Cohort Profile: WELL living laboratory in China (WELL-China)

Nonalcoholic fatty liver disease (NAFLD) detection and deep learning in a Chinese community-based population

OBJECTIVES

We aimed to develop and validate a deep learning system (DLS) by using an auxiliary section that extracts and outputs specific ultrasound diagnostic features to improve the explainable, clinical relevant utility of using DLS for detecting NAFLD.

METHODS

In a community-based study of 4144 participants with abdominal ultrasound scan in Hangzhou, China, we sampled 928 (617 [66.5%] females, mean age: 56 years ± 13 [standard deviation]) participants (2 images per participant) to develop and validate DLS, a two-section neural network (2S-NNet). Radiologists' consensus diagnosis classified hepatic steatosis as none steatosis, mild, moderate, and severe. We also explored the NAFLD detection performance of six one-section neural network models and five fatty liver indices on our data set. We further evaluated the influence of participants' characteristics on the correctness of 2S-NNet by logistic regression.

RESULTS

Area under the curve (AUROC) of 2S-NNet for hepatic steatosis was 0.90 for ≥ mild, 0.85 for ≥ moderate, and 0.93 for severe steatosis, and was 0.90 for NAFLD presence, 0.84 for moderate to severe NAFLD, and 0.93 for severe NAFLD. The AUROC of NAFLD severity was 0.88 for 2S-NNet, and 0.79-0.86 for one-section models. The AUROC of NAFLD presence was 0.90 for 2S-NNet, and 0.54-0.82 for fatty liver indices. Age, sex, body mass index, diabetes, fibrosis-4 index, android fat ratio, and skeletal muscle via dual-energy X-ray absorptiometry had no significant impact on the correctness of 2S-NNet (p > 0.05).

CONCLUSIONS

By using two-section design, 2S-NNet had improved the performance for detecting NAFLD with more explainable, clinical relevant utility than using one-section design.

KEY POINTS

• Based on the consensus review derived from radiologists, our DLS (2S-NNet) had an AUROC of 0.88 by using two-section design and yielded better performance for detecting NAFLD than using one-section design with more explainable, clinical relevant utility. • The 2S-NNet outperformed five fatty liver indices with the highest AUROCs (0.84-0.93 vs. 0.54-0.82) for different NAFLD severity screening, indicating screening utility of deep learning-based radiology may perform better than blood biomarker panels in epidemiology. • The correctness of 2S-NNet was not significantly influenced by individual's characteristics, including age, sex, body mass index, diabetes, fibrosis-4 index, android fat ratio, and skeletal muscle via dual-energy X-ray absorptiometry.

Cross-sectional and longitudinal associations between body flexibility and sarcopenia

BACKGROUND

The associations between body flexibility and sarcopenia were not well understood. This study aimed to explore the cross-sectional and longitudinal associations of flexibility with sarcopenia.

METHODS

Our study selected participants aged 50-80 from the WELL-China cohort and the Lanxi cohort. Participants from the urban area of the Lanxi cohort were followed up 4 years later. Body flexibility was measured by the sit-and-reach test. Muscle mass was evaluated by dual-energy X-ray absorptiometry. Muscle strength was evaluated using handgrip strength. Sarcopenia was defined as having both low muscle mass and low muscle strength. We used multivariable logistic regressions to assess the cross-sectional associations of body flexibility with low muscle mass, low muscle strength and sarcopenia. We also used multivariable logistic regressions to explore the associations of baseline flexibility and 4-year changes in flexibility with incident low muscle mass, low muscle strength and sarcopenia.

RESULTS

A total of 9453 participants were enrolled in the cross-sectional study, and 1233 participants were included in the longitudinal analyses. In the cross-sectional analyses, compared with low body flexibility, high body flexibility was inversely associated with low muscle mass (odds ratio [OR], 0.58; 95% confidence interval [CI], 0.50-0.68; P < 0.001), low muscle strength (OR, 0.62; 95% CI, 0.55-0.69; P < 0.001) and sarcopenia (OR, 0.52; 95% CI, 0.41-0.65; P < 0.001), and these associations did not differ in different age groups, sex or physical activity levels. In the longitudinal analyses, compared with participants with low body flexibility, participants with high body flexibility had lower risk of the incident low muscle strength (OR, 0.53; 95% CI, 0.38-0.74; P < 0.001) and sarcopenia (OR, 0.36; 95% CI, 0.21-0.61; P < 0.001), but not incident low muscle mass (OR, 0.59; 95% CI, 0.33-1.06; P = 0.076). Every 1-cm increase in flexibility during 4 years was associated with reduced risk of incident low muscle mass (OR, 0.96; 95% CI, 0.93-1.00; P = 0.025), low muscle strength (OR, 0.96; 95% CI, 0.94-0.98; P = 0.002) and sarcopenia (OR, 0.96; 95% CI, 0.93-0.99; P = 0.007).

