1 Measuring health-related quality of life of care home residents Age and Ageing 407 – 413 2019; 48: © The Author(s) 2019. Published by Oxford University Press on behalf of the British Geriatrics Society. doi: 10.1093/ageing/afy191 This is an Open Access article distributed under the terms of the Creative Commons Attribution- Published electronically 7 January 2019 NonCommercial-NoDerivs licence (http://creativ, which on of the work, in any medium, provided the original permits non-commercial reproduction and distributi work is not altered or transformed in any way, and that the work is properly cited. For commercial re- [email protected] use, please contact Measuring health-related quality of life of care home residents: comparison of self-report with staff proxy responses Downloaded from by guest on 11 May 2019 2 1 1 1 3 L U ,K ATHRYN H INSLIFF -S MITH ,S ,A NNABELLE ARAH ONG L ,G EMMA H OUSLEY EWIS , A DEELA SMAN 4 5 4 6,7,8 1,7,8,9 G AGE AKE ,T OM D ENING J ,J OHN RFG ORDAN LADMAN ,A DAM LG ORDON ,H EATHER J 1 Division of Medical Sciences and Graduate Entry Medicine, School of Medicine, University of Nottingham, Derby, UK 2 Division of Epidemiology and Public Health, School of Medicine, University of Nottingham, Nottingham, UK 3 East Midlands Academic Health Science Network, Nottingham, UK 4 Department of Clinical and Experimental Medicine, Surrey Health Economics Centre, University of Surrey, Guildford, UK 5 Division of Psychiatry and Applied Psychology, School of Medicine, University of Nottingham, Nottingham, UK 6 Division of Rehabilitation and Ageing, University of Nottingham, Nottingham, UK 7 East Midlands Collaboration for Leadership in Applied Health Research and Care, Nottingham, UK 8 Nottingham Biomedical Research Centre, University of Nottingham, Nottingham, UK 9 School of Health Sciences, City, University of London, London, UK Address correspondence to: Dr Adam Gordon, Derby Medical School, Royal Derby Hospital, Room 4113, Derby DE22 3NE, [email protected] 01332 724697; Email: + 01332 724668; Fax: UK. Tel: + Abstract care home residents are often unable to complete health-related quality of life questionnaires for themselves Introduction: because of prevalent cognitive impairment. This study compared care home resident and staff proxy responses for two mea- sures, the EQ-5D-5L and HowRU. ≥ 60 years across 24 care homes who were not receiving short stay, Methods: a prospective cohort study recruited residents respite or terminal care. Resident and staff proxy EQ-5D-5L and HowRu responses were collected monthly for 3 months. fi Weighted kappa statistics and intra-class correlation coef cients (ICCs) adjusted for clustering at the care home level were used to measure agreement between resident and proxies for each time point. The effect of staff and resident baseline vari- ables on agreement was considered using a multilevel mixed effect regression model. Results: at 1, 2 and 3 months, respectively. When clustering was 117, 109 and 104 matched pairs completed the questionnaires controlled for, agreement between resident and staff proxy EQ-5D-5L responses was fair for mobility (ICC: 0.29) and slight for all other domains (ICC ≤ 0.20). EQ-5D Index and Quality-Adjusted Life Ye ar scores (proxy scores higher than residents) showed better agreement than EQ-5D-VAS (residents scores hi gher than proxy). HowRU showed only slight agreement (ICC fl uence level of agreement for either index. 0.20) between residents and proxies. Staff and resident characteristics did not in ≤ Discussion: the levels of agreement for EQ-5D-5L and HowRU raise questions about their validity in this population. Keywords Residential facilities, nursing homes, quality of life, proxy, older people. 407

