0.27, 0.93) |
| Otherd | 64 (8) | 35 (55) | 29 (45) | 0.96 (0.54, 1.72) |
Note. OR = odds ratio; CI = confidence interval; HAART = highly active antiretroviral therapy; MSM = men having sex with men; IDU = injection drug use. Ellipses indicate variable was not analyzed or was too small for further analysis.
Hispanic, Asian, or Pacific Islander.
Including heterosexuals who had sexual intercourse with a high-risk individual (such as an injection drug user, a male bisexual, a transfused individual, or someone who was HIV positive).
Primarily the state corrections department, along with mental health facilities.
South Carolina residents reported from other states, blood banks/businesses, federal facilities, laboratories, or unknown sources.
TABLE 3.
Results of Multivariable Analysis of Initial Viral Loads and Indications for HAART (Relative Odds of Initial Viral Load of 100 000 copies/mL or Above) Among South Carolina Residents 13 Years or Older Diagnosed Between May 1, 2004, and April 30, 2005
| Initial Viral Load |
||||
| No. (%) | ≥ 100 000, No. (%) | < 100 000, No. (%) | Crude OR (95% CI) | |
| Total | 673 (100) | 193 (29) | 480 (71) | … |
| Gender | ||||
| Male | 467 (69) | 150 (32) | 317 (68) | 1.79 (1.22, 2.64) |
| Female (Ref) | 206 (31) | 43 (21) | 163 (79) | 1.00 |
| Race/ethnicity | ||||
| Black | 489 (73) | 130 (27) | 359 (73) | 0.64 (0.43, 0.94) |
| White (Ref) | 152 (23) | 55 (36) | 97 (64) | 1.00 |
| Othera | 30 (4) | 7 (23) | 23 (77) | 0.54 (0.22, 1.33) |
| Age at diagnosis, y | ||||
| 13–19 | 29 (4) | 5 (17) | 24 (83) | 0.89 (0.31, 2.53) |
| 20–29 (Ref) | 153 (23) | 29 (19) | 124 (81) | 1.00 |
| 30–39 | 210 (31) | 64 (30) | 146 (70) | 1.87 (1.14, 3.09) |
| 40–49 | 186 (28) | 68 (37) | 118 (63) | 2.46 (1.49, 4.07) |
| > 49 | 95 (14) | 27 (28) | 68 (72) | 1.70 (0.93, 3.10) |
| Exposure category | ||||
| MSM | 234 (35) | 72 (31) | 162 (69) | 1.57 (1.04, 2.38) |
| IDU | 34 (5) | 11 (32) | 23 (68) | 1.69 (0.77, 3.70) |
| MSM and IDU combined | 10 (1) | … | … | … |
| Heterosexualb (Ref) | 226 (34) | 51 (23) | 175 (77) | 1.00 |
| Unspecified | 169 (25) | 58 (34) | 111 (66) | 1.85 (1.19, 2.88) |
| Residence at diagnosis | ||||
| Rural (Ref) | 212 (31) | 55 (26) | 157 (74) | 1.00 |
| Urban | 461 (69) | 138 (30) | 323 (70) | 1.22 (0.85, 1.76) |
| Source of report | ||||
| County health department | 232 (34) | 31 (13) | 201 (87) | 0.32 (0.19, 0.53) |
| Hospital | 200 (30) | 95 (48) | 105 (52) | 1.85 (1.18, 2.90) |
| Private/group practice (Ref) | 140 (21) | 46 (33) | 94 (67) | 1.00 |
| State facilityc | 47 (7) | 7 (15) | 40 (85) | 0.36 (0.15, 0.86) |
| Otherd | 54 (8) | 14 (26) | 40 (74) | 0.72 (0.35, 1.45) |
Note. OR = odds ratio; CI = confidence interval; HAART = highly active antiretroviral therapy; MSM = men having sex with men; IDU = injection drug use. Ellipses indicate variable was not analyzed or was too small for further analysis.
Hispanic, Asian, or Pacific Islander.
Including heterosexuals who had sex with a high-risk individual (such as an injection drug user, a male bisexual, a transfused individual, or someone who was HIV positive).
Primarily the state corrections department, along with mental health facilities.
South Carolina residents reported from other states, blood banks/businesses, federal facilities, laboratories, or unknown sources.
