Printer Friendly

Evaluation of Seizure Observation and Documentation.

Abstract: This study evaluated seizure documentation after participants observed videotaped seizures to determine how their background (educational level employment position, years of experience, practice frequency, and familiarity with epilepsy and seizures) and attributes of the seizures themselves affected their skill. Observer variables did not show significant differences when mean seizure rating scores of 348 documented seizures for 58 participants were compared. Combinations of variables were significant; certified nursing assistants (CNAs) with clinical seizure experience had significantly higher mean seizure rating scores thane those without experience (p [is less than] .01). There were significant differences in the mean percentile scores for each seizure observed (p [is less than] .001). Seizures with excess motor activity had the highest scores. Most common observations for any seizure observed were location and description of movement. Participants had significantly higher scores when observing a seizure a second time (p [is less than] .001). Differences in education, employment position, and years of employment are not limiting factors in performing seizure observation and documentation for persons who have been trained and certified in the skill. Opportunities to periodically view seizures improved seizure observation and documentation. Because motor movement observations predominate seizure documents, seizure observation instruction should not only emphasize key observational details but also highlight less frequent observations, such as responsiveness, that may play a key role in seizure classification. Using videotaped seizures and a scoring tool are useful for initial training and certification as well as periodic retraining to maintain quality in the skill.


Seizure observation and documentation skills are an integral component of neuroscience nursing. At a children's hospital where an epilepsy program was based, mandatory inservice training on seizure observation and documentation skills was completed by all nursing and therapy staff members. At this facility any staff member involved with a patient having a seizure had the clinical responsibility to provide seizure first aid and to observe and document the seizure. Criteria were developed for Certifying staff members in this skill by using videotaped seizures. All staff members were required to have annual recertification. The basic seizure document comprised observation guidelines and a chronological, narrative description rather than a seizure flow sheet format.

Despite the above measures there continued to be variability in seizure observation and documentation skills. The clinical nursing specialists, nursing staff, and medical staff members continued to work with staff members on improving their skills and began to speculate about what measures might decrease this variability. Medical staff members began to request 1:1 registered nurse (RN) staffing to improve the documentation. The costs and implications of this would be prohibitive. The literature was reviewed. No studies related to this skill had been conducted.


In 1954, Penfield and Jasper established that within groups of seizures there were correlates between the attack patterns or behavioral responses of the patient and characteristics of the electroencephalogram (EEG) data.[14] The International League Against Epilepsy, first in 1969 and then with revision in 1981, developed and refined a classification system of seizures based on the area of brain involved at the onset of the seizure. Seizures were classified as either partial or generalized, according to whether or not consciousness was disturbed.[6] In 1998, Luders et al. proposed an epileptic seizure classification based exclusively on ictal semiology or symptomatology. Detailed correlations of symptoms would better define the pathophysiology of the seizure and would establish significance of different ictal symptoms in definition of epileptic syndromes.[13] Seizure classification focuses the observer's attention on the patient's clinical symptoms during a seizure.

Physicians utilize seizure details when determining the seizure type or classification. If medication is prescribed, the selection of antiepileptic drug is based mainly on the seizure classification.[12] Seizures are classified using EEG data from recorded seizures together with clinical symptoms. For the majority of patients, EEG data are not available and the physician must classify the seizure by using information from the patient and eyewitness accounts from an observer. Therefore, complete and accurate details about a seizure have great implications for seizure classification and choice of treatment.

Details observed during a seizure provide data for classifying the event and for evaluating treatment and seizure severity. Chee et al. established the usefulness of lateralizing signs for classifying partial seizures with secondary generalization.[4] The details included signs such as head turning, neck extension, and mouth deviation. They found that these small observational details may be the only lateralizing signs for classifying the seizure properly, because the EEG can often be obscured by movement and muscle artifact. Bodensteiner et al., in a retrospective study of 2219 seizure documents, found that the amount of details varied widely among physician seizure descriptions, resulting in 22-51% of seizures that were unclassifiable due to insufficient detail.[3] Duncan and Sander developed a scale that would identify seizure severity rather than just seizure numbers.[9] This scale included specific details as a means to score the severity of seizures.[9] This could be useful in evaluating treatment efficacy.

