Original Article Registered Nurses’ Knowledge about Adverse Effects of Analgesics when Treating Postoperative Pain in Patients with Dementia ---
From the *Department of Nursing Science, Faculty of Health Sciences, University of Eastern Finland, Kuopio, Finland; †School of Pharmacy, Faculty of Health Sciences, University of Eastern Finland, Kuopio, Finland. Address correspondence to Maija Rantala, MNSc, PhD, Department of Nursing Sciences, University of Eastern Finland, P. O. Box 1627, 70211 Kuopio, Finland. E-mail: [email protected] Received February 10, 2014; Revised October 7, 2014; Accepted October 10, 2014. This study was supported by the Finnish Concordia Fund, the Finnish Association for the Study of Pain, and the Finnish Nurses Association. The authors report no financial interests or potential conflicts of interest. The authors acknowledge the help of statistician M.L. Lamidi from the University of Eastern Finland and the nurses who participated in the study. 1524-9042/$36.00 Ó 2015 by the American Society for Pain Management Nursing http://dx.doi.org/10.1016/ j.pmn.2014.10.003
-
Maija Rantala, MNSc, PhD,* Sirpa Hartikainen, MD,† Tarja Kvist, PhD,* and P€ aivi Kankkunen, PhD*
ABSTRACT:
Registered nurses (RNs) play a pivotal role in treating pain and preventing and recognizing the adverse effects (AEs) of analgesics in patients with dementia. The purpose of this study was to determine RNs’ knowledge of potentially clinically relevant AEs of analgesics. A descriptive, cross-sectional study design was used. In all, 267 RNs treating orthopedic patients, including patients with dementia, in 7 university hospitals and 10 central hospitals in Finland, completed a questionnaire. Analgesics were defined according to the Anatomic Therapeutic Classification as strong opioids, weak opioids, nonsteroidal anti-inflammatory analgesics (NSAIDs), and paracetamol. Definitions of AEs were based on the literature. Logistic regression analysis was applied to analyze which variables predicted nurses’ knowledge. The RNs had a clear understanding of the AEs of paracetamol and strong opioids. However, the AEs of NSAIDs, especially renal and cardiovascular AEs, were less well known. The median percentage of correct answers was 87% when asked about strong opioids, 73% for weak opioids, and 60% for NSAIDs. Younger RNs had better knowledge of opioid-related AEs (odds ratio [OR] per 1-year increase, 0.97; 95% confidence interval [CI], 0.94-1.00) and weak opioids (OR, 0.96; 95% CI, 0.93-0.99). This study provides evidence of a deficiency in RNs’ knowledge, especially regarding the adverse renal and cardiovascular effects of NSAIDs. Such lack of knowledge indicates that hospitals may need to update the knowledge of older RNs, especially those who treat vulnerable patients with dementia. Ó 2015 by the American Society for Pain Management Nursing
Pain Management Nursing, Vol 16, No 4 (August), 2015: pp 544-551
Nurses’ Knowledge about Analgesics
BACKGROUND The medical treatment for pain in older adults is often suboptimal, ranging from failing to give analgesics to patients experiencing considerable pain to exposing older adults to potentially life-threatening toxicities, overdoses, or adverse effects (AEs) of analgesics (Arnstein, 2010). Older adults are particularly susceptible to the AEs of analgesics, due to age-related changes in pharmacokinetics and pharmacodynamics, and risk factors such as polypharmacy and comorbidities (Arnstein, 2010; Barber & Gibson, 2009; Jahr, Breitmeyer, Pan, Royal, & Ang, 2012; Macintyre, Shug, Scott, Visser, & Walker, 2010). On the other hand, there is no specific risk for AEs of analgesics in patients with dementia. Concerns have been raised about the development of delirium when administering strong opioids in patients with dementia, but previous studies have shown that this risk can be reduced in these patients by providing sufficient analgesia, so that severe pain is avoided (Lindesay, Rockwood, & Rolfson, 