Overall incidence of new HIV infections for adolescents was from the general survey 2·3% ([166/7,320], 95% CI: 1·94–2·63). Infections were notably greater in females (3·34%, 95% CI: 2·81–3·90) compared to male adolescents (0·86%, 95% CI: 0·56–1·24).
Characteristics of Adolescents Testing Newly Positive for HIV
Table 1 below shows detailed characteristics of adolescents testing newly positive to HIV. More females (83·7%, 95% CI: 77·2–89·0) than males (16·3%, 95% CI: 11·0–22·8) and more older adolescents (66·3%, 95% CI: 58·5–73·4) aged 20–24 compared to younger ones (33·7%, 95% CI: 26·6–41·5) aged 15–19 were diagnosed with recent HIV infections. Socioeconomic characteristics indicate larger proportions of unmarried adolescents (63.9%, 95% CI: 56·0–71·2) than married contemporaries (36·1%, 95% CI: 28·8–43·9) testing newly positive for HIV. New infections were observed to rise with wealth and education level. For instance, 28.0% (95% CI: 21·3–35·6) of participants in the highest wealth quintile compared to 7·3% (95% CI: 0·56–1·24) from the lowest tested positive. While 31·9% adolescents with primary education relative to 63.9% with secondary education had seroconverted. Behavioral attributes further show that two-thirds (67·6%) of adolescents that reported having had one sexual partner in 12 months prior to the survey compared to 19.9% with no partner had acquired the virus. Of all newly infected adolescents, only 14.5% were aware of their positive status, a consequence of low levels of HIV testing among young people (29–31).
Table 1
Characteristics of Adolescents Testing Newly Positive for HIV
Predictor | Total (n = 166) | Percentage of New HIV Infections | 95% CI |
Sex | | | |
Male | 27 | 16.3 (27/166) | 11.0–22.8 |
Female | 139 | 83.7 (139/166) | 77.2–89.0 |
Age | | | |
15–19 | 56 | 33.7 (56/166) | 26.6–41.5 |
20–24 | 110 | 66.3 (110/166) | 58.5–73.4 |
Residence | | | |
Rural | 70 | 42.2 (70/166) | 34.6–50.0 |
Urban | 96 | 57.8 (96/166) | 49.9–65.4 |
Marital status | | | |
Not married | 106 | 63.9 (106/166) | 56.0–71.2 |
Married | 60 | 36.1 (60/166) | 28.8–43.9 |
Wealth quintile | | | |
Lowest | 12 | 7.3 (12/164)* | 3.8–12.4 |
Second | 22 | 13.4 (22/164) | 8.6–19.6 |
Middle | 36 | 22 .0 (36/164) | 15.9–29.1 |
Fourth | 48 | 29.3 (48/164) | 22.4–36.9 |
Highest | 46 | 28.0 (46/164) | 21.3–35.6 |
Education level | | | |
No education | 2 | 1.2 (2/166) | 0.15–4.3 |
Primary | 53 | 31.9 (53/166) | 24.9–39.6 |
Secondary | 106 | 63.9 (106/166) | 56.0–71.2 |
Tertiary | 5 | 5.0 (5/166) | 0.9–6.8 |
Active syphilis | | | |
Positive | 12 | 7.2 (12/166) | 3.8–12.3 |
Negative | 154 | 92.8 (154/166) | 87.7–96.2 |
Hepatitis B virus | | | |
Positive | 11 | 6.6 (11/166) | 3.3–11.5 |
Negative | 155 | 93.4 (155/166) | 88.4–96.6 |
Number of sex partners in last 12 months | | | |
No partner | 30 | 19.9 (30/151)** | 13.8–27.1 |
One partner | 102 | 67.6 (102/151) | 59.5–74.9 |
More than one | 19 | 12.6 (19/151) | 7.7–18.9 |
Ever tested for HIV | | | |
Yes | 132 | 79.5 (131/166) | 71.9–84.5 |
No | 34 | 20.5 (34/166) | 15.1–28.0 |
Aware of HIV + status | | | |
Aware | 24 | 14.5 (24/166) | 9.5–20.7 |
Not aware | 142 | 85.5 (142/166) | 79.2–90.5 |
* “Wealth quintile” had 164 total observations instead of 166 because of missing values |
**“Number of sex partners in last 12 months” also had less observations (151) than the expected 166 due to missing values |
Virologic Outcomes of Adolescent with New HIV Infections
