It is important to measure the impact of your program. While the primary goal of collecting outcome data is to improve patient care, outcome data can serve many other purposes. Tracking appropriate outcome measures allows you to fine tune your intervention program to maximize successful implementation. It is important to demonstrate success to your multidisciplinary team to ensure members' ongoing active participation and compliance. In addition, the information may prove valuable in convincing your leadership to keep supporting your initiative administratively and financially.
5.1 Types of Outcome Measures
Outcome measures can be divided into three major categories:
a. Measuring the impact of your intervention
The goal of your intervention program is to reduce the horizontal transmission of KPC to prevent patient colonization. Ideally, you should track the prevalence of KPC colonization (number of KPC positive patients among all those screened) before and after your intervention. If baseline data are not available, track KPC prevalence rates for downward trends. In the prevalence rate, each patient can only count once in the numerator, even those who have been repeatedly positive on weekly sampling. A second helpful measure is to track the number of patients who initially tested negative for KPC and then tested positive. This outcome targets and measures the impact of disrupting horizontal transmission of KPC and preventing colonization. Finally, a crucial outcome measure to track is the percentage of total patient KPC colonization days spent in contact isolation. This is a good indication of how well your intervention is being executed and, in particular, will reflect the timeliness of patient sampling, turnaround testing times, feedback of results to clinical staff, and initiation of contact isolation. Institutions may also want to track actual KPC infections, but it is important to realize that the number of infected patients will be low, even in facilities that are hyperendemic for KPC. Once KPC is established in a facility, there will be approximately 50–100 KPC colonized patients for every detected KPC infection. Monitoring infections as the only outcome can be quite misleading in this setting.
b. Measuring potential confounders
Even a well-designed and executed intervention can fail if hospital staff do not comply with the requirements of contact isolation when it is initiated. You may consider tracking staff compliance with hand hygiene; proper use of barrier precautions, including gloves and gowns; and compliance with placement of patients in single rooms. This is best accomplished by periodic unannounced direct observation sessions to ascertain compliance with the elements of contact isolation. Tools for this purpose can be found in Tool 5A, Infection Control Observation Tool and at http://www.jointcommission.org/Measuring_Hand_Hygiene_Adherence_Overcoming_the_Challenges_.
While seemingly reviewing a patient chart, the observer documents compliance on a checklist tool for a predetermined number of staff-patient encounters. All types of staff need to be observed (physicians nurse, respiratory therapists, etc.). Typically, 30 staff-patient encounters a week is sufficient. This information should be shared with the health care team, and a plan to correct deficiencies should be implemented when necessary, keeping in mind that failures are often systemic and not solely attributable to individual noncompliance. For example, systemic failures can occur when—
- There is a lack of communication about a patient's colonization status.
- There is a lack of accessible personal protective equipment.
- There is a lack of understanding about the reasons for the contact isolation.
Staff members are frequently not aware of the clinical importance of KPC, even at endemic hospitals, and the fix for poor compliance with contact isolation may need to include staff education about the epidemiology and clinical importance of KPC.
c. Measuring unintended negative outcomes
Your KPC control program maybe highly successful but may simultaneously negatively impact other aspects of patient care and hospital operations. Many hospitals already struggle with other multiple MDRO organisms, and the availability of single rooms for contact isolation may already be at a premium. Additional demand for single rooms for the KPC program may result in difficulty in transferring patients within the hospital when the required level of care changes. Tracking delays in transfer related to the inability to find an appropriate single room for continued contact isolation may be important to continued acceptance of your intervention. Concerns about this issue should be explored with staff in targeted areas during the planning stages, and when appropriate, delays should be tracked and measured. Knowledge of KPC colonization status may inappropriately influence the selection of empiric antibiotic therapy. In addition, if PCR detection is used. It is important to remember that these assays are not FDA approved and results should not be used to guide patient therapy. Finally, it is not clinically appropriate to base therapy on colonization status. You may consider tracking polymyxin use before and after you initiate your intervention on the targeted clinical units.