2 A. Usman etal. Key points • Dementia is prevalent in UK care homes, limiting the usefulness of self-reported quality of life measures. This study found agreement between resident and proxy responses for EQ-5D-5L and HowRU quality of life measures • was inadequate. • Further work is required to better describe health-related quality of life as an outcome measure in the care home sector. asks respondents to indicate on a thermometer how they feel Introduction that day, with anchor points of 100 (best possible health) and Long-term care facilities in the UK are called care homes 0 (worst possible health). and classi fi ed as either care homes with or without nursing Downloaded from by guest on 11 May 2019 The prevalence of frailty and cognitive impairment in the based upon the availability of registered nurses on-site. The care home population means that collecting self-reported qual- fi cations of facility types of residents cared for in both classi ity of life measures from residents is challenging. In response are similar and all UK care homes are included in the inter- to this, proxy responses to quality of life items have sometimes nition of a nursing home [ fi national consensus de ]. 1 been used [ ]. For these, a consultee, drawn from care home 11 Around 425,000 people live in UK care homes [ ] with 2 staff, or a relative or friend who has regular ongoing visits, most residents requiring care due to disability from long- sbehalf.Usingproxy answers questions on the resident ’ term conditions. The majority of residents are aged over respondents can be unreliable in care home settings. There 85, 75 – 80% of residents live with dementia [ ], and over 3 may be lack of continuity of care home staff contact with indi- half of the residents die within 12 months of arrival [ 4 ]. vidual residents due to shift working and staff turnover, and Improving the quality of care for older people in long- family and friends may not be well placed to judge QoL term care has become a focus of attention both within the domains if they visit residents for only short periods [ 11 ]. UK and internationally [ 1 ], and an increasing number of There is limited evidence comparing self-reported and evaluative research studies are testing the effectiveness and proxy responses to the EQ-5D-5L in care home popula- cost-effectiveness of interventions in this setting [ 5 ]. tions [ ]. There is a particular paucity of data in UK care 12 Residents quality of life is frequently used as an outcome ’ home populations. measure in these studies to maintain a patient-centred focus ‘ HowRu ( ) is a patient-rated outcome meas- ’ How Are You and facilitate health economic evaluation. callydesignedforuseinlong-term ure which has been speci fi The EuroQoL EQ-5D questionnaires are widely used care settings to address quality of life in a way that is practical preference-based health-related quality of life measures suitable for older people [ ]. It records four variables (pain or dis- 13 , 14 for use in economic evaluations. They were speci fi cally designed d, limitation in activities and comfort, feeling low or worrie to be quick and easy to complete. The fi rst version of EQ-5D dependency on others) related to QoL at a fi xed point in time fi ve domains of quality of life on three levels (EQ- measured ( ’ ) on a four-point scale (none, How are you doing today? ‘ 5D-3L). EQ-5D-3L has been shown to have good construct slight, quite a lot, extreme). The HowRu score is calculated by validity for self-report [ 6 ] and has been used to measure quality summing up the values for each domain to give a value on a of life of older people living in their own homes and in care 13-point scale ranging from 0 (worst) to 12 (best). HowRU ]The ve-level version, EQ-5D-5L, was developed 7 fi homes [ may have greater cogency and immediacy than EQ-5D-5L. In ed issues with sensitivity and a fi subsequently to deal with identi a comparison with EQ-5D in patients attending a cardiovascu- ceiling effect on the EQ-5D-3L which limited its ability to dis- lar outpatient clinic, HowRu was reported to have better read- criminate between health states, particularly in those with higher ability, higher completion rate and report a wider range of quality of life [ 8 ]. The EQ-5D-5L version measures health- states [ ]. HowRu has not been evaluated for older people 13 related quality of life across fi ve domains (mobility, self-care, living in care homes. It is not known whether proxy responses usual activities, pain, and anxiety/depression) with the scale for in this setting may be useful