Human Factors: Best Practices for Scuba

The thoughts below were shared with me by a local diver, Michael Snow. I'm re-sharing them here with permission. Before retirement, Mike worked as a human performance specialist in the Aviation Safety Group at Boeing. Much of what Mike learned in decades of work on airline accident prevention also applies to scuba.  Enjoy! - Kirk

My professional ratings are from PADI and SSI. There's nothing I would take away from either syllabus, although I think the SSI material regarding human learning is a bit dated. There are a number of things I would recommend and have incorporated into my own diving habits, especially for solo divers [Please do not undertake solo diving without appropriate training, even then, it may increase your risk of injury or death - Kirk]. While I think the following would be helpful for all divers, it's perhaps best incorporated at the Stress & Rescue level and above.

1) Big data

During my career with Boeing, we entered the era of big data. We were able to build a database of 10s of millions of flights (essentially the Flight Data Recorder data) and then query that database for rare events. It was tremendously powerful in finding incidents that had the characteristics/precursors of accidents to generate preventive solutions. Similarly, in a different program, all stakeholders in the aviation system, everyone from pilots to maintenance people, are encouraged to submit narrative reports any time they notice an event with safety implications. This database was less structured and more difficult to use, but also valuable.

To the best of my knowledge, there are only two sources of data with similar safety value available to divers, both from DAN. The first isthe DAN Annual Diving Report. The second is DAN's Incident Report system, the results of which they incorporate into the Annual Diving Report and their Case Summaries. Neither is as extensive as those available to aviation safety professionals, of course. The Case Summaries are usually interesting, but rarely of broad applicability.

The Annual Report, on the other hand, I would recommend incorporating into training for students. The overlap between what students are taught, including emergencies, and what actually kills and injures divers is not 100%. I would recommend reviewing the report and including the most prevalent and/or injurious scenarios in student training. For example, we train divers for out-of-air scenarios, but when was the last time you simulated an entanglement? I've only had to cut my way out of an entanglement once (kelp while solo diving Skyline Wall), but that experience taught me never to be one dropped knife away from a bad outcome. I frequently practice deploying and stowing my knives (Recurrent training), and to always carry two, both on lanyards (Redundancy).

2) Best practices from aviation

a) Checklists

One human factors engineering best practice from aviation that has made its way into medicine and that I think should be more prevalent in the dive community is checklists. "Prospective memory" -- remembering what you need or want to remember -- is a well-known human failure mode, especially when distracted. Every Boeing and Airbus airplane designed since the 1990s has an electronic checklist. This is an electronic database of checklists, normal and non-normal, that flight crews proceed through in the course of a normal flight (e.g., pre-flight, taxi, takeoff, landing, etc.) and/or that pops up appropriately in the context of an emergency. If flight crews are distracted and put a checklist away without completing it, depending on how critical the task is (e.g., setting flaps for takeoff), an alert sounds and the checklist pops back up where they left off. Wherever possible, the system senses checklist steps and marks them completed as they are checked off (e.g., gear down for landing). This helps prevent failures of prospective memory from leading to safety-critical failures on the flight deck. The common mnemonic BWRAF for a pre-dive buddy check (Buoyancy, Weights, Releases, Air, Final check) is an example of a checklist used in diving. It's one I try never to skip, especially with a new buddy. Aside benefit of this particular checklist is that it familiarizes you with critical functions of your buddy's gear. Another common use of checklists in diving is preparing equipment, whether that's charging everything and testing it at home, or loading the car before you leave. While I don't use a traditional written checklist for these purposes, I do use what I would call a visual checklist. All my gear is stored in a designated location at home (affectionately known as the SCUBA Shed). If I get distracted while loading the car (resulting in a failure of prospective memory), I have only to look to see that there is nothing left in that location to know that I've packed it in the car: visual checklist complete, if you will.

