Anxious patients respond differently than nonanxious patients to anesthesia. The insertion of an intravenous catheter in the preoperative phase can be a difficult task as a result of anxiety-related vasoconstriction [1,2]. In anxious patients, larger doses of anesthetics are required to induce anesthesia [3,4] and the anesthesia itself may be associated with autonomic fluctuations [5,6]. Although the exact percentage of patients who are anxious preoperatively is not known, the literature suggests an incidence between 60% and 80% [7-10]. Although a sedative drug is often given as premedication to relieve fear and anxiety, anxious patients might also benefit from more attention and information from the anesthesiologist. In clinical practice, however, the anesthesiologist has very little time for preoperative consultation to identify the patients who are anxious and may benefit from extra attention.
There are many instruments for measuring the patient's level of preoperative anxiety [11,12]. The instrument most commonly used is Spielberger's State-Trait Anxiety Inventory (STAI) , which has been translated into Dutch and validated by van der Ploeg et al. . He also obtained norms for the Dutch population. The questionnaire consists of two separate, 20-item, self-report rating scales for measuring trait and state anxiety. The trait anxiety is a relatively stable personality disposition, while state anxiety is the situation-related anxiety and this may differ depending on the stress of the particular moment. The state scale is recommended for measuring patient anxiety in the preoperative phase  and has been used in several anesthesiologic studies [16-19]. Although this questionnaire is fairly short, it is still too long for use in busy outpatient clinics. Moreover, the questions are not related to the specific situation with which the patient is confronted.
A second aspect of preoperative care is the patient's need for information. Several studies [20-22] have shown that information given to patients before surgery may facilitate recovery. However, some patients like to shut themselves off from information, whereas others want to be informed as fully as possible . These different coping styles are almost never honored, as it is practically impossible for the anesthesiologist to discriminate between patients who would like to be informed as fully as possible from those who want to know as little as possible. It would be greatly beneficial to clinical practice if anesthesiologists knew whether they were dealing with a patient who wanted more than basic information which is routinely given, or a patient who would rather not be given any extra information.
As we could not find a clinically applicable instrument in the literature which fulfilled all our requirements, i.e., short, specifically attuned to the preoperative situation, and easy to interpret, we decided to develop a new instrument. Our point of reference was the work of Miller and Mangan [24,25], who studied the way patients cope with the stress of a threatening situation. They differentiated between "monitors" and "blunters," defining monitors as people who want to know as much as possible and search actively for information and blunters as those who have no need for information and even try to avoid it. This instrument should make it possible to distinguish anxious from nonanxious patients and patients who want information from those who do not.
A six-item questionnaire, the Amsterdam Preoperative Anxiety and Information Scale (APAIS) Table 1 was developed in a previous study1, covering both the "monitor" and "blunting" aspects. Four items represented fear of anesthesia and fear of the surgical procedure (Cronbach's alpha 0.86). Two items represented the need for information (Cronbach's alpha 0.72). The internal consistencies of both scales were sufficient for group comparison.
1 Moerman N, Dam van F, Boulogne-Abraham T, Hooff van M. The patient's need for information in the preoperative period. Proceedings of the 9th European Congress of Anaesthesiology. Jerusalem, Israel, 1994:257.
During a period of 3 mo, 320 consecutive patients visiting the anesthesiology outpatient department (patients who could not speak Dutch were excluded) were asked by the nursing staff to fill out this questionnaire Table 1. We noted the age and sex of the patients and whether they had had surgery previously. To ascertain that our instrument really measured anxiety, the last 200 patients were also asked to fill out the State version of Spielberger's STAI. This questionnaire consists of a 20-item self-report rating scale for measuring state anxiety. In the latter group we also examined the kind of procedure involved and the duration of the operation. We classified the operations as minor, intermediate, or major. "Minor" was defined as less invasive surgery of limited duration (minor orthopedic surgery, diagnostic procedures, arthroscopies, laparoscopies, inguinal hernia). Operations classified as "intermediate" had more impact for the patient (cholecystectomy, hysterectomy), and "major" were extensive operations with a high impact (laryngectomy, reconstructive and transplantation surgery).
Validity. To evaluate the validity of the APAIS Table 1, we performed several analyses. Attention was devoted to some aspects of construct (content) validity and criterion validity, too.
Construct validity was evaluated by factor analysis. Factor analysis is a statistical approach to reduce data by determining the relationships among variables and to determine the underlying structure which is formed by latent variables known as factors. The relation between variables and a certain factor is given by the so-called factor loadings, which indicate how much weight is assigned to each factor. Variables with high loadings for a factor are closely related to that particular factor. Rotation is the procedure used to make the factor solution more interpretable . The results of the factor analysis should reflect the concepts we put into our scales and should thus concur with the results of our previous study, i.e., two factors should emerge: anxiety and a need for information.
