Sindi et al. BMC Medicine (2020) 18:382 https://doi.org/10.1186/s12916-020-01846-w RESEARCH ARTICLE Open Access Sleep disturbances and the speed of multimorbidity development in old age: results from a longitudinal population- based study Shireen Sindi1,2*†, Laura Monica Pérez3,4†, Davide L. Vetrano3,5, Federico Triolo3, Ingemar Kåreholt3,6, Linnea Sjöberg3, Alexander Darin-Mattsson3, Miia Kivipelto1,2,7,8,9, Marco Inzitari4,10† and Amaia Calderón-Larrañaga3† Abstract Background: Sleep disturbances are prevalent among older adults and are associated with various individual diseases. The aim of this study was to investigate whether sleep disturbances are associated with the speed of multimorbidity development among older adults. Methods: Data were gathered from the Swedish National study of Aging and Care in Kungsholmen (SNAC-K), an ongoing population-based study of subjects aged 60+ (N = 3363). The study included a subsample (n = 1189) without multimorbidity at baseline (< 2 chronic diseases). Baseline sleep disturbances were derived from the Comprehensive Psychiatric Rating Scale and categorized as none, mild, and moderate–severe. The number of chronic conditions throughout the 9-year follow-up was obtained from clinical examinations. Linear mixed models were used to study the association between sleep disturbances and the speed of chronic disease accumulation, adjusting for sex, age, education, physical activity, smoking, alcohol consumption, depression, pain, and psychotropic drug use. We repeated the analyses including only cardiovascular, neuropsychiatric, or musculoskeletal diseases as the outcome. Results: Moderate–severe sleep disturbances were associated with a higher speed of chronic disease accumulation (ß/year = 0.142, p = 0.008), regardless of potential confounders. Significant positive associations were also found between moderate–severe sleep disturbances and neuropsychiatric (ß/year = 0.041, p = 0.016) and musculoskeletal (ß/year = 0.038, p = 0.025) disease accumulation, but not with cardiovascular diseases. Results remained stable when participants with baseline dementia, cognitive impairment, or depression were excluded. (Continued on next page) * Correspondence: shireen.sindi@ki.se † Shireen Sindi and Laura M Pérez contributed equally as first authors. † Marco Inzitari and Amaia Calderón-Larrañaga contributed equally as last authors. 1 Division of Clinical Geriatrics, Center for Alzheimer Research, Karolinska Institutet, Stockholm, Sweden 2 Neuroepidemiology and Ageing Research Unit, School of Public Health, Imperial College London, London, UK Full list of author information is available at the end of the article © The Author(s). 2020 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Sindi et al. BMC Medicine (2020) 18:382 Page 2 of 10 (Continued from previous page) Conclusion: The finding that sleep disturbances are associated with faster chronic disease accumulation points towards the importance of early detection and treatment of sleep disturbances as a possible strategy to reduce chronic multimorbidity among older adults. Keywords: Sleep disturbances, Multimorbidity, Aging, Cardiovascular, Neuropsychiatric, Musculoskeletal Background Findings from a US middle-aged sample showed that Sleep disturbances are a “public health problem,” as a among several health-promoting behaviors, only sleep third of adults report having insufficient sleep [1, 2], and and physical activity were associated with multimor- half of older adults report some type of sleep disturbance bidity [26]. Sex differences have also been demon- [3]. While the prevalence of diagnosed insomnia dis- strated, including differences in the relationship order ranges from 5 to 8%, 20–40% report insomnia, between sleep disturbances and commonly co- and up to 50% complain of sleep problems [4, 5]. It has occurring chronic conditions [27]. The association be- been estimated that the mean total healthcare costs were tween sleep disturbances and multimorbidity may be 75% higher for groups with moderate and severe insom- bidirectional. For example, one study showed that the nia compared to those with no insomnia, of a health number of physical conditions at baseline was associ- plan sample in the USA [6]. Similarly, those with un- ated with incident insomnia two years later [28]. treated insomnia have higher all-cause healthcare The aim of this study was to investigate the association utilization [7], providing evidence for the economic bur- between sleep disturbances and the speed of multimor- dens of (undiagnosed/untreated) sleep disturbances. bidity development, a proxy for loss of resilience and Chronic sleep disturbances have a range of negative multisystem homeostatic dysregulation in older adults health consequences, including hypertension, cardiovas- [29, 30]. cular diseases, stroke, obesity, diabetes, migraines, and mortality [8, 9]. Sleep disturbances are also associated Methods with an increased risk for dementia [10, 11], cognitive Study design and population impairment [12–14], and depressive symptoms [15, 16]. The study population was derived from the Swedish Another important consequence of sleep disturbances is National study of Aging and Care in Kungsholmen daytime sleepiness/drowsiness, which is associated with (SNAC-K), an ongoing population-based study, which falls, death, and accidents [17, 18]. In a National Sleep enrolled adults aged 60 years and older living in the Foundation survey, the frequency of sleep problem re- community or in institutions in the Kungsholmen dis- ports among older adults was 53% among those with no trict of Stockholm between 2001 and 2004 [31]. A total medical condition and escalated up to 80% among those of 4590 people from 11 age cohorts (60, 66, 72, 78, 81, with 4+ medical conditions, demonstrating the need for 84, 87, 90, 93, 96, and ≥ 99 years) were randomly invited further investigations [18, 19]. Similarly, insomnia was to participate in the study. Finally, 3363 accepted to par- associated with multiple medical and psychiatric comor- ticipate in the baseline examination (73.3% participation bidities in a large analysis of Medicare beneficiaries rate). Follow-up assessments are performed every (USA) [7]. six years for younger participants (60–78 years) and To understand the broad implications of chronic sleep every three years for older participants (≥ 78 years). In disturbances, it is important to simultaneously investi- the current study, we included a subsample of 1189 par- gate the associations between sleep disturbances and ticipants without multimorbidity at baseline (i.e., < 2 multiple health outcomes. Multimorbidity refers to the chronic disease) and with data for