Slightly more than one third of new infections (36·7% [61/166], 95% CI: 29·41–44·57) had HIV RNA viral loads clustered in the 10,001–50,000 copies/ml viral load category (see Fig. 2a.). HIV RNA viral load above 50,000 copies/ml is strongly associated with the risk of transmission (32). Findings show that the mean HIV RNA viral load for new HIV infections in adolescents was 164,183 copies/ml (95% CI: 64,886 − 263,480 copies/ml, Fig. 2b.), more than three times the 50,000 copies/ml threshold. Cumulatively, nearly half (47·6% [79/166], 95% CI: 39·80–55·47) of newly infected adolescents were diagnosed with HIV RNA viral load above 50,000 copies/ml (see red reference line in Fig. 2b.). Disproportionately high viral load levels beyond 500,000 copies/ml were detected only in younger females adolescents aged 15–19 (1·81% [3/166]), (Fig. 2c. and Fig. 2d.) suggesting a possibility of acute infection at time of first diagnosis. Dissimilar to the HIV incidence pattern, males had higher proportions (59·3% [16/27], 95% CI: 38·8–77·61) of HIV RNA ≥ 50,000 copies/ml than females (45·3% [63/139], 95% CI: 36·9–54·0). There was also an observed predisposition for slightly more (48·2% [27/56], 95% CI: 34·6–62·0, Fig. 2d.) younger adolescents (15–19) to have had viral loads greater than 50,000 copies/ml relative to the older cohort 20–24 years (45·5%; [50/110], 95% CI: 35·9–55·3]).
Figure 2: Viral Load and CD4 T-Cell Count for New HIV Infection in Adolescents
[Figure 2 Here]
Figure 2a:
Figure 2b:
*Red reference line (yline) indicates the cut-of at HIV RNA 50,000 copies/ml
Figure 2c:
Figure 2d:
Immunological Status among New HIV Seroconverts
Study results further show an average CD4 + T-cell count of 479 cells/mm3 (95% CI: 445–513·6, Figs. 2b., 2c., and 3.) among newly HIV infected persons. Close to two-thirds (62% [103/166], 95% CI 54·20–69·46) had CD4 + T-cell count ≤ 500 cells/mm3 signaling poor immune repertoire (see Fig. 3. and Table 2.). There was a marginal variation in proportions by sex among individuals presenting with CD4 + T-cell count ≤ 500 cells/mm3. Specifically, 63·0% (17/27, 95% CI: 42·4–80·6) males virtually at par with females at 61·9% (86/139, 95% CI: 53·3–70·0), see Table 2.
Figure 3. CD4 T-Cell Count of Newly Infected Adolescents
[Figure 3 Here]
* Red reference line (xline) shows CD4 + T-cell count cut-off at 500 cells/mm 3 . The larger area on the left of the reference line reveals a greater proportion of adolescents presenting with decreased immunity at first diagnosis of a positive HIV serostatus.
On the other hand, age differentials revealed larger proportions (64·5% [71/110], 95% CI: 54·9–73·4) of older adolescents ages 20–24 years exhibiting diminishing immunological profiles of CD4 T- cell counts ≤ 500 cells/mm3 than younger adolescents 15–19 years (57·1% [32/56], 95% CI 43·2–70·3). Analysis to further understand the immunity status among seroconverts revealed that 29·5% (7/139 + 34/139, Table 2) females and 37% (7/27 + 3/27,) males had CD4 + T-cell count ≤ 350 cells/mm3. In addition, slightly more females (38·1%, 95% CI: 30·0–46·7) presented with stronger immunity at CD4 + T-cell count greater than 500 cells/mm3 compared to males (37·0%, 95% CI: 19·4–57·6) males. Table 2 below illustrates further details of immunological statuses of adolescents by different characteristics.
