Effectiveness of Collecting Secondary Data Using Mobile Phone Image Capture: A Case Study of Immunization-History Data Among Children in Remote Areas of Thailand
|
If you are the presenter of this abstract (or if you cite this abstract in a talk or on a poster), please show the QR code in your slide or poster (QR code contains this URL). |
Abstract
Background: The entry of data onto paper-based forms, then digitizing them, is a traditional data-management method; it may involve inaccuracies, especially when it is necessary to collect secondary data because primary data are incomplete, illegible, or missing. Transcription errors from source documents to case record forms (CRFs) are common, and pass from the CRFs and source documents to the electronic database. This case study was based on immunization-history data collected in a Maternal-Child Health logbook. The logbooks (kept by parents) were updated whenever parents brought their children in for immunization. Health providers are supposed to key logbook data into the healthcare information system (HCIS); however, they usually delay it due to case-management workloads on prescheduled vaccination days. HCIS data may thus become incomplete/missing, and their integrity for analysis invalid/unreliable. This study aimed to demonstrate the use of mobile technologies to reduce incomplete data and transcription errors.
Method: The photographic functionality of mobile phones was used to capture page images directly from logbooks. Images of all immunization-history pages in each child’s logbook were captured and transcribed directly into the database, using a data-entry screen corresponding to logbook data fields. In the present study, data entry via phone image capture (DEPIC) was used to capture data for several groups of ethnic hilltribe children living in remote areas. DEPIC data were compared with HCIS data-points for quality, consistency, and completeness.
Results: A proof-of-concept was developed for the DEPIC-captured immunization history of 363 hilltribe children. Comparison of the two databases found differences in numbers of records and detected inconsistencies. The mean (min-max) for missing immunization records was 7.4 (2-19). The HCIS was missing about 42% for < 5 records, 34% for 6-10 records, and 23.7% for > 11 records. The mean (min-max) for inconsistent immunization dates was 1.4 (0-14). In the HCIS, 55.6% of dates were consistent, while 39.9% were inconsistent for < 5 records and 4.4% for > 6 records.
Conclusions: DEPIC has proven very useful for collecting secondary data. Compared with HCIS data, DEPIC can significantly improve the capture of missing data and the correction of inconsistent data. Most importantly, DEPIC halves data-capture and entry time; if data were collected using a paper-based method, transcription errors could occur at two stages--booklet extraction to paper-based CRF, and from CRF to database. In this study setting, the image-capture of immunization history data from logbooks is better than interviewing mothers who speak different ethnic languages. To comply with best data-entry practices in clinical studies, the next version of DEPIC will incorporate a double-entry function from image capture. DEPIC can serve as an effective automated data-collection tool for the timely transmission of data from remote areas to a central data-management center.
Method: The photographic functionality of mobile phones was used to capture page images directly from logbooks. Images of all immunization-history pages in each child’s logbook were captured and transcribed directly into the database, using a data-entry screen corresponding to logbook data fields. In the present study, data entry via phone image capture (DEPIC) was used to capture data for several groups of ethnic hilltribe children living in remote areas. DEPIC data were compared with HCIS data-points for quality, consistency, and completeness.
Results: A proof-of-concept was developed for the DEPIC-captured immunization history of 363 hilltribe children. Comparison of the two databases found differences in numbers of records and detected inconsistencies. The mean (min-max) for missing immunization records was 7.4 (2-19). The HCIS was missing about 42% for < 5 records, 34% for 6-10 records, and 23.7% for > 11 records. The mean (min-max) for inconsistent immunization dates was 1.4 (0-14). In the HCIS, 55.6% of dates were consistent, while 39.9% were inconsistent for < 5 records and 4.4% for > 6 records.
Conclusions: DEPIC has proven very useful for collecting secondary data. Compared with HCIS data, DEPIC can significantly improve the capture of missing data and the correction of inconsistent data. Most importantly, DEPIC halves data-capture and entry time; if data were collected using a paper-based method, transcription errors could occur at two stages--booklet extraction to paper-based CRF, and from CRF to database. In this study setting, the image-capture of immunization history data from logbooks is better than interviewing mothers who speak different ethnic languages. To comply with best data-entry practices in clinical studies, the next version of DEPIC will incorporate a double-entry function from image capture. DEPIC can serve as an effective automated data-collection tool for the timely transmission of data from remote areas to a central data-management center.
Medicine 2.0® is happy to support and promote other conferences and workshops in this area. Contact us to produce, disseminate and promote your conference or workshop under this label and in this event series. In addition, we are always looking for hosts of future World Congresses. Medicine 2.0® is a registered trademark of JMIR Publications Inc., the leading academic ehealth publisher.
This work is licensed under a Creative Commons Attribution 3.0 License.