Healthcare providers’ accuracy in assessing patients’ pain: A systematic review

G Model PEC 5100 No. of Pages 10

Patient Education and Counseling xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Patient Education and Counseling journal homepage: www.elsevier.com/locate/pateducou

Healthcare providers’ accuracy in assessing patients’ pain: A systematic review Mollie A. Rubena,* , Mara van Oschb , Danielle Blanch-Hartiganc a b c

US Department of Veterans Affairs, Center for Healthcare Organization and Implementation Research, Boston, MA, USA NIVEL (Netherlands Institute for Health Services Research), Utrecht, The Netherlands Department of Natural and Applied Sciences, Bentley University, Waltham, MA, USA

A R T I C L E I N F O

A B S T R A C T

Article history: Received 27 January 2015 Received in revised form 2 July 2015 Accepted 9 July 2015

Objective: Healthcare providers satisfy an important role in providing appropriate care in the prevention and management of acute and chronic pain, highlighting the importance of providers’ abilities to accurately assess patients’ pain. We systematically reviewed the literature on healthcare providers’ pain assessment accuracy. Methods: A systematic literature search was conducted in PubMed and PsycINFO to identify studies addressing providers’ pain assessment accuracy, or studies that compared patients’ self-report of pain with providers’ assessment of pain. Results: 60 studies met the inclusion criteria. Healthcare providers had moderate to good pain assessment accuracy. Physicians and nurses showed similar pain assessment accuracy. Differences in pain assessment accuracy were found according to providers’ clinical experience, the timing of the pain assessment, vulnerable patient populations and patients’ pain intensity. Conclusion: Education and training aimed at improving providers with poor pain assessment accuracy is discussed especially in relation to those with limited clinical experience (<4 years) or a great deal of clinical experience (>10 years) and those providing care for vulnerable patient populations. Practice implications: More research on characteristics that influence providers’ pain assessment accuracy and trainings to improve pain assessment accuracy in medical and continuing education may improve pain treatment for patients. Published by Elsevier Ireland Ltd.

Keywords: Pain assessment Provider accuracy Clinical experience Chronic pain Acute pain

1. Introduction Uncontrolled acute and chronic pain is a major healthcare challenge and public health problem [1]. Estimates suggest that more than one-third of American adults and one-fifth of European adults suffer from some type of chronic pain [1,2]. Although the majority of pain sufferers seek medical attention for their pain [3], pain is often undertreated. Considerable undertreatment of pain has been documented in patients with cancer [4], AIDS and HIV [5,6], emergency department patients [7], children [8], and older adults and dementia patients [9]. In a recent review, nearly 50% of patients with cancer had pain that was undertreated [4]. Undertreated pain can create physiological, psychological, social, and

* Corresponding author at: US Department of Veterans Affairs, Center for Healthcare Organization and Implementation Research, 150 S, Huntington Ave., Bldg. 9, Boston, MA 02130, USA. E-mail addresses: [email protected], [email protected] (M.A. Ruben).

economic burdens on sufferers, their families, and society at large [2,10,11]. Healthcare providers must be accurate in assessing patients’ pain in order to provide appropriate care and avoid undertreating pain. Accurately assessing pain refers specifically to the ability to correctly discriminate a patient’s level of pain. This is a crucial aspect of an effective patient-centered approach to clinical care for pain patients [12]. The Institute of Medicine recommends that healthcare providers complete consistent and comprehensive pain assessments so that patients receive appropriate pain care [1]. Accurate pain assessment is particularly important for those who cannot self-report their pain, such as infants and dementia patients [13]. For these populations, provider and caregivers’ assessments of nonverbal indicators of pain inform treatment and medication decision-making [14]. Despite recommendations and the need for accurate assessment of patients’ pain, the literature suggests that providers tend to underestimate and undertreat pain [15]. The purpose of the present article is to systematically review the literature on providers’ abilities to accurately assess their

http://dx.doi.org/10.1016/j.pec.2015.07.009 0738-3991/ Published by Elsevier Ireland Ltd.

Please cite this article in press as: M.A. Ruben, et al., Healthcare providers’ accuracy in assessing patients’ pain: A systematic review, Patient Educ Couns (2015), http://dx.doi.org/10.1016/j.pec.2015.07.009

G Model PEC 5100 No. of Pages 10

2

M.A. Ruben et al. / Patient Education and Counseling xxx (2015) xxx–xxx

patients’ pain by comparing providers’ ratings of patients’ pain to patients’ self-reported pain. More specifically, we aim to answer four main research questions: RQ1. How accurate are providers at assessing patients’ pain? RQ2. Are some providers more accurate at pain assessment than others (e.g. those with more clinical experience or certain types of providers)? RQ3. Do patient characteristics (e.g., gender, age, cognitive ability) impact providers’ ability to accurately assess pain? RQ4. Do characteristics of the pain itself (e.g., pain intensity, acute vs chronic) impact providers’ abilities to accurately assess pain? 1.1. Assessing pain in clinical contexts Accurately assessing patients’ pain is difficult because the perception of pain for the sufferer is a highly personal and subjective experience. The nociceptive input is influenced by biological and psychosocial aspects of pain (e.g., pathology, cultural background, memories, emotions, and cognitions) most of which are not easily accessible to a provider [11,16,17]. Moreover, the therapeutic context, which encompasses the doctor-patient relationship and the treatment regimen all have influences on patients’ pain experience [18–20]. The resultant pain experience and the nonverbal and verbal expression of pain is therefore not solely (or linearly) related to the nociceptive input, but shaped by a variety of aspects. By its nature, the complexity of the subjective experience of pain challenges the accurate assessment of pain by providers. Although the experience of pain for the sufferer is complex and multidimensional [21,22], in clinical practice and research, the standard method of assessing patients’ pain is a unidimensional selfreport of pain. Self-report instruments, written or verbal reports describing the sufferers’ pain intensity [16,23], are the most widely used way to measure patients’ pain [24]. Self-reports of pain are acknowledged as problematic for many reasons including the deliberate control of pain reports and the oversimplification of the multidimensional pain experience [25] and are not always the best reflection of patients’ actual pain. However, self-reports of pain are extremely efficient, especially in the clinical setting where time is limited. Therefore, most studies which assess provider accuracy, compare providers’ assessment to patients’ self-reports of pain. Although studies have explored biases in pain estimation, less is known about providers’ general pain assessment accuracy. It is important to note that providers’ pain assessment accuracy is independent from their pain assessment bias, or their overall tendency to underestimate or overestimate pain. The focus on pain assessment accuracy specifically has generated increasing interest within medical consultations. In the current systematic review, we summarize available evidence to provide an overall picture of how accurate providers are at assessing patients’ pain. In addition, this review also examines the patient, provider and pain-related characteristics that may influence providers’ pain assessment accuracy. Given the subjectivity of patients’ self-reports of pain, providers’ accuracy may be impacted by characteristics of their patients or characteristics of their patients’ pain. With a better understanding of pain assessment and the factors that influence accuracy, we can better target provider training to address the undertreatment of pain. 2. Methods 2.1. Study characteristics Inclusion and exclusion criteria were determined a priori. In order to be eligible for inclusion in the current review, studies had

to report providers’ pain assessment accuracy, or the direct comparison between patients’ self-report of pain and providers’ assessment of pain. Pain assessment accuracy was reported in studies as a Pearson correlation coefficient (r), intraclass correlation coefficient (ICC), or weighted kappa coefficient. The patient population included any patients who self-reported their pain, including children and older adults with dementia when a self-report was present. The provider population included any healthcare providers (i.e., physicians, nurses, midwives) who viewed patients in pain. Studies were excluded if, (1) they did not directly compare providers’ judgment of pain with patients’ verbal or written self-report of pain (the criterion), (2) providers had access to patients’ self-reported pain prior to inferring patients’ pain, or (3) patients were made-up vignettes or scenarios and not actually pain sufferers. 2.2. Search strategy We performed a broad systematic literature search for peerreviewed articles that contained the terms ‘pain assessment’, ‘judgments of pain’, ‘pain detection’, ‘pain’, and ‘pain intensity’ combined with terms related to providers and patients (including ‘provider’, ‘physician’, ‘nurse’, ‘clinician’, and ‘patient’). The following databases were searched up to January 2015: PubMed (coverage 1946-present) and PsycINFO (coverage 1894-present). The reference lists of relevant studies and systematic reviews were investigated. We also reviewed the reference lists of all articles identified. Although not an exclusion criteria, no non-English language publications satisfied the inclusion criteria. 2.3. Study selection The first author (MAR) independently reviewed all of the 819 titles and abstracts that met search criteria to determine their eligibility. Of these, 157 full texts versions were obtained and reviewed for inclusion. Ninety-seven studies did not meet the inclusion criteria. A total of 60 studies met all inclusion criteria and were included in the review. Any disagreements about inclusion were resolved by discussion with the third author (DBH). 2.4. Study extraction and management The first author (MAR) extracted the data from the included studies and a second author (DBH) reviewed data for accuracy and completeness. The following information was extracted for each study: 1 Study reference (author, year of publication, country of study completion) 2 Patient participants (number of participants, age group, gender) 3 Type of patient pain (pain intensity, acute or chronic, timing of pain assessment) 4 Provider participants (number of participants, gender, clinical experience) 5 Provider credentials (physician, nurse, other: midwives, healthcare administrators, physiotherapists) 6 Pain assessment accuracy level (coded as poor, fair, moderate, good, or excellent according to the established guidelines1 )

1 Cohen’s classifications were used to categorize correlational effect sizes [92]. Values less than 0.10 were considered poor, 0.10–0.30 fair, 0.3–0.5 moderate, 0.50– 0.75 good, and greater than 0.75 were considered excellent. Intraclass correlations were considered poor under 0.40, fair to good from 0.40 to 0.75, and excellent when greater than 0.75 [93]. In the categorization of the weighted Kappa coefficient, values under 0.20 were considered poor, 0.21–0.40 fair, 0.41–0.60 moderate, 0.61– 0.80 good, and greater than 0.81 excellent, based on guidelines from Landis and Koch [94].

