All Tied Up And No Place To Go!
Part Two of Three Parts

Prehospital Patient Restraint Application Techniques and Issues

Originally published on the EMS Web Site: in July, 1996; and updated on that site in August, 1998; THIS VERSION is even newer! (Last updated in June, 2001)
The 2002 Update coming in July or August, 2002!

By Charly D. Miller, Paramedic; EMS Author, Educator, & Consultant

Section II: Techniques for Physical Restraint of Patients:

Safety belts used only as "safety belts" do not need to be specifically documented because they are an "industry standard" safety precaution. If you DO NOT have them in place, however, and an accident results in additional patient injury (injury that would not have occurred had the safety belts been in use), you may be liable for a charge of NEGLIGENCE. When safety belts are used as "restraints," they should be documented as restraints.

PRONE POSITIONING IS CONTRAINDICATED FOR THE TRANSPORT OF ANY PATIENTS (with the exception of bizarre situations involving impaled objects or the like!): This will be explained in more detail later. Certainly, no patient should ever be restrained in a prone position.

* Author's note: "pram" is my term for the wheeled-stretcher (or bed) of the ambulance. The terms "cot" and "gurney" are synonymous with the term "pram."

To restrain any patient using any proven device or technique, only appropriate use of that device or technique will result in safe and effective restraint. For my technique, here are the basic rules:

RULE #1: Employ the pram "safety belts" as restraints before restraining a patient's limbs to the pram. This minimizes the amount of force required to safely apply limb restraints, minimizing the risk of patient or provider injury during the restraint process.

As our industry progresses, more and more "new contraptions" are developed to meet the needs of providers and patients. The use of a harness system of pram safety belts (straps over each shoulder, connecting with a center belt at the waist) is becoming more and more widespread. Indeed, it may eventually become an industry standard. This is because a harness system clearly represents a more effective method of patient protection in the event of an accident during transportation. Unfortunately, harness systems are not designed for use as patient restraints. If snugged tight enough to produce chest "restraint," harness systems can cause chest or abdominal injury, and may significantly impede a patient's respiratory effort, becoming a health-care-provider-induced life-threat! Thus, when using a harness system, an extra "chest restraint belt" must be available for when chest restraint is required. This extra belt can be kept buckled behind the adjustable torso-section of the pram, keeping it out of the way during transports not requiring restraint.

RULE #2: For safety belts to be used as safe and effective restraints, it is vitally important that they are attached to the pram in an area where they can easily be moved, easily slid along the pram's frame, enabling their positioning to be adjusted to each patient's particular body length. If their placement cannot be easily adjusted, their use as safe and effective restraints becomes compromised. Compromised restraints are always poor restraints - often even injurious restraints! Unfortunately, the newest (heavy-duty) pram styles - such as Stryker's "Rugged"(tm) or Ferno's "35-A+ Mobile Transporter"(tm) are designed in a manner that provides little-to-no room for adjustment of restraint-belt attachment, both above and below the patient's waist.** Providers who work with these prams will have to develop alternative methods of appropriately placed chest and lower body restraints. Perhaps they could use silver duct tape (running it around, and around the patient and the pram frame), or the like. At this writing, I haven't had opportunities to "play" with these prams, so I'm not sure of potential options.
** Since hearing my presentation, the Stryker company has made some frame modifications that improved the versatility of the Rugged's restraint-belt anchor sites. Although still not as versatile as standard prams, the Rugged is far better than Ferno's heavy-duty pram.

At any rate, to be safely and effectively used for restraints, safety belts must be multiply adjustable. Patients come in all different sizes, and restraint belts must be specifically placed upon a patient's body to be successfully effective, yet still safe.


FYI: Even "safety belts" used for routine transportation are ineffective if engaged below the patient's knees. Unless anchored above the patient's knees, a safety belt will fail to contain the patient during a rollover or violent accident. The patient's knees are allowed to flex, and his legs will simply slip and fall out of such a "safety belt." The only safe and effective lower limb "safety belt" or restraint belt is one that is anchored and engaged immediately above the knee. And, with an appropriately anchored and tightly engaged lower limb restraint belt, I rarely have to use ankle restraints.


