Abbreviated Injury Scale
The Abbreviated Injury Scale ( AIS ) or simplified injury scale was introduced at the end of the 1960s as a rating scale for the lethality of individual injuries . The driving force behind the development was accident research for motor vehicles in the USA . The description of injury patterns that was customary up to then led to inconsistent results when assessing injuries for car occupants: In the same mechanical-technical accident constellation, the recorded occupant injuries diverged greatly.
In AIS, the severity of an injury is classified with values from 0 (no injury) or 1 (slight injury) to 6 (fatal injury).
For the AIS coding developed by the Association for the Advancement of Automotive Medicine , injuries to a person are broken down into individual injuries that are included in the AIS catalog. For each injury a classified probability of survival is given there. The classification should only describe the actual injury, it is independent of the type of treatment, the quality of treatment or the duration of treatment. It has been shown here that the probability of survival can be used as a very valid proxy variable for the severity of an injury. The sensible merging of individual injuries into injury patterns or the aggregation of individual injuries in general, on the other hand, is a problem that has not yet been finally solved.
Nowadays, injury coding according to AIS is used in addition to its classic application in automotive accident research to describe the severity of injury in trauma . The modern polytrauma assessments are based on an individual injury assessment by the AIS. Outside of accident research, the AIS is usually not coded directly, but rather automatically collected using so-called "diagnosis lists" in which one or more AIS codes are assigned to a medical-clinical diagnosis.
history
With the rise of automobility in the United States in the 1950s and 1960s, the number of road users injured and died in road accidents rose dramatically. Even in the early years of automotive safety research, the main goal was to reduce the number of people killed in traffic. It was tacitly assumed that measures that led to a reduction in the number of fatalities would also lead to a shift from more serious injuries to lighter injuries. In the mid-1960s, the first so-called "Scaling Committees" were formed, which wanted to map the injuries in more detail. The template here was the technical documentation of the vehicle damage, which was based on the first, highly simplified classifications in front, side and rear collisions. The component orientation was later expanded by a classification according to the direction of impact and the energy converted in the collision, in the form of the amount of speed change .
By standardizing the technical accident surveys, the accident scenarios could be classified and categorized very well. When the injuries were recorded, however, it quickly became apparent that no two injuries to one person and no two injuries to two people are identical. However, when implementing the idea that injuries, i.e. injury severity and type of injury, could also be grouped together, there was a wide spread: The problem existed e.g. B. in making the head injury comparable to that of the chest in a frontal collision. This harmonization of injuries was achieved by evaluating each individual injury with a mortality risk, i.e. the risk of dying from this injury, ultimately leading to the "Abbreviated Injury Score" (AIS), ie a "reduced injury assessment". Initially, the mortality risk classification was based heavily on the preparatory work by Hugh DeHaven on the injuries suffered by military pilots during both world wars.
General
Injury classes
The basic idea of the AIS code is to summarize injuries with approximately the same mortality. From the numerous classes of the first systems, the system developed with six injury classes (1–6) and one class for injuries that were not diagnosed with sufficient certainty (9). If you expand the classes to include the uninjured (0) and exclude the insufficiently diagnosed patients, you get an ordinally scaled ranking according to survival probability: An injury with the value, also known as the AIS code, of 0 will almost certainly not lead to the death of the Patients, whereas an injury with the AIS code 6 is very likely to result in death.
What is special about AIS is that the survival probabilities for the individual injuries are not explicitly specified, but only groups of injuries with approximately the same risk are formed. In this way injuries from different parts of the body can be compared in terms of their lethality. Survival probabilities can be assigned to the AIS codes within a database with AIS-coded injuries. However, this is only possible without problems for individual injuries (see below).
Ordinal scaling
The choice of ordinal scaling must also be taken into account: An AIS code value of 2 is associated with a greater risk of mortality than a value of 1, but no statement can be made about the size of the difference due to the scaling. Thus, two injuries with the value 1 are not associated with the same mortality risk as an injury with an AIS code of 2. Generally speaking, the AIS assessment of injuries is an assessment of a single injury and the mortality class determined with the coding applies only for this single injury.
Injury coding
In the first versions of the AIS scale, only injuries with a morphological correlate could be coded, i.e. injuries for which an objective finding can be obtained using diagnostic methods and directly, e.g. B. by photography, X-ray or autopsy, can document. Using so-called “modifiers”, the treating or coding doctors could initially raise or lower the AIS code by 2 “points” and later by one “point”. As part of the objectification of the survey (from AIS 1990), this possibility of evaluating one's own case was deleted. Injuries with functional limitations without a direct morphological correlate (e.g. unconsciousness) can only be coded since the AIS 1990. To improve the objectivity of the coding, only findings diagnosed and documented by the physicians directly treating you may be used. A detailed coding instruction that is binding for all users precedes the AIS code books. The lethality risk classes created by the AIS met the requirements for an assessment of occupant risk at the time. It was only later that it became apparent that the probability of survival can be used as a very valid proxy variable for the severity of an injury. This is especially true if the technical accident severity and the survival probability are used as proxy variables for the injury severity.
Further development
The distribution and further development of the AIS was taken over by the American Association for Automotive Medicine (American Association for Automotive Medicine, AAAM) and its successor, the Association for the Advancement of Automotive Medicine ( Association for the Advancement of Automotive Medicine , AAAM).
The International Injury Scaling Committee (IISC) of the AAAM is the sponsor of the AIS and monitors its exploitation.
The current further development tries to link a measure for the long-term consequences of an injury ( Functional Capacity Index ) to the AIS identifier through a more precise anatomical structure of the AIS identifier .
