After something over a half-century of examination of the causation of slips and falls many erroneous concepts have crept into the "scientific" literature through the writings of well-intentioned investigators who have felt impelled to fill in voids in the information data base by drawing upon their intuition rather than careful observation. Some of those myths are dispelled here.

Unfortunately for the world's population, many of those who have set about to study slip/fall causation have not proceeded from a foundation of safety engineering knowledge based upon heavy experience in the investigation and prevention of slipping accidents. Some have impressive academic credentials and access to some of the finest laboratory facilities available, but their inexperience in the investigation of actual falling incidents has allowed them to come to conclusions that seem plausible to them, but which upon more careful examination are discovered to be unfounded.

This premise presents several problems. It is not primarily the coefficient of friction of a floor surface that determines its safety for walking. It is usually a localized spot that is slipperier than the rest of the floor that causes the unsuspecting and unprepared pedestrian to slip and fall. In normal walking, the maximum traction demand is in the "pushing off" phase when the toe of the backward foot is in contact with the floor. If the pedestrian's toe starts to slip backward, usually it does not present a fall hazard, but that occurrence tips off the human computer to the fact that there is a slippery surface under foot so the pedestrian applies a muscular bias against the slipping tendency and shortens up the stride so that the serious heel slide does not occur.

People can walk safely on surfaces that are slipperier than ice, if they know of the hazard (Vidal 1994.) In fact, some surfaces (such as dance floors and bowling alley approaches) are waxed to make them slippery, and nobody falls down from walking over them. But if an unexpected slippery spot occurs on a floor with an overall COF of .5, a classic heel slip could be expected to occur.

This is not to criticize the .5 COF as a threshold of safety. There is little controversy over the safety of the .5 number. Most people can walk on most surfaces with a slip index of less than .4, and the .5 gives a margin of safety. But it is not the overall traction performance of a floor that is usually the problem. It is that unexpected wet or greasy spot that can take the number down below .2 that causes most slip/fall accidents.

Therefore, in order to evaluate the degree of hazard of contaminated surfaces, it is necessary to be able to measure their traction properties in wet or lubricated conditions. If a floor becomes dangerously slippery when wet, it is necessary either to keep it clean and dry while people are expected to be walking on it, or it must be treated to make it safer under the contaminated conditions.

No inferences concerning the safety of a surface while in a wet or lubricated state can be made from traction measurements performed on it while it is dry.

It may be true that a toddler's gait amounts to a series of controlled falls, but it is observable that the mature human organism is a masterpiece of coordination with a mental computer that is able to control a marvelous musculoskeletal apparatus without conscious thought. People not only walk in almost perfect fluid action (Sussman n. d.), but it is a wonderful thing to behold a halfback who is expert in broken-field running so as to be able to sprint at high speed through a hostile defense that is trying to tackle or knock him down, all the while spinning, feinting and varying his speed. Anyone who has been a defensive line backer knows the problem it is to upset the equilibrium of the ball carrier so as to cause his fall to the ground. Any human falling tendency would appear to be very well controlled indeed among normal adults.

Less dramatic evidence of the phenomenal stability of the walking man is the ability of the physically fit to recover from most trips or minor slips by reflex action, without conscious thought.

Fall statistics published annually by the National Safety Council show that the portion of the population at greatest risk of falling injuries is post menopausal females, with the curve rising steeply after age 65. This suggests that people become more vulnerable to falls (1) as their reflexes slow and their muscle tone declines, and (2) the declining bone strength, particularly among those afflicted with osteoporosis, predisposes them to catastrophic injury from what would be a minor slip or trip to a younger, more athletic pedestrian (English 1994.)

The first time I read this, I looked up vault in the dictionary to make sure it meant what I thought it meant, and sure enough, it means jump. Careful observation of ambulation, especially with the aid of strobe photography or digitized video, shows that the motion of the foot in the landing phase is very smooth. The forward velocity of the foot drops to approximately zero with the edge of the heel within a small fraction of an inch of the walkway, and as the heel lands, the shoe bottom is at an angle of perhaps 30 degrees from the floor. The foot then quickly rotates about the center of the ankle joint until it is down flat on the floor. There is no jumping in this fluid motion.

In normal walking, the advancing foot swings forward attaining a velocity somewhat greater than twice the speed of the forward progress of the pedestrian. Then as it reaches the limit of swing dictated by the human computer, a powerful muscular bias is applied to the advancing leg to arrest its forward motion so that it is approximately zero just as the heel contacts the surface; and that muscular force is continued in a backward direction so that forward propulsion is maintained. This muscular force is a bias applied against the tendency of the heel to slide forward.

