Skip to main content
NCBI home page
Advances in Nutrition logoLink to Advances in Nutrition
. 2021 Sep 21;13(1):269–281. doi: 10.1093/advances/nmab115

Chronotype Differences in Energy Intake, Cardiometabolic Risk Parameters, Cancer, and Depression: A Systematic Review with Meta-Analysis of Observational Studies

Sofia Lotti 1,✉, Giuditta Pagliai 2, Barbara Colombini 3, Francesco Sofi 4,5, Monica Dinu 6
PMCID: PMC8803479  PMID: 34549270

ABSTRACT

Chronotype is a behavioral manifestation of the internal circadian clock system. It refers to the specific activity-rest preference of an individual over a 24-h period and can be assessed using different methodologies that classify individuals into morning or evening chronotype. In recent years, several studies have suggested a relation between individual chronotype, eating habits, and the risk of developing obesity and other conditions. Our aim was to evaluate the association between chronotype, energy intake, and health status through a meta-analytic approach. A comprehensive search of MEDLINE, Embase, Scopus, Web of Science, and Cochrane Database was conducted. Observational studies that reported a measure of association between chronotype, energy intake, and health indicators were considered eligible. Overall, 39 observational studies (37 cross-sectional studies, 2 prospective cohort studies) were included in the systematic review, with a total of 377,797 subjects. By comparing morning and evening subjects, pooled analyses of cross-sectional studies showed significantly (P < 0.001) higher concentrations of blood glucose [mean difference (MD): 7.82; 95% CI: 3.18, 12.45], glycated hemoglobin (MD: 7.64; 95% CI: 3.08, 12.21), LDL cholesterol (MD: 13.69; 95% CI: 6.84, 20.54), and triglycerides (MD: 12.62; 95% CI: 0.90, 24.35) in evening subjects. Furthermore, an association between evening type and the risk of diabetes (OR: 1.30; 95% CI: 1.20, 1.41), cancer (OR: 1.18; 95% CI: 1.08, 1.30), and depression (OR: 1.86; 95% CI: 1.20, 2.88) was reported. Regarding the other outcomes examined, no significant differences were observed between the groups in terms of energy intake, anthropometric parameters, blood pressure, insulin, total and HDL cholesterol, and hypertension risk. In conclusion, evening chronotype was associated with a worse cardiometabolic risk profile and higher risk of diabetes, cancer, and depression. Further studies are needed to confirm these results and to better elucidate the interplay between chronotype, nutrition, and health status.

This systematic review was registered at www.crd.york.ac.uk/prospero/ as CRD42021231044.

Keywords: chronotype, health, energy intake, risk factors, meta-analysis

Statement of Significance: This study includes all available observational studies to provide a comprehensive overview of the association between chronotype, nutritional parameters, and multiple health outcomes.

Introduction

Circadian rhythms, controlled by the master circadian clock located in the suprachiasmatic nuclei of the hypothalamus, regulate daily sleep/wake rhythms, feeding behavior, and hormone secretions (1). Individual circadian typology has been summarized under the concept of chronotype, which refers to the specific activity-rest preference of an individual over a 24-h period (2). Early risers who are preferentially active in the mornings are said to have a morning chronotype, whereas late risers with more nocturnal activities are said to have an evening chronotype (61). Different methods can be used to assess chronotype; the most common is the administering of validated questionnaires, such as the “Morningness-Eveningness Questionnaire” (MEQ) (3).

In recent years, chrono-nutrition—the science that combines elements of nutritional research with elements of chronobiology—has received increasing attention given the growing literature revealing a possible association between chronotype, dietary habits, and health (4–6). In fact, many studies suggest that evening subjects have worse eating habits and consume more alcoholic beverages and sweets and less whole grains, fish, vegetables, and fruit (4). Furthermore, a possible relation between the evening chronotype and higher BMI; higher concentrations of triglycerides, total cholesterol, LDL cholesterol, and glucose; and lower concentrations of HDL cholesterol has been reported (7, 8). Also, in terms of disease risk, the evening chronotype has been associated with an increased risk of metabolic disorders (5), type 2 diabetes (9), cardiovascular disease (10), and depression (11).

Although the most accredited hypothesis is that evening subjects have a higher risk of cardiometabolic and chronic diseases, no conclusive data have been obtained and no systematic reviews and meta-analyses have been conducted so far. The aim of this study was to carry out a comprehensive systematic review with meta-analysis of all cross-sectional and prospective cohort studies hitherto published in order to obtain an estimate of the association between chronotype, energy intake, and health status.