Different versions of the AIS
AIS 1969
In 1969 John D. States presented the "Abbreviated Injury Scale" at the STAPP conference as chairman of the Ad hoc Injury Scaling Committee. The severity of the injury ranges from “uninjured” with the value “0” to “minor”, “serious”, “severe (not life-threatening)”, “severe (life-threatening)” to “critical (survival uncertain)” with the value 5. A special feature was the differentiated consideration of accident fatalities.
| AIS69 code | patient | description |
|---|---|---|
| 6th | died within 24 hours | fatal injury in an AIS body region with simultaneous AIS code value of 3 or less in another body region |
| 7th | died within 24 hours | fatal injury in one AIS body region with simultaneous AIS code value of 4 or 5 in another body region |
| 8th | deceased | 2 fatal injuries in 2 AIS body regions |
| 9 | deceased | 3 or more fatal injuries |
It should be pointed out once again that these definitions of the AIS code versions 6, 7, 8 and 9 no longer correspond to the AIS code books from the 1976 revision.
For each severity of injury, the codebook describes the injuries or injured organ systems. A correlation of the AIS injury severity with the injury severity coded by the police was made possible by transferring the police coding to the AIS codebook.
With the publication of the first AIS codebook, the problem of injury aggregation, i.e. the summarizing assessment of all injuries suffered by an accident victim, was recognized as a problem that could not be directly solved using the AIS scale. In order to still be able to assign an overall injury severity to the individual persons, the 'Overall AIS' (OAIS) was proposed. However, its exact definition was only described in more detail in later publications, until it was finally discarded due to a lack of objectivity.
Comprehensive Injury Scale (CIS)
In addition to the AIS, States also presented a Comprehensive Injury Scale (CIS), which encoded separate assessments for the energy converted, permanent disability, length of treatment and frequency in addition to the life-threatening nature of an injury. However, the CIS could not prevail.
AIS 1971
The publication of the "Abbreviated Injury Scale" in the Journal of the American Medical Association (JAMA) is considered, due to the more technical orientation of the STAPP conference, for most medical professionals as the birth of the medical individual injury assessment.
The injury descriptions developed by doctors and statisticians up to that point primarily concerned the coding of the diagnosed diseases, the damaged structures and the cause of death or a description of the clinical course.
AIS 1976 revision (AIS76)
After the published intermediate steps in 1974 and 1975, the first manual on AIS coding was published in 1976. The main aim of the 1976 AIS edition was to consolidate the different AIS code versions for the deceased (7, 8 and 9). The assessment of an injury should no longer depend on the survival of the accident victim, since an increase in the AIS code in the event of death leads to a reduction in the mortality of the lower AIS code values. In this way, survival rates can also be determined for individual injuries. This separation of individual injury and survival serves to specify the cause of the injury while at the same time excluding medical treatment from the code. An example is a crush injury to the thorax: According to AIS71, it was rated with an AIS code of 5, but if the patient died within the first 24 hours, the AIS code 6 was assigned.
As of AIS76, the AIS code version 6 also includes injuries that inevitably lead to death with the given rescue system.
For injuries of unknown severity, the new version 9 was introduced. Assigning an AIS code of 6 just because the accident victim has died is no longer permitted or possible. Whether and when (immediately, after 24 hours or after 30 days) a patient died in an accident must be coded independently of the AIS: The AIS code only indicates a grouped probability that is not explicitly specified for the lethality of a single injury. The probabilities can be calculated for the types of accident and supply examined.
Furthermore, the revision focused on the distinction between the violation to be coded and the result of the injury. The consequence of the injury is not an injury, but can be used for a qualitative assessment of the injury. So the hemopneumothorax is not an injury. Rather, it is a chest trauma due to which the lungs collapse and bleeding into the pleural cavity has occurred.
In addition to the OAIS, the Injury Severity Score was also presented in the Codebook as an assessment method for multiple injuries.
AIS 1980 revision (AIS80)
In addition to bony skull injuries, injuries to the central nervous system (but without spinal cord injuries) also count as head injuries in AIS . Injuries to the brain are usually injuries without a morphological correlate, i.e., with the exception of the autopsy , the examiner can only determine the damage indirectly via clinical signs or, as in the case of the duration of unconsciousness, has to rely on the information left by third parties. This lack of direct orientation towards injured structures led to a wide variation in the quality and quantity of the coded head injuries. For a better illustration of the head injuries, these were divided into surgical-anatomical and clinical diagnoses. This procedure differs from the strictly morphological alignment of the coding in the other body regions. The use of clinical signs and symptoms for the classification of injuries was introduced as an intermediate step up to a sufficiently precise, objective diagnosis based on morphological changes. Thanks to the widespread use of CCT and, above all, cranial MRI recordings in the trauma centers, a large part of the clinical diagnoses could already be objectified using imaging methods with the AIS2005.
Although the number of burn injuries after traffic accidents was low, a simplified scheme for the coding of thermal injuries was included for a more extensive application of the AIS catalog. The classification is based on the degree of burn as well as Wallace's rule of nine, i.e. the relative proportion of the burned body surface.
The AIS coding of an injury depends on the documents used, such as rescue protocols, admission reports, x-rays, discharge letters, etc. In a study by the National Center for Health Statistics, the discharge letters turned out to be the most reliable source, so according to the AIS80 Codebook they should be used as the primary source for the AIS coding. However, it is explicitly pointed out that this should not be the only source and that it is not sufficient for some research projects to limit oneself to this information, be it for reasons of the timeliness of the accident, the quality of the survey or the orientation.
With the introduction of AIS in 1980, injury aggregation using OAIS was presented as an undesirable procedure: The coders tended to portray their patients as particularly severe cases, with the result that the OAIS injury severity increased. By relying on “mathematical procedures” such as MAIS and ISS, the aggregation of injuries should become more objective, i.e. more independent of the person responsible.