b) Recurrent training

Both in the military and in commercial aviation, pilots undergo recurrent training on a regular basis. Every few months they go into a simulator to be trained on and demonstrate proficiency in handling a variety of emergencies, everything from depressurization to engine fires. Their training does not stop when they get their initial licenses and type ratings. I think this is a best practice that every diver should apply. I frequently drill switching to my backup air supply. My son and I usually brief that a shared-air drill is fair game any time we're at safety stop depth or above. Most of my family are divers, but only my son and I dive in cold water. Usually when we go on a tropical dive vacation together, we plan a shore dive in pool-like conditions to check buoyancy and do a bunch of drills. IMO, the last time you deployed your SMB, removed and replaced your mask, or recovered your reg should be some time in the last few months -- not when you got certified.

c) Redundancy 

Commercial airplanes have two pilots, two engines, triple-redundant hydraulic systems, triple-redundant air data systems... just about everything flight-critical is designed and operated to have "redundant and dissimilar" backups so that no single failure can take out both the primary and backup systems simultaneously. Mathematically, flight-critical airplane systems are designed, maintained and certified with sufficient backup systems operating in parallel so that none can fail more than once in a billion flight hours. Now think about SCUBA diving and safety-critical equipment and procedures. Air supply is an obvious one. In theory, everyone diving with a buddy has a redundant and dissimilar air supply being carried on his buddy's back. However, in practice, are you always close enough to your buddy that if your next breath was your last (out of air, burst hose, whatever), you could reach him, get his attention, and execute shared-air procedures? This is something I brief with my usual buddies and I don't mind getting knocked into occasionally, especially in poor viz, if it means they're never much more than arm's reach away from me. I'm also a big fan of pony bottles, with their own 1st- and 2nd-stage regulators, of course (to be redundant and dissimilar). Most of the redundancies that are part of being trained and equipped for solo diving I carry even when I'm with a buddy: pony bottle with its own regs, extra mask, two knives both on lanyards. As with commercial airplanes, SCUBA diving equipment, operations and procedures should be designed, to the extent possible, so that you are never one failure away from a bad outcome.

d) Risk Management

In airplane design and aviation safety, we have a formal practice called risk management.In a nutshell, the objective is to discover all risks associated with a design or operation and assess the likelihood of their occurrence together with the severity of the outcome should they occur. Any risk that is identified as having too great a likelihood combined with too severe an outcome is managed, either operationally or by design, so that either the likelihood is reduced or the severity of the outcome is lessened (or both) to an acceptable level. Often this is managed via redundancy. I apply this to my own diving, especially when soloing: the right pocket of my BC has an SMB, a whistle, and a mirror. I carry a Nautilus Lifeline that I test at least annually. Finally, both my primary and back-up lights have an SOS mode should I need to signal a boat or light up my SMB at night. I have assessed the risks of being left behind by a boat, swept out to sea by current, or having a medical emergency all by myself as catastrophic. Therefore, I decrease the likelihood of these risks by researching dive operators and their procedures carefully before I dive with them, looking up tides and currents anywhere I think might be current sensitive and soliciting input from other divers on these sites (e.g., via pnwdiving.com), and trying to stay in good shape and dive within my limits. I decrease the severity of the outcome should these risks occur via equipment (all the stuff I just listed) and procedures: when I solo, I don't go at night, I don't go below 60', I don't solo anywhere I haven't dived with a buddy, I plan to surface with a third of my air left for contingencies, and I text my wife when I'm going in, where, and when she should expect an okay.

3) Findings from Loss Of Control studies

Toward the end of my career I served as the industry co-chair of a large, multidisciplinary committee analyzing the deadliest category of accidents in commercial aviation: Loss of Control In-flight. This generated some interesting findings that I think apply to SCUBA diving (see https://www.cast-safety.org/pdf/JSAT-ASA_FinalReport_June2014.pdf).

a) Emergency prevention vs. emergency management

In analyzing 18 fatal accidents over more than a decade, several themes emerged. You may be familiar with James Reason's error model, often portrayed as Swiss cheese, in which it takes several holes all lining up in safety barriers for an accident to happen. We noticed something consistent with this model: in these accidents, it was NEVER one thing that caused the accident. Indeed, of the twelve themes we identified, no accident occurred with less than six of those themes. The last link in the chain of events may have been a flight crew error, but there were always several other contributors. Similarly, when I review accidents and incidents in the DAN report or other media, I almost never see one error by one diver causing an incident. Even in the case of things like "Marine Envenomation" (getting stung by a jelly, lionfish, urchin or some other critter), while diver carelessness may seem like the obvious and proximate cause, things like training, a divemaster briefing that left out the hazard, buoyancy control, or distraction are often involved. 

