- Email: [email protected]
Determinants of pain assessment by nurses
ht. 1. Nws. Std. Vol. 21, Printed in Great Britain.
No.
1, pp. 43-49.
1990
0
ooze-7489ml f3.oo+ll.lm 1990 Pngamoa Press plc
Determinants of pain assessment by nurses RUUD HALFENS, Ph.D*t GEORGE EVERS, R.N., Drs.* HUDA ABU-SAAD, R.N., Ph.D.$ *Assistant Profaor. associate Professor, Department of Nursing Science, University of Limburg. P.O. Box 6I6, 6200 MD Maastricht, The Netherlands
Abstract-In this replication study pain assessment and patient perception by nurses was studied in relation to patient characteristics and nurse characteristics. Nurses were randomly assigned to one of 24 descriptions of a hypothetical patient of constant age and unspecified sex. The 24 descriptions varied by duration of pain, presence of a physical pathology, diagnosis category, and depression symptoms. Results showed that third- and fourth-year student nurses of a hospital based program, and registered nurses attributed more pain when test results of physical pathology were positive. Third- and fourth-year student nurses also attributed more pain to depressive patients. No differences were found with respect to pain duration and diagnostic category. Nursing experience seems also to be important in pain assessment. First-year student nurses attributed less pain to the hypothetical patient than third- and fourth-year student nurses and registered nurses. In addition they perceived the patient as more positive when physical pathology was present or when no symptoms of depression were present.
Introduction “Pain is whatever the experiencing person says it is, existing whenever he says it does” (McCaffery, 1979). This proposition has become the central message throughout the nursing literature on nursing the patient in pain (Casteldine, 1985). However, subjective verbal reports of painful sensations are difficult to quantify and analyse. A number of scales have been developed and tested in the measurement of pain, and increasing attention is given tAuthor to
whom correspondence should be addressed. 43
44
RUED HALFENS
et al.
to the use of subjective measures of pain as a basis for assessment and intervention (AbuSaad, 1984; Huskisson, 1976; Schmidt and Halfens, 1981). Hunt et al. (1977) reported that nurses tend to underestimate patients pain level in comparison with patients own estimates. In addition staff attitudes toward patients in relation to what desirable personality and behavioral traits they may exhibit, may influence the nurses assessment and treatment choices of these patients (Davitz and Davitz, 1975a, b; Davitz and Pendleton, 1%9). The characteristics of patients as important determinants in the assessment of pain were also supported by Taylor et al. (1984). In this study nurses were randomly assigned to one of 24 descriptions of a hypothetical patient of constant age and unspecified sex. These descriptions varied by duration of pain (3 years or 14 days), physical pathology (yes or no), diagnosis (headache, low back and joint pain), and signs of depression (yes or no). Nurses were asked to rate their pain-estimation of the patient using a IO-point scale. It was found that nurses assessed less intense pain when the hypothetical patient had no physical pathology and when pain was a long duration and chronic in nature. Further it was found that patients without pathology and patients with low back pain were perceived more negatively than other patients. A question that is left unanswered by Taylor’s study concerns the degree to which nurse assessment of patients pain is influenced by nurse characteristics. The findings of one study of nursing students showed that inferences of physical pain were greater among first-year nursing students than among second-year nursing students (Lenburg et al., 1970). It is assumed here that the nature of the education may contribute to the assessment process. In addition the views of nurses from different cultural backgrounds over the pain suffered by patients may be influenced by their own as well as the patient’s cultural background (Davitz et al., 1976). The purpose of this study is to replicate Taylor’s study and also to explore the influence of some nurse characteristics on patient pain assessment.
Research questions
Do nurses attribute less pain to a hypothetical patient: a. when findings of physical pathology are negative? b. who suffers from a chronic condition? Do nurses rate a hypothetical patient as less positive 2. a. when no physical pathology is present? b. when the diagnosis is low back pain? Is there a relationship between nurses characteristics and pain assessment by 3. nurses of a hypothetical patient? Is there a relationship between nurses characteristics and perceptions of nurses 4. of a hypothetical patient with a pain complaint? 1.
