Table 3 Measurement properties of the Doloplus-2
First author, year, country [reference] | Reliability | Validity | Responsiveness | Interpretability |
|---|---|---|---|---|
Akbarzadeh, 2007, Sweden [33] | Internal consistency Cronbach’s alpha for the total scale 0.84 for rater 1 and 0.82 for rater 2 Reliability (Inter-rater) Agreement between rater 1 and rater 2 for single items (Cohen’s Kappa coefficient) 0.31–0.69 No statistically significant difference between rater 1 and rater 2 for total score (Wilcoxon signed-rank test) p = 0.106 Spearman correlation 0.90 between rater 1 and rater 2 for total score | Criterion (Concurrent) Spearman correlation with the UAB as the ‘gold standard’ 0.70 for rater 1 and 0.72 for rater 2 Construct (Structural) EFA with the result of items loading on one factor | NR | NR |
Ando, 2010, Japan [23] | Reliability (Inter-rater) Matching scores by RN and researcher 77.5%, p = <0.01 The ICC for the agreement between RN and researcher was 0.90 (p = 0.001). Agreement by items 0.67–0.96 | Construct (cross-cultural) Semi-structured interviews with 14 nurses. Two items, ‘Protective body postures adopted at rest’ and ‘Sleep pattern’, were changed to more appropriate Japanese explanations | NR | NR |
Ando, 2016, Japan [54] | NR | NR | Before treatment, the mean total score was 9.8 (SD 4.2) for n = 10 patients, whereas their post-treatment score significantly decreased to 2.7 (SD 1.6); net change 7.1, 95% CI: 4.4–9.7 | NR |
Bauer, 2007, France [44] | NR | NR | NR | NR |
Chen, 2010a, Taiwan [22] | Internal consistency Cronbach’s alpha for the total scale 0.74. Subscales Somatic 0.79; Psychomotor 0.87; Psychosocial 0.74. The alpha coefficients did not increase when any of the items were deleted Reliability (Inter-rater) ICC for the agreement between RNs and RAs on the total scale 0.81. For the subscales; Somatic 0.79, Psychomotor 0.84 and Psychosocial 0.60 | Construct (Hypotheses testing) Pearson correlations with known correlates of pain. In moderate dementia, significant correlation with functional ability −0.38 (p < 0.01). In severe dementia, significant correlation with functional ability −0.22 (p < 0.01) and depression 0.12 (p < 0.05) Construct (Structural) A PCA showed three factors, accounting for 65% of the total variance. Factor 1: all five items of the Somatic subscale explained 27.43% of the variance. Factor 2: all three items of the Psychosocial subscale explained 19.86% of the variance. Factor 3: both items of the Psychomotor subscale, accounting for 19.99% of the variance Item-total and item-subtotal correlations: Each item was correlated with the originally belonged subscale, ranged from 0.6 to 0.94. Each item correlation with overall scale ranged from 0.42 to 0.65 Construct (Cross-cultural) Five experts examined the content of C-Doloplus-2 and rated each item on a 4-point Likert scale from relevant (4) to irrelevant (1). Only the option ‘Insomnia, affecting morning waking time’ of item 5 ‘Sleep pattern’ was recommended to be rephrased | NR | NR |
Chen, 2010b, Taiwan [38] | Reliability (inter-rater) Paired t-test for agreement of different pairs in assessing pain. No difference between mean total scores for RA-RN pairs (t = 0.28, p > 0.05), but a statistically significant differences between the mean total scores for RA-NA pairs (t = 6.70, p < 0.01). NAs tended to report more pain cues than RAs Logistic regression to examine factors influencing the extent of agreement for the different pair. For RA-RN pairs, OR increased when residents had stayed in the institution longer (OR 1.01, p = 0.01), had less physical dependency (OR 1.02, p = 0.00), and when RNs had received pain related training (OR 2.86, p = 0.04). For RA-NA pairs, OR increased when the patients had fewer medical diagnosis (OR 0.78, p = 0.01) and less physical dependency (OR 1.01, p = 0.04) | NR | NR | NR |
Chen, 2014, Taiwan [45] | Internal consistency Cronbach’s alpha for total scale 0.73 | NR | NR | NR |
Couilliot, 2013, France [55] | NR | NR | Statistically significant reduction on total and subscales scores after five acupuncture sessions: Total score mean variation −3.27, p < 0.01, effect size 0.77. Somatic score mean variation −2.08, p < 0.01, effect size 0.89. Psychomotor score mean variation −0.61, p < 0.05, effect size 0.33 Psychosocial score mean variation −0.59 points, p < 0.05, effect size 0.30 | NR |
