CLINICAL INVESTIGATIONS
Insomnia in Long-Term Care Facilities: A Comparison of Seven European Countries and Israel: The Services and Health for Elderly in Long TERm care Study Jacob Gindin, MD,a Tamar Shochat, DSc,a Angela Chetrit, MSc,a Shulamit Epstein, PhD,a Yehoshua Ben Israel, MD,b Sarah Levi, PhD,c Graziano Onder, MD, PhD,d Ian Carpenter, MD,e Harriet Finne-Soveri, MD, PhD,f Hein van Hout, PhD,g Jean-Claude Henrard, PhD,h Thorsten Nikolaus, MD,i Eva Topinkova, MD, PhD,j Daniela Fialova, PharmD,k,l and Roberto Bernabei, MD,d for the SHELTER project
OBJECTIVES: To assess insomnia and its correlates as part of the Services and Health for Elderly in Long TERm care (SHELTER) study, funded by the 7th Framework Programme of the European Union. DESIGN: Cross-cultural investigation. SETTING: Long-term care facilities (LTCFs) in eight European countries (Czech Republic, France, Finland, Germany, England, the Netherlands, Italy) and one non-European country (Israel). PARTICIPANTS: Elderly residents (N = 4,156) of 57 LTCFs. MEASUREMENTS: Information on insomnia, age, sex, activities of daily living (ADLs), cognitive status, depression, major stressful life events, physical activity, fatigue, pain, and sleep medication use was extracted from the International Resident Assessment Instrument (interRAI) LTCF instrument. Rates of insomnia and its correlates were analyzed. Multivariate logistic regression was used to
From the aCenter for Standards in Health and Disability, University of Haifa, Mt. Carmel, Haifa; bShoham Geriatric Medical Center, Pardes Hana; cGeriatrics Division, Ministry of Health, Jerusalem, Israel; dCentro Medicina dell’Invecchiamento, Universita Cattolica Sacro Cuore, Rome, Italy; eCentre for Health Services Studies, University of Kent, Canterbury, UK; fNational Institute for Health and Welfare, Helsinki, Finland; g Department of General Practice and Elderly Care Medicine, VU University/EMGO, Amsterdam, The Netherlands; hResearch Unit, Health Environment and Ageing Sainte Perine Hospital and Medical Faculty Paris-Ile-de-France-Ouest, Paris, France; iDepartment of Geriatric Medicine, University of Ulm, Ulm, Germany; jDepartment of Geriatrics, 1st Medical Faculty, Charles University and Institute of Postgraduate Medical Education, kDepartment of Social and Clinical Pharmacy, Faculty of Pharmacy, Hradec Kralove, Charles University, and lInstitute for Postgraduate Education in Healthcare, Prague, Czech Republic. Address correspondence to Associate Professor Tamar Shochat, The Center for Standards in Health and Disability and the Department of Nursing, University of Haifa, 199 Aba Khoushy Ave., Mt. Carmel, Haifa 3498838, Israel. E-mail: [email protected] DOI: 10.1111/jgs.13099
JAGS 62:2033–2039, 2014 © 2014, Copyright the Authors Journal compilation © 2014, The American Geriatrics Society
assess factors associated with insomnia, controlling for demographic variables. RESULTS: The prevalence of insomnia was 24% (range 13–30%), with significant differences between countries (P < .001). More insomnia complaints were reported in older than younger residents (P < .001). Higher rates of insomnia were associated with hypnosedatives and depression in all countries (P < .001) and with stressful life events, fatigue, and pain in most countries (P < .001). No associations were found between insomnia and ADLs, physical activity, or cognitive status. Age, depression, stressful life events, fatigue, pain and hypnosedatives were independent significant predictors of insomnia, controlling for all other variables and for country. CONCLUSION: Hypnosedatives and depression were strong predictors of insomnia beyond cultural differences. Overall, psychosocial variables were more strongly related to insomnia than functional and mental capacities. J Am Geriatr Soc 62:2033–2039, 2014.