As a measure of concurrent validity we determined the correlation of the APAIS with the STAI. We hypothesized that the State version of the STAI should correlate highly (>0.60) with the anxiety scale of the APAIS and should have a low correlation (<0.30) with the need-for-information scale.
For clinical use it is important to be able to identify those patients who can be considered as "anxiety cases." For this purpose we used Spielberger's STAI as the "gold standard." Auerbach  divided a group of surgical patients on the basis of their preoperative score on Spielberger's trait anxiety scale into a high-trait-anxiety group and a low-trait-anxiety group. The mean state anxiety score of the high-trait-anxiety group was 46. We used this score on the state scale as a reference point and considered patients with a score >or=to46 on the STAI-State as anxiety cases. Furthermore, this point concurs >or=to the 9th decile of a Dutch male reference group and >or=to the 8th decile of a Dutch female reference group . We determined for different cut-off points on the APAIS anxiety scale the sensitivity (proportion of correctly identified cases), the specificity (proportion of correctly identified noncases), and the positive predictive value (probability of a high scale score being a case) in relation to the STAI.
The validity was further evaluated by known-group comparison in three different ways. 1) We hypothesized that women should have a higher score than men on the APAIS anxiety scale. This hypothesis was based on data from the literature, where women are usually regarded as being more anxious than men [7,10,14,19,28]. 2) From the work of Miller and Mangan [24,25] it is known, that high monitors are also anxious people. In other words, in a threatening situation monitoring is mostly associated with higher anxiety and arousal than blunting . We therefore hypothesized that, in our instrument, patients with a high information requirement should have a higher score on the anxiety scale than patients with a low information requirement. 3) The effect of preanesthetic information is less valuable for patients who have previous anesthetic experience than for those who do not . We therefore hypothesized that patients with previous experience of anesthesia and surgery should have a lower information requirement than those who had never had surgery. No specific hypothesis was formulated regarding the difference between men and women with respect to their information requirement. Data were analyzed using the SPSS version 4.0. An analysis of variance (ANOVA) was used for group comparison. Student's t-tests were used to compare the mean scale scores for the subgroups at baseline. Statistical significance was considered at P < 0.05.
Reliability. Cronbach's alpha s were calculated as a measure for internal consistency of the scales. Reliability was considered acceptable when Cronbach's alpha s were >or=to70 .
Of the 322 patients who were asked to participate, 2 patients refused. Patient characteristics are presented in Table 2. Patients had no problem completing the APAIS and usually did so in less than 2 min. As was predicted, we found in a factor analysis with oblique rotation (see Table 3) two factors, which explained 72% of the variance: anxiety (questions 1, 2, 4, and 5, Table 1) and the need for information (questions 3 and 6, Table 1). The correlation between both factors was 0.31. The following step was to convert the two factors to scales and calculate Cronbach's alpha for the two scales separately. Cronbach's alpha for the four anxiety items (questions 1, 2, 4, 5) was 0.86. Cronbach's alpha for the need-for-information items (questions 3 and 6) was somewhat lower (0.68), as was to be expected with a scale consisting of only two items, but still sufficient for group comparisons.
Concurrent validity was determined by the correlation with the STAI. The correlation between the anxiety items of the APAIS and the STAI-State was high (0.74) and the correlation between the information items and the STAI-State was low (0.16).
The anxiety scale consists of four items (questions 1, 2, 4, 5), each of which could be scored from 1 to 5. The score of the anxiety scale is the sum of these four questions, with a scoring range from 4 to 20. There was a highly significant difference (P = <0.001) between men and women. The mean score of men was 7.5 (SD 3.5) and the mean score of women was 9.9 (SD 4.5). But an ANOVA indicated an interaction effect between previous experience of surgery and gender (P = 0.02); t-test for differences between means showed that men who had been operated on before had a lower score on the anxiety scale than those without previous experience of surgery. In women there was no such difference between those who had previous experience of surgery and those who had not Table 4.
There were no statistically significant relationships between age, type of operation, and the scores on the anxiety scale. As the number of patients who underwent major surgery was low, the results regarding the type of operation has to be interpreted with caution.
The need-for-information scale consists of two items (questions 3 and 6), each of which could be scored from 1 to 5. The sum of the need-for-information scale is the sum of these two questions, with a scoring range from 2 to 10. ANOVA indicated no interaction effect between type of operation and gender; only a statistically significant main effect for previous experience of surgery existed (P = 0.002). Patients with previous experience had a lower score (mean 6.6, SD 2.3) on the information scale, than those who had not been operated on before (mean 7.5, SD 2.2). There were no statistically significant relationships between age, type of operation, and the scores on the information scale.