2002; Sieber, Mears, Lee, & Gottschalk, 2011). It has been suggested that in severe pain an overly low dose of opioids is a risk factor for developing delirium in patients with dementia (Sieber et al., 2011). Previous studies have addressed nurses’ knowledge regarding the use of opioids and their AEs, especially addiction (Abdalrahim, Majali, Stomberg, & Bergholm, 2011; Eid, Manias, Bucknall, & Almazroo, 2014; Griffiths et al., 2012), respiratory depression (Abdalrahim et al., 2011), and dependence (Eid et al., 2014). According to a previous study, the nurses underestimated the pain intensity of postoperative patients and they lacked pharmacologic information concerning opioids and their AEs (Eid et al., 2014). Regimens of postoperative analgesia consider paracetamol as a basic analgesic (Griffiths et al., 2012; Jahr & Lee, 2010; Myles & Power, 2007). Paracetamol is recommended as a first-line analgesic for treatment of pain in older people because it is safe in doses smaller than 4 g per day (American Geriatrics Society [AGS], 2009). Although the AE associated with use of paracetamol is hepatic toxicity (Craig et al., 2011; Mort, Shiyanbola, Ndehi, & Stacy, 2011), it can be avoided by administering paracetamol in therapeutic doses (AGS, 2009). Adverse effects of nonsteroidal anti-inflammatory analgesics (NSAIDs) are significant and may limit their use in older people (AGS, 2009). NSAIDs should be used with caution in older people and the lowest doses should be provided for the shortest duration (Abdulla et al., 2013). The main AEs of NSAIDs include gastrointestinal (GI), cardiovascular, and renal effects (Ong,
545
Lirk, Tan, & Seymour, 2007; Vonkeman & van de Laar, 2010). Strong and weak opioids show similarities in their pharmacokinetics and dynamics, and both are m-opioid receptor agonists (Leppert, 2011) and thus have similar AEs in neurologic, cardiopulmonary, GI, and urologic systems (Harris, 2008). Because nurses play a pivotal role in advocating qualified pain treatment in vulnerable patients with dementia, and because they spend more time with patients with pain than any other health care team member, it is important that their knowledge regarding analgesics is as thorough as possible (McCaffery & Ferrell, 1997). Nurses also have direct responsibility related to the tailoring of analgesics and to preventing and recognizing their potential AEs, especially when treating postoperative pain in vulnerable patients with dementia. However, there is limited evidence of nurses’ knowledge concerning the AEs associated with analgesics and the factors that go with this knowledge. Overall, insufficient knowledge of pain management on the part of health care workers is a barrier to adequate treatment (Francis & Fitzpatrick, 2013; Tse & Ho, 2014).
RESEARCH QUESTIONS The purpose of this study was to determine nurses’ knowledge regarding potentially clinically relevant AEs of different types of analgesics when treating postoperative pain in patients with dementia. The research questions were as follows: 1. How did the nurses identify the potentially clinically relevant AEs of different types of analgesics? 2. Which variables predicted the nurses’ knowledge about potentially clinically relevant AEs of different types of analgesics?
MATERIALS AND METHODS Design and Sample This cross-sectional study was conducted in 7 university hospitals and 10 central hospitals between March and May 2011. The questionnaire was sent to 494 registered nurses (RNs) working in orthopedic units. The exclusion criteria were hospitals that had less than 100 first hip fractures by the year 2009. All the university hospitals and 10 of 11 central hospitals were included in the study. These hospitals admitted approximately 68% of the total number of first hip fracture patients in Finland by the year 2009. Data were from a larger study exploring nurses’ evaluations of postoperative pain management in patients with dementia (Rantala, 2014).
546
Rantala et al.