AIS 1985 revision (AIS85)
Numerical Injury Identifier
With the revision of the AIS in 1985, the “Numerical Injury Identifier”, or AIS85-ID for short, was introduced for each of the individual injuries that can be coded. The aim was to simplify a computer-based evaluation of injury locations and types and not just the severity of injuries specified in the AIS code. The six-digit AIS85 ID was divided into four blocks: "Body region", "Type of anatomical structure", "Specific anatomical structure or type of injury" and "Extent of injury". The most significant digit codes for the body region (see below), the second most significant for the "type of anatomical structure such. B. vessels, nerves, organs, skeletal system ”. The two-digit block “Specific anatomical structure or type of injury” should ensure the exact anatomical localization of the injury to be coded in a systematic manner, as well as coding the origin of the injury. The two lowest digits encode the "extent of the injury".
In order to prevent incorrect coding due to a lack of information, in which the encoder adapts the violations to the AIS IDs available for selection, AIS IDs with the AIS code "9" (NFS, "not further specified", German "not specified more precisely") introduced.
Each AIS85 ID is clearly assigned exactly one AIS85 code. In the nomenclature, the AIS code is separated from the Numerical Injury Identifier by a point. The AIS code is therefore also referred to as the post-dot value, i.e. the value after the point.
Other changes
The AIS code has been reduced for some individual injuries. The basis for this is the comparative evaluation of injuries over the years. It is practically a picture of the progress made in rescue and first aid for accident victims. The adjustments are mainly based on the US surveys and thus on the rescue and medical services there.
Due to the greater acceptance of AIS in traumatology, the anatomically shaped terminology was expanded to include clinical terminology. In order to avoid misunderstandings, especially in the case of evaluators who are not medically trained, colloquial descriptions have also been inserted for clarification (e.g. hematoma contusion, minor-superficial).
For the extended application, penetrating injuries were included as an injury mechanism.
ICD-9CM to AIS85 conversion
In the 9th edition of the International Statistical Classification of Diseases and Related Health Problems , ICD-9 for short, the World Health Organization (WHO) followed a disease-causative approach in addition to coding the diseases. That means that in addition to the actual diagnoses, the causes of the disease always had to be coded. The "Clinical Modification" (modification for clinical use) of the ICD-9 by the National Center for Health Statistics of the USA did not change this double recording of the disease.
The approach of coding the cause of injury in the ICD-9CM hardly differed from that of the AIS, so that a conversion table for the AIS85 could be created for the ICD-9CM, under some restrictive premises, which was published in 1986. In this way, AIS85 codes could be automatically determined for all trauma patients from the coded discharge diagnoses. The method was validated by the authors in 1989.
AIS 1990 revision (AIS90) and AIS 1990 update 1998 (AIS98)
In addition to a more detailed codability of the injuries required for the evaluation of injuries and, above all, injury patterns, coding guidelines and coding aids in the AIS codebook should limit incorrect coding due to ignorance or uncertain information as a source of error. The changes brought about by Update 98 are primarily aimed at better processing of the coding rules so that AIS90 and AIS98 can be dealt with together in the following.
The AIS90 / 98 scale is made up of a 6-digit "Numerical Injury Identifier", or AIS98-ID for short, and the single-digit AIS98 code that evaluates the severity of injury (actually the probability of survival), with the possible values 1, 2, …, 6 and 9, together. Each AIS98 ID is clearly assigned exactly one AIS98 code. In particular, brain injuries and, increasingly, extremity injuries could not be adapted to the rigid coding scheme, so that it was relaxed with the AIS of 1990 for the blocks “Specific anatomical structure or type of injury” and “Extent of injury”.
Age can have a dominant influence on injury mortality. For the adaptation of the AIS for children and infants, the AIS85 was assessed by pediatric trauma surgeons. The 15 of 2000 injury assessments for which a different injury severity was found for children were integrated into the AIS-90. It is mainly a description of the size of hematomas in the brain, the extent of blood loss in deep abrasions, and internal blood loss through thoracic or abdominal trauma.
There is no scope for evaluating a higher (lower probability of survival) or lower (higher probability of survival) AIS98 code by means of so-called “modifiers” for an AIS98 ID. Such an adaptation of the AIS98 code to the specific case would contradict the criteria of the objective survey.
The summary of all external injuries in the body region "External" has been abandoned: Skin injuries that are not thermal or electrical were distributed to the injured body parts. These are AIS 1 and 2 injuries. In addition to simpler coding, body region-specific evaluations can also include skin injuries. This change makes the formation of the Injury Severity Score (ISS) more difficult.
The assumption that severe brain injuries (AIS> = 3) are still coded too seldom could be objectified with the help of an evaluation of the MTOS database. To remedy this problem, new AIS IDs were introduced for bruises and hematomas in the brain, which also take into account the size, location and frequency of the injury. Furthermore, possible descriptions of injuries to the vessels in the brain and the cranial nerves have been included. From the AIS90 onwards, the base of the skull fracture may also be coded without imaging documentation, i.e. purely according to clinical signs.
In numerous studies on injuries to the brain, a predominance of low-grade concussions (Commotio cerebri, English Cerebral concussion, AIS98-ID: 161000) was found. This can be remedied by devaluing the corresponding AIS90 / 98 code from 2 to 1, according to the evaluation in AIS2005. Another reason for the devaluation of the concussion is the lack of objectification of the findings. A survey of the neurological status does not take place regularly, so that it cannot actually be coded. When evaluating retrospectively collected mental deficits, a missing coding or a bias to exaggerate the brain injury was found.
The expansion of the description of injuries, which had already begun with the AIS85, away from a purely anatomical language towards a surgical and slang description, was consistently continued with the AIS90. With the AIS90 Codebook, text blocks with explanations were integrated into the dictionary part of the Codebook. This has been expanded again with the AIS98 Codebook with the preceding, binding " Coding Rules " .
ICD-9CM to AIS90 / 98 conversion
The conversion list created for the AIS85 was transferred to the AIS90 and also the update from 1998. The conversion list is not published and the software based on it is not freely available.
