Over more than a decade, Assure360 has accumulated a substantial body of data on how licensed asbestos removal work is carried out. We’ve always used that information to help individual contractors understand their own performance. But it also gives us the opportunity to ask some much broader questions about the industry.

To begin that process, I asked Jack Garland to take a look at our data. Jack recently completed his master’s degree in Mathematics and Statistics at Oxford, and is now starting his PhD there. Importantly for this work, he approached the data as a statistician, rather than beginning with an industry assumption that he was seeking to prove.

What causes an enclosure to pass or fail?

The first question was a broad one. Could Jack identify any relationship between whether an enclosure passed or failed and the circumstances surrounding the work? He considered a wide range of variables, including enclosure size, airflow, project duration, working hours, rest periods, short-term workers, anticipated exposure and the experience of supervisors and contracts managers.

The answer was surprising. None of the variables examined provided statistically significant evidence of why an enclosure passed or failed. This did not mean these things were unimportant, it meant the data was not supporting the relationships we might have expected to find. So, we asked a different question: could the data tell us whether the clearance procedure itself was being followed correctly?

This produced a much more striking result.

The formal handover from the supervisor should mark the point at which the enclosure is presented to the analyst for the independent four-stage clearance (4SC). The analyst shouldn’t begin the 4SC process until that handover has been received. From then on, the 4SC is under the analyst’s control.

We examined 11,899 analyst sign-offs. Of these, 1,234 successful sign-offs occurred within five minutes of the supervisor’s formal handover. More than 800 were recorded no more than one minute later. In other words, more than one in ten successful analyst sign-offs was recorded within five minutes of the supervisor’s formal handover.

A four-stage clearance cannot credibly be started and successfully completed in five minutes. In these cases, one of three things must therefore have happened:

• The clearance work began before the formal handover was received
• The records were completed retrospectively
• The full procedure was not followed

The contractor is responsible for completing the handover properly. However, the analyst controls the clearance and should not start without it. Jack’s findings therefore raise a thorny question for analysts and analytical organisations: why are so many successful clearances being signed off almost immediately after the formal handover? None of the three above scenarios is compliant, and therefore none of them is acceptable – or particularly palatable.

This is only our first look at the data. We now want to establish how many of these cases represented first-time passes, and how many followed earlier failures. Among other things, we will examine whether enclosure size and complexity affect clearance times, whether practices have changed over time, and whether there is any difference between day and night shifts.

This will be the first in a series of articles extracting insights from our data, each asking a different question of the Assure360. Jack will undertake the statistical analysis, and I will consider what the findings mean in the context of regulations, guidance and real-world industry practice.

We intend to publish the research and methodology in full in the new year. In the meantime, we will share the findings one question at a time.