Table 2
Percentages of CD4 + T-Cell Count Levels for all Predictor Variables
Predictor | ≤ 200 cells/mm3 | 201–350 cells/mm3 | 351–500 cells/mm3 | > 500 cells/mm3 |
Total | 6.0 (10/166) | 24.7 (41/166) | 31.3 (52/166) | 38.0 (63/166) |
Sex | | | | | |
| Male | 11.2 (3/27) | 25.9 (7/27) | 25.9 (7/27) | 37.0 (10/27) |
| Female | 5.0 (7/139) | 24.5 (34/139) | 32.4 (45/139) | 38.1 (53/139) |
Age | | | | | |
| 15–19 | 7.1 (4/56) | 19.6 (11/56) | 30.4 (17/56) | 42.9 (24/56) |
| 20–24 | 5.5 (6/110) | 27.3 (30/110) | 31.8 (35/110) | 35.5 (39/110) |
Residence | | | | | |
| Rural | 5.7 (4/70) | 25.7 (18/70) | 31.4 (22/70) | 37.1 (26/70) |
| Urban | 6.3 (6/96) | 24.0 (23/96) | 31.3 (30/96) | 38.5 (37/96) |
Marital status | | | | | |
| Not married | 6.6 (7/106) | 18.9 (20/106) | 36.8 (39/106) | 37.7 (40/106) |
| Married | 5.0 (3/60) | 35.0 (21/60) | 21.7 (13/60) | 38.3 (23/60) |
Wealth quintile | | | | | |
| Lowest | 8.3 (1/12) | 33.3 (4/12) | 8.3 (1/12) | 50.0 (6/12) |
| Second | 4.5 (1/22) | 13.6 (3/22) | 50.0 (11/22) | 31.8 (7/22) |
| Middle | 2.8 (1/36) | 25.0 (9/36) | 33.3 (12/36) | 38.9 (14/36) |
| Fourth | 6.3 (3/48) | 35.4 (17/48) | 22.9 (11/48) | 35.4 (17/48) |
| Highest | 8.7 (4/46) | 17.4 (8/46) | 37.0 (17/46) | 37.0 (17/46) |
Education level | | | | | |
| No education | 0.0 (0/2) | 50.0 (1/2) | 50.0 (1/2) | 0.0 (0/2) |
| Primary | 1.9 (1/53) | 26.4 (14/53) | 24.5 (13/53) | 47.2 (25/53) |
| Secondary | 8.5 (9/106) | 23.6 (25/106) | 34.0 (36/106) | 34.0 (36/106) |
| Tertiary | 0.0 (0/5) | 20.0 (1/5) | 40.0 (2/5) | 40.0 (2/5) |
Active Syphilis | | | | | |
| Positive | 0.0 (0/12) | 25.0 (3/12) | 33.3 (4/12) | 41.7 (5/12) |
| Negative | 6.6 (10/154) | 24.7 (38/154) | 31.2 (48/154) | 37.7 (58/154) |
Hepatitis B Virus | | | | | |
| Positive | 0.0 (0/11) | 18.2 (2/11) | 36.4 (4/11) | 45.5 (5/11) |
| Negative | 6.5 (10/155) | 25.2 (39/155) | 31.0 (48/155) | 37.4 (58/155) |
Number of sex partners in last 12 months | | | | | |
| No partner | 3.3 (1/30) | 20.0 (6/30) | 33.3 (10/30) | 43.3 (13/30) |
| One partner | 5.9 (6/102) | 29.4 (30/102) | 31.4 (32/102) | 33.3 (34/102) |
| More than one | 5.3 (1/19) | 15.8 (3/19) | 42.1 (8/19) | 36.8 (7/19) |
Ever tested for HIV | | | | | |
| Yes | 5.3 (7/132) | 27.3 (36/132) | 31.8 (42/132) | 35.6 (47/132) |
| No | 8.8 (3/34) | 14.7 (5/34) | 29.4 (10/34) | 47.1 (16/34) |
Aware of HIV + status | | | | | |
| Aware | 12.5 (3/24) | 20.8 (5/24) | 33.3 (8/24) | 33.3 (8/24) |
| Not aware | 4.9 (7/142) | 25.4 (36/142) | 31.0 (44/142) | 38.7 (55/142) |
Co-Infection in New HIV-positives |
The study considered two common co-infections in HIV-positive adolescents; syphilis and HBV. Granted that syphilis, HBV and HIV can be acquired in similar ways, notwithstanding that syphilis and HBV are more infectious, co-infection occurs frequently (25, 26). Among adolescents with recent HIV infections, 7·2% (12/166), 95% CI: 3·8–12·3) and 6·6% (11/166), 95% CI: 3·3–11·5, Table 1) were diagnosed with syphilis and HBV co-infections, respectively. The risk of being diagnosed with a recent HIV infection significantly increased (IRR: 8·56, 95% CI: 2·94–24·93, P < 0·001, model 2) with a positive HBV test relative to those testing negative. To the contrary, in model 2, results suggest a near equal likelihood of HIV incidence among young people presenting with a seropositive syphilis test (IRR:1·23, 95% CI: 0·18–8·28, P = 0·84) compared with individuals testing negative, even though the statistic was not significant. We further observed that the two co-infections were uniquely detected in samples with HIV RNA above 10,000 copies/ml.