Methods

Search strategy

The review protocol has been registered on PROSPERO (CRD42021231044). According to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement (12), all potentially relevant articles were identified through a computerized search of the main electronic databases: MEDLINE, Embase, Scopus, Web of Science, and Cochrane Database, from inception to 1 April, 2021. Reference lists of the identified studies and previous reviews were also screened. Search terms included the following key words, used in combination as Medical Subject Headings (MeSH) terms and text words: “chronotype,” “diurnal preference,” “nocturnal preference,” “circadian typology,” “morningness,” “eveningness,” and their variants, in combination with words relating to dietary intake and health status: “energy intake,” “plasma lipids,” “cholesterol,” “glycemia,” “cardiovascular disease*,” “cancer,” “obesity,” “body mass index,” “diabetes,” “metabolic syndrome,” “depression,” “mortality,” “health,” “health status,” and their variants. Supplemental Table 1 provides a more exhaustive search strategy list, for each database. No language limitations were applied.

Data selection

Two authors (SL and MD) independently assessed potentially relevant articles for eligibility. Observational studies (cross-sectional and prospective cohort studies) that reported a measure of association between chronotype, energy intake, and/or health indicators were considered eligible for inclusion. Eligibility criteria are summarized in Supplemental Table 2, by following the PECOS (Population, Exposure, Comparator, Outcomes, Study design) framework. Inclusion criteria were as follows: 1) Population: adults (≥18 y old); 2) Exposure: evening chronotype; 3) Comparator: morning chronotype; 4) Outcomes: energy intake, cardiometabolic parameters [BMI, total cholesterol, LDL cholesterol, HDL cholesterol, triglycerides, glucose, insulin, glycated hemoglobin (HbA1c), systolic blood pressure, diastolic blood pressure], chronic degenerative diseases (e.g., cardiovascular diseases, cancer, depression); 5) Study design: cross-sectional and prospective cohort studies.

Exclusion criteria were as follows: 1) Population: nonadults (<18 y old), pregnancy, and postpartum; 2) Exposure: intermediate chronotype; 3) Outcomes: outcomes assessed with questionnaires (e.g., anxiety, depressive moods); 4) Study design: case-control studies (to minimize bias in recall and selection), review articles, letters to the editor, comments, case reports, and randomized controlled trials. Studies not reporting sufficient data to allow calculation of differences between subjects with evening preference and subjects with morning preference were excluded as well. When multiple articles for a single cohort were present, the most recent publication was considered. Missing data or necessary additional information were requested from the corresponding authors of the articles.

The decision to include studies was initially based on the title, the abstract, and full-text screening. In case of disagreement between the 2 reviewers, a third reviewer (FS) was consulted to reach consensus.

Data extraction

Two authors (SL and MD) independently extracted data from each study using a standardized form. Disagreements were resolved by consensus, or by a third investigator (FS) if consensus could not be reached. The spreadsheet was elaborated in Microsoft Excel® for Windows (2007) and was prepiloted, on 5 randomly selected articles, to ensure methodological concordance among the authors. The following data were extracted: first author and year of publication, study design, country of study population, age, sex, length of follow-up (when applicable), method used to assess individual chronotype, number of participants with morning chronotype, number of participants with evening chronotype, definition of outcome of interest, measures of effect size and CIs, and details of adjustment for confounding factors in the multivariate model (when available). If the results were reported separately for women and men, they were included in the analysis as separate populations.

Continuous outcomes were reported as follows: energy intake (kcal), body weight (kg), BMI (in kg/m2), fat percentage (%), systolic blood pressure (mm Hg), diastolic blood pressure (mm Hg), glucose (mg/dL), insulin (μU/mL), HbA1c (%), insulin resistance score (HOMA-IR), total cholesterol (mg/dL), LDL cholesterol (mg/dL), HDL cholesterol (mg/dL), and triglycerides (mg/dL). When data were provided in mmol/L, they were transformed into mg/dL for consistency of results.

Quality assessment

Two authors (SL and MD) independently assessed the methodological quality of the included studies using the Newcastle-Ottawa Scale (NOS) (13). Any incongruity was discussed and resolved with a third reviewer (FS). This scale assesses each study in 3 domains: the selection of the participants for each group, the comparability between the study groups, and the ascertainment of the outcome. We considered high-quality studies those that achieved ≥7 points, medium-quality studies those with 4–6 points, and poor-quality studies those with ≤3 points.

Statistical analysis

All data were analyzed using Review Manager (RevMan; version 5.4 for Macintosh). Pooled results were reported as mean difference (MD), OR, or HR and presented with 95% CIs with 2-sided P values. A random-effects model (DerSimonian and Laird method), which accounts for interstudy variation and provides a more conservative effect than the fixed model, was used. Meta-analysis was conducted if ≥2 studies were available for an outcome. When available, the results of the original studies with the most complete adjustment for potential confounders were used.

