2321. Prevalence and determinants of depression among healthcare workers during the COVID-19 pandemic in Khujand, Tajikistan, 2022

Open Forum Infectious Diseases, 2023

Полный текст Открытый доступ: Gold Метаданные (CSL-JSON)
DOI 10.1093/ofid/ofad500.1943
Достоверность данных 70 средняя
Как считается эта оценка

Складывается из проверяемых признаков самой записи: есть ли у статьи DOI, сколько независимых источников подтвердили её данные, заполнены ли обязательные поля и насколько свежо последнее подтверждение. Снижают оценку расхождения между источниками и отсутствие идентификатора. Значение от 80 означает, что запись опубликована без дополнительной проверки; от 60 до 79 — что она опубликована с пониженной достоверностью и ручная проверка желательна. Запись, выверенную редактором вручную, пересчёт не опускает ниже 90. Это оценка наших данных о статье, а не оценка самой статьи.

Аннотация

Abstract Background Healthcare workers experienced substantial work overload and stress during the COVID-19 response. Increase in prevalence of mental health conditions among healthcare workers have been reported globally. Low-resource settings, such as Tajikistan, experienced additional challenges related to the COVID-19 pandemic and have scare resources to mitigate mental health consequences. To inform program planning, we aimed to determine prevalence of depression, associated factors, and care gap among healthcare workers. Methods We conducted a cross-sectional study of all healthcare providers (physicians, nurses, and nurse assistants) working from July to August 2022 in three COVID-19 referral hospitals and three polyclinics in Khujand city, Tajikistan. We systematically sampled every fifth provider. Depression was assessed using the Hospital Anxiety and Depression Scale. We used multivariable logistic regression model adjusting for random effect of hospital to identify factors associated with depression. Results Among 400 healthcare workers, 277 (69%) had depression (Table 1). Half (44%) no longer enjoyed the things they used to (Figure 1). Prevalence of depression was significantly higher (p< 0.01) among females than males (71% vs 59%), nurses than physicians (77% vs 56%). Among providers with depression, 20% had received professional services, 32% had taken a vacation, 32% had severe COVID-19, and 53% experienced frequent respiratory issues, and 56% felt people avoided them because of work. In multivariable analysis depression was associated with being a nurse (adjusted odds ratio [aOR]=3.5; 95% confidence ratio [CI]: 1.7-7.2), having had severe COVID-19 disease (aOR=4.0; 95% CI: 1.4-11.8) and having felt that people avoided them because of work (aOR=1.6; 95% CI: 1.0-2.6) (Table 2). Providers who used respirators >3 hours had lower odds of depression (aOR=0.4; 95% CI: 0.2-0.9). General characteristics of healthcare providers by depression, Tajikistan, 2022 External pressures, physical conditions and supportive services among and depression among healthcare workers, Tajikistan, 2021 Factors associated with depression among healthcare workers, Tajikistan, 2021 Conclusion We found high prevalence of depression among healthcare workers in Khujand city, Tajikistan. Timely screening and increase in provision of supportive services for providers is needed. These services should be offered to all providers, and especially targeted towards those with prior severe COVID-19 disease and nurses. Disclosures All Authors: No reported disclosures