The nurse's role in seizure observation is to provide specific and intricate descriptions of a seizure or nonepileptic event so the physician can establish a proper diagnosis and seizure classification to form the basis of treatment.[1,2,16,17] This role could apply to anyone observing a seizure.

Factors inherent in seizures themselves may affect the quality and accuracy of the observation and documentation. The same type of seizure may manifest different symptoms between patients, so although some similarities exist, much variety is possible. Seizures occur unpredictably and may startle or surprise the observer at the onset. Frequently, observers may not see the actual beginning of the seizure and may not begin observation until the seizure progresses. If an observer is unaware of this, the details of the seizure may unknowingly be incomplete.[1] The need to provide seizure first aid can shift attention away from the observation task, thus affecting the accuracy of the observation.

Among observational research, the most important factors identified to optimize observer performance, accuracy, and consistency over time are care in initial selection of observers, training, and periodic retraining about definitions, techniques, and documentation format.[5,7,10] Observer training using videotape allows practice of the actual experience, comparisons and criterion scoring, and convenience and flexibility.

Observational data may be recorded inaccurately due to several factors. Scores of observed data will differ if observers are trained by different methods. Scores can increase due to expectation bias if the observer can anticipate what she or he will be watching or can anticipate the researcher's expectations about certain target behaviors.[11] Also, if the observer is aware of and anticipates the scoring or rating system, this form of "agreement checking" can increase scores.[11]

After observers have been trained, scores may decrease over time as they change the manner in which they apply definitions or develop their own system for observation and documentation.[11] Bodensteiner et al. analyzed the seizure classification process using pairs of neurologists and residents who reviewed the same seizure descriptions.[3] They found that more experienced neurologists were more likely to rely on their own judgment and less apt to closely follow classification criteria than the less experienced residents.[3]

Accuracy of scoring decreases as the complexity of coding increases and the task becomes more difficult for the observer. Observers tend to overrecord or have increased scores for negatively charged behaviors compared to positive or neutral behaviors. Finally, scores will vary if observers have opportunities to share their observations or confer or cue in any way with other observers' recording.[11]

In summary, the literature addressed the importance and difficulties of obtaining accurate seizure observation details. Methods for improving observer scoring and factors that can cause inaccuracy were also described. While much of this review helped elucidate the nature of the problem and what could be done to minimize variability, some questions remained unanswered. Would performance of seizure observation and documentation be affected by differences among the observers or factors inherent in seizures themselves?

The purpose of this study was to investigate the effect of observer characteristics and seizure attributes on seizure observation and documentation skills. This study had the following questions:

1. Is there a correlation between the detail and quality of seizure observation and documentation among observer characteristics such as educational level, professional versus nonprofessional position, years of experience, clinical experience with epilepsy, and frequency to practice seizure observation and documentation skills?

2. Is there a correlation between the detail and quality of seizure documentation and different seizure types?

3. What were the most common details documented for a given seizure type?


This study was conducted at a children's hospital with an epilepsy center. Human subjects' approval was obtained for the study. The study was discussed with all nursing department staff members. Participants were 60 volunteers, including RNs, CNAs licensed practical nurses (LPNs), and recreation therapists (RTs) from the main patient care units and the float staff. All participants had mandatory seizure observation and documentation training within the last two years and the clinical responsibility to observe and document any seizures they witnessed. On-call staff members were excluded from the study because of inconsistent work schedules. Two participants dropped out during the study because of change in employment. The study was completed with 58 participants each observing and documenting a total of 6 videotaped seizures, 1 seizure observed every 4 to 6 weeks, for a total of 348 seizure documents (event records).

Confidentiality of the participants was maintained by using social security numbers initially and then replacing them with randomly assigned two-digit participant code numbers for the remainder of the study. To protect participants' identity, their handwritten event records were mailed to a medical secretary outside of the institution for typing. The typed event records were compared to the original version for accuracy and completeness by a staff nurse not involved with data collection or ratting of the documents. Only the typed event records were used for scoring the observations.

Appointments for seizure observation and documentation were before or after scheduled shifts of work or at other designated times according to participant preference and convenience. The study site was a classroom adjacent to the inpatient units.