In a workshop conducted at a university ambulation lab where several athletic young men were hired to walk across a force plate at different speeds while wearing different shoes, an attempt was made to videotape an actual slip and fall occurrence. In hundreds of trials over several days, no slip/falls occurred, though many microslips were observed. The pedestrians had fast reflexes and good muscle tone, and there were no surprises encountered as they walked across the force plate. They were able to arrest every slip in its incipient phase by the rapid application of a muscular force that prevented a full uncontrolled heel slide.

Stride length must be considered in connection with other factors involved in foot traction during ambulation.

Particularly professors of physics or engineering who imagine themselves to be tribometric experts tend to remember an hour's lab in freshman physics where a block of steel is placed on a planar surface that is then tilted until the block slides. An alternative method is to set the block on a level surface and measure the force required to move it. From either of these experiments a static coefficient of friction number is calculable by applying a trigonometric or algebraic formula. The bad news is that neither of these traction measurement methods has anything to do with pedestrian activity.

These tribologic demonstrations are supposed to illustrate principles stated by Leonardo da Vinci and Charles Coulomb, principles that have been generally disproved by twentieth-century investigations (Derieux 1934), especially when rubber is the material of the shoe bottom.

When the heel comes down on the surface in walking, it presents a thrust vector that applies both normal and horizontal forces simultaneously. When forces are applied in this manner, the result is much different than when the "heel" is in contact for a period of time prior to application of the horizontal force. This is critical when wet surfaces are involved. The resulting sticktion produces misleading indications. Measurement of pedestrian slip-resistance is an incredibly complex exercise requiring certain knowledge and expertise in ergonomics and fall accident experience. Not just any engineering professor is so qualified.

As scientific as that sounds, existing testing protocols are not valid for application to most slipping incidents. As of this writing, all recognized US testing standards specify that the test surface must be dry for the test, except for the latest Underwriters Laboratories method (UL 410 1992,) which purports to take valid wet readings with the old James Machine, something the writers of ASTM standards for the James Machine (F 489, D 2047, F 695) know is impossible. Unfortunately, the overwhelming majority of falls occur because of some lubricating contaminant that makes the floor slipperier than it is while in a dry state.

Responsible investigators know that all of the recognized methods for measuring static coefficient of friction on dry surfaces will produce misleading results in the presence of lubricants, especially water (English 1994.) Typically, using existing ASTM methods to meter wet surfaces will "show" them to be more slip-resistant than when dry, even though everybody knows that is wrong.

All drag-sled meters and the traditional articulated strut instruments give misleading results on wet surfaces and are useless for safety engineering studies where surface contaminants are present. Similarly, their test results on wet surfaces cannot be expected to prevail in matters of litigation, if opposed by competent expert opinion.

While that may be true in some abstract sense in that various individuals do present minor nuances in walking style, analysis of force plate traces shows that the vast majority of people create similar forces, at least for those critically affecting traction. Studies (Hunter, Jablonsky and Merscher 1985, Schwartz et al 1947) show that a single person walking across the force plate many times presented about the same envelope of force traces as many different individuals did.

Since the vast majority of people apply similar force vectors in the heel-strike phase of walking, there is no technical reason why a practical slipmeter that mimics the human pedestrian's foot cannot measure the traction available to people walking on a given surface.

A special test grade of Neolite developed and produced for use in tribometry is gaining increasing acceptability as a standard slider pad for slipmeters.

Pedestrians who select shoes having unreasonably slippery bottoms present a risk largely beyond the control of the building owners. Those who slip because of shoe slipperiness should look to the shoe manufacturer for remedy rather than to the building owner, in most cases.

A transatlantic battle has raged for decades as to which is the relevant factor. US investigators have tended to favor the primacy of the static COF index, while the UK and European experts have stoutly maintained that it is not only the dynamic (or kinetic) figure that it significant but that there is no such thing as static COF.

A more careful consideration of the whole debate will suggest that the entire argument is wrong. What is significant is the actual slip-resistance values operating under the shoe bottoms of actual pedestrians. Therefore, what is needed is a slipmeter that mimics the action of the human foot in ambulation (Andres 1985, Adler and Pierman 1979.) What difference then would it make whether we theorize that it is static or kinetic friction we are trying to measure?

The engineers tell us that static coefficient of friction can only obtain on dry surfaces, and since we are concerned primarily with the metering of wet or lubricated surfaces (bearing in mind that it is the unexpected lubricated spot that produces most slip/falls), the argument becomes academic anyway.

