Management of Medicinal Demand Forecasting of the Drug Supply Chain during the COVID-19 Pandemic

Abstract

Background and Objectives: The COVID-19 pandemic has had an unprecedented impact on global supply chains, particularly in the healthcare sector. One of the most critical areas affected is the supply chain of medicinal drugs, especially in general medical clinics that serve as the last-mile service providers. This study focuses on general medical clinics in Thailand to explore how the pandemic has disrupted drug usage demand, and how various factors—demographic, service-related, and socio-economic—have influenced such disruptions. The primary objective is to develop a practical forecasting model to support more agile, data-driven inventory and procurement planning in clinical settings.

Methodology: This research employed a mixed-methods approach combining both qualitative and quantitative data collection and analysis techniques. In the qualitative phase, data were collected through participant observations and in-depth interviews with ten key stakeholders, including clinic administrators, physicians, pharmaceutical distributors, and representatives of drug companies. These participants were selected using purposive sampling based on their roles in the supply chain. The quantitative phase involved a structured survey distributed to 389 respondents in Nakhon Ratchasima province, located in the Northeast of Thailand, an area heavily affected by COVID-19. The survey was validated for reliability and content accuracy. Descriptive statistics, SIPOC process mapping, time-series analysis, and multiple regression techniques using SPSS (version 27) were employed to analyze the data and forecast trends.

Main Results: The findings reveal significant fluctuations in drug demand patterns during the pandemic period. Clinics, which represent the downstream end of the supply chain, experienced both over- and under-supply due to erratic demand. The regression analysis showed statistically significant correlations (p < 0.05) between drug usage and variables such as patient age, education level, underlying health conditions, clinic location, screening protocols, and medication pricing. Furthermore, time-series analysis identified trends and seasonal patterns in drug utilization, enabling the classification of drugs into high and low turnover categories. By applying the developed forecasting model, clinics could reduce their drug inventory costs by approximately 20%, from 218,510 baht to 173,950 baht.

Discussions: This study underscores the vulnerability of healthcare supply chains during crisis events. It emphasizes the importance of incorporating real-time, localized data into forecasting models to increase resilience. Price sensitivity emerged as a particularly influential factor, where higher prices discouraged patients from obtaining medications. The integration of time-series forecasting tools helped to anticipate future demand more accurately and manage stock levels more efficiently. The study also demonstrated that service attributes—such as convenience, physician trust, and promotional offers—play a significant role in shaping patient behavior.

Conclusions: This research contributes a valuable framework for demand forecasting in clinical supply chains, especially during health crises. By identifying key influencing factors and applying advanced analytical methods, this model offers a replicable strategy for other healthcare institutions seeking to enhance supply chain efficiency and responsiveness. In the long term, the findings can inform policy development and promote sustainable, resilient drug supply management practices in clinical contexts facing external shocks.