Provides staff and management with objective insight into the execution of data management processes and the effectiveness of associated work products.
Once sound data management processes are established and operational, it is important to ensure that established processes are being executed as intended and delivering expected benefits, also known as process quality assurance. Implementing this process supports highquality delivery at all levels by providing visibility into and feedback on processes and their associated work products.
Activities involved in performing quality assurance for processes include:
Objectivity in quality assurance evaluations is important to yield meaningful feedback and improvements, and it can be achieved by employing unambiguous criteria and independent evaluators. Less formal approaches, such as peer reviews, are appropriate to assist process performers day-to-day. For critical processes, a more formal approach may be employed periodically to ensure objectivity, including: audits by internal, separate quality assurance staff; in-depth analysis of process execution as it is performed; quality assurance reviews of work products; and process checks added into activity steps to prevent failures.
To consider an example of creating a quality assurance template for a data management process, take the case of an organization which is implementing its new data quality assessment process with the objective of improving the quality of patient demographic data. This process involves a number of activity steps and staff roles (See Data Quality Assessment). The table below depicts selected initial activity steps and indicates quality assurance considerations that can be applied:
|DQA Activity Steps||Work Products||Quality Assurance Factors|
|Convene working group of experts across the patient demographic lifecycle||
Documentation of current data quality issues
Fitness for purpose description
Verify if all relevant stakeholders were involved
Review list of current data quality issues
|Finalize initial set of key patient demographic data elements||Data element list for quality assessment||Review list and ensure that all identified data quality issues are aligned to one or more data elements|
|Define thresholds and targets based on business needs and selected data quality dimensions||For each data element, defined threshold (lowest acceptable quality) and target (aspirational level of quality)||
Verify relevant data elements thresholds and targets
Review quality dimensions applied and rationale provided
Review thresholds and targets for relevance
In organizations that have succeeded in evolving a quality-oriented culture (See Data Quality Planning), the process quality assurance role can be performed completely by peers, and the quality assurance function can be an activity step in process performance. Those conducting the peer review should understand the process and be familiar with the work products used or produced by it. Reviewers of work products should not be the same individuals involved in creating or maintaining them.
It is advised to create a review template or checklist to ensure that the same steps, standards, and approach are employed on every occasion. Once the quality assurance process has been developed and employed, it quickly becomes embedded into the organization’s culture and typically is valued by staff at all levels.
Data management process quality assurance results should be communicated to governance, and if significant issues have been observed, they can be escalated as necessary. When noncompliance issues are identified, it is best to first address and resolve them, as possible, within the performing team. Quality assurance results often lead to discovery of useful improvements to the process itself or enhancements to standards and work products.
It is ideal to consider quality assurance when developing a new process or directly following rollout to ensure that applicable policies, standards, work products, and procedures are incorporated into the plan for assuring effective execution.
Implementing process quality assurance will enable the organization to realize these benefits:
Issues in process performance and corresponding work products can be triggered by a variety of internal and external circumstances. In some cases, poor quality of data inputs to the process may be a root cause of process quality issues. In other cases, the data inputs may be sufficient, but the process corrupts the quality of the data outputs. Whatever the case, process quality and data quality are interdependent and should be considered together.
For example, duplicate patient records can result from inadequate controls during the original registration of a patient. Without considering the quality of the process, the root issue may go undetected and result in repeated data cleansing by downstream consumers of the data.
Alternatively, process quality issues may result from lower throughput of patients in the registration process, holding up the delivery of care due to redundant manual steps in the patient registration process. However, the redundant steps may have been performed to reduce negative impacts on patient identity integrity. In this case there is a risk that process improvements in the registration process may result in more duplicate records inadvertently impacting downstream consumers of the data.
Example Work Products
Process improvements should be made through careful analysis of the steps performed ensuring that they accomplish the objectives for the process. While the specific purpose of data management processes varies along the continuum of patient care, many will share steps at a high-level, due to the commonality of the data. If these processes are performed differently, the quality of the data can be negatively impacted.
Process standards support consistent performance along the lifecycle of patient demographic data promoting resource efficiency, resource mobility, and fewer data defects. Documentation of process standards, descriptions, and procedures enables the objective evaluation of important processes impacting data quality, such as registration intake, checking for duplicates, informing originating sources about duplicates, changes to standards, etc.
Any variation in the performance of a process will impact downstream activities along the patient demographic data lifecycle – existing process defects, as well as new improvements, are no exception. Accordingly, communicating process changes to all relevant stakeholders across the patient demographic data lifecycle is essential.
In some cases, negative impacts on other processes may be incurred to support the greater goal of patient safety (e.g., writing ‘Jane Doe’ as the name of an unconscioius emergency room patient). Without a a mechanism for monitoring process improvement compliance, adoption of changes is likely to be compromised. This illustrates the role of governance to collectively review process improvement recommendations and indicate potential negative impacts.
Example Work Products
Establishing and adopting process standards, descriptions, and procedures is not enough to ensure that results are improved across the patient demographic data lifecycle. The one-to-many relationships between standards and the implemented environment pose risks for any changes that may result in unanticipated negative impacts on the quality of patient data and ultimately on patient care.
Standard process quality measures and metrics support the objective evaluation of the effects of process improvements, especially in a complex chain of linked activities where little informal communication may occur between the various stages of care. In such cases, standard measures and metrics clearly communicate performance expectations that support the overall objective of patient safety for participants in every related process, as well as fostering achievement of data quality objectives.
Example Work Products
1.1 Are issues with data management processes and work products identified and addressed in the context of the entire patient demographic data lifecycle?
2.1 Are selected data management processes and corresponding work products objectively evaluated against process descriptions, standards, and procedures?
2.2 Are quality issues with processes and work products communicated to stakeholders and are noncompliance issues resolved?
3.1 Does the organization maintain and apply standard measures and metrics to process quality issues?