CD4+ T-Cell Counts
Among the 759 individuals who had data available on CD4+ T-cell count at diagnosis, the overall median was 306 cells/mm3 (interquartile range = 131–491 cells/mm3); 259 (34%) of these individuals had an initial count of 200 cells/mm3 or less, and 424 (56%) had an initial count of 350 cells/mm3 or less (Tables 1 and 2). Median CD4+ T-cell counts differed according to gender, age at diagnosis, and source of report; men, older patients, and patients whose diagnoses had been reported by hospitals had lower CD4+ counts.
There were strong associations between lower CD4+ T-cell counts and gender, age, and source of HIV test report (Tables 2 and 3). The multivariable analysis demonstrated a statistically significant gender difference, with men having higher odds of low CD4+ T-cell counts (Table 4). Also, in comparison with patients aged 20 to 29 years, older patients had significantly greater odds of presenting with a CD4+ T-cell count of 200 cells/mm3 or less (Table 1). This association remained after adjustment for gender, race/ethnicity, mode of exposure, area of residence, and source of HIV report (Table 4). Similarly, separate comparisons between the 20- to 29-year age group and the other age groups showed that the odds of presenting with an initial CD4+ T cell count of 350 cells/mm3 or below were significantly greater among older patients than among younger patients (Tables 2 and 4).
TABLE 4.
Results of Multiple Logistic Regression Analysis of Initial CD4+ T-Cell Counts and Viral Loads Among South Carolina Residents 13 Years or Older Diagnosed Between May 1, 2004, and April 30, 2005
| CD4 ≤ 200, AOR (95% CI) | CD4 ≤ 350, AOR (95% CI) | Viral Load ≥ 100 000, AOR (95% CI) | |
| Gender | |||
| Male | 2.07 (1.36, 3.16) | 2.77 (1.86, 4.12) | 2.01 (1.25, 3.24) |
| Female (Ref) | 1.00 | 1.00 | 1.00 |
| Race/ethnicity | |||
| Black | 1.32 (0.88, 1.99) | 1.53 (1.05, 2.23) | 0.74 (0.48, 1.14) |
| White (Ref) | 1.00 | 1.00 | 1.00 |
| Othera | 2.56 (1.11, 5.91) | 2.19 (0.95, 5.09) | 0.59 (0.22, 1.58) |
| Age at diagnosis, y | |||
| 13–19 | 0.12 (0.02, 0.96) | 0.85 (0.38, 1.92) | 1.10 (0.36, 3.33) |
| 20–29 (Ref) | 1.00 | 1.00 | 1.00 |
| 30–39 | 2.45 (1.51, 3.96) | 1.89 (1.24, 2.89) | 1.72 (1.01, 2.96) |
| 40–49 | 2.04 (1.22, 3.40) | 1.66 (1.05, 2.60) | 2.08 (1.19, 3.66) |
| > 49 | 3.01 (1.67, 5.40) | 2.51 (1.45, 4.36) | 1.24 (0.63, 2.45) |
| Exposure category | |||
| MSM | 0.93 (0.58, 1.50) | 0.69 (0.44, 1.08) | 1.06 (0.63, 1.80) |
| IDU | 1.05 (0.50, 2.23) | 1.13 (0.52, 2.45) | 0.96 (0.41, 2.25) |
| MSM and IDU combined | … | … | … |
| Heterosexualb (Ref) | 1.00 | 1.00 | 1.00 |
| Unspecified | 0.67 (0.43, 1.06) | 0.62 (0.41, 0.94) | 1.24 (0.76, 2.01) |
| Residence at diagnosis | |||
| Rural (Ref) | 1.00 | 1.00 | 1.00 |
| Urban | 1.11 (0.77, 1.58) | 1.04 (0.74, 1.46) | 1.08 (0.72, 1.62) |
| Source of report | |||
| County health department | 0.51 (0.32, 0.82) | 0.71 (0.47, 1.07) | 0.36 (0.21, 0.61) |
| Hospital | 2.17 (1.41, 3.36) | 2.01 (1.29, 3.13) | 1.94 (1.21, 3.11) |
| Private/group practice (Ref) | 1.00 | 1.00 | 1.00 |
| State facilityc | 0.22 (0.09, 0.54) | 0.41 (0.21, 0.80) | 0.32 (0.13, 0.80) |
| Otherd | 0.90 (0.48, 1.71) | 0.91 (0.50, 1.68) | 0.67 (0.32, 1.37) |
Note. AOR = adjusted odds ratio; CI = confidence interval; MSM = men having sex with men; IDU = injection drug use. Ellipses indicate variable was not analyzed or was too small for further analysis.