The study design was a situational test approach utilizing videotape of five seizures and one nonepileptic event. Upon recruitment for the study, participants gave written consent and completed a background information sheet. Each observation session included the following:

* Completion of data collection about seizure observation and documentation frequency over the past month

* Observation of a videotaped seizure shown once

* Documentation of the observation on an event record

Participants were instructed they could take as much time as they needed to complete the documentation and were asked not to discuss their observations with anyone else. The longest testing session lasted 15 minutes. Participants were not informed about the seizure events to be witnessed or about the rating system for scoring the documentation.

Each participant observed the same six seizure events. However, to minimize problems with higher scores due to a practice effect with later observations and to minimize possible contamination if participants violated the study protocol and talked about their observations with other participants, a randomized seizure event sequence was developed for each participant.

One seizure (long-duration complex partial seizure) was shown twice to all participants. This seizure was shown once in the beginning and then repeated toward the end of the sequence. Whether this seizure was shown first or second or repeated fifth or sixth in the sequence was randomized among the participants. The repeated seizure score would be used to examine intraparticipant variability and practice effect.

Two raters evaluated and scored the event records--the investigator and one research assistant. At six evenly spaced intervals during the study, both raters independently scored 10 records and compared scores for interrater reliability. Minimum score comparisons were set at 80%. A t-test was used to compare scores between the raters, and this showed no significant difference in the two raters' scores for a random sample of 60 records. Correlation coefficient r was 0.932, or a prediction of 86.49%.

Instruments and Testing Materials

Materials developed for the study were a seizure event videotape, background information sheet, observation/documentation practice frequency form, event record, and event record score sheet.

Seizure Event Videotape

A videotape of five seizures (one shown twice) and one nonepileptic event was compiled by using videoelectroencephalography (video EEG) archives from another hospital. This would eliminate possible recognition of a given seizure from the hospital's own video EEG archives for staff members who had clinical epilepsy experience. The criteria used by the investigator to select seizures for the study included videotape quality, quantity of observational details, and variety. Written permission was obtained from the patients to use these video EEG archives for research purposes. No identifying information, seizure classification, or histories about the patients were given to the volunteer participants.

The videotape included the following types of seizures that had been previously classified: simple partial, short-duration complex partial, long duration complex partial, partial seizure with secondary generalization, and a nonepileptic event. The nonepileptic event was included in the study, because these were often observed in the clinical setting. This event consisted of lost responsiveness and alternating clonic movements. The long duration complex partial seizure was the seizure shown twice to all participants.

Background Information

Each participant completed a background information sheet at the beginning of the study. This included data on educational background, employment position, nursing unit assignment, and length of employment.

Observation/Documentation Practice Frequency

The data included the participant's clinical experience with patients having epilepsy in the past month, frequency of direct care for patients having epilepsy in the past month, frequency of seizure observation and documentation in the past month, and an approximate count of seizures observed in the past 2 weeks of work. The data were collected from the participants just before each seizure event observation.

Event Record

The event record (Fig 1) used by the participants to document their observations, was a formalized version of what was used in clinical practice at the institution to document seizures. Participants were instructed to document a chronological description of the event observed using observation guidelines available on the form.


Event Record Score Sheet

An event record score sheet (Fig 2) was developed using the observation guidelines as the framework. A numerical rating scale was added for the quality of each detail: satisfactory (2), needs improvement (1), and not described (0). One master event record score sheet was developed for each seizure event. Each master score sheet included details of the seizure event to facilitate rating each record. Thus, a total score was based on both detail and quality found in each seizure event record document.


Three experts (two physicians, one clinical nurse specialist) reviewed each seizure event to document all possible details that could be observed from the videotape. Then a master score sheet was developed for each seizure event to facilitate scoring.

A pilot scoring trial was conducted to evaluate the event record score sheets. Two raters scored 20 event records for each seizure event. Results between the raters were compared and score sheets revised. A second trial documented consistent scoring with interrater reliability of better than 80%.

Statistical Analyses

Descriptive statistics were used to characterize the sample. Total mean seizure rating scores were used to compare educational background, employment position, nursing unit assignment, and length of employment among the participants using one way analysis of variance. Multivariate analysis with multiple regressions was used to further analyze combinations of observer variables. A two sample t-test was used to evaluate the two largest groups of participants (RN and CNA) according to experience with epilepsy. It was not feasible to compare LPNs and RTs because of the very small numbers of these participants. Mean percentile scores were calculated for each seizure event and compared using one way analysis of variance. Mean scores for each observer were used to analyze the repeated seizure using a paired t-test. The level of statistical significance for the study was set at p [is less than] .05.