Key words: Insomnia; elderly; long-term care facility; InterRAI; hypnosedatives
nsomnia is common in the general elderly population.1–4 It has been associated with medical and psychiatric conditions3,5–7 and with impaired daily functioning,3,8,9 cognitive decline,10–13 poor health-related quality of life,7,14 risk of falls and hip fractures,15,16 and all-cause mortality.17,18 Furthermore, the use of hypnotic medications for the treatment of insomnia is most prevalent and most chronic in elderly adults19–21 and has been associated with depression and traumatic life events,20 poor functional status,20,22 and risk of injury.23,24 Although sleep disturbances are highly prevalent in nursing homes,25 few studies have specifically investigated
I
0002-8614/14/$15.00
2034
GINDIN ET AL.
insomnia and its correlates in long-term care facilities (LTCFs).15,26,27 Thus, in a Canadian study of more than 2,000 LTCF residents, only 6% had insomnia based on Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) criteria, although 17% had at least one insomnia symptom, yet more than half used benzodiazepines.26 Insomnia was associated with disruptive behaviors and psychological distress. In a study of more than 2,500 LTCF residents in Germany using the Minimum Data Set (MDS) 2.0, the prevalence of insomnia based on the two items referring to insomnia was 37% for difficulty falling and staying asleep and nearly 30% for nonrestful sleep or tired in the morning.27 Insomnia was associated with low levels of activities and social engagement and high levels of communication difficulties and interpersonal conflicts. In a 6-month longitudinal U.S. study of more than 34,000 LTCF residents using the MDS, insomnia was associated with greater risk for falls (odds ratio (OR) = 1.52, 95% confidence interval (CI) = 1.38–1.66) after adjusting for several sociodemographic and clinical confounders.15 The International Resident Assessment Instrument (InterRAI) for LTCFs is one component of a suite of instruments designed to provide ongoing standard assessment for a wide variety of healthcare settings and has been described elsewhere in detail.28 Validity and reliability of the instrument have been established previously29 and more recently by the Services and Health for Elderly in Long TERm care (SHELTER) study as part of the 7th Framework Programme of the European Union.28 The SHELTER study included three waves of longitudinal data from 4,156 nursing home residents from seven European countries and one non-European (Israel) country. Several publications based on the SHELTER database have demonstrated the effectiveness of the instrument for evaluation and comparison of various health concerns, such as polypharmacy,30 inappropriate drug use,31 pain and associated psychiatric symptoms,32 pneumococcal vaccine coverage,33 and depression34 in nursing homes. The aims of this investigation, which was based on the InterRAI LTCF evaluation baseline intake in the SHELTER study, were to evaluate the prevalence of insomnia in LTCFs of seven European countries and Israel, controlling for age, sex, and country, and to investigate the associations between insomnia and measures of depression, stress, fatigue, pain, functional status, cognitive status, physical activity, and hypnosedative medication use.
METHODS Participants Participants were elderly individuals (mean age 83.4 9.4; N = 4,156) residing in nursing homes from eight countries (Czech Republic, France, Finland, Germany, England, the Netherlands, Italy, and Israel) who agreed to participate in the project, as previously reported.28
NOVEMBER 2014–VOL. 62, NO. 11
JAGS
Primary Outcome Insomnia was defined as difficulty falling asleep or staying asleep, waking up too early, restlessness, or nonrestful sleep.