for HowRU. each domain ranging from level 1 (no problems) to level 5 This study was conducted as a part of a programme of fi ve domains are (extreme problems). The responses from the work focussing on improving quality of care in UK care homes converted to QoL index scores (utilities) generated from a given which used EQ-5D-5L and HowRu as outcome measures. To country ’ s general population [ ]. These index scores can be 9 inform our use of proxy measures, we set out to establish the used to calculate quality-adjusted life years (QALYs), which reliability of staff proxy responses for both indices. s state of health. One QALY ’ are a measure of the person equates to one year in perfect health. The cost per QALY gained from an intervention when compared to usual care is Method the chosen cost-utility measure for determining eligibility for funding support through the UK National Health Service The full study protocol has been published [ 15 ]. [ 10 ]. EQ-5D-5L also includes a visual analogue scale which Participants were a sub-population of care home residents 408

3 Measuring health-related quality of life of care home residents staff and directly addressed any questions whilst they com- recruited as a part of the Proactive Health Care in Care pleted their responses but did not otherwise direct them. Homes (PEACH) study. The PEACH study includes an Responses for both self-reported and proxy questionnaires open cohort stepped wedge randomised trial to assess the were completed on the same day. impact of Comprehensive Geriatric Assessment implemen- Analysis was based on cross-sectional analysis of agree- ted by Quality Improvement Collaborative. Comprehensive ment at each time point. For the EQ-5D-5L and HowRu Geriatric Assessment is widely recognised as a gold- domain levels, the levels of agreement between self- standard way to deliver care for older people with frailty reported and staff responses were calculated using percent [ ]. PEACH uses EQ-5D-5L and HowRu as outcome 16 agreement and weighted kappa statistics at 1, 2 and 3 measures and understanding their measurement properties months. Weighted kappa helps to distinguish between small represented important preparatory work. The measure- and large differences in agreement ratings assigned to the ments for this study coincided with months 3 – 6 of the different levels of each domain but with equal importance PEACH study, with data collection for PEACH continuing given to disagreement [ ndings of this study were fi for a further 6 months. The , 18 19 ]. We used linear weights for Downloaded from by guest on 11 May 2019 therefore able to inform the PEACH analysis. the weighted kappa: this assigns the same importance to the All residents of care homes participating in PEACH difference between any two categories within the response who were aged ≥ 60 years were eligible for inclusion. Those scale [ fi dence interval for the weighted ]. The 95% con 20 who were admitted for short term respite or immediately kappa was calculated at each time by bootstrapping using approaching end of life were excluded. Informed consent Stata 15 (Statacorp, LLC, 2015) with 1000 replications. was obtained from residents who had mental capacity and The kappa statistic ranges from − 1 to 1, and the from an appropriate consultee when residents lacked mental strength of the agreement was interpreted with regards to capacity. Capacity assessments were based on the guidelines published guidelines [ 21 ] with agreement being: in the 2005 Mental Capacity Act for England and Wales. • Poor, if kappa ≤ 0.00 For this study, in addition to the routine collection of – • Slight, if kappa = 0.01 0.20 EQ-5D-5L and HowRu from residents undertaken as part 0.40 – Fair, if kappa = 0.21 • of PEACH, proxy responses to EQ-5D-5L and HowRu 0.60 – Moderate, if kappa = 0.41 • were gathered from staff. We included staff such as care Substantial, if kappa = 0.61 – 0.80 • assistants, care home managers and registered nurses, who 0.81 ≥ Almost perfect, if kappa • were identi fi ed by the care home manager as most familiar with the resident. This placed emphasis on staff providing For the EQ-5D visual