However, with the mandatory use of the ICD-10, which is oriented towards illness and billing, in Germany, the possibility of linking ICD and AIS was lost. In the USA, the causes of illness or injury must still be coded according to the ICD-9CM according to the specifications of the National Center for Health Statistics (NCHS), a division of the Centers for Disease Control (CDC).
AIS 2005 (AIS2005) and AIS 2005 update 2008 (AIS2008)
After the AIS code had established itself as a very valid means of describing the probability of survival of individual injuries and also showed stable results as a proxy variable for the medical injury severity, an attempt was made with the AIS2005 to provide a more detailed "Numerical Injury Identifier" that enables the direct assessment of the long-term impairment of the injured person. Analyzes showed that the brain and extremity injuries in particular were not sufficiently detailed for an assessment of the long-term consequences of an injury. Adapted versions of the Functional Capacity Index (FCI) for most AIS-IDs were published with the AIS2005 Update 2008 (AIS2008) Codebook. Attempts to use this FCI for a holistic view of the injured person are currently still failing due to a suitable aggregation function across the body regions. The AIS2008-FCI coupling can already be used successfully within individual body regions (see e.g. foot injuries).
In addition to the Functional Capacity Index for long-term impact assessment, it should also be possible to link the disease diagnoses coded using the ICD-10 and the causes of the disease coded according to ICD-9CM and later according to ICD-10CM with the AIS-ID. In addition to the compatibility of the AIS-ID with the coding approaches of the trauma-related medical departments, attention was also paid to the needs of biomechanical research.
By expanding the so-called bilateral injuries, here injuries that can be found simultaneously on the right and left side of the body are combined into one injury, the severity of the injury can be described much more precisely. Combining several injuries under one AIS-ID changes the number of injuries: This makes the recoding of older cases more difficult, the comparability between the AIS generations more complicated and the characteristics of some injury aggregation masses change. For multiple injury or injury causes a bilateral violation of the newly introduced is cause of injury coding (COI, engl. Cause Of Injury) absurdity, because each AIS ID only a COI can be assigned.
Two localizers , L1 and L2 , were introduced for more precise anatomical localization of the injury . While L1 describes the position and characteristics in more detail in accordance with the common medical nomenclature, L2 explicitly specifies the damaged body structure. Standard variants, however, cannot be described. These new variables are supplemented by a 4-digit code for coding the cause of the injury . Both the localizers and the COI can be seen as an optional addition to the AIS identifier, the coding of which is not required for an AIS codebook-compliant classification of injuries.
In addition to the structural adjustments, the corresponding AIS codes of the current clinical care have been adapted for some AIS IDs.
Even if the AIS2005 ID is a superset of the AIS98 ID, recoding is not possible in all cases: The information reduction by the original AIS98 coding cannot be canceled without additional information for the more detailed AIS2005 coding. Furthermore, the technical progress in trauma diagnostics leads to more differentiated diagnoses, which can be represented in AIS2005, but cannot be ascertained for old trauma cases. An example is the use of magnetic resonance imaging (MRT) for brain injuries, which has only become routine in recent years . Despite this limitation, the AIS2005 / AIS2008 code books offer the best possible proposal for recoding from and to AIS98. However, conversion, particularly automated conversion, is still the subject of current research.
ICD-10 GM to AIS2005 / 2008 conversion
Due to the great importance of the ICD-10 in trauma care, especially in terms of accounting, attempts have been made at various points to automatically convert ICD-10GM-coded injuries into AIS2005 / 2008 coding. An independent validation of the recoding is still pending.
Specification of the injured body region
Since AIS85, the injured body region has been specified by the most significant digit in the AIS ID:
| AIS98 ID (first digit) | Body region | Body region | Number of AIS98 IDs | Number of AIS2008 IDs |
|---|---|---|---|---|
| 1 | Head | Head (without face and facial skull) | 236 | 281 |
| 2 | Face | Facial skull, face (including eyes and ears) | 87 | 175 |
| 3 | Neck | Neck (without spinal cord) | 80 | 111 |
| 4th | thorax | Rib cage | 175 | 191 |
| 5 | abdomen | Abdomen | 233 | 250 |
| 6th | Spine | Spine | 208 | 216 |
| 7th | Upper extremity | Arms (including shoulder) | 125 | 325 |
| 8th | Lower Extremity | Legs (including hips and pelvic bones) | 164 | 402 |
| 9 | External and other trauma | External and other injuries | 33 | 48 |
Overall, there are different Numerical Injury Identifiers in AIS98 1341 and AIS2008 1999, there is no direct checksum , but the code is over-specified because there is exactly one AIS98 code for each localizer.
Specification of the type of injury
Since AIS85, the type of violation has been specified by the second-highest digit in the AIS-ID:
| AIS98 ID (second digit) | Type of injury | Type of injury |
|---|---|---|
| 1 | Whole Area | Entire surface |
| 2 | Vessels | Vessels |
| 3 | Nerves | annoy |
| 4th | Organs (including muscles and ligaments) | Organs (including muscles and ligaments) |
| 5 | Skeletal (including joints) | Bones (including joints) |
| 6th | Head, Loss of Consciousness (LOC) | Head, loss of consciousness |
AIS code
There are 6 degrees of severity of injury (AIS98 codes) and a placeholder (AIS98 code = 9) for the general description of injuries, which cannot be specified in more detail due to a lack of diagnostics:
| AIS code | AIS Injury Severity | AIS Injury Severity | Number of AIS98 codes | Number of AIS2008 codes |
|---|---|---|---|---|
| 1 | Minor | Low | 258 | 447 |
| 2 | Moderate | Seriously | 404 | 729 |
| 3 | Serious | Heavy | 339 | 419 |
| 4th | Severe | Very difficult | 154 | 172 |
| 5 | Critical | Critical | 141 | 155 |
| 6th | maximum | Maximum (not treatable) | 24 | 33 |
| 9 | NFS (Not Further Specified) | Not specified | 21st | 44 |
The linguistic description should help to roughly classify the severity of the injury, it is not used for coding. "Minor" injuries include superficial abrasions, cuts and hematomas. Injuries that were considered “untreatable” and therefore fatal when the codebook was drawn up were given an AIS code of 6. Codes of 9 are a special case: Here the coder has information about an injury, but this is not sufficient to precisely determine the anatomical structure involved (for example, leg injury when rolling over and fatally without an autopsy).
