526. Treatment Inconsistencies of Patients Hospitalized with COVID-19 during Four Waves in Almaty, Kazakhstan, 2020-2021

2023

Толық мәтін Ашық қол жеткізу: Gold Метадеректер (CSL-JSON)
DOI 10.1093/ofid/ofad500.595
Деректер сенімділігі 70 орташа
Бұл баға қалай есептеледі

Ол жазбаның өзінің тексерілетін белгілерінен құралады: мақаланың DOI-і бар ма, оның деректерін қанша тәуелсіз дереккөз растады, міндетті өрістер толтырылған ба және соңғы растау қаншалықты жаңа. Дереккөздер арасындағы алшақтық пен идентификатордың болмауы бағаны төмендетеді. 80-нен бастап мән жазбаның қосымша тексерусіз жарияланғанын, 60-тан 79-ға дейін — оның сенімділігі төмен күйде жарияланғанын және қолмен тексеру ұсынылатынын білдіреді. Редактор қолмен тексерген жазбаны қайта есептеу 90-нан төмен түсірмейді. Бұл — мақаланың өзіне емес, ол туралы деректерімізге берілген баға.

Аңдатпа

Abstract Background Clinical management guidelines were continually changing during the COVID-19 pandemic to reflect best available evidence for a novel virus. In Kazakhstan, the treatment guidelines have been modified 16 times since the pandemic began. We assessed compliance with guidelines during four waves of COVID-19 in Kazakhstan. Methods We conducted a cross-sectional study among people hospitalized with COVID-19 in an infectious disease hospital in Almaty during 4 waves: T1 (June 1-August 30, 2020); T2 (October 1-31 December 2020); T3 (April 1-May 31, 2021); and T4 (July 1-October 26, 2021). Changes in COVID-19 diagnostic and treatment guidelines published during these periods were studied. We abstracted data from patient electronic medical records. Results Seven of 16 COVID-19 diagnostic and treatment guidelines adopted in Kazakhstan were updated during this time. In T1, all people with COVID-19 including those with asymptomatic and mild disease. Antibiotics were recommended only for bacterial infections in T1-T4. Glucocorticosteroids were recommended only to patients with severe disease in T1-T4. Anticoagulants were recommended prophylactically in T1-T3 but only empirically in T4. Of 1,146 COVID-19 patients hospitalized, 14% were in T1, 14% in T2, 22% in T3, and 50% in T4. Mean age was 57 years (range 18-96 years) and 59% were female. In T1-T4 respectively: Antibiotics were given to 78%, 64%, 71%, and 67% of patients with no documentation of bacterial infections; glucocorticosteroids were given to 36%, 31%, 33%, and 28% of patients with non-severe disease; non-COVID-19-specific antivirals were given to 28%, 10%, 4%, and 8% of patients; and anticoagulants given to 18%, 24%, 6%, and 7% of patients with mild disease. Treatment of hospitalized COVID-19 patients inconsistent with national treatment guidelines, Almaty, Kazakhstan, 2020-2021 T1: June 1-August 30, 2020; T2: October 1-31 December 2020; T3: April 1-May 31, 2021; T4: July 1-October 26, 2021 Conclusion We identified inconsistencies with patient treatment and national guidelines. Inappropriate use of antibiotics and non-specific antivirals can result in adverse treatment outcomes and rise in resistance. Improved communication with additional training is needed when guidelines are updated. Disclosures All Authors: No reported disclosures