analogue scale, EQ-5D-5L index personal care to the resident on the day of data collection scores, QALYs and HowRu scores, the levels of agreement because both EQ-5D-5L and HowRu ask about the resi- between the self-reported and proxy responses were assessed dent ’ s health today. We excluded staff employed in a sup- by calculating the intra-class correlation coef fi cient (ICC) at portive role, such as activity coordinators, since their each time point using a two-way mixed effect analysis of vari- orientation to supporting residents is more variable and ance (ANOVA) model [ ]. Although the ANOVA model 22 they are less likely to be involved in personal care. has been reported to be robust to deviation in normality, Data were collected from proxies in three consecutive bootstrapping was run to assess if it made any difference to months during 2017 and matched with resident data for the estimated ICCs. The same benchmarks used for kappa those months. The EQ-5D-5L questionnaire, including the were used for the intra-cluster correlation coef fi cients. EQ-5D visual analogue score, was used. Responses from ICCs were calculated for EQ-5D-Self, EQ-5D-5L- ve domains were transformed into utilities (index the fi Proxy, HowRu-Self and HowRu-Proxy. Since the calcula- scores) derived from the UK general population. This was tion of kappa and ICC assumes independence of observa- done using the crosswalk value set [ ing. In our study, clustering tions, we adjusted for cluster 17 ]. For residents with may have occurred at three levels. First, at the care home proxy and self-reported responses at all three time points, level where residents within the same care home have QALYs were calculated using the area under the curve. similar characteristics and are different from those in other HowRu has four domains scaled from 0 to 3. Values for care homes. Second, at staff level where staff members each domain were none = 3, slight = 2, quite a lot = 1 and within a care home respond on behalf of multiple resi- extreme = 0. These were summed to give a 13-point scale dents, and third at the individual level where responses are ranging from 0 (worst health) to 12 (best health). clustered within each resident. To standardise responses, taking account of residents For the ICCs, clustering was adjusted for using a multi- who were unable to read or write, the researchers read tting a two-level random fi level mixed effect model by the questionnaire for both EQ-5D-5L and HowRu to effect model with a random effect for care home and indi- participants and then recorded their responses. For staff viduals. For the kappa statistics, clustering was adjusted for s ’ resident – responses, we asked them to consider the proxy using a variance formula [ perspective when completing the questionnaire using the 15 ]. ‘ Please rate how you (staff) think the following statement: The study sample size was 160 residents, based upon a resident will rate his/her own health-related quality of life, kappa of 0.145 and a con fi dence level width of 0.153 taken ’ if the resident was to communicate . A researcher sat with from a previous study [ 23 ]. 409

4 A. Usman etal. adjusted ICC: 0.24) was less than for index EQ-5D-5L Results scores (cluster adjusted ICC: 0.55), and agreement for the 117, 109 and 104 matched pairs completed the question- latter was less than for QALYs (cluster adjusted ICC: 0.70). naires at 1, 2 and 3 months, respectively. The mean (SD) At all-time points EQ-5D visual analogue scale showed a age of the residents was 86.8 (7.6) years and 68% were slight to fair agreement, EQ-5D-5L index scores showed female. Forty-four percentage of participants had a docu- substantial agreement, and QALYs showed sub- moderate – mented diagnosis of dementia or cognitive impairment in stantial agreement between residents and proxy responses. their care home record. The characteristics of staff who When regression analysis was conducted on resident and provided proxy responses are reported in Table . 