Hispanic, Asian, or Pacific Islander. Ellipses indicate variable was not analyzed or was too small for further analysis.
Including heterosexuals who had sexual intercourse with a high-risk individual (such as an injection drug user, a male bisexual, a transfused individual, or someone who was HIV positive).
Primarily the state corrections department, along with mental health facilities.
South Carolina residents reported from other states, blood banks/businesses, federal facilities, laboratories, or unknown sources.
In addition, the odds of presenting with an initial CD4+ T-cell count of 350 cells/mm3 or less were significantly greater among patients whose diagnosis was reported by a hospital than among patients whose diagnosis was reported by a state facility (Tables 2 and 4). Also, the strong association of late-stage disease with source of report remained in the multivariable analysis (Table 4). Race/ethnicity, mode of exposure, and area of residence at initial HIV diagnosis did not exhibit significantly strong associations with CD4+ T-cell count at time of HIV diagnosis (Tables 1, 2, and 4).
Plasma Viral Loads
Of the 673 individuals for whom data on viral loads were reported to SC-DHEC, the median count was 27 728 copies per ml (interquartile range = 7279–114 116 copies/mL). Among these 673 individuals, 193 (29%) had an initial viral load of 100 000 copies/mL or above, whereas 480 (71%) had an initial viral load below 100 000 copies/mL (Table 3). Median counts differed according to gender, age at diagnosis, mode of exposure, and source of HIV report. Overall, men, those in older age groups, MSM, and hospital patients had significantly higher viral loads.
The multivariable analysis demonstrated strong associations between viral load and gender, age, and source of HIV test report (Table 4). Men were twice as likely as women to present with an initial viral load of 100 000 copies/mL or greater. Also, in comparison with individuals aged 20 to 29 years, older individuals had significantly greater odds of presenting with an initial viral load of 100 000 copies/mL or greater (Table 4). In each of these analyses, the observed associations remained after adjustment for all other covariates.
Analyses of source of positive HIV test report showed that the odds of an initial viral load of 100 000 copies/mL or above were approximately twice as high among hospital patients (Table 4) and less than half as high among individuals whose diagnosis was reported by a state health facility, in comparison with private physician patients (Table 4). In the regression models used in these analyses, we also adjusted for gender, race/ethnicity, age at diagnosis, mode of exposure, and area of residence at HIV diagnosis (Table 4).
Median viral loads were 33 405 and 20 974 copies/mL for MSM and heterosexuals, respectively (data not shown). Odds of an initial viral load of 100 000 copies/mL or above were more than 1.5 times higher among MSM as among heterosexuals (Table 3), whereas odds were 1.9 times as high among those in the unspecified risk group (Table 3). However, these associations did not remain after adjustment for the other covariates (Table 4). Also, race/ethnicity and area of residence at initial HIV diagnosis did not exhibit strong associations with viral load at time of HIV diagnosis.
DISCUSSION
For the most part, the characteristics of our study population were similar to the characteristics of the 2004 South Carolina population diagnosed with HIV.1 Sixty-seven percent of our participants were men (identical to the 67% figure reported by SC-DHEC1); 73% were Black (again, identical to the 73% figure reported by SC-DHEC1); and 34% fell in the same-gender sexual activity (MSM) exposure category (as compared with 31% reported by SC-DHEC1).
Our data confirm that many adult and adolescent South Carolina residents are diagnosed with HIV infection at an advanced stage. An analysis of first laboratory values reported after initial diagnosis showed that 56% of our study population had a CD4+ T-cell count of 350 cells/mm3 or below, 34% had a T-cell count of 200 cells/mm3 or below, and 29% had a viral load of 100 000 copies/mL or above. Further analysis revealed significant associations between late presentation and male gender, older age, and hospital as source of HIV test report. Disease stage and medication need at initial diagnosis were not related to participants' area of residence, race/ethnicity, or mode of exposure.