In order to investigate whether there was a relationship between the patient's need for information and the level of anxiety, we divided patients according to their score on the information scale into three groups. Patients with a score of 2-4 on the information scale can be classified as having no or little information requirement and can be considered as blunters. Patients with a score of 5-7 can be classified as having an average information requirement, and those with a score of 8-10 as having a high information requirement. The latter can therefore be considered "monitors." Using this three-group classification for the information scale, it turned out that in the population of 320 patients the percentages of patients with low, medium, and high information requirements were 16.9%, 39.7%, and 43.4%, respectively. After correction for sex and experience with previous operations, patients with a high information requirement turned out to be the ones who were most anxious Table 5.
We also investigated whether we could use the APAIS for detecting "anxiety cases." We wanted to know at what score on the APAIS anxiety scale patients could be considered anxious and would therefore benefit from more attention. As mentioned in Methods, we used the STAI-State as a "gold standard" and chose the score of 46 as a reference point. Using this reference point, the sensitivity, specificity, and the predictive value were calculated at different cutoff points on the anxiety scale (APAIS). Table 6 shows that the cutoff point of 11 leads to a good balance. Sensitivity and specificity are good and the predictive value is 71%. At the score of 11, 37 patients are misclassified (18 false-positives and 19 false-negatives). This means that 9% (n = 18) of the patients are anxious on the anxiety scale (APAIS) but not on the STAI-State, and 9.5% (n = 19) are not anxious on the APAIS although they are on the STAI-State. At the cutoff score of 10 sensitivity increases but, because of the lower specificity, the predictive value is lower, resulting in a higher number of false-positive patients (anxious on the APAIS but not on the STAI) than at the score of 11. At the scores >11 sensitivity decreases and specificity increases. The cutoff scores 11-13 produce approximately the same amount of misclassified patients (false-positives and false-negatives together), varying from 34 to 39. However, an increasing predictive value reduces the number of false-positive patients.
Using the score of 46 on the STAI-State as a reference point, the prevalence of anxiety cases in the population (n = 200) was 32%.
The purpose of the study was to develop a screening instrument for use in the preoperative period. For this reason, a six-item questionnaire was developed: the Amsterdam Preoperative Anxiety and Information Scale (APAIS). The APAIS was easily and very quickly completed by patients. Two clear factors emerged: anxiety and information requirements. The anxiety scale correlated highly with the standard questionnaire for measuring anxiety: Spielberger's STAI-State (0.74). Both the anxiety and the need-for-information scale showed good psychometric properties and were feasible in clinical practice. According to the literature we found that 1) women have a higher score on the anxiety items than men, and 2) there is a positive relationship between anxiety and information requirement. Patients with a greater need for information were patients with a higher anxiety level than those with a low information demand. 3) Patients without previous experience of surgery had a higher information requirement than those who had been operated on before.
As already mentioned, there was a difference in anxiety levels between men and women. Women scored higher on the anxiety scale than men. It was, however, striking that men who had not been operated on before were just as anxious as women. However, men who had undergone previous surgery scored lower on the anxiety scale than men who had not. In other words men who have been operated on before cope differently with their fear of anesthesia and surgery than women. This finding warrants further research.
In contrast to our expectations, it emerged that the questionnaire did not distinguish well between fear of anesthesia and fear related to surgery, which means that feelings of anxiety in the preoperative period are diffuse and are not really focused on either surgery or anesthesia. In this respect, the STAI-State should be a good method for measuring preoperative anxiety, as noted by Spielberger et al. . A great advantage of our questionnaire, however, is that it is much shorter. The anxiety scale of the APAIS consists of only four questions, while in the STAI-State 20 questions must be answered. Because the APAIS is specifically attuned to the preoperative situation, patients can complete it without further explanation.