Please cite this article in press as: M.A. Ruben, et al., Healthcare providers’ accuracy in assessing patients’ pain: A systematic review, Patient Educ Couns (2015), http://dx.doi.org/10.1016/j.pec.2015.07.009

G Model PEC 5100 No. of Pages 10

Authors

Year

Patient N

Pain type

Patient age group

Provider total N

Provider population

Main findings

Akin & Durna [95] Bergh & Sjostrom [53]

2013 1999

119 39

Chronic

Adults Older adults

126 39

Nurses Nurses

Blomquist & Hallberg [96]

1999

29

Chronic

29

Bondestam, Hovgren, Johansson, Jern, Herlitz, & Holmberg [97] Broberger, Tishelman, & von Essen [98] Choiniere, Melzack, Girard, Rondeau, & Paquin [99] Colwell, Clark, & Perkins [30] Cremeans-Smith, Stephens, Franks, Martire, Druley, & Wojno [32] de Bock, van Marwijk, Kapkin, & Mulder [100] Drayer, Henderson, & Reidenberg [39]

1987

47

Acute

Older adults Adults

Nurses, nursing assistants Nurses

Nurses showed fair accuracy for patients’ pain. Turkey Sweden Nurses showed moderate accuracy for patients’ pain. Higher patient pain negatively related to nurses’ accuracy. Nurses who had received pain management training had a higher correlation with patients' pain ratings compared to nurses who had not completed the training. Nurses showed good accuracy for patients’ pain. Sweden

2005 85

Chronic

Adults

83

1990

Acute

Adults

1996 124 2003 114

Acute

1994

198

1999

45

42

1994 49 Everett, Patterson, Marvin, Montgomery, Ordonez, & Campbell [101] Ewing, Rogers, Barclay, McCabe, 2006 126 Martin, Campbell, & Todd [40] Favaloro & Touzel [48] 1990 22

Country

Nurses showed excellent accuracy for patients’ pain.

Sweden

Nurses

Nurses showed moderate accuracy for patients’ pain.

Sweden

42

Nurses

Nurses showed moderate accuracy for patients’ pain.

Canada

Children Older adults

44 119

Nurses Physicians

Nurses showed good to excellent accuracy for patients’ pain. Physicians showed poor accuracy for patients’ pain.

USA USA

Chronic

Adults

40

Physicians

Physicians showed moderate accuracy for patients’ pain.

Netherlands

Acute and chronic Acute

Adults

Physicians, nurses

Physicians and nurses showed moderate accuracy for patients’ pain.

USA

Adults

Nurses

Nurses showed good to excellent accuracy for patients’ pain.

USA

Chronic

Adults

Physicians, nurses

Both physicians and nurses showed good accuracy for patients’ pain.

UK

Acute

Children

Nurses

Nurses showed good accuracy for patients' pain. Nurses with < 1 year of experience showed Australia moderate accuracy while nurses with > 1 year of experience showed excellent accuracy for patients' pain. When patients experienced higher levels of pain, nurses showed fair accuracy for patients' pain. Physicians showed good accuracy for patients’ pain. Denmark

27

Forrest, Hermann, & Andersen [102] Fridh & Gaston-Johansson [52]

1989

52

Acute

Adults

8

Physicians

1990

138

Acute

Adults

12

Midwives

Fromme, Eilers, Mori, Hsieh, & Beer [36] Grossman, Sheidler, Swedeen, Mucenski, & Piantadosi [41] Harrison [103]

2004 37

Chronic

Adults

1

Physician

Midwives showed moderate accuracy for patients’ labor pain at the beginning and middle of labor, Sweden but when the labor was most intense at the end, midwives showed poor accuracy for patients’ pain. The physician showed poor accuracy for patients’ pain. USA

Fellows, residents, nurses Nurses

All providers showed moderate accuracy for patients’ pain, but when patients experienced high levels of pain, accuracy was poor. Nurses showed moderate accuracy for patients’ pain.

Kuwait

Heikkinen, Salantera, Kettu, & Taittonen [104] Heuss, Sughanda, & Degen [42]

2005 45

Finland

2012

Higginson & McCarthy [105] Hodgkins, Albert, & Daltroy [106] Horgas & Dunn [33]

1993 1985

103

Adults

1993

199

Acute

Children and adults Adults

2

Nurses

Nurses showed excellent accuracy for patients’ pain on two types of pain assessments.

222

Acute

Adults

21

Physicians, nurses

84 21

Chronic Acute

Adults Adults

84

Teams Physicians

Both physicians and nurses showed moderate accuracy for patients’ pain. The physicians and nurses' Switzerland ratings were better correlated with each other than with the patients. Teams of providers showed moderate accuracy for patients’ pain. UK Physicians showed good accuracy for patients’ pain. USA

2001 45

Chronic

16

Nurses

Nurses showed very poor accuracy for patients’ pain.

USA

2005 209

Acute

Older adults Adults

97

Nurses

Nurses showed good accuracy for patients’ pain.

Sweden

1991

Acute

Children

Physicians, nurses

Both nurses and physicians showed good accuracy for patients’ pain.

USA

13

USA

3

Idvall, Berg, Unosson, & Brudin [107] LaMontagne, Johnson, & Hepworth [43]

1991

M.A. Ruben et al. / Patient Education and Counseling xxx (2015) xxx–xxx

Please cite this article in press as: M.A. Ruben, et al., Healthcare providers’ accuracy in assessing patients’ pain: A systematic review, Patient Educ Couns (2015), http://dx.doi.org/10.1016/j.pec.2015.07.009

Table 1 Pain assessment accuracy studies with patient and provider characteristics.

Main findings

Country

Physicians or nurses

Physicians and nurses showed moderate accuracy for patients’ pain.

Norway

42

Nurses, other

Children Adults

1

Other Physicians

Nurses' showed moderate accuracy for patients' pain when the patient was female and old, but did Germany not have dementia. Nurses showed poor accuracy for patients’ pain when patients were a female with dementia, a young male or a young female. Accuracy was highest and moderate prior to and during the painful event but poor during recovery. USA Physicians showed very poor accuracy for patients’ pain. USA

Acute

Adults

200

Physicians

Acute Acute and chronic

Children Adults

85 28

Nurses Physicians

Patient N

Laugsand, Sprangers, Bjordal, Skorpen, Kaasa, & Klepstad [44] Lautenbacher, Niewelt, & Kunz [31]

2010

1918

2013

60

Acute

Adults

LeBaron & Zeltzer [55] Litwin, Lubeck, Henning, & Carroll [34] Maguire, Morrell, Westhoff, & Davis [37] Manne, Jacobsen, & Redd [27] Mantyselka, Kumpusalo, Ahonen, & Takala [50]

1984 1998

50 1920

Acute Chronic

2014

200

McGee, Howie, Ryan, Moss, & Holubowycz [54] McKinley & Botti [49]

2002 1117

Melotti, Dekel, Carosi, Ricchi, Chiari, D'Andrea, & DiNino [38] Miller [56]

2009 869

Nekolaichuk, Bruera, Spachynski, MacEachern, Hanson, & Maguire [45] Oi-Ling, Man-Wah, & Kam-Hung [108] Olden, Jordan, Sakima, & Grass [109] Paice, Mahon, & Faut-Callahan [110] Perreault & Dionne [111] Powers [112] Rhondali, Hui, Kim, Kilgore, Kang, Nguyen, & Bruera [113] Rundshagen, Schnabel, Standl, & Schulte am Esch [114] Salmon & Manyande [59] Schneider & LoBiondo-Wood [28] Sheiner, Sheiner, Shohan-Vardi, Mazor, & Katz [115] Singer, Gulla, & Thode [26] Singer, Richman, Kowalska, & Thode [116] Sjostrom, Haljamae, Dahlgren, & Lindstrom [46] Sneeuw, Aaronson, Sprangers, Detmar, Wever, & Schornagel [47] Stephenson [117]

1992 85 2001 738

1991

115

1996

20

1999

49

Pain type

Patient age group

Provider total N

Adults

Adults Acute and chronic

Acute

Physicians showed poor accuracy when judging patients’ maximum rated pain during intrauterine USA device insertion. Nurses showed good accuracy for patients’ pain. USA Physicians showed moderate accuracy for patients with acute pain but fair accuracy for patients with Finland chronic pain.

