Send Orders for Reprints to
[email protected] 194
Current Clinical Pharmacology, 2015, 10, 194-203
Prevalence of Analgesic Use and Pain in People with and without Dementia or Cognitive Impairment in Aged Care Facilities: A Systematic Review and Meta-Analysis Edwin C.K. Tan1,*, Natali Jokanovic1,2, Marjaana P.H. Koponen3, Dennis Thomas1, Sarah N. Hilmer4,5,6 and J. Simon Bell1,3,6,7 1
Centre for Medicine Use and Safety, Faculty of Pharmacy and Pharmaceutical Sciences, Monash University, Melbourne, Australia; 2Pharmacy Department, Alfred Hospital, Melbourne, Australia; 3 Kuopio Research Centre of Geriatric Care, School of Pharmacy, University of Eastern Finland, Kuopio, Finland; 4Department of Clinical Pharmacology, Royal North Shore Hospital, St Leonards, Sydney, Australia; 5Kolling Institute of Medical Research, Northern Clinical School, Sydney Medical School, The University of Sydney, St Leonards, Australia; 6NHMRC Cognitive Decline Partnership Centre, Hornsby Ku-ring-gai Hospital, Hornsby, NSW; 7Sansom Institute, School of Pharmacy and Medical Sciences, University of South Australia, Adelaide, Australia
E.C.K. Tan
Abstract: Pain is a frequent cause of discomfort and distress in residents in residential aged care facilities (RACFs). Despite the benefits of adequate pain management, there is inconsistency in the literature regarding analgesic use and pain in residents with dementia. The aim of this systematic review was to determine the prevalence of analgesic drug use among residents with and without dementia or cognitive impairment in RACFs. A systematic search of MEDLINE and EMBASE (inception to January 2014) was conducted using Medical Subject Headings and Emtree terms, respectively. Studies were included if they reported prevalence of analgesic use for residents both with and without dementia within the same study. Data extraction and quality assessment was performed independently by two investigators. Data on the prevalence of analgesic use, pain and painful conditions were extracted. Meta-analyses were performed using random effect models. The 7 included studies were of high quality (≥5 out of 7 on the adapted Newcastle-Ottawa Scale). Analgesic use in residents with and without dementia or cognitive impairment ranged from 20.2% to 61.2% and 38.8% to 79.6%, respectively. Paracetamol was the most prevalent analgesic in people with and without dementia. Residents with dementia or cognitive impairment had a significantly lower prevalence of analgesic use (odds ratio [OR] 0.576, 95% confidence interval [CI] = 0.406-0.816) and of self-reported and clinician-observed pain (OR 0.355, 95% CI = 0.278-0.454) than residents without cognitive impairment, despite a comparable prevalence of painful conditions. These findings may indicate under-reporting and under-detection of pain in persons with dementia, and subsequent suboptimal treatment.
Keywords: Analgesics, dementia, homes for aged, long-term care, nursing home, pain, systematic review. BACKGROUND
Pain is under-recognised and under-treated among older people [7]. This may be particularly true among people with
dementia and cognitive impairment [8]. People with dementia may have a low placebo response to analgesics and hence typical doses of analgesics may be less effective in this population that in those without dementia [9]. Although experimental research has suggested people with and without dementia experience a similar prevalence and intensity of pain [9, 10], people with dementia may have a lower overall prevalence of analgesic use. There are several possible reasons. Firstly, pain may be expressed via facial grimacing, body movements, and behavioural changes in individuals with dementia. Clinicians may not recognise these signs, and fail to prescribe appropriate analgesia [11, 12]. People with dementia or cognitive impairment may be particularly susceptible to analgesic-related ADEs, including daytime sedation, falls and fractures, delirium and respiratory depression [3].
*Address correspondence to this author at the Centre for Medicine Use and Safety, Faculty of Pharmacy and Pharmaceutical Sciences, Monash University (Parkville campus), 381 Royal Parade, Parkville, VIC 3052, Australia; Tel: +61-(0)3-9903 9244; Fax: +61-(0)3-9903 9629; E-mail:
[email protected]
There is evidence to support structured approaches to pain management in residential aged care facilities (RACFs). This may include routine screening and assessment of pain, and stepped approaches to analgesic use [13, 14]. Poorly managed pain has been linked with increased aggression and
Pain is a frequent cause of discomfort and distress in residents in aged care facilities [1, 2]. Pain is associated with reduced quality of life, impaired physical function, sleep disturbance, and risk of falls. Optimising pain management requires consideration of pain intensity, resident frailty, disability and cognitive status [3]. Older people are underrepresented in clinical trials evaluating pain management interventions, resulting in limited evidence on which to base analgesic prescribing [4]. Analgesic selection is challenging due to the prevalence of frailty, multimorbidity and polypharmacy [3, 5, 6].
