A total of 633 participants completed the online survey. One participant was excluded from the database for not meeting the inclusion criteria, and a further 12 participants (4 females, 8 male) were removed from the data set due to having missing data values for the HADS. The remaining 620 participants (308 females, 306 male) had a mean age of 22.47 years (SD=3.29; range 18-51). Table 1 shows the full demographic information for the sample.
Table 1. Socio-demographic characteristics of the sample
Gender
|
|
Male
|
306 (49.40%)
|
Female
|
308 (49.70%)
|
Other
|
1 (0.20%)
|
Prefer not to say
|
5 (0.80%)
|
Age
|
|
18 – 19
|
39 (6.30%)
|
20 – 29
|
511 (82.40%)
|
30+
|
17 (3.00%)
|
Prefer not to say
|
53 (8.50%)
|
Mean age (SD)
|
22.47 (SD 29.00)
|
Marital status
|
|
Single
|
596 (96.10%)
|
Married
|
17 (2.70%)
|
Divorced
|
1 (0.20%)
|
Prefer not to say
|
6 (1.00%)
|
Year of study
|
|
Second year
|
256 (41.30%)
|
Third year
|
217 (35.00%)
|
Fourth year
|
83 (13.40%)
|
Sixth year
|
29 (4.70%)
|
Prefer not to say
|
35 (5.60%)
|
Department of study
|
|
Agriculture UNZA
|
19 (3.10%)
|
Engineering UNZA
|
50 (8.10%)
|
Education UNZA
|
103 (16.60%)
|
Humanities and social sciences UNZA
|
90 (14.50%)
|
Law UNZA
|
55 (8.90%)
|
Mines UNZA
|
5 (0.80%)
|
Medicine UNZA
|
129 (20.80%)
|
Natural sciences UNZA
|
64 (10.30%)
|
Veterinary medicine UNZA
|
9 (1.50%)
|
Medicine LAMU
|
81 (13.10%)
|
Prefer not to say
|
15 (2.40%)
|
Mean Material Affluence Scale score (SD)
|
9 (SD 1.79)
|
Internal consistency was acceptable for the HADS depression subscale (α =0.68) and good for the HADS anxiety subscale (α=0.81) and the Protective Behaviours Scale (α=0.89).
The two-item self-efficacy scale had only moderate internal consistency (α=0.51). The study found that 335 (54%) scored above the threshold for possible depressive disorder and 460 students (74.2%) scored above the threshold for possible anxiety disorder. Females had significantly higher scores for depression (t=2.01, df=612, p=0.045) and anxiety (t=-3.27, df=612, p<0.001). Compared to males, a higher proportion of females scored above the cutoff for anxiety (X2=11.27, df=1, p<0.001) and above the cutoff for depression X2=5.48, df=1, p=0.019). There was no difference in self-efficacy between males and females (p=0.05). (Table 2)
Table 2. Mental health outcomes and self-efficacy by gender
|
Male (n=306)
|
Female (n=308)
|
Total (n=620)
|
P value
|
Mean HADS-D (SD)
|
7.6 (SD 3.64)
|
8.26 (3.71)
|
7.99 (3.67)
|
0.045
|
Case depression (%)
|
150 (49.00%)
|
180 (58.40%)
|
335 (54%)
|
0.019
|
Mean HADS-A (SD)
|
10.18 (SD 4.60)
|
11.36 (SD 4.31)
|
10.81 (4.52)
|
<0.001
|
Case anxiety (%)
|
208 (68.00%)
|
246 (79.90%)
|
460 (74.2%)
|
<0.001
|
Mean self-efficacy (SD)
|
11.00 (2.45)
|
11.02 (2.32)
|
11.00 (2.34)
|
Ns
|
Factors influencing health-protective behaviours
Students reported a mean protective behaviour score (PBS) of 74.09 (out of 105). Just over a quarter (n=168, 27.00%) said they would accept vaccination against COVID-19, n=333 (54.20%) were unsure and n=113 (18.40%) would refuse a vaccination if they were offered one. There was no significant difference in total PBS scores between males and females (p=>0.05) but over a third of males (n=105, 34.30%) reported they would accept a vaccination, compared to only a fifth (n= 63, 20.50%) of female students (X2 =14.83, d=1, P<0.001; OR 2.03, CIs 1.41- 2.92). Thus, males were twice as likely to agree that they would accept a COVID-19 vaccination if they were offered one.