Methods
Procedure The study was designed as quasi-experimental. At three university hospitals in different areas of the Netherlands, three groups of subjects-student nurses in the first year of their hospital based program, student nurses in the last two years of their hospital based program, and registered nurses-were asked to answer a brief questionnaire about a hypothetical
DETERMINANTS
OF PAIN ASSESSMENT
BY NURSES
45
patient (Taylor et al., 1984). For each group 24 subjects from each hospital were randomly selected and assigned to one of the 24 descriptions of the hypothetical patient of constant age and unspecified sex for a’total Nof 216. One hundred and thirty-three subjects (62Yo) returned a completed questionnaire. Forty-four (33qo) first-year student nurses, 40 (30%) third- and fourth-year student nurses and 49 (37%) registered nurses. They ranged in age from 18 to 48 years (x = 24 years). Experience in nursing of registered nurses ranged from two to 30 years (x= 8.4 years, SD = 5.2). The 24 descriptions of the hypothetical patient varied by duration of pain (3 years or 14 days) physical pathology (yes or no), diagnosis (headache, low back pain and joint pain), and signs of depression (yes or no). The description were constructed using the following model: Patient A, 45years-old, is admitted for evaluation of a 1Cday (three-year) history of low back (headache/joint) pain. The patient notes feeling sad and losing interest in most activities, including eating (sentence omitted in half the questionnaires). Physical examination and lab data reveal (no) significant objective signs of pathology. As you enter the room, the patient says: “I am in severe pain” (Taylor et al., 1984). Instruments
The questionnaire covered the following areas: 1. Biographical data, including age, sex, marital status, and level of education. 2. A IO-point scale: participants were asked to assess the amount of pain they believed the hypothetical patient was having using a IO-point scale (1 = no pain; 10 = worst pain imaginable; x = 6.3; SD = 1.7). 3. Patient perception: Using a nine-trait dimension scale, participants were asked to estimate “the sort of person this patient is” on a 7-point Likert-scale. The nine trait-dimensions were: strong/weak, intelligent/not-intelligent, lazy/ambitious, unpleasant/pleasant, happy/sad, anxious/calm, demanding/not demanding, honest/dishonest, complaining/not-complaining. Scores were recorded with high values representing positive poles. Internal consistency of the nine dimensions was 0.70. The nine-trait dimension scale ranged from 21 to 50 points (x = 34.9; SD = 5.8).
Results
For the first research question the pain assessment of nurses was entered into a 2 (duration) x 2 (physical pathology) x 2 (depression symptoms) x 3 (diagnostic category) between groups analysis of variance (ANOVA). Results indicated a main significant effect, for the physical pathology factor (F= 20.57; p < 0.001). Thus, nurses attributed less pain to the hypothetical patient when test results of physical pathology were negative (x = 5.7) than when test results were positive (x = 7.0). Results also indicated a small main effect for the depression factor (F = 3.19; p c 0.10). Nurses tended to attribute more pain to the hypothetical patient with symptoms of depression (x = 6.6) than to the patient without symptoms of depression (x = 6.1). No main effects or significant interaction effects were found for the pain duration factor and medical diagnostic factor. To answer the second research question, the nine-trait dimension-scale was entered into a 2 (duration) x 2 (psysical pathology) x 2 (depression symptoms) x 3 (diagnostic category) ANOVA. Results showed a main effect for the physical pathology factor (F= 7.03;
46
RUUD HALFENS
et al.