1 staff characteristics to consider their impact on the differ- The agreement between proxies and residents for ence between EQ-5D-5L and HowRu scores, no statistic- individual domains of the EQ-5D-5L and HowRU, respect- fi fi ed. cant associations were identi ally signi ively, are summarised in Tables 2 , respectively. The 3 and strength of agreement found between staff and residents Discussion for HowRu measure was weaker than for EQ-5D-5L. The Downloaded from by guest on 11 May 2019 intra-cluster correlation showed clustering of measures This study compared UK care home resident and staff within care homes. When kappa values were adjusted for proxy responses to the EQ-5D-5L and HowRu. Agreement clustering, agreement was fair for the mobility domain of for the domains of both the EQ-5D-5L and HowRu were EQ-5D-5L and slight for all other domains. Agreement slight when clustering was accounted for, with the exception was slight for all domains of HowRU when clustering was of the EQ-5D-5L mobility domain where agreement was accounted for. still only fair. EQ-5D-5L total and HowRu scores reported Mean total resident EQ-5D-5L (0.57, 0.50, 0.58) and by residents were higher than when reported by proxies, yet HowRu (9.4, 9.2, 9.6) scores were higher than proxy EQ- the mean EQ-5D visual analogue scale was higher in prox- 5D-5L (0.43, 0.42, 0.42) and HowRu (8.4, 8.3, 9.0) scores ies than residents, indicating further concern about the at all three time points. By contrast, the mean EQ-5D-VAS measurement of health-related quality of life in care home was higher in proxies (68, 74, 72) compared to residents residents. (65, 63, 69) across all time points. A strength of this study was that analyses controlled The strength and magnitude of agreement between resi- for clustering, important due to the heterogeneity of UK dents and proxies for EQ-5D visual analogue scale (cluster care homes, where the resident case-mix and staff skill- mix vary substantially between institutions [ 3 ]. We were further able to understand how resident and staff attri- Characteristics of care home staff Table 1. butes in uenced agreement by using regression analysis fl to consider their impact and found that they had no (%) at N Characteristic in fl uence by doing so. baseline ... The main limitation of the study was in the low preva- Age group ( n = 117) lence of dementia reported in the cohort. Other studies 35 years 39 (33.3%) 18 – have shown the prevalence of dementia in UK care homes 55 years 36 50 (42.7%) – to be close to 80% [ ]. We found that a formal diagno- , 24 3 28 (23.9%) Aged 56 or older sis of cognitive impairment did not impact on the level of Sex ( = 117): numbers of female staff 103 (88.0%) n n Role/rank ( = 117) agreement between residents and proxies and therefore, 89 (76.1%) Care worker or health care assistant even though the sample here was unrepresentative of the 6 (5.1%) Registered nurse care home population as a whole, we do not think this Other 22 (18.8%) uences the validity of our in fl fi ndings about EQ-5D-5L and Care home assistant practitioner 2 (1.7%) HowRU proxy measurements in this setting. Nursing assistant 6 (5.1%) Senior care assistant 13 (11.1%) The main reason for wanting to use staff proxy Deputy Manager 1 (0.9%) responses for EQ-5D-5L in the care home population is Length of time working in study care home ( n = 117) the high prevalence of cognitive impairment. However, the Less than 6 months 15 (12.8%) ndings here match those in studies using EQ-5D-3L, and fi 5 (4.3%) 11 months – 6 done in long-term care sectors in other countries, which – 1 5 years 42 (35.9%) More than 5 years 55 (47.0%) suggest that staff proxy ratings consistently differ from Length of time working in care of older people ( n = 117) those of residents, for residents with or without cognitive Less than 6 months 7 (6.0%) impairment [ ]. The reason for these differences is not 25 , 11 6 – 11 months 3 (2.6%) clear. It may be that staff and residents understand the 31 (26.5%) 1 – 5 years domains included in health-related quality of life measures 76 (65.0%) More than 5 years Frequency of delivering care to resident ( n = 107) differently. It may, alternatively, be that indices developed in 86 (73.5%) Most/all of the time non-care home settings, do not include the sort of domains 19 (16.2%) Sometimes upon which care home staff feel they can reliably comment. 2 (1.7%) Rarely Further work is required to understand the ways in which 410