the main exposure. coexistence of two or more chronic medical conditions SNAC-K participants undergo a comprehensive clin- in an individual [20]. Although multimorbidity is the ical and functional assessment by trained personnel at norm among older adults [21, 22], there are inter- baseline and follow-ups. Data on past medical history individual differences in the speed of multimorbidity de- and vital status were also available through the National velopment and progression. Patient Register and the Swedish Death Register. Cross-sectional studies showed that short sleep dur- ation, feelings of drowsiness, and difficulties sleeping were associated with multimorbidity [23, 24]. In a Ethics, consent, and permissions multi-national cross-sectional study, a linear dose- SNAC-K was approved by the Regional Ethical Review dependent association was found between the number Board in Stockholm in Sweden. Written informed con- of chronic conditions and sleep disturbances [25]. sent was collected from all participants or, if the Sindi et al. BMC Medicine (2020) 18:382 Page 3 of 10 participant was not capable of providing such consent, it in the last month before baseline was also considered. was obtained from a proxy. Depressive symptoms were assessed through the Mont- gomery–Åsberg Depression Scale (MADRS) [36], a 10- Multimorbidity assessment item subscale driven from the CPRS rating depressive The total number of chronic conditions at baseline and symptom intensity, frequency, and duration. The MADR at the follow-ups was operationalized according to the S scores range between 0 and 60, and scores > 9 reflect comprehensive list proposed by Calderon-Larrañaga the presence of depression [37]. et al., described elsewhere [32]. All the diagnoses were coded according to the International Classification of Statistical analyses Diseases, 10th revision (ICD-10) and then classified as The cross-sectionaly associations between baseline chronic or not chronic; then, those classified as chronic sociodemographic, clinical and lifestyle characteristics were grouped into 60 broader disease categories. Clinical and sleep disturbances were examined using the chi- information, laboratory test results, current treatment, square test for proportions and the Kruskal–Wallis test and data from primary care and hospital medical records for continuous variables. were also used to identify specific chronic conditions. The longitudinal associations between sleep distur- Chronic conditions considered cardiovascular risk fac- bances at baseline and the speed of chronic disease accu- tors (i.e., hypertension, dyslipidemia, and obesity) were mulation throughout the 9-year follow-up were analyzed excluded from the total count consistent with previous through linear mixed models. We included the inter- studies [33]. Sleep disorders (ICD-10 codes F510-13 and action term between time and the main exposure as a G47) were also excluded to avoid exposure–outcome fixed effect. The β coefficients for such interactions can circularity. be interpreted as the effect of each level of sleep disturb- ance on the average annual increase in the number of Sleep disturbances chronic diseases [29, 30]. Random effects were defined The presence of sleep disturbances at baseline was for the intercept and slope, unstructured covariance was assessed by physicians through the Comprehensive assumed, and restricted maximum likelihood estimation Psychiatric Rating Scale (CPRS), a scale rating the sever- was applied. Models were adjusted for all covariates in a ity of psychiatric symptoms and observed behavior and cumulative manner: (1) sex, age, and education; (2) change in symptoms over time [34]. The degree of sleep additionally by physical activity, smoking, and alcohol disturbances was categorized into three groups accord- consumption; and (3) additionally by the presence of de- ing to participants’ answers to the CPRS items on sleep: pression (MADRS score > 9), presence of any pain, use no sleep disturbances, mild sleep disturbances (grouping of psychotropic medication, and presence of one chronic the categories 1–2 “Slight difficulty dropping off to sleep disease at baseline. Three-way interactions between age, or slightly reduced, light or fitful sleep”), and moderate– sex, or psychotropic drug use and sleep disturbances × severe sleep disturbances (grouping the categories 3–4 time were tested. To investigate the role of sleep distur- “Sleep reduced or broken by at least two hours” and 5–6 bances on the development of specific clusters of “Less than two or three hours of sleep”). chronic diseases [38], we repeated the analyses including only cardiovascular, neuropsychiatric, and musculoskel- Covariates etal diseases as the outcome in three different models. Education was measured as the highest level of formal Last, we explored if specific types of sleep disturbances education and was categorized as elementary school, were associated with multimorbidity development, either high school, and university or above. Physical activity among participants reporting any type of sleep distur- was categorized into three groups of intensity: inad- bances or in the whole study sample. equate, health-enhancing, and fitness-enhancing [35]. Smoking habit was categorized as never, former, or Sensitivity analyses current smoker, and alcohol consumption as never, To evaluate if the associations were driven by any light/moderate, and heavy consumption. Body mass specific chronic conditions either at baseline or at index (BMI) was calculated by dividing weight in follow-ups, we first compared the distribution of the kilograms by the square of height in meters and then most prevalent cardiovascular, neuropsychiatric, and categorized as underweight (< 18.5 kg/m2), normal musculoskeletal chronic diseases by the presence and se- weight (18.5–24.9 kg/m2, reference category), overweight verity of sleep disturbances at baseline using the chi- (25–29.9 kg/m2), or obese (≥ 30 kg/m2). We created a square test. Second, we repeated the analyses concerning variable identifying people taking psychotropic drugs cardiovascular, neuropsychiatric, and musculoskeletal (i.e., anxiolytics, hypnotics, sedatives, antipsychotics, and diseases after removing the three most common chronic antidepressants). The self-reported presence of any pain conditions from each group of diseases, one at a time. Sindi et al. BMC Medicine (2020) 18:382 Page 4 of 10 Third, we repeated the analyses excluding those partici- osteoarthritis and other degenerative joint diseases were pants with a diagnosis of dementia or a Mini-Mental removed, the association was no longer significant. Re- State Examination (MMSE) score < 24 at