Of 279 responses, 12 were excluded because they did not answer the research question concerning their knowledge about possible AEs of analgesics. In all, 267 RNs responded to the questionnaire, giving a response rate of 54%. The characteristics of the RNs are reported in Table 1. Contact persons, one head nurse from each orthopedic unit, distributed the questionnaires and informed the participants. A cover information letter detailing the procedure was attached to the questionnaire, to which participants were asked to respond. In April 2011, participants were reminded by the contact persons about filling out the questionnaire. Questionnaire forms were returned to the researcher in prepaid envelopes. Instrument The questionnaire included a demographic section (Table 1), a survey of the different types of analgesics, and a list of different AEs (Table 2). The demographic section included the type of hospital (university or central), contract (permanent or deputy), employment arrangements (full vs. part time), work shifts (daytime or two-shift work vs. three-shift work or night work), age, work experience in health care, and work experience on current unit (Table 1). The types of analgesics were categorized according to the Anatomical Therapeutic Chemical Classification System recommended by the World Health Organization (WHO, 2010). In the WHO
TABLE 1. Characteristics of RNs Variable (n ¼ 267)
Mean, SD, or %
Hospital University hospital (n ¼ 126) 47.2% Central hospital (n ¼ 141) 52.8% Age (n ¼ 266) 40.9 y 10.9 y Sex Female (n ¼ 257) 96.3% Male (n ¼ 10) 3.7% Work experience in current unit (n ¼ 263) 9.7 y 8.2 y Work experience in health care (n ¼ 265) 15.4 y 9.8 y Contract Permanent (n ¼ 215) 82.1% Deputy (n ¼ 47) 17.9% Employment arrangement Fully time (n ¼ 234) 88.3% Part time (n ¼ 31) 11.7% Work shifts (n ¼ 266) Three-shift or night work (n ¼ 229) 86.1% Daytime or two-shift work (n ¼ 37) 13.9%
study, analgesics were defined as strong opioids (N02AA01-55, N02AB), weak opioids (N02AA59, N02AE, and N02AX), NSAIDs (M01AB-AH), and paracetamol (N02BE01). The questionnaire consisted of a table of 21 AEs concerning four different types of analgesics (Table 2). The potentially clinically relevant AEs associated with the types of analgesics were defined based on previous literature (AGS, 2009; Elia, Lysakowski, & Tramer, 2005; Fishbain, Cole, Lewis, Rosomoff, & Rosomoff, 2008; Moore, Derry, McQuay, & Wiffen, 2011; Gnjidic, Murnion, & Hilmer, 2008; Harris, 2008; Leppert, 2011; Ong et al., 2007; Plante & VanItallie, 2010; Puopolo et al., 2007; Vonkeman & van de Laar, 2010) and an expert panel of authors. The databases Medline, Cochrane, Cinahl, and Science Direct were searched for relevant articles published between 2005 and 2011. Additional references were identified from bibliographies of the retrieved reports. Strong opioids were associated with 15 potentially clinically relevant AEs, weak opioids with 11, NSAIDs with 5, and paracetamol with 1. Table 2 presents the associated potentially clinically relevant AEs. A pretest of the questionnaire was conducted in October and November 2010 with 19 nursing staff in a surgical department that was not involved in the main study. Each participant received a folder containing participant information, a questionnaire, and a postage-stamped envelope. The nursing staff and one surgeon stated on a separate form that the items in the questionnaire were adequate and clearly expressed. No items in the tool were eliminated or modified based on the pretest. After that, the face validity was established by asking two pain experts (one a docent in nursing science and one professor of geriatric pharmacotherapy), one professor of nursing science, and eight doctorial students to review the questionnaire. The RNs were asked to circle the correct option for each question. For example, in the case of strong opioids, the potential number of AEs was 15, with 15 correct choices representing thorough knowledge of the associated AEs. The respondents were divided into two groups to determine how many correct answers were given by the group with number of the nurses with best overall knowledge (approximately 20% of the nurses) of potentially clinically relevant AEs of analgesics. The cutoff point of ‘‘number of nurses with best knowledge’’ was defined as 20% by an expert panel of authors who hypothesized that 20% is a sufficient level of knowledge. The RN group with ‘‘the number of nurses with best knowledge of adverse effects of analgesics’’ produced the following results:
547
Nurses’ Knowledge about Analgesics
All analgesics: 18% (n ¼ 49) of RNs with at least 29 correct answers (range 0-32); Strong opioids: 19% (n ¼ 51) of RNs with 15 correct answers (range 0-15); Weak opioids: 22% (n ¼ 59) of RNs with at least 10 correct answers (range 0-11); NSAIDs: 26% (n ¼ 69) of RNs with 5 correct answers (range 0-5); and Paracetamol: 91% (n ¼ 243) of RNs with 1 correct answer (range 0-1).