MECHANICAL RESTRAINTS (Handcuffs, Ankle Shackles) STINK!
They pose a serious impediment to thorough examination and performance of treatment, and risk patient injury. Law enforcement officers (and inadequately restraint-trained care providers) may feel that they are occasionally required - but they are NOT. After an officer's initial use of mechanical restraint, a well trained EMS provider may slowly and surely coordinate the patient's transfer into soft-but-secure-and-safe "medical restraints" (restraints that enable the care provider to adequately evaluate and treat the patient, without interference from combative behavior or poorly positioned restraints).

If the patient's wrists are restrained in front, mechanical wrist restraints become a weapon! So, even for the brief period of time that mechanical wrist restraints "must" be used, the patient's hands must be restrained behind his back. Consequently, you cannot position a patient SUPINE when mechanical wrist restraints are used. Such restraints require prone positioning. PRONE POSITIONING WILL MAKE IT IMPOSSIBLE FOR YOU TO MANAGE THE PATIENT'S AIRWAY (should the need acutely arise), AND WILL SIGNIFICANTLY COMPROMISE THE PATIENT'S ABILITY TO BREATHE.

Since restraints are applied for "patient safety," restraint circulation impairment can never be "excused." Thankfully, it is entirely unlikely that restraint circulation impairment will actually harm a patient when transport times are short. Nevertheless, it is very poor form to arrive at the emergency department (ED) with your patient sporting bilaterally cyanotic hands and feet because of restraint circulation impairment! If patient transport time exceeds 15 minutes, restraints that become tight enough to impair distal circulation should be loosened as often as necessary - or a non-tightening form, such as a "full knot" or "locked" clove hitch, should subsequently be employed to replace the tightening restraint. Circulation checks should be documented at least every 15 minutes.(8)

An article reporting two case studies involving this occurrence was published in Annals of Emergency Medicine's May, 1995 issue.(13) "Hobble restraint" ("Hog Tied" restraint) traditionally consists of a person's wrists restrained together behind the back, his ankles tied together, his knees flexed, and then his restrained ankles tied to his restrained wrists. Earliest studies of hobble restraints causing death were limited to law enforcement transport. This Annals article, however, focused specifically on paramedic transport of hobbled - and forcefully-prone-restrained - patients, presenting two case studies of Restraint Asphyxia death occurring during paramedic transport. In each study the patient was forcefully restrained in a prone position, with wrists & ankles tightly tied together ("hobbled") behind the back. Each study patient went into cardiopulmonary arrest, but received rapid and aggressive ACLS intervention (initiated enroute to the ED). All prehospital and inhospital resuscitation efforts failed to revive EITHER of these patients.

Each study patient had non-lethal post-mortem toxicological evidence of drugs or alcohol. Each study patient's autopsy ruled-out all natural and unnatural causes of death. Each study patient's cause of death was determined to be Restraint-Related Positional Asphyxia. After reviewing all the information, the case studies' authors agreed that the prone "hobble" restraint used during paramedic transportation resulted in positional asphyxia, directly contributing to each patient's death.

SPECIAL NOTE: On the RESTRAINT ASPHYXIA NEWZ directory is a link that will take you to the full text of this Annals case study. Additionally, a link to my three-part article about "Restraint Asphyxia - Silent Killer" is provided on that directory!

POSITIONAL ASPHYXIA occurs when a person's body is TRAPPED in a position that prevents their ability to breathe - either due to an essential upper or lower airway obstruction, OR due to prevention of chest/abdominal expansion and/or relaxation. It "is important to consider the contribution of the chest wall and abdomen to the process of ventilation."(13) The act of breathing requires the ability to inspire and expire. INSPIRATION requires the development of a negative intrathoracic pressure, achieved by rib cage expansion and/or downward contraction of the diaphragm. (Since the diaphragm is the largest muscle of respiration, the diaphragm is the most important respiratory-muscle-function to protect!) EXPIRATION requires the development of a positive intrathoracic pressure, achieved by chest wall and diaphragmatic relaxation. In positional asphyxia deaths unrelated to restraints, unconsciousness due to acute alcohol intoxication is the most frequent explanation of the victim's inability to escape from the asphyxiating position.(4)

RESTRAINT ASPHYXIA (also called, "Restraint-Related Positional Asphyxia") occurs when RESTRAINT is responsible for an individual being TRAPPED in a position that interferes with their ability to breathe. Restraint Asphyxia is most often caused by forceful-PRONE-restraint. However, Restraint Asphyxia has incorrectly been most often considered to be associated with hobble restraint. If hobble restraint is achieved without the use of forceful-prone-restraint to apply the hobble, and the hobble restrained individual is kept on his SIDE (leaving his diaphragm free to "breathe"), Restraint Asphyxia will not occur!