Physicians

Adults

97

Adults and children Children

115

20

Adults

30

Physicians showed good accuracy for patients’ pain. Physicians showed lower accuracy when the Australia patient had other comorbid health problems and when patients experienced higher pain levels. Nurses' showed moderate accuracy for patients’ pain. Students were significantly more accurate at Australia Nursing students, staff nurses, associate charge assessing pain compared to staff nurses. A curvilinear relationship emerged where nurses in their first year and >10 years of experience had poorer accuracy than those with 1-10 years of experience. nurses Chronic pain patients were assessed more accurately than acute pain patients. Nurses Nurses showed moderate accuracy for patients’ pain. This was true of all age groups, except the 91- Italy 99 age group, where there was very poor accuracy for patients’ pain.

Nurses Physicians, nurses

Nurses showed moderate accuracy for patients’ pain on days 1 and 2 after surgery. On day 3, nurses' USA accuracy declined. Physicians and nurses showed good accuracy for patients’ pain. Canada

Physicians

Physicians showed moderate accuracy for patients’ pain.

China

Nurses

Nurses showed poor accuracy for patients’ pain.

USA

2005 30

Chronic

Adults

1995

20

Acute

Adults

1991

30

Acute

Adults

68

Physicians, nurses

Physicians showed fair accuracy while nurses showed poor accuracy for patients’ pain.

USA

2005 78 1987 50 2012 118

Chronic Acute Chronic

Adults Children Adults

9 33 20

Physio-therapists Nurses Nurses

Physiotherapists showed good accuracy for patients’ pain. Nurses showed moderate accuracy for patients’ pain. Nurses showed good accuracy for patients’ pain.

Canada Canada USA

1999

60

Acute

Adults

Nurses

Nurses showed moderate to good accuracy for patients' pain.

Germany

1996 1992

56 40

Acute Acute

Adults Children

Nurses Nurses

Nurses showed good accuracy for patients’ pain. Nurses showed good accuracy for patients’ pain.

UK USA

1999

447

Acute

Adults

Physicians, midwives

Physicians and midwives showed good to excellent accuracy for Jewish and Bedouin patients’ pain. Israel

2002 57 1999 1104

Acute Acute

Children Adults

114

Physicians, nurses Physicians

Physicians and nurses showed poor accuracy for patients’ pain. Physicians showed a variable accuracy for patients’ pain from fair to good.

1997

180

Acute

Adults

60

Physicians, nurses

1999

90

Chronic

Adults

140

Physicians, nurses

More experienced nurses and physicians showed excellent accuracy while less experienced nurses Sweden showed moderate accuracy and less experienced physicians showed good accuracy for patients’ pain. Nurses and physicians showed good accuracy for patients’ pain with nurses showing higher accuracy Netherlands than physicians.

1994

25

Acute

Adults

11

Nurses

15 40

Nurses showed moderate accuracy for patients’ pain.

USA USA

USA

M.A. Ruben et al. / Patient Education and Counseling xxx (2015) xxx–xxx

Provider population

Year

G Model PEC 5100 No. of Pages 10

4

Please cite this article in press as: M.A. Ruben, et al., Healthcare providers’ accuracy in assessing patients’ pain: A systematic review, Patient Educ Couns (2015), http://dx.doi.org/10.1016/j.pec.2015.07.009

Table 1 (Continued) Authors

G Model PEC 5100 No. of Pages 10

USA Nurses showed moderate accuracy for both acute pain and chronic pain patients. Nurses showed slightly more accuracy for chronic patients' pain compared to acute patients' pain.

Nurses showed moderate accuracy for patients’ pain. Nurses showed poor accuracy for patients’ pain.

Nurses showed moderate accuracy for patients’ pain. Nurses with less than two years of experience USA or more than six years of experience showed lower accuracy for patients' pain, while those with 35 years showed moderate accuracy for patients' pain. Nurses showed moderate accuracy for patients’ pain. USA

Nurses showed fair accuracy for patients’ pain.

Nurses

Nurses Nurses

Nurses

Nurses

Nurses

USA

USA Physicians showed good accuracy for patients’ pain. Physicians

USA Physicians showed fair accuracy for patients’ pain. Physicians

Netherlands Netherlands

M.A. Ruben et al. / Patient Education and Counseling xxx (2015) xxx–xxx

5

We report results from studies that reported providers’ pain assessment accuracy compared across these different characteristics including patient age, patient and provider gender, provider type, provider clinical experience, type of pain, pain intensity, and timing of pain assessment. The findings are summarized in Table 1. 3. Results 3.1. Study characteristics Sixty studies reported providers’ accuracy in assessing patients’ pain. Studies were published from 1982 to 2014. Forty-five percent of studies were conducted in the United States of America, 12% in Sweden, 7% in the Netherlands, 7% in Canada, 5% in the UK, 5% in Australia, with the remaining studies conducted in China, Denmark, Finland, Germany, Israel, Italy, Kuwait, Norway, Switzerland, or Turkey. The number of patients in each of the studies ranged from 13 to 1920 patients, with a mean of 209.92 patients (SD = 399.37). The majority of studies examined assessments of adult pain (72%). Fifteen percent of studies reported providers’ pain assessment accuracy for pain in children or adolescents, 10% specifically in older adults, and 3% for adults and children in the same study. The number of providers making assessments of patients’ pain ranged from 1 to 293, with a mean of 55.41 providers (SD = 58.84). Seventy-two percent of studies examined nurses’ accuracy in assessing patients’ pain. Fifty-two percent examined physicians’ accuracy, and 8% of studies examined other types of clinicians (midwives, physiotherapists, or administrators in healthcare).2 Fifty-three percent of studies measured accuracy in assessments of acute pain, 27% of chronic pain, 7% of studies included a mix of acute and chronic pain, and 14% did not distinguish between acute or chronic pain. Patient pain was most often assessed using a Numeric Rating Scale (NRS; 35%) or Visual Analog Scale (VAS; 57%) for the patient and the provider, while 5% of studies with children in pain employed a faces pain scale [26–28]. Three percent of studies used a color rating scale [29,30]. All assessments were made after viewing the patient in-person except one study, which had providers assess the pain of patients videotaped undergoing acute pressure pain [31].

119

42

119 Zalon [122]

1993

Acute

Older adults Adults 42 Weiner, Peterson, & Keefe [121] 1999

Acute 15 1982

Van der Does [120] van Herk, van Dijk, Biemold, Tibboel, Baar, & de Wit [35] Walkenstein [29]

1989 126 2009 174

Chronic

8

1 Adults

Acute and chronic Acute Chronic

Adults Older adults Adults

22 Adults Acute

Suarez-Almazor, Conner-Spady, 2001 105 Kendall, Russell, & Skeith [118] Sutherland, Wesley, Cole, 1988 Nesvacil, Daly, & Gepner [119] Teske, Daut, & Cleeland [51] 1983 71

5 Adults Chronic

30 293

3.2. Overall provider pain assessment accuracy Fewer than five percent of studies documented that providers had excellent pain assessment accuracy, showing the difficulty of this skill for providers. Seventy-seven percent of studies reported that providers had moderate to good accuracy when assessing patients’ pain (see Table 2). Thirteen percent of studies documented that providers had very poor or poor pain assessment accuracy [26,28,32–37]. 3.3. Characteristics of the patient and provider 3.3.1. Patient age Two studies examined providers’ pain assessment accuracy as a function of patient age. Melotti et al. [38] reported in their study of nurses that accuracy was moderate for patients of all ages. However, for patients aged 91–99 years old, nurses’ ability to accurately assess patients’ pain was poor [38]. In a similar study by Lautenbacher et al. [31], nurses were moderately accurate at assessing patients’ pain when the patient was an older female who

2 Percentages sum to more than 100% because some studies examined two or more provider types.

Please cite this article in press as: M.A. Ruben, et al., Healthcare providers’ accuracy in assessing patients’ pain: A systematic review, Patient Educ Couns (2015), http://dx.doi.org/10.1016/j.pec.2015.07.009

G Model PEC 5100 No. of Pages 10

6

M.A. Ruben et al. / Patient Education and Counseling xxx (2015) xxx–xxx

Table 2 Frequency and percentage of each level of pain assessment accuracy by provider type across studies. Type of provider Level of accuracy

Physicians

Nurses

Other

Total

Very poor Poor Fair Moderate Good Excellent Total

N (%) 1 (3.85) 4 (15.38) 2 (7.69) 9 (34.62) 10 (38.46) 0 (0) 26

N (%) 1 (2.33) 4 (9.30) 2 (4.65) 19 (44.19) 14 (32.56) 3 (6.98) 43

N (%) 0 (0) 0 (0) 0 (0) 3 (60.00) 2 (40.00) 0 (0) 5

N (%) 2 (2.70) 8 (10.81) 4 (5.41) 31 (41.89) 26 (35.14) 3 (4.05) 74

Note: “Other” provider type includes midwives, physiotherapists, and healthcare administrators. N is number of studies, % is percent of studies out of total number of studies for given provider type.