baseline (n = sults remained the same when participants with a base- 13), to avoid exposure measurement error in these par- line diagnosis of dementia or MMSE < 24 were excluded ticipants. We also repeated the analyses excluding par- from the study sample (Supplementary Table 5). When ticipants with a diagnosis of depression or a MADRS participants with depression and mood diseases or a score > 9 at baseline (n = 43). All analyses were per- MADRS score > 9 were excluded, the direction of the as- formed using Stata v.14. sociation between moderate–severe sleep disturbances and the rate of chronic disease accumulation remained Results the same, even if it did not reach statistical significance The mean age of participants was 67.5 years (SD = 7.9) (Supplementary Table 6). and 58% were female (Supplementary Table 1). Those with missing exposure data (n = 28) were more likely to Discussion be widow and have an elementary education compared This study sought to investigate the association between with the study population. Participants with moderate– sleep disturbances and the speed of multimorbidity severe sleep disturbances were more likely to take development in older adults. Results showed that mod- psychotropic drugs, suffer from depressive symptoms, erate–severe sleep disturbances at baseline were inde- and self-report pain compared to those with no or mild pendently associated with a higher speed of chronic disturbances (Table 1). There was also a higher propor- disease accumulation during the 9-year follow-up. With tion of women among participants with moderate–se- regard to specific groups of chronic diseases, a signifi- vere disturbances, and this group had higher levels of cant positive association was found between moderate– inadequate physical activity (Table 1). Among the popu- severe sleep disturbances and neuropsychiatric and lation with sleep disturbances, the three most prevalent musculoskeletal disease accumulation, but not with types of disturbances were waking up during the night cardiovascular diseases. To the best of our knowledge, (86.5%), waking up early (51.6%), and problems to fall this is the first study to longitudinally investigate sleep asleep (39.1%) (Table 2). and speed of multimorbidity development after a long Moderate–severe sleep disturbances at baseline were follow-up duration in a large population-based study. associated with a higher speed of chronic disease accu- Considering that this is the first study to show such mulation throughout the 9-year follow-up, and the associations, direct comparisons cannot be made with direction and magnitude of the association remained previous literature. stable regardless of potential confounders (Table 3). No Our finding on the association between sleep distur- significant interactions were found with age, sex, or bances and accumulation of neuropsychiatric diseases psychotropic drug use. When focusing on specific supports previous cross-sectional results from nationally groups of chronic diseases, a significant positive associ- representative samples, where of several conditions, de- ation was found between moderate–severe sleep distur- pression was the only one consistently associated with bances and neuropsychiatric and musculoskeletal disease sleep disturbances [25]. In a Korean study, baseline in- accumulation, but not with cardiovascular diseases somnia was associated with incident depression after (Table 4, Fig. 1). two years [28]. In our study, even when depression was When focusing on specific types of sleep disturbances, removed from the analyses, the associations remained “feeling tired more than 2 hours during the day” was robust. This may highlight the additional strong associa- consistently associated with the rate of total, neuro- tions between sleep disturbances and dementia and mi- psychiatric, and musculoskeletal disease accumulation graine and facial pain syndromes. Our findings are (Supplementary Tables 2A and B). consistent with studies showing that sleep disturbances Except for ischemic heart disease and depression and are associated with dementia risk [10, 11]. This study mood diseases (borderline statistical significance), no additionally supports a recent meta-analysis showing other differences were found in the baseline distribution that obstructive sleep apnea is associated with cerebro- of the most prevalent cardiovascular, neuropsychiatric, vascular diseases [39]. The current findings provide fur- and musculoskeletal chronic diseases by presence and ther evidence regarding the association between sleep severity of sleep disturbances (Supplementary Table 3). disturbances and incidence of neuropsychiatric condi- When the analyses in Table 4 were repeated after re- tions, but also a faster accumulation of neuropsychiatric moving one condition at a time (only the individual con- conditions. dition specified) from each group of chronic diseases Regarding our finding on the association between (Supplementary Table 4), similar findings were obtained. sleep disturbances and speed of musculoskeletal disease The one exception was musculoskeletal disease; when accumulation, when osteoarthritis and other Sindi et al. BMC Medicine (2020) 18:382 Page 5 of 10 Table 1 Baseline distribution of sociodemographic, clinical, and lifestyle characteristics by presence and severity of sleep disturbances No disturbances, n = 955 Mild disturbances¥, n = 175 Moderate–severe disturbances§, n = 59 p value* Age, mean (SD) 67.6 (7.8) 67.9 (8.3) 65.9 (8.1) 0.504 Sex, % (n) 0.032 Female 56.3 (538) 64.0 (112) 69.5 (41) Male 43.7 (417) 36.0 (63) 30.5 (18) Education, % (n) 0.332 Elementary 10.4 (99) 8.0 (14) 17.0 (10) High school 46.1 (439) 44.6 (78) 39.0 (23) University 43.5 (414) 47.4 (83) 44.0 (26) Physical activity, % (n) 0.001 Inadequate 18.2 (174) 16.0 (28) 33.9 (20) Health-enhancing 50.3 (480) 57.7 (101) 54.2 (32) Fitness-enhancing 31.5 (301) 26.3 (46) 11.9 (7) Smoking, % (n) 0.214 Never 43.3 (410) 43.7 (76) 42.4 (25) Former 37.7 (358) 43.1 (75) 45.8 (27) Current 19.0 (180) 13.2 (23) 11.8 (7) Alcohol, % (n) 0.097 Never 20.6 (195) 20.6 (36) 33.8 (20) Light–moderate 61.8 (589) 57.7 (101) 49.2 (29) Heavy 17.6 (169) 21.6 (38) 17.0 (10) BMI, % (n) 0.380 Underweight 1.0 (9) 0.0 (0) 1.7 (1) Normal weight 45.3 (426) 47.7 (83) 39.0 (23) Overweight 42.3 (398) 37.3 (65) 42.3 (25) Obesity 11.4 (107) 15.0 (26) 17.0 (10) Psychotropic drugsa, % (n) 7.1 (68) 22.3 (39) 25.4 (15) < 0.001 b Depression , % (n) 1.1 (10) 5.8 (10) 14.3 (8) < 0.001 Painc, % (n) 22.8 (217) 34.5 (60) 47.5 (28) < 0.001 *Obtained through the chi-squared test or the Kruskal–Wallis test, as appropriate a Psychotropic drugs include anxiolytics, hypnotics, sedatives, antipsychotics, and