Survival probabilities of the AIS code values
Variations 1 to 6 of the AIS code represent, by definition, grouped survival probabilities without specifying the probabilities behind them. The reason lies in the AIS approach to documenting injuries, which does not want to assess the rescue and treatment of injuries. In order to be able to assign a survival probability to the characteristics of the AIS code, only injured persons are considered for whom exactly one individual injury was coded. For injuries in the National Trauma Data Base (NTDB) this is 38.3% of the coded injuries. A corresponding approach led to the values of the trauma register and the GIDAS survey. GIDAS (MAIS) denotes the survival probabilities of the individual injuries to an injured person aggregated using MAIS, regardless of their number. The different values for the probabilities of survival are shown in the following table:
| AIS98 code | AIS Injury Severity | NTDB | Trauma.org | GIDAS (AIS) | GIDAS (CORN) |
|---|---|---|---|---|---|
| 0 | Unharmed | 100.0 | 100.0 | 100.0 | 100.0 |
| 1 | Low | 99.3 | 100.0 | 99.8 | 99.9 |
| 2 | Seriously | 99.2 | 99.3 | 99.4 | 99.6 |
| 3 | Heavy | 96.5 | 97.1 | 98.3 | 95.8 |
| 4th | Very difficult | 85.4 | 93.1 | 74.6 | 74.9 |
| 5 | Critical | 60.4 | 67.7 | 61.5 | 42.0 |
| 6th | Maximum (not treatable) | 21.0 | 0.0 | 0.0 | 0.0 |
Injury Severity not listed with an AIS98 code of 9 is either incompatible injuries or incompletely diagnosed injuries. Thus, an indication of the survival probability for this class would say something about the distribution of injuries between these two subgroups, but not something about the actual lethality of the injuries depicted in the class.
Scale level of the AIS code
According to the definition in the AIS98 codebook, the AIS code is nominally scaled with its versions 1,…, 6 and 9 . If one excludes the “violations not specified more precisely”, ie all violations with an AIS code of 9, the scale level rises to ordinal , and there is a ranking among the AIS codes . When considering the severity of injuries to persons with multiple injuries, persons with even an AIS code of 9 must be excluded if the ordinal scale level is to be retained (see e.g. ISS ). The ordinal scale level remains with the introduction of the value “0” for uninjured persons or body regions.
The probabilities of survival behind the AIS codes are interval-scaled .
Aggregation of injuries
The AIS code describes individual injuries that have to be aggregated in order to consider the severity of the injury to a person or even just a body region. Aggregated injury assessments of injuries coded using different AIS Codebook revisions cannot be compared with one another. On the one hand, this is due to the fact that the options for coding injuries have become more and more detailed due to the expansion of the field of application of the scale and the development of diagnostic imaging methods ( ultrasound , CT , MRT ). Furthermore, numerous AIS codes had to be adapted via the codebook revisions, even though the AIS measuring system does not want to evaluate the treatment of the injuries, but only their severity, or better, their risk of mortality, when they arise. Thus, for long-term studies, all violations must be carried out according to the rules and with the AIS code evaluation of an AIS codebook revision. Due to the medical diagnostic technology integrated in the AIS codebook and therefore prescribed, for example when differentiating between brain swelling and brain edema, retrospective coding of old and very old injuries according to the newer AIS code books is usually not possible.
Overall AIS (OAIS)
The "Overall AIS" (OAIS) is the clinical assessment of the injury severity of all injuries to a person by a doctor experienced in the treatment of accident victims. When creating the OAIS, mutual influences of the individual injuries with regard to the severity of the injury should be taken into account. The OAIS scale is identical to the AIS scale on which it is based.
It was noted in the codebook that there is no formula for creating the OAIS; it is not the sum of the AIS codes, but a subjective assessment of the overall severity of the injury. It was explicitly pointed out that the OAIS can also be greater than the largest AIS-rated individual injury, but the OAIS can never be smaller than the MAIS. The tendency of the coders to rate the overall injury pattern as high as possible led to a bias (shift) of the coded OAIS values in the direction of higher OAIS values. Due to the lack of objectivity of the violation aggregation by the OAIS, its use has not been supported since the AIS revision of 1980.
The OAIS was the first attempt to aggregate AIS-coded single injuries.
Maximum AIS (MAIS)
In medical and, above all, technically oriented presentations, according to a suggestion from the AIS codebook from 1980, the severity of a patient's injury is often given as the maximum AIS value (MAIS, more rarely also maxAIS). A maximum value formation requires at least an ordinal scaling of the characteristics. With regard to AIS injury codes, this means that only injury characteristics between 1 and 6 may be aggregated if none of the injuries considered was coded with a 9 at the same time. When considering injuries to body regions, it is customary to specify a MAIS value for them (e.g. MAIS thorax). Correspondingly, no injuries affecting the body region may be coded with an AIS code of 9. The definition of the body regions does not necessarily have to be based on that of the AIS Codebook, see for example the body regions of the Injury Severity Score (ISS).
Injury Severity Score (ISS)
The Injury Severity Score (ISS) is an injury aggregation of injuries assessed with the AIS code, the main aim of which is to derive a limit for multiple trauma from the severity of individual injuries to individual body regions. The body regions are identical to those of the AIS before 1990, i.e. H. the body regions for the calculation of the ISS do not correspond to those of the AIS90ff. The AIS identifiers must be regrouped accordingly before the calculation.