Mean seizure rating scores did not vary significantly when the following variables in the participants were compared using one-way analysis of variance: educational level, employment position, years of employment, and clinical experience with epilepsy. Table 1 identifies the numbers of participants for each observer variable, mean scores, standard deviation, standard error, and F-stat values. Employment position was analyzed for the two largest groups, RN and CNA, and scores were not significantly different. Although not statistically significant, an inverse relationship was observed between mean seizure rating scores and years of experience for all participants.

Table 1

Comparison of Observer Variables
Observer Mean Standard
Variable n Scores Deviation

Educational Level

High School or
 One-Year Technical 20 77.00 15.94
Two-Year College Diploma 14 81.57 15.89
Four-Year College Plus 24 78.66 13.68

Employment Position

Registered Nurses 38 81.23 14.40
Nursing Assistants 17 74.18 15.82

Years of Employment

1 month-2 years 18 83.38 12.90
2 years-10 years 23 80.08 14.35
10 years plus 17 72.17 15.93

Nursing Unit

Unit A Epilepsy 20 84.15 8.73
Unit B Orthopedics 17 71.17 18.93
Unit C Acute 14 80.21 15.86
Float Staff 7 79.14 9.56

Observer Standard F-Stat
Variable Error Values

Educational Level 0.3925

High School or
 One-Year Technical 3.56
Two-Year College Diploma 4.24
Four-Year College Plus 2.79

Employment Position 2.6569

Registered Nurses 2.33
Nursing Assistants 3.84

Years of Employment 2.7778

1 month-2 years 3.04
2 years-10 years 2.99
10 years plus 3.86

Nursing Unit 2.5996

Unit A Epilepsy 1.95
Unit B Orthopedics 4.59
Unit C Acute 4.24
Float Staff 3.61

Multivariate analysis with multiple regressions using combinations of observer variables was completed. Only two comparisons were statistically significant. Those participants who had observed more than 10 seizures in the clinical setting in the 2 weeks previous to the testing session had higher mean percentile test scores when observing the nonepileptic event (p [is less than] .05) (Table 2). This finding was not seen in the other types of seizure events. Registered nurses who had clinical experience with epilepsy did not show significantly higher seizure rating scores than RNs without clinical experience. However, this was not true for CNAs (Table 3). CNAs who had clinical experience with epilepsy had significantly higher mean seizure rating scores than those without epilepsy experience using a two sample t-test (p [is less than] .01).

Table 2

Comparison of Mean Seizure Rating Scores by Practice Frequency
 2-Week Mean
 Observation Scores
Type of Seizure Frequency n (%)

Long complex partial None 19 51.47
(initial observation) 1 to 9 22 56.44
 10+ 13 57.69
Long complex partial seizure None 22 63.64
(repeat observation) 1 to 9 22 60.36
 10+ 12 59.03
Short complex partial seizure None 25 63.23
 1 to 9 18 62.38
 10+ 14 68.12
Simple partial motor seizure None 29 55.18
 1 to 9 14 56.82
 10+ 12 59.09
Partial secondarily generalized
 seizure None 20 65.00
 1 to 9 22 59.08
 10+ 13 63.62
Nonepileptic event(*) None 25 65.09
 1 to 9 17 57.77
 10+ 15 70.91

(*) p < .05

Table 3

Comparison of Mean Seizure Rating Scores Based on Clinical Experience
 Mean q Standard
Clinical Experience n Scores Deviation Error
 With epilepsy experience 12 84.5 9.2 2.67
 With no epilepsy experience 21 79.9 17.5 3.81

 With epilepsy experience(*) 8 83.6 8.5 2.99
 With no epilepsy experience 9 65.8 16.4 5.46

(*) p < 0.01

Mean seizure rating scores for the seizure that was shown twice were significantly higher when repeated, compared to the initial observation using a paired t-test (p [is less than] .001). This result was not related to whether the initial observation was observed first or second in the sequence and the repeated observation was fifth or sixth in the sequence or any other variables.

Seizure Attributes

There were significant differences (p [is less than] .001) in the mean percentile seizure rating scores for different types of seizures using a one-way analysis of variance (Table 4). These differences were not related to the duration of the seizure. Although this finding was not quantifiable, seizures that had more dramatic movement had higher mean percentile seizure rating scores and seizures with the least amount of motor activity had the lowest mean percentile seizure rating scores.

