Skip to main content
BMC Geriatrics logoLink to BMC Geriatrics
. 2015 Sep 9;15:109. doi: 10.1186/s12877-015-0085-1

Erratum to: Association between pain, neuropsychiatric symptoms, and physical function in dementia: a systematic review and meta-analysis

Annelore H van Dalen-Kok 1,, Marjoleine Pieper 1,2, Margot de Waal 1, Albert Lukas 3, Bettina S Husebo 4,5, Wilco P Achterberg 1
PMCID: PMC4564980  PMID: 26354753

The original version of this article unfortunately contained some mistakes. The presentation of Table 2, Table 5 and Table 6 was incorrect. The corrected tables are given below.

Table 2.

Measurements of pain, neuropsychiatric symptoms and physical function

Measurement of pain Measurement of neuropsychiatric symptoms Measurement of function
First author Rating scale Method of detection Rating scale Method of detection Rating scale Method of detection
Ahn 201336 MDS pain severity scale, combining pain frequency and pain intensity Self-report, if not possible staff report based on proxy reports MDS subscales; wandering-item, aggression behaviour scale (ABS), challenging behaviour profile (CBP) agitation subscale Patient self-report, proxy and professional MDS-ADL long form (7 items) Staff observation
Bartels 20038 No use of rating scale Data collection instrument (3-month period), raters unknown MDS for depression Medical records MDS (number of ADLs) Medical records
Black 200639 No use of rating scale Medical records, preceding 6 months, interview surrogate and physician No use of rating scales Medical records, preceding 6 months, interview proxy and staff No use of rating scale Medical records, preceding 6 months, interview proxy and staff
Brummel-Smith 200240 1 out of 3 scales: faces or line scale, or word-based pain intensity scale self-report, assessed by trained research assistants No use of rating scales Trained research assistants No use of rating scale Trained research assistants
Cipher 20044 GMPI pain and suffering subscale Part of neuropsychological evaluation by a licensed clinical geropsychologist -GDS-15 “-26 dysfunctional behaviours with scores “1-7” Part of neuropsychological evaluation by a licensed clinical geropsychologist PRADLI Part of neuropsychological evaluation by a licensed clinical geropsychologist
Cipher 200641 GMPI Part of neuropsychological evaluation by a licensed clinical geropsychologist and each instrument was administered after interviewing the resident, nursing staff and family members GLDS, 19 categories with scores 1-7 Part of neuropsychological evaluation by a licensed clinical geropsychologist and each instrument was administered after interviewing the resident, nursing staff and family members, Medical records, preceding 6 to max 26, Months GLDS Part of neuropsychological evaluation by a licensed clinical geropsychologist and each instrument was administered after interviewing the resident, nursing staff and family members
D’Astolfo 200644 No use of rating scale Medical records, preceding 6 to max 26 months No use of rating scales No use of rating scale Medical records Ambulatory status: independent, requires assistance, wheel chair (or bedridden n?=?1)
Gruber-Baldini 200545 PGC-PIS, score ≥ 2 Rating by supervisory staff member CSDD Rating by supervisory staff member MDS; activities of daily living scale, SMOI Rating/observation by supervisory staff member
CMAI
Kunik 200530 PGC-PIS, item on level of pain in previous week, scores 1-6 Interview with patient and proxy by trained interviewer/research assistant CMAI Interview with patient and proxy by trained interviewer/research assistant - -
HAM-D
NPI (subdomains delusion/hallucinations)