Statistical Analyses A descriptive analysis of the questionnaire data was conducted using SPSS 19.0 for Windows (SPSS Inc., Chicago, IL, USA). Logistic regression analysis using the Wald Forward method was conducted to determine which variables were associated with the best knowledge (Fig. 1) of potentially clinically relevant AEs in relation to all AEs, strong opioids, weak opioids, NSAIDs, and paracetamol. The odds ratio (OR, 95% confidence interval [CI]) was used to analyze deviations in knowledge of clinically significant AEs of analgesics. The variables included the type of hospital (university or central), contract (permanent or deputy), employment arrangements (full time vs. part time), work shifts (daytime or two-shift work vs. three-shift work or night work), age, work experience in health care, and work experience in current unit. A p value of
From the *Department of Nursing Science, Faculty of Health Sciences, University of Eastern Finland, Kuopio, Finland; †School of Pharmacy, Faculty of Health Sciences, University of Eastern Finland, Kuopio, Finland. Address correspondence to Maija Rantala, MNSc, PhD, Department of Nursing Sciences, University of Eastern Finland, P. O. Box 1627, 70211 Kuopio, Finland. E-mail: [email protected] Received February 10, 2014; Revised October 7, 2014; Accepted October 10, 2014. This study was supported by the Finnish Concordia Fund, the Finnish Association for the Study of Pain, and the Finnish Nurses Association. The authors report no financial interests or potential conflicts of interest. The authors acknowledge the help of statistician M.L. Lamidi from the University of Eastern Finland and the nurses who participated in the study. 1524-9042/$36.00 Ó 2015 by the American Society for Pain Management Nursing http://dx.doi.org/10.1016/ j.pmn.2014.10.003
-
Maija Rantala, MNSc, PhD,* Sirpa Hartikainen, MD,† Tarja Kvist, PhD,* and P€ aivi Kankkunen, PhD*
ABSTRACT:
Registered nurses (RNs) play a pivotal role in treating pain and preventing and recognizing the adverse effects (AEs) of analgesics in patients with dementia. The purpose of this study was to determine RNs’ knowledge of potentially clinically relevant AEs of analgesics. A descriptive, cross-sectional study design was used. In all, 267 RNs treating orthopedic patients, including patients with dementia, in 7 university hospitals and 10 central hospitals in Finland, completed a questionnaire. Analgesics were defined according to the Anatomic Therapeutic Classification as strong opioids, weak opioids, nonsteroidal anti-inflammatory analgesics (NSAIDs), and paracetamol. Definitions of AEs were based on the literature. Logistic regression analysis was applied to analyze which variables predicted nurses’ knowledge. The RNs had a clear understanding of the AEs of paracetamol and strong opioids. However, the AEs of NSAIDs, especially renal and cardiovascular AEs, were less well known. The median percentage of correct answers was 87% when asked about strong opioids, 73% for weak opioids, and 60% for NSAIDs. Younger RNs had better knowledge of opioid-related AEs (odds ratio [OR] per 1-year increase, 0.97; 95% confidence interval [CI], 0.94-1.00) and weak opioids (OR, 0.96; 95% CI, 0.93-0.99). This study provides evidence of a deficiency in RNs’ knowledge, especially regarding the adverse renal and cardiovascular effects of NSAIDs. Such lack of knowledge indicates that hospitals may need to update the knowledge of older RNs, especially those who treat vulnerable patients with dementia. Ó 2015 by the American Society for Pain Management Nursing
Pain Management Nursing, Vol 16, No 4 (August), 2015: pp 544-551
Nurses’ Knowledge about Analgesics
BACKGROUND The medical treatment for pain in older adults is often suboptimal, ranging from failing to give analgesics to patients experiencing considerable pain to exposing older adults to potentially life-threatening toxicities, overdoses, or adverse effects (AEs) of analgesics (Arnstein, 2010). Older adults are particularly susceptible to the AEs of analgesics, due to age-related changes in pharmacokinetics and pharmacodynamics, and risk factors such as polypharmacy and comorbidities (Arnstein, 2010; Barber & Gibson, 2009; Jahr, Breitmeyer, Pan, Royal, & Ang, 2012; Macintyre, Shug, Scott, Visser, & Walker, 2010). On the other hand, there is no