did not have dementia. Pain assessment accuracy was poor for patients with dementia. 3.3.2. Patient or provider gender No studies in this review reported provider accuracy in assessing pain separately for male and female patients or providers. 3.3.3. Provider type Of the 10 studies that documented or compared pain assessments made by both physicians and nurses of the same patients [26,39–47], only one study reported a significant difference in pain assessment accuracy by provider type [46]. Sjöström et al. [46] found that of the nurses and physicians that were less experienced (< 4 years of clinical experience), physicians showed good pain assessment accuracy while nurses showed only moderate pain assessment accuracy. There was no difference in pain assessment accuracy for nurses and physicians practicing for more than 4 years [46]. 3.3.4. Provider clinical experience In total, four studies examined the relationship between providers’ clinical experience and providers’ pain assessment accuracy. Two studies demonstrated a linear relationship between clinical experience and accuracy in assessing patients’ pain and two studies documented a curvilinear relationship. Sjöström et al. [46] documented that as both nurses and physicians gained experience, their accuracy in assessing pain improved. Favaloro and Touzel [48] showed that nurses with less than one year of experience had only moderate accuracy in assessing adolescent patients’ pain while nurses with more than one year of healthcare experience showed excellent accuracy when assessing patients’ pain [48]. Two studies indicated a curvilinear relationship between provider experience and pain assessment accuracy, such that providers with a moderate amount of clinical experience showed the highest pain assessment accuracy, while less experienced and very experienced healthcare professionals had lower pain assessment accuracy. Walkenstein [29] found that nurses with less than two years of experience or more than six years of experience had poor pain assessment accuracy, while those with 3–5 years had moderate pain assessment accuracy. McKinley and Botti [49] showed nurses in their first year and nurses with more than 10 years of experience had poorer pain assessment accuracy than those with 1–10 years of experience. 3.4. Characteristics of the pain 3.4.1. Type of pain: acute vs chronic Four studies examined pain assessment accuracy in both acute and chronic pain patients [39,49–51], but only two studies

reported pain assessment accuracy separately for chronic pain and acute pain patients. McKinley and Botti [49] found that chronic pain patients were assessed more accurately by nurses than acute pain patients. Teske et al. [51] found no difference in nurses’ pain assessment accuracy for acute and chronic pain patients. Pain assessment accuracy for both acute and chronic pain patients was in the moderate range [51]. 3.4.2. Pain intensity Four studies that examined the relationship between patients’ pain intensity and providers’ pain assessment accuracy yielded a consistent pattern; when patients were in more pain compared to less pain, providers’ accuracy was poorer. For example, Fridh and Gaston-Johansson [52] documented that midwives had moderate pain assessment accuracy when assessing patients’ pain in the earlier stages of labor, but when the labor was most intense, the midwives showed poor pain assessment accuracy [52]. Bergh and Sjöström’s [53] study showed similar findings among nurses. McGee, Howie, Ryan, Moss and Holubowycz [54] documented that on average physicians showed good pain assessment accuracy, but when patients experienced higher levels of pain or other comorbid health problems, physicians showed poorer pain assessment accuracy. Finally, Favaloro and Touzel [48] showed that nurses had good pain assessment accuracy overall, but poor pain assessment accuracy for a subset of adolescents who experienced higher levels of pain. 3.4.3. Timing of pain assessment Two studies indicated how the timing of providers’ pain assessment affected providers’ pain assessment accuracy. LeBaron and Zeltzer [55] showed that providers’ pain assessment accuracy was moderate prior to and during a painful procedure, bone marrow aspiration, but poor during patient recovery. Similarly, Miller [56] documented that nurses’ pain assessment accuracy for children’s post-surgical pain was moderate on day 1 and 2 immediately following the surgery but nurses became less accurate on post-surgical day 3. 4. Discussion and conclusion 4.1. Discussion The literature examining providers’ pain assessment accuracy showed that overall, healthcare providers had moderate to good levels of accuracy when assessing their patients’ pain. Nurses and physicians showed similar levels of pain assessment accuracy. Some characteristics of the providers, patients and pain itself were shown to influence providers’ pain assessment accuracy. Providers were less accurate at assessing pain in older patients. Future research should explore the mechanism responsible for this age difference in pain assessment accuracy. It may be that older patients are less expressive about their pain or display their pain in different ways. For example, pain expressions in older adults are sometimes interpreted as confusion, social withdrawal, or apathy [57]. Conversely, providers may be less attuned to the needs of their older patients or less sensitive to their pain. It is unclear why higher pain intensity made it more difficult for providers to accurately assess pain. Providers may be unwilling to assess patients’ pain at a high level because they are concerned about patients taking legal action for aggressive treatment of pain or they are concerned about the potential for addiction to or abuse of prescription opioids [58]. It is also possible that providers have a tendency to assume intense pain is exaggerated pain. For example, in Salmon and Manyande’s [59] study, patients who experienced the highest pain were evaluated by nurses as unpopular or demanding compared to patients who experienced lower levels of

Please cite this article in press as: M.A. Ruben, et al., Healthcare providers’ accuracy in assessing patients’ pain: A systematic review, Patient Educ Couns (2015), http://dx.doi.org/10.1016/j.pec.2015.07.009

G Model PEC 5100 No. of Pages 10

M.A. Ruben et al. / Patient Education and Counseling xxx (2015) xxx–xxx

pain. Finally, providers may have poorer pain assessment accuracy when assessing higher intensities of patients’ pain because they have less experience or training with patients experiencing higher intensities of pain. Providers also had poorer pain assessment accuracy when the assessment was made days after an acute pain experience compared to immediately after the acute pain experience. Providers may assume recovery means some patients are feeling less intense pain than they actually are, leading to poorer pain assessment accuracy. The literature in this review suggests a curvilinear relationship between providers’ clinical experience and accuracy of pain assessment. There are several possibilities as to why providers who have been in practice for many years show decreased pain assessment accuracy. One possibility is that these providers have a larger patient load and increased time constraints, which leads them to miss patient cues pertinent to pain assessment accuracy [60]. More experienced providers may be attuned to the wrong cues of pain. Another possibility is that more experienced providers minimize the pain that they observe [61] as a defense mechanism against continued exposure to patients’ suffering or negative emotions. The relationship between pain assessment accuracy and clinical experience dovetails with studies finding declines in empathy and attention to feelings as healthcare training progresses [62–64]. However, these studies generally demonstrate a decline in empathy across years in medical school, not years in practice. Providers with more clinical experience may also be older. Future studies should examine how provider age and clinical experience interact to influence pain assessment accuracy. We did not find systematic differences in pain assessment accuracy for acute and chronic pain patients. This finding is in line with research showing that acute and chronic pain patients’ facial expressions of pain during a physical examination appear to be similar [65]. It is important to note, however, that the experience of acute and chronic pain for patients can be quite different. In one study, although chronic pain patients and acute pain patients did not differ on the sensory, affective, and evaluative dimensions of the McGill Pain Questionnaire (MPQ), chronic pain patients with higher levels of pain on all three of the MPQ dimensions had significantly higher levels of state anxiety and depression [66]. Although the sensory, affective, and evaluative experience of pain did not differ among acute and chronic pain patients, higher levels of anxiety and depression may make it more difficult for providers to be accurate when assessing chronic pain patients’ pain. Chronic pain patients’ expressions of pain may be complicated by expressions related to their depressive symptoms. 4.1.1. Limitations of the current review and future research directions This systematic review suggests that additional research is needed to fully understand how characteristics of the patient, provider and pain experience influence providers’ pain assessment accuracy. Only a small proportion of studies included in this review specifically examined the relationship between these characteristics and pain assessment accuracy. Future studies should examine providers’ pain assessment accuracy by characteristics of the patient including their gender and race. Female patients are known to report and experience higher levels of acute and chronic pain compared to male patients [67,68] and there are known racial disparities in pain management [69]. The studies included in this review assessed pain using various instruments (e.g., NRS, VAS, faces and color pain scales). All of these scales generally measure pain intensity but may tap different aspects of pain or may be used differently by patients depending on the scale. For example, the midpoint of the faces pain scale may be used or interpreted differently by children compared to the midpoint of an adult VAS or NRS. Although there is preliminary

7

work showing that these scales are often well correlated with each other [70], it is possible that they are measuring or being used differently. Even more standardization of pain assessment tools in clinical interactions and research studies would not only help researchers understand factors that contribute to pain assessment accuracy but would also help providers and patients as they would become familiar with the standardized pain assessment instrument across the lifespan and across their clinical training. In addition to the use of different scales to measure pain intensity, a variety of scales measure different aspects of the pain experiencing including the duration, location, and sensation of pain. No studies examined these aspects of pain in relation to accuracy, thus this review is limited to understanding provider accuracy in pain intensity, as it is the most widely used component of pain assessed in clinical and research settings. Additional studies should examine providers’ pain assessment accuracy for these additional dimensions of patient pain. In addition, none of the studies in this review examined what specific cues providers used when making their assessments of pain (e.g., patients’ nonverbal behavior, patients’ verbal behavior, patients’ injury/disease, physiological indicators) and whether this influenced accuracy. Nonverbal behaviors are powerful indicators of the pain experience and for infants appear to be the most consistent expression of pain [71]. Perhaps providers are more accurate at assessing pain when they focus on nonverbal expressions of pain, as nonverbal behaviors may be more spontaneous and less under the volitional control of the sufferer [72]. Studies should also examine providers’ pain assessment accuracy for nonverbal patient populations (e.g., patients with dementia [73], Alzheimer’s [74], or seniors with limited communication abilities [75]). Coding systems and behavioral checklists have been designed to measure pain using nonverbal behaviors in infants and for patients with cognitive impairments [76–78]. This review focused exclusively on accuracy as defined as the direct comparison between providers’ assessments of patients’ pain and patients’ self-reported pain (the criterion), given that this is how the literature measures patients’ pain. However, future research on providers’ pain assessment accuracy would benefit from exploring other criterion besides patients’ self-report. Providers’ pain assessments could be compared to patients’ physiological responses to pain, nonverbal expressions of pain, fMRI responses to pain, or caregiver/proxy reports of pain. This review was also limited to studies written in English because the literature search did not identify studies written in other languages. Also, the majority of studies took place in the United States or Western Europe. Therefore, results may not generalize to providers in other countries. Although pain assessment accuracy was classified from poor to excellent based on the overall effect size, existing research has not determined what is a clinically relevant pain assessment accuracy score. Many studies showed moderate to good pain assessment accuracy; however, there is room for improvement as we know problems of undertreatment [79] and misinterpretation of pain cues still exist for many patients [80]. 4.1.2. Training to improve pain assessment accuracy With only a handful of studies showing excellent pain assessment accuracy and probable decline in accuracy over time, there is a need for trainings to improve providers’ pain assessment accuracy. The Institute of Medicine [1] recommended the development of educational opportunities for primary care practitioners and other providers to improve their knowledge and skills in pain assessment and treatment, including safe and effective opioid prescribing. Pain assessment accuracy is a skill amenable to training [81,82]. Short trainings focused on nonverbal facial expressions of pain