The chi-square Cochran's Q test with the I2 statistic was calculated to assess the statistical heterogeneity between studies. The I2 value determined the appropriateness of pooling the individual study results and provided a variance estimation across studies based on heterogeneity rather than chance (14). Where I2 was >50%, heterogeneity was defined as substantial and subgroup analyses were performed to establish the source of the heterogeneity (15). We assessed whether there were differences related to geographical region [Northern compared with Southern countries, as defined by Brandt (16)], study populations (clinically healthy subjects compared with patients, i.e., subjects with a clinical diagnosis of disease), and study quality (low = studies with scores ≤ 3 on the NOS; moderate = studies with scores ranging from 4 to 6 on the NOS; high = studies with scores ≥ 7 on the NOS). To establish the robustness of the results, a sensitivity analysis was conducted by removing each study one-by-one from the meta-analysis and recalculating the summary estimate (the “leave-one-out” approach). If >10 studies were available, the possibility of publication bias was investigated by visual inspection of a funnel plot of effect size against SE. A P value < 0.05 was considered statistically significant.

Results

Literature search and study characteristics

Figure 1 shows the selection process, in accordance with PRISMA guidelines. Initial databases and other searches yielded 7194 articles. After review and elimination of duplicates, 251 articles were identified as potentially relevant for analysis. Of these, 212 were excluded based on full-text evaluation. At the end of the selection process, 39 articles with a total of 377,797 subjects (75% morning types, 25% evening types) met the inclusion criteria and were included in the analysis. Most of the studies had a cross-sectional design (n = 37), and Tables 1 and 2 summarize their characteristics. As for the prospective cohort studies, only 2 studies were identified, and Table 3 shows their characteristics.

FIGURE 1.

FIGURE 1

Open in a new tab

Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram for search strategy.

TABLE 1.

Characteristics of cross-sectional studies evaluating chronotype, energy intake, and cardiometabolic risk factors1

