Potential Problems With Audits

While audits and the audit process can be a very effective tool for collecting information and evaluating safety performance, they can also pose some potential problems for an organization. Some examples of potential problems with audits include... 

Although a thorough validation cannot rule out all potential problems, the process of method development and validation would address the most common ones. Examples of typical problems that can be minimized or avoided include interferences that coelute with the analyte in liquid chromatography (LC), a particular type of column that no longer produces the separation needed because the supplier of the column has changed the manufacturing process, an assay method that is unable to achieve the same detection limit after a few weeks, or a quality assurance audit of a validation report that finds no documentation on how the validation was performed. 

For studies that have delegated responsibilities to a PI, the study director will rely on that individual to ensure that relevant phase)s) of the study are conducted in accordance with the study plan, relevant SOP, and principles of GLP. The PI should contact the study director when event(s) occur that may affect the objectives defined in the study plan. All communications should be documented. Communication between the study director and the QA is required at different stages of the study. For instance (1) to review study plans (2) to review new and revised SOPs (3) attendance of QA personnel at study initiation meetings and in resolving potential problems related to GLP (4) by responding to inspection and audit reports promptly (5) by indicating corrective action and (6) by necessary liaison with QA staff, and scientific and technical personnel. In certain unforeseen conditions, change of study director becomes essential. Under such conditions, replacement of the study director should take immediate effect. 

Clause 8.5 requires the organization to plan for continual improvement of the quality system through the use of quality pohcy, objectives, audit results, analysis of data, corrective and preventive action, and management review. The last two subclauses in clause 8.5 are entitled Corrective and Preventive Action. These two requirements were a single clause in the 1994 edition of the standard. This led to some confusion among users of the standard as to the difference between them. Corrective action includes the correction of problems that have occurred, whereas preventive action deals with potential problems that may have never occurred. 

While this does seem to be a comprehensive suite of risk control measures, the real reason for the problem emerged during a potential human error audit and safety management system review which was being rmdertaken at the mine - not one of the 150+ vehicles in the undergrotmd fleet at the mine was fitted with a speedometer. 

Measures to reduce human error are often implemented at an existing plant, rather than during the design process. The decision to conduct an evaluation of the factors that can affect error potential at an existing plant may be taken for several reasons. If human errors are giving rise to unacceptable safety, quality or production problems, plant management, with the assistance of the workforce, may wish to carry out a general evaluahon or audit of the plant in order to identify the direct causes of these problems. 

Despite these potential benefits, there are possible problems associated with the use of generic safety audit systems. Questions that need to be considered in the case of such standardized audits include ... 

The principles of GLP require an independent quality assurance (QA) program to ensure that the study is being conducted in compliance with GLP. The QA personnel cannot overlap with those of the study because of the potential conflict of interest, but they may be part-time staff if the size of the study does not warrant a full-time QA section. The responsibilities of the QA unit are to maintain copies of plans, standard operating procedures, and in particular the master schedule of the study, and to verily, in writing, that these conform to GLP. The QA unit is responsible for inspections and audits, which must be documented and the results made available to the study director and the principal investigator. The QA unit also signs off on the final report. Any problems discovered or corrective action that is recommended by the unit must be documented and followed up. 

Often auditing selected partial data sets can indicate if the data system is in good shape or if there is potentially a systemic problem. It is also possible to determine if problems exist with the data collection system or with execution of system-defined tasks. 

Another option is to have the organization s environmental health and safety staff conduct inspections. In a smaller organization, these types of inspections may adequately address the ILCI frequency recommendations. In larger institutions, the safety staff may be limited to semiannual or even annual walk-throughs. A more practical use of safety staff for inspections may be to target certain operations or experiments. Or the safety staff could focus on a particular type of inspection, such as safety equipment and systems. Finally, they could perform "audits" to check the work of other inspectors or look specifically at previous problem areas. It is important for the safety staff to address noted deficiencies with appropriate reminders and/ or additional training. Punitive measures should be employed, but only for chronic offenders or deliberate problems that pose a serious potential hazard. 

Detector builders worry that poor test results are really due to test set problems, and the customer will want to be sure that good results are real. These concerns are sometimes justified it is quite common to find errors in test setups and data, particularly with new or recently modified equipment. It is incumbent on test engineers to audit and monitor their setups to discover and eliminate potential errors. 

---