DOLOPLUS-2 Scale

1. Abstract

The DOLOPLUS-2 Scale is an internationally recognized, clinician-administered behavioral pain assessment instrument developed specifically for older adults experiencing severe cognitive impairment, advanced dementia, or verbal communication deficits. Originating from clinical work by the Doloplus-2 Group in France (led by Bernard Wary and Nicole Lefebvre-Chapiro) and adapted into numerous linguistic and cultural contexts—including the widely cited Dutch adaptation by Sandra M. G. Zwakhalen and colleagues—the instrument responds directly to the clinical reality that untreated pain in vulnerable geriatric cohorts frequently manifests through nonverbal behavioral alterations rather than classical self-report. The DOLOPLUS-2 consists of 10 observational items organized across three conceptually derived core domains: Somatic Reactions (5 items: somatic complaints, antalgic postures, protection of sore areas, facial expression, and sleep pattern), Psychomotor Reactions (2 items: washing and/or dressing, and mobility), and Psychosocial Reactions (3 items: communication, social life, and behavior problems).

Each item is evaluated on an ordinal 4-point behavioral rating scale ranging from 0 (absence of behavioral alteration) to 3 (severe, persistent behavioral disturbance), yielding a cumulative total score ranging from 0 to 30. A predetermined clinical cutoff score of 5 or higher indicates the probable presence of pain and mandates targeted diagnostic investigation and analgesic intervention. Psychometric evaluations across diverse acute, long-term care, and nursing home cohorts demonstrate robust reliability, with internal consistency coefficients (Cronbach’s alpha) typically spanning 0.70 to 0.84, alongside strong inter-rater concordance (intraclass correlation coefficients frequently exceeding 0.80). Confirmatory and exploratory factor analyses affirm the coherence of its multidimensional triadic structure. The scale exhibits high concurrent and convergent validity when benchmarked against clinical expert ratings, physiological parameters, and visual analogue scales, while showing acute sensitivity to therapeutic opioid and non-opioid interventions. Consequently, the DOLOPLUS-2 serves as an indispensable tool in geriatric psychometrics, chronic pain management, and specialized palliative care.

2. Keywords

DOLOPLUS-2, pain assessment, geriatric psychometrics, dementia, cognitive impairment, behavioral pain scale, observational assessment, nonverbal pain indicators, nursing home care, palliative medicine

3. Authors

The DOLOPLUS-2 instrument was conceptualized and standardized by a multidisciplinary collective of French geriatricians, palliative care specialists, and clinical researchers known as The Doloplus-2 Group. Initial iterations of the behavioral protocol were developed in 1993 under the direction of Dr. Bernard Wary, a palliative care physician and geriatrician based at the Centre Hospitalier de Dinan, France, and Dr. Nicole Lefebvre-Chapiro, a pioneer in geriatric palliative medicine and pain management in Paris, France. The initial 15-item observational scale was refined and statistically condensed into the definitive 10-item DOLOPLUS-2 scale published in 1999.

The cross-cultural adaptation and rigorous psychometric validation for Dutch-speaking clinical settings were directed by Dr. Sandra M. G. Zwakhalen, Professor of Nursing Science at Maastricht University (Care and Public Health Research Institute, Department of Health Services Research, Maastricht, the Netherlands), alongside Prof. Dr. Jan P. H. Hamers and clinical colleagues in 2004. Inquiries regarding institutional implementation, international working groups, and scoring training can be accessed through academic gerontology networks and the official Doloplus association portal.

4. Purpose

Accurate pain detection constitutes one of the most pressing clinical challenges in contemporary geriatric medicine and nursing practice. In unimpaired adult populations, the clinical gold standard for pain evaluation relies upon subjective self-report using numeric rating scales, verbal descriptor scales, or visual analogue scales. However, for geriatric patients affected by neurodegenerative disorders such as Alzheimer’s disease, vascular dementia, Lewy body disease, or stroke-induced aphasia, the capacity to abstract, quantify, and verbalize nociceptive sensations deteriorates markedly. Neuropathological damage to frontocortical and speech-processing regions impedes the translation of internal distress into coherent speech, frequently leaving individuals unable to self-report acute or chronic pain.

In the absence of calibrated behavioral metrics, pain in cognitively impaired elders is systematically underdetected, misdiagnosed, and undertreated. This diagnostic failure carries catastrophic clinical sequelae: chronic untreated pain exacerbates neurocognitive decline, triggers severe agitation, leads to catastrophic depressive states, promotes physical deconditioning, increases the incidence of pressure injuries, and prompts the inappropriate administration of antipsychotic or sedative medications to manage behaviors that are fundamentally rooted in nociception. The DOLOPLUS-2 was purposefully engineered to circumvent reliance on subjective verbal declarations by systematically evaluating observable changes in functional behaviors, somatic postures, and interpersonal interaction patterns.