Hadjistavropolous, 2008, Canada [46] | NR | Construct (Hypotheses testing) Item 10 (‘Problems of Behavior’) was related to dementia severity (β = − .25, p < .003), depression (β = .31, p < .001) and presence of delirium (β = .25, p < .003) Item 6 (‘Washing and dressing’) was related to delirium severity (β = .42, p < .004) and dementia severity (β = − .39, p < .005) Other items related to depression were item 5 (‘Sleep pattern’) (β = .22, p < .003), item 9 ‘Social life’ (β = .25, p < .001) | NR | NR |
Hølen, 2005, Norway [21] | NR | Content (Face) Results from a questionnaire, completed by the 11 administrators of the Doloplus-2, was the Doloplus-2 was instructive regarding observation indicating pain, and includes important pain clues Construct (cross-cultural) The translation was approved by all administrators. No item was pointed out as confusing, difficult to understand or elsewhere problematic Criterion (concurrent) Experts’ pain rating with NRS-11 was used as a pain criterion. The experts rated 25 patients as pain free where the Doloplus-2 made five false positive with scores of 5 and 6. Of the 59 cases, the Doloplus-2 made false negatives on 10 occasions: a Doloplus-2 ≥ 5 at the same time as the expert rated above 0 on the NRS-11. In five of these cases, the expert’s score was one half (usually 0 at rest and 1 in movement), three had a score of 1 and the remaining two were rated with 2 and 3 on the NRS-11 The Doloplus-2 explained 62% (R2) of the pain distribution. For 85% of the assessments, the Doloplus-2 score (0–30) multiplied by 0.25 (beta) corresponded to the expert score ± 1 unit on the 0–10 NRS scale Facial expression explained 48% (R2 = 0.48) of the experts scores alone. When including items Protective body postures at rest, Communication and Somatic complaints, these four items explained 68% of the total variability in the experts’ scores | NR | NR |
Hølen, 2007, Norway [47] | Reliability (inter-rater) Agreement between a geriatric specialist nurse and an enrolled nurse on the total score was 0.77 (ICC), with a 95% CI of 0.47–0.92. Assessed in the 16 patients included at the geriatric hospital unit | Criterion (concurrent) The pain criterion was the specialist nurse (pain expert) who made a single evaluation of each patient’s pain level on NRS-11. Doloplus-2 scores against the expert scores produced an R2 = 0.023, implying poor criterion validity of the Doloplus-2 when compared to pain experts evaluation. Association was found between the pain expert and the geriatric expert nurse who administered the Doloplus-2 in 16 patients in the Hospital, R2 = 0.54 | NR | NR |
Monacelli, 2013, Italy [53] | NR | NR | Reduction of total mean score between the first assessment and after 1 year of follow up (Wilcoxon rank test) R2 = 0.216, p < 0.001 | NR |
Neville, 2014, Australia [48] | Internal consistency Cronbach’s alpha for the two rater groups on the two assessment occasion was 0.86 and 0.87 Reliability (test–retest) Agreement for the two testing occasions occurring two weeks apart. Pearson correlation 0.71 for both rater groups Reliability (inter-rater) ICC for the agreement between nurse raters for the total score at first 0.73 and second testing occasion 0.81. Weighted Kappa to compare pain level categorizations (no pain, mild, moderate, severe pain) across raters at first 0.42 and second testing occasion 0.50 | Criterion (concurrent) Pain criterion was RNs initial yes/no rating of the residents’ pain. Pearson correlation for each rater group at the first testing occasion showed moderate correlations at 0.43 (rater group 1) and 0.45 (rater group 2) Construct (Structural) EFA showed a 1-factor solution was the best description of the factor structure of the Doloplus-2 EFA showed a single factor model best described the correlation among all the total scale scores for the Doloplus-2, CNPI and APS, each score loading highly (>0.60) on that single factor, indicating that all of the scales measures essentially the same single construct |  |  |
Pautex, 2007, Switzerland [49] | Internal consistency Cronbach’s alpha was adequate for all items, lower in patients with dementia (0.67) compared to cognitively intact patients (0.84). The lowest internal consistency scores were found for the items ‘Expression’ (0.82) and ‘Mobility’ (0.82) Reliability (test-retest) Performed in a subsample of 20 patients hospitalized in the same units with the same characteristics and stable chronic pain. The second testing occasion happened the day after the first one. ICC indicated excellent agreement at 0.96 | Criterion (concurrent) Spearman 0.46 indicated a moderate correlation with the pain criterion; patients’ self-assessment (VAS). The correlation was better in patients without dementia compared to patients with dementia (0.68 vs. 0.38) Doloplus-2 predicted 41% of the variability of pain intensity measured by VAS. The somatic dimension explained 36% of the variability, the psychomotor and psychosocial dimension 5% each. The intensity of pain (VAS) was mainly associated with the somatic dimension of Doloplus-2. Two items of the psychosocial reaction were also statistically significant (p < 0.05) | NR | NR |
