Integrating a social media chatbot with DHIS2 for event-based surveillance by private sector during the COVID-19 pandemic in Vietnam

Part of the Event-based surveillance & rumor monitoring DAC2021 Session: Tuesday 22nd June 15:00

If you are attending the DHIS2 Annual Conference, more details on this session are available here.

If you have not yet registered to the DHIS2 Annual Conference, do not hesitate to visit the DHIS2 Annual Conference website and secure your spot!
Integrating a social media chatbot with DHIS2 for event-based surveillance by private sector during the COVID-19 pandemic in Vietnam

Hoa H. NGUYEN, Phuong M. LA & Bram PIOT
Population Service International Vietnam (PSI Vietnam)

Event-based surveillance (EBS) for early warning and response to disease outbreaks was implemented in Vietnam in the early 2010s, and was gradually expanded to all 63 provinces. EBS received increased interest during the COVID-19 pandemic and is considered a potentially effective tool in detecting unusual events that might signal a new outbreak. Available COVID-19 EBS signals in Vietnam come from various sources: official point-of-entry quarantine units, media monitoring, individual health declaration, close-contact exposure history, etc. While the Ministry of Health encourages healthcare providers to report suspected COVID-19 cases, this largely focuses on public facilities. Private health facilities including pharmacies in particular, despite often being the first points of contact for people seeking care, barely participate in disease surveillance. Moreover, there is limited information on how EBS signals are consolidated, analyzed, and shared among stakeholders such as health authorities at different levels.

In August 2020, building on our experience with a malaria case reporting chatbot, PSI Vietnam launched similar social media chatbot for reporting fever cases by private outlets in the context of COVID-19. Using automated conversation to prompt health care providers to report fever cases and other symptoms, the chatbot is connected to DHIS2 for analysis and visualization. Our aims are to (1) encourage private health outlets to report possible COVID-19 cases, (2) build a network for reporting, (3) collect and analyze data for tracking trends, and (4) support local health authorities with COVID-19 surveillance.

The preliminary results after 9 months of implementation are promising: as of April 2021, 2,926 private providers (340 clinics, 2586 pharmacies) are engaged in 5 provinces (Hanoi, Ha Nam, Quang Binh, Bac Ninh, Thai Nguyen). In total, 77,744 client visits were reported by 1,768 (61%) outlets. Heatmaps show 10 districts with highest number of fever cases. Most common reported symptoms were cough (35%), fever (28%), and sore throat (29%); 65% of outlets participating in an acceptability survey submitted reports within 24 hours. More than half (58%) prefer using this chatbot over the traditional health declaration thanks to its ease of use, speed, and flexibility.

PSI provided technical assistance and capacity building to provincial Centers for Disease Control and Prevention and district health centers to increase the availability of EBS data for verification and assessment, potentially contributing to their response actions. Each institution has access to their data and a DHIS2 dashboard demonstrating outlet distribution, daily number of reported events by location, gender, age and symptom.


This is interesting!..


@KetshabileT thank you! For more information on our use of social media chatbots for disease surveillance, check out this blog post from late last year, and feel free to reach out directly for further details.


Thank you @bram_piot , i will surely do reach out!

1 Like

very interesting, would love to know more about this thing :+1: :+1:

1 Like


1 Like

Video session can be found here:


Hi @bram_piot and @honghoa do we have WhatsApp integrated in the chatbot yet?

Hi @bridget - we’re working on it! WhatsApp is being tested in Laos at the moment, and Telegram in Myanmar. Feel free to reach out directly if you are interested in more details. @jose or @ctejo may also help answer any technical questions you may have.