What To Do: Updating Vestigial Practices

Phrenology copy

By Phil La Duke

 Several weeks ago I posted “Ending Vestigial Practices In Safety” an article in which I pointed out that organizations looking to rapidly change their cultures run the very real risk of creating vestigial practices, that is, practices that don’t make any sense but linger around sucking up time and resources as safety professionals continue, zombie-like, to go through the motions.  In response to that article, a long-time reader challenged me to focus on what safety professionals should be doing instead.  So in response to that comment I thought I would take this week’s post to do just that.


Let’s start with the one vestigial practice that the mere thought of giving it up caused so many of you so much heartburn: audits.  A careful read of the original article will show that I wasn’t advocating for giving up audits, but I do think that their needs to be substantial re-engineering the way we conduct audits and how we interpret the data that we glean from doing them. Audit results are lagging indicators, and lagging indicators have been vilified by a lot of safety professionals. This criticism of lagging indicators is largely unfair. Lagging indicators tell us whether or not our strategies are working.  Think of an audit as analogous to stepping on the scale.  You can do all the exercise you want and change your diet, but unless you step on the scale once in a while you really can’t tell if you’re doughnut and ice cream diet is working, or if your 4 squats a day is enough to make a difference. Lagging indicators (not just audits but most lagging indicators) can provide us with critical information on our progress and can tell us (if properly interpreted) where we need to adjust our tactics to get better results.

Sadly, most audits conducted today focus almost exclusively on compliance to the exclusion of performance and, more importantly, risk.  While compliance is important, it’s not the only important element of our safety management system.  Audits need to be more balanced and more focused on the practices that put people at risk, irrespective of whether or not a law has been broken.  When I am conducting a performance audit too often people ask, “what’s the rule?” or “what is the government requirement?” These people are missing the point, it is more important to understand the areas of the operation that pose the greatest threat to worker safety than it is to check the compliance box.  Compliance audits are lagging indicators, but performance audits are both lagging AND leading indicators.  The amount of useful information gleaned from performance audits are exponentially higher than audits that focus too heavily on compliance.

Body Maps

I don’t have a lot of good things to say about body maps.  In a one of the books on human error that I read recently, (I honestly can’t remember which of the four books I’ve read in the last two week) the author described the colorful images produced by MRI’s as “brain porn” and dismissed many of the findings of the researcher’s as over-reaching speculation.  Even in doing so, the author admitted that brightly colored images of the brain tend to impress people and lend credence to the researcher’s claims.  I think there is something similar going on with body maps.  A map of the human body with little red dots where injuries have occurred really piques the interest of those who see it.  To an unsophisticated operations manager or site leader the body map creates the illusion that the safety professional knows more than he or she does.  It is a collection of data points that leads to a specious conclusion—most injuries are to the hands so we need to focus our efforts on the hand.  Unfortunately, since the severity of the injuries isn’t included, the conclusion that we need to focus on hand injuries isn’t quite right. (Yes, I get it, it COULD be, but we can’t make that call based on the information before us.)  If we use the body map in this way we risk channel scarce resources to protect workers from scrapes and bruises on their hands while allowing the occasional decapitation. Clearly no organization would be happy with this kind of trade off.

So what can we do to the body map? Well, it would be a lot of work, but if we are hell-bent and determined to use a body map, why not have three (or even four) body maps: one for first aid cases, one for recordable injuries, one for lost-time injuries, and one for fatalities.  Creating multiple maps would allow individuals to delineate between the truly serious injuries and those that are relatively minor and that carry far less risk; it’s more work, but without it the body map doesn’t really tell us much, in fact, it often misleads us.

Area Maps

Area maps are another practice that takes far more time to build than it could ever hope to provide in terms of a reasonable benefit.  Frankly, I’ve never seen one that is particularly well done from a graphics perspective (a pie chart is a better graphic, is far easier to create, and can be used to provide the same information.) Unfortunately I don’t have a “do this instead” tip for you; if you stop doing area maps I seriously doubt that anyone would notice or miss them.  As for what could you be doing instead? Virtually anything would add more value.  Some of you will continue to defend the area map, but to you I say this: if you need an area map to tell you where your trouble spots are then you probably don’t know your business well enough to be effective.

There are plenty of good ways to occupy your time instead of making area maps, one is to track the location of injuries by the type of work performed, or by individual job families. Knowing that you have 35% of your injuries during welding operations, or that 65% of your injuries are traffic related is far more valuable than having a fancy graphic to show that information.

This is not to say that knowing where on the site injuries happened isn’t valuable, quite the contrary, knowing that vehicle-pedestrian interaction is a particular problem at a given intersection can be vital information, but again, I’m not sure an area map is the best way to learn this information.

Green Cross for Safety

Okay, this practice really needs to be dumped.  I never saw any value in it; it’s one of those warm fuzzy, cutesy relics from the early 1990’s that seemed to helpful but in the final estimation provides no value whatever.  Instead of posting the Green Cross for safety why not chart the number of hazards found in the area, the average time to correct hazards, and containment measures? at least these measures shed some light on the relative risk level of a department. If we want to keep people focused on safety, isn’t it better to keep them focused on doing something positive in pursuit of safety?

Safety Observations

Safety Observations may be the single biggest waste of time in all of safetydom. Safety Observations cost a lot—we pay someone to watch someone else work.  The person observe changes his or her behavior because he or she is being watched and if the individual has the brains God gave geese the person does his or her best to do the job safely.  The observer than provides feedback on what he or she has observed.  It’s a pointless and futile gesture.  Instead, organizations should observe the work holistically.  First line supervision should be asking questions like “what is different today than it was yesterday?” or “what is out of process?” or “what could go wrong?” This kind of departmental observation is far more valuable than merely watching someone work, because it identifies issues beyond safety.

Root Cause Analysis

Certainly we need to analyze the cause of injuries, but I’m not so sure that Root Cause Analysis is the best way to do it.  Root Cause Analysis tends to presuppose that there is one (and only one) “root” cause. A root cause is the singular cause from which all other causative factors spring.  Root cause analysis is a wonderful tool for eliminating a single cause. Unfortunately, injuries seldom result from a single root cause.  Instead of using root cause analysis consider using situation analysis or a similar tool.  Situation analysis is used to determine multiple, inter-related causes that grow over time (the kind of causes that create an elevated risk level which can cause numerous injuries.) This may seem like I am making a semantic argument here, but I’m not.  Different problem structures (in this case specific and broad) with different structures (in this case sudden occurrence versus gradual occurrences) require different tools.

So there you have it, my attempt at “what to do” instead of “what not to do”  I’m sure it will still be unsatisfying to some (no harm/no foul you can’t please everyone) and even more sure that many of you will continue vestigial practices.  But consider this: as budgets get tight and you are forced to fight for every penny, shedding vestigial practices or re-engineering them so that they are useful is a good way to save money while making progress.