Further injury aggregations
Another injury aggregator is the New Injury Severity Score (NISS) .
In addition to these aggregators based purely on the severity of the injury, there are numerous that also take physiological parameters into account. With this additional information, the patient's injury severity can generally be better estimated, but the disadvantage is that the emergency treatment has a major impact on the score result: For a possibly more precise injury severity, the medical treatment, even if it is only the emergency treatment, is linked to the injury severity. Another problem is the different data situation depending on the type of treatment (outpatient, inpatient), both in terms of quantity and quality.
Aggregation of the severity of the injury
For many areas, 6 levels of injury severity (plus '0' for uninjured and '9' for a classification not specified sufficiently) are too detailed: Either because the frequency of occurrence of individual injury severity levels is too low, making statistical processing difficult or even impossible, or because a division of the medical consideration of a connection in 8 cases is out of proportion to the technical accuracy.
CORN x +
The injury aggregation via MAIS x + converts the ordinally scaled MAIS variable into a binary variable, the value of which for MAIS is below x, 0 and for values of x and greater, 1. As a rule, persons with an injury with an AIS code of 9 are excluded from the 'MAIS x +' considerations. An assignment of the injuries coded with AIS 9 to the different MAISx + characteristics is often possible after an individual case analysis.
The table shows the injury severity aggregators used in practice:
| AIS code | AIS Injury Severity | AIS Injury Severity | ISO injury severity | Probability of survival | Injury severity | Universal |
|---|---|---|---|---|---|---|
| 0 | Not injured | Unharmed | S0 | P1 | I1 | E1 |
| 1 | Minor | Low | S1 | P1 | I1 | E1 |
| 2 | Moderate | Seriously | S1 | P1 | I2 | E2 |
| 3 | Serious | Heavy | S2 | P2 | I2 | E3 |
| 4th | Severe | Very difficult | S2 | P3 | I3 | E4 |
| 5 | Critical | Critical | S3 | P3 | I3 | E4 |
| 6th | maximum | Maximum (not treatable) | S3 | P3 | I3 | E4 |
| 9 | NFS (Not Further Specified) | Not specified | - | - | - | - |
Probability of survival from an injury
This classification is based on the lethality of the injuries. AIS values of 0, 1 and 2 have a very low lethality and are therefore grouped together. By including the AIS code value of 0, it is possible to compare the number of injured persons with the population of the surveyed persons (or body regions) also retrospectively. So not only injured people / body regions are considered, but also people who are raised. Injuries with an AIS code of 3 have a not inconsiderable risk of death, but this is significantly lower than that of AIS 4, 5 and 6 injuries. MAIS3 + and MAIS4 + can be determined from the aggregated data.
Severity of an injury
A distinction was made according to the severity of the injury from the point of view of the injured person. Cases with no or only minor injuries (AIS code 0 and 1) are grouped together. Various surveys have shown that there is no clear line between “uninjured” and “slightly injured”: simply because of the possibility of financial compensation for an injury, a previously “uninjured” person leads to injuries with an AIS code of 1 need to be assessed and vice versa. By including the AIS code value of 0, this aggregation can also be related to the total number of persons surveyed. Injuries with an AIS score of 2 and 3 are severe and painful injuries from the patient's point of view, but they are clearly differentiated from those with an AIS score of 4, 5 or 6. MAIS2 + and MAIS4 + can be calculated from the aggregated data.
Universal aggregation
The form often found in the literature results from the statements on the aggregation of the severity of injuries: Uninjured and slightly injured persons (AIS 0 and 1) are summarized. The numerically frequent occurrences of 2 and 3 are coded separately. By summarizing the values 4, 5 and 6, in most of the surveys there are enough cases in the group of the very seriously injured for a statistical analysis. In addition to MAIS2 +, MAIS3 + and MAIS4 +, the multiple trauma criterion 'ISS> = 16' can also be calculated from the aggregated data.
ISO aggregation
In the ISO proposal for injury severity aggregation, uninjured and individual injury severities are recoded from 1 to 6. Situations that lead to an AIS 9-rated injury are excluded from consideration. With this coding, which is frequently used in consideration of functional safety , the 7 AIS98 versions (0, 1–6) are assigned to the 4 states S0 to S3 (“no injury”, “practically no hazard”, “not insignificant hazard” and “ Risk to life and limb ”) reduced. This limitation of the characteristics simplifies the assignment of ISO-assessed injury patterns with the "technical failures" to be assessed.
In addition to the severity of the injury, the probability of the severity of the injury occurring is also taken into account. With the value S3 , situations are coded in which more than 10% of the cases lead to injury severity of AIS5 or 6, i.e. AIS5 +. S2 depicts situations in which AIS3 + injuries are to be expected in more than 10% of cases and S3 coding overestimates the severity of the injury. Accordingly, S1 depicts situations in which AIS1 + injuries are to be expected in more than 10% of the cases and neither the code S2 nor S3 describe the situation correctly. The value S0 is reserved for situations in which no injuries occur in at least 90% of the cases.
Coupling the situation and the severity of the injury prevents exceptional cases from being overestimated: For example, the situation of “crossing a zebra crossing” is possible in over 90% of cases without injuries, but in exceptional cases people crossing a zebra crossing are seriously injured or even killed .
literature
- John D. States: The Abbreviated and the Comprehensive Research Injury Scales . In: STAPP Car Crash Journal . tape 13 . Society of Automotive Engineers, New York 1969, p. 282-294 , doi : 10.4271 / 690810 , JSTOR : 44644251 .
- Thomas A. Gennarelli (Ed.): The Abbreviated Injury Scale . 1990 Revision Update 1998. Association for the Advancement of Automotive Medicine (AAAM), Barrington IL 2001.
- Thomas A. Gennarelli, Elaine Wodzin (Eds.): The Abbreviated Injury Scale 2005 . Update 2008. American Association for Automotive Medicine (AAAM), Des Plaines IL 2008.