Forceful-prone-restraint "physically interferes with diaphragmatic motion by restricting downward displacement of the abdominal contents"(13) - significantly restricting inspiration. FORCEFUL PRONE POSITIONING: With restraint belts tightened across the posterior chest and mid-thigh, prone positioning is equally as lethal a situation as the use of hobble restraints. Indeed, Restraint Asphyxia deaths have occurred solely from prone positioning, even with short transport times.(14) When a patient is restrained prone, his abdomen is impeded from assisting respirations. When such abdominal impedance is compounded by "safety belts" tightened across the posterior chest and thighs, positional asphyxia can easily occur.

One group of researchers write, "Considering this physiologic information, methods to avoid possible (positional) asphyxia should include placing a restrained individual in the lateral or supine position rather than in the prone position. When hobble-type techniques are used, there should be slack in the restraints to allow for ventilatory motion of the chest wall muscles."(13) Unfortunately, these researchers are clueless when it comes to providing safe and effective restraint. "Slack" restraints, or laterally-positioned-patient restraints, are completely ineffective restraints for medical purposes! Slack restraints allow the patient freedom to interfere with examination, evaluation, and treatment. They also allow an increased risk of patient injury, increased risk of provider injury, and an increased risk of successful litigation.

ROLLER GAUZE!!! A thick-yet-strong version of roller gauze (such as 6-ply versions of KERLIX(tm) or KLING(tm)) will effectively and safely restrain any sort of patient, but ONLY IF IT IS APPLIED AND ANCHORED CORRECTLY.

FIRST, do not cut the roll into two pieces! "Saving money" is not as important as providing effective and safe restraint. When you cut the roll in half (in an effort to use one roll for two restraints), you diminish the strength of soft roller gauze restraint by 50 percent. Additionally, when used only in single-thickness, roller gauze compacts and becomes abrasive. Although single-thickness might be strong enough to secure small pediatric or geriatric patients, the abrasiveness will result in unsightly and painful skin abrasions. Do you really want to deliver bloody-wristed patients to the ED?

Use the entire 6-ply roll, DOUBLED, so that you have a 12-ply, non-abrasive restraint.

SECONDLY, the roller gauze restraint (or any type of restraint) must be appropriately anchored before it will be effective restraint. We'll discuss appropriate anchoring later.

How restraints are anchored is vitally important to EFFECTIVE and SAFE patient restraint. There is a significant distinction between whether a limb is "tied up," or whether a limb is effectively, "medically restrained." A limb that is merely "tied up" is one that still has - even a "minor" - amount of mobility. A limb that is "tied up" to a bar is a limb that can move - sliding up or down along the bar.

When "tied up" with even a minor amount of slack between the knot at the patient's limb and the knot at the site of anchor, the limb:

A "medically restrained" limb is one that is prevented from any type of movement whatsoever - one that is prevented from interfering (in any manner) with medical evaluation and treatment. It is mandatory that there be no "play" between the restrained limb and the point of anchor. If even an inch or two are "loose," the patient can move the limb, can continue to interfere with care procedures, and may additionally pose a risk of self injury (an injury that may be considered the health-care-provider's "fault"). Additionally, if the IV limb is not adequately immobilized by effectively anchored restraint, limb movement will likely result in multiple IV attempts (needle sticks), further aggravating the patient (and the provider).

A "T-joint" is any place where two bars connect and form a "T." By anchoring the restraint directly to a T-junction, the restraint anchor cannot slide back or forth, up or down, in or out. When a fully-extended limb is immediately secured to a T-junction, it cannot move. If a T-junction is slightly within the flexion range of the patient's limb, the patient can still move the limb. Thus, the next T-junction, distal to the patient's wrist, should be employed for restraint anchor. When a T-junction slightly beyond the full extension of the patient's limb is used, simply traction the restrained limb and anchor it at the distal T-junction. This will prevent the patient from moving that limb up or down, back or forth, in or out. The only truly and effectively, medically restrained limb is one that cannot move - one that cannot interfere with evaluation or treatment procedures.

