Access to the PPM Journal and newsletters is FREE for clinicians.
11 Articles in Volume 14, Issue #1
The WHO Pain Ladder: Do We Need Another Step?
History of Pain: The Psychosocial Assessment of Pain
Lyme Disease: A Short Primer for Pain Practitioners
Opioid Prescribing Part 1: A Practical Guide to Appropriate Documentation
Pain, Impairment, Whiplash, and the New AMA Guides: What Clinicians Need to Know
The 5 Coping Skills Every Chronic Pain Patient Needs
Demystifying Benzodiazepine Urine Drug Screen Results
Practical Pain Management: The Nation’s Premier Teaching Journal for Pain Practitioners
PPM’s Editorial Board Weighs In on WHO Ladder
Are patients taking acetaminophen (Tylenol) at risk for developing serious skin conditions?
What are some home exercises and tips to help patients manage rotator cuff injuries and pain?

Pain, Impairment, Whiplash, and the New AMA Guides: What Clinicians Need to Know

Close to 3 million Americans suffer whiplash each year. Whiplash now accounts for 45% of chronic neck pain in adults. The author discusses how new car design prevents fatal crashes but increases whiplash injuries.

Every year, approximately 3 million people in the United States suffer from whiplash,1 resulting in an estimated $43 billion in comprehensive costs.2 Over the past few decades, these injuries have been on the rise worldwide.3 Studies have reported rises in whiplash injuries from a low of 7.2% in 1982 to 30% by 1997.4

Seat belt and shoulder harness use, while providing life-saving protection in most crashes, can increase the risk for whiplash injury.5-7 The substantial rise in seatbelt use in the US since the introduction of primary- and secondary-use laws has contributed to the increased incidence of whiplash injury. Increased stiffness of seat backs also has been a contributing factor.8-10 A large population-based European study found that the increase in whiplash from 1989 through 1995 was associated with stable seatbelt usage.11 Krafft et al, showed that the relative risk of being injured in a crash was related to model year—that is, the risk of being injured in a 1990s era model was 2.7 times that of being injured in a 1980s model.12,13

As the incidence of whiplash has increased, so have the rates of disability: whiplash may account for as much as 45% of chronic neck pain in adults.1 A significant portion of people who suffer from whiplash injuries seek some form of compensation. For the past half century, various editions of the American Medical Association’s Guides to the Evaluation of Permanent Impairment (Guides), have served as a tool to help physicians evaluate injuries and rate the degree of permanent injury. Although most jurisdictions in the US do not currently use the Guides for matters of personal injury impairment rating and litigation, the latest version of the Guides, the 6th Edition,14 features a diagnosis-based impairment (DBI) assessment that mentions whiplash by name. This may portent a push by insurers to require whiplash injury claims to be rated using the Guides.

This review will discuss substantive changes to the new Guides and what clinicians need to know about these changes for the diagnosis of whiplash.

Why the Increased Incidence of Whiplash?

A significant factor in the increase in whiplash claims has been the automotive industry’s attempt to manage the increasing crash energy recommended by the National Highway Traffic Safety Administration’s (NHTSA) New Car Assessment Program (NCAP) and the Insurance Institute for Highway Safety’s (IIHS) 40% offset deformable barrier crash tests. The NCAP and IIHS tests are non-obligatory tests, designed to induce automakers to enhance efforts in crash worthiness using a “name-and-shame” approach. The results of both tests are then promulgated to the public.

While the current government-mandated compliance test (FMVSS 208) requires a frontal crash into a rigid barrier at 30 mph, the NCAP is conducted at a more rigorous 35 mph. The IIHS tests are 40% offset frontal tests into a deformable barrier on the driver’s side to simulate a typical head-on collision in which only 40% of the vehicle’s frontal area makes contact. These are run at 40 mph. To perform well on these crash tests, manufacturers have increased the stiffness of passenger cars by using stronger alloys of steel and aluminum, and have made other structural modifications. While this has led to a reduction in injuries occurring at higher speeds (30-40 mph), it has increased the stiffness of cars in lower velocity crashes. Since model year 1982 (three years after the inception of NCAP), vehicle stiffness has increased by approximately 34%.15

