Predicting COVID-19 incidence in Pakistan: It's time to act now!

Dear Editor,

We read with great interest the article entitled "A quick prediction tool for unfavorable outcome in COVID-19 inpatients: Development and internal validation" by Salto-Alejandre et al. Authors of this article forecast the outcomes of COVID-19. The current paper highlights the prediction of pandemics and potential catastrophes of COVID-19 in Pakistan.

COVID-19 is a severe and ongoing public health emergency in Pakistan that has infected over 01 million people with COVID-19, leading to 23,000 deaths so far. , The number of confirmed cases varies according to each country's epidemiological surveillance and detection capacities. As a result, an estimate of the total number of confirmed cases and anticipated future cases is integral to maintaining demand and resource allocation to the health system. Short- and long-term case estimates require mathematical and statistical modeling tools to determine the magnitude and type of measures necessary to confined an outbreak.

Auto-Regressive Integrated Moving Average (ARIMA) models effectively simulate the time-dependent structure of a time series by accounting for evolving trends, regular changes, and random distortions. The ARIMA approach is mainly devoid of mathematics and statistics. Predictive models for end-users have been established in this manner and may be used further in the decision-making process.

This study evaluates the fourth mode of COVID-19 epidemic propagation in Pakistan, which is the deadliest. The data considered in this study spans the period 1st July 2021 to 31st July 2021 (31 days) and is used to forecast the next 30 days (August month 2021). The data set was utilized to implement and analyze cases and an estimating model for fatalities using several ARIMA models. Thus, in addition to providing insight into the epidemic's transmission patterns, its objective is to use models based on basic quantitative models to give authorities realistic estimations of the epidemic's peak period and severity. These models can aid in forecasting future medical infrastructure and material requirements for patients in these countries.

Fig. 1, Fig. 2 depict predicting graphs of COVID-19 cases and deaths for the next 30 days, respectively. The findings indicate that the expected daily infection rate for the next 30-days (end of August 2021) might reach 8320 (CI 95%: 3289–13,350), while daily mortality could get 47 (CI 95%: −61–156). The anticipated number indicates an increase in infected cases and deaths during the next 30 days. Prediction errors were used to validate the model. We considered the mean absolute error (MAE) and R2 parameters to determine the ARIMA model's significance for COVID-19 pandemic data in Pakistan. ARIMA (0,1,0) daily registered cases models with R-squared (0.86), RMSE (445.40), and MAE (351.66), as well as ARIMA (0,1,0) daily deaths models with R-squared (0.81), RMSE (435.30), and MAE (332.45), were verified and adequately predicted.

Fig. 1

Fig. 2

Pakistan's deteriorating healthcare system may become overwhelmed by the growing number of fourth-wave patients, and the country may become an epicenter in Asia. Medical, public health and policymakers' responses will need to be led by statistics that determine the locations, dates, and populations affected by new instances. This data can be used to drive the country's health advocacy and the kind and extent of government actions to require public health behaviors or managing commercial activities to contain the spread. Forecasting infected cases are crucial for organizing healthcare resources and ensuring that communities confronting the unpredictability of a quickly growing infectious disease throughout a pandemic response have access to care and the best possible results. Careful management and distribution of COVID-19 patients are critical; otherwise, the consequences will be more severe and catastrophic.

CRediT authorship contribution statement

Muhammad Imran Khan: Visualization, Formal analysis, Data curation, Writing – original draft, Formal analysis, Writing – review & editing. Humera Qureshi: Writing – original draft, Formal analysis, Visualization, Data curation, Writing – review & editing. Aamer Ali Khattak: Supervision, Writing – review & editing. Usman Ayub Awan: Writing – original draft, Formal analysis, Visualization, Data curation, Supervision, Writing – review & editing.

Declaration of Competing Interest

The authors declare that there is no conflict of interest or financial disclosure about this publication.