Equipment, Calibration and the Quiet Failure

Part of Research Assistant (ASSIST) Training: Preparing Visits, Rooms and Supplies.

About 1 minutes. Free, in your language, no account needed.

A machine that is wrong does not say so

Scales, blood pressure monitors, thermometers, centrifuges and ECG machines all drift. A machine reading consistently two kilograms light produces perfectly plausible numbers that are all wrong.

Nothing about the reading looks unusual. That is what makes it a quiet failure, and why calibration is a scheduled requirement rather than something done when a device seems off.

What is usually required

A record showing each piece of equipment used for study measurements has been calibrated or serviced, when, and when it is next due. Some studies require the calibration certificate to be filed.

You may be asked to check these dates as part of setting up. If something is out of date, it is not used for study measurements until it is dealt with.

Use the equipment the protocol specifies

If a study says measurements are taken on a particular device, using a different one because it is nearer produces a number the study cannot compare with the others.

The same goes for using the same device across visits for one participant where the protocol asks for it. Consistency is what makes a change over time meaningful.

Report a fault rather than working around it

If a device seems wrong, stop using it for study measurements and tell someone. Do not take a second reading on another machine and use whichever looks more sensible.

A fault reported is one repeated measurement. A fault worked around silently affects every measurement taken since it started, and nobody will know when that was.