In an earlier study, Luby et al.8 used data on CD4+ T-cell counts collected for HIV/AIDS surveillance reporting to evaluate the stages and epidemiology of recent HIV infection among patients seen in South Carolina public health clinics. They found that 12% of their study population had a CD4+ T-cell count below 200 cells/mm3 and that 46% had a count below 500 cells/mm3 (the cell count recommendations for management of HIV at the time their study was conducted). In addition, they found that women accounted for an increased percentage of newly infected patients and that individuals who were older than 24 years and men were much more likely to present with late-stage disease. Luby et al. did not find any associations between mode of exposure, race, or area of residence and presentation with late-stage HIV.
Our findings confirm and expand on those of the Luby et al. study. However, unlike that investigation, which included laboratory data from only consenting public health clinic patients, our study was population based and included all newly diagnosed patients reported by all health care providers in South Carolina. Thus, our results provide a comprehensive view of statewide population data in South Carolina. These data can be used to assess disease stage and need for medication use at the time of initial HIV diagnosis and as a baseline to track trends in the timeliness of HIV diagnoses of South Carolina patients in the future. Similar studies in other states will also enable better characterization of the HIV epidemic and better assessment of HIV/AIDS prevention and control programs across the United States.
Implications of Delayed HIV Diagnoses
Timely identification of HIV infection is important from both a clinical and a public health perspective. Late diagnoses may be associated with irreversible immunological damage and related comorbidities.9 Early diagnosis and commencement of antiretroviral therapy have contributed immensely to transforming HIV from a disease associated with high mortality rates to a chronic disease requiring long-term management. Typically, individuals with HIV infection now live at least 13 to 14 years longer if they have undergone potent antiretroviral therapy than if they have not completed therapy.10
Early HIV diagnoses also provide opportunities to reduce transmission of HIV through modifications in individual risk behaviors.9 Moreover, it is easier to educate and thereby empower patients when diagnoses are made before the need for commencement of antiretroviral therapy. In the clinical setting, such a scenario will improve patients' long-term adherence by providing an opportunity to adequately prepare them for a lifetime of daily use of HIV-related medications.11 Early diagnoses are even more important for patients with social, physical, and mental comorbidities (e.g., depression, hepatitis C infection, and drug dependence), who may require greater preparation to prevent poor adherence and drug interactions once they begin using antiretroviral medications.
From a public health perspective, a late diagnosis of HIV/AIDS suggests missed opportunities to control the spread of the HIV epidemic.11 Natural history studies have shown that patients diagnosed with laboratory AIDS (as was the case with 34% of our sample) have been infected with HIV for an average of 7 to 10 years before their initial diagnosis. A recent Ugandan study12 revealed that the highest per-sexual-act rates of heterosexual transmission, as well as the highest percentages of transmission overall, involved early-stage infection in index partners, when few seroconverters knew their HIV status or had undergone HAART. In most other studies, there has been a U-shaped HIV transmission pattern, with the highest rates of transmission occurring in the very early and late disease stages.
If patients were diagnosed earlier, the HIV epidemic could potentially be curbed through effective patient education on ways to decrease transmission and through commencement of antiretroviral therapy, which has been shown to reduce infectivity rates among affected patients.9 Sanders et al. estimated that routine 1-time HIV screening would reduce the annual transmission rate in the United States by slightly more than 20%.9 Accordingly, the stage at which HIV-positive patients are first diagnosed could have a significant impact on prevention and treatment initiatives.13
Finally, late diagnoses of HIV infection are also more costly to the health care system. Several studies have shown that it is more costly to treat patients at advanced stages of HIV, who are more likely to have comorbid conditions and to be hospitalized.9,11,14
Significance of Our Findings
We found significant differences in disease stage according to gender, age at diagnosis, and source of HIV test report. There are several possible reasons for the gender differences observed here. For example, women may be diagnosed earlier because they have more opportunities for contact with the health care system, such as during family planning, prenatal, or reproductive health care visits during which HIV screening is provided. Also, recent data from the South Carolina Behavioral Risk Factor Surveillance System (BRFSS) show that higher percentages of women than men are tested.15 The BRFSS results revealed that more women than men believed it was important for people to undergo testing (95.9% vs 90.9%, respectively).15 The findings of other studies suggest that men tend not to undergo an HIV test until they are symptomatic, and thus they are more likely to be diagnosed at a later disease stage.16,17
The disparities associated with source of HIV test report may reflect different testing strategies in the facilities examined. Indigent patients often use hospitals for regular care and may not be routinely offered HIV testing. South Carolina county health departments routinely offer HIV testing in their sexually transmitted disease clinics, and in the state corrections department testing is mandated for all individuals at their point of entry into the system.