Table 4

Comparison of Mean Seizure Rating Scores by Seizure Type(*)
 Duration Mean Score
Type of Seizure (seconds) (%)

Nonepileptic event 52 64.34
Short complex partial seizure 20 64.32
Partial secondarily generalized seizure 55 61.94
Simple partial motor seizure 12 55.96
Long complex partial seizure 120 54.17

(*) n = 58, p < .001 secondarily

Most Common Seizure Details Documented

The most frequent details documented for all seizure events were the location and description of movement. The relative frequency of documented details for each seizure event are shown in Table 5.

Table 5

Relative Frequency of Documented Details for Seizure Events
 Relative Frequency

 Simple Short
 Partial Complex
Response Motor Partial
Category Description Seizure Seizure

 A Circumstances before seizure onset 52 84
 B Patient's description of aura 0 0

 C Description of appearance at onset 26 50
 D Location of movement 100 98
 E Description of movement 100 98
 F One or both sides of body involved 55 64
 G Rhythmic or nonrhythmic movement 18 0
 H Symmetrical or asymmetrical
 appearance 0 50
 I Response to name 0 0
 J Response to command/interactions 0 0
 K Response to visual stimuli 0 0
 L Response to tactile stimuli 0 0
 M Patient's verbalization during
 seizure 0 0
 N Duration of seizure 98 100
 O Duration of seizure portions 5 0

 P Resumes activity 51 18
 Q Response of patient to
 commands/interaction 0 83
 R Appropriate/inappropriate responses 0 72
 S Patient's recall of seizure 0 28
 T Post-ictal duration 0 0

 Long Complex Partial
Description First Time Repeat

Circumstances before seizure onset 41 55
Patient's description of aura 48 62

Description of appearance at onset 97 98
Location of movement 100 100
Description of movement 100 100
One or both sides of body involved 24 60
Rhythmic or nonrhythmic movement 37 22
Symmetrical or asymmetrical
 appearance 0 0
Response to name 19 19
Response to command/interactions 67 71
Response to visual stimuli 41 37
Response to tactile stimuli 41 58
Patient's verbalization during
 seizure 88 86
Duration of seizure 92 97
Duration of seizure portions 0 0

Resumes activity 0 0
Response of patient to
 commands/interaction 72 67
Appropriate/inappropriate responses 72 71
Patient's recall of seizure 0 0
Post-ictal duration 28 28

 Generalize Nonepileptic
Description Seizure Event

Circumstances before seizure onset 81 58
Patient's description of aura 0 0

Description of appearance at onset 76 18
Location of movement 98 100
Description of movement 98 100
One or both sides of body involved 98 98
Rhythmic or nonrhythmic movement 33 41
Symmetrical or asymmetrical
 appearance 45 45
Response to name 0 0
Response to command/interactions 55 0
Response to visual stimuli 0 0
Response to tactile stimuli 0 0
Patient's verbalization during
 seizure 0 0
Duration of seizure 98 95
Duration of seizure portions 12 0

Resumes activity 48 37
Response of patient to
 commands/interaction 0 0
Appropriate/inappropriate responses 0 0
Patient's recall of seizure 0 0
Post-ictal duration 0 0

Incidental Findings

Omission errors (missing details) were expected by the investigator, because observers are not able to observe or recall all the details during an observation. Commission errors (documentation of details not present on videotape) were not expected by the investigator yet were found in some of the documents (Table 6). There was no correlation between commission errors and seizure duration.

Table 6

Commission Errors in Seizure Documentation(*)
 Observers with Observers
Type of Seizure 1 Error with 2 Errors

Long complex partial seizure 14 4
Simple partial seizure 5 1
Repeat long complex partial seizure 4 0
Short complex partial seizure 3 0
Nonepileptic event 2 0
Partial secondarily generalized
 seizure 0 0

 Percentage Duration
 of Observers of Seizure
Type of Seizure with Errors (seconds)

Long complex partial seizure 31% 120
Simple partial seizure 10% 12
Repeat long complex partial seizure 7% 120
Short complex partial seizure 5% 20
Nonepileptic event 3% 52
Partial secondarily generalized
 seizure 0 55

(*) n = 58


All participants had the same training and certification in seizure observation and documentation skills prior to study enrollment. According to the results, differences in education, employment position, and years of experience are not limiting factors in performing seizure observation and documentation once persons have received basic instruction and certification in seizure observation and documentation skills.