The thickness of leather restraint anchor straps make effective pram anchoring entirely unlikely. It is almost impossible to securely snug a leather-restrained limb to a single point. Mechanical restraints (handcuffs or ankle shackles) are also almost impossible to securely anchor in a medically-restrained manner. Either of these types of restraint will allow the patient plenty of "play" with which to harm her/himself or to interfere with care.

SPECIAL NOTE: DO NOT ANCHOR SOFT ROLLER GAUZE RESTRAINTS TO SHARP METAL. Obviously, a patient could saw through the soft roller gauze if given enough time, enough determination, and a sharp edge.

LONG BACK BOARDS have plenty of anchor sites that are smooth and can provide secure (usually distal to the full limb extension) restraint anchor sites. SCOOP STRETCHERS are okay, but beware of sharp edges. The pram usually has a large variety of smooth T-joint sites to effectively anchor restraints to. Yet, if a long back board is used, the patient will not require freedom from the prehospital restraints before fully transferring care to the ED.

Picture this scenario: Your patient requires restraint but is in a location that prevents safe access with your pram or a long back board. Perhaps the terrain is filled with debris or can only be reached by traversing an incredibly uneven surface. Perhaps the physical structure of the scene prevents your access with a pram or long back board. Whatever the reason, you must safely bring your patient out to a location that allows utilization of your pram or long back board. Carrying a manually-restrained patient across dangerous terrain (even when you have one-person-per-limb) is risky, potentially harmful to either the patient or the care providers. Thankfully, the patient's body can provide temporary restraint anchor sites.

Now the patient's arms have been secured to each corresponding leg and have become "handles" for safe, controlled, and relatively painless transportation. If the patient's ankles were tied together, you have a third "handle" to use for carrying the patient safely.

This restraint technique of a combative or resistive patient is only designed to be temporary. Neither arm is completely prevented from movement, thus IV access is not controlled with this technique. Additionally, circulatory impairment may occur if this restraint method remains in use for a prolonged period of time. Once safe transit to a pram or backboard has been accomplished, full and effective soft safety bracelet restraint techniques should be employed.

For cardiac arrest patients, however, the crotch restraint method may be employed (with less snugging) and left in place. This keeps the patient's arms extended (allowing IVs to continue running) and out of the way, and is especially handy for transfer of the patient to the ED's bed.

The decision to use restraints should be entirely based upon the patient's needs. If restraint is needed, it should be employed without hesitation. But, first you must INFORM the patient as to the need for, and manner of restraint that will be used.

Avoid uttering negative- or punitive-sounding terms or phrases such as "restraint," "tie you up," "tie you down," and the like. These terms and phrases only serve to threaten, irritate, and agitate people. We certainly don't need patients to become more irritated or agitated than they already are!

Instead, use the terms, "safe," "secure," and "comfortable" as often as possible when referring to the need for and application of restraints. Feel free to use my phrase, "SOFT SAFETY BRACELETS," when referring to soft roller gauze restraints!

And so on....

Lastly, the topic of restraint should never be used as a threat. If a patient requires involuntary treatment, "threatening" them with restraint in the false belief that you can "bargain" them into cooperating (and avoid restraint) is not only inappropriate, it's completely ineffective. Patients who are incompetent to refuse treatment cannot be anticipated to "bargain" competently. Bargaining only leads to arguments and aggravation.

If you're uncomfortable with any patient, even one that hasn't been actively "combative," the patient requires restraint. If you reasonably anticipate a patient becoming combative (secondary to being stuck with a needle, for example), you have the right and the duty to provide the patient, and everyone else involved, with the "security" of patient restraint. Remember that a threat to "third parties" (verbal or physical) constitutes a "state" reason for restraint.(2) Since you can also be considered a "third party," you also have a right to safety.

Simply advise the patient that you're uncomfortable with her/him being without "soft safety bracelets," that you're fearful she/he may hurt themselves or others - even though she/he may not mean to do so.

"Oh! I'll cooperate! You don't need to tie me up!" That's bargaining, and cannot be accepted or relied upon.

A cooperative patient will understand your concern and cooperate with soft safety bracelets. Additionally, a cooperative patient frequently needs only minimal restraint: chest, lower limb, and unilateral wrist restraint. If the patient subsequently demonstrates the need for additional restraint (the other wrist), it can be safely applied with a minimum of struggle.