5 Measuring health-related quality of life of care home residents Table 2. proxy agreement using the exact percent agreement and Kappa values for the EQ-5D-5L at three points Resident – in time Time point (month) Kappa coef Domain cient (95% CI) Kappa adjusted for Number of clusters fi clustering (95% CI) (range in cluster size) ... 0.48 0.22 Mobility 1 16 (0.36,0.59) (2 – 15) = 117) ( n (0.01, 0.46) 11 2 0.56 0.33 (0.44 – 0.65) (0.16, 0.50) (3 – 14) n ( = 109) 0.33 14 3 0.48 = 104) (0.16, 0.50) – 0.60) n (2 – 14) ( (0.35 0.36 14 1 Self-care 0.19 = 117) (0.26 – 0.46) (0.04, 0.35) (2 – 15) ( n 0.10 0.25 2 14 = 109) (0.15 – n (0.00, 0.21) (2 – 14) ( 0.38) Downloaded from by guest on 11 May 2019 0.23 12 3 0.33 (0.21 = 104) (0.12,0.35) (2 – 14) n – ( 0.44) 14 0.15 1 Usual activities 0.02 0.28) – − 0.17, 0.21) (2 n 15) = 117) ( (0.02 ( – 13 0.13 2 0.26 – 0.39) (0.00, 0.25) (2 – ( n = 109) (0.12 14) 11 3 0.17 0.09 – – ( − 0.02, 0.20) (2 ( 14) 0.30) n = 104) (0.02 14 0.14 0.22 1 Pain/discomfort ( – 0.34) ( − 0.02,0.30) (2 – 15) n = 117) (0.11 0.16 2 0.20 11 (0.08 = 109) (0.05,0.28) (5 – 14) – 0.31) n ( 10 0.11 3 0.14 14) – 0.26) ( − 0.00,0.23) (5 – = 104) ( n (0.03 0.05 10 Anxiety/depression 1 0.08 15) ( 0.03 = 117) 0.21) (0.09,0.19) (2 – − ( n – 0.08 9 0.10 2 = 109) ( − 0.03 – 0.23) ( − 0.02 – 0.18) (2 – 14) ( n 0.14 12 3 0.24 (2 (0.09 ( − 0.07 – 0.35) 0.42) – 14) n ( = 104) – – proxy agreement using the exact percent agreement and Kappa values for HowRu at three points in time Table 3. Resident Kappa coef Time point (month) cient Domain Kappa adjusted for clustering fi Number of clusters (range in cluster size) (95% CI) (95% CI) ... 1 0.25 0.14 14 Pain/discomfort = 117) (0.14 – 0.35) ( − 0.02,0.30) (2 – 15) ( n 0.16 11 2 0.18 (0.06 = 109) (0.05,0.28) (5 – 14) – n ( 0.31) 10 0.11 0.16 3 0.00,0.23) 0.29) ( − n (5 – 14) = 104) ( – (0.05 0.16 13 Feeling low or worried 1 0.22 (0.09 n 0.37) (0.02,0.30) (3 – 15) ( = 117) – 11 0.16 2 0.20 (0.07 – 0.35) (0.00,0.35) (3 ( 14) n = 109) – 0.09 11 0.14 3 = 104) (0.01 – 0.28) ( − 0.18,0.36) (2 – 14) ( n 0.15 13 1 Limited in what you can do 0.03 = 117) (0.02 – 0.27) ( − 0.10,0.17) (3 – 15) ( n 0.00 14 2 0.09 (2 ( – 0.21) ( − 0.13,0.14) 0.02 – 14) ( − n = 109) 12 3 0.15 0.18 (2 (0.03 0.29) ( − 0.04, 0.40) – – 14) = 104) n ( 0.17 0.11 13 Dependent on others 1 = 117) – n 0.27) ( − 0.01,0.24) (3 ( (0.07 15) – 2 0.13 12 0.20 = 109) (0.10 – 0.30) (0.03,0.23) (3 – 14) ( n 0.21 0.10 3 11 0.05,0.26) = 104) – 0.33) n − (0.09 (2 – 14) ( ( 411

6 A. Usman etal. (London Bromley REC ref: 205840; University of proxy respondents understand existing measures, whether Nottingham ref: LT07092016). The PEACH study protocol these measures can be adapted to take account of current has been reviewed as a part of good governance by the fi fi c measures culties, or whether new care home-speci dif Nottinghamshire Healthcare Foundation Trust. are required. Several other health-related quality of life indices have been developed to take speci fi c account of dementia. The most notable amongst these are QUALIDEM [ 26 ], References DeMQoL [ ]. DeMQoL was developed 28 ] and QoL-AD [ 27 in non-institutional community settings and the proxy 1. Sanford AM, Orrell M, Tolson D . An international def- etal responses within it have not been validated for care homes, nursing home ‘ inition for . J Am Med Dir Assoc 2015; 16: ’ where carer relationships are different to those for patients 181 – 4. cared for in their own homes. QUALIDEM was developed 2. Laing & Buisson. Care of the Elderly People Market Survey, in the long-term care setting and is an observation-based 2009. London: Laing and Buisson, 2010. Gordon AL, Franklin M, Bradshaw L, Logan P, Elliott R, 3. Downloaded from by guest on 11 May 2019 – retest reliability but some issues measure, with good test Gladman JR. Health status of UK care home residents: a with inter-observer agreement, with four out of the nine cohort study. Age Ageing 2014; 43: 97 – 103. subscales showing poor – moderate agreement only [ 29 ]. etal . The provision of care for 4. Kinley J, Hockley J, Stone L Work comparing QoL-AD with EQ-5D in people with residents dying in UK nursing care homes. Age Ageing 2014; dementia suggested that patients and their carer proxies 43: 375 – 9. fl uenced by differ- applied different constructs and were in National Institute of Health Research Dissemination Centre. 