Please cite this article in press as: M.A. Ruben, et al., Healthcare providers’ accuracy in assessing patients’ pain: A systematic review, Patient Educ Couns (2015), http://dx.doi.org/10.1016/j.pec.2015.07.009

G Model PEC 5100 No. of Pages 10

8

M.A. Ruben et al. / Patient Education and Counseling xxx (2015) xxx–xxx

have shown promise in improving pain assessment accuracy in non-clinical settings [83–85]. And Bergh and Sjöström [53] demonstrated that nurses who had received pain management training had higher pain assessment accuracy scores compared to nurses who had not completed pain management training. The probable curvilinear relationship between years of clinical experience and accuracy in pain assessment suggest that training in pain assessment should occur both in medical school for those providers with minimal clinical experience [46,48] and in continuing medical education to provide a refresher for those with many years of clinical experience [49]. Only one study in this review evaluated pain assessment accuracy with a videotaped test of patients in pain [31]. Although it does not occur during a clinical interaction, this method of measuring accuracy is valuable to understanding providers’ pain assessment accuracy as a skill. In clinical studies, pain assessment accuracy is measured across many different patients making it difficult to disentangle whether providers with high pain assessment accuracy were actually more accurate or were interacting with patients’ who were better at expressing their pain. Videotaped tests of pain assessment accuracy use a standard set of patients so accuracy can be compared across providers. Standardized tests using videotaped patients in other domains (e.g., assessing patients’ emotions) have been used and accepted in medical education and medical training [86,87], yet a validated videotaped test of pain assessment accuracy does not currently exist in the healthcare field and could be valuable for assessment and training purposes. 4.2. Conclusion This review showed that healthcare providers had moderate to good levels of accuracy when assessing patients’ pain. The type of provider did not seem to influence pain assessment accuracy, although having limited clinical experience or a great deal of clinical experience was associated with poorer accuracy. Furthermore, providers had poorer pain assessment accuracy when assessing older patients and patients with cognitive disabilities. Providers also showed poorer pain assessment accuracy when assessing patients experiencing higher pain intensities. Findings suggest that there is some room for improvement to address providers’ pain assessment accuracy by implementing education and training in medical schools and continuing medical education, and also by promoting the use of available and appropriate methodological instruments tailored to specific vulnerable patient populations. 4.3. Practice implications Higher levels of pain assessment accuracy in healthcare providers may mean more appropriate pain management, less undertreatment of pain, and better mental and physical health outcomes for patients. Although the research on clinical outcomes associated with providers’ pain assessment accuracy has not been conducted, two studies examined the role of caregivers’ pain assessment accuracy on patients’ mental and physical health outcomes. Miaskowski, Zimmer, Barrett, Dibble, and Wallhagen [88] found that when caregivers had lower levels of pain assessment accuracy, patients had more mood disturbance and poorer quality of life. Riemsma, Taal, and Rasker [89] similarly found that both over- and underestimation of pain by partners was related to patients’ poorer mental health status. Treating patients’ pain requires the provider to perceive the cues to pain, interpret those cues to make an accurate assessment of the subjective experience of that pain and then respond accordingly [82]. Pain assessment accuracy is one aspect of providers’ ability to

accurately perceive the characteristics of their patients, a skill known as person perception accuracy. Person perception accuracy in the clinical context has become a growing area of research because providers who are more accurate at assessing their patients’ verbal and nonverbal cues have patients who are more satisfied with their care and who have better health outcomes [90,91]. Like most types of person perception, pain assessment is a skill that some providers are more accurate at than others. By targeting providers who show poor pain assessment accuracy and by targeting vulnerable patient populations, we may be able to improve providers’ abilities to assess and treat their patients’ pain and begin to ameliorate the burden of acute and chronic pain that plagues our healthcare system. Funding sources there were no funding sources for the current research. Conflict of interests There are no conflicts of interest. References [1] P.A. Pizzo, N.M. Clark, O. Carter-Pokras, Relieving Pain in America: A Blueprint for Transforming Prevention, Care, Education, and Research, The National Academies Press, Washington, D.C, 2011, doi:http://dx.doi.org/10.3109/ 15360288.2012.678473. [2] H. Breivik, B. Collett, V. Ventafridda, R. Cohen, D. Gallacher, Survey of chronic pain in Europe: prevalence, impact on daily life, and treatment, Eur. J. Pain 10 (2006) 287–333, doi:http://dx.doi.org/10.1016/j. ejpain.2005.06.009. [3] C.D. Fox, D. Berger, P.G. Fine, G.F. Gebhardt, M. Brabois, R.I. Kulich, et al., Pain Assessment and Treatment in the Managed Care Environment. A Position Statement from the American Pain Society, American Pain Society, Glenview, IL, 2000. [4] S. Deandrea, M. Montanari, L. Moja, G. Apolone, Prevalence of undertreatment in cancer pain. A review of published literature, Ann. Oncol. 19 (2008) 1985–1991, doi:http://dx.doi.org/10.1093/annonc/mdn419. [5] W. Breitbart, B.D. Rosenfeld, S.D. Passik, M.V. McDonald, H. Thaler, R.K. Portenoy, The undertreatment of pain in ambulatory AIDS patients, Pain 65 (1996) 243–249, doi:http://dx.doi.org/10.1016/0304-3959(95) 00217-0. [6] F. Larue, A. Fontaine, S.M. Colleau, Underestimation and undertreatment of pain in HIV disease: multicentre study, Br. Med. J. 314 (1997) 23–28, doi: http://dx.doi.org/10.1136/bmj.314.7073.23. [7] R. Stalnikowicz, R. Mahamid, S. Kaspi, M. Brezis, Undertreatment of acute pain in the emergency department: a challenge, Int. J. Qual. Health Care 17 (2005) 173–176, doi:http://dx.doi.org/10.1093/intqhc/mzi022. [8] N.L. Schechter, The undertreatment of pain in children: an overview, Pediatr. Clin. N. Am. 36 (1989) 781–794. [9] R.S. Morrison, A.L. Siu, A comparison of pain and its treatment in advanced dementia and cognitively intact patients with hip fracture, J. Pain Symptom Manag. 19 (2000) 240–248, doi:http://dx.doi.org/10.1016/S0885-3924(00) 00113-5. [10] F. Brennan, D.B. Carr, M. Cousins, Pain management: a fundamental human right, Anesth. Analg. 105 (2007) 205–221, doi:http://dx.doi.org/10.1213/01. ane.0000268145.52345.55. [11] I. Tracey, P.W. Mantyh, The cerebral signature for pain perception and its modulation, Neuron 55 (2007) 377–391, doi:http://dx.doi.org/10.1016/j. neuron.2007.07.012. [12] W. Kwok, T. Bhuvanakrishna, The relationship between ethnicity and the pain experience of cancer patients: a systematic review, Indian J. Palliat. Care 20 (2014) 194–200. [13] T. Hadjistavropoulos, L.M. Breau, K.D. Craig, Assessment of pain in adults and children with limited ability to communicate, in: D.C. Turk, R. Melzack (Eds.), Handbook of Pain Assessment, Guilford Press, New York, 2011, pp. 260–280. [14] K. Herr, K. Bjoro, S. Decker, Tools for assessment of pain in nonverbal older adults with dementia: a state-of-the-science review, J. Pain Symptom Manag. 31 (2006) 170–192, doi:http://dx.doi.org/10.1016/j. jpainsymman.2005.07.001. [15] M. Botti, T. Bucknall, E. Manias, The problem of postoperative pain: issues for future research, Int. J. Nurs. Pract. 10 (2004) 257–263, doi:http://dx.doi.org/ 10.1111/j.1440-172x.2004.00487.x. [16] D.C. Turk, R. Melzack, The measurement of pain and the assessment of people experiencing pain, in: D.C. Turk, R. Melzack (Eds.), Handbook of Pain Assessment, Guilford Press, New York, 2011, pp. 3–16.