Since A COOPERATIVE PATIENT WILL COOPERATE WITH RESTRAINTS - a patient who doesn't want to cooperate with restraints must be considered to be "UNCOOPERATIVE!" If the patient becomes combative secondary to your desire to secure her/him, then YOU WERE RIGHT TO FEEL UNCOMFORTABLE WITH THIS PATIENT BEING LOOSE! THIS IS NOT A "COOPERATIVE" PATIENT!

If at ANY time a patient requires physical restraint, the patient should remain restrained - in some manner - until delivered to the ED! Do NOT allow a patient to bargain with you, or talk you into release of restraints because; "I've calmed down now," or "Hey, I'll cooperate now, you don't need to keep me tied up," "Just let me loose and I'll be good!" Remember, A COOPERATIVE PATIENT WILL COOPERATE WITH RESTRAINTS.

Even a patient who was combative secondary to HYPOGLYCEMIA or a POSTICTAL seizure patient, should remain at least minimally restrained. After the patient receives dextrose - or "recovers" from the postictal phase - and becomes "AAOX3," do not feel as though you must remove all of the restraints. Hypoglycemic patients may rapidly utilize the single dose of dextrose and succumb to hypoglycemia again (becoming combative again). Seizure patients may experience another seizure, followed by another combative postictal phase. And, remember, A COOPERATIVE PATIENT WILL COOPERATE WITH RESTRAINTS.

As YOU feel comfortable, remove the ankle restraints (leaving the lower limb safety belt in place). As YOU feel comfortable, remove the wrist restraint from the IV arm. And if that is as much as YOU are comfortable with releasing, then that's fine! ONCE RESTRAINED, ALWAYS RESTRAINED - at least until someone else is responsible for patient and third party safety; then let them make the decision to remove all safety measures.

It is vitally important, and enormously therapeutic, for all patients to be given as many "options" as possible. Giving patients choices and options helps them to begin regaining a sense of self control, a sense of "normalcy." Emergency situations inflict an extremely frightening and threatening sense of "loss of control" upon the patients who experience them. Think about it. If our patients had retained control over themselves, their bodies, their situations, would they have allowed the emergency to occur?! Of course not! By giving the patient options, assisting the patient to start regaining control over his person, you help the patient to feel better - even help the patient to respond more positively and more rapidly to your care provision. This recognition of the need to "allow options," to restore a patient's sense of self control, however, does NOT CHANGE THE FACT that restraint is GOING TO OCCUR! Once you recognize the need for restraint, restraint must be employed. Restraint application is not a patient "option."

But, as soon as possible, begin to allow patients as much control as YOU are comfortable with allowing.

If you are uncomfortable giving your patient such restraint limb or position options, find other options that you can offer. The "secret" to offering patients options is this; as long as YOU DON'T CARE what the patient chooses, offer them the option. And so on....

The fact that restraint will occur does not change. But, to improve the patient's condition and encourage cooperation, the patient should be given as many opportunities as possible to exercise some sense of control over the situation.

All of these combinations occur after the chest and lower limb safety belts have been engaged as restraints! Additionally, all of these combinations assume that the patient has mobility of all four limbs.

SINGLE LIMB (WRIST) RESTRAINT: is indicated for use as

Although disoriented patients aren't always "combative," the rarely ever appreciate being stuck with a needle. Thus, non-combative disoriented patients require restraint - at least one wrist. When only one wrist requires restraint, secure the dominant wrist above the patient's head (at the central T-junction at the top of the pram). Start your IV in the unrestrained arm. Some providers make the mistake of securing the IV arm's wrist only. It's easy for the patient to use the "loose" arm and pull out the IV. A lot easier than having to sneak the IV arm way up above his head to the restrained wrist! Also, by restraining the dominant wrist up and out of the patient's view (above and behind the head) the patient will be less preoccupied by the restraint ("out of sight, out of mind").

