Depression and PrEP uptake, interruption, and adherence among young women in Uganda

Figures & data

Figure 1. Distribution of PHQ-9 categories over time.

Table 1. Baseline characteristics of women in the Kampala Women’s Bone Study, by PHQ-9 depression score category and overall.

	No/Minimal (Score 0–4) (N = 330)	Mild (Score 5–9) (N = 122)	Moderate/Severe (Score 10+) (N = 47)	Total (N = 499)
Age (years), median (IQR)	20.0 (18.2, 21.0)	20.0 (19.0, 21.8)	20.0 (18.0, 21.0)	20.0 (18.5, 21.0)
Relationship status, n (%)
Single with no steady partner	41 (12.4%)	16 (13.1%)	5 (10.6%)	62 (12.4%)
Single with a steady partner	285 (86.4%)	104 (85.2%)	41 (87.2%)	430 (86.2%)
Married	3 (0.9%)	1 (0.8%)	1 (2.1%)	5 (1.0%)
Widowed, divorced or separated	1 (0.3%)	1 (0.8%)	0 (0.0%)	2 (0.4%)
Lives with partner, n (%)	14 (4.2%)	7 (5.7%)	3 (6.4%)	24 (4.8%)
Contraceptive choice, n (%)
DMPA	165 (50.0%)	79 (64.8%)	24 (51.1%)	268 (53.7%)
Condoms	165 (50.0%)	43 (35.3%)	23 (48.9%)	231 (46.3%)
Years of education completed, median (IQR)	11.0 (8.2, 13.0)	10.0 (7.0, 11.0)	11.0 (7.0, 12.0)	11.0 (8.0, 12.0)
Earns own income, n (%)	166 (50.3%)	69 (56.6%)	26 (55.3%)	261 (52.3%)
Partner provides financial support, n (%)	305 (92.4%)	113 (92.6%)	39 (83.0%)	457 (91.6%)
Travel time to research clinic, n (%)
<1 h	52 (15.8%)	19 (15.6%)	5 (10.6%)	76 (15.2%)
1–2 h	265 (80.3%)	98 (80.3%)	40 (85.1%)	403 (80.8%)
>2 h	13 (3.9%)	5 (4.1%)	2 (4.3%)	20 (4.0%)
Number of times had sex in past 3 months, median (IQR)	18.0 (8.0, 60.0)	24.0 (9.0, 71.5)	24.0 (9.0, 72.0)	20.0 (8.0, 60.0)
Any condomless sex in past 3 months, n (%)	210 (63.6%)	93 (76.2%)	34 (72.3%)	337 (67.5%)
Condom used at last sex, n (%)	199 (60.3%)	64 (52.5%)	27 (57.4%)	290 (58.1%)
Number of sexual partners in last 3 months, median (IQR)	2.0 (1.0, 3.0)	2.0 (1.0, 4.0)	2.0 (1.0, 4.0)	2.0 (1.0, 3.0)
Any new partners in last 3 months, n (%)	88 (26.7%)	56 (45.9%)	15 (31.9%)	159 (31.9%)
Any partner living with HIV, n (%)
Yes	3 (0.9%)	5 (4.1%)	1 (2.1%)	9 (1.8%)
No	125 (37.9%)	39 (32.0%)	13 (27.7%)	177 (35.5%)
Not Sure	201 (60.9%)	78 (63.9%)	33 (70.2%)	312 (62.5%)
Any alcohol in past month, n (%)	63 (19.1%)	41 (33.6%)	13 (27.7%)	117 (23.4%)
Any cigarette smoking in past month, n (%)	4 (1.2%)	1 (0.8%)	0 (0.0%)	5 (1.0%)
Any recreational drugs in past month, n (%)	8 (2.4%)	5 (4.1%)	3 (6.4%)	16 (3.2%)
Any abuse from any partner in past 3 months, n (%)	6 (1.8%)	14 (11.5%)	8 (17.0%)	28 (5.6%)
Experienced any social harm from current sexual partner(s), n (%)	2 (0.6%)	4 (3.3%)	4 (8.5%)	10 (2.0%)

Table 2. Association of depression with declining PrEP at enrollment.

	PrEP declined (n = 68)	PrEP initiated (n = 431)	Unadjusted analysis
	N (%) or median (IQR)	N (%) or median (IQR)	OR (95% CI)	p-value
Depression severity
None or minimal (score 0–4)	55 (16.7%)	275 (83.3%)	REF	–
Mild (score 5–9)	10 (8.2%)	112 (91.8%)	0.45 (0.22–0.91)	0.55
Moderate/severe (score 10+)	3 (6.4%)	44 (93.6%)	0.34 (0.10–1.14)	0.28
Alcohol use
No	60 (15.7%)	322 (84.3%)	REF	–
Yes	8 (6.8%)	109 (93.2%)	0.39 (0.18–0.85)	0.02
Recreational drug use
No	67 (13.9%)	416 (86.1%)	REF	–
Yes	1 (6.2%)	15 (93.8%)	0.41 (0.05–3.19)	0.40
Experienced IPV^a
No	68 (14.5%)	402 (85.5%)	REF	–
Yes	0 (0.0%)	28 (100%)	0.13 (0.00–0.57)	0.01

Note: OR: odds ratio; CI: confidence interval; PrEP: pre-exposure prophylaxis; IQR: interquartile range; PHQ-9: participant health questionnaire-9; IPV: intimate partner violence.