p < 0.01). Thus perception of the hypothetical patient was significantly more positive when physical pathology was present (X = 36.1) than when no physical pathology was present (X = 33.3). Further a small main effect was found for the depression factor (F= 3.33; p < 0.10). The hypothetical patient with symptoms of depression was perceived less positive (x = 33.8) than the patient with no symptoms of depression (x = 35.8). No main effects were found for the pain duration factor and the medical diagnostic factor. The relationship between nurses characteristics and pain assessment (research question 3) was studied with a 2 (sex) x 2 (marital status) x 3 (level of education) between groups analysis of variance (ANOVA). Age was not used as a nurse characteristic, because age and level of education were highly correlated (r = 0.57). Results showed one main effect and no interaction effects. The level of pain assessment is influenced by nurses level of education (F= 5.87; p < 0.01). Student nurses in the first year of their education assess less pain to the hypothetical patient (x = 5.7) than student nurses in the last two years of their education (x = 6.9). Registered nurses came in the middle (x = 6.4). To answer the fourth research question the nine-trait dimension scale was entered into a 2 (sex) x 2 (marital status) x 3 (level of education) between groups analysis (ANOVA). Results showed no main and interaction effects. Thus, sex, marital status, and level of education were not reiated to nurses perception of the hypothetical patient. Because level of education was a significant variable, the analysis of the replication section, were repeated for the first-year student nurses, third- and fourth-year student nurses, and registered nurses separately. Pain assessment scores were entered into a 2 (duration) x 2 (physical pathology) x 2 (depression symptoms) x 3 (diagnosis category) between groups analysis of variance (ANOVA) for each group separately. For the first-year student nurses no main and interaction effects were significant, whereas for the third- and fourth-year student nurses, physical pathology (F= 10.04, p < 0.01) and depression symptoms (F= 6.01, p < 0.05) were significant. Thus third- and fourth-year student nurses attributed more pain to the hypothetical patient, when test results were positive (x = 7.8) than when test results were negative (x = 6.1). They also attributed more pain to the hypothetical patient with symptoms of depression (x = 7.6) than to the patient without symptoms of depression (x = 6.3). For the registered nurses only the physical pathology factor was significant (F= 19.90; p c 0.001). Registered nurses attributed more pain to the hypothetical patient when test results were positive (x = 7.2), than when test results were negative (x = 5.5). The mean pain scores attributed to the hypothetical patient with regard to physical pathology and depression symptoms are shown in Fig. 1. The relationship between patient characteristics and patient perception was also studied with a 2 (duration) x 2 (physical pathology) x 2 (depression symptoms) x 3 (diagnosis category) between groups analysis of variance (ANOVA) for the first year student nurses, third and fourth year student nurses and registered nurses separately. Results showed two main factors, physical pathology (F= 6.32; p < 0.05) and depression symptoms (F= 6.38; p < 0.05) to be significant for the first-year student nurses. First-year student nurses perceived the hypothetical patient as significantly more positive when physical pathology was present (x = 36.4) than when no physical pathology was present (x = 31.21). They also perceived the hypothetical patient without symptoms of depression as more positive (x = 36.0) than the patient with symptoms of depression (x = 30.8). However for the thirdand fourth-year student nurses, and the registered nurses no main and interaction effects were found.
DETERMINANTS
OF PAIN ASSESSMENT
BY NURSES
47
Fig. I Main pain scores attributed to hypothetical patient by level of education and by physical pathology and depression symptoms. 1 = first year student nurses 2 = third and fourth year student nurses 3 = registered nurses
A = physiological pathology B = depression C = no depression D = no physiological pathology
Discussion
In this study it was found that: Study subjects attributed less pain to a hypothetical patient when findings of physical pathology are negative. Study subjects attributed no less pain to a hypothetical patient who suffers from a chronic condition than to a hypothetical patient whose complaint is not chronic in nature. Study subjects rate a hypothetical patient as less positive when no physical pathology is present. Study subjects do not rate a hypothetical patient with low back pain as less positive than hypothetical patients with headache or joint pain. Level of education of study subjects is related to pain assessment by nurses. No relationship was found between nurses characteristics and nurses perceptions of a hypothetical patient. The two main findings of this study namely the presence or absence of a physical pathology for both the assessment of patient’s pain level and the perception of the patient by nurses, are in line with Taylor’s et al. (1984) findings. No confirmation was found for Taylor’s findings that the presence of a chronic condition influences nurses assessment of the patient’s pain level and for the finding that patients with low back pain were perceived as less positive than patients with headache or joint pain. In this study two small effects were found, namely nurses assess the pain level of patients with symptoms of depression higher than that of the patients without symptoms of depression, while they rate patients with symptoms of depression as less positive than patients without symptoms of depression. It is important to note that the differences in findings between Taylor’s study and this study should be interpreted in relation to two main factors. First, in Taylor’s study only registered nurses were included, whereas in this study registered nurses as well as student
48
RUUD HALFENS
et al.