Knowledge of HIV-positive Status among Newly HIV Infected Adolescents
Further analysis shows that one in every five (19·3% [32/166], 95% CI: 13·57–26·11) adolescents testing with recent HIV infections in the survey were already aware of their HIV seropositive status. In spite of this verity, laboratory diagnostics did not detect ARVs in their blood samples an indication they had not initiated antiretroviral treatment. Statistically significant associations (= 5·659; P < 0·02) between knowledge of one’s HIV-positive status and age were also established. Older adolescents aged 20–24 years had greater awareness of their HIV-positive status (12·7% [21/166]; 95% CI 8·0–18·7) than younger adolescents aged 15–19 years (1·8% [3/166]; 95% CI 0·37–5·2) on awareness of an individual’s HIV-positive status.
[Figure 4a and 4b Here]
Even if 85·5% of participants did not know they had HIV, older adolescents were relatively less aware(38·7% [89/166]; 95% CI: 45·7–61·4) than younger adolescents (31·9% [53/166]; 95% CI: 24·9–39·6, Figs. 4a. and 4b.). Within age and sex categories, more younger males (88·9% [24/27], 95% CI 70·84–97·65, Fig. 4a.) and younger females (84·9% [118/139], 95% CI: 77·83–90·40) were unaware of their HIV-positive serostatus than their older ones. Besides sex, Figure:4b shows that adolescents in urban areas were less inclined to knowing they had the virus at 50% (83/166; 95% CI: 42·2–57·8) than those residing in rural areas (35·5%, 95% CI: 28·3–43·3).
Model Estimation Results
Table 3
Generalized Negative Binomial Regression Models for New HIV Infections in Adolescents
| Model 1† | Model 2†† | Model 3††† |
Predictors | | | |
IRRa | P-Value | 95% CI | IRR | P-Value | 95% CI | IRR | P-Value | 95% CI |
Age | 1·00069 | 0·000 | 1·000–1·001 | 1·06343 | 0·369 | 0·930–1·216 | 0·984041 | 0·837 | 0·844–1·147 |
Sex | | | | | | | | | |
Female | 2·27921 | 0·000 | 2·277–2·281 | 0·39227 | 0·011 | 0·190–0·809 | 0·486252 | 0·044 | 0·241–0·980 |
Male | Ref.b | .. | .. | Ref. | .. | .. | Ref. | | |
Active syphilisc | | | | | | | | | |
Positive | 0·292607 | 0·162 | 0·0522–1·640 | 1·21842 | 0·840 | 0·179–8·283 | 0·795605 | 0·735 | 0·212–2·990 |
Negative | Ref. | .. | .. | Ref. | .. | .. | Ref. | .. | .. |
Hepatitis B status | | | | | | | | | |
Positive | 7·66655 | 0·018 | 6·594 − 41·371 | 8·56390 | 0·000 | 2·942 − 24·931 | 6·86583 | 0·007 | 1·702 − 27·695 |
Negative | Ref. | .. | .. | Ref. | .. | .. | Ref. | | |
Marital status | | | | | | | | | |
Married | 4·07027 | 0·000 | 4·067 − 4·073 | 3·15757 | 0·037 | 1·071 − 9·313 | 2·76422 | 0·117 | 0·774–9·869 |
Not married | Ref. | .. | .. | Ref. | | | Ref. | | |
Residence | | | | | | | | | |
Rural | 2·37048 | 0·000 | 1·524–3·687 | 3·89909 | 0·001 | 1·762–8.631 | 3·10161 | 0·003 | 1·488–6·465 |
Urban | Ref. | .. | .. | Ref. | .. | .. | Ref. | | |
Wealth quintile | | | | | | | | | |
Lowest | Ref. | .. | .. | Ref. | .. | .. | - | | |
Second | 0·59644 | 0·182 | 0·279–1·274 | 1·01981 | 0·982 | 0·194–5·366 | - | | |
Middle | 0·24992 | 0·002 | 0·106–0·589 | 0·70233 | 0·65 | 0·153–3·233 | - | | |
Fourth | 0·33639 | 0·003 | 0·164–0·690 | 1·12396 | 0·874 | 0·264–4·779 | - | | |
Highest | 0·31899 | 0·002 | 0·156–0·654 | 2·42966 | 0·235 | 0·562 − 10·506 | - | | |
Education | | | | | | | | | |
No education | Ref. | .. | .. | Ref. | .. | .. | Ref. | | |
Primary | 5·14169 | 0·027 | 1.202–21·985 | 10·3383 | 0·001 | 2·681 − 39·859 | 5·891361 | 0·026 | 1·236 − 28·080 |
Secondary | 1·74790 | 0·442 | 0.421–7·260 | 7·35867 | 0·004 | 1·922 − 28·169 | 6·285652 | 0·014 | 1·449 − 27·263 |
Higher than secondary | 1·40491 | 0·671 | 0.292–6·761 | 8·99769 | 0·014 | 1·557 − 51·990 | 12·84462 | 0·006 | 2·091 − 78·997 |
Number of sex partners in past 12 months | | | | | | | | | |