The APAIS can be used for clinical practice and for research purposes. The scores on the anxiety scale of the APAIS range from 4 (not anxious) to 20 (highly anxious). The cutoff points chosen depend on the purpose for which the scale is to be used, i.e., clinical use or research purpose. Based on a comparison with the STAI as a gold standard, it is clear from the results that for clinical practice the cutoff score of 11 produces a good predictive value with an acceptable balance between false-positive and false-negative patients. So far, the score of 11 seems a useful and efficient score for identifying anxious patients in clinical practice. A score of 10 would result in a lower predictive value and a higher number of false-positive patients (anxious on the APAIS but not on the STAI) than the score of 11 (14.5% vs 9%). Whether the anesthesiologist will accept a score of 10 as an indication for anxiety cases and accept a relatively high number of false-positive patients, or prefers a score of 11 with a relatively low number of false-positive patients depends on the amount of time the anesthesiologist wants to devote to a patient's preoperative stress and anxiety. With scores higher than 11 the predictive value increases but because of the higher percentage of specificity the number of false-negative patients (not anxious on the APAIS, but anxious on the STAI) also increases. For the purposes of clinical practice, it is important to identify the patients who are anxious, and a high number of false-negative patients is not acceptable. On the basis of these results, we recommend for the purposes of clinical practice that patients with a score of >or=to11 on the anxiety scale should be considered as anxiety cases. Future research should be conducted to clarify whether it is useful to distinguish between anxiety cases and nonanxiety cases.
When the list is used for research purposes, the number of false-positive patients is more important. The score of 11 produces 9% false-positives, that is to say, 18 patients have a high score on the APAIS, but not on the STAI. With a score of 13 the number of false-positives decreases to 4. If the list is used for research purposes, where anxiety reduction is an important outcome criterion, we recommend a score of 13. At the score of 13 there are hardly any falsepositives (2%) and thus the database is less polluted.
The scores on the information scale of the APAIS range from 2 (no need for information) to 10 (high need for information). The results of our study show that over 80% of patients have a positive attitude toward receiving information (score >or=to 5). This Figure concurswith data from other countries [31,32]. Given the implications for daily medical practice, this is an important point of which every anesthesiologist should be aware. It also emerged that the patients with an extremely high information requirement (score >or=to 8) are anxious patients. The relationship between anxiety and information requirement has already been underlined by Janis , who was the first to conduct systematic research on preoperative anxiety. It is important to realize that anxious patients might derive great benefit from more attention and information. However, extensive information is not always useful and may even induce anxiety . Particularly patients with a "blunting" coping style may become anxious when confronted with extensive information. By contrast, patients with a monitoring coping style become anxious when they are not provided with as much information as they want [24,35]. In our population almost 17% of the patients had a negative or uninterested attitude toward information (score <or=to 4). The law requires that patients be given information but, as mentioned above, it is important to realize that not everyone wants to be fully informed. We therefore advise that patients with a score of 5 and higher should be given information on the topics about which they wish to be informed and in accordance to their score. A score below 5 should be a signal for providing no more information than is legally required.
Consideration of patient's preoperative fears and anxieties is of paramount importance in the quality of anesthesiologic care. However, devoting attention to a patient's fear takes time, and time is in short supply. Fortunately, not all patients are equally anxious and in need of additional support. We have developed a simple screening instrument which, if used during preoperative assessment, may facilitate the identification of those patients who are in need of extra support. The extent to which the APAIS will be useful in clinical practice has to be verified in future research.
The authors wish to thank Benno Bonke, PhD, and Kommer Sneeuw, MA, for their valuable and constructive comments, the nursing staff of the anesthesiologic outpatient department of the Academic Medical Centre for their assistance, Marjolein Porsius, research nurse, for collecting the data about the operations, and Marion Alhadeff for her expertise and support as a translator.
1. Thyer BA, Papsdorf JD, Davis R, Vallecorsa S. Autonomic correlates of the subjective anxiety scale. J Behav Ther Exp Psychiatry 1984;15:3-7.
2. Wallin BG. Neural control of human skin blood flow. J Auton Nerv Syst 1990;30:S185-90.
3. Williams JGL, Jones JR. Psychophysiological responses to anesthesia and operation. JAMA 1968;203:127-9.
4. Goldmann L, Ogg TW, Levey AB. Hypnosis and daycase anaesthesia: a study to reduce pre-operative anxiety and intraoperative anaesthetic requirements. Anaesthesia 1988;43:466-9.
5. Tolksdorf W, Schmollinger U, Berlin J, Rey ER. Das praoperative psychische Befinden-Zusammenhange mit anasthesierelevanten psychophysiologischen Parametern (The preoperative psychological state and its correlation to psychophysical parameters important to anaesthesia). Anasth Intensivther Notfallmed (Stuttgart) 1983;18:81-7.
6. Tolksdorf W, Berlin J, Rey ER, et al. Der praoperative Stre beta: Untersuchung zum Verhalten psychischer und physiologischer Stre beta parameter nichtpramedizierter Patienten in der praoperativen Phase (Preoperative stress: investigation of psychological and physiological stress parameters in unpremedicated patients). Anaesthesist 1984;33:212-7.