^a Calculated with exact methods due to small cell count.

Table 3. Association of depression with participant-driven PrEP interruption.

Depression category	PrEP interrupted	PrEP continued	Unadjusted analysis		Adjusted analysis^a
12 months of PrEP use
	Total Visits (n = 18)	Total Visits (n = 2017)	OR (95% CI)	p-value	OR (95% CI)	p-value
	N (%)	N (%)
None or minimal (score 0–4)	15 (1.0%)	1534 (99.0%)	REF	–	REF	–
Mild (score 5–9)	3 (0.8%)	399 (99.3%)	0.77 (0.22–2.67)	0.68	0.99 (0.38–2.56)	0.99
Moderate/ severe (score 10+)	0 (0%)	109 (100%)	0.00 (0.00–0.00)	<0.0001	0.00 (0.00–0.00)	*
24 months of PrEP use
	Total Visits n = 46	Total Visits n = 3128	OR (95% CI)	p-value	OR (95% CI)	p-value
	N (%)	N (%)
None or minimal (score 0–4)	42 (1.7%)	2464 (98.3%)	REF	–	REF	–
Mild (score 5–9)	4 (0.7%)	555 (99.3%)	0.42 (0.15–1.18)	0.1	0.54 (0.22–1.29)	0.17
Moderate/severe (score 10+)	0 (0%)	109 (100%)	0.00 (0.00–0.00)	<0.0001	0.00 (0.00–0.00)	*

Note: OR: odds ratio; CI: confidence interval; PrEP: pre-exposure prophylaxis.

^a Adjusted for whether partner’s HIV status is positive or unknown and whether participant has had new partners in past 3 months

*Unestimable due to small sample size.

Table 4. Association of depression with lower PrEP adherence detected via Wisepill and DBS during follow-up.

Depression Category	Total Occurrences N (%)		Unadjusted analysis		Adjusted analysis^a
12 months of PrEP use
	Lower PrEP adherence (n = 1065 visits)	Higher PrEP adherence (n = 385 visits)	OR (95% CI)	p-value	OR (95% CI)	p-value
None or minimal (score 0–4)	770 (72.4%)	293 (27.6%)	REF	–	REF	–
Mild (score 5–9)	243 (76.9%)	73 (23.1%)	1.26 (0.92, 1.75)	0.15	1.21 (0.87, 1.69)	0.63
Moderate/severe (score 10+)	52 (73.3%)	19 (26.8%)	1.04 (0.59, 1.83)	0.89	0.86 (0.47, 1.59)	0.25
24 months of PrEP use
	Lower PrEP adherence (n = 1548 visits)	Higher PrEP adherence (n = 479 visits)	OR (95% CI)	p-value	OR (95% CI)	p-value
None or minimal (score 0–4)	1198 (75.9%)	381 (24.1%)	REF	–	REF	–
Mild (score 5–9)	292 (78.7%)	79 (21.3%)	1.18 (0.87, 1.58)	0.28	1.10 (0.81, 1.49)	0.54
Moderate/severe (score 10+)	58 (75.3%)	19 (24.7%)	0.97 (0.56, 1.70)	0.92	0.73 (0.40, 1.33)	0.3
Subset with DBS
	DBS Low PrEP adherence (n = 342 visits)	DBS High PrEP adherence (n = 4 visits)	OR (95% CI)	p-value	OR (95% CI)	p-value
None or minimal (score 0–4)	277 (99.3%)	2 (0.7%)	REF	–	REF	–
Mild (score 5–9)	59 (96.7%)	2 (3.3%)	1.02 (0.98–1.07)	0.3	1.03 (0.98–1.08)	0.3
Moderate/ severe (score 10+)	6 (100%)	0 (0%)	0.99 (0.98–1.00)	0.16	0.98 (0.97–1.00)	0.09

Note: OR: odds ratio; CI: confidence interval; PrEP: pre-exposure prophylaxis; DBS: dried blood spot.

^a Adjusted for whether participant has reported condomless sex with primary partner in past 3 months.

Table 5. Comparison between adherence detected via DBS and Wisepill.

	Wisepill Adherent	Wisepill Non-Adherent
DBS Adherent	4	0
DBS Non-Adherent	66	241

Data availability statement

Data are available upon request from [email protected].