One sex partner | Ref. | .. | .. | Ref. | .. | .. | Ref. | | |
No sex partner | 0·28871 | 0·011 | -1.296–0·164 | 0·79571 | 0·500 | 0·409–1·547 | 0·29033 | 0·477 | 0·362–1·608 |
More than one sex part. | -2·0569 | 0·000 | -3.081–1·0326 | 0·19566 | 0·002 | 0·068 − 0·559 | 0·20708 | 0·001 | 0·084 − 0·511 |
Ever tested for HIV | | | | | | | | | |
Yes | 1·56962 | 0·055 | 0.991–2·486 | 0·95721 | 0·915 | 0·429–2·135 | - | | |
No | Ref. | .. | .. | Ref. | .. | .. | - | | |
Aware of one's HIV + status | | | | | | | | | |
Yes | 1·25719 | 0·840 | 0·137 − 11·542 | - | | | - | | |
No | Ref. | .. | .. | - | | | - | | |
†Model 1 is a bivariate model unadjusted for other variables. It examines association between each predictor variable and the outcome variable. |
††Model 2 is a full controlled model with robust standard errors (SEs) in which inclusion into the model for all variables and covariates was contingent on passing the 0.2 p-value inclusion criterion in model 1are included. Hence, the variable, aware of one’s HIV status was excluded. |
†††Model 3 with robust SEs estimation include only statistically significant covariates. The restriction is relaxed for the four main predictors. Three variables, wealth quintile, ever tested for HIV and aware of one’s HIV-positive status are not included. |
-
a. IRR – These are the incidence rate ratios for the negative binomial regression, showing the rate at which new HIV infections occurred.
-
b. Ref – refers to the variable category used as reference in the specified regression model.
-
c. Active syphilis variable meant the result for the simultaneous presence of antibodies against non-Treponemal and Treponema pallidum antigens clinical test.
We found evidence that incidence of new HIV infections is significantly different between males and females. Increased likelihood of new HIV infection was significantly associated with adolescent girls than boys (IRR 2·30; 95% CI: 2·27–2·28), p < 0·001; Table 2, model1).
Younger participants had lower relative risk of new infection than older participants (IRR 1·00; 95% CI: 1·00–1·001), p < 0·001; Table 2, model1). However, associations between age and new HIV infections were completely attenuated when adjustment for covariates was performed (Table 2, model 2 and model 3). Further independent associations were seen in individuals with active syphilis infection (IRR 0·29; 95% CI: 0·05–1·64; model 1), individuals in lower brackets of the wealth quintile (IRR 0·59; 95% CI: 0·28–1·27; model 1) and those with more than one sexual partner in past 12 months (IRR 0·26; 95% CI: 0·09–0·77; model 1). Although mainly consistent in showing decreased likelihood of incidence of HIV infection, but most of these associations did not attain statistical significance.
High educational attainment was associated with high relative risk of new HIV infection by up to 12.8 times (95% CI: 2·09–79·00, p < 0·01; model 3) among those with higher than secondary education compared to individuals with no education. Other socio-behavioural covariates strongly associated with elevated relative risk were; adolescents aware of their HIV-positive status (IRR 1·26; 95% CI: 0·014–11·54; model 1), individuals that had previously tested for HIV (IRR 1·57; 95% CI: 0·99–2·49; model 1), those reporting one sex partner in past 12 months (IRR 2·07; 95% CI: 1·18–3·65; in all three models), those married or cohabiting (IRR 3·16; 95% CI: 1·07–9·31; model 2) and adolescents residing in rural areas (IRR 3·90; 95% CI: 1·76–8·63; model 2 and model 3). With an exception of the variables “aware of one’s HIV + status” and “ever tested for HIV”, all estimates indicate statistically significant associations.