1574-8847/15 $58.00+.00
© 2015 Bentham Science Publishers
Prevalence of Analgesic Use and Pain in People with and without Dementia
healthcare service utilisation [11, 15, 16]. Optimising the use of simple analgesics such as paracetamol may reduce the need for opioids and thus minimise potential for opioidrelated ADEs [3]. This may be important because communitydwelling people with Alzheimer’s disease (AD) have a lower overall prevalence of analgesic use but higher prevalence of strong opioid use [17]. Pain expressed as behavioural symptoms may prompt prescribing of psychotropic medications, which in turn are associated with ADEs including cognitive impairment, falls and fractures, and death [11, 18, 19]. Structured approaches to detect and manage pain may reduce agitation and minimise psychotropic use in RACFs [13]. Despite the benefits of adequate pain management in residents with dementia in RACFs, there is inconsistency in the literature regarding pain and analgesic use in this population [8]. The aim of this systematic review was to determine the prevalence of analgesic drug use among people with and without dementia or cognitive impairment who live in RACFs. METHODS Search Strategy A search of the literature was undertaken using MEDLINE and EMBASE (from inception to January 2014). Medical Subject Headings (MeSH), Emtree subject headings and key words related to analgesics (analgesics, pain, paracetamol, NSAIDs, opioids) AND dementia (dementia, Alzheimer’s disease, cognitive impairment) AND aged care facilities (nursing homes, long term care, homes for the aged, residential aged care, aged care facility, skilled nursing facilities) were used (see Box 1). Searches were limited to English-language articles and excluded conference abstracts. Reference lists of studies identified and other review articles were screened for additional relevant studies.
Box 1.
Current Clinical Pharmacology, 2015, Vol. 10, No. 3
195
Inclusion and Exclusion Criteria Studies were included in the review if they reported: •
original research involving cross-sectional assessment of analgesic prevalence among residents of aged care facilities with and without dementia or cognitive impairment;
•
prevalence data separately for residents with and without dementia or cognitive impairment drawn from the same sample and presented within the same study; and
•
prevalence data for residents of aged care facilities separately in studies that included residents across multiple settings.
Studies were excluded if they met any of the following conditions: •
studies focused on a single class of analgesics (e.g. opioids);
•
all study participants had specific disease or disorder (other than dementia); and
•
publications that only reported the prevalence of analgesic use in residents with dementia or cognitive impairment.
Study Selection The titles and abstracts of studies were screened for relevance by one author (ET). Full-text copies were obtained if a study appeared to meet the inclusion criteria or it was unclear whether it would meet the criteria. Two authors (ET and NJ) independently reviewed the full texts to assess each study’s suitability for inclusion. Disagreements or uncertainties about study inclusion were resolved by discussion in the presence of a third author.
Search strategy.