Leonard 200650 MDS pain burden using a 4-level composite score based on pain frequency and intensity - MDS (Physical aggression: MDS item 'others were hit, shoved, scratched, sexually abused'; Depression: MDS score ≥3 on sum of 9 items, e.g. 'being sad', 'making negative statements', 'persistent anger with self or others', 'pained facial expressions'. (At least once in week before) - - -
Leong 200735 PAINAD for non-communicative patients Interviews with patient and staff member by professionals for communicative patients Depression with GDS-15 or STAI Self-report or staff report AAS Not reported
Anxiety with Cornell
Lin 201146 PAINAD-Chinese version Observation immediately following instances of routine care by principal investigator and research assistant No use of rating scales Medical records and observations by professional No use of rating scale Medical records and observation by professional
Morgan 201247 PGC-PIS worst pain item Not reported CMAI aggression subscale Not reported - -
CMAI non-aggressive physical agitation subscale
HAM-D depression
Norton 201042 PPQ, intensity item, 10–14 day baseline Primary CNA and data used from medical records RMBPC-NH, selection of 3 need driven behaviours, BEHAVE-AD Primary CNA and unit staff PSMS Nurses and trained research assistants
Shega 200548 VDS, 1 item on presence and severity of pain ‘right now’ Interviews with patients and caregivers by trained research assistant GDS-15 Interview patient and proxy KATZ Interview patient and proxy
CMAI IADL
Shega 201049 VDS, 5 point, ‘pain past 4 weeks’ Interviews with patient by trained research assistant Mental Health screening questionnaire; 5-item and 6 point scale Interview with patient by trained research assistant OARS/IADL; 3 point scale Interview patient by trained research assistant
Torvik 201048 VRS, 4 point, ‘pain right now’ Patient self-report DQoL, 29-items on 5 domains: self-esteem, aesthetics, positive affect, negative affect, belonging Not reported Barthel Self-report and medical records
Tosato 20123 InterRAI LTCF InterRAI LTCF questions and observation of behaviour, any type of pain or discomfort of the body in previous 3 days by trained (research) staff InterRAI LTCF 5 behavioural symptoms, previous 3 days Not reported MDS ADL Hierarchy Scale Data recorded by study physicians
Volicer 200937 MDS-RAI pain frequency (item J2a) Combination of physical examination, patient history, observation, consultation caregiver and medical records by staff MDS Depression Rating Scale Combination of physical examination, patient history, observation, consultation caregiver and medical records by staff - -
MDS item J1e for delusions MDS item J1i for hallucinations
Volicer 201151 MDS Combination of physical examination, patient history, observation, consultation caregiver and medical records by staff MDS items I1ee, E1a, E1d, E1f, E1b, E1i, E1l, E1m for depression Combination of physical examination, patient history, observation, consultation caregiver and medical records by staff - -
MDS for delusions and hallucinations
MDS items B5b, E1b, E4aa, E4da for agitation
Williams 200543 PGC-PIS, score =2, and 0–10 pain numeric rating scale Registered nurses or licensed practical nurses and interview with overseeing supervisor CSDD, score =7 Rating by care supervisors, registered nurses and licensed practical nurses MDS-ADL Rating by care supervisors, registered nurses and licensed practical nurses
CMAI, any behaviour at least weekly APAS
SMOI
Zieber 200538 DS-DAT, and a 7-point pain rating scale Trained facility nurses, palliative care nurse consultants PAS Trained facility nurses - -