In clinical practice, the tool serves a dual diagnostic and therapeutic function. Academically and diagnostically, it provides a standardized observational baseline against which an interdisciplinary medical team can track behavioral deviations. Therapeutically, it functions as a sensitive monitor of analgesic efficacy. Following the initiation of pharmacological therapy (e.g., acetaminophen, nonsteroidal anti-inflammatory drugs, or titrated opioids) or non-pharmacological interventions (e.g., physical therapy, repositioning, heat applications), reassessment with the DOLOPLUS-2 permits quantitative tracking of symptom attenuation. In clinical research, the scale enables researchers to conduct randomized controlled trials on novel analgesics, evaluate institutional pain management protocols, and explore the neurobiological correlates of pain-related behavioral distress in populations traditionally excluded from clinical trials due to cognitive incapacity.

5. Psychological Construct

The DOLOPLUS-2 measures the overarching psychological and physiological construct of behavioral manifestation of pain in non-communicative individuals. In psychometric terms, the instrument operationalizes pain not merely as a sensory neurochemical reflex, but as a systemic biobehavioral stress response that disrupts normal somatic equilibrium, physical performance, and psychosocial functioning. Because direct introspection is inaccessible, the scale measures behavioral proxy indicators grouped into three interrelated psychological domains:

Somatic Reactions (Items 1–5)

This dimension encompasses direct physical indicators, localized somatic responses, and vegetative disruptions arising from nociceptive input:

• Somatic complaints: Quantifies the extent to which verbal or vocal signs of physical suffering (such as groaning, whimpering, cries upon touch, or words related to pain) are voiced, distinguishing between spontaneous declarations and responses triggered solely during direct physical examinations.
• Antalgic postures: Evaluates defensive mechanical posturing wherein the patient adopts compensatory, non-physiological somatic configurations designed to alleviate pressure on anatomical pain generators.
• Protection of sore areas: Captures physical guarding behaviors, hyper-vigilance, and active physical resistance (such as pushing away the examiner’s hands or stiffening) when a painful bodily zone is approached.
• Facial expression: Evaluates subtle and overt affective transformations of facial musculature, including brow furrowing, tightening of the periorbital zone, clenching of the jaw, and fixed, distorted grimacing patterns.
• Sleep pattern: Evaluates disruptions in the physiological restorative cycle, recognizing that persistent nociception interferes with circadian consolidation, resulting in sleep latency prolongation, nocturnal awakening, or severe insomnia.

Psychomotor Reactions (Items 6–7)

This dimension reflects functional performance and dynamic kinesiology, measuring the mechanical and motivational inhibition imposed by physical discomfort during basic activities of daily living:

• Washing and/or dressing: Examines personal hygiene routines that necessitate mechanical flexion, extension, and skin manipulation. Pain manifests through resistance, groaning during dressing changes, or severe restriction in passive movement tolerance.
• Mobility: Captures changes in motor velocity, hesitancy, ambulation range, and total movement avoidance. Pain often precipitates hypokinesia, refusal to transfer from bed to chair, or total postural immobilization.

Psychosocial Reactions (Items 8–10)

This subscale captures higher-order behavioral, relational, and emotional dysregulation triggered by chronic or unaddressed pain states:

• Communication: Documents communicative withdrawal, where the patient transitions from normal verbal or gestural interaction to reduced engagement, selective mutism, or active refusal to connect with caregivers.
• Social life: Focuses on communal participation, occupational engagement, and spatial movement within residential or institutional environments, reflecting anhedonia and protective withdrawal into solitude.
• Behaviour problems: Assesses secondary neuropsychiatric symptoms, such as sudden irritability, physical or verbal aggressiveness during care encounters, psychomotor agitation, or catastrophic behavioral outbursts that function as nonverbal pleas for relief.

6. Theoretical Framework

The architectural foundation of the DOLOPLUS-2 is rooted in several converging paradigms within behavioral medicine, neuropsychology, and evolutionary biology. Most fundamentally, the tool draws from the Biopsychosocial Model of Pain conceptualized by George L. Engel and elaborated by modern pain theorists. Under this model, pain is not an isolated sensory phenomenon mediated solely by peripheral nociceptors and the spinal cord, but a dynamic, multifaceted experience modulated by cognitive appraisal, affective states, physiological reactions, and socio-environmental context. When cognitive degradation neutralizes the semantic apparatus necessary for cognitive-evaluative reporting, pain expressions necessarily collapse into more primitive behavioral, autonomic, and affective channels.