Study Country Sex Age, y2 Study population Morning type, n Definition of morning type Evening type, n Definition of evening type Outcome Mean difference (95% CI) Study quality
Energy intake
 Iwasaki et al. (17) Japan M 53.9 ± 7.1 Diabetic population 32 MES > 59 points (MEQ) 11 MES < 41 points (MEQ) Energy intake, kcal/d −29.90 (−373.20, 313.40) Medium
 Kanerva et al. (4) Finland M/F 25–74 General population 995 5th quintile of rMEQ 826 1st quintile of rMEQ Energy intake, kcal/d 0.00 (−27.68, 27.68) Medium
 Lucassen et al. (18) USA M/F 41.7 ± 5.9 Obese population 80 MES > 50 points (MEQ) 39 MES < 49 points (MEQ) Energy intake, kcal/d 147.00 (−143.77, 437.77) Medium
 Reutrakul et al. (19) USA M/F 18–85 Diabetic population 51 1st quartile of MSF (MCTQ) 48 4th quartile of MSF (MCTQ) Energy intake, kcal/d −19.00 (−256.49, 218.49) Low
 Osonoi et al. (20) Japan M/F 57.8 ± 8.6 Diabetic population 117 MES > 65 points (MEQ) 184 MES < 52 points (MEQ) Energy intake, kcal/d 31.00 (−87.96, 149.96) Medium
 Maukonen et al. (21) Finland M 25–74 General population 839 MES 19–27 points (MEQ) 568 MES 5–12 points (MEQ) Energy intake, kcal/d 42.75 (26.99, 58.51) High
 Maukonen et al. (21) Finland F 25–74 General population 816 MES 19–27 points (MEQ) 669 MES 5–12 points (MEQ) Energy intake, kcal/d −23.88 (−34.52, −13.24) High
 Munoz et al. (22) Spain M/F 30–60 General population 80 MES ≥ 52 points (MEQ) 91 MES ≤ 51 points (MEQ) Energy intake, kcal/d 165.00 (106.38, 223.62) Medium
 Ruiz-Lozano et al. (23) Spain M/F 52 ± 11 Obese population 124 MES > 64 points (MEQ) 128 MES < 53 points (MEQ) Energy intake, kcal/d 31.00 (−225.47, 287.47) Medium
 Basnet et al. (24) Finland M/F 51.6 ± 13.8 General population 1935 MES 19–27 points (MEQ) 595 MES 5–12 points (MEQ) Energy intake, kcal/d −17.59 (−42.00, 6.82) High
 Maukonen et al. (25) Finland M/F 25–74 General population 904 MES 19–27 points (MEQ) 224 MES 5–12 points (MEQ) Energy intake, kcal/d 17.44 (−12.21, 47.09) High
 Teixeira et al. (26) Brazil M/F ≥18 General population 151 MES > 59 points (MEQ) 124 MES < 41 points (MEQ) Energy intake, kcal/d 140.10 (88.98, 191.22) High
 Vera et al. (7) Spain M/F 40 ± 13 General population 1110 MES > 59 points (MEQ) 1016 MES < 41 points (MEQ) Energy intake, kcal/d −54.26 (−121.50, 12.98) High
 Yoshizaki et al. (27) Japan F 20–59 Nurses 270 MES > 60 points (MEQ) 336 MES < 53 points (MEQ) Energy intake, kcal/d −29.00 (−105.35, 47.35) Medium
 Yazdinezhad et al. (28) Iran F ≥20 General population 25 MES > 52 points (MEQ) 16 MES < 51 points (MEQ) Energy intake, kcal/d −73.00 (−249.62, 103.62) Medium
 Yazdinezhad et al. (28) Iran F ≥20 Obese population 39 MES > 52 points (MEQ) 16 MES < 51 points (MEQ) Energy intake, kcal/d −114.00 (−322.89, 94.89) Medium
 Kayacan and Tokay (29) Turkey M 39.7 ± 14.9 Patients with IBD 8 MES > 59 points (MEQ) 6 MES < 41 points (MEQ) Energy intake, kcal/d 520.60 (306.07, 735.13) Medium
 Kayacan and Tokay (29) Turkey F 39.7 ± 14.9 Patients with IBD 4 MES > 59 points (MEQ) 8 MES < 41 points (MEQ) Energy intake, kcal/d −220.60 (−305.38, −135.82) Medium
Cardiometabolic risk factors
 Gaspar-Barba et al. (30) Mexico M/F 34.0 ± 11.7 Patients with major depressive disorder 21 MES > 59 points (MEQ) 18 MES < 41 points (MEQ) Body weight, kg −1.76 (−8.96, 5.44) Medium
BMI, kg/m2 −2.50 (−4.91, −0.09)
 Selvi et al. (31) Turkey M/F ≥18 Patients after AMI 63 MES > 59 points (MEQ) 40 MES < 41 points (MEQ) Total cholesterol, mg/dL −2.08 (−18.18, 14.02) Medium
HDL-C, mg/dL −1.40 (−5.24, 2.44)
LDL-C, mg/dL 2.32 (−18.07, 22.71)
Triglycerides, mg/dL 14.98 (−25.12, 55.08)
 Iwasaki et al. (17) Japan M 53.9 ± 7.1 Diabetic population 32 MES > 59 points (MEQ) 11 MES < 41 points (MEQ) BMI, kg/m2 0.70 (−1.82, 3.22) Medium
Systolic BP, mm Hg −7.00 (−17.57, 3.57)
Diastolic BP, mm Hg −0.60 (−6.26, 5.06)
HbA1c, % 12.00 (1.01, 22.99)