- Thomas A. Gennarelli, Elaine Wodzin: AIS 2005: A Contemporary Injury Scale . In: Injury . International Journal of the Care of the Injured. tape 37 , no. 12 . Elsevier, December 2006, pp. 1083-1091 , doi : 10.1016 / j.injury.2006.07.009 .
- Carl Haasper, Mirko Junge, Antonio Ernstberger et al .: The Abbreviated Injury Scale (AIS) . Potential and problems in use. In: The trauma surgeon . tape 113 , no. 5 . Springer, Berlin May 2010, p. 366-372 , doi : 10.1007 / s00113-010-1778-8 .
Individual evidence
- ↑ Walied Abdulla: Interdisciplinary Intensive Care Medicine. Urban & Fischer, Munich et al. 1999, ISBN 3-437-41410-0 , p. 466 f.
- ↑ Walied Abdulla: Interdisciplinary Intensive Care Medicine. Urban & Fischer, Munich et al. 1999, ISBN 3-437-41410-0 , p. 466 f.
- ^ DJ van Kirk, WA Lange: A Detailed Injury Scale for Accident Investigation . In: STAPP Car Crash Journal . tape 12 . Society of Automotive Engineers, 1968, ISSN 1532-8546 , LCCN 67-022372 , pp. 240-259 .
- ^ Accident Investigation And Reconstruction Practices Committee: Collision Deformation Classification . SAE J224. SAE, March 1980 ( can be ordered from sae.org ).
- ^ WD Nelson: The History and Evolution of the Collision Deformation Classification SAE J224 Mar80 . SAE, February 1981 ( can be ordered from sae.org ).
- ^ A. Howard Hasbrook: The historical development of the crash-impact engineering point of view. In: Clinical orthopedics . tape 8 . Lippincott, 1956, ISSN 0095-8654 , LCCN sf85-003079 , p. 268-274 .
- ^ A. Howard Hasbrook: The historical development of the crash-impact engineering point of view . In: William Hardon, Edward A. Suchman, David Klein (Eds.): Accident Research . Methods and Approaches. 1st edition. Harper & Row, New York 1964, chap. 9 , p. 547-554 (Abridged reprint from Hasbrook 1956).
- ^ A b c d John D. States: The Abbreviated and the Comprehensive Research Injury Scales . In: STAPP Car Crash Journal . tape 13 . Society of Automotive Engineers, 1969, ISSN 1532-8546 , LCCN 67-022372 , pp. 282-294 .
- ↑ a b c d Joint Committee of AMA, AAAM, SAE (Ed.): The Abbreviated Injury Scale . 1976 revision. American Association for Automotive Medicine (AAAM), Morton Grove IL 1976.
- ↑ a b c d e f g John D. States, Donald F. Huelke (Ed.): The Abbreviated Injury Scale . 1980 revision. American Association for Automotive Medicine (AAAM), Morton Grove IL 1980, Part I / III, pp. 6 .
- ^ A b John D. States, David C. Viano: Injury Impairment and Disability Scales to Assess the Permanent Consequences of Trauma . In: Accident Analysis and Prevention . tape 22 , no. 2 . Pergamon Press, 1990, ISSN 0001-4575 , pp. 151-159 , doi : 10.1016 / 0001-4575 (90) 90066-T .
- ^ William K. Keller for the AMA Committee on Medical Aspects of Automotive Safety: Rating the Severity of Tissue Damage . I. The Abbreviated Injury Scale. In: The Journal of the American Medical Association (JAMA) . tape 215 , no. 2 . American Medical Association, Jan 1971, ISSN 0098-7484 , pp. 277–280 ( jama.ama-assn.org can be ordered).
- ^ William K. Keller for the AMA Committee on Medical Aspects of Automotive Safety: Rating the Severity of Tissue Damage . II. The Comprehensive Scale. In: The Journal of the American Medical Association (JAMA) . tape 220 , no. 5 . American Medical Association, May 1972, ISSN 0098-7484 , pp. 717–720 ( jama.ama-assn.org can be ordered).
- ↑ World Health Organization [WHO] (Ed.): The International Statistical Classification of Disease, Injuries and Cause of Death . Geneva 1967.
- ↑ William Bouzarth: Neurosurgical Watch Sheet for craniocerebral trauma . In: The Journal of Trauma . tape 8 , no. 1 . Lippincott William & Wilkins, January 1968, ISSN 0022-5282 , pp. 29-31 ( journals.lww.com ).
- ↑ John D. States, DF Huelke, LN Hames: 1974 Revision of the AMA-SAE-AAAM Abbreviated Injury Scale (AIS) . In: Proceedings . American Association for Automotive Medicine (AAAM), Toronto 1974.
- ^ Joint Committee on Injury Scaling of SAE-AAAM-AMA: The Abbreviated Injury Scale (1975 Revision) . In: Proceedings . American Association for Automotive Medicine (AAAM), San Diego CA 1975.
- ^ A b c d Susan P. Baker, Brian O'Neill, William Haddon Jr., William B. Long: The Injury Severity Score . A method for describing patients with multiple injuries and evaluating emergency care. In: The Journal of Trauma . tape 14 , no. 3 . Lippincott Williams & Wilkins, March 1974, ISSN 0022-5282 , pp. 187-196 , PMID 4814394 .
- ↑ a b c d e f g Thomas A. Gennarelli, Elaine Wodzin (eds.): The Abbreviated Injury Scale 2005 . American Association for Automotive Medicine (AAAM), Barrington IL 2006.
- ↑ Elaine J. MacKenzie, EA Garthe, G. Gibson: Evaluating the Abbreviated Injury Scale . In: Proceedings . American Association for Automotive Medicine, 1978.
- ^ The Abbreviated Injury Scale . 1985 revision. American Association for Automotive Medicine (AAAM), Des Plains IL 1985.
