Of a total of 3,555 complete responses, the majority were from the North district of Taiwan (66.3%), were from females (78.2%), and were from respondents aged 31-40 years (36.9%) or 41-50 years (28.4%). A summary of the characteristics of the respondents is provided in the first and second columns of Table 1.
Anxiety symptoms
Using a cut-off score of 44 for the STAI score, a total of 52.1% (95%CI 50.4–53.7) of the respondents reported moderate to severe levels of anxiety symptoms in the PW, whereas 48.8% (95%CI 47.2–50.5) reported moderate to severe levels of anxiety symptoms in the (BO) (Table 1). The highest proportions of moderate to severe cases of anxiety in the PW were observed in the 31-40 years age group (57.1%; 95%CI 54.4–59.8), among women (55.2%; 95%CI 53.4–57.1), among those with an education level of high school and below (56.0%; 95%CI 49.6–62.2), and among respondents from the North district (53.1%; 95%CI 51.1–55.1). Respondents with an average household income of below NT$20,000 (54.4%; 95%CI 50.1–58.6) and those with an average household income between NT$20,000 and NT$50,000 (54.4%; 95%CI 51.8–56.9) showed a similar proportion of moderate to severe cases of anxiety in the PW.
The highest proportion of moderate to severe cases of anxiety during the BO were observed in those aged 31-40 years (53.7%; 95%CI 51.0–56.4), females (51.9%; 95%CI 50.0–53.7), those with a college education level (51.2%; 95%CI 49.1–53.4), and respondents with a household income between NT$20,000 and NT$50,000 (51.6%; 95%CI 49.0–54.2). Among respondents from the North (49.4%; 95%CI 47.3–51.4) and Central (49.7%; 95%CI 45.8–53.7) districts there was a very similar proportion of moderate to severe cases of anxiety during the BO. On the whole, those in the older age group, males, those whose highest educational level was a degree and above, those in the higher income group and those in the South and East districts reported a lower level of anxiety symptoms during both the PW and the BO.
A comparison of the proportion of moderate to severe cases of anxiety during the PW and the BO revealed that significantly the highest increase in the proportion of moderate to severe cases of anxiety was among those in the 41-50 years age group (5.2%; 95%CI 4.0–6.7). In contrast, among the respondents in the 20-30 year age group there was a decrease in the proportion with moderate to severe anxiety (-2.8%; 95%CI -9.2–3.6). The group of female participants showed a significantly higher increase in the proportion of moderate to severe cases of anxiety (3.3%; 95%CI 2.7–4.1) than the group of male participants (2.7%; 95%CI 1.8–4.1). The respondents whose highest educational level was high school or below (6.0%; 95%CI 3.7–9.7) and those in the income group below NT$20,000 (5.7%; 95%CI 4.0–8.0) showed the highest increase in the proportion of those with moderate to severe anxiety. Respondents from the North (3.7%; 95%CI 3.0–4.6) and Central (3.2%; 95%CI 2.0–4.9) districts reported a higher increase in the proportion of those with moderate to severe anxiety than those from other districts.
The difference in the mean STAI score for the PW (46.2, SD=11.68) and the BO (45.0, SD=12.13) was statistically significant (Wilcoxon signed rank test = 6.414, p < 0.001). The median scores for anxiety symptoms during the PW and the BO were 46.7 (IQR 36.7–53.3) and 43.3 (IQR 36.7–53.3), respectively. The scale for both scores ranges from 20 to 80. In total, 1,229 (34.6%; 95%CI 33.0–36.2) of the respondents recorded an increase in their anxiety score during the PW compared with the BO (PW score > BO score), and 2,326 (65.4%; 95%CI 63.9–67.0) recorded a similar or lower anxiety score (PW score ≤ BO score).
Preventive measures
The summary of responses in relation to preventive measures during the PW and BO is shown in Table 2. Among the personal protection measures carried out during the PW, the highest proportion was for respondents who frequently washed their hands with soap (49.0%; 95%CI 47.3–50.7), followed by those who wore a mask (31.8%; 95%CI 30.3–33.4). Under cough etiquette, the most commonly practised preventive measure was disposing of tissues immediately after coughing or sneezing (80.1%; 95%CI 78.9–81.4), and covering one’s mouth and nose when coughing or sneezing (65.2%; 95%CI 63.6 – 66.7). Performing hand hygiene was reported always practised by a considerable proportion of the participants (60.3%; 95%CI 58.6–61.9). Over a third of respondents reported that they always avoided proximity with others (34.4%; 95%CI 32.8–36.0) and avoided gathering in groups (41.2%; 95%CI 39.5–42.8). Near two thirds reported that they would always voluntarily quarantine themselves if they were feeling unwell (61.6%; 95%CI 60.0–63.2), but fewer than half said they would always report promptly to the health authorities if they were feeling unwell (45.4%; 95%CI 43.8–47.1).
Similar findings were reported for preventive measures at the BO, with the highest proportion reporting that they had always disposed of tissues immediately after coughing and sneezing (66.5%; 95%CI 64.9–68.0), followed by those who had covered their mouths and noses when coughing or sneezing (62.2%; 95%CI 60.6–63.8). The proportion who practised contact precautions was relatively low. A comparison of the proportions who always practised particular preventive measures during the PW and the BO shows that disposing of tissues immediately after coughing and sneezing recorded the highest increase (13.6%; 95%CI 12.6–14.8), followed by avoiding group gatherings (10.3%; 95%CI 9.4–11.4) and frequently washing hands (10.8%; 95%CI 9.8–11.8). A lower increment in the proportion can be seen for using eye protection (1.9%; 95%CI 1.5–2.4) and covering one’s mouth and nose when coughing or sneezing (3.0%; 95%CI 2.4–3.6).
The difference in the mean total scores for preventive measures during the PW (25.4, SD=5.85) and the BO (23.4, SD=6.56) was statistically significant (Wilcoxon signed rank test = 23.917, p < 0.001). The median scores for preventive measures during the PW and the BO were 26.0 (IQR 21.0–30.0) and 24.0 (IQR 19.0–28.0), respectively. The total score for preventive measures during the PW ranges from 2 to 36, whereas the total score for preventive measures during the BO ranges from 0 to 36. For the respondents overall, a total of 1,696 (47.7%, 95%CI: 46.1–49.4) reported an increase in their total preventive measures score during the PW compared with their score during the BO and 1,859 (52.3%, 95%CI: 50.6–53.9) reported similar or lower total preventive measures scores.
Table 3 shows the univariate and multivariable analyses of the factors associated with a higher total preventive measures score when comparing the PW with the BO. Having an increased anxiety score during the PW compare with the score during the BO (adjusted OR = 7.38, 95%CI 6.28–8.66) was a strongly significant determinant of an increase in the preventive measure score from the BO to the PW. Respondents with a college level of education (adjusted OR = 1.58, 95% 1.17–2.13) or graduate level and above (adjusted OR = 1.74, 95% 1.28–2.38) were more likely to have an increased preventive measure score from the BO to the PW compared to those whose highest education level was high school and below.