nurses of hospital-based programs participated. Second, these findings may be looked at in terms of cultural differences in how patients pain level is perceived and assessed. It is clear that Dutch subjects in this study were not influenced by the acuity and chronicity of the pain condition in their assessment of the patients pain level. In addition, they do not perceive patients with low back pain as less positive as in the case in the American sample. Hence the culture may be an important characteristic in pain assessment. Thus one may reasonably conclude that attitudes and stereotypes about certain types of patients may very well be culturally bound. The results demonstrated a strong and consistent influence of the physical pathology dimension on all classes of variables including nurses level of education. This clearly illustrates the role that the biomedical model approach plays in the assessment and treatment of pain. The view that in order for pain to be present tissue damage should have occurred is unfortunately still widely accepted in the health care arena. The striking evidence that student nurses in the last two years of their nursing program and registered nurses in contrast to first year student nurses consistently used physical pathology as an important criteria in their assessment of patients pain demonstrates the overriding influence of the medical model in their educational and training programs. This approach to pain assessment obviously discounts the role of emotional and psychosociological factors in the pain experience. The evidence that student nurses in the last years of their nursing program and registered nurses, in contrast to first year students, consistently used physical pathology as an important criteria in assessment has other important clinical implications. Reports of pain by patients are often discounted or ignored when a physical cause can not be found erroneously suggesting that the physical cause for all symptoms can be identified. As a result, patients’ reports of pain many times are ignored and go untreated until a physical diagnosis is made. Examples of cancer patients whose symptoms in many cases are not treated until a physiological cause is found are numerous. Other important phenomenon such as pain reported below the level of a spinal cord injury and phantom limb pain which until recently were ignored and untreated are now better understood due to new scientific discoveries and developments in the field. Because pain is one of the most frequently encountered phenomenon in health care, hence pain assessment becomes one of the major tasks performed by nurses. It is therefore of utmost importance that nurses and other health professionals who are directly involved in patient care are introduced in their educational programs to current pain theories and kept up-to-date with new developments in the ever-expanding field of pain. This will allow for effective and appropriate assessment and rational decision-making in the treatment of pain. References Abu-Saad. H. (1984). Assessing Children’s Responses to Pain. Puin 9, 163-171. McCaffery, M. (1979). Nursing Munagement 01 the Patient with Pain. Lippincott, Philadelphia. Casteldine, G. (1985). Pam: a decade of nursing Nurs. Practice 1, 16-19. Davitz, 1. J. and Davitz, J. R. (1975a). How do nurses feel when patients suffer. Am. J. Nurs.75, 1505-1510. Davitz. I. J. and Davitz, J. R. (1975b). How nurses view patients suffering. R.N. 38, 69-74. Davitz. I. J. and Pendleton, S. H. (1969). Nurses Inferences of Suffering. Nun. Res. 18, 100-110. Davitz, L. N.. Sameshima. Y. and Davitz, J. (1976). Suffering as viewed in six different cultures. Am. J. Nurs. 76. 1296-1297. Hunt, J., Stofler. T., Littlejohns. D., Twycross, R. and Vere, D. (1977). Patients with protrac-pain: A survey conducted at the London Hospital. 1. Med. Ethics 3, 61-73.
DETERMINANTS
OF PAIN ASSESSMENT
BY NURSES
49
Huskisson. E. C. (1976). Measurement of Pain. Lurrce~ 2, 1127-1131. Lenburg, C. B., Bumside. H. and Davitz. L. J. (1970). Inferences of Physical Pain and Psychological Distress, III. In relation to length of time in the nursing educational program. Nun. Res. 19, 399-401. Schmidt, T. and Halfens. R. (1981). Pijnmeting in de kfinixhe pruktijk. Medische Psycholgie, University of Limburg. Maastricht. Taylor, A. G., Sk&on, J. A. and Butscher. J. (1984). Duration of Pain Condition and Physical Pathology as Determinants of Nurses Assessments of Patients in Pain. Nurs. Res. 33, 4-8. (Received 27 April 1989; accepted for publication 8 September 1989)
No.