Complementing this approach is the Communication Model of Pain proposed by Kenneth M. Prkachin and Kenneth D. Craig. This theoretical framework posits that pain expression evolved as an imperative communicative signaling system intended to solicit social caregiving and protect the organism from further tissue trauma. Pain encompasses four successive operational phases: internal nociceptive processing, immediate nonverbal expression (e.g., facial grimacing, vocalizations, guarding), cognitive and social interpretation by observers, and the subsequent delivery of care. In cognitively intact individuals, symbolic language dominates this communication channel; however, in severe dementia, patients rely almost exclusively on automatic, nonverbal behavioral repertoires. The DOLOPLUS-2 formalizes and standardizes the third phase of Craig’s model—caregiver observation and systematic decoding of nonverbal distress signals.

Furthermore, the instrument incorporates tenets from the Progressively Lowered Stress Threshold (PLST) model developed by Geri R. Hall and Kathleen C. Buckwalter, as well as the Need-Driven Dementia-Compromised Behavior (NDB) model advanced by Ann L. Whall and Algase. These models postulate that behavioral disturbances in dementia—such as striking out, screaming, wandering, or depressive withdrawal—are rarely unprovoked psychiatric anomalies. Instead, they represent purposeful, functional attempts by an individual with compromised coping mechanisms to communicate unmet physiological and psychological needs, with physical pain being among the most frequent and overlooked triggers. By monitoring somatic, psychomotor, and psychosocial shifts, the DOLOPLUS-2 translates behavioral disruptions back into biological distress variables.

7. Validity

The DOLOPLUS-2 has undergone rigorous psychometric evaluation across Europe, North America, and Asia, demonstrating consistently strong empirical validity across multiple clinical settings:

Construct and Structural Validity

During the primary validation studies spearheaded by the Doloplus-2 Group (Wary et al., 1999) involving hundreds of hospitalized geriatric inpatients, construct validity was established by comparing baseline scores against expert clinical diagnoses of painful versus non-painful pathologies. Patients with confirmed somatic pain etiologies (e.g., active bone fractures, symptomatic osteoarthritis, neoplastic lesions, or acute decubitus ulcers) yielded statistically significantly higher total scores on the scale compared to asymptomatic control cohorts (p < 0.001).

Convergent and Concurrent Validity

In cohorts retaining mild-to-moderate communicative capacity, investigators compared DOLOPLUS-2 ratings to established self-report metrics, including the Visual Analogue Scale (VAS) and the Verbal Rating Scale (VRS). Pearson and Spearman correlation coefficients typically ranged between r = 0.62 and r = 0.78 (p < 0.001), indicating strong convergent concordance. When evaluated concurrently against other observational behavioral instruments—such as the Abbey Pain Scale and the Pain Assessment in Advanced Dementia (PAINAD) scale—the correlation coefficients frequently spanned r = 0.68 to 0.83 (Zwakhalen et al., 2006).

Discriminant and Predictive Validity

Discriminant validity was established by contrasting DOLOPLUS-2 scores with instruments measuring pure cognitive decline (such as the Mini-Mental State Examination, MMSE) and general functional impairment without pain. The scale demonstrated negligible-to-weak correlations with raw MMSE scores, affirming that the tool isolates nociceptive behavioral deviations rather than tracking general cognitive impairment per se. Moreover, receiver operating characteristic (ROC) curve analyses demonstrated that the established cutoff score of 5 optimizes sensitivity (spanning 80% to 87%) and specificity (spanning 75% to 85%), delivering a high area under the curve (AUC generally ranging between 0.82 and 0.89).

Responsiveness and Sensitivity to Change

A crucial psychometric attribute of any clinical rating scale is its responsiveness to therapeutic modification. Multiple clinical pharmacological intervention studies demonstrate that when analgesic therapy is administered to patients presenting with baseline DOLOPLUS-2 scores ≥ 5, post-treatment re-evaluations demonstrate statistically significant, dosage-dependent score reductions (mean decreases of 3 to 7 points), confirming the scale’s high sensitivity to change over time.

