PERFORMANCE OF GOOGLE TRENDS TO DETECT THE OUTBREAK OF DENGUE IN YOGYAKARTA AND JAKARTA

ABSTRACT Background: Early detection of disease outbreak is among the most critical role of the sub-national authorities as mandated by health decentralization policy. Given the continuous growth of Internet penetration and dependencies of the society to the digital ecosystem, it is essential to investigate the potentials innovation to improve the existing surveillance system using digital epidemiology. Several studies have assessed the roles of Google Trends (GT) to improve dengue surveillance systems, including in Indonesia. However, studies focusing on comparing the performance of Google Trends in dengue-high burdened provinces were limited. Aims: This study aims to examine the correlation between GT data on dengue-related query terms with the official dengue surveillance report in Jakarta and Yogyakarta Province. Methods: We [j1] Relative Search Volume data for dengue were collected from the area of Jakarta and Yogyakarta between 2012 to 2016. Those data are compared to the official dengue report using Pearson correlation and Time-lag correlation, performed with Stata version 13. Results: GT data is highly correlated with the routine surveillance report in Jakarta ( r = 0.723, p-value= 0.000) and Yogyakarta Province ( r = 0.715, p-value= 0.000). High correlation of lag-1 ( r =0.828, p-value= 0.000) was occurred in Jakarta only. This finding indicates that GT data could possibly detect the dengue outbreak a month earlier. Hence, GT data can be used to monitor disease dynamic and improving the public awareness of potential outbreak in near-real-time. Conclusion: GT data are statistically correlated with the routine surveillance report in Jakarta and Yogyakarta Province. Early warning system utilizing GT data is potentially performed in Jakarta Province. Therefore, GT is potential tool to support the surveillance program at sub-national level.  Further studies involving other digital data sources, for example, Twitter, online news, and administrative data from the national health insurance are essential to strengthen the current surveillance system with digital epidemiology approach. Keywords : dengue, digital epidemiology, Google Trends, Indonesia ABSTRAK Latar Belakang: Belum ada penelitian yang mengkaji penggunaan Google Trends (GT) sebagai sumber data potensial untuk surveilans termasuk demam berdarah dengue pada level sub-nasional atau provinsi. Hal ini penting dikarenakan kebijakan terkait penentuan KLB merupakan otoritas pemerintah daerah di era desentralisasi. Tujuan: Penelitian ini bertujuan untuk menilai korelasi data GT dan data rutin surveilans dengue di Daerah Khusus Ibukota Jakarta dan Daerah Istimewa Yogyakarta. Metode: Penelitian ini menggunakan data GT terkait dengue dari 2012 hingga 2016. Data tersebut selanjutnya dikomparasikan dengan data surveilans dengue menggunakan korelasi Pearson dan korelasi Time-lag menggunakan Stata versi 13. Hasil: Korelasi Pearson menunjukkan hubungan yang kuat antara data GT dengan laporan surveilans di Provinsi Jakarta (r= 0.723, p-value= 0.000) dan Yogyakarta (r= 0.715, p-value= 0.000). Korelasi Time-lag mengindikasikan korelasi yang lebih tinggi pada lag-1 untuk Provinsi Jakarta (r=0.828, p-value= 0.000). Temuan ini menunjukkan adanya kemungkinan penggunaan data GT untuk mendeteksi kenaikan kasus dengue satu bulan lebih awal. Oleh karena itu, GT berpotensi digunakan untuk monitoring dinamika penyakit dan kesadaran publik secara tepat waktu dan lebih cepat dari pada laporan rutin surveilans. Kesimpulan: GT berkorelasi secara statistik dengan data rutin surveilans dengue di Jakarta dan Yogyakarta. Sistem deteksi dini menggunakan GT berpotensi digunakan di Jakarta. Penelitian lebih lanjut yang memanfaatkan sumber data digital seperti Twitter, berita daring dan data administratif BPJS Kesehatan merupakan langkah penting untuk memperkuat sistem surveilans kesehatan menggunakan pendekatan epidemiologi digital. Kata kunci : dengue, epidemiologi digital, Google Trends, Indonesia [j1] Do not use the “personal” pronoun in this study. Please, check the other sections