HDL-C, mg/dL −7.70 (−14.65, −0.75)
LDL-C, mg/dL 29.00 (12.78, 45.22)
Triglycerides, mg/dL 24.70 (−24.97, 74.37)
 Kanerva et al. (4) Finland M/F 25–74 General population 995 5th quintile of rMEQ 826 1st quintile of rMEQ BMI, kg/m2 −0.20 (−0.76, 0.36) High
 Meule et al. (32) Germany M/F 23.1 ± 2.7 General population 35 MES ≥ 55 points (MEQ) 31 MES ≤ 44 points (MEQ) BMI, kg/m2 −0.80 (−2.92, 1.32) Low
 Roeser et al. (33) Germany F 23.2 ± 4 General population 27 MES > 59 points (MEQ) 28 MES < 41 points (MEQ) BMI, kg/m2 0.07 (−1.36, 1.50) Medium
Systolic BP, mm Hg 4.48 (−0.02, 8.98)
Diastolic BP, mm Hg 2.62 (−0.55, 5.79)
 Lucassen et al. (18) USA M/F 41.7 ± 5.9 Obese population 80 MES > 50 points (MEQ) 39 MES < 49 points (MEQ) BMI, kg/m2 0.90 (−1.59, 3.39) High
Blood glucose, mg/dL 1.50 (−2.29, 5.29)
Insulin, μU/mL 1.00 (0.13, 1.87)
Total cholesterol, mg/dL −1.80 (−15.54, 11.94)
HDL-C, mg/dL 1.00 (−0.24, 2.24)
LDL-C, mg/dL 10.00 (6.02, 13.98)
Triglycerides, mg/dL −16.00 (−21.58, −10.42)
 Merikanto et al. (9) Finland M/F 25–74 General population 3242 MES 19–27 points (MEQ) 809 MES 5–12 points (MEQ) Body weight, kg −0.20 (−1.49, 1.09) High
Fat percentage, % −0.20 (−0.89, 0.49)
 Reutrakul et al. (19) USA M/F 18–85 Diabetic population 51 1st quartile of MSF (MCTQ) 48 4th quartile of MSF (MCTQ) BMI, kg/m2 2.30 (−0.76, 5.36) Low
HbA1c, % 14.00 (12.01, 15.99)
 Osonoi et al. (20) Japan M/F 57.8 ± 8.6 Diabetic population 117 MES > 65 points (MEQ) 184 MES < 52 points (MEQ) BMI, kg/m2 2.20 (1.30, 3.10) Medium
Systolic BP, mm Hg 1.00 (−2.30, 4.30)
Diastolic BP, mm Hg 2.00 (−1.08, 5.08)
Blood glucose, mg/dL 10.00 (2.38, 17.62)
HbA1c, % 6.00 (3.63, 8.37)
Total cholesterol, mg/dL −3.00 (−9.35, 3.35)
HDL-C, mg/dL −5.00 (−8.07, −1.93)
Triglycerides, mg/dL 18.00 (13.65, 22.35)
 Kim et al. (34) Brazil M/F 20–80 General population 446 MES > 59 points (MEQ) 72 MES < 41 points (MEQ) BMI, kg/m2 −0.50 (−2.00, 1.00) Medium
 Merikanto et al. (35) Finland M/F 25–74 General population 4851 MES 19–27 points (MEQ) 1204 MES 5–12 points (MEQ) Body weight, kg −1.10 (−2.15, −0.05) Medium
BMI, kg/m2 −0.60 (−0.92, −0.28)
Systolic BP, mm Hg −8.90 (−10.03, −7.77)
Diastolic BP, mm Hg −3.30 (−3.99, −2.61)
 Vetter et al. (36) USA F 25–42 Nurses 22,089 Question no. 19 of MEQ 7029 Question no. 19 of MEQ BMI, kg/m2 2.20 (2.03, 2.37) Medium
 Yu et al. (5) Republic of Korea M/F 47–59 General population 480 MES > 59 points (MEQ) 95 MES < 41 points (MEQ) BMI, kg/m2 0.20 (−0.54, 0.94) High
Systolic BP, mm Hg −3.60 (−6.73, −0.47)
Diastolic BP, mm Hg 3.40 (1.32, 5.48)
Blood glucose, mg/dL 2.00 (−0.54, 4.54)
HbA1c, % 1.00 (0.62, 1.38)
HOMA-IR 0.20 (0.01, 0.39)
Total cholesterol, mg/dL 8.00 (6.05, 9.95)
HDL-C, mg/dL 0.00 (−0.70, 0.70)
Triglycerides, mg/dL 27.00 (26.56, 27.44)
 Antypa et al. (37) Netherlands M/F 19–68 Depressed patients 409 1st quintile of MSF (MCTQ) 387 5th quintile of MSF (MCTQ) BMI, kg/m2 −1.50 (−2.15, −0.85) Medium
 Dickerman et al. (38) Finland M/F 40 ± 12.1 Twins 3159 Question no. 19 of MEQ 1117 Question no. 19 of MEQ BMI, kg/m2 −1.00 (−1.21, −0.79) High
 Maukonen et al. (21) Finland M 25–74 General population 839 MES 19–27 points (MEQ) 568 MES 5–12 points (MEQ) BMI, kg/m2 −0.30 (−0.71, 0.11) High
Fat percentage, % −0.20 (−0.81, 0.41)
 Maukonen et al. (21) Finland F 25–74 General population 816 MES 19–27 points (MEQ) 669 MES 5–12 points (MEQ) BMI, kg/m2 −0.10 (−0.57, 0.37) High
Fat percentage, % 0.10 (−0.55, 0.75)
 Munoz et al. (22) Spain M/F 30–60 General population 80 MES ≥ 52 points (MEQ) 91 MES ≤ 51 points (MEQ) Body weight, kg 5.00 (−0.27, 10.27) Medium
BMI, kg/m2 0.30 (−1.50, 2.10)
Fat percentage, % 0.40 (−1.93, 2.73)
Systolic BP, mm Hg −4.00 (−8.50, 0.50)
Diastolic BP, mm Hg −4.00 (−6.55, −1.45)
Blood glucose, mg/dL 3.00 (0.29, 5.71)
HDL-C, mg/dL −6.00 (−6.98, −5.02)
Triglycerides, mg/dL 20.00 (18.46, 21.54)