5. ent baseline variables, when providing quality of life rat- Advancing Care: Research within Care Homes. Available at ings [ 30 ]. Our work here, which questions the utility of EQ-5D-5L and HowRU in care home residents with themed-review.pdf (20 July 2018, date last accessed). more advanced cognitive impairment, underlines the etal Janssen MF, Pickard AS, Golicki D 6. . Measurement prop- inability of current health-related quality of life indices to erties of the EQ-5D-5L compared to the EQ-5D-3L across eight patient groups: a multi-country study. Qual Life Res accurately inform research and practice in this group. 27. – 2013; 22: 1717 Further work is needed. Borowiak E, Kostka T. Predictors of quality of life in older 7. In conclusion, we recommend that staff proxy responses people living at home and in institutions. Aging Clin Exp Res for EQ-5D-5L are treated with caution in care home stud- 2004; 16: 212 20. – ies. Staff responses for HowRU are not a good proxy for 8. Devlin NJ, Brooks R. EQ-5D and the EuroQol group: fi cult to envisage a scenario resident responses and it is dif past, present and future. Appl Health Econ Health Policy in which they would be useful. – 2017; 15: 127 37. Devlin NJ, Shah KK, Feng Y, Mulhern B, van Hout B. 9. Valuing health-related quality of life: an EQ-5D-5L value set Acknowledgements: The authors would like to acknow- for England. Health Econ 2018; 27: 7 – 22. ledge the broader PEACH study team: Mr Zimran Alam, 10. McCabe C, Claxton K, Culyer AJ. The NICE cost- Ms Anita Astle, Professor Tony Avery, Dr Jaydip Banerjee, effectiveness threshold: what it is and what that means. Professor Clive Bowman, Dr Neil H Chadborn, Mr Pharmacoeconomics 2008; 26: 733 – 44. etal Kane RL, Kane RA, Bershadsky B 11. . Proxy sources for Michael Crossley, Dr Reena Devi, Professor Claire ’ quality of life. information on nursing home residents Goodman, Professor Pip Logan, Professor Finbarr Martin, – 25. J Gerontol B Psychol Sci Soc Sci 2005; 60: S318 Professor Julienne Meyer, Professor Dominick Shaw, 12. Bulamu NB, Kaambwa B, Ratcliffe J. A systematic review of Professor Rowan Harwood and Dr Maria Zubair. instruments for measuring outcomes in economic evaluation None. fl Declaration of Con icts of Interest: within aged care. Health Qual Life Outcomes 2015; 13: 179. 13. Benson T, Potts HWW, Whatling JM, Patterson D. This work has been Declaration of Sources of Funding: Comparison of howRU and EQ-5D measures of health- conducted as a part of the Proactive Healthcare of Older related quality of life in an outpatient clinic. Inform Prim People in Care Homes (PEACH) study, supported by the Care 2013; 21: 12 – 7. Dunhill Medical Trust, award number FOP1/0115. 14. Benson T, Williams DH, Potts HWW. Performance of EQ- 5D, howRu and Oxford hip & knee scores in assessing the outcome of hip and knee replacements. BMC Health Serv Res 2016; 16: 512. Ethics 15. etal . Measuring health Usman A, Lewis S, Hinsliff-Smith K related quality of life of care home residents, comparison of This study is a part of preparatory work for the larger self-report with staff proxy responses for EQ-5D-5L and Proactive Healthcare of Older People in Care Homes HowRu: protocol for assessing proxy reliability in care home (PEACH) study. PEACH was reviewed by both the UK outcome testing. BMJ Open 2018; 8: e022127. Health Research Authority and the University of 16. Welsh TJ, Gordon AL, Gladman JR. Comprehensive geriatric Nottingham Research Ethics Committee and determined a guide for the non-specialist. Int J Clin Pract — assessment 2014; 68: 290 – 3. by both to be a service development and evaluation project 412