1, pp. 43-49.
1990
0
ooze-7489ml f3.oo+ll.lm 1990 Pngamoa Press plc
Determinants of pain assessment by nurses RUUD HALFENS, Ph.D*t GEORGE EVERS, R.N., Drs.* HUDA ABU-SAAD, R.N., Ph.D.$ *Assistant Profaor. associate Professor, Department of Nursing Science, University of Limburg. P.O. Box 6I6, 6200 MD Maastricht, The Netherlands
Abstract-In this replication study pain assessment and patient perception by nurses was studied in relation to patient characteristics and nurse characteristics. Nurses were randomly assigned to one of 24 descriptions of a hypothetical patient of constant age and unspecified sex. The 24 descriptions varied by duration of pain, presence of a physical pathology, diagnosis category, and depression symptoms. Results showed that third- and fourth-year student nurses of a hospital based program, and registered nurses attributed more pain when test results of physical pathology were positive. Third- and fourth-year student nurses also attributed more pain to depressive patients. No differences were found with respect to pain duration and diagnostic category. Nursing experience seems also to be important in pain assessment. First-year student nurses attributed less pain to the hypothetical patient than third- and fourth-year student nurses and registered nurses. In addition they perceived the patient as more positive when physical pathology was present or when no symptoms of depression were present.
Introduction “Pain is whatever the experiencing person says it is, existing whenever he says it does” (McCaffery, 1979). This proposition has become the central message throughout the nursing literature on nursing the patient in pain (Casteldine, 1985). However, subjective verbal reports of painful sensations are difficult to quantify and analyse. A number of scales have been developed and tested in the measurement of pain, and increasing attention is given tAuthor to
whom correspondence should be addressed. 43
44
RUED HALFENS
et al.
to the use of subjective measures of pain as a basis for assessment and intervention (AbuSaad, 1984; Huskisson, 1976; Schmidt and Halfens, 1981). Hunt et al. (1977) reported that nurses tend to underestimate patients pain level in comparison with patients own estimates. In addition staff attitudes toward patients in relation to what desirable personality and behavioral traits they may exhibit, may influence the nurses assessment and treatment choices of these patients (Davitz and Davitz, 1975a, b; Davitz and Pendleton, 1%9). The characteristics of patients as important determinants in the assessment of pain were also supported by Taylor et al. (1984). In this study nurses were randomly assigned to one of 24 descriptions of a hypothetical patient of constant age and unspecified sex. These descriptions varied by duration of pain (3 years or 14 days), physical pathology (yes or no), diagnosis (headache, low back and joint pain), and signs of depression (yes or no). Nurses were asked to rate their pain-estimation of the patient using a IO-point scale. It was found that nurses assessed less intense pain when the hypothetical patient had no physical pathology and when pain was a long duration and chronic in nature. Further it was found that patients without pathology and patients with low back pain were perceived more negatively than other patients. A question that is left unanswered by Taylor’s study concerns the degree to which nurse assessment of patients pain is influenced by nurse characteristics. The findings of one study of nursing students showed that inferences of physical pain were greater among first-year nursing students than among second-year nursing students (Lenburg et al., 1970). It is assumed here that the nature of the education may contribute to the assessment process. In addition the views of nurses from different cultural backgrounds over the pain suffered by patients may be influenced by their own as well as the patient’s cultural background (Davitz et al., 1976). The purpose of this study is to replicate Taylor’s study and also to explore the influence of some nurse characteristics on patient pain assessment.
Research questions
Do nurses attribute less pain to a hypothetical patient: a. when findings of physical pathology are negative? b. who suffers from a chronic condition? Do nurses rate a hypothetical patient as less positive 2. a. when no physical pathology is present? b. when the diagnosis is low back pain? Is there a relationship between nurses characteristics and pain assessment by 3. nurses of a hypothetical patient? Is there a relationship between nurses characteristics and perceptions of nurses 4. of a hypothetical patient with a pain complaint? 1.
