Maintaining healthy sleep patterns and frailty transitions: a prospective Chinese study | BMC Medicine

Study design and participants

The CKB study is a prospective cohort study that included 512,725 adults aged 30 to 79 living in 10 regions across China. Extensive questionnaire data, body measurements and blood samples were collected at the baseline assessment in 2004-2008 under the direction of trained health workers. In addition, periodic follow-up surveys were conducted in 2008 and 2013-2014 with a random sample of 5% of surviving participants. The second resurvey added several detailed items to the questionnaire and physical examination. Details of the CKB design have been previously described [11]. The CKB study was approved by the Ethics Review Committee of the Chinese Center for Disease Control and Prevention (Beijing, China, 005/2004) and the Oxford Tropical Research Ethics Committee, University of Oxford (Cambridge, United Kingdom, 025-04). All participants in the CKB study gave written informed consent.

In the present study, we included participants who completed both baseline and second post-assessment with no missing values ​​on the variables used to construct the Frailty Index (FI). We excluded frail participants at baseline from the present analyzes to avoid reverse causality.

Sleep pattern assessment

Detailed information on sleep patterns and other covariates was collected through a laptop-based baseline and post-survey questionnaire, including sociodemographic characteristics (age, gender, study areas, highest education level), smoking status, and alcohol consumption.

The questionnaire for sleep patterns in CKB surveys was shown in Supplementary File 1: Text S2. The present study defined the following sleep patterns. (a) Short/long habitual sleep duration: Respondents were asked, “How many hours do you typically sleep per day (including naps)?” Sleep duration could only be reported on an hourly basis. According to the American National Sleep Foundation, short sleep duration was defined as 6 hours/day or less, while long sleep duration was defined as 9 hours or more [12]. (b) insomnia: Participants were asked about the symptoms of insomnia for at least 3 days/week in the past month, having difficulty falling asleep or maintaining sleep (DIMS), early morning awakenings (EMA), daytime dysfunction (DDF), and taking medication (herbal or sleeping pills). ) at least once a week to aid sleep. According to the Diagnostic and Statistical Manual of Mental Disorders (Version 4) (DSM-4), those who reported either DIMS or EMA and DDF, or those who reported taking sleep aid medications were classified as suffering from insomnia [13, 14]. (c) snoring: Participants self-reported their usual snoring status during sleep as often, sometimes and never/don’t know. Those who chose the first two options were placed in the snoring group, others in the non-snoring group. In our study, sleep duration (Spearman’s correlation coefficient = 0.63) and snoring (weighted kappa = 0.69) showed good reproducibility in 15,720 participants who completed a repeat questionnaire survey within 1 to 2 weeks of baseline. (d) Healthy sleep baseline scores: Previously, Fan and colleagues developed an index value for healthy sleep patterns from several sleep behaviors (e.g. sleep duration, insomnia, snoring). It has been independently replicated in a large cohort [15, 16]. In our study, participants received one point for each of the three sleep patterns at baseline (sleep duration: 7 or 8 h/d, no insomnia, no snoring). (e) maintaining healthy sleep patterns: For each sleep pattern, participants were divided into 2 groups (yes or no) based on whether the participants maintained the healthy sleep pattern at both baseline and second re-examination, and were assigned 1 or 0 points for each sleep pattern. The overall scores for consistently healthy sleep were the number of times that healthy sleep patterns were maintained.

Also Read :  Thousands of Kiwi with severe asthma given access to newly funded medicine

FI construction and weak transitions

The Frailty Index (FI) was one of the widely used measures of frailty status developed by Mitnitski and colleagues [17]. FI encompassed multiple health deficits in different physiological systems. Based on the CKB baseline survey, we constructed FI using a standard procedure in which 28 deficits were selected. And accelerated aging, as measured by the FI, was associated with mortality risk [18]. The deficits included 14 self-reported diagnoses of disease (e.g. coronary artery disease, diabetes, cancer), 10 self-reported symptoms (e.g. symptoms of insomnia, frequent cough, bowel movements), and 4 physical measurements (e.g. body mass index). ). [BMI]Waist to waist circumference [WHR]heartbeat) [18]. Including BMI (kg/m2) was calculated by dividing weight (kg) by the square of standing height (m), WHR was the ratio of waist circumference to hip circumference, and daily physical activity was calculated by dividing the metabolic equivalent of tasks (METs) for each type of task activity was multiplied by the number of hours spent per day on the corresponding activity, and then summing the MET hours for all activities [19].

Also Read :  UnityPoint Health Family Medicine Clinic hosts open house | News, Sports, Jobs

The FI construction for the second repeat study was slightly different in terms of medical history. For example, chronic heart disease (CHD) was included in the baseline survey, but more detailed categories in the second replicate survey, including acute myocardial infarction, angina pectoris, or other ischemic heart disease, individuals with any of the 3 subtypes were categorized as CHD patients. In addition, we included the incidents during the follow-up period. All other variables were measured in the same way.

In the present study, we removed the insomnia symptoms from the FI construct, leaving 27 items in the FI construct for both surveys. Then we divide frailty according to FI into three statuses: robust (FI ≤ 0.10), prefrail (0.10 < FI < 0.25), and frail (≥ 0.25). [18]. Accordingly, the frailty transitions were defined by the change in frailty status from baseline to the second reassessment, which remained robust, robust deterioration (from robust to prefrail or frail), remained prefrail, prefrail improvement (from prefrail to robust), and prefrail comprised deterioration (from prefrail to frail).

Also Read :  Clearmind Medicine Featured in "Psychedelics Tonight,"

Statistical analysis

The present study first described baseline characteristics and sleep patterns after frailty transitions, adjusted for gender, age, and study area. We then examined the associations between basic sleep patterns and frailty transitions. Participants who remained in their frailty status at baseline were each designated as a reference group (ie Remained Robust vs. Deteriorating, Prefrail Remained vs. Improved, Prefrail Remained vs. Deteriorating). Noting that the odds ratio may overestimate the effect in high prevalence outcomes [20]we used log-binomial regression to estimate the prevalence ratios (PR) and 95% CI [21]adjusted for sex, age, area of ​​study (urban or rural), highest level of education (above middle school or not), smoking status (current smoker or not), and alcohol consumption (daily drinker or not).

In addition, we examined whether maintaining healthy sleep patterns was associated with frail transitions. We also assessed the effects of baseline and consistently healthy sleep scores on frailty transitions and tested their linear trend by treating the scores as continuous variables. All models were log-binomial regression with the same fitting of the above covariates. In addition, the differences in sleep patterns and frailty status between older and younger participants were taken into account [12, 22]we performed the sensitivity analyzes by examining the above associations in individuals < 60 years of age (n= 18,995). Also, we further adjusted the major diseases and medication status at baseline (illness and medication, illness without medication, without illness) separately for cardiovascular diseases (including coronary artery disease, stroke or transient ischemic attack, hypertension) and their medication (including aspirin, angiotensin converting enzyme inhibitors, βblockers, diuretics, statins and calcium channel blockers) and diabetes and its drugs (including chlorpropamide, metformin and insulin) in the sensitivity analysis. Statistical analysis was performed in Stata 16.0 and charts were drawn with R 4.0.5. two-tailed P<0.05 indicated statistical significance.


Leave a Reply

Your email address will not be published.