In addition, South Carolina law mandates that SC-DHEC notify the partners of all individuals reported with HIV. Partner notification allows identification of those at greater risk of infection at a time when they may not be aware of their risk. These testing strategies may account for our finding that patients whose HIV test reports originated from county health departments and state facilities were diagnosed at an earlier disease stage. Finally, the risk-based testing strategy typically used by private physician practices may account for the finding that patients from these settings were diagnosed at a later stage. Broadening universal “opt-out” screening for HIV (all patients, except those who decline, undergo an HIV test during routine clinical visits) at primary care facilities, in emergency department units, and at other sites (such as hospitals and state facilities) may result in earlier HIV diagnoses.18
Previous studies have shown that because White patients with HIV have better access to antiretroviral therapy and opportunistic infection prophylaxis, their survival rates are better than those of Black patients.19–21 In our study, no racial disparities in regard to initial CD4+ T-cell counts or viral loads were observed, nor were any differences found according to patients' mode of exposure or their area of residence at the time of their diagnosis.
A limitation of this study was that not all of the patients reported to HARS had initial laboratory data available for this analysis. Of the individuals who satisfied our inclusion criteria, data on viral load were missing for 21%, and data on CD4+ T-cell count were missing for 11%. However, analyses revealed that individuals with missing values did not differ in regard to any significant characteristics from those with available initial laboratory values (data not shown). Also, because this study was cross sectional, it was limited in its ability to reveal any trends in laboratory values over the past several years.
Implications for Public Health Practice
Our results suggest that, at least in South Carolina, HIV diagnoses are delayed in a significant proportion of patients. Late-stage diagnoses are associated with increased morbidity and mortality, prolonged transmission, less-robust immune restoration after therapy, and significantly increased treatment costs. To promote earlier detection of HIV infection, CDC recommends that all young people and adults aged 13 to 64 years be screened for HIV during routine health care visits.22 The potential benefits of routine and universal HIV testing, which would allow individuals to be linked to treatment and preventive services at an early disease stage, could be substantial. Given the differences we found in the screening practices of private providers, they should be an important focus of expansions in routine screening efforts.
Our study provides policymakers and medical providers with population-based data that can serve as a baseline assessment of an entire state's ongoing epidemic, allowing future trends in timely diagnosis and control of HIV/AIDS to be monitored. In addition, our findings can be used to support local efforts to promote and adopt CDC's routine testing recommendations. Future assessments should address the effects of expanded routine testing on earlier detection of HIV. Such efforts will assist in identifying health care settings in which opportunities for early diagnosis are missed and thereby provide opportunities for education and training to promote routine screening.
Human Participant Protection
No protocol approval was needed for this study.
Acknowledgments
We are indebted to Eric Brenner, Dana Giurguitiu, Terri Stephens, Jerry Gibson, and Lillian Smith for facilitating the collaborations that led to this article and for offering valuable suggestions at the early stages of development of this work.