specific risk for AEs of analgesics in patients with dementia. Concerns have been raised about the development of delirium when administering strong opioids in patients with dementia, but previous studies have shown that this risk can be reduced in these patients by providing sufficient analgesia, so that severe pain is avoided (Lindesay, Rockwood, & Rolfson, 2002; Sieber, Mears, Lee, & Gottschalk, 2011). It has been suggested that in severe pain an overly low dose of opioids is a risk factor for developing delirium in patients with dementia (Sieber et al., 2011). Previous studies have addressed nurses’ knowledge regarding the use of opioids and their AEs, especially addiction (Abdalrahim, Majali, Stomberg, & Bergholm, 2011; Eid, Manias, Bucknall, & Almazroo, 2014; Griffiths et al., 2012), respiratory depression (Abdalrahim et al., 2011), and dependence (Eid et al., 2014). According to a previous study, the nurses underestimated the pain intensity of postoperative patients and they lacked pharmacologic information concerning opioids and their AEs (Eid et al., 2014). Regimens of postoperative analgesia consider paracetamol as a basic analgesic (Griffiths et al., 2012; Jahr & Lee, 2010; Myles & Power, 2007). Paracetamol is recommended as a first-line analgesic for treatment of pain in older people because it is safe in doses smaller than 4 g per day (American Geriatrics Society [AGS], 2009). Although the AE associated with use of paracetamol is hepatic toxicity (Craig et al., 2011; Mort, Shiyanbola, Ndehi, & Stacy, 2011), it can be avoided by administering paracetamol in therapeutic doses (AGS, 2009). Adverse effects of nonsteroidal anti-inflammatory analgesics (NSAIDs) are significant and may limit their use in older people (AGS, 2009). NSAIDs should be used with caution in older people and the lowest doses should be provided for the shortest duration (Abdulla et al., 2013). The main AEs of NSAIDs include gastrointestinal (GI), cardiovascular, and renal effects (Ong,
545
Lirk, Tan, & Seymour, 2007; Vonkeman & van de Laar, 2010). Strong and weak opioids show similarities in their pharmacokinetics and dynamics, and both are m-opioid receptor agonists (Leppert, 2011) and thus have similar AEs in neurologic, cardiopulmonary, GI, and urologic systems (Harris, 2008). Because nurses play a pivotal role in advocating qualified pain treatment in vulnerable patients with dementia, and because they spend more time with patients with pain than any other health care team member, it is important that their knowledge regarding analgesics is as thorough as possible (McCaffery & Ferrell, 1997). Nurses also have direct responsibility related to the tailoring of analgesics and to preventing and recognizing their potential AEs, especially when treating postoperative pain in vulnerable patients with dementia. However, there is limited evidence of nurses’ knowledge concerning the AEs associated with analgesics and the factors that go with this knowledge. Overall, insufficient knowledge of pain management on the part of health care workers is a barrier to adequate treatment (Francis & Fitzpatrick, 2013; Tse & Ho, 2014).
RESEARCH QUESTIONS The purpose of this study was to determine nurses’ knowledge regarding potentially clinically relevant AEs of different types of analgesics when treating postoperative pain in patients with dementia. The research questions were as follows: 1. How did the nurses identify the potentially clinically relevant AEs of different types of analgesics? 2. Which variables predicted the nurses’ knowledge about potentially clinically relevant AEs of different types of analgesics?
MATERIALS AND METHODS Design and Sample This cross-sectional study was conducted in 7 university hospitals and 10 central hospitals between March and May 2011. The questionnaire was sent to 494 registered nurses (RNs) working in orthopedic units. The exclusion criteria were hospitals that had less than 100 first hip fractures by the year 2009. All the university hospitals and 10 of 11 central hospitals were included in the study. These hospitals admitted approximately 68% of the total number of first hip fracture patients in Finland by the year 2009. Data were from a larger study exploring nurses’ evaluations of postoperative pain management in patients with dementia (Rantala, 2014).
546
Rantala et al.