 Ruiz-Lozano et al. (23) Spain M/F 52 ± 11 Obese population 124 MES > 64 points (MEQ) 128 MES < 53 points (MEQ) Body weight, kg 6.20 (1.43, 10.97) Medium
BMI, kg/m2 2.30 (0.84, 3.76)
 Suh et al. (39) Republic of Korea M/F 58 ± 7.1 General population 1138 MES > 59 points (MEQ) 146 MES < 41 points (MEQ) BMI, kg/m2 −0.18 (−0.87, 0.51) High
 Basnet et al. (24) Finland M/F 51.6 ± 13.8 General population 1935 MES 19–27 points (MEQ) 595 MES 5–12 points (MEQ) Body weight, kg −1.00 (−2.45, 0.45) High
BMI, kg/m2 −0.70 (−1.16, −0.24)
Fat percentage, % −0.06 (−0.88, 0.76)
Systolic BP, mm Hg 0.99 (−0.77, 2.75)
Diastolic BP, mm Hg 1.31 (0.28, 2.34)
 Maukonen et al. (25) Finland M/F 25–74 General population 904 MES 19–27 points (MEQ) 224 MES 5–12 points (MEQ) BMI, kg/m2 0.50 (−0.21, 1.21) High
 Teixeira et al. (26) Brazil M/F ≥18 General population 151 MES > 59 points (MEQ) 124 MES < 41 points (MEQ) BMI, kg/m2 −0.40 (−1.21, 0.41) High
 Knutson and Schantz (10) United Kingdom M/F 38–73 General population 117,224 Question no. 19 of MEQ 38,867 Question no. 19 of MEQ BMI, kg/m2 0.30 (0.24, 0.36) High
 Vera et al. (7) Spain M/F 40 ± 13 General population 1110 MES > 59 points (MEQ) 1016 MES < 41 points (MEQ) BMI, kg/m2 0.32 (−0.12, 0.76) High
Fat percentage, % −0.05 (−0.57, 0.47)
Systolic BP, mm Hg 0.00 (−1.26, 1.26)
Diastolic BP, mm Hg −0.30 (−1.10, 0.50)
Blood glucose, mg/dL 0.25 (−1.00, 1.50)
Insulin, μU/mL 0.22 (−0.40, 0.84)
HOMA-IR 0.07 (−0.09, 0.23)
HDL-C, mg/dL −1.50 (−2.80, −0.20)
Triglycerides, mg/dL 4.37 (3.88, 4.86)
 Yoshizaki et al. (27) Japan F 20–59 Nurses 336 MES > 60 points (MEQ) 400 MES < 53 points (MEQ) BMI, kg/m2 −0.10 (−0.50, 0.30) Medium
 Kwon et al. (8) Republic of Korea M/F 19–81 General population 145 MES > 59 points (MEQ) 145 MES < 41 points (MEQ) BMI, kg/m2 0.00 (−0.77, 0.77) High
Systolic BP, mm Hg −1.00 (−4.14, 2.14)
Diastolic BP, mm Hg −0.80 (−3.38, 1.78)
Blood glucose, mg/dL 0.20 (−3.53, 3.93)
HbA1c, % 1.00 (−0.63, 2.63)
Total cholesterol, mg/dL 9.90 (2.11, 17.69)
HDL-C, mg/dL −1.40 (−4.29, 1.49)
LDL-C, mg/dL 9.20 (1.80, 16.60)
Triglycerides, mg/dL 18.70 (4.23, 33.17)
 Sun et al. (40) USA M/F 48.2 ± 5.3 General population 498 3rd tertile of rMEQ 131 1st tertile of rMEQ BMI, kg/m2 2.80 (1.02, 4.58) Medium
 Tomizawa et al. (41) Japan M 28.2 Shift workers 32 MES > 59 points (MEQ) 17 MES < 41 points (MEQ) BMI, kg/m2 −0.50 (−2.03, 1.03) Medium
Total cholesterol, mg/dL 5.50 (−12.07, 23.07)
HDL-C, mg/dL 0.30 (−6.74, 7.34)
LDL-C, mg/dL 8.00 (−8.96, 24.96)
 Yazdinezhad et al. (28) Iran F ≥20 General population 25 MES > 52 points (MEQ) 16 MES < 51 points (MEQ) BMI, kg/m2 1.10 (−4.68, 6.88) Medium
 Yazdinezhad et al. (28) Iran F ≥20 Obese population 39 MES > 52 points (MEQ) 16 MES < 51 points (MEQ) BMI, kg/m2 2.20 (−3.32, 7.72) Medium
 De Amicis et al. (42) Italy M/F 50 ± 13 General population 135 3rd tertile of rMEQ 38 1st tertile of rMEQ BMI, kg/m2 −0.30 (−2.46, 1.86) Medium
 Hashemipour et al. (43) Iran M/F 40.7 ± 3.3 Diabetic population 42 MES > 59 points (MEQ) 58 MES < 41 points (MEQ) BMI, kg/m2 1.90 (0.66, 3.14) Low
Blood glucose, mg/dL 31.80 (20.58, 43.02)
HbA1c, % 20.00 (15.46, 24.54)
 Henson et al. (44) United Kingdom M/F 63.8 ± 8.4 Diabetic population 159 MES ≥ 65 points (MEQ) 146 MES ≤ 52 points (MEQ) BMI, kg/m2 1.80 (0.59, 3.01) Low
HbA1c, % 3.00 (0.06, 5.94)
 Muscogiuri et al. (45) Italy M/F 51.8 ± 15.7 Obese population 100 MES > 59 points (MEQ) 22 MES < 41 points (MEQ) Body weight, kg 0.80 (−5.61, 7.21) Medium
BMI, kg/m2 1.20 (−1.36, 3.76)
 Thapa et al. (46) Republic of Korea M/F ≥70 General population 56 MES > 59 points (MEQ) 46 MES < 41 points (MEQ) BMI, kg/m2 3.00 (1.94, 4.06) Low
 Kayacan and Tokay (29) Turkey M/F 39.7 ± 14.9 Patients with IBD 12 MES > 59 points (MEQ) 14 MES < 41 points (MEQ) Blood glucose, mg/dL 27.70 (21.68, 33.72) Medium
Open in a new tab
1