antidepressants b Depression defined according to the Montgomery–Åsberg Depression Rating Scale (MADRS); scores > 9 indicate depression c Self-reported presence of any pain in the last month ¥ Categories 1–2 “Slight difficulty dropping off to sleep or slightly reduced, light or fitful sleep” of the CPRS question on sleeping problems § Categories 3–4 “Sleep reduced or broken by at least two hours” and 5–6 “Less than two or three hours of sleep” of the CPRS question on sleeping problems Table 2 Type of disturbances among subjects reporting sleep disturbances in the study sample Total, < 78 years ≥ 78 years n = 234 Male, n = 70 Female, n = 123 Male, n = 11 Female, n = 30 Problems to fall asleep, % (n) 39.1 (90) 30.9 (21) 42.5 (51) 45.5 (5) 43.3 (13) Waking up during night, % (n) 86.5 (199) 89.7 (61) 82.6 (100) 90.9 (10) 93.3 (28) Not being able to fall back asleep, % (n) 38.4 (84) 27.7 (18) 43.4 (49) 27.3 (3) 46.7 (14) Waking up too early, % (n) 51.6 (116) 51.5 (34) 52.1 (62) 45.5 (5) 51.7 (15) Feeling tired > 2 h during the day, % (n) 4.4 (10) 6.0 (4) 4.1 (5) 9.1 (1) 100.0 (30) Taking sleeping drugs, % (n) 28.6 (65) 16.4 (11) 30.0 (36) 27.3 (3) 50.0 (15) Total sleep duration < 6 h*, % (n) 27.1 (61) 27.3 (18) 25.0 (30) 27.3 (3) 35.7 (10) *Defined as short sleep duration by Gildner et al. [9] Sindi et al. BMC Medicine (2020) 18:382 Page 6 of 10 Table 3 Association between sleep disturbances at baseline and rate of chronic diseases accumulation throughout the 9-year follow-up Model 1 Model 2 Model 3 ß (SE) p value ß (SE) p value ß (SE) p value No disturbances Ref Ref Ref Mild disturbances¥ 0.018 (0.032) 0.574 0.019 (0.032) 0.539 0.022 (0.032) 0.496 § Moderate–severe disturbances 0.145 (0.053) 0.006 0.147 (0.053) 0.005 0.142 (0.054) 0.008 ß coefficients for the interaction term between time and the exposure, obtained through linear mixed models. Model 1: adjusted by sex, age, and education level Model 2: adjusted additionally by physical activity, smoking, alcohol consumption, and BMI Model 3: adjusted additionally by the presence of depression (MADRS score > 9), presence of any pain, psychotropic medication, and presence of one chronic disease at baseline ¥ Categories 1–2 “Slight difficulty dropping off to sleep or slightly reduced, light or fitful sleep” of the CPRS question on sleeping problems § Categories 3–4 “Sleep reduced or broken by at least two hours” and 5–6 “Less than two or three hours of sleep” of the CPRS question on sleeping problems degenerative joint diseases were removed from the ana- proposed as central mechanisms in these associations, lyses, the associations were no longer significant, sug- and fatigue may also play a role [41, 42]. gesting a central role of these conditions. A recent In the current study, although the lack of association be- systematic review and meta-analysis has shown that hav- tween sleep disturbances and cardiovascular disease accu- ing osteoarthritis was associated with the presence of mulation may appear to contradict previous findings, other chronic conditions and multimorbidity in a dose- comparisons are in fact non-feasible, considering the dependent manner [40]. Our results add to this litera- cross-sectional study designs, differences in age groups, ture by highlighting the role of osteoarthritis and degen- definitions of multimorbidity, sleep disturbance measure- erative joint disease in the associations between sleep ments, and non-included potential confounders (e.g., disturbances and musculoskeletal disease accumulation. BMI, hypnotics, education, smoking, alcohol consump- These findings are also consistent with cross-sectional tion, exercise) in the previous studies reporting positive results from Brazil demonstrating a U-shaped associ- associations [9, 23, 25, 27]. These methodological differ- ation; both short sleep and long sleep duration were as- ences may contribute to the observed differences. More- sociated with rheumatism/arthritis/arthrosis and over, in our study, the non-significant association between osteoporosis [23]. Short sleep duration was also associ- sleep disturbances and cardiovascular disease may reflect ated with back pain problems. Similarly, in the afore- a selection bias; in this sample of older adults, those with mentioned multi-national study, sleep disturbances were cardiovascular diseases may have not participated in this associated with arthritis [25]. Another cross-sectional study or its follow-up assessments due to earlier adverse study among older adults in Germany showed that in- outcomes (e.g., mortality and disability). Also, participants somnia and daytime tiredness were associated with joint with cardiovascular diseases tend to have other coexisting diseases [27]. Increased pain and inflammation are chronic conditions [43] and may have been excluded from Table 4 Association between sleep disturbances at baseline and rate of cardiovascular (CV), neuropsychiatric (NP), and musculoskeletal (MSK) chronic disease accumulation throughout the 9-year follow-up CV diseases NP diseases MSK diseases ß (SE) p value ß (SE) p value ß (SE) p value No disturbances Ref Ref Ref Mild disturbances¥ − 0.012 (0.011) 0.267 0.008 (0.009) 0.419 − 0.003 (0.010) 0.735 § Moderate–severe disturbances 0.001 (0.018) 0.995 0.041 (0.016) 0.008 0.038 (0.017) 0.025 ß coefficients for the interaction term between time and the exposure, obtained through linear mixed models. Models adjusted by sex, age, education level, physical activity, smoking, alcohol consumption, BMI, presence of depression (MADRS score > 9) except for the model with NP diseases as the outcome, presence of pain, psychotropic medication, and presence of one chronic disease at baseline. Cardiovascular diseases: ischemic heart disease, heart failure, atrial fibrillation, cerebrovascular disease, cardiac valve diseases, bradycardias or conduction diseases, peripheral vascular disease, and other cardiovascular diseases. Neuropsychiatric diseases: depression and mood diseases, dementia, neurotic or stress-related and somatoform diseases, migraine and facial pain syndromes, peripheral neuropathy, Parkinson or parkinsonism, epilepsy, schizophrenia and delusional diseases, multiple sclerosis, other psychiatric or behavioral diseases, and other neurological diseases. Musculoskeletal diseases: dorsopathies, inflammatory arthropathies, osteoarthritis and other degenerative joint diseases, osteoporosis, and other musculoskeletal and joint diseases ¥ Categories 1–2 “Slight difficulty dropping off to sleep or slightly reduced, light or fitful sleep” of the CPRS question on sleeping problems § Categories 3–4 “Sleep reduced or broken by at least two hours” and 5–6 “Less than two or three