Of 279 responses, 12 were excluded because they did not answer the research question concerning their knowledge about possible AEs of analgesics. In all, 267 RNs responded to the questionnaire, giving a response rate of 54%. The characteristics of the RNs are reported in Table 1. Contact persons, one head nurse from each orthopedic unit, distributed the questionnaires and informed the participants. A cover information letter detailing the procedure was attached to the questionnaire, to which participants were asked to respond. In April 2011, participants were reminded by the contact persons about filling out the questionnaire. Questionnaire forms were returned to the researcher in prepaid envelopes. Instrument The questionnaire included a demographic section (Table 1), a survey of the different types of analgesics, and a list of different AEs (Table 2). The demographic section included the type of hospital (university or central), contract (permanent or deputy), employment arrangements (full vs. part time), work shifts (daytime or two-shift work vs. three-shift work or night work), age, work experience in health care, and work experience on current unit (Table 1). The types of analgesics were categorized according to the Anatomical Therapeutic Chemical Classification System recommended by the World Health Organization (WHO, 2010). In the WHO
TABLE 1. Characteristics of RNs Variable (n ¼ 267)
Mean, SD, or %
Hospital University hospital (n ¼ 126) 47.2% Central hospital (n ¼ 141) 52.8% Age (n ¼ 266) 40.9 y 10.9 y Sex Female (n ¼ 257) 96.3% Male (n ¼ 10) 3.7% Work experience in current unit (n ¼ 263) 9.7 y 8.2 y Work experience in health care (n ¼ 265) 15.4 y 9.8 y Contract Permanent (n ¼ 215) 82.1% Deputy (n ¼ 47) 17.9% Employment arrangement Fully time (n ¼ 234) 88.3% Part time (n ¼ 31) 11.7% Work shifts (n ¼ 266) Three-shift or night work (n ¼ 229) 86.1% Daytime or two-shift work (n ¼ 37) 13.9%
study, analgesics were defined as strong opioids (N02AA01-55, N02AB), weak opioids (N02AA59, N02AE, and N02AX), NSAIDs (M01AB-AH), and paracetamol (N02BE01). The questionnaire consisted of a table of 21 AEs concerning four different types of analgesics (Table 2). The potentially clinically relevant AEs associated with the types of analgesics were defined based on previous literature (AGS, 2009; Elia, Lysakowski, & Tramer, 2005; Fishbain, Cole, Lewis, Rosomoff, & Rosomoff, 2008; Moore, Derry, McQuay, & Wiffen, 2011; Gnjidic, Murnion, & Hilmer, 2008; Harris, 2008; Leppert, 2011; Ong et al., 2007; Plante & VanItallie, 2010; Puopolo et al., 2007; Vonkeman & van de Laar, 2010) and an expert panel of authors. The databases Medline, Cochrane, Cinahl, and Science Direct were searched for relevant articles published between 2005 and 2011. Additional references were identified from bibliographies of the retrieved reports. Strong opioids were associated with 15 potentially clinically relevant AEs, weak opioids with 11, NSAIDs with 5, and paracetamol with 1. Table 2 presents the associated potentially clinically relevant AEs. A pretest of the questionnaire was conducted in October and November 2010 with 19 nursing staff in a surgical department that was not involved in the main study. Each participant received a folder containing participant information, a questionnaire, and a postage-stamped envelope. The nursing staff and one surgeon stated on a separate form that the items in the questionnaire were adequate and clearly expressed. No items in the tool were eliminated or modified based on the pretest. After that, the face validity was established by asking two pain experts (one a docent in nursing science and one professor of geriatric pharmacotherapy), one professor of nursing science, and eight doctorial students to review the questionnaire. The RNs were asked to circle the correct option for each question. For example, in the case of strong opioids, the potential number of AEs was 15, with 15 correct choices representing thorough knowledge of the associated AEs. The respondents were divided into two groups to determine how many correct answers were given by the group with number of the nurses with best overall knowledge (approximately 20% of the nurses) of potentially clinically relevant AEs of analgesics. The cutoff point of ‘‘number of nurses with best knowledge’’ was defined as 20% by an expert panel of authors who hypothesized that 20% is a sufficient level of knowledge. The RN group with ‘‘the number of nurses with best knowledge of adverse effects of analgesics’’ produced the following results:
547
Nurses’ Knowledge about Analgesics
All analgesics: 18% (n ¼ 49) of RNs with at least 29 correct answers (range 0-32); Strong opioids: 19% (n ¼ 51) of RNs with 15 correct answers (range 0-15); Weak opioids: 22% (n ¼ 59) of RNs with at least 10 correct answers (range 0-11); NSAIDs: 26% (n ¼ 69) of RNs with 5 correct answers (range 0-5); and Paracetamol: 91% (n ¼ 243) of RNs with 1 correct answer (range 0-1).
Statistical Analyses A descriptive analysis of the questionnaire data was conducted using SPSS 19.0 for Windows (SPSS Inc., Chicago, IL, USA). Logistic regression analysis using the Wald Forward method was conducted to determine which variables were associated with the best knowledge (Fig. 1) of potentially clinically relevant AEs in relation to all AEs, strong opioids, weak opioids, NSAIDs, and paracetamol. The odds ratio (OR, 95% confidence interval [CI]) was used to analyze deviations in knowledge of clinically significant AEs of analgesics. The variables included the type of hospital (university or central), contract (permanent or deputy), employment arrangements (full time vs. part time), work shifts (daytime or two-shift work vs. three-shift work or night work), age, work experience in health care, and work experience in current unit. A p value of