AMI, acute myocardial infarction; BP, blood pressure; HbA1c, glycated hemoglobin; HDL-C, HDL cholesterol; IBD, inflammatory bowel disease; LDL-C, LDL cholesterol; MCTQ, Munich Chronotype Questionnaire; MEQ, Morningness-Eveningness Questionnaire; MES, Morningness–Eveningness Score calculated from the Morningness-Eveningness Questionnaire items; MSF, Mid Sleep on Free Days; rMEQ, reduced Morningness-Eveningness Questionnaire.2Values are reported as mean ± SD

TABLE 2.

Characteristics of cross-sectional studies evaluating chronotype and diseases1

Study Country Sex Age, y2 Study population Morning type, n/total n Definition of morning type Evening type, n/total n Definition of evening type Outcome OR (95% CI) Adjustment Study quality
Selvi et al. (31) Turkey M/F ≥18 Patients after AMI 24/6310/63 MES > 59 points (MEQ) 16/407/40 MES < 41 points (MEQ) HypertensionDiabetes 1.08 (0.48, 2.44)1.35 (0.39, 2.39) — Medium
Merikanto et al. (9) Finland M/F 25–74 General population NA/3242 MES 19–27 points (MEQ) NA/809 MES 5–12 points (MEQ) HypertensionDiabetesMyocardial infarctionStroke 1.3 (1.0, 1.8)2.6 (1.5, 4.4)0.8 (0.3, 2.1)1.4 (0.7, 2.8) Sex, age, education level, civil status High
Ramin et al. (47) USA F 52.9 Nurses 647/25,038 Question no. 19 of MEQ 238/8542 Question no. 19 of MEQ Breast cancer 1.15 (0.98, 1.34) Age, family history of breast cancer, age at menarche, history of rotating night-shift work, smoke, BMI, alcohol, history of benign breast disease, oral contraceptive use, menopausal status, age at menopause, parity, and age at first birth and postmenopausal hormone use High
Yu et al. (5) Republic of Korea M/F 47–59 General population 97/480148/480111/480 MES > 59 points (MEQ) 27/9522/9529/95 MES < 41 points (MEQ) HypertensionDiabetesMetS 0.68 (0.40, 1.13)1.73 (1.01, 2.95)1.74 (1.05, 2.87)3 Age, sex, BMI, smoking, alcohol, exercise, occupation, sleep duration, use of antihypertensive, antidiabetic, or antilipid drugs High
Antypa et al. (37) Netherlands M/F 19–68 Depressed patients 139/409 1st quintile of MSF (MCTQ) 171/387 5th quintile of MSF (MCTQ) Depression 1.54 (1.15, 2.05) — Medium
Basnet et al. (24) Finland M/F 51.6 ± 13.8 General population NA/1935NA/1935NA/1935NA/1935 MES 19–27 points (MEQ) NA/595NA/595NA/595NA/595 MES 5–12 points (MEQ) HypertensionDiabetesCancerDepression 1.05 (0.72, 1.51)1.01 (0.51, 1.98)1.51 (0.53, 4.33)2.43 (1.52, 3.90) Age, BMI, sex, marital status, education, region, smoking, alcohol intake, physical activity High
Knutson and Schantz (10) United Kingdom M/F 38–73 General population 6330/117,22445,014/117,224 Question no. 19 of MEQ 2417/38,86713,309/38,867 Question no. 19 of MEQ DiabetesCVD 1.30 (1.24, 1.36)1.07 (1.04, 1.10) Age, sex High
Hurley et al. (48) USA F 40–89 Postmenopausal women 955/15,175 5th quintile of rMEQ 392/5027 1st quintile of rMEQ Breast cancer 1.20 (1.06, 1.35) Age, race, family history of breast cancer, age at menarche, smoking pack-years, BMI, alcohol consumption, physical activity, age at first full-term pregnancy, breastfeeding history, age at menopause, ever use of hormone therapy Medium
Kwon et al. (8) Republic of Korea M/F 19–81 General population 38/14510/145 MES > 59 points (MEQ) 31/14510/145 MES < 41 points (MEQ) HypertensionDiabetes 0.77 (0.44, 1.32)1.00 (0.40, 2.48) Age, sex, BMI High
Tan et al. (49) Sweden M/F 40–69 General population NA/79,955 Question no. 19 of MEQ NA/26,016 Question no. 19 of MEQ Diabetes 1.25 (1.17, 1.33) Age, sex, self-reported chronotype, self-reported sleep duration, insomnia, BMI, systolic BP, smoking, alcohol intake, test center, principal components of ancestry, Townsend index Medium
Open in a new tab
1

AMI, acute myocardial infarction; BP, blood pressure; CVD, cardiovascular disease; MCTQ, Munich Chronotype Questionnaire; MEQ, Morningness-Eveningness Questionnaire; MES, Morningness–Eveningness Score calculated from the Morningness-Eveningness Questionnaire items; MetS, metabolic syndrome; NA, not available.2Values are reported as mean ± SD.3Not adjusted.