BILATERAL WRIST RESTRAINT: is indicated for use as

When both wrists are restrained, anchor one above the head (at the central T-junction at the top of the pram), and one below the waist at the patient's side. This "splitting of the arms" interferes with the patient's ability to use his strong abdominal muscles in an effort to defeat the chest restraint. The patient's dominant wrist should be placed above the head, with the IV arm's wrist securely tractioned to the thigh area of the pram/board/scoop. When the chest belt and both wrists are secured appropriately, you can immobilize the entire IV arm by simply cupping the patient's elbow with the palm of your hand and hyperextending the patient's arm (pulling it toward you). Remember to use only enough force to immobilize the patient's arm - exerting excessive force becomes painful. Only if a patient is "divinely inspired" to be combative, will you need your partner to maintain shoulder control during IV access.


RESTRAIN ONLY THE NON-DEFICIT-SIDE LIMBS! First, it's unnecessary to restrain the side that can't move. Secondly, having the deficit side loose allows for notice of any return of mobility. Having the deficit side loose may also provide some warning of the onset of a seizure: CVA patients who seize frequently begin with a focal-motor seizure of the deficit side. Thus, you can quickly cut loose the restrained arms before the patient has a full-body seizure.


What would be the point? Ankle restraints are the last, "most restrictive," part of "4-POINT" restraints (chest and lower limb safety belt restraint, both arms, both ankles = body and all four limbs). After securing the chest and lower safety belts, and after both wrists are restrained, if the patient's efforts to defeat the lower safety belt restraint appear determined, then 4-point restraint is required. Secure the ankles TOGETHER FIRST. Then anchor the ankles to the end of the pram/board/scoop at a T-junction.

Go To "All Tied Up and No Place to Go" Part 3 NOW
(Another link to Part 3 is provided at the end of this page, if you want to wait.)

REFERENCES (for all parts):

  1. Lavoie FW: Consent, Involuntary Treatment, and the Use of Force in an Urban Emergency Department. Ann Emerg Med January 1992;21:25-32.
  2. Rice MM; Moore GP: Management of the Violent Patient, Therapeutic and Legal Considerations. Emerg Med Clin North Am February 1991;9(1):13-30.
  3. Appelbaum PS; Grisso T: Assessing Patients' Capacities to Consent to Treatment. N Engl J Med 1988;319:1635-8.
  4. West's Colorado Revised Statutes Annotated, 1989;13-22-101 to 103.
  5. Webster's Encyclopedic Unabridged Dictionary of the English Language, 1989.
  6. Shanaberger CJ: Escaping the Charge of False Imprisonment. J Emerg Med Serv (JEMS) 1990 Mar;15(3):58-61.
  7. Nixon RG: Restraints and Prehospital Care. Emerg Med Serv 1986 Jan/Feb;15:24,26,46.
  8. Rund DA; Keller MD: To Restrain or Not to Restrain. Emerg Med Serv 1986 Jan/Feb;15:24,46-9.
  9. Northrop CE: A Question of Restraints. Nursing 1987 Feb;17(2):41.
  10. Leisner K: Managing the Pre-Violent Patient. Emerg Med Serv 1989 Aug;18(7):18-20, 23, 26, 28-9.
  11. Simoneau JK: Medicolegal Aspects of Restraint. Emerg Med Serv 1989 Aug;18(7):24.
  12. Richmond PW; Fligelstone LJ; Lewis E: Injuries Caused by Handcuffs. BMJ 1988 Jul 9;297(6641):111-2.
  13. Stratton SJ, Rogers C, Green K: Sudden Death in Individuals in Hobble Restraints During Paramedic Transport. Ann Emerg Med May 1995; 25:5, pages 710-12.
  14. Ambulance transport death results in questioning of techniques. EMS Professionals, July-August 1997;6-8.
  15. Stewart CE: Of Trolls and Control. Emerg Med Serv 1986 Jan/Feb;15:25,52-3.
  16. Bell MD, Rao VJ, Wetli CV, Rodriquez RN: Positional asphyxia in adults. Am J Forensic Med Pathol, 1992;13(2):101-107

** To obtain the "JEMS EMS Pocket Guide" for rapid on-scene reference needs, visit your local medical book store or call 1-800-240-0703 (JEMS Bookstore) and order it (suggested retail price of $16.95) - OR: CLICK HERE to GO to for the JEMS POCKET GUIDE

To personally Email ME (Charly) with your feedback about this article,
use my Email link at the bottom of this page.


Return to the Restraint Asphyxia Newz Directory


Email Charly at:
Those are hyphens/dashes between the "c" and "d" and "miller"

This counter was started on April 13th, 2001.