Abbreviations: MDS Minimum Dataset, ADL Activities of Daily Living, GMPI Geriatric Multidimensional Pain and Illness Inventory, GDS-15 Geriatric Depression Scale-15 short version, PRADLI Psychosocial Resistance to Activities of Daily Living Index, GLDS Geriatric Level of Dysfunction Scale, PGC-PIS Philadelphia Geriatric Centre Pain Intensity Scale, CSDD Cornell Scale for Depression in Dementia, CMAI Cohen-Mansfield Agitation Inventory, SMOI Structured Meal Observational Instrument, HAM-D Hamilton Rating Scale for Depression, NPI Neuropsychiatric Inventory, PAINAD Pain Assessment in Advanced Dementia, STAI State-Trait Anxiety Inventory, AAS Adjusted Activity Scale, PPQ Proxy Pain Questionnaire, CNA Certified Nursing Assistant, RMBPC-NH Revised Memory and Behaviour Problems Checklist-Nursing Home, BEHAVE-AD Behavioural Pathology in Alzheimer’s disease, PSMS Physical Self Maintenance Scale, VDS Verbal Descriptor Scale, KATZ Index of Independence in Activities of Daily Living, IADL Instrumental Activities of Daily Living, OARS/IADL Older Americans Recourses and Services/Instrumental Activities of Daily Living, VRS Verbal Rating Scale, DQol Dementia Quality of life, APAS Albert Patient activity Scale, DS-DAT Discomfort Scale - Dementia of Alzheimer Type, PAS Pittsburgh Agitation Scale

Table 5.