hours of sleep” of the CPRS question on sleeping problems Sindi et al. BMC Medicine (2020) 18:382 Page 7 of 10 1.4 No disturbances 1.4 No disturbances Slight disturbances Slight disturbances 1.2 Moderate-severe disturbances 1.2 Moderate-severe disturbances Number of MSK diseases Number of NP diseases 1.0 1.0 0.8 0.8 0.6 0.6 0.4 0.4 0.2 0.2 0.0 0.0 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 Follow-up time Follow-up time Fig. 1 Predicted number of neuropsychiatric (NP) and musculoskeletal (MSK) chronic diseases over the 9-year follow-up in relation to the presence and severity of sleep disturbances at baseline ß coefficients for the interaction term between time and the exposure, obtained through linear mixed models. Models adjusted by sex, age, education level, physical activity, smoking, alcohol consumption, BMI, presence of depression (MADRS score > 9) except for the model with NP diseases as the outcome, presence of pain, psychotropic medication, and presence of any chronic disease. Neuropsychiatric diseases: depression and mood diseases, dementia, neurotic or stress-related and somatoform diseases, migraine and facial pain syndromes, peripheral neuropathy, Parkinson or parkinsonism, epilepsy, schizophrenia and delusional diseases, multiple sclerosis, other psychiatric or behavioral diseases, and other neurological diseases. Musculoskeletal diseases: dorsopathies, inflammatory arthropathies, osteoarthritis and other degenerative joint diseases, osteoporosis, and other musculoskeletal and joint diseases. our study sample, which may have prevented us from neurotransmitter that plays a role in regulating sleep-wake finding such an association. states, and also regulates mood and tends to be depleted The results showed that there were no significant in- in depression, representing an important component teractions between sleep disturbances and age or sex. A in the pathophysiology of the disorder [55–57]. An- previous study did find sex-specific associations; insom- other potential mechanism may be through the nia and specific insomnia symptoms were associated neuropeptide orexin, which regulates sleep-wake cy- with multimorbidity among women, but not among men cles and the circadian rhythm and plays a role in cog- [27]. This study differed from the current one as it was nitive function and the cardiovascular system [58]. cross-sectional and it did not adjust for depression, Orexin receptor antagonists have been investigated chronic diseases, and pain. Also, the sample was slightly for their potential therapeutic effects for the treat- older in age with lower education levels. Considering the ment of insomnia [59]. Sleep disturbances are also as- previously reported sex differences in sleep, this topic sociated with compromises in metabolic health and warrants further investigation among older adults [44]. can increase the risk for obesity, dyslipidemia, hyper- Several mechanisms may underlie the associations ob- tension, and impaired glucose metabolism, all of served. First, an important mechanism underlying neuro- which increase the risk for cardiovascular conditions psychiatric and cardiovascular conditions is dysregulations [60]. Lifestyle factors, beyond those taken into ac- in the hypothalamic–pituitary–adrenal (HPA) axis [45, 46], count in our models (e.g., changes in appetite, diet, which are also implicated in sleep disturbances. Poor sleep and polypharmacy), which also increase cardiometa- quality enhances the reactivity of the HPA axis and is asso- bolic risk and degrade sleep quality [60–62]. ciated with higher daytime cortisol levels [47, 48], and al- This study has a few limitations. The measure of sleep tered cortisol awakening response [49]. The relationship was self-reported and measured at a single time point. between sleep disturbances and HPA axis dysregulations Objective measures of chronic sleep disturbances may may be bidirectional, as HPA axis hyperactivation is also as- provide more reliable data. While the current study in- sociated with poor sleep [50]. Second, sleep disturbances vestigates sleep in late life and its association with multi- are connected with dysregulations in inflammatory morbidity, evidence from other studies highlights the markers. For example, sleep disturbances are associated importance of midlife sleep and adverse outcomes [11, with higher levels of C-reactive protein (CRP) and 14, 63]. Regarding the feelings of tiredness during the interleukin-6 (IL-6) [51], which are also commonly elevated day, we could not examine the source of tiredness, as in musculoskeletal and neuropsychiatric diseases [52, 53], this additional information was not collected. We also and among people with a fast accumulation of chronic dis- did not have clinical measures of sleep apnea, which is eases [54]. These mechanisms may also impact serotonin, a associated with multimorbidity [64, 65], and some Sindi et al. BMC Medicine (2020) 18:382 Page 8 of 10 parameters such as sleep efficiency and sleep latency are Supplementary Information not included. The external validity of the study is limited The online version contains supplementary material available at https://doi. org/10.1186/s12916-020-01846-w. because the sample only included healthy participants from SNAC-K population, which is wealthier than average Additional file 1: Supplementary Table 1. Baseline in Sweden, and may have better lifestyle and fewer vascu- sociodemographic, clinical and lifestyle characteristics of the study lar risk factors (e.g., obesity). Attrition due to mortality population by age and sex. Supplementary Table 2. Association between type of sleep disturbance at baseline and rate of total and and dropout might have affected the results. Although we specific groups of chronic disease accumulation throughout the nine- excluded subjects with multimorbidity at baseline, and ad- year follow-up. Supplementary Table 3. Baseline distribution of most justed our models for several confounders, the possibility prevalent cardiovasculara (CV), neuropsychiatricb (NP) and musculoskele- talc (MSK) chronic diseases by presence and severity of sleep distur- of reverse causality or residual confounding cannot be bances. Supplementary Table 4. Association between presence and fully discarded given the potential heterogeneity in sub- severity of sleep disturbances at baseline and rate of cardiovascular (CV), clinical health states leading to multimorbidity. Finally, neuropsychiatric (NP) and musculoskeletal (MSK) chronic disease accumu- lation throughout the nine-year follow-up. Analyses performed after re- sleep disturbances may in fact represent the symptoms of moving most prevalent chronic conditions, one at a