Search

From paper to digital data collection: some insights

Updated: Nov 4

A 2022 paper* in the Tropical Medicine and Infectious Disease journal raises a number of key points about the transition from paper and pencil interviewing (PAPI) to computer-aided personal interviewing using digital tools including phones and tablets (CAPI). As handheld digital devices and data collection tools become more accessible, even in areas with lower internet access, it’s worth bearing in mind the challenges and advantages of digital technology for disease control, specifically TB.



EPCON's AI model

Why PAPI is popular It seems a no-brainer to make a transition to digital tools for data collection - yet PAPI retains its popularity in many low-resourced areas, for various reasons:

  • PAPI is familiar and easy to use

  • It enjoys cultural acceptance

  • There are suspicions about the security of digital data

  • Some areas have poor phone and internet network coverage (though CAPI tools have been developed to function offline)

  • Local power outages make data entry and syncing to a server trickier - this is specifically true for finding TB active cases due to the unique workflow of TB chest camps, which take place at multiple locations, and require a specific cascade of care (e.g. verbal screening, sputum collection, x-ray, etc.)

Despite these factors, Mercy Corps piloted a hybrid digital and paper-based data collection process between September 2020 and March 2021 as part of their TB active case finding strategy (as described in the paper above). Data was collected by paper at 140 TB chest camps in low internet access areas in two provinces in Pakistan, a country with the 5th highest global burden of TB. After the camps, this data was entered digitally. To assess the hybrid approach, the quality of the digital records were measured against the “gold standard” paper records. In addition, user acceptance was evaluated via focus group discussions.


A desire for digital data collection

Although the study reported high data completion during the first two months and positive overall user feedback of the system, there was a general desire for the input system to be solely digital. The double entry system - paper plus digital - is a safe temporary solution, but not ultimately acceptable for many users, as it is overly time consuming.

The introduction of digital technology data collecting ideally needs to be accompanied by several other vital resources, including:

  • Record linking technology

  • Clear protocols

  • Supportive supervision

  • Data mentoring

There may be a temporary dip in data quality when transitioning to a digital-only data entry method, but supportive supervision can help users adapt. In addition, because a digital data collection system in this setting requires the capacity for working offline, individual barcodes, QR codes or patient IDs need to be used and then matched against diagnostic or treatment data such as sputum collection or x-rays. This may require potential updates to the workflow at chest camps.


Benefits of digital collection systems

A digital collection system offers a critical benefit: the precise location of a TB case can be captured and available in real-time. This valuable data point is then uploaded to the server, allowing program implementers to understand where the demand for TB services is located, and how TB burden is distributed geographically (“TB hotspots”). EPCON has developed such a system via integrating WhatsApp into its state-of-the-art AI epidemiology models for tracking disease. The provision of more efficient and cost effective TB services is a powerful step in disrupting TB transmission. Data quality, including accuracy and completeness, should also be improved with a digital collection system.


In low-internet access settings, a digital tool which can work offline is an important investment. Digital tools should result in a single central geospatial repository, thereby drastically improving estimates of TB prevalence at the community level.



* Mergenthaler, C, Mathewson, J.D, Latif, A, Tahir, H, Meurrens, V, van Werle, A, Rashid, A, Tariq, M, Ahmed, T, Naureen, F and Rood, E. "Factors Affecting the Transition from Paper to Digital Data Collection for Mobile Tuberculosis Active Case Finding in Low Internet Access Settings in Pakistan" Tropical Medicine and Infectious Disease22,7(8): 201.