Higher reported total PBS scores were associated with higher self-efficacy (r=0.24, n=620, p<0.001) and lower levels of anxiety (r=-0.103, n=620, p=0.01). Age, family affluence and depression were not significantly related to PBS (all p=>0.05). Both depression (r=-0.106, p<0.001) and anxiety (r=-0.1, p=0.013) were negatively associated with perceptions of self-efficacy for protecting against COVID-19. Participants scoring above the cut-off for possible anxiety had lower PBS scores (t=2.4, df=618, p=0.017) and lower self-efficacy scores (t=2.25, df=218, p=0.025). Participants scoring above the cut-off for possible depressive disorder had lower self-efficacy scores (t=2.29 df=218, p=0.02) but not lower PBS scores (Table 3).
Table 3. PBS scores and self-efficacy by anxiety caseness.
|
Possible case (n=460)
|
Non-case
(n=160)
|
P value
|
Mean PBS (SD)
|
72.95 (19.98)
|
77.33 (19.65)
|
0.017
|
Mean self-efficacy (SD)
|
10.87 (2.32)
|
11.36 (2.37)
|
0.025
|
A three-way ANOVA found main effects for anxiety group (F=5.00, df=1,618, p=.024, eta squared=0.008) and self-efficacy group (F=13.1, df =1,618, p<0.001, eta squared =0.021). Students scoring above the cut-off for possible anxiety and students with low-self efficacy had lower self-reported total COVID-19 protective behaviour scores (Table 3). There were no significant interactions between the variables. There was no relationship between willingness to be vaccinated and either anxiety or depression. Mean scores for individual COVID-19 protective behaviours are shown in Table 4. The anxiety group had lower scores for mask-wearing in class, mask-wearing outside when socially distancing was not possible, and mask-wearing when visiting family and friends. They also had lower scores for keeping social distance in class and keeping social distance outside but higher scores for isolating if they suspected they had COVID-19 symptoms.
Table 4. Health protective behaviours (n=620) and differences in mental health status.
|
Mean behaviour (SD)
|
Never carried out behaviour (%)
|
Case anxiety vs Non-anxiety Case
|
Frequently washed my hands with soap and water
|
5.43 (1.76)
|
31 (5.00%)
|
ns
|
Avoided touching my eyes, nose and mouth with unwashed hands
|
4.45 (1.99)
|
73 (11.80%)
|
ns
|
Used disinfectants to clean hands when soap and water were not available
|
2.104 (5.19)
|
76 (12.30%)
|
ns
|
Avoided mixing with large groups of friends
|
4.94 (2.10)
|
72 (11.60%)
|
ns
|
Wore a mask during classes
|
5.27 (2.39)
|
126 (20.30%)
|
t=2.11, p=0.03
|
Wore a mask on the bus
|
5.77 (2.0)
|
67 (10.80%)
|
ns
|
Wore a mask in shops
|
5.96 (1.83)
|
49 (7.90%)
|
ns
|
Wore a mask outside when I couldn’t socially distance
|
5.30 (2.12)
|
80 (12.90%)
|
t= 2.42, p= 0.016
|
Wore a mask to visit family or friends
|
4.39 (2.19)
|
114 (18.40%)
|
t=3.23, p=0.001
|
Kept my distance with a group of friends outside
|
4.33 (2.14)
|
100 (16.10%)
|
t=2.91, p=0.004
|
Kept my distance with a group of friends inside
|
3.90 (2.17)
|
135 (21.80%)
|
ns
|
Isolated myself when I thought I might have COVID-19 symptoms
|
3.99 (2.58)
|
226 (36.50%)
|
t=-2.57, p=0.01
(Anx>Non-Anx)
|
Wore a mask outside in a large group
|
5.53 (2.10)
|
79 (12.70%)
|
ns
|
Kept my distance in shops
|
5.39 (2.00)
|
67 (10.800%)
|
ns
|
Kept my distance in class
|
4.25 (2.43)
|
167 (26.90%)
|
t=2.04, p=0.041
|
Isolated myself when I had a positive COVID-19 test *
|
4.73 (2.56)
|
51 (26.40%)
|
ns
|
ns = not significant
*n = 193 (Not included in PBS total)