time, from each already established disorders (e.g., depression, arthritis). group of chronic diseases. Supplementary Table 5. Association be- These could be the single disorders that people with only tween presence and severity of sleep disturbances at baseline and rate of total and specific groups of chronic disease accumulation throughout the one disease suffer from (who were not excluded from the nine-year follow-up. Analyses excluding participants with dementia or study sample) or disorders that were undiagnosed yet at cognitive impairment at baseline. Supplementary Table 6. Association baseline. In both cases, this could potentially lead to re- between presence and severity of sleep disturbances at baseline and rate of total and specific groups of chronic disease accumulation throughout verse causality regarding the associations between sleep the nine-year follow-up. Analyses excluding participants with depression disturbances and multimorbidity. at baseline*. Despite the limitations, this study has several strengths. First, this study is population-based, longitu- Acknowledgements dinal, with baseline measures of sleep disturbances in The authors would like to thank all the SNAC-K participants as well as the SNAC-K study team for their time and their valuable contributions. old age, and a 9-year follow-up. Second, chronic condi- tions were either objectively measured by physicians Authors’ contributions within the SNAC-K study or retrieved in medical health Study concept and design: S. Sindi, L.M. Pérez, M. Inzitari, and A. Calderón- records, including a wide range of diseases. Third, this Larrañaga. Drafting of the manuscript: S. Sindi and L.M. Pérez. Data interpretation and critical revision of the manuscript: S. Sindi, L.M. Pérez, D.L. study provides data on the speed of multimorbidity ac- Vetrano, F. Triolo, I. Kåreholt, L. Sjöberg, A. Darin-Mattsson, M. Kivipelto, M. cumulation, a proxy for the speed of biological aging Inzitari, and A. Calderón-Larrañaga. Statistical analysis: L.M. Pérez and A. Cal- [66]. Fourth, individuals with multimorbidity at baseline derón-Larrañaga. Obtained funding: S. Sindi, M. Kivipelto, and A. Calderón- Larrañaga. Study supervision: M. Kivipelto, M. Inzitari, and A. Calderón-Larra- were excluded from analyses, which limits the risk of re- ñaga. All authors read and approved the final manuscript. verse causality. Fifth, sleep was measured using a vali- dated scale including several parameters of sleep Funding This work was supported by the funders of The Swedish National study of disturbances. Finally, we have adjusted for potential con- Aging and Care (SNAC): the Ministry of Health and Social Affairs, Sweden; the founders including medications. participating county councils and municipalities; and the Swedish Research Sleep disturbances and circadian rhythm disruptions Council. Shireen Sindi receives support from Alzheimerfonden, Demensförbundet, Karolinska Institute Foundation and Funds (KI Stiftelser may be a proxy of biological vulnerability or early disease och Fonder) and Loo and Hans Osterman Foundation for Medical Research. stage [67] and therefore may help identify individuals with Amaia Calderón-Larrañaga received grants from the Swedish Research potentially worse multimorbidity trajectories. Sleep distur- Council (2016-00981) and the Swedish Research Council for Health, Working Life and Welfare (2017-01764). Miia Kivipelto receives research support from bances tend to be chronic, and interventions need to tar- the Alzheimer’s Research & Prevention Foundation, Academy of Finland get a comprehensive range of behavioral/lifestyle factors (SALVE and 278457, 305810, 317465), Finnish Social Insurance Institution, that impact sleep. Evidence from randomized controlled Finnish Ministry of Education and Culture, Juho Vainio Foundation (Finland), Joint Programme - Neurodegenerative Disease Research (MIND-AD and trials offers support for cognitive behavioral therapy for EURO-FINGERS), Alzheimerfonden (Sweden), Swedish Research Council, Cen- insomnia (CBT-I) as an effective and comprehensive non- ter for Innovative Medicine (CIMED) at Karolinska Institutet, Region Stockholm pharmacological approach [61, 62, 68]. Considering the (ALF, NSV), AXA Research Fund, Knut and Alice Wallenberg Foundation (Sweden), Stiftelsen Stockholms sjukhem (Sweden), Konung Gustaf V:s och potential side effects of pharmacological treatments, they Drottning Victorias Frimurarstiftelse (Sweden), and Swedish Research Council are recommended as a last resort [61]. for Health, Working Life and Welfare (FORTE). Open Access funding provided by Karolinska Institute. Availability of data and materials Conclusion Data are from the SNAC-K project, a population-based study on aging and In conclusion, for the first time, this study showed that dementia (http://www.snac-k.se/). Access to these original data is available to moderate–severe sleep disturbances were associated the research community upon approval by the SNAC-K data management and maintenance committee. Applications for accessing these data can be with a higher speed of overall, neuropsychiatric and submitted to Maria Wahlberg (Maria.Wahlberg@ki.se) at the Aging Research musculoskeletal chronic disease accumulation over time. Center, Karolinska Institutet, Stockholm, Sweden. Sindi et al. BMC Medicine (2020) 18:382 Page 9 of 10 Ethics approval and consent to participate 15. Sun Y, Shi L, Bao Y, Sun Y, Shi J, Lu L. The bidirectional relationship SNAC-K was approved by the Regional Ethical Review Board in Stockholm in between sleep duration and depression in community-dwelling middle- Sweden. Written informed consent was collected from all participants. If the aged and elderly individuals: evidence from a longitudinal study. Sleep participant was not capable of providing such consent, it was obtained from Med. 2018;52:221–9. a close relative or, eventually, from the legal guardian. 16. Zhai L, Zhang H, Zhang D. Sleep duration and depression among adults: a meta-analysis of prospective studies. Depress Anxiety. 2015; Consent for publication 32(9):664–70. Not applicable 17. Leger D, Bayon V, Ohayon MM, Philip P, Ement P, Metlaine A, Chennaoui M, Faraut B. Insomnia and accidents: cross-sectional study (EQUINOX) on sleep- Competing interests related home, work and car accidents in 5293 subjects with insomnia from None of the authors has any competing interests. 10 countries. J Sleep Res. 2014;23(2):143–52. 18. Miner B, Kryger MH. Sleep in the aging population. Sleep Med Clin. 2017; Author details 12(1):31–8. 1 Division of Clinical Geriatrics, Center for Alzheimer Research, Karolinska 19. NSF: National Sleep Foundation. Sleep in Americal Poll. https://www. Institutet, Stockholm, Sweden. 