Statistical quality control practices in a sample of Spanish Blood Establishments: a mixed-methods exploratory study.

BACKGROUND

The European Directorate for the Quality of Medicines & HealthCare (EDQM) provides guidance for statistical quality control in blood component manufacturing, including recommendations for sampling and statistical process control. However, the extent to which these recommendations are implemented in routine practice remains unclear.

MATERIALS AND

METHODS

We conducted a sequential explanatory mixed-methods study among professionals working in blood establishments, recruited through the Spanish Society of Blood Transfusion blood processing working group. Phase 1 was a cross-sectional survey (10 responding centers across 10 autonomous communities; response rate 77%). Phase 2 comprised semi-structured interviews with a purposive sample of survey respondents (No.=7). Survey findings informed the interview topic guide. Thematic analysis followed the Braun and Clarke approach. Findings from both phases were integrated using a joint display.

RESULTS

Respondents represented centers with varied production volumes, and most had more than 6 years of experience (80%). Use of statistical quality control methods was limited: 90% reported using descriptive statistics only and 10% used control charts. For sample size determination, 80% applied fixed production percentages without formal statistical justification. Spreadsheet-based workflows with custom formulas predominated (90%). Key challenges included determining sample size and laborious report generation (both 80%). Interviews highlighted uncertainty about sampling adequacy, potential for earlier trend detection, perceived complexity of guidance, and the need for interpretive decision support. All participants expressed interest in automated approaches.

DISCUSSION: Among participating centers, variation in the implementation of statistical quality control practices relative to EDQM recommendations suggests opportunities for improved harmonization. Automated tools that embed recommended sampling and monitoring methods may support implementation, although larger studies are needed to estimate prevalence at national level.