Data Quality

Complete, consistent, accurate, and timely data enables policymakers to identify service gaps, design targeted interventions, and monitor progress toward Universal Health Coverage and the Sustainable Development Goals. The comparator in eligible studies is low-quality routine health data, such as data that is incomplete, delayed, inconsistent, or inaccurate. Studies that do not report data quality using the WHO definition, or that assess it only subjectively, will be excluded from the review. Eligible studies may assess one, some, or all data quality dimensions including completeness, accuracy, timeliness, and consistency. SSNAP records rarely stand alone: only 3.6% of non-fatal strokes and 0.6% of haemorrhagic strokes appeared in SSNAP without corroboration from at least one other source. PPV for stroke coding varies substantially by setting and subtype, and coded diagnoses could not be verified against written clinical records. The high proportion of unknown stroke subtype in death records and primary care reflects weaker diagnostic coding compared with hospital and audit data. Recurrent stroke could not be reliably distinguished from follow-up contacts for the index stroke, lim…