Enabling Data-Driven Exploration: A Novel Integration of DHIS2 and OpenCPU for Statistical Analysis within an Interactive Application

Implementation - Implémentation Interoperability
,
1

This abstract has been accepted at the 2024 DHIS2 Annual Conference


click to view session
session link: Extending DHIS2

Enabling Data-Driven Exploration: A Novel Integration of DHIS2 and OpenCPU for Statistical Analysis within an Interactive Application

This abstract introduces a project on data exploration and analysis in the DHIS2 ecosystem. Our intervention integrates DHIS2 data retrieval via API and OpenCPU-powered real-time R code execution within a web application. Utilizing DHIS2 API: Leveraging the DHIS2 API, our intervention ensures efficient data extraction directly from the DHIS2 database, circumventing the laborious process of individual dataset downloads. Prioritizing data security, the app’s architecture maintains confidentiality, providing users a streamlined, error-free experience. Interactive Statistical Analysis with OpenCPU: A key feature of our application is the integration of OpenCPU, offering users a dynamic platform for real-time execution of R code. Emulating RStudio’s layout, the app’s GUI includes dropdown menus for dataset selection and a text box for code entry. Addressing Challenges: This application mitigates challenges associated with traditional methods, such as time-consuming downloads and potential data security risks. By consolidating DHIS2 data retrieval and statistical analysis, our approach simplifies workflows, fostering an environment where users can interactively explore datasets without compromising security. Impact Assessment: Our intervention significantly enhances data exploration efficiency and promotes user-driven statistical analyses. By empowering users to interactively select datasets, compose R code, and witness real-time results, the application facilitates a dynamic and responsive analytical process. This shift towards user autonomy is expected to elevate the overall impact of DHIS2 utilization, enabling more informed decision-making and yielding meaningful outcomes. Conclusion: In summary, our project introduces an application that integrates DHIS2 data retrieval and OpenCPU for interactive statistical analysis. Addressing challenges in data exploration, our approach emphasizes efficiency, security, and user autonomy. Beyond technical enhancements, this intervention promises an impact on how DHIS2 is leveraged for dynamic data analysis.

Primary Author: Shikhar Shukla


Keywords:
DHIS2, OpenCPU, Integration, Custom App, R code, Statistical Analysis, User Interface, GUI, API, Web application, data exploration

6 Likes
2

I am looking forward to presenting this information at the annual conference on June 10th at 1pm (Oslo time). Please share any questions you have by commenting below, and I will make sure to address them during my presentation.

1 Like
3

Nice presentation. I use a similar approach

4

Me too, and thanks Alan for the presentation. I have a shiny app that helps user pull data, clean it, and do time-series modeling. I often pull several years of data for several data elements and it can take several hours, especially for large countries (DRC, Nigeria, etc.). Perhaps my process is not as efficient as yours. Basically I make API requests for each data element category combination, for each month, at each orgUnit level. I found that for large countries the request may time out if I did not limit request somehow, and that is how I ended up using each orgUnit level. How do you guys do it?