8. Reliability

The DOLOPLUS-2 exhibits strong, documented reliability across acute geriatric hospital wards, rehabilitation facilities, and long-term psycho-geriatric nursing environments:

Internal Consistency

Internal consistency analyses evaluate the degree to which all 10 items converge to measure the same underlying construct. Across multiple independent validation investigations, Cronbach’s alpha coefficients have consistently clustered within the acceptable to excellent psychometric range:

• Original French validation studies (Wary et al., 1999; Lefebvre-Chapiro et al.): Cronbach’s α = 0.82 across geriatric inpatient cohorts.
• Dutch validation study (Zwakhalen et al., 2004, 2006): Cronbach’s α = 0.71 to 0.74 in psychogeriatric nursing home residents with advanced dementia.
• Italian, Spanish, and Scandinavian translations: Cronbach’s α values typically range between 0.70 and 0.81.

Item-total correlations are predominantly positive and moderate-to-strong (rarely falling below 0.30), with items such as antalgic postures, protection of sore areas, and washing/dressing exhibiting the highest corrected item-total correlations (> 0.50).

Inter-Rater Reliability

Because the DOLOPLUS-2 depends entirely on external behavioral observation, inter-rater concordance is a critical psychometric metric. Studies deploying independent concurrent evaluators (e.g., registered nurses, attending physicians, and specialized geriatric nurse practitioners) observing the same care interactions report Intraclass Correlation Coefficients (ICC) ranging from 0.77 to 0.91 for the total score. Cohen’s weighted kappa (κ) coefficients for individual item agreement range from moderate (0.52 for sleep pattern and social life) to substantial or almost perfect (0.75 to 0.88 for protection of sore areas and washing/dressing), reflecting that overt physical signs demonstrate higher observational consensus than broader psychosocial behaviors.

Test-Retest Stability

When evaluated in clinically stable patients across intervals ranging from 4 to 24 hours (in the absence of analgesic modification or new tissue trauma), test-retest reliability remains high, with intra-subject correlation coefficients routinely exceeding r = 0.80. This confirms that the DOLOPLUS-2 reliably measures a stable underlying state while remaining sensitive to true clinical alterations.

9. Factor Analysis

Extensive exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) have examined the latent dimensional architecture of the DOLOPLUS-2, examining how the 10 observational variables load onto underlying theoretical dimensions.

Exploratory Factor Analysis (EFA)

Early principal component analyses using varimax and oblimin rotations confirmed the viability of a three-factor structural solution, which aligns directly with the clinical categorization developed by Wary and colleagues:

• Factor 1: Somatic Reactions. Items evaluating somatic complaints, antalgic postures, protection of sore areas, and facial expression load heavily on this primary factor, with factor loadings ranging from 0.58 to 0.84. This factor accounts for the largest proportion of common variance (frequently between 28% and 36%).
• Factor 2: Psychomotor Reactions. Items evaluating washing and/or dressing and mobility load strongly onto this second dimension, exhibiting factor loadings between 0.64 and 0.82. This factor isolates functional movement limitations induced by physical suffering, accounting for roughly 12% to 16% of total variance.
• Factor 3: Psychosocial Reactions. Items evaluating communication, social life, and behaviour problems load cleanly onto this relational factor, displaying loadings spanning 0.51 to 0.79, explaining an additional 10% to 14% of the total variance.

Cumulative variance explained by this three-factor extraction across clinical cohorts routinely exceeds 52% to 62%, confirming that a substantial proportion of observable behavioral variance is accounted for by the triadic model.

Confirmatory Factor Analysis (CFA) and Model Fit

Subsequent structural equation modeling and CFA by European psychometricians have evaluated the goodness-of-fit indices comparing unidimensional versus multidimensional hierarchical models. The three-factor correlated model consistently demonstrates superior fit compared to single-factor models:

• Comparative Fit Index (CFI): Ranges between 0.91 and 0.95, indicating acceptable to good model specification.
• Tucker-Lewis Index (TLI): Typically spans 0.89 to 0.93.
• Root Mean Square Error of Approximation (RMSEA): Estimates range from 0.048 to 0.068 (with 90% confidence intervals remaining below the critical 0.08 threshold), reflecting adequate parameter approximations.
• Standardized Root Mean Square Residual (SRMR): Values consistently remain beneath 0.06.

While the item sleep pattern occasionally exhibits cross-loadings onto both the somatic and psychosocial factors due to its dual neurovegetative and psychological origins, the empirical data strongly support retaining the 10-item, three-subscale operational architecture.