Please cite this article in press as: M.A. Ruben, et al., Healthcare providers’ accuracy in assessing patients’ pain: A systematic review, Patient Educ Couns (2015), http://dx.doi.org/10.1016/j.pec.2015.07.009

G Model PEC 5100 No. of Pages 10

M.A. Ruben et al. / Patient Education and Counseling xxx (2015) xxx–xxx [17] I. Tracey, Getting the pain you expect: mechanisms of placebo, nocebo and reappraisal effects in humans, Nat. Med. 16 (2010) 1277–1283, doi:http://dx. doi.org/10.1038/nm.2229. [18] J.A. Turner, R.A. Deyo, J.D. Loeser, M. Von Korff, W.E. Fordyce, The importance of placebo effects in pain treatment and research, J. Am. Med. Assoc. 271 (1994) 1609–1614, doi:http://dx.doi.org/10.1001/ jama.1994.03510440069036. [19] D.D. Price, D.G. Finniss, F. Benedetti, A comprehensive review of the placebo effect: recent advances and current thought, Annu. Rev. Psychol. 59 (2008) 565–590, doi:http://dx.doi.org/10.1146/annurev.psych.59.113006.095941. [20] D.G. Finniss, T.J. Kaptchuk, F. Miller, F. Benedetti, Biological, clinical, and ethical advances of placebo effects, Lancet 375 (2010) 686–695, doi:http://dx. doi.org/10.1016/S0140-6736(09)61706-2. [21] P.A. McGrath, Psychological aspects of pain perception, Arch. Oral Biol. 39 (Suppl. 5) (1994) 55–62, doi:http://dx.doi.org/10.1016/0003-9969(94)901899. [22] R. Melzack, The McGill pain questionnaire: major properties and scoring methods, Pain 1 (1975) 277–299, doi:http://dx.doi.org/10.1016/0304-3959 (75)90044-5. [23] T. Hadjistavropoulos, K.D. Craig, A theoretical framework for understanding self-report and observational measures of pain: a communications model, Behav. Res. Ther. 40 (2002) 551–570, doi:http://dx.doi.org/10.1016/S00057967(01)00072-9. [24] M. Schiavenato, K.D. Craig, Pain assessment as a social transaction: beyond the gold standard, Clin. J. Pain 26 (2010) 667–676, doi:http://dx.doi.org/ 10.1097/AJP.0b013e3181e72507. [25] U. Nworah, From documentation to the problem: controlling postoperative pain, Nurs. Forum 47 (2012) , doi:http://dx.doi.org/10.1111/j.17446198.2012.00262.x. [26] A.J. Singer, J. Gulla, H.C. Thode, Parents and practitioners are poor judges of young children’s pain severity, Acad. Emerg. Med. 9 (2002) 609–612, doi: http://dx.doi.org/10.1197/aemj.9.6.609. [27] S.L. Manne, P.B. Jacobsen, W.H. Redd, Assessment of acute pediatric pain: do child self-report, parent ratings, and nurse ratings measure the same phenomenon? Pain 48 (1992) 45–52, doi:http://dx.doi.org/10.1016/03043959(92)90130-4. [28] E.M. Schneider, G. LoBiondo-Wood, Perceptions of procedural pain: parents, nurses, and children, Child Health Care 21 (1992) 157–162, doi:http://dx.doi. org/10.1207/s15326888chc2103_5. [29] M.D. Walkenstein, Comparison of burned patients’ perception of pain with nurses’ perception of patients’ pain, J. Burn Care Rehabil. 3 (1982) 233–236. [30] C. Colwell, L. Clark, R. Perkins, Postoperative use of pediatric pain scales: children’s self-report versus nurse assessment of pain intensity and affect, J. Pediatr. Nurs. 11 (1996) 375–382. [31] S. Lautenbacher, B.G. Niewelt, M. Kunz, Decoding pain from the facial display of patients with dementia: A comparison of professional and nonprofessional observers, Pain Med (U. S.) 14 (2013) 469–477, doi:http://dx.doi.org/10.1111/ pme.12050. [32] J.K. Cremeans-Smith, M.A.P. Stephens, M.M. Franks, L.M. Martire, J.A. Druley, W.C. Wojno, Spouses’ and physicians’ perceptions of pain severity in older women with osteoarthritis: dyadic agreement and patients’ well-being, Pain 106 (2003) 27–34, doi:http://dx.doi.org/10.1016/S0304-3959(03)00268-9. [33] A.L. Horgas, K. Dunn, Pain in nursing home residents. Comparison of residents’ self-report and nursing assistants’ perceptions. Incongruencies exist in resident and caregiver reports of pain; therefore, pain management education is needed to prevent suffering, J. Gerontol. Nurs. 27 (2001) 44–53. [34] M.S. Litwin, D.P. Lubeck, J.M. Henning, P.R. Carroll, Differences in urologist and patient assessments of health related quality of life in men with prostate cancer: results of the CaPSURE database, J. Urol. 159 (1998) 1988–1992, doi: http://dx.doi.org/10.1016/S0022-5347(01)63222-1. [35] R. Van Herk, M. Van Dijk, N. Biemold, D. Tibboel, F.P. Baar, R. De Wit, Assessment of pain: can caregivers or relatives rate pain in nursing home residents, J. Clin. Nurs. 18 (2009) 2478–2485, doi:http://dx.doi.org/10.1111/ j.1365-2702.2008.02776.x. [36] E.K. Fromme, K.M. Eilers, M. Mori, Y.C. Hsieh, T.M. Beer, How accurate is clinician reporting of chemotherapy adverse effects? A comparison with patient-reported symptoms from the quality-of-life questionnaire C30, J. Clin. Oncol. 22 (2004) 3485–3490, doi:http://dx.doi.org/10.1200/JC.O 200403. [37] K. Maguire, K. Morrell, C. Westhoff, A. Davis, Accuracy of providers’ assessment of pain during intrauterine device insertion, Contraception 89 (2014) 22–24, doi:http://dx.doi.org/10.1016/j.contraception.2013.09.008. [38] R.M. Melotti, B.G.S. Dekel, F. Carosi, E. Ricchi, P. Chiari, R. D’Andrea, et al., Categories of congruence between inpatient self-reported pain and nurses evaluation, Eur. J. Pain 13 (2009) 992–1000, doi:http://dx.doi.org/10.1016/j. ejpain.2008.11.019. [39] R.A. Drayer, J. Henderson, M. Reidenberg, Barriers to better pain control in hospitalized patients, J. Pain Symptom Manag. 17 (1999) 434–440, doi:http:// dx.doi.org/10.1016/S0885-3924(99)00022-6. [40] G. Ewing, M. Rogers, S. Barclay, J. McCabe, A. Martin, M. Campbell, et al., Palliative care in primary care: a study to determine whether patients and professionals agree on symptoms, Br. J. Gen. Pract. 56 (2006) 27–34. [41] S.A. Grossman, V.R. Sheidler, K. Swedeen, J. Mucenski, S. Piantadosi, Correlation of patient and caregiver ratings of cancer pain, J. Pain Symptom Manag. 6 (1991) 53–57, doi:http://dx.doi.org/10.1016/0885-3924 (91)90518-9.