Predicting Variables Three summary InterRAI scales were used to evaluate functional status, cognitive status, and presence of depressive symptoms (for more information on scales, see28). Assessment of predicting variables included functional status (based on four categories of activities of daily living (ADLs): 0 = no impairment, 1–2 = limited impairment, 3–4 = extensive impairment, 5–6 = dependent), cognitive impairment (based on the cognitive performance scale (CPS): 0 = no impairment, 2–4 = moderate impairment, 5–6 = severe impairment), presence of depressive symptoms (based on the Depression Rating Scale (DRS): 0–2 = not depressed, >3 = depressed), major life stress (yes, no), physical activity (based on total hours of exercise per day: 0,
Insomnia in Long-Term Care Facilities: A Comparison of Seven European Countries and Israel: The Services and Health for Elderly in Long TERm care Study Jacob Gindin, MD,a Tamar Shochat, DSc,a Angela Chetrit, MSc,a Shulamit Epstein, PhD,a Yehoshua Ben Israel, MD,b Sarah Levi, PhD,c Graziano Onder, MD, PhD,d Ian Carpenter, MD,e Harriet Finne-Soveri, MD, PhD,f Hein van Hout, PhD,g Jean-Claude Henrard, PhD,h Thorsten Nikolaus, MD,i Eva Topinkova, MD, PhD,j Daniela Fialova, PharmD,k,l and Roberto Bernabei, MD,d for the SHELTER project
OBJECTIVES: To assess insomnia and its correlates as part of the Services and Health for Elderly in Long TERm care (SHELTER) study, funded by the 7th Framework Programme of the European Union. DESIGN: Cross-cultural investigation. SETTING: Long-term care facilities (LTCFs) in eight European countries (Czech Republic, France, Finland, Germany, England, the Netherlands, Italy) and one non-European country (Israel). PARTICIPANTS: Elderly residents (N = 4,156) of 57 LTCFs. MEASUREMENTS: Information on insomnia, age, sex, activities of daily living (ADLs), cognitive status, depression, major stressful life events, physical activity, fatigue, pain, and sleep medication use was extracted from the International Resident Assessment Instrument (interRAI) LTCF instrument. Rates of insomnia and its correlates were analyzed. Multivariate logistic regression was used to
From the aCenter for Standards in Health and Disability, University of Haifa, Mt. Carmel, Haifa; bShoham Geriatric Medical Center, Pardes Hana; cGeriatrics Division, Ministry of Health, Jerusalem, Israel; dCentro Medicina dell’Invecchiamento, Universita Cattolica Sacro Cuore, Rome, Italy; eCentre for Health Services Studies, University of Kent, Canterbury, UK; fNational Institute for Health and Welfare, Helsinki, Finland; g Department of General Practice and Elderly Care Medicine, VU University/EMGO, Amsterdam, The Netherlands; hResearch Unit, Health Environment and Ageing Sainte Perine Hospital and Medical Faculty Paris-Ile-de-France-Ouest, Paris, France; iDepartment of Geriatric Medicine, University of Ulm, Ulm, Germany; jDepartment of Geriatrics, 1st Medical Faculty, Charles University and Institute of Postgraduate Medical Education, kDepartment of Social and Clinical Pharmacy, Faculty of Pharmacy, Hradec Kralove, Charles University, and lInstitute for Postgraduate Education in Healthcare, Prague, Czech Republic. Address correspondence to Associate Professor Tamar Shochat, The Center for Standards in Health and Disability and the Department of Nursing, University of Haifa, 199 Aba Khoushy Ave., Mt. Carmel, Haifa 3498838, Israel. E-mail: [email protected] DOI: 10.1111/jgs.13099
JAGS 62:2033–2039, 2014 © 2014, Copyright the Authors Journal compilation © 2014, The American Geriatrics Society
assess factors associated with insomnia, controlling for demographic variables. RESULTS: The prevalence of insomnia was 24% (range 13–30%), with significant differences between countries (P < .001). More insomnia complaints were reported in older than younger residents (P < .001). Higher rates of insomnia were associated with hypnosedatives and depression in all countries (P < .001) and with stressful life events, fatigue, and pain in most countries (P < .001). No associations were found between insomnia and ADLs, physical activity, or cognitive status. Age, depression, stressful life events, fatigue, pain and hypnosedatives were independent significant predictors of insomnia, controlling for all other variables and for country. CONCLUSION: Hypnosedatives and depression were strong predictors of insomnia beyond cultural differences. Overall, psychosocial variables were more strongly related to insomnia than functional and mental capacities. J Am Geriatr Soc 62:2033–2039, 2014.
