Variability in Perceived Healthcare Data Quality Across Tanzanian Regional Referral Hospitals: A Hospital-Based...

Healthcare data quality is critical for informed decision-making and effective healthcare management. However, variability in perceived data quality exists across different healthcare facilities. This study aimed to explore the differences in healthcare data quality perceptions among healthcare workers across four dimensions: data accuracy, completeness, timeliness, and consistency.

Between August and October 2024, we conducted a cross‑sectional survey of 336 healthcare workers, selected via simple random sampling from the eight RRHs. Participants completed a validated questionnaire (Cronbach's α  ≥  0.83 across dimensions). We used descriptive statistics to summarize mean scores by hospital, and dimension scores were analyzed using one‑way ANOVA, with partial η² as an effect‑size measure and Tukey HSD post‑hoc tests to identify specific inter‑hospital differences.

Significant hospital‑level differences emerged for accuracy (F(7, 328) = 5.03, p = 0.026, η² = 0.097), completeness (F(7, 328) = 8.65, p < 0.001, η² = 0.156), timeliness (F(7, 328) = 6.07, p = 0.006, η² = 0.115), and consistency (F(7, 328) = 5.41, p < 0.001, η² = 0.103). Post‑hoc analyses showed that H8 and H2 consistently outscored other hospitals: both led on completeness (vs. all peers), and both exceeded H6 (the lowest‑rated) on accuracy, timeliness, and consistency.

Perceptions of data quality vary markedly across RRHs in Tanzania, with medium‑to‑large effect sizes indicating substantive differences. However, the cross‑sectional, perception‑based design precludes causal inference and lacks objective data-quality verification. Future work should integrate objective metrics (e.g., audit logs, error rates) and apply a guiding framework such as Donabedian's structure-process-outcome to identify the facility‑level determinants driving these perceptual gaps and their impact on patient‑care outcomes.