Multivariate Analysis of COVID-19 for Countries with Limited and Scarce Data: Examples from Nepal

Area of Science:

• Epidemiology
• Biostatistics
• Public Health

Background:

• Understanding COVID-19 transmission dynamics is crucial for effective public health interventions, especially in developing countries with limited data.
• Multivariate statistical approaches can uncover complex relationships influencing pandemic spread beyond easily measurable factors.

Purpose of the Study:

• To analyze the dynamics of COVID-19 spread in Nepal using advanced statistical techniques.
• To identify and model hidden variables affecting pandemic progression.
• To evaluate the impact of government-imposed lockdowns on disease transmission.

Main Methods:

• Multivariate statistical techniques including multinomial logistic regression, odds ratio, linear mixed-effect models, and principal component analysis.
• Nonlinear regression for modeling daily new cases and ARIMA models for short-term forecasting.
• Analysis of daily COVID-19 data (cases, deaths, recoveries) and government restriction periods.

Main Results:

• Identified significant hidden variables and their interrelationships impacting COVID-19 spread in Nepal.
• Developed and discussed optimal nonlinear models for daily incidence and forecasted trends using ARIMA.
• Quantified the effect of different lockdown strategies (strict, relaxed) on curbing the pandemic.

Conclusions:

• Multivariate analysis provides valuable insights into pandemic dynamics, particularly for data-limited settings.
• Government interventions like lockdowns significantly influenced COVID-19 spread.
• The findings offer a framework applicable to other South Asian and African countries facing similar challenges.