HIV fatalism and engagement in transactional sex among Ugandan fisherfolk living with HIV

Figures & data

Table 1. Sample characteristics and descriptive statistics, Uganda 2015, N = 91.

	n (%)	Mean (SD)
Demographics
HIV testing type
Home	60 (65.9%)
Event	31 (34.06%)
Sex
Female	51 (56.00%)
Male	40 (44.00%)
Age		31.93 (8.06)
Education
Secondary	14 (15.40%)
Primary or less	77 (84.60%)
Monthly household expenditure
15 USD or greater	52 (57.10%)
14 USD or less	39 (42.90%)
Occupation
Fisherman	28 (30.76%)
Businessman/woman	25 (27.47%)
Fishing support services	7 (7.69%)
Sex worker	4 (4.39%)
Other	18 (19.78%)
Unemployed	9 (9.89%)
Marital status
Married and/or living together	61 (67.00%)
Not married, not living with someone	20 (22.00%)
Separated/divorced	10 (11.00%)
Psychosocial factors
Internalised HIV stigma (possible range: 0–4)		0.98 (0.81)
Anticipated HIV stigma (possible range: 0–4)		1.26 (0.95)
Perceived poor health (possible range: 0–4)		1.14 (1.50)
Structural factors
Food insecurity in the last 30 days
Yes	57 (62.60%)
No	34 (37.40%)
Mobility (travel away from community)
Yes	51 (56.0%)
No	40 (43.90%)
Structural barriers to HIV care (possible range: 0–4)		1.15 (0.49)
HIV risk behaviour
Condom use at last sex
Yes	14 (15.40%)
No	77 (84.60%)
Engaged in transactional sex in the last year (gave money or received money for sex)
Yes	25 (27.50%)
No	66 (72.50%)
HIV fatalism (possible range: 0–3)		1.50 (0.51)
Life is risky and getting HIV is just one of those things.	59 (64.84%)
It is better not to think about HIV and enjoy your life.	45 (49.45%)
I do not care that I have HIV because it is just another disease.	41 (45.05%)
I am more worried about dying from drowning, hunger or something other than HIV.	43 (47.25%)
It was just a matter of time before I got HIV.	35 (38.46%)
I will only worry about having HIV when I start to get ill.	31 (34.07%)
Before testing positive, I had given up trying to protect myself from HIV.	27 (29.67%)

Notes: SD = Standard deviation; Mean score of HIV Fatalism scale is reported, and the percentage of participants reporting ‘agree’ or ‘strongly agree’ for each individual item.

Table 2. Results of bivariate generalised logistic regression model for correlates of engagement in transactional sex in the last year, adjusted for HIV test site and multivariate logistic regression model for correlates of engagement in transactional sex in the last year, Uganda 2015 (N = 91).

	Bivariate logistic regression			Multivariate logistic regression
	AOR (95% CI)	x^2	p	AOR (95% CI)	x^2	p
HIV fatalism	2.65 (1.01–6.92)	3.96	0.05	3.07 (1.02–9.23)	3.98	0.04
Demographic factors
Sex
Female	0.67 (0.26–1.72)	0.69	0.41	0.21 (0.04–1.14)	3.26	0.07
Male (reference)
Age	1.00 (0.94–1.06)	0.03	0.87	0.95 (0.88–1.03)	1.60	0.21
Education
Secondary	4.08 (0.48–34.54)	1.66	0.20	0.65 (0.13–3.27)	0.28	0.60
Primary or less (reference)
Monthly household expenditure
15 USD or greater	1.26 (0.41–3.92)	0.16	0.69	0.71 (0.17–3.06)	0.21	0.65
14 USD or less (reference)
Occupation
Fishermen	0.82 (0.30–2.28)	0.14	0.70	0.29 (0.05–1.72)	1.86	0.17
Other (reference)
Psychosocial factors
Internalised HIV stigma	1.01 (0.57–1.77)	0.00	0.98	1.04 (0.48–2.24)	0.01	0.93
Anticipated HIV stigma	0.82 (0.49–1.39)	0.55	0.46	0.62 (0.31–1.28)	1.67	0.20
Structural factors
Food insecurity
Yes	2.88 (0.98–8.45)	3.70	0.06	2.57 (0.76–8.71)	2.31	0.13
No (reference)
Mobility (travel away from community)
Yes	0.78 (0.30–1.99)	0.28	0.60	0.49 (0.16–1.52)	1.51	0.22
No (reference)
Covariate
HIV testing type
Home	2.60 (0.87–7.79)	2.91	0.09	3.16 (0.61–16.24)	1.89	0.17
Event (reference)

Notes: HIV test site refers to the type of testing event in which participants were recruited for participation (home-based HIV testing or community-based event testing); AOR = Adjusted odds ratio; x² = Wald chi square; Bivariate models are adjusted for HIV testing site only, multivariate model are adjusted for variables included in the model; Bolded text is used to indicate statistically significant associations at the p < 0.10 level.

Table 3. Results of bivariate generalised linear regression model for correlates of HIV fatalism, adjusted for HIV test site and multivariate generalised linear regression for correlates of HIV fatalism, Uganda 2015 (N = 91).

	Bivariate linear regression			Multivariate linear regression
Model	β	Std. Error	p	Β	Std. Error	p
Demographic factors
Sex
Female	0.12	0.11	0.25	−0.01	0.17	0.96
Male (reference)
Age	−0.001	0.01	0.93	0.00	0.01	0.58
Education
Secondary	−0.01	0.15	0.94	−0.08	0.16	0.61
Primary or no schooling (reference)
Monthly household expenditure
15 USD or greater	0.30	0.12	0.01	0.24	0.14	0.09
14 USD or less (reference)
Occupation
Fishermen	−0.18	0.12	0.14	−0.67	0.19	0.73
Other (reference)
Psychosocial factors
Internalised HIV stigma	0.14	0.07	0.03	0.08	0.08	0.32
Anticipated HIV stigma	0.05	0.05	0.37	−0.00	0.07	0.97
Perceived poor health	0.11	0.04	0.008	0.07	0.04	0.11
Structural factors
Food insecurity (prior 30 days)
Yes	0.17	0.11	0.13	0.10	0.12	0.37
No (reference)
Mobility (travel away from community)
Yes	0.22	0.11	0.03	0.19	0.11	0.10
No (reference)
Structural barriers to HIV care	0.35	0.10	0.001	0.26	0.12	0.04
Covariate
HIV testing type
Home	0.14	0.11	0.21	−0.81	0.17	0.64
Event (reference)

Notes: HIV test site refers to the type of testing event in which participants were recruited for participation (home-based HIV testing or community-based event testing); β = Beta, Std. error = standard error; Bivariate models are adjusted for HIV testing site only, multivariate model are adjusted for all variables statistically significant at the p < 0.10 level in bivariate models; Bolded text is used to indicate statistically significant associations at the p < 0.10 level.