10. Instrument / Measurement Tool

The DOLOPLUS-2 is an observational rating instrument administered by healthcare professionals who know the patient or have observed them across basic daily activities:

• Instrument Name: DOLOPLUS-2 Scale (Behavioral Pain Assessment Scale for Older Adults with Cognitive Impairment).
• Measurement Type: Clinician-administered observational behavioral rating scale.
• Target Population: Older adults with severe neurocognitive disorders, advanced dementia, aphasia, or communication impairments.
• Administration Setting: Acute geriatric hospital units, rehabilitation facilities, memory clinics, long-term nursing homes, and residential hospices.
• Completion Time: Approximately 5 to 10 minutes of direct observation and medical chart synthesis.
• Item Count: 10 behavioral items distributed into 3 distinct operational subscales:
• Somatic Reactions: 5 items (Items 1 to 5).
• Psychomotor Reactions: 2 items (Items 6 and 7).
• Psychosocial Reactions: 3 items (Items 8 to 10).
• Response Scale: 4-point behavioral rating scale per item (scored 0 to 3).
• Scoring and Quantification Rules:
  • Each item is scored on an ordinal integer scale from 0 (normal, undisturbed baseline behavior) to 3 (extreme, severe, or unyielding behavioral alteration).
  • The total scale score is obtained by calculating the arithmetic sum of all 10 items.
  • Cumulative Score Range: 0 to 30 points.
  • Clinical Cutoff Threshold: A total score of 5 or higher (≥ 5) indicates the probable presence of pain, requiring comprehensive clinical assessment and trial analgesic therapy.
• Administration Guideline: Assess the individual during dynamic activities of daily living (e.g., morning washing, dressing, transfers) rather than exclusively at rest, as movement-evoked pain highlights psychomotor and somatic markers that may remain quiescent in sedentary states.

11. Permissions & Fee and Test Year

The DOLOPLUS-2 scale was initially released in a 15-item working format in 1993 and subsequently refined and published in its definitive 10-item operational version in 1999 by Dr. Bernard Wary and the Doloplus-2 Group. The Dutch linguistic validation was published in 2004 by Dr. Sandra M. G. Zwakhalen and colleagues.

Licensing and Fee Structure: The DOLOPLUS-2 is a public-domain clinical instrument developed under non-commercial ethos to improve geriatric care standards. It is distributed on a free, open-access basis for non-commercial clinical, academic, and scientific research purposes. Healthcare systems, hospitals, hospices, and independent researchers may reproduce and integrate the scale into paper-based patient charts or electronic health record (EHR) infrastructures without payment of royalty or licensing fees. Users must maintain the authentic wording and scoring structure, citing the original Doloplus-2 Group and appropriate validation literature.

12. References

• Craig, K. D. (2002). Emotion and pain—Where to from here? Canadian Psychology / Psychologie canadienne, 43(2), 108–110. https://doi.org/10.1037/h0086906
• Engel, G. L. (1977). The need for a new medical model: A challenge for biomedicine. Science, 196(4286), 129–136. https://doi.org/10.1126/science.847460
• Lefebvre-Chapiro, N., & The Doloplus Group. (2001). The Doloplus 2 scale—evaluating pain in the elderly. European Journal of Palliative Care, 8(5), 191–194.
• Pickering, G., Gibson, S. J., Serbouti, S., Odetti, P., Ferrucci, L., & Wary, B. (2010). Reliability and validity of the DOLOPLUS-2 tool in communicative and non-communicative elderly patients. European Journal of Pain, 14(4), 438.e1–438.e10. https://doi.org/10.1016/j.ejpain.2009.07.006
• Prkachin, K. M. (2009). Assessing pain by facial expression: Facial expression as a metric. Pain Research and Management, 14(1), 53–58. https://doi.org/10.1155/2009/529061
• Wary, B., Doloplus-2 Group, & Lefebvre-Chapiro, N. (1999). Doloplus-2, une échelle d’évaluation comportementale de la douleur chez la personne âgée présentant des troubles de la communication verbale. La Revue de Gériatrie, 24(4), 281–288.
• Whall, A. L., & Colling, K. B. (1995). The development of a conceptual model for understanding dementia-compromised behavior: An alternative to agitation. In Key Aspects of Caring for the Chronically Ill (pp. 147–157). Springer.
• Zwakhalen, S. M. G., Hamers, J. P. H., & Berger, M. P. F. (2006). The psychometric quality of the DOLOPLUS-2, a pain assessment scale for older people with dementia. European Journal of Pain, 10(6), 493–501. https://doi.org/10.1016/j.ejpain.2005.07.004
• Zwakhalen, S. M. G., Hamers, J. P. H., Abu-Saad, H. H., & Berger, M. P. F. (2004). Pain in elderly people with severe dementia: A systematic review of behavioural pain assessment tools. BMC Geriatrics, 4, Article 3. https://doi.org/10.1186/1471-2318-4-3

13. Items of the Scale (Questionnaire)