References
- 1.STD Division Surveillance Report, June 30, 2005. Columbia: South Carolina Dept of Health and Environmental Control; 2005 [Google Scholar]
- 2.Centers for Disease Control and Prevention Cases of HIV and AIDS in the United States, 2004. HIV/AIDS Surveill Rep. 2004;16:29–30 [Google Scholar]
- 3.Anastos K, Barron Y, Miotti P, et al. Risk of progression to AIDS and death in women infected with HIV-1 initiating highly active antiretroviral treatment at different stages of disease. Arch Intern Med. 2002;162:1973–1980 [DOI] [PubMed] [Google Scholar]
- 4.Jacobson LP, Li R, Phair J, et al. Evaluation of the effectiveness of highly active antiretroviral therapy in persons with human immunodeficiency virus using biomarker-based equivalence of disease progression. Am J Epidemiol. 2002;155:760–770 [DOI] [PubMed] [Google Scholar]
- 5.Phillips AN, Staszewski S, Weber R, et al. HIV viral load response to antiretroviral therapy according to the baseline CD4 cell count and viral load. JAMA. 2001;286:2560–2567 [DOI] [PubMed] [Google Scholar]
- 6.Hogg RS, Yip B, Chan KJ, et al. Rates of disease progression by baseline CD4 cell count and viral load after initiating triple-drug therapy. JAMA. 2001;286:2568–2577 [DOI] [PubMed] [Google Scholar]
- 7.Guidelines for the Use of Antiretroviral Agents in HIV-1-Infected Adults and Adolescents. Washington, DC: US Dept of Health and Human Services; 2005 [Google Scholar]
- 8.Luby S, Jones J, Horan J. Using CD4 counts to evaluate the stages and epidemiology of HIV infection in South Carolina public clinic patients. Am J Public Health. 1994;84:377–381 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Sanders GD, Bayoumi AM, Sundaram V, et al. Cost-effectiveness of screening for HIV in the era of highly active antiretroviral therapy. N Engl J Med. 2005;352:570–585 [DOI] [PubMed] [Google Scholar]
- 10.Walensky RP, Paltiel AD, Losina E, et al. The survival benefits of AIDS treatment in the United States. J Infect Dis. 2006;194:11–19 [DOI] [PubMed] [Google Scholar]
- 11.Bozzette SA. Routine screening for HIV infection—timely and cost-effective. N Engl J Med. 2005;352:620–621 [DOI] [PubMed] [Google Scholar]
- 12.Wawer MJ, Gray RH, Sewankambo NK, et al. Rates of HIV-1 transmission per coital act, by stage of HIV-1 infection, in Rakai, Uganda. J Infect Dis. 2005;191:1403–1409 [DOI] [PubMed] [Google Scholar]
- 13.Gupta SB, Gilbert RL, Brady AR, Livingstone SJ, Evans BG. CD4 cell counts in adults with newly diagnosed HIV infection: results of surveillance in England and Wales, 1990–1998. AIDS. 2000;14:853–861 [DOI] [PubMed] [Google Scholar]
- 14.Paltiel AD, Weinstein MC, Kimmel AD, et al. Expanded screening for HIV in the United States—an analysis of cost-effectiveness. N Engl J Med. 2005;352:586–595 [DOI] [PubMed] [Google Scholar]
- 15.South Carolina Behavioral Risk Factor Surveillance System Report: 2003. Columbia: South Carolina Dept of Health and Environmental Control; 2005 [Google Scholar]
- 16.Biber CL, Jaker MA, Kloser P, Auerbach SB, Rhoads GG. A study of sex differences in presentation for care of HIV. AIDS Patient Care STDs. 1999;13:103–110 [DOI] [PubMed] [Google Scholar]
- 17.Klein D, Hurley LB, Merrill D, Quesenberry CP., Jr Review of medical encounters in the 5 years before a diagnosis of HIV-1 infection: implications for early detection. J Acquir Immune Defic Syndr. 2003;32:143–152 [DOI] [PubMed] [Google Scholar]
- 18.Centers for Disease Control and Prevention Missed opportunities for earlier diagnosis of HIV infection—South Carolina, 1997–2005. MMWR Morb Mortal Wkly Rep. 2006;55:1269–1272 [PubMed] [Google Scholar]
- 19.Moore RD, Stanton D, Gopalan R, Chaisson RE. Racial differences in the use of drug therapy for HIV disease in an urban community. N Engl J Med. 1994;330:763–768 [DOI] [PubMed] [Google Scholar]
- 20.Shapiro MF, Morton SC, McCaffrey DF, et al. Variations in the care of HIV-infected adults in the United States: results from the HIV Cost and Services Utilization Study. JAMA. 1999;281:2305–2315 [DOI] [PubMed] [Google Scholar]
- 21.Melnick SL, Sherer R, Louis TA, et al. Survival and disease progression according to gender of patients with HIV infection. JAMA. 1994;272:1915–1921 [PubMed] [Google Scholar]
- 22.Centers for Disease Control and Prevention Revised recommendations for HIV testing of adults, adolescents, and pregnant women in health-care settings. MMWR Morb Mortal Wkly Rep. 2006;55:1–17 [PubMed] [Google Scholar]