7 Sarcopenia and 1-year mortality in cancer Selai C, Pickard S, Johnston JA. The collection of EuroQol 25. van Hout B, Janssen MF, Feng Y-S . Interim scoring for etal 17. EQ-5D data via proxies. EQ Plenary Meet York Sept 2002. the EQ-5D-5L: mapping the EQ-5D-5L to EQ-5D-3L value Available at: 15. – sets. Value Health 2012; 15: 708 http// fi leadmin/user_upload/ Vanbelle S, Albert A. A note on the linearly weighted kappa 18. Documenten/PDF/Selai_Proxy_Guidance.pdf (June 27 2011, fi cient for ordinal scales. Stat Methodol 2009; 6: 157 – coef 63. date last accessed) 2002. 19. Brenner H, Kliebsch U. Dependence of weighted kappa coef- Ettema TP, Dröes R-M, de Lange J, Mellenbergh GJ, Ribbe 26. fi cients on the number of categories. Epidemiology 1996; 7: WM. QUALIDEM: development and evaluation of a demen- 199 – 202. tia speci fi c quality of life instrument –– validation. Int J Cicchetti DV, Allison T. A new procedure for assessing reli- 20. Geriatr Psychiatry 2007; 22: 424 – 30. ability of scoring EEG sleep recordings. Am J EEG Technol Chua K-C, Brown A, Little R 27. . Quality-of-life assessment etal 1971; 11: 101 – 10. in dementia: the use of DEMQOL and DEMQOL-Proxy Landis JR, Koch GG. The measurement of observer agree- 21. 18. – total scores. Qual Life Res 2016; 25: 3107 ment for categorical data. Biometrics 1977; 33: 159 – 74. 28. Logsdon RG, Gibbons LE, McCurry SM, Teri L. Assessing 22. Harwell MR, Rubinstein EN, Hayes WS, Olds CC. quality of life in older adults with cognitive impairment. Summarizing Monte Carlo results in methodological research: – 9. Psychosom Med 2002; 64: 510 Downloaded from by guest on 11 May 2019 the one- and two-factor fi xed effects ANOVA cases. J Educ Aspden T, Bradshaw SA, Playford ED, Riazi A. Quality-of- 29. Stat 1992; 17: 315 – 39. life measures for use within care homes: a systematic review 23. Devine A, Taylor SJC, Spencer A, Diaz-Ordaz K, Eldridge of their measurement properties. Age Ageing 2014; 43: S, Underwood M. The agreement between proxy and self- – 603. 596 completed EQ-5D for care home residents was better for 30. . Patient and proxy meas- etal Sheehan BD, Lall R, Stinton C index scores than individual domains. J Clin Epidemiol urement of quality of life among general hospital in-patients 43. 2014; 67: 1035 – with dementia. Aging Ment Health 2012; 16: 603 – 7. . Prevalence of etal Livingston G, Barber J, Marston L 24. and associations with agitation in residents with dementia liv- Received 27 July 2018; editorial decision 9 November 2018 ing in care homes: MARQUE cross-sectional study. BJPsych 8. doi:10.1192/bjpo.bp.117.005181. Open 2017; 3: 171 – 48: 413 418 Age and Ageing 2019; – © The Author(s) 2019. Published by Oxford University Press on behalf of the British Geriatrics doi: 10.1093/ageing/afy212 Society. All rights reserved. For permissions, please email: [email protected] Published electronically 4 January 2019 Impact of sarcopenia on 1-year mortality in older patients with cancer 2 1 3 1 1 S TOBÄUS L ,K INDSEY F RANZ O ,L AURENCE G ENTON TTEN ,U RSULA M ÜLLER -W ERDAN RISTINA , ,N ICOLE 4 1,5 W ,K RISTINA N ORMAN IRTH R AINER 1 — Universitätsmedizin Berlin, Humboldt-Universität zu Berlin and Berlin Institute of Health, Department of Geriatrics, Charité Research Group Geriatrics, Berlin, Germany 2 — Universitätsmedizin Berlin, Campus Mitte, Berlin Institute of Health, Clinical Research Unit, Germany Charité 3 Geneva University Hospitals, Clinical Nutrition, Geneva, Switzerland 4 Marien Hospital Herne, University Hospital Ruhr-Universität Bochum, Germany 5 Department of Nutrition and Gerontology, German Institute of Human Nutrition Potsdam-Rehbrücke, Nuthetal, Germany + Address correspondence to: K. Norman. Tel: ( 49) 304 5056 5139, Email: [email protected] Abstract sarcopenia is common especially in hospitalised older populations. The aim of this study was to assess the Objectives: fi ned as low skeletal mass and muscle strength, and its impact on 1-year mortality in older prevalence of sarcopenia, de patients with cancer. 2 skeletal muscle mass was estimated using bioelectric impedance analysis and related to height Methods: (SMI; Janssen etal. 2002). Grip strength was measured with the JAMAR dynamometer and the cut-offs suggested by the European 413

Related documents