2Neuroepidemiology and Ageing Research sleepfoundation.org/sites/default/files/inline-files/2003SleepPollExecSumm. Unit, School of Public Health, Imperial College London, London, UK. 3Aging pdf Accessed 1 Sept 2019 2003. Research Center, Department of Neurobiology, Care Sciences and Society, 20. Johnston MC, Crilly M, Black C, Prescott GJ, Mercer SW. Defining and Karolinska Institutet and Stockholm University, Stockholm, Sweden. 4REFiT measuring multimorbidity: a systematic review of systematic reviews. Eur J Barcelona Research Group, Vall d’Hebrón Research Institute and Parc Sanitari Pub Health. 2019;29(1):182–9. Pere Virgili, Barcelona, Spain. 5Centro di Medicina dell’Invecchiamento, IRCCS 21. Marengoni A, Angleman S, Melis R, Mangialasche F, Karp A, Garmen A, Fondazione Policlinico “A. Gemelli” and Catholic University of Rome, Rome, Meinow B, Fratiglioni L. Aging with multimorbidity: a systematic review of Italy. 6Institute of Gerontology, School of Health and Welfare, Aging Research the literature. Ageing Res Rev. 2011;10(4):430–9. Network – Jönköping (ARN-J), Jönköping University, Jönköping, Sweden. 22. Violan C, Foguet-Boreu Q, Flores-Mateo G, Salisbury C, Blom J, Freitag M, 7 Institute of Public Health and Clinical Nutrition and Institute of Clinical Glynn L, Muth C, Valderas JM. Prevalence, determinants and patterns of Medicine, Neurology, University of Eastern Finland, Kuopio, Finland. 8Theme multimorbidity in primary care: a systematic review of observational studies. Aging, Karolinska University Hospital, Stockholm, Sweden. 9Research and PLoS One. 2014;9(7):e102149. Development Unit, Stockholms Sjukhem, Stockholm, Sweden. 10Department 23. Lima MG, Bergamo Francisco PM, de Azevedo Barros MB. Sleep duration of Medicine, Universitat Autònoma de Barcelona, Barcelona, Spain. pattern and chronic diseases in Brazilian adults (ISACAMP, 2008/09). Sleep Med. 2012;13(2):139–44. Received: 7 April 2020 Accepted: 6 November 2020 24. Eckerblad J, Theander K, Ekdahl A, Unosson M, Wirehn AB, Milberg A, Krevers B, Jaarsma T. Symptom burden in community-dwelling older people with multimorbidity: a cross-sectional study. BMC Geriatr. 2015;15:1. References 25. Koyanagi A, Garin N, Olaya B, Ayuso-Mateos JL, Chatterji S, Leonardi M, 1. Hafner M, Stepanek M, Taylor J, Troxel WM, van Stolk C. Why sleep matters- Koskinen S, Tobiasz-Adamczyk B, Haro JM. Chronic conditions and sleep the economic costs of insufficient sleep: a cross-country comparative problems among adults aged 50 years or over in nine countries: a multi- analysis. Rand Health Q. 2017;6(4):11. country study. PLoS One. 2014;9(12):e114742. 2. Liu Y, Wheaton AG, Chapman DP, Cunningham TJ, Lu H, Croft JB. 26. Loprinzi PD. Health-enhancing multibehavior and medical multimorbidity. Prevalence of healthy sleep duration among adults - United States, 2014. Mayo Clin Proc. 2015;90(5):624–32. Mmwr-Morbid Mortal W. 2016;65(6):137–41. 27. Helbig AK, Stockl D, Heier M, Thorand B, Schulz H, Peters A, Ladwig KH, 3. Neikrug AB, Ancoli-Israel S. Sleep disorders in the older adult - a mini- Meisinger C. Relationship between sleep disturbances and multimorbidity review. Gerontology. 2010;56(2):181–9. among community-dwelling men and women aged 65-93 years: results 4. Gooneratne NS, Vitiello MV. Sleep in older adults: normative changes, sleep from the KORA Age Study. Sleep Med. 2017;33:151–9. disorders, and treatment options. Clin Geriatr Med. 2014;30(3):591–627. 28. Kim JM, Stewart R, Kim SW, Yang SJ, Shin IS, Yoon JS. Insomnia, depression, 5. Li J, Vitiello MV, Gooneratne NS. Sleep in normal aging. Sleep Med Clin. and physical disorders in late life: a 2-year longitudinal community study in 2018;13(1):1–11. Koreans. Sleep. 2009;32(9):1221–8. 6. Sarsour K, Kalsekar A, Swindle R, Foley K, Walsh JK. The association between 29. Dekhtyar S, Vetrano DL, Marengoni A, Wang HX, Pan KY, Fratiglioni L, insomnia severity and healthcare and productivity costs in a health plan Calderón-Larrañaga A. Association between speed of multimorbidity sample. Sleep. 2011;34(4):443–50. accumulation in old age and life experiences: a cohort study. Am J 7. Wickwire EM, Tom SE, Scharf SM, Vadlamani A, Bulatao IG, Albrecht JS. Epidemiol. 2019;188(9):1627–36. Untreated insomnia increases all-cause health care utilization and costs 30. Pérez LM, Hooshmand B, Mangialasche F, Mecocci P, Smith AD, Refsum H, among Medicare beneficiaries. Sleep. 2019;42(4):zsz007. Inzitari M, Fratiglioni L, Rizzuto D, Calderón-Larrañaga A. Glutathione serum 8. Itani O, Jike M, Watanabe N, Kaneita Y. Short sleep duration and health levels and rate of multimorbidity development in older adults. J Gerontol: outcomes: a systematic review, meta-analysis, and meta-regression. Sleep Series A. 2020;75(6):1089–94. Med. 2017;32:246–56. 31. Lagergren M, Fratiglioni L, Hallberg IR, Berglund J, Elmstahl S, Hagberg B, 9. Wang YM, Song M, Wang R, Shi L, He J, Fan TT, Chen WH, Wang L, Yu LL, Holst G, Rennemark M, Sjolund BM, Thorslund M, et al. A longitudinal study Gao YY, et al. Insomnia and multimorbidity in the community elderly in integrating population, care and social services data. The Swedish National China. J Clin Sleep Med. 2017;13(4):591–7. study on Aging and Care (SNAC). Aging Clin Exp Res. 2004;16(2):158–68. 10. Shi L, Chen SJ, Ma MY, Bao YP, Han Y, Wang YM, Shi J, Vitiello MV, Lu L. 32. Calderon-Larranaga A, Vetrano DL, Onder G, Gimeno-Feliu LA, Coscollar- Sleep disturbances increase the risk of dementia: a systematic review and Santaliestra C, Carfi A, Pisciotta MS, Angleman S, Melis RJF, Santoni G, et al. meta-analysis. Sleep Med Rev. 2018;40:4–16. Assessing and measuring chronic multimorbidity in the older population: a 11. Sindi S, Kareholt I, Johansson L, Skoog J, Sjoberg L, Wang HX, Johansson B, proposal for its operationalization. J Gerontol A Biol Sci Med Sci. 2017; Fratiglioni L, Soininen H, Solomon A, et al. Sleep disturbances and dementia 72(10):1417–23. risk: a multicenter study. Alzheimers Dement. 2018;14(10):1235–42. 33. Vetrano DL, Rizzuto D, Calderon-Larranaga A, Onder G, Welmer AK, Bernabei 12. Yaffe K, Falvey CM, Hoang T. Connections between sleep and cognition in R, Marengoni A, Fratiglioni L. Trajectories of functional decline in older older adults. Lancet Neurol. 2014;13(10):1017–28. adults with neuropsychiatric and cardiovascular multimorbidity: a Swedish 13. Fortier-Brochu E, Beaulieu-Bonneau S, Ivers H, Morin CM. Insomnia and cohort study. PLoS Med. 2018;15(3):e1002503. daytime cognitive performance: a meta-analysis. Sleep Med Rev. 2012; 34. Asberg M, Montgomery SA, Perris C, Schalling D, Sedvall G. A comprehensive 16(1):83–94. psychopathological rating scale. Acta Psychiatr Scand Suppl. 