Methods
Procedure The study was designed as quasi-experimental. At three university hospitals in different areas of the Netherlands, three groups of subjects-student nurses in the first year of their hospital based program, student nurses in the last two years of their hospital based program, and registered nurses-were asked to answer a brief questionnaire about a hypothetical
DETERMINANTS
OF PAIN ASSESSMENT
BY NURSES
45
patient (Taylor et al., 1984). For each group 24 subjects from each hospital were randomly selected and assigned to one of the 24 descriptions of the hypothetical patient of constant age and unspecified sex for a’total Nof 216. One hundred and thirty-three subjects (62Yo) returned a completed questionnaire. Forty-four (33qo) first-year student nurses, 40 (30%) third- and fourth-year student nurses and 49 (37%) registered nurses. They ranged in age from 18 to 48 years (x = 24 years). Experience in nursing of registered nurses ranged from two to 30 years (x= 8.4 years, SD = 5.2). The 24 descriptions of the hypothetical patient varied by duration of pain (3 years or 14 days) physical pathology (yes or no), diagnosis (headache, low back pain and joint pain), and signs of depression (yes or no). The description were constructed using the following model: Patient A, 45years-old, is admitted for evaluation of a 1Cday (three-year) history of low back (headache/joint) pain. The patient notes feeling sad and losing interest in most activities, including eating (sentence omitted in half the questionnaires). Physical examination and lab data reveal (no) significant objective signs of pathology. As you enter the room, the patient says: “I am in severe pain” (Taylor et al., 1984). Instruments
The questionnaire covered the following areas: 1. Biographical data, including age, sex, marital status, and level of education. 2. A IO-point scale: participants were asked to assess the amount of pain they believed the hypothetical patient was having using a IO-point scale (1 = no pain; 10 = worst pain imaginable; x = 6.3; SD = 1.7). 3. Patient perception: Using a nine-trait dimension scale, participants were asked to estimate “the sort of person this patient is” on a 7-point Likert-scale. The nine trait-dimensions were: strong/weak, intelligent/not-intelligent, lazy/ambitious, unpleasant/pleasant, happy/sad, anxious/calm, demanding/not demanding, honest/dishonest, complaining/not-complaining. Scores were recorded with high values representing positive poles. Internal consistency of the nine dimensions was 0.70. The nine-trait dimension scale ranged from 21 to 50 points (x = 34.9; SD = 5.8).
Results
For the first research question the pain assessment of nurses was entered into a 2 (duration) x 2 (physical pathology) x 2 (depression symptoms) x 3 (diagnostic category) between groups analysis of variance (ANOVA). Results indicated a main significant effect, for the physical pathology factor (F= 20.57; p < 0.001). Thus, nurses attributed less pain to the hypothetical patient when test results of physical pathology were negative (x = 5.7) than when test results were positive (x = 7.0). Results also indicated a small main effect for the depression factor (F = 3.19; p c 0.10). Nurses tended to attribute more pain to the hypothetical patient with symptoms of depression (x = 6.6) than to the patient without symptoms of depression (x = 6.1). No main effects or significant interaction effects were found for the pain duration factor and medical diagnostic factor. To answer the second research question, the nine-trait dimension-scale was entered into a 2 (duration) x 2 (psysical pathology) x 2 (depression symptoms) x 3 (diagnostic category) ANOVA. Results showed a main effect for the physical pathology factor (F= 7.03;
46
RUUD HALFENS
et al.