- ↑ Lily C. Huang, Joseph C. March: AIS and threat to life . In: Proceedings . tape 22 . American Association for Automotive Medicine, 1978, p. 242-254 .
- ↑ Susan P. Baker, Brian O'Neill: The Injury Severity Score: An Update . In: The Journal of Trauma . tape 16 , no. 11 . The Williams & Wilkins Co., 1976, ISSN 0022-5282 , pp. 882-885 ( journals.lww.com ).
- ↑ a b Thomas A. Gennarelli (Ed.): The Abbreviated Injury Scale . 1990 Revision Update 1998. Association for the Advancement of Automotive Medicine (AAAM), Barrington IL 2001 ( slc.westdc.net can be ordered).
- ↑ Ellen J. MacKenzie, Donald M. Steinwachs, Belavadi S. Shankar: An ICD-9CM to AIS Conversion Table . Development and Application. In: Proceedings of the AAAM . tape 30 , 1986, pp. 135 ff .
- ↑ Ellen J. MacKenzie, Donald M. Steinwachs, Belavadi S. Shankar: Classifying severity of trauma based on hospital discharge diagnoses . Validation of an ICD-9CM to AIS-85 conversion table. In: Medical Care . tape 27 , no. 4 . Lippincott Williams & Wilkins, 1989, pp. 412-422 , JSTOR : 3765189 .
- ↑ a b c d e f g Thomas A. Gennarelli (Ed.): The Abbreviated Injury Scale . 1990 revision. Association for the Advancement of Automotive Medicine (AAAM), Des Plains IL 1990.
- ↑ Thomas A. Gennarelli, HR Champion, WJ Sacco, WS Copes, WM Alves: Mortality of patients with head injury and extracranial injury treated in trauma centers . In: The Journal of Trauma . tape 29 , no. 9 . The Williams & Wilkins Co., September 1989, ISSN 0022-5282 , pp. 1193-1202 ( journals.lww.com ).
- ↑ Wayne S. Copes, Howard R. Champion, William J. Sacco, Mary M. Lawnick, Susan L. Keast, Lawrence W. Bain: The Injury Severity Score revisited . In: The Journal of Trauma . tape 28 , no. 1 , January 1988, ISSN 0022-5282 , pp. 69-77 ( journals.lww.com ).
- ↑ a b c d e Thomas A. Gennarelli, Elaine Wodzin (Ed.): The Abbreviated Injury Scale 2005 . Update 2008. American Association for Automotive Medicine (AAAM), Des Plaines IL 2008 ( slc.westdc.net can be ordered).
- ↑ a b c d e f g Thomas A. Gennarelli, Elaine Wodzin: AIS 2005: A Contemporary Injury Scale . In: Injury . International Journal of the Care of the Injured. tape 37 , no. 12 . Elsevier, December 2006, ISSN 0020-1383 , p. 1083-1091 , doi : 10.1016 / j.injury.2006.07.009 .
- ↑ ICD-9CM homepage of the NCHS / CDC
- ↑ Carl Haasper, Mirko boy Antonio Ernst Berger et al .: The Abbreviated Injury Scale (AIS) . Potential and problems in use. In: The trauma surgeon . tape 113 , no. 5 . Springer, May 2010, ISSN 0177-5537 , p. 366-372 , doi : 10.1007 / s00113-010-1778-8 .
- ^ Rene Schubert: Foot injuries in front-passenger cars . An analysis of the GIDAS database (dissertation). 2010 ( qucosa.de [PDF; 2.4 MB ; accessed on August 15, 2011]).
- ^ Fracture and Dislocation Compendium . Orthopedic trauma association committee on coding and classification. In: Orthopedic Trauma Association (ed.): J Orthopedic Trauma . tape 10 , Supplement 1, 1996, pp. 1–155 ( 2007 version - meanwhile replaced by the 2007 version). Version 2007 ( Memento of the original from July 28, 2011 in the Internet Archive ) Info: The archive link was inserted automatically and has not yet been checked. Please check the original and archive link according to the instructions and then remove this notice.
- ^ Organ Injury Scales (OIS) of the American Association for the Surgery of Trauma .
- ↑ Derk Adams, Astrid B. Schreuder, Kristin Salottolo, April Settell, J. Richard Goss: Validation of the “HAMP” Mapping Algorithm: A Tool for Long-Term Trauma Research Studies in the Conversion of AIS 2005 to AIS 98 . In: The Journal of Trauma . tape 71 , no. 1 , July 2011, ISSN 0022-5282 , p. 85-89 , doi : 10.1097 / TA.0b013e3181f9f873 ( journals.lww.com [accessed August 20, 2011]).
- ↑ Boyan Nikolov: Translation of the Injury-Related Codes of the XIX. Chapter of ICD-10-GM 2008 in AIS 2005 . Münster 2010, p. 402 (dissertation in the medical department).
- ^ A b Sean O'Brien: Measurement and Assessment of Passenger Vehicle Compatibility in Front and Side Collisions . 2010 (Dissertation, RMIT University ).
- ↑ John M. Cavanaugh: Biomechanics of Thoracic Trauma . In: Alan M. Nahum, John W. Melvin (Eds.): Accidental Injury (Biomechanics and Prevention) . Springer, New York 1993, ISBN 3-540-97881-X , chap. 15 , p. 362-390 .
- ↑ Turner Osler, Susan P. Baker, William Long: A modification of the injury severity score that both improves accuracy and simplifies scoring . In: The Journal of Trauma . tape 43 , no. 6 . Lippincott Williams & Wilkins, 1997, ISSN 0022-5282 , pp. 922-925, 925-926 , PMID 9420106 .
- ↑ a b Road Vehicles - Functional Safety . Part 3: Concept phase. In: ISO (Ed.): ISO 26262 . International Organization for Standardization (ISO), 2010, p. 6.4.4.3.2 .