Correlates of pain and neuropsychiatric symptoms

Correlates of pain and specified NPS
First author N Pain: prevalence Neuropsychiatric symptoms: prevalence Correlates of pain with NPS Quality of study
Ahn 201336 56577 Not reported Wandering 9 % AOR 0.77 (95 % CI: 0.73-0.81) with wandering 10
Subsample without psychotropic medication
AOR 0.72 (95 % CI: 0.63-0.83) with wandering
(Adjusted for cognition, ADL, sociodemographics)
Kunik 200534 99 Pain mean 2.4 (SD 1.2) Delusions/hallucinations mean 0.35 (SD 0.48) r = 0.15 (p > 0.05) with psychosis 8.5
Leong 200735 225 Pain 44 %, chronic pain 34 % Anxiety 48 % SOR 1.8 (95 % CI: 1.0-3.0) with anxiety 8.5
Norton 201042 161 Not reported BEHAVE-AD mean 6..4 (SD 29.2) r = 0.15 (p = 0.08) for pain intensity and emotional behaviour problems 9
RMBPC-NH mean 1.45 (SD 0.64) r = 0.05 (p = 0.58) for pain intensity and resistiveness to care
Torvik 201052 106 Current pain in total group 55 %, in cognitive impaired group 52 % Negative affect index (DQoL) mean 2.0 (SD 0.75), positive affect/humour index (DQoL) mean 3.4 (SD 0.9) p < 0.01 for current pain and negative affect 6.5
p = 0.11 for current pain and with positive affect/humour
Tosato 20123 2822 Any pain 19 % (moderate/severe/excruciating pain 13 %) Behavioural symptoms 37 % Psychiatric symptoms 21 % AOR = 0.74 (95 % CI: 0.55-1.0) with wandering 11.5
AOR = 1.4 (95 % CI: 1.08-1.8) with resistance to care
AOR 1.5 (95 % CI: 1.07-2.03) with delusions
AOR 1.06 (95 % CI: 0.80-1.41) with verbal abuse
AOR 1.08 (95 % CI: 0.75-1.55) with physical abuse
(Adjusted for age, gender, country, cognitive impairment, number of diseases, ischemic heart disease, stroke, falls, communication problems, and a flare-up of a chronic or recurrent condition)
Volicer 200937 929 Daily pain 29 %, less than daily pain 19 % Verbally abusive not easily altered 2 %, physically abusive not easily altered 12 % r = 0.07 (p = 0.03) for pain frequency and verbal abuse 11
AOR = 0.9 (p = 0.53) with resisting care
AOR = 0.7 (p = 1.2) with verbal abuse
AOR = 0.7 (p = 0.16) with physical abuse
Delusions 8 % (Both multivariate models among others controlled for resisting care)
Hallucinations 9 %
Zieber 200538 58 Not reported Not reported r = 0.46 (p < 0.01) for DS-DAT scores and resisting care 8
r = 0.42 (p < 0.01) for DS-DAT scores and aberrant vocalization
Pain rating by palliative care nurse consultants:
r = 0.51 (p < 0.01) with resisting care
r = 0.40 (p < 0.01) with aberrant vocalizations
Pain rating by facility nurse:
r = 0.48 (p < 0.01) with resisting care
r = 0.065 (p < 0.63) with aberrant vocalizations
Correlates of pain and unspecified NPS
First author N Pain: prevalence Neuropsychiatric symptoms: prevalence Correlates of pain with unspecified NPS Quality of study
Black 200639 123 Pain 63 % Psychiatric disorders or behaviour problems 85 %, behaviour problems 67 % SOR 1.9 (95 % CI: 0.7-5.3) with psychiatric/behaviour problems 6.5
SOR 1.2 (95 % CI: 0.5-2.5) with behaviour problems
Brummel-Smith 200240 104 (excluding those unable to self-report pain) Moderate-severe pain 60 % ≥1 disruptive behaviours (wandering, verbal disruption, physical aggression, regressive behaviour, hallucinations) SOR 1.8 (95 % CI: 0.8-4.0) with ≥ 1 disruptive behaviour 7
No-mild pain 40 %
50 subject unable to answer
70 % in dementia sample n = 154
Cipher 20044 234 Persistent pain 72 % Dysfunctional behaviours mean 4.4 (SD 0.76) r = 0.22 (p < 0.05) with dysfunctional behaviours 7.5
Cipher 200641 277 Acute pain 29 % - r = 0.18 (p < 0.05) with GLDS mean behavioural intensity 7.5
Chronic pain 59 %
Norton 201042 161 Not reported BEHAVE-AD mean 61.4 (SD 29.2) r = 0.18 (p = 0.03) for pain intensity and disruptive behaviour problems 9
RMBPC-NH mean 1.45 (SD 0.64) r = 0.05 (p = 0.53) for pain intensity and global need driven behaviours
Tosato 20123 2822 Any pain 19 % (moderate/severe/excruciating pain 13 %) Behavioural symptoms 37 % AOR = 1.4 (95 % CI: 1.04-1.8) with socially inappropriate behaviour 11.5
Psychiatric symptoms 21 % (Adjusted for age, gender, country, cognitive impairment, number of diseases, ischemic heart disease, stroke, falls, communication problems, and a flare-up of a chronic or recurrent condition)
Williams 200539 331 Pain 21 %, in nh 23 %, in rc/al 20 % (self-report for subgroup mmse > 10 was higher: 39 % and 25 %) Behavioural symptoms 58 % OR = 1.1 (95 % CI: 0.49-2.29) and AOR = 1.2 (95 % CI: 0.57-2.36) with behavioural symptoms 10
(Adjusted for: sex, race, age, cognitive status, number of 10 comorbidities, impairments of 7 activities of daily living)