Participants who had performed seizure observation skills more than 10 times in the previous 2 weeks before an observation session had significantly higher seizure rating scores when observing the nonepileptic event. It is difficult to understand why this occurred for one seizure type and not for others. The meaning of this finding is uncertain. Perhaps a more experienced observer could more easily discern the details that occurred in this nonepileptic event where there were side-to-side lateral head movements and alternating see-saw flexion extension movements in the extremities.

RNs having clinical experience with epilepsy had higher mean seizure rating scores than those without epilepsy experience; however, this did not reach statistical significance. The CNAs who had clinical experience with epilepsy had higher mean seizure rating scores than those without experience (p [is less than] .01). At this institution, the CNAs generally had more practice experience with seizure observation because they were frequently assigned to care for one patient with epilepsy. This indicates the importance for nursing staff members to provide CNAs consistent assignments to patients with epilepsy. Based upon this study, CNAs continued to perform seizure observation and documentation skills as part of their job. This CNA role may represent a significant cost savings to an institution compared with 1:1 RN staffing for those patients who had frequent daily seizures.

Significantly higher mean seizure rating scores were seen when the long duration complex partial seizure was observed a second time. The increase in mean scores most likely represents improvement in seizure observation and documentation skills over time because of a practice effect. The higher score was unlikely to be related to familiarity with this seizure because it had been 3 to 5 months since it was first observed. Also, after the study ended, participants were asked whether they thought one of the seizures was seen twice and only a few participants recalled this.

In a clinical setting, repeated observations of a patient's seizures provide an opportunity to increase the detail and quality of both observation and documentation. For example, if the observer did not determine the patient's responsiveness to stimuli when observing the initial seizure, then observing the patient having another seizure may allow for additional interaction, observation, and documentation of this detail.

Statistically significant differences were seen in mean percentile seizure rating scores among the different seizure types, but this finding did not correlate with seizure duration. This may indicate that trained observers do not observe all seizure events in the same way. Although this was not quantifiable, the seizures that involved more dramatic movement had higher mean scores compared to those with the least amount of motor activity.

Seizures with subtle symptoms and less dramatic movements may have a less recognizable seizure onset compared to those with dramatic movements. Perhaps more dramatic movements captured the observer's attention and made the seizure easier to recall when it was documented.

The most frequently documented details for all videotaped seizure events were the location and description of movement. Why did this occur? It may be that describing movement is easier to remember and document than the responsiveness of a patient to stimuli. Research literature on observational methods report that as coding complexity or amount of details observed increases, accuracy decreases.[10,11,15]

Both omission and commission errors may occur and affect the accuracy and details of seizure documentation. These types of errors need to be considered by physicians, clinical nurse specialists, nurse practitioners, and other care providers who review seizure documents, because they have implications for seizure classification, drug treatment, and evaluation of effectiveness of therapy.


This study may have limited direct application for observers in settings in which there is no consistent systematic staff training or certification in this skill. Details gathered about the seizure events that were part of this study may not pertain to other types of seizures. Observing videotaped seizure events were necessary for the study. However, this method did not allow the observer to move and interact with the patient, which is an important technique in seizure observation.

The videotaped seizure events did show staff members creatively interacting with the patient, and this may have been instructive to some observers. For example, a variety of tactile and visual stimuli to different parts of the body were used to determine responsiveness, rather than calling the person's name repeatedly. Descriptions of a patient who cannot answer a question or respond to his name yet can follow a simple command with the left side may help the physician classify the seizure as partial due to altered consciousness rather than generalized with loss of consciousness.

The two largest groups of participants analyzed were the CNAs (n = 17) and RNs (n = 33). The small number of CNA participants is a limitation in the study. However, each person was scored on 6 documents, providing a total of 102 documents for CNAs and 198 documents for RNs whose scores were compared.


Observing a seizure is a unique skill requiring, as a final result, documentation of details rather than a classification, analysis, or summary. In contrast, for many other skills, the product is a summary or synthesis of the details that have been gathered about a phenomenon. This distinction has relevance for training and educating staff or observers about the skill of seizure observation and documentation.