The increase in whiplash injury risk caused by the increased stiffness of cars and the increased use of seat belts led the IIHS and a number of other international insurance consortia to form the International Insurance Whiplash Protection Group (IIWPG).16 They developed a special seat/head restraint testing program using specially developed rear-impact crash test dummies and sophisticated sled systems. Test results are available at www.iihs.org. Rear-impact dummies have been validated in human subject biofidelity crash testing.17,18 The validation of the crash test dummies has meant that there now is a standardized dummy that reliably represents a human response in a rear impact collision, which is the most common type of collision. That allows the industry to develop more effective seat- and head-restraint designs, and to evaluate the effect of seatbelt load limiters and pretensioners. Airbags do not have any effect on whiplash injuries caused by rear impact collisions; they only deploy in frontal- or side-impact (or roll-over) collisions.

The AMA 6th Edition: How It Differs

In the development of the latest edition of the Guides, the editors employed a modified Delphi Panel. In most cases, the panel relied on published works, grading those works based on a standard hierarchy of evidence, with meta-analysis of randomized controlled trials (RCT) at the top, followed by RCTs, non-randomized interventional studies, observational studies, and so on. The authors noted the failings of earlier editions and asserted that this edition is a “paradigm shift” that will rectify earlier shortcomings.

All of the impairment areas included in the Guides share a generic template comprised of 5 classes of impairment (0-4). The percentage of impairment initially is based on these classes and varies with the diagnosis. In this sense, the system resembles the old 5th Edition diagnosis-related estimate (DRE) classes. However, unlike the old system, the DRE has been replaced with the DBI estimate. The DBI includes severity grades A through E, with A being the least severe and E being the most severe. The severity grade is determined by subtracting the class number from a number related to grade modifiers for functional history, physical examination, and clinical studies.

In the most current version of the Guides, there is now a DBI for whiplash. When a DBI is not available or appropriate, clinicians can use a pain-related impairment (PRI). The authors of the PRI chapter willingly accepted and discussed the controversy of this practice. Some of the controversy lies in the difficulty in objectifying and quantifying pain. It also was argued that empirical data was lacking on the role of pain in impairment. And while some pain experts believe that pain should be given a more prominent role in impairment rating, others argue that it should be disregarded completely.

Some experts discourage PRI, while others believe it is a necessary adaptation. In any event, the maximum rating for pain is capped at 3% of whole person impairment (WPI). One simply doesn’t get much credit for pain without a more objective DBI. Readers interested in finding out more about this are referred to Chapter 17 of the Guides, entitled The Spine and Pelvis.14

 Applying the New Guides

We will only discuss the cervical spine here; however, the other spinal levels are similar with respect to the mechanics of the rating. The first thing to do in the cervical impairment rating is to choose the appropriate DBI. For the cervical spine there are 7 DBIs.

Once you have settled on a DBI, then you select the appropriate class. For example, in the case of chronic neck pain due to whiplash or strain/sprain injury, only class 0 and class 1 are available. In general, class 1 allows for a range of WPI from 1% to 8%, but in the case of chronic neck pain, no more than 3% can be assigned. That is also the cap based on the PRI, so it is at least logical, if questionable on clinical grounds and real-world experience.

The next order of business is to consider the modifiers or the “adjustment grid.” There are 3 for the spine—functional capacity, physical examination, and clinical studies. Each of these has 5 modifier levels (0-4), which correspond to: no problem (0), mild problem (1), moderate problem (2), severe problem (3), and very severe problem (4). The Guides provide examples of each of these.14 For example, a functional capacity level of 2 implies pain and symptoms with normal activity. In the case of functional capacity, a clinician can also use the pain disability questionnaire (PDQ), which is provided in the Guides as an appendix. It also is permissible to use an “alternative validated assessment,” although none were specifically mentioned that relate to the cervical spine. An alternative assessment that probably would be acceptable is the neck disability index (NDI) since, as we will discuss later, it has become the de facto instrument in the research community. Examples of physical examination modifiers include the straight leg raising test, sensory changes, reflexes, etc. Examples of clinical studies modifiers include needle electromyography or imaging studies. However, if the imaging study was used to place the person in the DBI (eg, MRI to diagnose disc lesion), it cannot be used again as a modifier.