Pickering, 2010, multinational [50] | Reliability (test-retest) Patients was assess at initial contact and again 4 h later. When evaluated with ICC, agreement ranged from 0.62 for the Dutch version to 0.98 for the Italian version (0.98). Evaluated with Pearson correlation, the results ranged from 0.57 for the Dutch version to 0.99 for the Portuguese version Reliability (inter-rater) ICC for the agreement between physicians for the total score ranged from 0.75 (Dutch version) to 0.97 (Italian version) Pearson correlation indicated excellent agreement ranging from 0.75 (Dutch version) to 0.97 (Italian and Portuguese version) Kappa was used to compare agreement for each of the 10 items across language version. The agreement ranged from fair to excellent (0.51–0.84) for the English version, excellent (0.79–0.96) for the Italian version, good to excellent (0.65–0.82) for the Portuguese version, fair to excellent (0.47–0.87) for the Spanish version and poor to excellent (0.19–1) for the Dutch version | NR | NR | NR |
RodrÃguez-Mansilla, 2015, Spain [56] | NR | NR | The best improvement in the mean total score was reached in the last (third) month of ear acupressure. The average improvement was 8.55 points (SD 4.39), 95% CI: 7.14–9.95 | NR |
Sheu, 2011, Canada [51] | Reliability (inter-rater) Three clips indicative of mild, moderate and severe pain intensities were selected for study for each participant. The mean of criterionvalues for each intensity level was 0.04 (−0.20–0.38) for mild pain, 0.20 (−0.07–0.46), for moderate pain, and 0.38 (0.11–0 .68) for severe pain | Criterion (concurrent) Pain criterion used was FACS-scores. No significant correlations were observed with the FACS at any of the pain intensities. Pearson correlation for mild pain was −0.13 (an inverse relationship between scores), 0.16 for moderate pain, and 0.10 for severe pain Construct (Hypotheses testing) Examined whether the scale differentiated the 3 levels of facial expression of pain by a pairwise comparison of the mean between each intensity level of the scale. The Doloplus-2 did not distinguish the 3 levels of pain: Mild-moderate: 0.37, p = 0.488 Mild-severe: 0.03, p = 0.955 Moderate-severe: 0.40, p = 0.481 | NR | NR |
Stacpoole, 2014, UK [52] | NR | NR | NR | NR |
Torvik, 2009, Norway [43] | NR | NR | NR | NR |
Torvik, 2010, Norway [42] | Internal consistency Cronbach’s alpha for the total scale was 0.71, and 0.60 (Somatic) 0.80 (Psychomotor) and 0.78 (Psychosocial) for the subscales. After excluding individual items, the alpha values for the subscales were comparable to alpha for the overall scale, except for the Somatic subscale where the alpha score decreased from 0.60 to 0.47 when item ‘Somatic complaint’ deleted | Criterion (concurrent) Pain criterion used was RNs proxy assessment answering the question ‘Do you believe that this patient is experiencing pain?’ Response options were ‘no’, ‘yes’ or ‘don’t know’. Nursing staff evaluated significantly more patients as experiencing pain when using Doloplus-2 compared with proxy-rated pain (p = 0.01) When pain was proxy rated, 36 of 40 (90%) cases where the RNs assessed ‘yes, pain’, scored ≥5 on Doloplus-2. 11 of 15 (73.3%) assessed as ‘no pain’ by RNs scored <5 on Doloplus-2 | NR | NR |
Voyer, 2008, Canada [41] | NR | NR | NR | NR |
Voyer, 2009, Canada [39] | NR | NR | NR | NR |
Voyer, 2011, Canada [40] | NR | NR | NR | NR |
Zwakhalen, 2006, the Netherlands [24] | Internal consistency Internal consistency for the total and subscales at different assessment points (T1 and T3). Cronbach’s alpha was 0.75 for the total scale, 0.70 for Somatic reactions, 0.80 for Psychomotor reactions, and 0.63 for Psychosocial reactions at T1 At T3, Cronbach’s alpha was 0.74 for the total scale, 0.63 for Somatic reactions, 0.77 for Psychomotor reactions, and 0.58 for Psychosocial reactions | Construct (Hypotheses testing) Used the known-groups technique by comparing Doloplus-2 scores between a ‘non-pain’ group’ and a ‘daily pain’ group. The mean score in the ‘Daily pain’ (mean 9.8; SD 6.0; range 2–23) was obviously higher compared to mean score in ‘no pain group’ (mean 5.1; SD 3.9; range 0–16) Pearson correlation was 0.29 for VAS by rater 1, 0.33 for VAS by nurse, 0.36 for the VRS, 0.29 for the PACSLAC and 0.34 for the PAINAD |  | NR |