9

[42] L.T. Heuss, S.P. Sughanda, L.P. Degen, Endoscopy teams’ judgment of discomfort among patients undergoing colonoscopy—how bad was it really? Swiss Med. Wkly. 142 (2012) , doi:http://dx.doi.org/10.4414/ smw.2012.13726. [43] L.L. LaMontagne, B.D. Johnson, J.T. Hepworth, Children’s ratings of postoperative pain compared to ratings by nurses and physicians, Issues Compr. Pediatr. Nurs. 14 (1991) 241–247. [44] E.A. Laugsand, M.A.G. Sprangers, K. Bjordal, F. Skorpen, S. Kaasa, P. Klepstad, Health care providers underestimate symptom intensities of cancer patients: a multicenter European study, Health Qual. Life Outcomes 8 (2010) 104, doi: http://dx.doi.org/10.1186/1477-7525-8-104. [45] C.L. Nekolaichuk, E. Bruera, K. Spachynski, T. MacEachern, J. Hanson, T.O. Maguire, A comparison of patient and proxy symptom assessments in advanced cancer patients, Palliat. Med. 13 (1999) 311–323, doi:http://dx.doi. org/10.1191/026921699675854885. [46] B. Sjöström, H. Haljamäe, L.O. Dahlgren, B. Lindström, Assessment of postoperative pain: impact of clinical experience and professional role, Acta Anaesthesiol. Scand. 41 (1997) 339–344, doi:http://dx.doi.org/10.1111/ j.1399-6576.1997.tb04695.x. [47] K.C. Sneeuw, N.K. Aaronson, M.A. Sprangers, S.B. Detmar, L.D. Wever, J.H. Schornagel, Evaluating the quality of life of cancer patients: assessments by patients, significant others, physicians and nurses, Br. J. Cancer 81 (1999) 87– 94, doi:http://dx.doi.org/10.1038/sj.bjc.6690655. [48] R. Favaloro, B. Touzel, A comparison of adolescents’ and nurses’ postoperative pain ratings and perceptions, Pediatr. Nurs. 16 (1990) 414–424. [49] S. McKinley, M. Botti, Nurses’ assessment of pain in hospitalised patients, Aust. J. Adv. Nurs. 9 (1991) 8–14. [50] P. Mäntyselkä, E. Kumpusalo, R. Ahonen, A. Kumpusalo, J. Kauhanen, H. Viinamäki, et al., Pain as a reason to visit the doctor: a study in Finnish primary health care, Pain 89 (2001) 175–180, doi:http://dx.doi.org/10.1016/ S0304-3959(00)00361-4. [51] K. Teske, R.L. Daut, C.S. Cleeland, Relationships between nurses’ observations and patients’ self-reports of pain, Pain 16 (1983) 289–296, doi:http://dx.doi. org/10.1016/0304-3959(83)90117-3. [52] G. Fridh, F. Gaston-Johansson, Do primiparas and multiparas have realistic expectations of labor, Acta Obstet. Gynecol. Scand. 69 (1990) 103–109. [53] I. Bergh, B. Sjöström, A comparative study of nurses’ and elderly patients’ ratings of pain and pain tolerance, J. Gerontol. Nurs. 25 (1999) 30–36. [54] M.A. McGee, D.W. Howie, P. Ryan, J.R. Moss, O.T. Holubowycz, Comparison of patient and doctor responses to a total hip arthroplasty clinical evaluation questionnaire, J. Bone Joint Surg. Am. 84-A (2002) 1745–1752. [55] S. LeBaron, L. Zelter, Assessment of acute pain and anxiety in children and adolescents by self-reports, observer reports, and a behavior checklist, J. Consult Clin. Psychol. 52 (1984) 729–738, doi:http://dx.doi.org/10.1037/ 0022-006X.52.5. [56] D. Miller, Practice applications of researchComparisons of pain ratings from postoperative children, their mothers, and their nurses, Pediatr. Nurs. 22 (1996) 145–149. [57] M. Wallace, Assessment and management of pain in the elderly, Medsurg. Nurs. 3 (1994) 293–298. [58] C.C. Upshur, R.S. Luckmann, J.A. Savageau, Primary care provider concerns about management of chronic pain in community clinic populations, J. Gen. Intern. Med. 21 (2006) , doi:http://dx.doi.org/10.1111/j.15251497.2006.00412.x. [59] P. Salmon, A. Manyande, Good patients cope with their pain: postoperative analgesia and nurses’ perceptions of their patients’ pain, Pain 68 (1996) 63– 68, doi:http://dx.doi.org/10.1016/S0304-3959(96)03171-5. [60] E.I. Schafheutle, J.A. Cantrill, P.R. Noyce, Why is pain management suboptimal on surgical wards? J. Adv. Nurs. 33 (2001) 728–737, doi:http://dx.doi.org/ 10.1046/j.1365-2648.2001.01714.x. [61] L. Goubert, K.D. Craig, A. Buysse, J. Decety, W. Ickes, Perceiving others in pain: experimental and clinical evidence on the role of empathy, Soc. Neurosci. Empathy (2009) 153–165. [62] M. Neumann, F. Edelhäuser, D. Tauschel, M.R. Fischer, M. Wirtz, C. Woopen, et al., Empathy decline and its reasons: a systematic review of studies with medical students and residents, Acad. Med. 86 (2011) 996–1009, doi:http:// dx.doi.org/10.1097/ACM.0b013e318221e615. [63] M. Hojat, M.J. Vergare, K. Maxwell, G. Brainard, S.K. Herrine, G. Isenberg a, et al., The devil is in the third year: a longitudinal study of erosion of empathy in medical school, Acad. Med. 84 (2009) 1182–1191. [64] T.D. Stratton, J.A. Saunders, C.L. Elam, Changes in medical students’ emotional intelligence: an exploratory study, Teach. Learn. Med. 20 (2008) 279–284, doi:http://dx.doi.org/10.1080/10401330802199625. [65] H.D. Hadjistavropoulos, K.D. Craig, Acute and chronic low back pain: cognitive, affective, and behavioral dimensions, J. Consult. Clin. Psychol. 62 (1994) 341–349, doi:http://dx.doi.org/10.1037//0022-006X.62.2. [66] M.D. Ackerman, M.J. Stevens, Acute and chronic pain: pain dimensions and psychological status, J. Clin. Psychol. 45 (1989) 223–228. [67] P. Croft, F.M. Blyth, D. van der Windt, Chronic Pain Epidemiology: From Aetiology to Public Health, Oxford University Press, Oxford, 20109–18. [68] R.B. Fillingim, C.D. King, M.C. Ribeiro-Dasilva, B. Rahim-Williams, J.L. Riley III, Sex, gender, and pain: a review of recent clinical and experimental findings, J. Pain 10 (2009) 447–485, doi:http://dx.doi.org/10.1016/j.jpain.2008.12.001. [69] R.C. Tait, J.T. Chibnall, Racial/ethnic disparities in the assessment and treatment of pain: psychosocial perspectives, Am. Psychol. 69 (2014) 131– 141, doi:http://dx.doi.org/10.1037/a0035204.

Please cite this article in press as: M.A. Ruben, et al., Healthcare providers’ accuracy in assessing patients’ pain: A systematic review, Patient Educ Couns (2015), http://dx.doi.org/10.1016/j.pec.2015.07.009

G Model PEC 5100 No. of Pages 10

10

M.A. Ruben et al. / Patient Education and Counseling xxx (2015) xxx–xxx

[70] M.A. Ferreira-Valente, J.L. Pais-Ribeiro, M.P. Jensen, Validity of four pain intensity rating scales, Pain 152 (2011) 2399–2404, doi:http://dx.doi.org/ 10.1016/j.pain.2011.07.005. [71] K.D. Craig, The facial expression of pain better than a thousand words? APS J. 1 (1992) 153–162, doi:http://dx.doi.org/10.1016/1058-9139(92)90001-S. [72] B.M. DePaulo, Nonverbal behavior and self-presentation, Psychol. Bull. 111 (1992) 203–243, doi:http://dx.doi.org/10.1037/0033-2909.111.2.203. [73] V. Warden, A.C. Hurley, L. Volicer, Development and psychometric evaluation of the pain assessment in advanced dementia (PAINAD) scale, J. Am. Med. Dir. Assoc. 4 (2003) 9–15, doi:http://dx.doi.org/10.1097/01. JAM.0000043422.31640.F7. [74] P. Lane, M. Kuntupis, S. MacDonald, P. McCarthy, J.A. Panke, V. Warden, et al., A pain assessment tool for people with advanced Alzheimer’s and other progressive dementias, Home Healthc. Nurse 21 (2003) 32–37, doi:http://dx. doi.org/10.1097/00004045-200301000-00007. [75] S. Fuchs-Lacelle, T. Hadjistavropoulos, Development and preliminary validation of the pain assessment checklist for seniors with limited ability to communicate (PACSLAC), Pain Manag. Nurs. 5 (2004) 37–49, doi:http://dx. doi.org/10.1016/j.pmn.2003.10.001. [76] K. Bjoro, K. Herr, Assessment of pain in the nonverbal or cognitively impaired older adult, Clin. Geriatr. Med. 24 (2008) 237–262, doi:http://dx.doi.org/ 10.1016/j.cger.2007.12.001. [77] D. Li, K. Puntillo, C. Miaskowski, A review of objective pain measures for use with critical care adult patients unable to self-report, J. Pain 9 (2008) 2–10, doi:http://dx.doi.org/10.1016/j.jpain.2007.08.009. [78] L.A. Hatfield, E.A. Ely, Measurement of acute pain in infants: a review of behavioral and physiological variables, Biol. Res. Nurs. 17 (2015) 100–111. [79] K.E. Bakkelund, E. Sundland, S. Moen, G. Vangberg, S. Mellesmo, P. Klepstad, Undertreatment of pain in the prehospital: setting a comparison between trauma patients and patients with chest pain, Eur. J. Emerg. Med. 20 (2013) 428–430, doi:http://dx.doi.org/10.1097/MEJ.0b013e32835c9fa3. [80] A. Werner, K. Malterud, It is hard work behaving as a credible patient: encounters between women with chronic pain and their doctors, Soc. Sci. Med. 57 (2003) 1409–1419, doi:http://dx.doi.org/10.1016/S0277-9536(02) 520-8. [81] D. Blanch-Hartigan, S.A. Andrzejewski, K.M. Hill, The effectiveness of training to improve person perception accuracy: a meta-analysis, Basic Appl. Soc. Psych. 34 (2012) 483–498, doi:http://dx.doi.org/10.1080/ 01973533.2012.728122. [82] D. Blanch-Hartigan, M.A. Ruben, Training clinicians to accurately perceive their patients: current state and future directions, Patient Educ. Couns. 92 (2013) 328–336, doi:http://dx.doi.org/10.1016/j.pec.2013.02.010. [83] M.L. Hill, K.D. Craig, Detecting deception in pain expressions: the structure of genuine and deceptive facial displays, Pain 98 (2002) 135–144, doi:http://dx. doi.org/10.1016/S0304-3959(02)00037-4. [84] K.E. Galin, B.E. Thorn, Unmasking pain: detection of deception in facial expressions, J. Soc. Clin. Psychol. 12 (1993) 182–197, doi:http://dx.doi.org/ 10.1521/jscp.1993.12.2.182. [85] P.E. Solomon, K.M. Prkachin, V. Farewell, Enhancing sensitivity to facial expression of pain, Pain 71 (1997) 279–284, doi:http://dx.doi.org/10.1016/ S0304-3959(97)03377-0. [86] D. Blanch-Hartigan, Measuring providers’ verbal and nonverbal emotion recognition ability: reliability and validity of the patient emotion cue test (PECT), Patient Educ. Couns. 82 (2011) 370–376. [87] J.A. Hall, A.N. Ship, M.A. Ruben, E.M. Curtin, S. Clever, D.L. Roter, et al., The test of accurate perception of patients’ affect (TAPPA): an ecologically valid tool for assessing interpersonal perception accuracy in clinicians, Patient Educ. Couns. 94 (2014) 218–223. [88] C. Miaskowski, E.F. Zimmer, K.M. Barrett, S.L. Dibble, M. Wallhagen, Differences in patients’ and family caregivers’ perceptions of the pain experience influence patient and caregiver outcomes, Pain 72 (1997) 217– 226, doi:http://dx.doi.org/10.1016/S0304-3959(97)00037-7. [89] R.P. Riemsma, E. Taal, J.J. Rasker, Perceptions about perceived functional disabilities and pain of people with rheumatoid arthritis: differences between patients and their spouses and correlates with well-being, Arthritis Care Res. 13 (2000) 255–261. [90] J.A. Hall, F.J. Bernieri, Interpersonal Sensitivity: Theory and Measurement, Lawrence Erlbaum Associates, Inc., 2001. [91] J.A. Hall, Clinicians’ accuracy in perceiving patients: Its relevance for clinical practice and a narrative review of methods and correlates, Patient Educ. Couns. 84 (2011) 319–324, doi:http://dx.doi.org/10.1016/j.pec.2011.03.006. [92] J. Cohen. Statistical power analysis for the behavioral sciences. vol. 2nd. 1988. doi:10.1234/12345678. [93] K.O. McGraw, S.P. Wong, Forming inferences about some intraclass correlation coefficients, Psychol. Methods 1 (1996) 30–46, doi:http://dx.doi. org/10.1037/1082-989X.1.1.30. [94] J.R. Landis, G.G. Koch, The measurement of observer agreement for categorical data, Biometrics 33 (1977) 159–174, doi:http://dx.doi.org/ 10.2307/2529310. [95] S. Akin, Z. Durna, A comparative descriptive study examining the perceptions of cancer patients, family caregivers, and nurses on patient symptom severity in Turkey, Eur. J. Oncol. Nurs. 17 (2013) 30–37. [96] K. Blomqvist, I.R. Hallberg. Pain in older adults living in sheltered accommodation—agreement between assessments by older adults and staff. vol. 8. 1999.

