Table II.
Bivariate and multivariate logistic regression models predicting Unknown HIV Status among men who have sex with men (MSM)
| Bivariate models | Final adjusted model* | |||||
|---|---|---|---|---|---|---|
| Participant Characteristic | OR | 95% CI | p-value | OR | 95% CI | p-value |
| Age (years) | 0.97 | 0.96–0.97 | <.001 | 0.96 | 0.96–0.97 | <.001 |
| Race | ||||||
| White | Ref. | -- | -- | Ref. | -- | -- |
| American Indian/Native Alaskan | 0.85 | 0.42–1.70 | .638 | 0.66 | 0.31–1.39 | .273 |
| Asian/Hawaiian/Pacific Islander | 1.35 | 0.95–1.90 | .091 | 1.21 | 0.83–1.77 | .332 |
| Black | 0.88 | 0.64–1.21 | .431 | 0.56 | 0.39–0.79 | .001 |
| Hispanic/Latino | 1.02 | 0.86–1.21 | .857 | 0.72 | 0.59–0.87 | .001 |
| Other/multiple | 0.72 | 0.50–1.03 | .070 | 0.48 | 0.32–0.71 | <.001 |
| Region | ||||||
| Northwest | 1.03 | 0.85–1.25 | .739 | |||
| Midwest | 1.37 | 1.12–1.63 | <.001 | |||
| South | 1.02 | 0.87–1.19 | .858 | |||
| West | Ref. | -- | --- | |||
| US Territory | 0.65 | 0.08–5.10 | .684 | |||
| Education | ||||||
| Less than high school graduate | 9.75 | 7.15–13.32 | <.001 | 5.45 | 3.86–7.70 | <.001 |
| High school graduate/GED | 4.09 | 3.41–4.91 | <.001 | 2.58 | 2.11–3.15 | <.001 |
| Some college/Technical degree | 1.99 | 1.74–2.28 | <.001 | 1.64 | 1.42–1.89 | <.001 |
| College grad graduate and beyond | Ref. | -- | -- | Ref. | -- | -- |
| Rural Residence | ||||||
| Yes | 1.81 | 1.60–2.04 | <.001 | 1.54 | 1.35–1.75 | <.001 |
| No | Ref. | -- | -- | Ref. | -- | -- |
| Health Insurance | ||||||
| Private insurance | Ref. | -- | -- | |||
| Public insurance | 1.49 | 1.22–1.82 | <.001 | |||
| Other/Multiple insurance | 1.44 | 1.18–1.77 | <.001 | |||
| No insurance | 1.04 | 0.84–1.29 | .725 | |||
| Visited healthcare provider in last 12 months | ||||||
| Yes | 0.56 | 0.47–0.65 | <.001 | 0.57 | 0.48–0.67 | <.001 |
| No | Ref. | -- | -- | Ref. | -- | -- |
| Sexual Identity | ||||||
| Heterosexual | 2.45 | 1.74–3.43 | <.001 | |||
| Homosexual | Ref. | -- | -- | |||
| Bisexual | 0.45 | 0.39–0.52 | <.001 | |||
| Disclosure status (“outness”) | ||||||
| Disclosed | Ref. | -- | -- | Ref. | -- | -- |
| Not disclosed | 4.79 | 3.96–5.79 | <.001 | 3.70 | 2.99–4.59 | <.001 |
| Ever had sex with a female | ||||||
| Yes | 1.77 | 1.51–2.07 | <.001 | |||
| No | Ref. | -- | -- | |||
| Sexual position | ||||||
| Identifies as top | 0.90 | 0.76–1.06 | .194 | 1.01 | 0.84–1.20 | .957 |
| Identifies as bottom | 1.41 | 1.22–1.64 | <.001 | 1.45 | 1.24–1.70 | <.001 |
| Identifies as versatile | Ref. | -- | -- | Ref. | -- | -- |
| Number of sex partners last 12 months | 0.97 | 0.97–0.98 | <.001 | |||
| Type of sex last 12 months | ||||||
| Oral only | 0.77 | 0.63–0.94 | .011 | 0.83 | 0.66–1.04 | .107 |
| Anal only | 0.74 | 0.33–1.63 | .450 | 0.43 | 0.17–1.12 | .085 |
| Both oral and anal | 0.32 | 0.27–0.38 | <.001 | 0.28 | 0.23–0.34 | <.001 |
| None | Ref. | -- | -- | Ref. | -- | -- |
| Condomless anal intercourse (CAI) in last 12 months (n=8,191) | ||||||
| CAI with partner of unknown HIV status | 1.28 | 1.02–1.61 | .034 | |||
| No CAI with partner of unknown status | Ref. | -- | -- | |||
| CAI with HIV-positive partner | 0.34 | 0.20–0.55 | <.001 | |||
| No CAI with HIV-positive partner | Ref. | -- | -- | |||
| CAI with HIV-negative partner | 0.68 | 0.42–1.10 | .115 | |||
| No CAI with HIV-negative partner | Ref. | -- | -- | |||
| Sexually transmitted infection in last 12 months | ||||||
| Yes | 0.35 | 0.25–0.49 | <.001 | |||
| No | Ref. | -- | -- | |||
| Serosorting - only sex with HIV-negative men | ||||||
| Yes | 0.76 | 0.67–0.85 | <.001 | |||
| No | Ref. | -- | -- | |||
| Ever injected drugs | ||||||
| Yes | 0.43 | 0.27–0.67 | <.001 | |||
| No | Ref. | -- | -- | |||
| Non-injection drug use in last 12 months | ||||||
| Yes | 0.76 | 0.66–0.87 | <.001 | |||
| No | Ref. | -- | -- | |||
| Peers get HIV tested every 3–6 months | ||||||
| Neither Agree/Disagree | Ref. | -- | -- | |||
| Strongly Agree | 0.46 | 0.32–0.67 | <.001 | |||
| Agree | 0.70 | 0.57–0.87 | <.001 | |||
| Disagree | 0.75 | 0.62–0.90 | .003 | |||
| Strongly Disagree | 1.05 | 0.81–1.37 | .712 | |||
*
logistic regression model adjusting for all other covariates using Full Information Maximum Likelihood estimation