Unless you have been hiding far too long in the basement, you should have heard of evidence-based medicine (EBMed). For more than 20 years, and especially in the last 5, physicians have been migrating from the traditional practice of medicine to EBMed. Should we clinical engineering (CE) professionals also consider a migration to evidence-based maintenance (EBMaint)?

Traditionally, physicians have used four sets of knowledge to make clinical decisions when caring for patients: scientific principles, medical education and training, personal clinical experience, and clinical guidelines. The rapid advance of medicine, including the introduction of new technologies, has proven that the traditional four sets are no longer enough to ensure that patients are getting the best possible care. In particular, many new ideas based on sound principles that initially encouraged randomized clinical trials (RCTs) have proven to be suboptimal or even dangerous when applied to large populations.

Physicians practicing EBMed are trained to search for results of current clinical studies online, evaluate the applicability of the results found for the case at hand, and then take appropriate actions.

Today, medical equipment maintenance suffers from the same ailments that traditional medicine was suffering from before. We rely on four sets of knowledge to make maintenance decisions when managing equipment: scientific and engineering principles, such as mechanics, electronics, and medical instrumentation; biomedical engineering education and training; personal maintenance experience; and recommendations from manufacturers and experts.

The rapid advance of medical technologies has proven that the traditional four sets are no longer enough to ensure that equipment is getting the best possible maintenance. Although there have been few studies on maintenance outcomes for medical equipment, experience from other industries has shown that the traditional preventive maintenance (PM)—periodic replacement of parts and calibrations—is often counterproductive. Recommendations from manufacturers and even “experts” have typically been excessively conservative, due to a combination of liability concerns and revenue desires. Consensus guidelines are rare and seldom up to date.

Many of us are still holding on to process measures rather than analyzing the outcomes, or evidence. Granted, it is not possible to judge the outcomes if an established process is not carefully followed and completed. However, to judge maintenance outcome solely by PM completion rate is tantamount to measuring students’ learning by their classroom attendance record. In order to have maintenance data comparable to the RCTs, we need to start carefully measuring the outcomes, such as uptimes and failure rates, as we tweak the frequencies and procedures of PMs and safety and performance inspections (SPIs), as well as different strategies for corrective maintenance. Furthermore, we need to start comparing notes among institutions so the sample sizes are statistically significant to evaluate maintenance impact.

To reduce equipment failures, we need to trace back to their root causes. This can help us assess whether it is technically feasible and cost effective to deploy PMs and SPIs for each group or brand/model of equipment, or if it is better to adopt the “run to failure” strategy. Making maintenance decisions without knowing the root causes and their respective failures versus time patterns is analogous to physicians dispensing aspirin indiscriminately to all patients who have a fever, without knowing what is causing it.

To migrate into EBMaint, we will have to train CE professionals to search and critically review maintenance literature, evaluate the applicability of the methodologies, and pilot test the new ideas. The test results will be compared and discussed so consensus can be reached before widespread adoption. EBMed experience shows that the basic principles can be learned by people from different backgrounds and at any stage of their professional careers, so everyone can participate and contribute to the success of EBMaint.

What’s on Your Mind?

Got a gripe? A recommendation? Does someone or something deserve praise? Share your opinions and insights with your peers. Soapbox columns should be 700–750 words in length and can be e-mailed to .

Coming back now to my original question: Should we consider a migration to EBMaint? My answer is a resounding yes! After all, I would not go to a doctor who does not practice EBMed. So why should equipment owners want to continue to employ us or pay for our services if we do not practice EBMaint, or cannot provide concrete proof of the validity and value of our services?

Binseng Wang, ScD, CCE, is the senior director, program support and quality assurance, ARAMARK Healthcare’s Clinical Technology Services, Charlotte, NC. For more information, contact .