Unadjusted and adjusted associations with recent violence across a range of perpetrators reported by street-based sex workers
| Physical or sexual violence from clients | Any violence from others | Depression/Anxiety (cut-point >5) | ||||
| Unadjusted | Adjusted | Unadjusted | Adjusted | Unadjusted | Adjusted | |
| Individual/Partnership-level factors | GEE OR (95% CI) | GEE aOR (95% CI) | GEE OR (95% CI) | GEE aOR (95% CI) | GEE OR (95% CI) | GEE aOR (95% CI) |
| People of colour (vs white) | 0.62 (0.26 to 1.49) | – | 1.02 (0.45 to 2.33) | – | 0.60 (0.25 to 1.48) | – |
| Bisexual, homosexual, any other term (vs heterosexual) | 1.73 (0.67 to 4.49) | – | 1.20 (0.49 to 2.97) | – | 2.82 (1.14 to 6.96) | 3.55 (1.30 to 9.71) |
| Unstable residency status (vs permanent/UK resident)†† | 0.70 (0.17 to 2.92) | – | 0.68 (0.16 to 2.79) | – | 0.77 (0.18 to 3.33) | – |
| Primary or secondary education (vs further education) | 1.17 (0.54 to 2.50) | – | 0.86 (0.40 to 1.85) | – | 1.30 (0.53 to 3.16) | – |
| Partner supplies drugs | 1.12 (0.48 to 2.60) | – | 2.11 (0.94 to 4.72) | |||
| Daily crack or heroin use | 1.56 (0.66 to 3.69) | 1.18 (0.41 to 3.43) | 2.18 (0.91 to 5.21) | 2.78 (0.91 to 8.55) | 1.40 (0.51 to 3.83) | 0.68 (0.22 to 2.06) |
| Any intimate partner violence in last 6 months | 2.12 (0.91 to 4.96) | – | 2.87 (1.35 to 6.11) | 4.00 (1.64 to 9.72) | – | – |
| Physical/Sexual violence from clients in last 6 months | Not included | 2.80 (1.16 to 6.80) | – | 2.30 (1.18 to 4.48) | 2.55 (1.10 to 5.91) | |
| Limited/Severely limited by disability (vs no disability) | Not included* | Not included* | 3.21 (1.52 to 6.78) | 3.85 (1.49 to 9.95) | ||
| Workplace factors | ||||||
| Provides services in vehicles | 2.93 (1.19 to 7.20) | – | Not included* | Not included* | ||
| Duration in sex work ≥15 years† | 0.61 (0.26 to 1.41) | 0.43 (0.16 to 1.15) | 0.89 (0.42 to 1.88) | 1.02 (0.41 to 2.53) | 0.77 (0.37 to 1.62) | 0.73 (0.28 to 1.88) |
| Number of days worked in the last week >5 | 4.55 (1.89 to 10.97) | 3.04 (1.16 to 7.96) | 1.23 (0.57 to 2.67) | – | Not included* | |
| Always work alone | 0.93 (0.35 to 2.42) | – | Not included* | 0.87 (0.41 to 1.85) | – | |
| Always work in areas where there is CCTV | 0.99 (0.36 to 2.77) | – | Not included* | Not included* | ||
| Always screen and refuse clients | 1.12 (0.51 to 2.42) | – | Not included* | Not included* | ||
| Always work in well-lit areas | 0.89 (0.34 to 2.35) | – | Not included* | Not included* | ||
| Structural variables— law enforcement‡ | ||||||
| Displaced from area by police in last 6 months | 5.24 (2.18 to 12.6) | 4.35 (1.36 to 13.90) | ||||
| Arrest for any reason in last 6 months | 2.72 (1.18 to 6.27) | 2.77 (1.11 to 6.94) | ||||
| Denied access to criminal justice§ | 2.91 (1.23 to 6.86) | 1.95 (0.74 to 5.15) | ||||
| Structural variables—other | ||||||
| Homeless in the last 4 weeks‡‡ | 2.75 (1.32 to 5.71) | 1.87 (0.74 to 4.74) | 1.48 (0.70 to 3.14) | 0.88 (0.36 to 2.14) | 2.47 (1.05 to 5.80) 0 | 2.19 (0.81 to 5.88) |
| Accessed sex worker support project in last 6 months | 2.55 (1.13 to 5.72) | 3.54 (1.27 to 9.89) | 1.08 (0.50 to 2.34) | – | 0.83 (0.42 to 1.64) | – |
| Unmet mental health support need in last 6 months | 1.65 (0.69 to 3.91) | – | 2.20 (1.03 to 4.68) | – | 1.11 (0.62 to 2.00) | – |
| Find it difficult to pay usual expenses (at time of survey) | 2.86 (1.32 to 6.23) | 4.66 (1.64 to 13.24) | 0.86 (0.39 to 1.87) | – | 1.80 (0.88 to 3.70) | – |
| Sensitivity analyses¶ | ||||||
| Ever experienced violence from police | 3.24 (1.60 to 6.57) | 3.77 (1.46 to 9.73) | 1.67 (0.83 to 3.33) | – | ||
| Client arrested in last 6 months | 2.63 (0.99 to 6.96) | 3.61 (1.11 to 11.78) | ||||
| Referred to services in last 6 months | 2.05 (0.73 to 5.73) | 2.91 (1.01 to 8.37) | ||||
| Police presence deters punters** | 2.65 (1.27 to 5.51) | 2.32 (1.04 to 5.16) | ||||
| Police presence causes rushed negotiations** | 3.00 (1.41 to 6.37) | 4.15 (1.84 to 9.39) | ||||
– denotes variables excluded in adjusted models not significant in adjusted models at p<0.05 and not a priori confounders.
*Not considered a potential confounder.
†Age or duration of sex work were selected as a priori confounders on the basis of quasi-AIC (QIC) evaluations.
‡Multivariable models adjust for one policing variable.
§Denied access to justice was defined as not reporting an episode of violence to police or reported violent episode but the police either arrested the sex worker or failed to take the report seriously.
¶Sensitivity analyses tested all policing variables in separate models adjusted for the variables shown excluding other policing variables. Only those with significant effect sizes (p<0.05) in adjusted analyses are displayed.
**Response to two-part question about how police activity in the last 6 months in the area affects work.
††Unstable defined as overseas national/asylum seeker/unknown or illegal migration status (vs permanent residency/UK national/settled status).
‡‡Homelessness defined as sleeping rough or living in unstable accommodation (eg, parent’s or friend’s home, sheltered or homeless accommodation).
GEE, generalised estimating equation.