7. Norris W, Baird WLM. Preoperative anxiety: a study of the incidence and aetiology. Br J Anaesth 1967;39:503-9.
8. Ramsay MAE. A survey of pre-operative fear. Anaesthesia 1972;27:396-402.
9. Mackenzie JW. Daycase anaesthesia and anxiety: a study of anxiety profiles amongst patients attending a day bed unit. Anaesthesia 1989;44:437-40.
10. Shevde K, Panagopoulos G. A survey of 800 patients' knowledge, attitudes and concerns regarding anesthesia. Anesth Analg 1991;73:190-8.
11. Johnston M. Pre-operative emotional state and post-operative recovery. Adv Psychosom Med 1986;15:1-22.
12. Millar K, Jelecic M, Bonke B, Asbury AJ. Assessment of preoperative anxiety: comparison of measures in patients awaiting surgery for breast cancer. Br J Anaesth 1995;75:180-3.
13. Spielberger CD, Gorsuch RL, Lushene RE. State trait anxiety inventory manual. Palo Alto, CA: Consulting Psychologists Press, 1970.
14. Ploeg HM van der, Defares PB, Spielberger CD. Handleiding bij zelfbeoordelingsvragenlijst STAI-DY. Lisse-Amsterdam: Swets & Zeitlinger, 1980.
15. Spielberger CD, Auerbach SM, Wadworth AP, et al. Emotional reactions to surgery. J Consult Clin Psychol 1973;40:33-8.
16. Lanz E, Schafter M, Brunisholz V. Midazolam zur oralen Pramedikation vor Regional-Anaesthesie (Oral premedication with midazolam for local anesthesia). Anaesthesist 1987;36:197-202.
17. Antrobus JHL. Anxiety and informed consent: does anxiety influence consent for inclusion in a study of anxiolytic premedication? Anaesthesia 1988;43:267-9.
18. Arellano R, Cruise C, Chung F. Timing of the anesthetist's preoperative outpatient interview. Anesth Analg 1989;68:645-8.
19. Badner NH, Nielson WR, Munk S, et al. Preoperative anxiety: detection and contributing factors. Can J Anaesth 1990;37:444-7.
20. Suls J, Wan CK. Effects of sensory and procedural information on coping with stressful medical procedures and pain: a meta-analysis. J Consult Clin Psychol 1989;57:372-9.
21. Webber GC. Patient education: a review of the issues. Med Care 1990;28:1089-1103.
22. Devine EC. Effects of psychoeducational care for adult surgical patients: a meta-analysis of 191 studies. Patient Educ Couns 1992;19:129-42.
23. Miller SM, Combs C, Stoddard E. Information, coping and control in patients undergoing surgery and stressful medical procedures. In: Steptoe A, Appels A, eds. Stress, personal control and health. New York: Wiley, 1989;107-30.
24. Miller SM, Mangan CE. Interacting effects of information and coping style in adapting to gynecologic stress: should the doctor tell all? J Pers Soc Psychol 1983;45:223-36.
25. Miller SM. Monitoring and blunting: validation of a questionnaire to assess styles of information seeking under threat. J Pers Soc Psychol 1987;52:345-53.
26. Nunnally JC, Bernstein IH. Psychometric theory. 3rd ed. New York: McGraw-Hill, 1994.
27. Auerbach SM. Trait-state anxiety and adjustment to surgery. J Consult Clin Psychol 1973;40:264-71.
28. Johnston M. Anxiety in surgical patients. Psychol Med 1980;10:145-52.
29. Murris P. Monitoring and blunting: coping styles and strategies in threatening situations [thesis]. Amsterdam, The Netherlands: Univ. of Amsterdam, 1994.
30. Elsass P, Eikard B, Junge J, et al. Psychological effect of detailed preanesthetic information. Acta Anaesthesiol Scand 1987;31:579-83.
31. Londsdale M, Hutchison GL. Patients' desire for information about anaesthesia: Scottish and Canadian attitudes. Anaesthesia 1991;46:410-2.
32. Farnill D, Inglis S. Patients' desire for information about anaesthesia: Australian attitudes. Anaesthesia 1993;48:162-4.
33. Janis IL. Psychological stress: psychoanalytic and behavioral studies of surgical patients. New York: Wiley, 1958.
34. Lankton JW, Batchelder BM, Ominsky AJ. Emotional responses to detailed risk disclosure for anesthesia: a prospective randomized study. Anesthesiology 1977;46:294-6.
35. Watkins LO, Weaver L, Odegaard V. Preparation for cardiac catheterization: tailoring the content of instruction to coping style. Heart Lung 1986;15:382-9.