Instruction should include initial and periodic retraining on seizure observation and documentation skills. Instruction should also emphasize to observers that they identify key observational details because these can influence seizure classification and the selection and evaluation of treatment. Instruction should assist the observer in developing an increased awareness that, with experience, observers are more likely to rely on their own judgment rather than observational criteria or details. Instructors should strongly encourage observers to resist classifying the seizure and to increase the breadth and depth of details they observe and recall for documentation. Perhaps less emphasis should be placed on the seizure classification system during observer training.

More seizure observation and documentation training time should be devoted to less frequently observed, yet important aspects such as determining a person's responses to sounds, sights, touch and commands, perceptual abilities, circumstances surrounding the seizure, and postictal responses. Because seizure movements often capture the observer's attention, other subtle details at the beginning of the seizure, such as mouth, head, or eye deviation, may be overlooked. These details can be critical in distinguishing a partial from a primary generalized seizure, thus determining the course of treatment for the patient.

The seizure observation guidelines can be a helpful tool for staff members. Fig 3 shows details that should be considered for seizure observation and documentation. This information can easily be taught to persons who might need to observe seizures outside of an inpatient setting.

Describing the Beginning

1. What were the circumstances?

2. Were there any precipitating factors?

3. Was the seizure onset observed?

4. Did the person state or give any indication that the seizure was beginning? (An aura is the beginning of the seizure onset)

5. Did they attempt to continue, stop or slow down in their activities?

6. What happened first? Then in order, describe how the rest of the seizure proceeded.

Assessing Responsiveness

1. Describe patient's observed response to you, self, environment.

2. Describe whether responses were rote or more complex and how much they were affected (e.g. partially, completely)

3. Assess response to tactile stimuli (light touch, tickle, blow on face, a mild shake of an extremity, attempt to open eyes or move extremities, ice, etc.).

4. Assess response to auditory stimuli (clap hands, call name, give a command, state a word and ask for recall later).

5. Assess response to visual stimuli (note visual flinch, throw an object to person unexpectedly). Check pupil reactions.

Assessing Movements, Mobility or Tone

1. Was there any movement or change in posture? Give location and description. Be as specific as possible. Consider a head-to-toe approach.

2. Was there any movement of the head or eyes at the beginning of the seizure?

3. Did this affect one or both sides of the body? If both sides involved, did they look the same or different?

4. Assess whether tone is increased (tonic, spastic, rigid), decreased (flaccid or limp), or normal.

5. Were there any automatisms (repetitive, purposeless movements)? Were there any purposeful movements?

Assessing Sensation and Perception

1. What does the person describe or state? Ask them during or after seizure for a detailed description if possible.

2. Are there any autonomic signs and symptoms (e.g., skin temperature change, change in color, sweating)?

3. Did they say or do anything unusual (mumbling, speaking inappropriately, cursing, wandering, climbing up or under objects, fumbling, resisting or combating touch, agitation, etc?

Assessing Postictal Responses

1. What were they like after the seizure? Describe their behavior.

2. How long did it take before resumption of previous activities?

3. Could they recall the event in general or remember what happened at the beginning or throughout the seizure?

4. Were there any temporary deficits (memory loss, aphasia, paresis)?

5. Was there any confusion or disorientation? Describe and give duration.

6. Was there any bowel or bladder incontinence?

Adapted from Wolf J: Page 1751 in: Family Centered Nursing Care of Children, 2nd ed, CL Betz, M Hunsberger, S Wright (editors). WB Saunders, 1999.

Fig 3. Seizure observation and documentation.

The use of specific criteria for categorizing seizure symptoms has been shown to greatly improve reliability.[8] Seizure guidelines and narrative charting may allow more details to be documented than a seizure flowsheet with fixed choices of details to code or check off. The chronological evolution of seizure details can be lost in a seizure flowsheet.

The event record score sheet used in this study may be a potential tool for future studies of seizure observation and documentation skills. It may have utility as a tool to help individuals examine patterns in their seizure observation and documentation skills. For example, frequent omission of documenting a patient's postictal responses by an observer may be more obvious when a score sheet is used. This feedback to the observer would identify areas that require more work.