Abbreviations: AOR Adjusted Odds Ratio, ADL Activities of Daily Living, SD Standard Deviation, r correlation coefficient, SOR Self-Calculated Odds Ratio, BEHAVE-AD Behavioural Pathology in Alzheimer’s disease, RMBPC-NH Revised Memory and Behaviour Problems Checklist-Nursing Home, DQoL Dementia Quality of life, DS-DAT Discomfort Scale - Dementia of Alzheimer Type, GLDS Geriatric Level of Dysfunction Scale, rc/al residential care/assisted living, MMSE Mini Mental State Examination, OR Odds Ratio

Table 6.

Correlates of pain with physical function

Correlates of pain and ADL or IADL
First author N Pain: prevalence Physical function: prevalence Correlates of pain with ADL or IADL Quality of study
Brummel-Smith 200236 104 (excluding those unable to self-report pain) Moderate-severe pain 60 %, no-mild pain 40 % (50 subject unable to answer) ≥1 ADL limitations SOR 1.9 (95 % CI: 0.6-6.0) with ≥ 1 ADL limitation 7
92 % in dementia sample (n = 154)
Cipher 20044 234 Persistent pain 72 % ADL independency mean 0.09 (SD 0.99) Correlations with GMPI ’pain and suffering’ 7.5
r = −0.04 (α > 0.05) with ADL independency
Shega 200544 115 Any current pain self-report 32 %, caregiver report 53 % KATZ mean 8.5 (SD 2.7), IADL mean 15.3 (SD 3.9) For self-report pain 9.5
No association ADL and IADL (p > 0.05)
For caregiver pain report
No association with ADL or IADL (p > 0.05)
Shega 201045 5549 Moderate or greater pain: 35.8 % Any IADL impairment: 66.5 % OR = 1.74 (95 % CI: 1.15-2.62) with any iADL impairment 9
(Adjusted for demographics)
Torvik 201048 106 Current pain in total group 55 %, in cognitive impaired group 52 % Highly or moderate ADL dependent 36 % p = 0.20 for current pain and ADL 6.5
SOR = 0.5 (95 % CI: 0.2-1.2) for current pain and ADL high/medium v.s. low
Tosato 20123 2822 Any pain 19 % (moderate/severe/excruciating pain 13 %) No disability 8 %, assistance required 43 %, dependent 49 % SOR 1.0 (95 % CI: 0.9-1.2) with ADL-dependent 11.5
SOR 0.9 (95 % CI: 0.75-1.09) with ADL assistance required
(Adjusted for age, gender, country, cognitive impairment, number of diseases, ischemic heart disease, stroke, falls, communication problems, and a flare-up of a chronic or recurrent condition)
Correlates of pain and other functional impairments
First author N Pain: prevalence Physical function: prevalence Correlates of pain with ADL or IADL Quality of study
Black 200639 123 Pain 63 % Nutrition/hydration problems total sample 85 % SOR 1.9 (95 % CI: 0.7-5.3) with nutrition/hydration problems 6.5
Brummel-Smith 200240 104 (excluding those unable to self-report pain) Moderate-severe pain 60 %, no-mild pain 40 % (50 subject unable to answer) ≥1 ADL limitations SOR 1.6 (95 % CI: 0.6-4.2) with bladder incontinence 7
92 % in dementia sample (n = 154)
D’Astolfo 200644 140 Pain 64 % (musculoskeletal pain 40 %) Use of wheel chair 60 % SOR 1.5 (95 % CI: 0.7-3.0) with use of wheel chair or bedridden 7
Requires assistance 34 % SOR 1.0 (95 % CI: 0.5-2.0) with requires assistance
(Analyses in sample of no dementia-severe dementia)
Lin 201146 112 Observed pain 37 % (PAINAD > =2) Being restrained 46 %; observed care activities: bathing 43 %, assisted transfer 31 %, self-transfer 26 % OR = 5.4 (95 % CI: 2.3-12.5) and AOR = 3.0 (95 % CI: 1.0-8.7) with being restrained 12
OR = 23.4 (95 % CI: 3.0-188) and AOR = 19.2 (95 % CI: 2.3-162) with bathing
OR = 29.7 (95 % CI: 3.6-242) and AOR = 11.3 (95 % CI: 1.2-102) with assisted transfer, both compared to self-transfer
(Adjusted for gender, age, wound, restraint, tube present in body, recent fall, severity of dementia and type of activity)
Williams 200543 331 Pain 21 %, in nh 23 %, in rc/al 20 % (self-report for subgroup MMSE > 10 was higher: 39 % and 25 %) Low activity 47 %, immobile 12 % OR = 0.65 (95 % CI: 0.38-1.11) and AOR = 0.64 (95 % CI: 0.37-1.10) with low activity 10
Low food intake 53 % OR = 1.1 (95 % CI: 0.49-2.29) and AOR = 0.8 (95 % CI: 0.37-1.69) with immobility
Low fluid intake 51 % OR = 1.18 (95 % CI: 0.64-2.17) and AOR = 1.03 (95 % CI: 0.56-1.87) with low food intake
OR = 1.20 (95 % CI: 0.67-2.15) and AOR 1.14 (95 % CI: 0.66-1.99) with low fluid intake
(Adjusted for: sex, race, age, cognitive status, number of 10 comorbidities, impairments of 7 activities of daily living)

Abbreviations: SOR Self-Calculated Odds Ratio, ADL Activities of Daily Living, SD Standard Deviation, r correlation coefficient, GMPI Geriatric Multidimensional Pain and Illness Inventory, PAINAD Pain Assessment in Advanced Dementia, OR Odds Ratio, AOR Adjusted Odds Ratio, KATZ Index of Independence in Activities of Daily Living, IADL Instrumental Activities of Daily Living, nh nursing home, rc/al residential care/assisted living, MMSE Mini Mental State Examination

Footnotes

The online version of the original article can be found under doi:10.1186/s12877-015-0048-6.

Contributor Information

Annelore H. van Dalen-Kok, Email: a.h.van_dalen@lumc.nl

Marjoleine Pieper, Email: m.j.c.pieper@lumc.nl.

Margot de Waal, Email: m.w.m.de_waal@lumc.nl.

Albert Lukas, Email: albert.lukas@malteser.org.

Bettina S. Husebo, Email: bettina.husebo@igs.uib.no

Wilco P. Achterberg, Email: w.p.achterberg@lumc.nl


Articles from BMC Geriatrics are provided here courtesy of BMC

RESOURCES