1978;271:5–27. 14. Sindi S, Johansson L, Skoog J, Mattsson AD, Sjoberg L, Wang HX, Fratiglioni 35. Rydwik E, Welmer AK, Kareholt I, Angleman S, Fratiglioni L, Wang HX. L, Kulmala J, Soininen H, Solomon A, et al. Sleep disturbances and later Adherence to physical exercise recommendations in people over 65--the cognitive status: a multi-centre study. Sleep Med. 2018;52:26–33. SNAC-Kungsholmen study. Eur J Pub Health. 2013;23(5):799–804. Sindi et al. BMC Medicine (2020) 18:382 Page 10 of 10 36. Montgomery SA, Asberg M. A new depression scale designed to be 60. Schmid SM, Hallschmid M, Schultes B. The metabolic burden of sleep loss. sensitive to change. Br J Psychiatry. 1979;134:382–9. Lancet Diabetes Endocrinol. 2015;3(1):52–62. 37. Zimmerman M, Chelminski I, Posternak M. A review of studies of the 61. Cunnington D, Junge M. Chronic insomnia: diagnosis and non- Montgomery-Asberg Depression Rating Scale in controls: implications for pharmacological management. BMJ. 2016;355:i5819. the definition of remission in treatment studies of depression. Int Clin 62. van der Zweerde TB, L.; Kyle, S.D.; Lancee, J.; van Straten, A.: Cognitive Psychopharmacol. 2004;19(1):1–7. behavioral therapy for insomnia: a meta-analysis of long-term effects in 38. Prados-Torres A, Calderon-Larranaga A, Hancco-Saavedra J, Poblador-Plou B, controlled studies. Sleep Medicine Reviews (In Press) 2019. van den Akker M. Multimorbidity patterns: a systematic review. J Clin 63. Virta JJ, Heikkila K, Perola M, Koskenvuo M, Raiha I, Rinne JO, Kaprio J. Epidemiol. 2014;67(3):254–66. Midlife sleep characteristics associated with late life cognitive function. 39. Wu Z, Chen F, Yu F, Wang Y, Guo Z. A meta-analysis of obstructive sleep Sleep. 2013;36(10):1533–41 1541A. apnea in patients with cerebrovascular disease. Sleep Breath. 2018;22(3): 64. Robichaud-Halle L, Beaudry M, Fortin M. Obstructive sleep apnea and 729–42. multimorbidity. BMC Pulm Med. 2012;12:60. 40. Swain S, Sarmanova A, Coupland C, Doherty M, Zhang W. Comorbidities in 65. Ruel G, Martin SA, Levesque JF, Wittert GA, Adams RJ, Appleton SL, Shi Z, osteoarthritis: a systematic review and meta-analysis of observational Taylor AW. Association between multimorbidity and undiagnosed studies. Arthritis Care Res (Hoboken). 2020;72(7):991–1000. obstructive sleep apnea severity and their impact on quality of life in men 41. Smith MT, Quartana PJ, Okonkwo RM, Nasir A. Mechanisms by which sleep over 40 years old. Glob Health Epidemiol Genom. 2018;3:e10. disturbance contributes to osteoarthritis pain: a conceptual model. Curr 66. Fabbri E, Zoli M, Gonzalez-Freire M, Salive ME, Studenski SA, Ferrucci L. Pain Headache Rep. 2009;13(6):447–54. Aging and multimorbidity: new tasks, priorities, and frontiers for integrated 42. Vitiello MV, McCurry SM, Shortreed SM, Baker LD, Rybarczyk BD, Keefe FJ, gerontological and clinical research. J Am Med Dir Assoc. 2015;16(8):640–7. Von Korff M. Short-term improvement in insomnia symptoms predicts long- 67. Leng Y, Musiek ES, Hu K, Cappuccio FP, Yaffe K. Association between term improvements in sleep, pain, and fatigue in older adults with circadian rhythms and neurodegenerative diseases. Lancet Neurol. 2019; comorbid osteoarthritis and insomnia. Pain. 2014;155(8):1547–54. 18(3):307–18. 43. Forman DE, Maurer MS, Boyd C, Brindis R, Salive ME, Horne FM, Bell SP, 68. Dzierzewski JM, Martin JL, Fung CH, Song Y, Fiorentino L, Jouldjian S, Fulmer T, Reuben DB, Zieman S, et al. Multimorbidity in older adults with Rodriguez JC, Mitchell M, Josephson K, Alessi CA. CBT for late-life insomnia cardiovascular disease. J Am Coll Cardiol. 2018;71(19):2149–61. and the accuracy of sleep and wake perceptions: results from a 44. Baker F.C. YD, de Zambotti M.: Sex differences in sleep. In: Sleep disorders in randomized-controlled trial. J Sleep Res. 2019;28(4):e12809. women edn. Edited by Attarian H. V-SM: Humana Cham; 2020. 45. Belvederi Murri M, Pariante C, Mondelli V, Masotti M, Atti AR, Mellacqua Z, Publisher’s Note Antonioli M, Ghio L, Menchetti M, Zanetidou S, et al. HPA axis and aging in Springer Nature remains neutral with regard to jurisdictional claims in depression: systematic review and meta-analysis. published maps and institutional affiliations. Psychoneuroendocrinology. 2014;41:46–62. 46. Burford NG, Webster NA, Cruz-Topete D. Hypothalamic-pituitary-adrenal axis modulation of glucocorticoids in the cardiovascular system. Int J Mol Sci. 2017;18(10):2150. 47. van Dalfsen JH, Markus CR. The influence of sleep on human hypothalamic- pituitary-adrenal (HPA) axis reactivity: a systematic review. Sleep Med Rev. 2018;39:187–94. 48. Morgan E, Schumm LP, McClintock M, Waite L, Lauderdale DS. Sleep characteristics and daytime cortisol levels in older adults. Sleep. 2017;40(5): zsx043. 49. Elder GJ, Wetherell MA, Barclay NL, Ellis JG. The cortisol awakening response--applications and implications for sleep medicine. Sleep Med Rev. 2014;18(3):215–24. 50. Buckley TM, Schatzberg AF. On the interactions of the hypothalamic- pituitary-adrenal (HPA) axis and sleep: normal HPA axis activity and circadian rhythm, exemplary sleep disorders. J Clin Endocrinol Metab. 2005; 90(5):3106–14. 51. Irwin MR, Olmstead R, Carroll JE. Sleep disturbance, sleep duration, and inflammation: a systematic review and meta-analysis of cohort studies and experimental sleep deprivation. Biol Psychiatry. 2016;80(1):40–52. 52. Gallo J, Raska M, Kriegova E, Goodman SB. Inflammation and its resolution and the musculoskeletal system. J Orthop Translat. 2017;10:52–67. 53. Smith KJ, Au B, Ollis L, Schmitz N. The association between C-reactive protein, Interleukin-6 and depression among older adults in the community: a systematic review and meta-analysis. Exp Gerontol. 2018;102:109–32. 54. Fabbri E, An Y, Zoli M, Simonsick EM, Guralnik JM, Bandinelli S, Boyd CM, Ferrucci L. Aging and the burden of multimorbidity: associations with inflammatory and anabolic hormonal biomarkers. J Gerontol A Biol Sci Med Sci. 2015;70(1):63–70. 55. Brown RE, Basheer R, McKenna JT, Strecker RE, McCarley RW. Control of sleep and wakefulness. Physiol Rev. 2012;92(3):1087–187. 56. Daut RA, Fonken LK. Circadian regulation of depression: a role for serotonin. Front Neuroendocrinol. 2019;54:100746. 57. Melancon MO, Lorrain D, Dionne IJ. Exercise and sleep in aging: emphasis on serotonin. Pathol Biol (Paris). 2014;62(5):276–83. 58. Sieminski M, Szypenbejl J, Partinen E. Orexins, sleep, and blood pressure. Curr Hypertens Rep. 2018;20(9):79. 59. Clark JW, Brian ML, Drummond SPA, Hoyer D, Jacobson LH. Effects of orexin receptor antagonism on human sleep architecture: a systematic review. Sleep Med Rev. 2020;53:101332.