p < 0.01). Thus perception of the hypothetical patient was significantly more positive when physical pathology was present (X = 36.1) than when no physical pathology was present (X = 33.3). Further a small main effect was found for the depression factor (F= 3.33; p < 0.10). The hypothetical patient with symptoms of depression was perceived less positive (x = 33.8) than the patient with no symptoms of depression (x = 35.8). No main effects were found for the pain duration factor and the medical diagnostic factor. The relationship between nurses characteristics and pain assessment (research question 3) was studied with a 2 (sex) x 2 (marital status) x 3 (level of education) between groups analysis of variance (ANOVA). Age was not used as a nurse characteristic, because age and level of education were highly correlated (r = 0.57). Results showed one main effect and no interaction effects. The level of pain assessment is influenced by nurses level of education (F= 5.87; p < 0.01). Student nurses in the first year of their education assess less pain to the hypothetical patient (x = 5.7) than student nurses in the last two years of their education (x = 6.9). Registered nurses came in the middle (x = 6.4). To answer the fourth research question the nine-trait dimension scale was entered into a 2 (sex) x 2 (marital status) x 3 (level of education) between groups analysis (ANOVA). Results showed no main and interaction effects. Thus, sex, marital status, and level of education were not reiated to nurses perception of the hypothetical patient. Because level of education was a significant variable, the analysis of the replication section, were repeated for the first-year student nurses, third- and fourth-year student nurses, and registered nurses separately. Pain assessment scores were entered into a 2 (duration) x 2 (physical pathology) x 2 (depression symptoms) x 3 (diagnosis category) between groups analysis of variance (ANOVA) for each group separately. For the first-year student nurses no main and interaction effects were significant, whereas for the third- and fourth-year student nurses, physical pathology (F= 10.04, p < 0.01) and depression symptoms (F= 6.01, p < 0.05) were significant. Thus third- and fourth-year student nurses attributed more pain to the hypothetical patient, when test results were positive (x = 7.8) than when test results were negative (x = 6.1). They also attributed more pain to the hypothetical patient with symptoms of depression (x = 7.6) than to the patient without symptoms of depression (x = 6.3). For the registered nurses only the physical pathology factor was significant (F= 19.90; p c 0.001). Registered nurses attributed more pain to the hypothetical patient when test results were positive (x = 7.2), than when test results were negative (x = 5.5). The mean pain scores attributed to the hypothetical patient with regard to physical pathology and depression symptoms are shown in Fig. 1. The relationship between patient characteristics and patient perception was also studied with a 2 (duration) x 2 (physical pathology) x 2 (depression symptoms) x 3 (diagnosis category) between groups analysis of variance (ANOVA) for the first year student nurses, third and fourth year student nurses and registered nurses separately. Results showed two main factors, physical pathology (F= 6.32; p < 0.05) and depression symptoms (F= 6.38; p < 0.05) to be significant for the first-year student nurses. First-year student nurses perceived the hypothetical patient as significantly more positive when physical pathology was present (x = 36.4) than when no physical pathology was present (x = 31.21). They also perceived the hypothetical patient without symptoms of depression as more positive (x = 36.0) than the patient with symptoms of depression (x = 30.8). However for the thirdand fourth-year student nurses, and the registered nurses no main and interaction effects were found.
DETERMINANTS
OF PAIN ASSESSMENT
BY NURSES
47
Fig. I Main pain scores attributed to hypothetical patient by level of education and by physical pathology and depression symptoms. 1 = first year student nurses 2 = third and fourth year student nurses 3 = registered nurses
A = physiological pathology B = depression C = no depression D = no physiological pathology
Discussion
In this study it was found that: Study subjects attributed less pain to a hypothetical patient when findings of physical pathology are negative. Study subjects attributed no less pain to a hypothetical patient who suffers from a chronic condition than to a hypothetical patient whose complaint is not chronic in nature. Study subjects rate a hypothetical patient as less positive when no physical pathology is present. Study subjects do not rate a hypothetical patient with low back pain as less positive than hypothetical patients with headache or joint pain. Level of education of study subjects is related to pain assessment by nurses. No relationship was found between nurses characteristics and nurses perceptions of a hypothetical patient. The two main findings of this study namely the presence or absence of a physical pathology for both the assessment of patient’s pain level and the perception of the patient by nurses, are in line with Taylor’s et al. (1984) findings. No confirmation was found for Taylor’s findings that the presence of a chronic condition influences nurses assessment of the patient’s pain level and for the finding that patients with low back pain were perceived as less positive than patients with headache or joint pain. In this study two small effects were found, namely nurses assess the pain level of patients with symptoms of depression higher than that of the patients without symptoms of depression, while they rate patients with symptoms of depression as less positive than patients without symptoms of depression. It is important to note that the differences in findings between Taylor’s study and this study should be interpreted in relation to two main factors. First, in Taylor’s study only registered nurses were included, whereas in this study registered nurses as well as student
48
RUUD HALFENS
et al.