Finally, the number corresponding to the DBI class is subtracted from each modifier number, and the 3 values obtained in this way for the 3 modifier classes are summed. If the sum is equal to 0, then there is no net movement within the class. If the sum is 1, the severity within the class is increased one increment. If the sum is negative, the severity is decreased. Take Class 2 in the cervical spine as an example. It carries a potential WPI of 9% to 14% across all 7 of the DBIs (with the exception of chronic neck pain for which only class 0 and class 1 are possible). This range, 9% to 14%, is represented by the letters A-E. One starts in the middle of the range, C, which, in the case of Class 2, corresponds to 11% (9, 10, 11, 12, 14). Mathematically astute persons might notice that the only way 11 could be right in the middle would be if one were to omit the 13. In any case, 11 (or C) is the default value. If the sum of the modifiers is 2, then you move 2 places to the right and go from a C severity rating to an E. No matter what the modifier number is, you never migrate out of the original class. These modifiers can only reduce the severity to as low as A or increase it to as high as E.

Important New Feature

An important feature of the new Guides is the assessment of AOMSI (Alteration of Motor Segment Integrity). In the cervical spine, one indication of AOMSI is a translation of greater than 20% of the anterior-to-posterior (AP) diameter of the body of the vertebra above, measured on either flexion or extension radiographs (Figure 1).

In the older 5th edition Guides DRE IV category, which could be assigned if there was 3.5 mm or more of horizontal intervertebral translation of a vertebrae, critics questioned precisely how to measure this translation. Some have assumed the posterior and anterior translation can be summed, but this practice will often overrate the severity of the lesion and misinterpret the research that formed the basis for this theory. White et al clarified the question and found that adding the anterior translation to the posterior translation was incorrect; they recommended the use of one or the other, whichever is greater than 20.19

At first blush, it appears that the authors of the Guides tried to settle this controversy in the 6th edition. However, under a section entitled Cervical Spine AOMSI, they describe the 20% A/B ratio method just mentioned. However, they note that an AP translation >2.5 mm for the thoracic spine, >4.5 mm for the lumbar spine, and >3.5 mm for cervical spine all indicate segmental instability or AOMSI.

Thus, some ambiguity remains after all. One thing that is clearly different in the 6th edition is that this 3.5 mm or greater translation now merits only a class 2 rating (4%-8%) if there is a resolved radiculopathy at that level, or a class 3 rating (9%-14%) if there is an ongoing radiculopathy at that level. The value of the finding has been significantly downgraded.

The new Guides include the 11-degree angulation rule from the 5th edition—also given to us by White et al—as an indicator of instability.14,19 Alternatively, a loss or near-loss of motion due to developmental fusion, successful or unsuccessful surgical fusion intervention, or preserved motion with disc arthroplasty, also qualifies as AOMSI.

The authors state that AOMSI is to be measured only by “plain film radiographs.” This restriction may have been leveled at users of videofluoroscopy (VF) or upright MRI bending studies. Of course, VF is radiography; thus, as long as one can capture high quality coplanar still images at the extremes of motion, the other differences between plain film radiography and C-arm VF are technically irrelevant for this application because the 20% A/B measurement is a simple ratio and is thus unaffected by differential geometric magnification. Similarly, the angulation method would be valid using VF.

The Banishment of Range Of Motion

One of the biggest changes in the 6th edition of the Guides is the elimination of range of motion (ROM) assessment. The rationale for this change was that ROM is not a reliable indicator of pathology or functional status. This may indeed be true with the broadest interpretation of neck disorders. For example, in cases of 3-level spinal fusion with good outcome, there may be relatively little or no impairment, even though cervical spine ROM may be measurably diminished. In cases of cervical spine trauma with ongoing disability, ROM may be normal or even greater than normal. In the case of whiplash injury, however, a growing body of evidence suggests that ROM is an important prognostic factor and that it correlates well with ongoing impairment.

Kasch et al reported that whiplash patients with a total ROM 2 standard deviations (SD) less than the mean of the sum of ranges of flexion/extension, lateral flexion, and rotation (adding to approximately 360 degrees) were at higher risk for developing long-term symptoms.20 The authors have more recently developed a predictive risk model for long-term whiplash, which includes initial marked limited cervical ROM, female sex, a high number of non-painful complaints (dizziness, cognitive features, blurred vision, nausea, etc.), and high self-reported pain on a visual analog scale (VAS).21,22 As defined by Kasch’s group, high-risk patients had a VAS >4, were female, had a greater number of non-painful complaints, and a total ROM of under 240 degrees.