Importantly, in my knowledge of aviation safety and human factors, many of the factors that contribute to an accident occur before the airplane even leaves the ground: training, safety culture, maintenance, communication, and knowledge of systems are examples. My reading of SCUBA diving accidents and incidents leads me to conclude that, similarly, many factors contribute to an incident before a diver even gets in the water: training, maintenance, familiarity with equipment, communication (especially, adequacy of a dive brief), proper completion of buddy checks, knowledge and assessment of conditions, and diver fitness and/or impairment are ones I've noticed frequently in these reports. Adhering to published maintenance schedules, making sure you are properly trained for the diving you're planning to do, ensuring you're familiar with the equipment you're using (especially rental equipment, especially your dive computer), asking your buddy for a buddy check, asking the divemaster about hazardous wildlife... these are all ways of mitigating contributing factors before you get in the water. 

Lastly, humans are much better at establishing situation awareness, diagnosing failures, and managing emergencies when they are sitting somewhere reading about it than when they are actually in one. Life-threatening emergencies are almost always unexpected. Situation awareness has usually been lost, attention has narrowed, there is usually significant time pressure (real or perceived), and there are often significant physical stressors (e.g., cold, noise, exertion, G, breathing difficulty). All of these have detrimental impacts on human cognitive performance. Practice can mitigate these effects somewhat. 

Also, a conscious strategy of stopping to assess the situation, rebuild situation awareness, and formulating a plan can be helpful. Interestingly, in my training as a medic in the Army, as a pilot, and as a SCUBA diver, I've been taught the same mantra: Stop, Think, Act.Good advice. By far the best time to manage an emergency is before youare in one.

b) Mismanaged distraction. 

One of the themes we identified in every single accident we analyzed was "Mismanaged Distraction". Pilots actually train for distraction, both individually and as a crew. It's something you get tested on when you get a private pilot's license. Despite this, our committee of subject-matter experts (including several airline pilots) was able to identify some critical distraction in each accident sequence that took the flight crew's attention away from their fundamental task of keeping the airplane upright and within its flight envelope. These distractions ranged from communication with air traffic control, to preoccupation with weather, to the autoflight system, to a system failure: a system failure that would not have threatened flight safety had it not distracted the flight crew for some critical period of time (a few seconds to a few minutes, depending on phase of flight). I have noticed a similar theme in my reading of SCUBA accidents and incidents. 

Many of us carry a distraction with us on most dives: cameras. My wife has claimed that diving with me and my camera is like diving with half a buddy. As a result, I take extra care to cross-check my buddy and my gauges when diving with a camera. I make a point of consciously checking on my buddy, my air, my depth, etc. regularly and frequently. I do not take my camera on the first dive of a dive trip. I also leave it behind when I'm diving with a new buddy for the first time. 

Another common distraction is gear issues. A mask fogging up is a classic example. Safety critical? No, but not being able to see clearly and having to frequently clear your mask throughout a dive are definitely a distraction. I would contend that this takes time and attention away from checking your air, checking your depth, checking your NDL, checking on your buddy... tasks that are critical to safety.

c) Again, training

Half of the accidents we analyzed were associated with some deficiency in training. Typically, this was failure torecall and act in accordance with training, training that was counterproductive in the situation actually encountered, or training that was nonexistent: it existed, but the flight crew in question had never had it. In human factors engineering, we consider training a weak safety barrier: for it to prevent an accident, the operator has to have had the training, recall it correctly in the moment, and apply it in the correct context. It's not as effective as simply designing ahazard out of a system. However, it is important and many examples exist in which flight crew training has saved the day (e.g., the Miracle On The Hudson). Recalling it correctly and applying it in the correct context can be improved with, you guessed it, more training.

To the best of my knowledge, all human knowledge, skills and abilities decay with disuse. This applies to algebra, foreign languages, and how to do a shared-air ascent. The more likely a situation is to occur, and the more safety-critical the situation is, the more handling it effectively should be practiced.