nurses of hospital-based programs participated. Second, these findings may be looked at in terms of cultural differences in how patients pain level is perceived and assessed. It is clear that Dutch subjects in this study were not influenced by the acuity and chronicity of the pain condition in their assessment of the patients pain level. In addition, they do not perceive patients with low back pain as less positive as in the case in the American sample. Hence the culture may be an important characteristic in pain assessment. Thus one may reasonably conclude that attitudes and stereotypes about certain types of patients may very well be culturally bound. The results demonstrated a strong and consistent influence of the physical pathology dimension on all classes of variables including nurses level of education. This clearly illustrates the role that the biomedical model approach plays in the assessment and treatment of pain. The view that in order for pain to be present tissue damage should have occurred is unfortunately still widely accepted in the health care arena. The striking evidence that student nurses in the last two years of their nursing program and registered nurses in contrast to first year student nurses consistently used physical pathology as an important criteria in their assessment of patients pain demonstrates the overriding influence of the medical model in their educational and training programs. This approach to pain assessment obviously discounts the role of emotional and psychosociological factors in the pain experience. The evidence that student nurses in the last years of their nursing program and registered nurses, in contrast to first year students, consistently used physical pathology as an important criteria in assessment has other important clinical implications. Reports of pain by patients are often discounted or ignored when a physical cause can not be found erroneously suggesting that the physical cause for all symptoms can be identified. As a result, patients’ reports of pain many times are ignored and go untreated until a physical diagnosis is made. Examples of cancer patients whose symptoms in many cases are not treated until a physiological cause is found are numerous. Other important phenomenon such as pain reported below the level of a spinal cord injury and phantom limb pain which until recently were ignored and untreated are now better understood due to new scientific discoveries and developments in the field. Because pain is one of the most frequently encountered phenomenon in health care, hence pain assessment becomes one of the major tasks performed by nurses. It is therefore of utmost importance that nurses and other health professionals who are directly involved in patient care are introduced in their educational programs to current pain theories and kept up-to-date with new developments in the ever-expanding field of pain. This will allow for effective and appropriate assessment and rational decision-making in the treatment of pain. References Abu-Saad. H. (1984). Assessing Children’s Responses to Pain. Puin 9, 163-171. McCaffery, M. (1979). Nursing Munagement 01 the Patient with Pain. Lippincott, Philadelphia. Casteldine, G. (1985). Pam: a decade of nursing Nurs. Practice 1, 16-19. Davitz, 1. J. and Davitz, J. R. (1975a). How do nurses feel when patients suffer. Am. J. Nurs.75, 1505-1510. Davitz. I. J. and Davitz, J. R. (1975b). How nurses view patients suffering. R.N. 38, 69-74. Davitz. I. J. and Pendleton, S. H. (1969). Nurses Inferences of Suffering. Nun. Res. 18, 100-110. Davitz, L. N.. Sameshima. Y. and Davitz, J. (1976). Suffering as viewed in six different cultures. Am. J. Nurs. 76. 1296-1297. Hunt, J., Stofler. T., Littlejohns. D., Twycross, R. and Vere, D. (1977). Patients with protrac-pain: A survey conducted at the London Hospital. 1. Med. Ethics 3, 61-73.
DETERMINANTS
OF PAIN ASSESSMENT
BY NURSES
49
Huskisson. E. C. (1976). Measurement of Pain. Lurrce~ 2, 1127-1131. Lenburg, C. B., Bumside. H. and Davitz. L. J. (1970). Inferences of Physical Pain and Psychological Distress, III. In relation to length of time in the nursing educational program. Nun. Res. 19, 399-401. Schmidt, T. and Halfens. R. (1981). Pijnmeting in de kfinixhe pruktijk. Medische Psycholgie, University of Limburg. Maastricht. Taylor, A. G., Sk&on, J. A. and Butscher. J. (1984). Duration of Pain Condition and Physical Pathology as Determinants of Nurses Assessments of Patients in Pain. Nurs. Res. 33, 4-8. (Received 27 April 1989; accepted for publication 8 September 1989)