Key words: Insomnia; elderly; long-term care facility; InterRAI; hypnosedatives
nsomnia is common in the general elderly population.1–4 It has been associated with medical and psychiatric conditions3,5–7 and with impaired daily functioning,3,8,9 cognitive decline,10–13 poor health-related quality of life,7,14 risk of falls and hip fractures,15,16 and all-cause mortality.17,18 Furthermore, the use of hypnotic medications for the treatment of insomnia is most prevalent and most chronic in elderly adults19–21 and has been associated with depression and traumatic life events,20 poor functional status,20,22 and risk of injury.23,24 Although sleep disturbances are highly prevalent in nursing homes,25 few studies have specifically investigated
I
0002-8614/14/$15.00
2034
GINDIN ET AL.
insomnia and its correlates in long-term care facilities (LTCFs).15,26,27 Thus, in a Canadian study of more than 2,000 LTCF residents, only 6% had insomnia based on Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) criteria, although 17% had at least one insomnia symptom, yet more than half used benzodiazepines.26 Insomnia was associated with disruptive behaviors and psychological distress. In a study of more than 2,500 LTCF residents in Germany using the Minimum Data Set (MDS) 2.0, the prevalence of insomnia based on the two items referring to insomnia was 37% for difficulty falling and staying asleep and nearly 30% for nonrestful sleep or tired in the morning.27 Insomnia was associated with low levels of activities and social engagement and high levels of communication difficulties and interpersonal conflicts. In a 6-month longitudinal U.S. study of more than 34,000 LTCF residents using the MDS, insomnia was associated with greater risk for falls (odds ratio (OR) = 1.52, 95% confidence interval (CI) = 1.38–1.66) after adjusting for several sociodemographic and clinical confounders.15 The International Resident Assessment Instrument (InterRAI) for LTCFs is one component of a suite of instruments designed to provide ongoing standard assessment for a wide variety of healthcare settings and has been described elsewhere in detail.28 Validity and reliability of the instrument have been established previously29 and more recently by the Services and Health for Elderly in Long TERm care (SHELTER) study as part of the 7th Framework Programme of the European Union.28 The SHELTER study included three waves of longitudinal data from 4,156 nursing home residents from seven European countries and one non-European (Israel) country. Several publications based on the SHELTER database have demonstrated the effectiveness of the instrument for evaluation and comparison of various health concerns, such as polypharmacy,30 inappropriate drug use,31 pain and associated psychiatric symptoms,32 pneumococcal vaccine coverage,33 and depression34 in nursing homes. The aims of this investigation, which was based on the InterRAI LTCF evaluation baseline intake in the SHELTER study, were to evaluate the prevalence of insomnia in LTCFs of seven European countries and Israel, controlling for age, sex, and country, and to investigate the associations between insomnia and measures of depression, stress, fatigue, pain, functional status, cognitive status, physical activity, and hypnosedative medication use.
METHODS Participants Participants were elderly individuals (mean age 83.4 9.4; N = 4,156) residing in nursing homes from eight countries (Czech Republic, France, Finland, Germany, England, the Netherlands, Italy, and Israel) who agreed to participate in the project, as previously reported.28
NOVEMBER 2014–VOL. 62, NO. 11
JAGS
Primary Outcome Insomnia was defined as difficulty falling asleep or staying asleep, waking up too early, restlessness, or nonrestful sleep.
Predicting Variables Three summary InterRAI scales were used to evaluate functional status, cognitive status, and presence of depressive symptoms (for more information on scales, see28). Assessment of predicting variables included functional status (based on four categories of activities of daily living (ADLs): 0 = no impairment, 1–2 = limited impairment, 3–4 = extensive impairment, 5–6 = dependent), cognitive impairment (based on the cognitive performance scale (CPS): 0 = no impairment, 2–4 = moderate impairment, 5–6 = severe impairment), presence of depressive symptoms (based on the Depression Rating Scale (DRS): 0–2 = not depressed, >3 = depressed), major life stress (yes, no), physical activity (based on total hours of exercise per day: 0,
