Kidney Week

Abstract: FR-PO0021

Effects of Macroeconomic Indicators on Kidney Transplant Awareness in the United States: A SARIMAX Analysis of Google Search Trends, 2004-2025

Session Information

• Artificial Intelligence: Clinical Prediction, Decision Support, and Digital Kidney Care
  October 23, 2026 | Location: Exhibit Hall A, Convention Center
  Abstract Time: 10:00 AM - 12:00 PM

Category: Artificial Intelligence, Digital Health, and Data Science

• 300 Artificial Intelligence, Digital Health, and Data Science

Authors

• Kookanok, Chutawat, Interfaith Medical Center, New York, New York, United States

Chutawat Kookanok, MD

• Poochanasri, Methavee, Phramongkutklao Hospital, Bangkok, Bangkok, Thailand

Methavee Poochanasri, MD

• Kulthamrongsri, Narathorn, University of Hawai'i System, Honolulu, Hawaii, United States

Narathorn Kulthamrongsri, MD

• Yongkiatkan, Panchanit, University of California Irvine, Irvine, California, United States

Panchanit Yongkiatkan, MD

• Sackey, Jeffrey Afrah, Interfaith Medical Center, New York, New York, United States

Jeffrey Afrah Sackey

• Phichedwanichskul, Kitinan, Phramongkutklao Hospital, Bangkok, Bangkok, Thailand

Kitinan Phichedwanichskul, MD

• Tangprasittichok, Nathasith, Phramongkutklao Hospital, Bangkok, Bangkok, Thailand

Nathasith Tangprasittichok, MD

• Aiumtrakul, Noppawit, Mayo Clinic Minnesota, Rochester, Minnesota, United States

Noppawit Aiumtrakul, MD

Background

Kidney transplantation(KT) remains the optimal treatment for kidney failure; however, transplant volumes are influenced by social and economic conditions. This study evaluates the performance of a Seasonal Autoregressive Integrated Moving Average with Exogenous Regressors(SARIMAX) model in forecasting KT searching volumes and examining the impact of economic indicators on transplant trends.

Methods

The monthly Google Search Index(0–100) for the term “Kidney Transplant” and FRED(Federal Reserve Economic Data) economic indicators(Table 1) were analyzed from January 2004 to August 2025. Time-series decomposition, auto.arima function in R program and data training procedure were used to guide appropriate model selection.

Results

The lowest monthly search volume(score = 32) occurred in December 2011, while the highest search volume was observed in July 2025(score = 100)(Fig. 1). Time-series decomposition confirmed a upward trend and seasonal patterns(Fig. 1). The final model, selected via auto.arima demonstrated performance with MAE = 3.778, RMSE = 5.711, MAPE = 7.34%, and no autocorrelation(ACF1 < 0.01; Ljung–Box test p-value = 0.138). Unemployment and Industrial Production were identified as negative predictors, whereas Disposable Personal Income showed a positive association(Table 1). The KT prediction plot is shown in Fig. 2.

Conclusion

SARIMAX model effectively predicted KT search volume. Increased unemployment and decreased industrial production were associated with reduced search activity, while increased real disposable income was linked to higher engagement. Integrating these indicators strengthens forecasting accuracy, helping health systems anticipate resource needs and potential economic barriers to patient access.

Table 1. Coefficients of Macroeconomic Predictors in the SARIMAX Model.