In another study, reduced ROM allowed differentiation between non-symptomatic and long-term whiplash victims, with a sensitivity of 86.2% and a specificity of 95.3%.23 Using computerized kinematic analysis of ROM, a significantly reduced ROM was reported for all movements in whiplash subjects compared with an age-matched population.24 After whiplash injury, cervical ROM had predictive value for outcome and correlates with long-term disability. Thus, its measurement is an important part of the initial work-up and impairment rating, despite its absence from the 6th edition of the AMA Guides.


The NDI, which is the cervical version of the Revised Oswestry Disability Questionnaire, was originally validated in a study of whiplash patients.25 It has a high degree of internal consistency, reliability, and responsiveness. It is considered the most highly validated outcome assessment instrument in the evaluation of whiplash, and it is the gold standard and most widely used instrument for neck pain used in whiplash research.26-38 In patients with chronic neck pain, 2-digit disability scores are common. In contrast, the methodologies recommended in the 6th edition of the AMA Guides for the assessment of impairment of whiplash have not been subjected to validation. They do not allow uncomplicated whiplash injuries to exceed a single digit (3%) WPI level.

The authors of the Guides openly acknowledge that previous consensus-based editions were flawed, and that the new methodology, “must await further empirical testing.” This has important medicolegal relevance because in 1993, a critical ruling was handed down by the Supreme Court of the United States in the case of Daubert v Merrell Dow.39 This ruling redefined the landscape of medicolegal testimony of experts. As a result of this watershed case, the Supreme Court effectively made trial judges the gatekeepers of scientific expert testimony on the basis on 4 touchstone criteria:

  • Whether the theory or technique can be, and has been, tested.
  • Whether the theory or technique has been subjected to peer review.
  • The known or potential rate of error of the method used.
  • The degree of the methods or conclusion’s acceptance within the relevant scientific community.

The Daubert ruling applies both to assessment tools and medical testimony, and experts are frequently challenged in FRE 402 hearings prior to being allowed to testify. Thus, the best evidence-based recommendations in the evaluation of whiplash would be to include the NDI with the comprehensive history and physical examination, and to carefully record cervical ROM in all 3 major planes. Radiographic motion assessment of the cervical spine for motor segment integrity also is fundamental in the investigation of chronic whiplash injury. I recommend measurement of ROM using dual inclinometers, preferably electrogoniometers for greatest precision.

It is noteworthy that the method recommended by the AMA Guides for detecting malingering is significantly less sensitive than the method of applying coefficient of variation (CV) statistics,40 and the software supplied with the more sophisticated electrogoniometers incorporate this analysis during testing.


While the terms impairment and disability are not synonymous, they are related, and several clinical outcome studies of whiplash patients have reported the proportions of disability among their study cohorts. In most cases, “disability” was not operationally defined by the authors, but we can assume it referred to some form of permanent work or activity modification, because it was reported as a late-stage percentage of cohorts separately from the proportions who had chronic pain; ie, most with chronic pain were not “disabled.” In this context, disability has been reported from 2% to 24% of the whiplash-injured populations studied.41,42 These figures do not relate to the percentage of WHI as assigned during impairment ratings, but, coupled with the common 2-digit NDI scores found within the chronic whiplash population, they attest to the fact that the current Guides may underestimate a large fraction of chronic whiplash cases.

The Guides are new and improved, although characterizing them as a “paradigm shift” is perhaps premature. From a practical standpoint, I suspect that the Guides will not be particularly sensitive in the assessment of whiplash impairment. The notion of WPI is useful within the context of the workers’ compensation system in which the key players—workers’ compensation judges, lawyers, doctors, and disability raters—all understand the actual meaning and impact of these numbers. For the public at large, however, a 3% WPI is likely to bias towards a defense jury verdict, and it is unlikely that the concept of WPI can be effectively communicated to a lay jury in the time typically available. Using it may thus unfairly marginalize legitimate claims.


Last updated on: April 8, 2015
close X