The opioid crisis is driving mortality among under-served people living with HIV in British Columbia, Canada

BMC Public Health

Table 2 LISA study participants mortality outcomes as of June 30, 2017 (n = 910)

Variable	Alive (n = 716)	Died (n = 194)	p-value
Age, at survey
Median (Q1, Q3)	45 (39, 50)	47 (41, 53)	0.001*
Calendar year of ART initiation			0.149
< 1996	29 (4.1%)	< 5^a
1996–2004	419 (58.5%)	124 (63.9%)
> 2004	268 (37.4%)	65–70^a
Gender
Women	197 (27.5%)	60 (30.9%)	0.349
Men	519 (72.5%)	134 (69.1%)
Ethnicity
White/Caucasian	474 (66.2%)	117 (60.3%)	0.009*
Asian	20 (2.8%)	< 5^a
Indigenous	180 (25.2%)	69 (35.6%)
Other	42 (5.9%)	< 5^a
Current tobacco use
Yes	442 (62.0%)	154 (79.8%)	< 0.001*
No	271 (38.0%)	39 (20.2%)
Current excessive alcohol use (CAGE)			< 0.001*
Yes	338 (47.7%)	121 (63.4%)
No	370 (52.3%)	70 (36.6%)
Current injection drug use			< 0.001*
Yes	137 (19.2%)	65 (33.7%)
No	576 (80.8%)	128 (66.3%)
Stable housing
Yes	519 (72.6%)	101 (52.1%)	< 0.001*
No	196 (27.4%)	93 (47.9%)
Food security
Yes	258 (36.1%)	46 (23.8%)	0.001*
No	456 (63.9%)	147 (76.2%)
Location of residence
Rural	20 (2.8%)	< 5^a
Urban/Suburban	686 (97.2%)	185–195	0.443
CD4 cell count at ART initiation
cells/mm³	370 (240, 540)	270 (180, 450)	< 0.001*

Median (Q1, Q3) or n (%) presented
*indicates statistical significance at p ≤ 0.05
^a For confidentiality and disclosure control measures, exact counts for cells with values n < 5 were suppressed. In accordance with strategies employed by Lofters et al. [30], if only one cell was n < 5 then one additional category is masked by providing a range (with no percentages available). Disclosure controls are only applied for descriptive tabulations and exact counts were used for analyses

ISSN: 1471-2458