Use of videotaped seizures in this study allowed the investigator to use criteria for scoring and eliminated many conflicting variables that would have otherwise made the study impossible. Periodic observations of videotaped seizures may provide a useful method for assisting staff members in improving their skills. However, observing videotaped seizures should not imply that the process of seizure observation is simply a passive observation. Good seizure observation is a creative, interactive process. Much can be learned by interacting with the patient and observing responses to stimuli during a seizure. The Seizure Assessment guidelines published by the American Association of Neuroscience Nurses can be a resource for more information on useful techniques.[1]


Seizure observation and documentation skills are an integral component of neuroscience nursing. This study evaluated seizure documents of trained observers to determine how observer characteristics and seizure attributes might influence the skill of seizure observation and documentation. Recognition of problem areas and skill training are important considerations in providing complete and accurate seizure documentation.


[1.] American Association of Neuroscience Nurses: Seizure Assessment Clinical Guideline Series. Author, 1998.

[2.] Barry K, Teixira S: The role of the nurse in the classification and management of epileptic seizures. J Neurosurg Nurs 1983; 15(4):243-249.

[3.] Bodensteiner JG, Brownsworth RD, Knapik JR et al: Inter-observer variability in the ILAE classification of seizures in childhood. Epilepsia 1988; 29(2):123-128.

[4.] Chee MWL, Kotogal P, VanNess PC et al: Lateralizing signs in intractable partial epilepsy: Blinded multiple-observer analysis. Neurology 1993; 43:2519-2525.

[5.] Cipani E, McLaughlin T: Procedural analysis for conducting observer training for observational research. Corrective and Social Psychiatry and Journal of Behavior Technology, Methods and Therapy 1981; 27(3):125-129.

[6.] Commission on Classification and Terminology of the International League Against Epilepsy: Proposal for revised clinical and electroencephalographic classification of epileptic seizures. Epilepsia 1981; 22:489-501.

[7.] Cone JD: Issues in functional analysis in behavioral assessment. Behav Res Ther 1997; 35(3):259-275.

[8.] Disabato J, Wulf JA: Altered neurologic function. Pages 1748-1754 in: Family-Centered Nursing Care of Children, 2nd ed, Betz CL, Hunsberger M, Wright S (editors). WB Saunders, 1994.

[9.] Duncan JS, Sander JWAS: The Chalfont seizure severity scale. J Neurol Neurosurg Psych 1991; 54:873-876.

[10.] Fradenburg LA, Harrison RJ, Baer DM: The effect of some environmental factors on the interobserver agreement. Res Dev Disabil 1995; 16(6):425-437.

[11.] Harris C, Lahey S: Recording system bias in direct observational methodology: A review and critical analysis of factors causing inaccurate coding behavior. Clin Psychol Rev 1982; 2(4):539-556.

[12.] Leppik IE: Contemporary Diagnosis and Management of the Patient with Epilepsy, 3rd ed. Handbooks in Health Care, 1997.

[13.] Luders H, Acharya J, Baumgartner C et al: Semiological seizure classification. Epilepsia 1998; 39(9): 1006-1013.

[14.] Penfield W, Jasper HH: Epilepsy and the Functional Anatomy of the Human Brain. Little, Brown and Co, 1954.

[15.] Polit DF, Hungler BP: Nursing Research: Principles and Methods, 5th ed. JB Lippincott Company, 1995.

[16.] Tucker C: Nursing care of patients with temporal lobe epilepsy during phases I and II on telemetry recording. J Neurosurg Nurs 1979; 11(4):227-234.

[17.] Woodward E: The total patient: Implications for nursing care of the epileptic. J Neurosurg Nurs 1982; 14(4):166-169.

Questions or comments about this article may be directed to: Judith A. Wulf, RN CNRN ANP-C, Medical Advanced Pain Specialist, 2104 Northdale Boulevard NW, Suite 220, Minneapolis, MN 55433, e-mail She is a nurse practioner.
COPYRIGHT 2000 American Association of Neuroscience Nurses
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2000 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Wulf, Judith A.
Publication:Journal of Neuroscience Nursing
Geographic Code:1USA
Date:Feb 1, 2000
Previous Article:Living in Limbo: The Low-Level Brain-Injured Patient and the Patient's Family.
Next Article:`What Happened to Me': Rural African American Elders' Experiences of Stroke.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters