Step Strategies When Recovering Balance During a Forward Fall

This summer I have been doing biomechanical engineering research under Dr. Darryl Thelen, Ph.D., in which we study the movements and reactions of the human body from a mechanics standpoint. This research is building on that of past years and of other people, and our ultimate goal in this series of studies is to develop a program to prevent falls in the elderly. In the past two months, we have been exploring the results of the young and the old, and of males and females. From our data, we hope to be able to quantify the biomechanics of balance recovery and determine if an exercise intervention program improves the ability of the elderly to recover balance.

During the first few weeks of research, we spent time designing and building some new equipment to use when testing subjects in the lab. First, we designed a new triggering mechanism for the experiment that needed to be accurate, computer-controlled, durable, relatively quiet, and self-contained. We decided to make the core of the design an archery trigger, and we extended the trigger with two metal plates and built a metal base to hold the trigger securely in place. To pull the trigger, we attached a solenoid and a relay to the computer.

Once the trigger was up and running, and connected to the necessary mechanical, electrical, and computer equipment, we moved on to the Biomechanics Research Laboratory of the University of Michigan in Ann Arbor.

The equipment we used consists of the triggering mechanism, which has several hook-ups to the computer, a jacket-like harness for the subject to wear, and a padded belt for around the subject's waist that attaches to a rope of adjustable length. The other end of the rope attaches to the trigger. The subject is also connected by two cables to a load cell (force measuring device) above them, and this is attached to a metal track in the ceiling. The track allows the cables and force measuring device to slide forward as the subject moves, and does not restrict forward range of motion. The cable system insures safety because if the subject starts to fall, the harness and cable will catch and support the weight, preventing the fall from occurring. Since the cable is also connected to a force measuring device above, we can determine how much of the subject's body weight is being supported by the harness and cable. If a subject places a certain force or percentage of their body weight on the cable support system, it is considered a "fail" by our standards, but the subject can never touch hands to the ground or acquire serious injuries. Also, to record the subject's position at any given time during the test, sixteen infra-red sensors are placed in specified locations on the body and provide data for the computer.

In our trials, the subject wears the harness and padded belt along with all of the infra-red sensors and is attached to our trigger by a rope. After all safety precautions are taken, the subject is told to lean forward into the padded belt around their waist with the rope supporting them. We then lean them forward until a certain percentage of their body weight is being supported by the rope, and then, at a random time, the computer sends a signal through the relay to the solenoid, the trigger is pulled, data collection begins, the subject is released, and we observe how the subject regains balance. We do this several times at specified angles.

In the first series of tests, we give the subject no guidance on how to regain their balance, but simply observe their natural patterns of recovery. In the second series of tests, we instruct the subjects to regain balance in only one rapid step, if possible, and observe how this is attempted. We also record the time it takes each subject to take this first step after being released, and we take strength measurements of the muscles around the subject's ankle, knee, and hip joints.

From the data we collected during our four weeks of testing at the University of Michigan, we found that:

1. young recover in one step more often and more successfully than do the old

2. males recover in one step more often than females do and can do so when leaned further forward

3. when given a choice, most elderly take more than one step to recover balance, while most young take only one. [Now, we are speculatindistance) - balance problems - flexibility factors - weight considerations (e.g. our elderly females were heavier on average than our young females) - psychological factors (background, how were raised, afraid of falling? not confident in ability to recover? feel insecure? unsafe?)

4. subjects that succeed in regaining balance from large leans in a single rapid step are likely to succeed in large leans when they can choose their own strategy of recovery.

5. the elderly have a significantly longer step time than the young, but there is no significant difference in step times between genders.

6. it does not necessarily require more strength to recover in only one step, but the strength must be able to be developed quickly. [So, weaker subjects could surpass stronger ones in recovery from a maximum lean if they moved more quickly.]

Throughout the coming year and possibly next summer, we will continue to analyze the data, test our hypotheses on why the results turned out the way they did, look for significant data correlations, and make comparisons to past studies. To acquire a more versatile visual model to simulate the experiment, we will animate the acts of walking and stepping using a Working Model computer program, and eventually quantify the biomechanics of recovering balance in a fall.

In addition to this study, a psychological study may be run on the subjects we tested to determine their attitudes and feelings about the test (i.e., if they were confident and motivated to do well, or if they were frightened so they did not give their best effort), and these results will be compared to the data we collected to see if there are any significant correlations.

Also, the data we collect is being used in conjunction with a study being done on how strength training affects the stepping capabilities of the elderly. In this study, half of the elderly subjects are involved in six weeks of strength training after our balance-recovery test, after which, they are re-tested in the same balance-recovery set-up. The other half of the elderly subjects do not complete strength training, but are also re-tested six weeks later. The data from the elderly not involved in the strength training is used as "control" data and is not expected to show any change from their first set of data. The re-test data from the elderly that are involved in strength training is then compared to the "control" data, and also to their own initial balance-recovery data. It is expected that the elderly that complete six weeks of strength training will more successfully recover balance and be able to recover when leaned farther forward than the elderly that do not complete strength training.