CONCLUSIONS

High flexibility was associated with reduced risk of incident low muscle strength and sarcopenia. Increases in flexibility were associated with reduced risk of incident low muscle mass, low muscle strength and sarcopenia. Flexibility exercises and monitoring the dynamic change of flexibility might be helpful in preventing sarcopenia among adults aged 50 years or over.

Contemplative Practices Behavior Is Positively Associated with Well-Being in Three Global Multi-Regional Stanford WELL for Life Cohorts

Positive associations between well-being and a single contemplative practice (e.g., mindfulness meditation) are well documented, yet prior work may have underestimated the strength of the association by omitting consideration of multiple and/or alternative contemplative practices. Moreover, little is known about how contemplative practice behavior (CPB) impacts different dimensions of well-being. This study investigates the relationship of CPB, consisting of four discrete practices (embodied somatic-observing, non-reactive mindfulness, self-compassion, and compassion for others), with multiple dimensions of well-being. As with other canonical lifestyle behaviors, multiple contemplative practices can be integrated into one's daily routine. Thus, it is critical to holistically consider these behaviors, extending them beyond a simple uni-dimensional measure (e.g., daily mindfulness meditation practice). We developed an integrative measure of four types of contemplative practice and found it to be significantly associated with a multi-dimensional measure of well-being. Importantly, our findings were from three large global multi-regional cohorts and compared against better-understood lifestyle behaviors (physical activity). Data were drawn from California/San Francisco Bay Area, (n = 6442), Hangzhou City (n = 10,268), and New Taipei City (n = 3033). In all three cohorts, we found statistically significant (p < 0.05) positive associations between CPB and well-being, both overall and with all of the constituent domains of well-being, comparable to or stronger than the relationship with physical activity across most well-being outcomes. These findings provide robust and cross-cultural evidence for a positive association between CPB and well-being, illuminate dimensions of well-being that could be most influenced by CPB, and suggest CPB may be useful to include as part of fundamental lifestyle recommendations for health and well-being.

Changes in lifestyles and depressive symptom among patients with chronic diseases during COVID-19 lockdown

This study aims to investigate the impact of COVID-19 lockdown on lifestyle behaviors and depressive symptom among patients with NCDs (noncommunicable diseases). We incorporated a COVID-19 survey to the WELL China cohort, a prospective cohort study with the baseline survey conducted 8–16 months before the COVID-19 outbreak in Hangzhou, China. The COVID-19 survey was carried out to collect information on lifestyle and depressive symptom during lockdown. A total of 3327 participants were included in the COVID-19 survey, including 2098 (63.1%) reported having NCDs at baseline and 1457 (44%) without NCDs. The prevalence of current drinkers decreased from 42.9% before COVID-19 lockdown to 23.7% during lockdown, current smokers from 15.9 to 13.5%, and poor sleepers from 23.9 to 15.3%, while low physical activity increased from 13.4 to 25.2%, among participants with NCDs (P < 0.05 for all comparisons using McNemar's test). Participants with NCDs were more likely than those without to have depressive symptom (OR, 1.30; 95% CI 1.05–1.61), especially among those who need to refill their medication during the COVID-19 lockdown (OR, 1.52; 95% CI 1.15–2.02). Our findings provide insight into the development of targeted interventions to better prepare patients with NCDs and healthcare system to meet the challenge of future pandemic and lockdown.

Association between napping status and depressive symptoms in urban residents during the COVID-19 epidemic

: To explore the association between napping status and depressive symptoms in urban residents during the coronavirus disease 2019 (COVID-19) epidemic. : The survey was embedded in the Wellness Living Laboratory-China (WELL China) cohort study. Health and lifestyle information during the COVID-19 epidemic were obtained via the telephone interview from April 8, 2020 to May 29, 2020. A total of 3075 residents aged 18 to from Gongshu district of Hangzhou city with complete data were included in the analyses. The World Health Organization-Five Well-being Index (WHO-5) was used to measure depressive symptoms. Multiple logistic regression model was used to assess the association between napping status and depressive symptoms in the participants. : The prevalence of depressive symptoms was 20.6% in the participants during the epidemic. Daytime napping behavior, especially napping time ≤30 min, was associated with a lower risk of prevalent depressive symptoms (=0.61, 95%: 0.47-0.79, <0.01) and incident depressive symptoms in the population (=0.66, 95%: 0.50-0.88, <0.01). Among those with depressive symptoms at baseline, napping time ≤ was beneficial for the outcome of depressive symptoms (=0.42, 95%: 0.21-0.82, <0.05). : One in five urban residents have depressive symptoms during the COVID-19 epidemic, and a short nap during the day may be a protective factor against depressive symptoms.