Apparently some eminent investigators have made trigonometric calculations based on stride length and have formulated a theorem that a pedestrian's foot requires less slip-resistance while walking on stairs. But experienced accident investigators know that this theory is false. More slip-resistance is required on stairs than on floors (English 1989.)

This is because most people descend steps by placing the ball of the foot just behind the nosing of the treads, and as the weight is transferred to the forward foot, the retreating foot rocks forward and slips off of the step edge. It is this rocking/slipping action that determines the friction requirement. It is particularly significant because if the foot unexpectedly slips even an inch, a fall is likely, with a much smaller chance of recovery than on a level surface.

The academic investigators are right, however, that stair nosings should be uniformly slip-resistant. As the pedestrian's human computer factors the slip-resistance of the first couple of treads into its control of ambulation, and an unexpectedly slipperier nosing is very likely to produce the uncontrolled slip.

Stair nosings should be more slip-resistant than floors also because stair falls tend to produce more severe injury than slip/falls on the same level.

Concluding Comments
Technological evolution of the tribometric art has been progressing rapidly. Many practitioners who were on the leading edge only a decade ago have lost their competence because they are still applying premises now known to be false and obsolescent test methods, not knowing that progress has passed them by.

Since falls on the same level and falls on stairs and platforms are the dominant loss type in the vast majority of industries, a great opportunity for accident reduction can be found in applying new scientific methods to this age-old problem. Those persisting in the application of anachronistic theories will not be successful in protecting the people and other assets in the enterprises utilizing their services.

References
Adler, S. C. and Pierman, B. C. 1979, A History of Walkway Slip-Resistance Research at the National Bureau of Standards, NBS

Andres, R. O. 1985, Essential Design Criteria for an Ergonomically Sound Portable Slip-Resistance Tester, and Other Field Measurement Considerations, University of Massachusetts Department of Exercise Science

Derieux, J. B. 1934, "The Coefficient of Friction of Rubber, Journal of the Elisha Mitchell Scientific Society, 50, 53-55

English, W. 1994, "The Validation of Slipmeters," Proceedings of the Annual Conference of Ergonomics Society, University of Warwick, Coventry, England, April 1994

English, W. 1989, Slips, Trips and Falls: Safety Engineering Guidelines for the Prevention of Slip, Trip and Fall Occurrences, Hanrow press, Del Mar, CA

Hunter, R. J., Jablonsky, R. D. and Merscher, J. H. 1985, "Development of Coefficient of Friction Measurement Methods by Consensus-Standards-Making Organizations," Ergonomics, 28, 1055-1063

Perkins, P. J. 1978, "Measurement of Slip Between the Shoe and Ground During Walking," In Anderson/Senne (ed.) ASTM STP 649 Walkway Surfaces: Measurement of Slip Resistance, American Society for Testing and Materials, West Conshohocken, PA, 71-87

Schwartz, R. P., Heath, A. L., Morgan, D. W. and Towns, R. C. 1964, "A Quantitative Analysis of Recorded Variables in the Walking Pattern of 'Normal' Adults," The Journal of Bone and Joint Surgery, 46, No. 2, 324-334

Sherman, R. M. 1992, "Preventing Slips that Result in Falls," Professional Safety, March 1992, 23-25

Sigler, P. A., Geib, M. N. and Boone, T. M. 1948, "Measurement of the Slipperiness of Walkway Surfaces," J. Res. National Bureau of Standards, as quoted by R. Brungraber in An Overview of Floor Slip-Resistance Research with Annotated Bibliography, National Bureau of Standards 1976, 63

Sussman, H. and Goode, R. 1976, The Magic of Walking, as quoted by R. Brungraber in An Overview of Floor Slip-Resistance Research with Annotated Bibliography, National Bureau of Standards, 1976, 31

Underwriters Laboratories, Inc. 1992, Slip Resistance of Floor Surface Materials, UL 410, First Edition

Vidal, Keith 1994, "The Americans with Disabilities Act and Slip/Falls," Proceedings of the National Conference of National Academy of Forensic Engineers, Kansas City, July, 1994

Winter, David A. 1990, Bio mechanics and Motor Control of Human Movement, Second Edition, Wiley-Interscience, 605 Third Avenue, New York, NY 10158-0012, 1990

Note: The material contained in this article (plus a number of figures and illustrations omitted here out of respect for bandwidth limitations) appears in an expanded form as Chapter 5, "Myths about Pedestrian Slip Resistance" in my book Pedestrian Slip Resistance, How to Measure It and How to Improve It as well as in the Second Edition.