Skip to main content

Table 2 Performance of several risk scores for AHIa

From: Development and validation of a risk score to assist screening for acute HIV-1 infection among men who have sex with men

Column 1

2

3

4

5

6

7

8

Risk score

Cutoffb

Seroconversion visits among visits with a score of at least the cutoffc

Seroconversion visits among visits with a score below the cutoffc

Sensitivity %

(95% CI)

Specificity %

(95% CI)

Overall AUC

(95% CI)

% to be tested

Development of a risk score in Amsterdam Cohort Studies

 A. Initial risk scored,e

  -CLAI a conditionf

2

80/1888

95/15555

45.7 (38.2–53.4)

89.5 (89.1–90.0)

0.70 (0.66–0.74)

10.8

 B. 14 symptoms scored as beta coefficient in symptom based modeld,g

  -CLAI a conditionf

0.1

90/2342

85/15101

51.4 (43.8–59.0)

87.0 (86.4–87.5)

0.71 (0.67–0.74)

13.4

 C. 14 symptoms scored as beta coefficient in symptom based modelg,h

  -CLAI not a condition

0.2

116/4430

59/13012

66.3 (58.8–73.2)

75.0 (74.4–75.7)

0.74 (0.70–0.78)

25.4

 D. 14 symptoms and 3 risk factors, scored as beta coefficient in combined modeli,j

1.3

111/4334

24/10858

82.2 (74.7–88.3)

72.0 (71.2–72.7)

0.83 (0.80–0.87)

28.5

 E. 4 symptoms and 3 risk factors scored as beta coefficient in combined modelk,j

1.5

103/3675

32/11517

76.3 (68.2–83.2)

76.3 (75.6–77.0)

0.82 (0.79–0.86)

24.2

Validation of optimal risk score in Multicenter AIDS Cohort Study

 E. 4 symptoms and 3 risk factorsk, scored as beta coefficient

1.5

77/3779

60/29274

56.2 (47.5–64.7)

88.8 (88.4–89.1)

0.78 (0.74–0.82)

11.4

  1. AHI acute HIV-1 infection, CI confidence interval, CLAI condomless anal intercourse
  2. aBased on imputed data using the last observation carried forward method
  3. bBased on Youden-index, with the exception for risk score A: pre-defined cutoff was 2
  4. cDue to missing values, the denominators of columns 3 and 4 do not add up to 17,446. Data were missing for the following variables: fever, 1 missing; genital ulcers, 3 missings; gonorrhea, 2072 missings; receptive condomless anal intercourse, 162 missings; >5 sexual partners, 56 missings.
  5. dDue to 3 records with missing values, these risk score evaluations are based on 17,443 records
  6. e14 symptoms scored as case definition, but genital warts and oral ulcers not included due to large number of missing variables; score of 1: fatigue, headache, myalgia, nausea, night sweats, pharyngitis, or rash; score of 2: diarrhea, fever, genital ulcers, lymphadenopathy, oral thrush, oral ulcers, vomiting, or weight loss
  7. f112/175 seroconverters reported condomless anal intercourse in the preceding 6 months
  8. gIn a multivariable model including 14 symptoms (Table 3, column 6): diarrhea, fatigue, fever, genital ulcers, headache, lymphadenopathy, myalgia, nausea, night sweats, oral thrush, rash, sore throat, vomiting, weight loss
  9. hDue to 4 records with missing values, these risk score evaluations are based on 17,442 records
  10. iIn a multivariable model including 14 symptoms and 3 risk factors (Table 3, column 8): diarrhea, fatigue, fever, genital ulcers, headache, lymphadenopathy, myalgia, nausea, night sweats, oral thrush, rash, sore throat, vomiting, weight loss, and in the preceding 6 months: gonorrhea, receptive condomless anal intercourse, >5 sexual partners
  11. jDue to 2254 records with missing values, these risk score evaluations are based on 15,192 records
  12. kFever, lymphadenopathy, oral thrush, weight loss, and in the preceding 6 months: gonorrhea, receptive condomless anal intercourse, >5 sexual partners
Back to article page

BMC Infectious Diseases

ISSN: 1471-2334

Contact us