Customizing General Risk Assessment Results for Pediatric Patients
It is possible to adjust the outcomes of a risk assessment conducted with an adult (or non-specific) population to account for the safety risks specific to the pediatric population. This method allows for modification the risk results (frequency, consequence) based on a review of the relevant Child Specific Risk Factors to better specify the risks presented by pediatric patients

Example: Research shows that the potential for serious adverse drug events within the pediatric inpatient population is about three times as high as among hospitalized adults in large part due to variations in physical characteristics such as weight. Therefore, two categories must be adjusted:

• Increase frequency/likelihood score
• Increase consequence score

After these scores have been adjusted, the risk bin may need to be revised as well. This adjustment allows for prioritizing risks specifically for pediatric patient safety improvement and a focus on the nature of the risks for pediatric patients.  In addition this method can help to identify aspects of an improvement intervention that may need to be customized in order to avoid increasing risks to pediatric patients from a general safety intervention.

Common Fail-points for Process Steps  [back to top]
Your institution may have an interest in understanding the risks and risk contributors for a particular set of process steps to understand generic risks.  This analysis enables a review of a process step across existing risk results.

Use the Risk Assessment Review Sheet with risk assessment results entered from the studies of interest and “Risk Binned” to reveal the High and Medium risks.  Add a column, Process Step Type.  Categorize process steps across the included risk assessments into groups of like types.  This can be done for similar or divergent FMEA studies.  Organized the process steps into like groups, including, the related Fail-points and Failure Mode Causes and Risk Contributors.  This will provide you with an array of the fail-points and fail point causes for a specific process step.  This analysis can help to direct toward interventions for a type of process step, that if standardized across other general processes, may improve safety across a variety of processes.  It is important to assess if there are divergent features or additional considerations for each particular context in which the process is applied.

Common Cause Analysis  [back to top]
Through this analysis underlying causes and risk contributors that contribute to varied processes, process steps and fail-points can be identified.

Use the Risk Assessment Review Sheet with risk assessment results entered from the studies of interest and “Risk Binned” to reveal the High and Medium risks.  Add a column,  Cause Type.  Categorize failure mode causes across the included risk assessments into groups of like types.  This can be done for similar or divergent FMEA studies.  Organized the process steps into like groups, including, the related Process steps, Fail-points and Risk Contributors.  This will provide you with an array of the most frequent and highest consequence causes and their related process steps and specific fail-points.  This analysis can help to direct toward interventions to address specific underlying causes that may cut across processes and fail-points.  It is important to assess if there are specific features related to these identified causes that must be addressed as well in order to eliminate or reduce their occurrence or mitigate their influence on the process. 

Categorizing Failure Modes  [back to top]
Review all of the high and medium risk process steps and developed a list of generalizable failure categories. Within these categories, the group will develop representative cases that are emblematic of the risk scenarios identified. Selection of representative cases will be based on the child-specific factors (e.g., medication ordering, accident category, variable size and weight, and immature physiology of a young child), the emergency medicine performance-shaping factors (e.g., distractions and noise, verbal communication of orders, and lack of pediatric trained staff) and underlying risk conditions of high or medium risk fail points. 

After the members of the LEARN team completed categorizing the fail points the coding was reviewed for discrepancies across reviewers. Any discrepancies in team member evaluations are reconciled to consensus based on discussions of the elements, the specifics of the clinical context, and the risk assessment methodologies.

Through the use of these categories specific features of the risk context were described that could inform “Representative Cases” (emblematic risk scenarios).

Developing/Selecting “Representative Cases”  [back to top]
Developing/Selecting “Representative Cases” (emblematic scenarios) representing the characteristics of risk, including specific risk types, features, and complexity is a key part of the LEARN project. Apply categories of characteristics of the process step and failure modes. For our project analysis the LEARN Team applied several types of categories to the process steps and fail points to illuminate important contexts of interacting risks including:

• Event Type
• Child Specific Risk Factors
• Performance Shaping Factors
• Problem Type
• Omission/Commission

The results of this analysis can be used to construct and assess potential interventions and solutions to address the assessed risks and risk contributors.

The results of this analysis can also be applied to assist with the construction of simulation scenarios and scenario triggers.




Development of Interventions Based on Representative Cases  [back to top]
The identification of representative cases from the results of FMEAs of healthcare processes can inform the development of interventions to reduce the risk to patients. This is accomplished by using of a set of preferences for intervention development along with the specific information contained in the descriptive information which defines the representative case.

The preference for interventions is as follows:

1. Eliminate the possibility for the failure (change the process to eliminate the activity or equipment that is postulated to fail)
2. Prevention over mitigation (change to reduce the likelihood of the failure rather than mitigate the consequences)
3. Engineered intervention over administrative intervention (design a process feature to address the failure rather than add a process step to check on the failure)
4. Passive intervention over active intervention (have an inherent safe state of a proves step rather than require an activity to make a process safe)
5. Intervention closest to the failure (provide a check or correction as soon after the failure could occur rather than at the end of a process)

Using the intervention preferences defined above, the information describing the representative case is examined to determine the potential for changes to the process.  This information includes the description of the process step, failure mode, failure cause, and failure outcome.  Examine the characteristics of the failure mode and associated information and develop potential interventions that address the type of failure, causes, and/or outcomes.  Keep in mind that the aspects of the failure can be addressed by eliminating them or reducing the likelihood and/or consequence of the failure.  However, to understand the risk impact of any intervention it must also be analyzed to determine what the risk impact of it would be with the intervention present in the process. A few examples of this activity are:

1. If the process step is “fax a reply to the sending institution” and the failure mode is “fax machine out of paper”, this failure could be eliminatedby replacing the fax machine with another means of communication.  However, to reduce risk associated with this process step, the replacement communication means would need to be more reliable (e.g., have a smaller likelihood of failure). 
2. If the process step is “check patient vitals” and the failure “patient’s vitals incorrectly recorded” adding technology that directly records a patient’s vital signs would reduce the likelihood of this failure.

If the process step is “label the sample” and the failure mode “sample not labeled”, adding a process step to check for a label before letting the patient leave the lab would add an intervention close to the failure.

Assessing Systemic Vulnerabilities  [back to top]
Using a qualitative emergent themes approach across a set of risk assessments either of the same process or of disparate processes underlying systemic vulnerabilities can be identified for intervention.  This review can reveal large themes of risk across the practice and organization of medicine.

Themes are inductively developed from review of the whole assembled set.  The process of this analysis of risks include iterative steps: 1) overall review of the fail-points and causes; 2) detailed review of a few process fail-points, causes and risk contributors to formulate meaning; 3) review of additional process fail-points, causes and risk contributors to develop preliminary themes; 4) review the assembled process fail-points, causes and risk contributors to identify additional themes and 5) finalize themes and theme definitions. Definitions should establish the boundary of the theme - what is included in the theme and what is excluded.

Risk Binning  [back to top]
Risk Binning id done as the first step after assembling the Risk Assessment Review Sheet for each of the following assessment methods.

Risk Binning involves using the frequency, consequence and safeguard scores to assess the level of risk of a Fail-point using the following table.  After assessing the level of risk – Low, Medium or High – create a Risk Assessment Review Sheet with only the High and Medium fail-points remaining. This is the risk data on which you can conduct further analyses.

Table 4. Risk Binning Table