MEDLINE Exp analgesics/ or Pain/dt [Drug Therapy] or analgesic$.mp. or acetaminophen$.mp. or paracetamol$.mp. or coxib$.mp. or NSAID$.mp. or opioid$.mp. AND exp dementia/ or dement$.ti, ab. or alzheimer$.mp. or cognitive$ impair$.mp AND Nursing homes/ or Nursing home$.ti. or Long term care.ti. orLongtermcare.ti or Homes for the aged/ or Residential aged care.mp. or LTCF.mp. or Aged care facilit$.mp. or Skilled nursing facilities/ EMBASE 'analgesic agent'/exp OR analgesic* OR acetaminophen* OR paracetamol* OR coxib* OR nsaid* OR opioid* AND 'dementia'/exp OR dement*.ti,ab. OR alzheimer* OR cognitive* AND impair* AND 'nursing home'/exp OR 'nursing home patient'/exp OR 'long term care'/exp OR 'home for the aged'/exp OR 'residential care'/exp OR nursing AND home*.ti OR long AND term AND care.ti OR longterm AND care.ti OR residential AND aged AND care OR ltcf OR aged AND care AND facility*
196 Current Clinical Pharmacology, 2015, Vol. 10, No. 3
Data Extraction and Validity Assessment Data were extracted independently by two authors (ET and NJ) using a standardised data abstraction form. Data extracted included year of publication, study design, study setting, study duration, study sample characteristics (age, gender and cognitive function), sample size, year of data collection, definition of analgesics and method used to determine analgesic use, and method used to determine presence of dementia and/or cognitive impairment. Regarding the prevalence of analgesic use, data extracted included: use of at least one analgesic drug, use of specific analgesic categories, and report of regular or as needed (prn) use of analgesics. Of the articles reporting analgesic drug use for participants with pain, the prevalence of pain and/or painful conditions (as categorised in the original studies), definition of pain and data collection method used were also abstracted. Data were extracted separately for both participants with and without dementia or cognitive impairment. When this stratification was clearly defined in the original article, the definition of dementia or cognitive impairment provided by the authors was used. When it was not clearly defined but stratification was possible using the available data, cognitive status categories were regrouped into “dementia/cognitive impairment” and “non-dementia/ non-cognitive impairment” groups for the purposes of this review (e.g. cognitive impairment defined as Mini Mental State Examination (MMSE) ≤ 23). Methodological quality of
Fig. (1). Flow chart of study selection. *after contacting authors to obtain additional data where possible RACF = residential aged care facility
Tan et al.
studies was assessed using an adapted version of the Newcastle-Ottawa Scale modified for cross-sectional studies (See Appendix 1). Attempts were made to contact 15 authors to clarify details or obtain additional data as needed. Meta-Analysis Meta-analysis was performed using Comprehensive Meta-analysis (Biostat, Inc, Englewood, NJ). Data were pooled using a random effects model and heterogeneity was explored using I2 statistics. As only one study reported odds ratios (ORs) adjusted for pain and pain-related diagnoses [20], unadjusted ORs were used for consistency. If ORs were not reported, these were calculated using data reported in the paper or obtained directly from authors. RESULTS Search and Study Selection The electronic database searches retrieved 686 articles. An additional 5 articles were identified by a manual search of relevant review articles and reference lists. After removal of duplicates, the titles and abstracts of 446 studies were reviewed, of which 413 were excluded because they clearly did not meet the inclusion criteria. The full text versions of 33 articles were obtained and scrutinised, of which 26 were
Prevalence of Analgesic Use and Pain in People with and without Dementia
excluded after independent review by at least two investigators (Fig. 1). A total of seven studies were included Table 1.
Current Clinical Pharmacology, 2015, Vol. 10, No. 3
197
in the final review and are summarised below and in Tables 1 and 2.
Methodological characteristics of studies of analgesic use in persons with and without dementia in residential aged care facilities.
Author, Year
Study Design and Setting
Dementia Definition and Data Collection Method
Pain definition and Data Collection Method
Analgesic Definition and Data Collection Method
Time of Data Collection
Achterberg, 2007 [22]
Crosssectional Netherlands
de Souto Barreto, 2013 [20]
Crosssectional France
Formal diagnosis of dementia reported in resident's medical chart or subjectively made by NH medical staff by answering 'do you believe this patient has dementia?'
Recognized pain complaints: NH medical staff was asked “Does the patient currently complain about pain?”
ATC codes: any analgesic (N02), NSAIDs (M01A), Analgesic use: participants who had taken 1 or more analgesics (N02) in the week they were included in the study
2011
Haasum, 2011 [23]
Crosssectional Sweden
Dementia diagnosis determined according to DSM-IV criteria by 2 physicians independently, with a third consulted if disagreement occurred. MMSE used to control for dementia severity
Patient self-report: patients were asked whether they had experienced any pain during the last 4 weeks.
ATC codes: any analgesic (N02), paracetamol (N02BE01), opioids (N02A), NSAIDs (M01A); Medical records
2001-2004
Kölzsch, 2012 [24]
Crosssectional Germany
MMSE via interview with resident
Patient self-report: pain intensity assessed with 6 point verbal descriptor scale (for patients MMSE≥10) Observation: quantitative scale (for MMSE