Metabolomic Profiles of Plasma Retinol-Associated Dyslipidemia in Men and Women

Background and Aims: Studies of both animals and humans show that a high intake of vitamin A is associated with a lower risk of dyslipidemia. However, an association of plasma retinol levels with dyslipidemia is unclear. Therefore, the aim of this study is to investigate an association between plasma retinol and dyslipidemia and to identify related metabolites and pathways in the general population. Methods: We included 250 participants aged 20-80 years from the Wellness Living Laboratory (WELL) China cohort. Associations between plasma retinol levels and dyslipidemia were analyzed using adjusted logistic models. Related metabolites were identified using ANCOVA, adjusted for the false discovery rate (FDR) and used for pathway analyses. Because there are sex differences in plasma retinol levels, all analyses were conducted separately by sex. Results: Plasma retinol was significantly higher in men than in women. A positive association between plasma retinol and dyslipidemia was found in both sexes. In men, the 2nd and 3rd tertiles showed significantly higher proportions of dyslipidemia than the 1st tertile (1st tertile vs. 2nd tertile: p = 0.026; 1st tertile vs. 3rd tertile: p = 0.003). In women, the 3rd tertile showed a significantly higher proportion of dyslipidemia than the 1st and 2nd tertile (3rd tertile vs. 1st tertile: p = 0.002, 3rd tertile vs. 2nd tertile: p = 0.002). Overall, 75 and 30 metabolites were significantly associated with retinol levels in men and women, respectively. According to these metabolites, lipid metabolic pathways, including glycerophospholipid, arachidonic acid, linoleic acid, alpha-linolenic acid, and glycosylphosphatidylinositol (GPI), as well as steroid hormone biosynthesis pathways were found to overlap across the sexes. These pathways showed that elevated retinol levels might be associated with hormone metabolism and inflammation status. Conclusions: We found a positive association between plasma retinol levels and dyslipidemia. Related metabolomic profiles and interrupted pathways showed that such an increase might be associated with steroid hormone synthesis and inflammation. In addition, large, population-based longitudinal studies and intervention studies are needed to confirm the role of retinol in lipid metabolism and the prevention of cardiovascular disease (CVD).

Association between contact with a general practitioner and depressive symptoms during the COVID-19 pandemic and lockdown: a large community-based study in Hangzhou, China

OBJECTIVES

To determine the association of general practitioner (GP) contact with depressive symptoms during the COVID-19 pandemic and lockdown in China.

DESIGN

In April 2020, a follow-up survey was conducted on the basis of a baseline survey conducted between October 2018 and May 2019.

SETTING

The survey was embedded in the Stanford Wellness Living Laboratory-China (WELL China) study, an ongoing prospective community-based cohort study during 2018-2019.

PARTICIPANTS

The survey was conducted by telephone interview among 4144 adult urban residents participating in the WELL China study at baseline. We collected information on sociodemographic characteristics, depressive symptoms and GP contact during the lockdown period (February to March 2020).

PRIMARY AND SECONDARY OUTCOME MEASURES

Depressive symptoms were measured using the WHO-Five Well-being Index, comprising five questionnaire items that briefly indicate psychological well-being. Logistic regression models were applied to assess the association between GP contact and depressive symptoms.

RESULTS

In total, 3356 participants responded to the survey; 203 were excluded owing to missing data on depressive symptoms, leaving 3153 participants in the present study. During lockdown, 449 participants had GP contact. GP contact was significantly negatively associated with prevalent depressive symptoms (OR, 0.67; 95% CI 0.51 to 0.89; p<0.01) and incident depressive symptoms (OR 0.68; 95% CI 0.51 to 0.93; p<0.05). Stratified analysis showed a significant negative association between depressive symptoms and GP contact in individuals who were 45-64 years old (p<0.01), had a middle or high education (p<0.01) and had self-reported non-communicable diseases (p<0.05).

CONCLUSIONS

Contact with GPs during the COVID-19 pandemic and lockdowns may have a negative association with depressive symptoms in community-dwelling populations. Given the possibility of further surges in COVID-19 infections, GPs' contact in the community should be enhanced.