[97] E. Bondestam, K. Hovgren, F. Gaston Johansson, S. Jern, J. Herlitz, S. Holmberg, Pain assessment by patients and nurses in the early phase of acute myocardial infarction, J. Adv. Nurs. 12 (1987) 677–682. [98] E. Broberger, C. Tishelman, L. Von Essen, Discrepancies and similarities in how patients with lung cancer and their professional and family caregivers assess symptom occurrence and symptom distress, J. Pain Symptom Manag. 29 (2005) 572–583, doi:http://dx.doi.org/10.1016/j. jpainsymman.2004.11.006. [99] M. Choinière, R. Melzack, N. Girard, J. Rondeau, M.J. Paquin, Comparisons between patients’ and nurses’ assessment of pain and medication efficacy in severe burn injuries, Pain 40 (1990) 143–152, doi:http://dx.doi.org/10.1016/ 0304-3959(90)90065-L. [100] G.H. De Bock, H.W. van Marwijk, A.A. Kaptein, J.D. Mulder, Osteoarthritis pain assessment in family practice, Arthritis Care Res. 7 (1994) 40–45. [101] J.J. Everett, D.R. Patterson, J.A. Marvin, B. Montgomery, N. Ordonez, K. Campbell, Pain assessment from patients with burns and their nurses, J. Burn Care Rehabil. 15 (1994) 194–198. [102] M. Forrest, G. Hermann, B. Andersen, Assessment of pain: a comparison between patients and doctors, Acta Anaesthesiol. Scand. 33 (1989) 255–256. [103] A. Harrison, Comparing nurses’ and patients’ pain evaluations: a study of hospitalized patients in Kuwait, Soc. Sci. Med. 36 (1993) 683–692, doi:http:// dx.doi.org/10.1016/0277-9536(93)90065-C. [104] K. Heikkinen, S. Salantera, M. Kettu, M. Taittonen, Prostatectomy patients’ postoperative pain assessment in the recovery room, J. Adv. Nurs. 52 (2005) 592–600 doi: JAN3631 [pii]\r10.1111/j.1365-2648.2005.03631.x [doi]. [105] I.J. Higginson, M. McCarthy, Validity of the support team assessment schedule: do staffs’ ratings reflect those made by patients or their families? Palliat Med. 7 (1993) 219–228, doi:http://dx.doi.org/10.1177/ 026921639300700309. [106] M. Hodgkins, D. Albert, L. Daltroy, Comparing patients’ and their physicians’ assessments of pain, Pain 23 (1985) 273–277, doi:http://dx.doi.org/10.1097/ 00132586-198606000-00040. [107] E. Idvall, K. Berg, M. Unosson, L. Brudin, Differences between nurse and patient assessments on postoperative pain management in two hospitals, J. Eval. Clin. Pract. 11 (2005) 444–451, doi:http://dx.doi.org/10.1111/j.13652753.2005.00555.x. [108] K. Oi-Ling, D.T.S.E. Man-Wah, D.N.G. Kam-Hung, Symptom distress as rated by advanced cancer patients, caregivers and physicians in the last week of life, Palliat. Med. 19 (2005) 228–233, doi:http://dx.doi.org/10.1191/ 0269216305pm1001oa. [109] A.J. Olden, E.T. Jordan, N.T. Sakima, J.A. Grass, Patients’ versus nurses’ assessments of pain and sedation after cesarean section, J. Obstet. Gynecol. Neonatal. Nurs. 24 (1995) 137–141. [110] J.A. Paice, S.M. Mahon, M. Faut-Callahan, Factors associated with adequate pain control in hospitalized postsurgical patients diagnosed with cancer, Cancer Nurs. 14 (1991) 298–305, doi:http://dx.doi.org/10.1097/00002820199112000-00003. [111] K. Perreault, C.E. Dionne, Patient–physiotherapist agreement in low back pain, J. Pain 6 (2005) 817–828, doi:http://dx.doi.org/10.1016/j. jpain.2005.07.008. [112] D.M. Powers, Ratings of pain from postoperative children and their nurses, Nurs. Pap. 19 (1987) 49–58. [113] W. Rhondali, D. Hui, S.H. Kim, K. Kilgore, J.H. Kang, L. Nguyen, et al., Association between patient-reported symptoms and nurses’ clinical impressions in cancer patients admitted to an acute palliative care unit, J. Palliat. Med. (2012) 120217115201002 doi: 10.1089/jpm.2011.0403. [114] I. Rundshagen, K. Schnabel, T. Standl, J. Schulte am Esch. Patients’ vs nurses’ assessments of postoperative pain and anxiety during patient- or nursecontrolled analgesia. vol. 82. 1999. [115] E.E.K. Sheiner, E.E.K. Sheiner, I. Shoham-Vardi, M. Mazor, M. Katz, Ethnic differences influence care giver’s estimates of pain during labour, Pain 81 (1999) 299–305, doi:http://dx.doi.org/10.1016/S0304-3959(99)00019-6. [116] A.J. Singer, P.B. Richman, A. Kowalska, J. Thode, Comparison of patient and practitioner assessments of pain from commonly performed emergency department procedures, Ann. Emerg. Med. 33 (1999) 652–658, doi:http://dx. doi.org/10.1016/S0196-0644(99)70194-0. [117] N.L. Stephenson, A comparison of nurse and patient: perceptions of postsurgical pain, J. Intraven. Nurs. 17 (1994) 235–239. [118] M.E. Suarez-Almazor, B. Conner-Spady, C.J. Kendall, A.S. Russell, K. Skeith, Lack of congruence in the ratings of patients’ health status by patients and their physicians, Med. Decis. Mak. 21 (2001) 113–121, doi:http://dx.doi.org/ 10.1177/0272989X0102100204. [119] J.E. Sutherland, R.M. Wesley, P.M. Cole, L.J. Nesvacil, M.L. Daly, G.J. Gepner, Differences and similarities between patient and physician perceptions of patient pain, Fam. Med. 20 (1988) 343–346. [120] A.J.W. Van der Does, Patients’ and nurses’ ratings of pain and anxiety during burn wound care, Pain 39 (1989) 95–101, doi:http://dx.doi.org/10.1016/03043959(89)90179-6. [121] D. Weiner, B. Peterson, F. Keefe, Chronic pain-associated behaviors in the nursing home: resident versus caregiver perceptions, Pain 80 (1999) 577– 588, doi:http://dx.doi.org/10.1016/S0304-3959(98)00249-8. [122] M.L. Zalon, Nurses’ assessment of postoperative patients’ pain, Pain (1993) 329–334, doi:http://dx.doi.org/10.1016/0304-3959(93)90033-L.

Please cite this article in press as: M.A. Ruben, et al., Healthcare providers’ accuracy in assessing patients’ pain: A systematic review, Patient Educ Couns (2015), http://dx.doi.org/10.1016/j.pec.2015.07.009