Risk and evidence-based decision making to help improve quality
1. Is there ‘a body of peer reviewed and published evidence’ i.e. several papers, which broadly support the diagnostic methods in use? (validation)
2. Has the lab been able to reproduce the available evidence in its own 'mini study'? (verification)
3. Do the methods provide reliable and reproducible data?
4. In the absence of any verifiable/traceable calibration materials for semen parameters (sperm number, motility, morphology), how do we define accuracy/reliability?
5. How reproducible is the diagnosis if the same patient has multiple tests?
6. Does the evidence below from our national quality assessment scheme (and the spread of results for the same sperm count by more than 200 laboratories) say anything about industry standards as a whole ?
Sperm count from 253 laboratories analysis the same ‘pooled’ sample
Taking these factors into account, what is the associated measurement uncertainty and what is the risk of providing a misleading diagnosis? Bear in mind the overall diagnostic uncertainty is not just about the 'bench measurement' - its the whole process
‘measurement uncertainty’ is an important consideration, especially for those laboratories going through the accreditation process. In particular, look at:
§ ‘bench measurement’ variables (reproducibility) - application of methods, imported uncertainty e.g. staffing, levels of experience/training, QC, equipment, temperature control
§ Variability and potential for error associated with specimen collection/delivery, especially duration of sexual abstinence
§ Potential for error when transcribing results
§ Biological variability including level of sperm heterogeneity, agglutination/aggregation, viscosity