During the first wave of the coronavirus disease (COVID-19) epidemic in Thailand, 3 Jan to 22 Jun 2020, there were 3,151 confirmed cases and 58 related deaths. This study aimed to describe epidemiological characteristics of the deaths and explore risk factors using a retrospective cohort study design. A COVID-19 related death was defined as a confirmed COVID-19 case who died from a clinically compatible illness. We collected data from investigation reports and medical records using a semi-structure questionnaire and retrieved secondary data from the Department of Disease Control’s database. Of the 58 deaths, the median age was 58 years (interquartile range (IQR) 50-70), 44 were male, and underlying disease was found in 44, hypertension being the most common. The median time from onset to diagnosis date was 7 days (IQR 5-9) compared to 4 days (IQR 2-7) in recovered cases. Six were nosocomial infections and of the remaining 52, 36 had visited a medical facility at least once before they were hospitalized. Male, elderly, and delayed diagnosis were found to be positively associated with death. Early detection of COVID-19 cases should be strengthened in health care facilities throughout Thailand.

On 30 Jan 2014, a company employee in Samut Prakan Province developed severe pneumonia. Since many other employees also developed influenza-like illness (ILI), an investigation was conducted to describe the situation, identify risk factors and provide recommendations. Medical records and company employees’ medical notes were reviewed. Case definition for ILI was based on the guideline of World Health Organization. Probable ILI cases were randomly sampled to confirm influenza A(H1N1)pdm09 using real-time polymerase chain reaction. A retrospective cohort study was performed using a self-administered questionnaire and a walk-through survey of the company was conducted. Total 102 respondents (18.8%) reported having ILI and among them, two were diagnosed with pneumonia. Seven of 21 throat swab specimens were positive for an influenza virus strain that appeared to be influenza A(H1N1)pdm09. The highest reproductive number (R0) of this outbreak was 2.7 (95% CI=1.9-3.8). Risk factors for illness included attending the company party (adjusted OR = 9.1, 95% CI = 2.73-56.35, PAF = 0.86) and having contact with persons who developed ILI (adjusted OR = 2.7, 95% CI = 1.46-4.93, PAF = 0.24). This outbreak showed that the pandemic strain of influenza in 2009 became the circulating strain during 2014.