Table 4B.
Zero-inflated negative binomial regression for the count of risk exposures in the past 30 days for housed individuals.
| Outcomes | Variables | Coefficient (SE) | Odds | t | p-value |
|---|---|---|---|---|---|
| Number of times used crack, coke or heroin in last 30 days | |||||
| Zero count logistic model | Male | −1.14 (0.31) | 0.32 | −3.68 | <.001 |
| Less than HS education | −1.17 (0.32) | 0.31 | −3.71 | <.001 | |
| Ever told by a doctor that I have HIV/AIDS | 0.75 (0.32) | 2.12 | 2.35 | .019 | |
| Number of others live here | 0.15 (0.05) | 1.16 | 3.16 | .002 | |
| Drug users live with you | −1.85 (0.65) | 0.16 | −2.86 | .004 | |
| Positive count negative binomial model | Male | 0.39 (0.21) | 1.86 | .06 | |
| Contribute money to household expenses | −0.75 (0.21) | −3.61 | <.001 | ||
| Living with family, sex partner of friend | 0.76 (0.21) | 3.68 | <.001 | ||
| Total number of sex partners in last 30 days1 | |||||
| Positive count negative binomial model | Age (years) | −0.03 (0.01) | −3.15 | .002 | |
| Likelihood of living in the same place in 3 months | 0.18 (0.08) | 2.41 | .016 | ||
| Contribute money to household expenses | 0.33 (0.16) | 2.00 | .045 | ||
| Live with drug users | 0.03 (0.30) | 0.12 | .91 | ||
| Longest prison sentence | −0.01 (0.01) | −1.43 | .15 | ||
| Living with family, sex partner of friend | 0.04 (0.23) | 0.19 | .85 | ||
| Live with drug users * Longest prison sentence | 0.04 (0.01) | 3.15 | .002 | ||
| Living with family, sex partner of friend * Longest prison sentence | 0.03 (0.01) | 3.04 | .002 | ||
| Number of different people having sex with without condoms in last 30 days2 | |||||
| Positive count negative binomial model | Age (years) | −0.04 (0.01) | −3.32 | <.001 | |
| Ever told by a doctor that I have HIV/AIDS | −0.71 (0.30) | −2.39 | .017 | ||
| Number of moves in the past 6 months | 0.74 (0.47) | 1.55 | .12 | ||
| Number of others live here | −0.12 (0.06) | −1.94 | .052 | ||
| Conflict over household expenses | 0.84 (0.29) | 2.87 | .004 | ||
| Age * Number of moves in the past 6 months | −0.03 (0.02) | −2.23 | .026 |
Outcome variables: Number of times smoked crack, or sniffed/injected coke, or sniffed/injected heroin in the last 30 days (Fitted model: Likelihood Ratio Chi-square test=137.98, df=8; p=<.0001; Schwarz Bayesian Criterion(SBC)=1846); Total number of sex partners in last 30 days (Fitted model: Likelihood Ratio Chi-square test=113.39, df=8; p<.0001; SBC=754); Number of different people having sex with without condoms in the last 30 days (Fitted: Likelihood Ratio Chi-square test=51.90, df=6; p<.0001; SBC=421).
SE = Standard Error of coefficient estimate.
Odds = Exponentiated coefficients of the logistic model for zero inflation represent the odds of a zero count.
Number of cases =293 (excludes 81 living on street or in shelter and 18 living in SRO or half-way houses).
Housing status groups: Living with family, sex partner of friend; living in own apartment with a rental subsidy or supportive housing; living in own apartment without a rental subsidy (reference).
Predictors considered at beginning of the model: age (years); black race (y/n); male gender (y/n); less than HS education (y/n); diagnosed with mental illness (y/n); told by doctor that I have HIV/AIDS (y/n); feel safe in the place I live (scale); perceived gang activity in neighborhood (scale); experienced violence at the location I live (y/n); likelihood of living at the same place in 3 months (scale); number of moves in the last 6 months; number of positive reasons for moving; number of negative reasons for moving; did you pay rent (y/n); housing paid for by subsidies (y/n); rent-to-income ratio; evicted (y/n); live with non-relatives (y/n); number of others who live here; contributes money toward household expenses (y/n); conflicts over household expenses (y/n); drug users live with you (y/n); shares drugs to stay at place I live (y/n); have sex to stay at place I live (y/n), longest prison sentence (months); longest homeless period (months); living with family, sex partner of friend (y/n); living in own apartment with a rental subsidy or in supportive housing (y/n).
A zero count logistic regression component was not fit because there is no evidence of an excess in number of zero counts for the number of sex partners in the last 30 days.
A zero count logistic regression component was not fit because there is no evidence of an excess number of zero counts for number of partners having condomless sex with in the last 30 days.