Leveraging routine data in impact evaluation: Understanding data systems in primary healthcare prior to a matched cluster-randomised evaluation of adherence guidelines in South Africa

Abstract

Background. An essential part of providing high-quality patient care and a means of efficiently conducting research studies relies upon high-quality routinely collected medical information.Objectives. To describe the registers, paper records and databases used in a sample of primary healthcare clinics in South Africa (SA) with the view to conduct an impact evaluation using routine data.Methods. Between October 2015 and December 2015, we collected information on the presence, quality and completeness of registers, clinical stationery and databases at 24 public health facilities in SA. We describe each register and type of clinical stationery we encountered, their primary uses, and the quality of completion. We also mapped the ideal flow of data through a site to better understand how its data collection works.Results. We identified 13 registers (9 standard, 4 non-standard), 5 types of stationery and 4 databases as sources of medical information within a site. Not all clinics used all the standardised registers, and in those that did, registers were kept in various degrees of completeness: a common problem was inconsistent recording of folder numbers. The quality of patient stationery was generally high, with only the chronic patient record being considered of varied quality. The TIER.Net database had high-quality information on key variables, but national identification (ID) number was incompletely captured (42% complete). Very few evaluation sites used electronic data collection systems for conditions other than HIV/AIDS.Conclusion. Registers, databases and clinical stationery were not implemented or completed consistently across the 24 evaluation sites. For those considering using routinely collected data for research and evaluation purposes, we would recommend a thorough review of clinic data collection systems for both quality and completeness before considering them to be a reliable data source.

Record linkage without patient identifiers: proof of concept using data from South Africa’s national HIV program

“…We were unable to accurately link 28% of CD4 count and viral load results in TIER.Net to NHLS. We cannot be sure why they were not linked; however, other studies have noted inconsistencies between information recorded in patient les and that captured in TIER.Net (12,13,34,44). Fifth, our approach requires the availability of data on the same patient characteristics -here, laboratory test results -to facilitate linkage, and would not be suitable for linking databases that do not contain some shared data points.…”

“…While laboratory tests (CD4 counts and HIV viral loads) information are also captured, the process is manual and inconsistent resulting in incomplete f (13). TIER.Net is not nationally networked (9), and lab results preceding HIV diagnosis and ART initiation are largely unavailable on TIER.Net (33,34). The TIER.Net patient ID is allocated by facilities, and patients who seek care at alternative facilities may receive a new TIER.Net patient ID, creating duplicate records and hindering tracking of patients across facilities (13).…”

Trajectories of re‐engagement: factors and mechanisms enabling patient return to HIV care in Zambia

“…Patients’ self‐perception of engagement status often does not match EMR‐based characterization as lost‐to‐follow‐up [ 73 ]. This is due to differences in definitions of “in‐care” and well‐documented EMR inaccuracies regionally [ 31 , 74 , 75 ]. Interventions using EMR to identify patients with care gaps, such as PEPFAR Tracking and Tracing [ 76 ], may need improved linkage data [ 77 ] and/or may benefit from allowing patients to self‐describe care status.…”