TABLE 3.

Characteristics of prospective cohort studies evaluating chronotype and diseases1

Study Country (cohort, length of follow-up) Sex Age, y2 Study population Morning type, n/total n Definition of morning type Evening type, n/total n Definition of evening type Outcome HR (95% CI) Adjustment Study quality
Vetter et al. (36) USA (NHSII, 6 y) F 25–42 Nurses 93/22,089 Question no. 19 of MEQ 49/7029 Question no. 19 of MEQ Diabetes 1.01 (0.73, 1.38) Age, family history of diabetes, smoking, alcohol intake, physical activity, diet score, oral contraceptive use, menopausal status, postmenopausal hormone use, sleep duration, median annual household income, depressive symptoms, cumulative rotating night-shift work exposure since 1989, BMI Medium
Dickerman et al. (38) Finland (OFTC, 30 y) M 40 ± 12.1 Twins 208/3159 Question no. 19 of MEQ 181/1117 Question no. 19 of MEQ Prostate cancer incidence 1.3 (1.1, 1.6) Age, education, BMI, physical activity, social class, smoking status, alcohol use, snoring, zygosity High
Open in a new tab
1

MEQ, Morningness-Eveningness Questionnaire; NHSII, Nurses’ Health Study II; OFTC, Older Finnish Twin Cohort.2Values are reported as mean ± SD.

Overall, 18 studies were conducted in Europe (of which 8 were in Scandinavian countries), 12 studies in Asia, and 9 studies in the United States. Most of the studies (n = 37; 95%) used the MEQ (in its full or reduced form) to assess chronotype, whereas only 2 studies used the Munich Chronotype Questionnaire (MCTQ). Six studies (15%) were conducted only in women and 2 (5%) only in men. Nine studies (22.5%) were conducted on subjects with a clinical diagnosis of disease. Regarding adjustment for possible confounders, only 10 studies (25.6%) reported adjusted results, and in some cases only few factors were considered. Based on the NOS assessment, 9 studies (23.1%) were of high quality, 25 (64.1%) of medium quality, and 5 (12.8%) of low quality. Supplemental Tables 3 and 4 give a detailed description of the quality assessment of the cross-sectional and prospective cohort studies, respectively.

Chronotype, energy intake, and cardiometabolic risk factors

Table 1 shows the characteristics of cross-sectional studies that investigated the association between chronotype, energy intake, and cardiometabolic risk factors. In particular, the following outcomes were examined: energy intake (n = 16), body weight (n = 7), BMI (n = 33), fat mass percentage (n = 5), systolic blood pressure (n = 9), diastolic blood pressure (n = 9), fasting blood glucose (n = 8), HbA1c (n = 7), insulin (n = 2), HOMA-IR (n = 2), total cholesterol (n = 6), HDL cholesterol (n = 9), LDL cholesterol (n = 6), and triglycerides (n = 8).

Figure 2 shows the forest plot of cross-sectional studies summarizing the association between chronotype, energy intake, and cardiometabolic risk factors. By comparing evening and morning subjects, pooled analysis of 8 studies (5, 7, 8, 18, 19, 21, 28, 42) showed significantly higher concentrations of fasting blood glucose in evening subjects than in morning subjects (MD: 7.82; 95% CI: 3.18, 12.45), with substantial heterogeneity between studies (I2 = 94%, P < 0.00001). Another significant difference between groups was observed for HbA1c, where the pooled analysis of 7 studies (1, 5, 8, 17, 19, 42, 43) showed significantly higher concentrations of HbA1c in evening subjects than in morning subjects (MD: 7.64; 95% CI: 3.08, 12.21). Again, heterogeneity was high (I2 = 98%, P < 0.00001). Finally, evening subjects reported significantly higher concentrations than morning subjects of LDL cholesterol (MD: 13.69; 95% CI: 6.84, 20.54) and triglycerides (MD: 12.62; 95% CI: 0.90, 24.35) in 6 (8, 18, 21, 17, 30, 40) and 8 studies (5, 7, 8, 17–19, 21, 30), respectively, with substantial heterogeneity between studies in both cases (I2 = 85%; P < 0.00001 and I2 = 100%; P < 0.00001, respectively). No significant differences between the groups were reported for energy intake, anthropometric parameters, blood pressure, insulin, HOMA-IR, total cholesterol, and HDL cholesterol.

FIGURE 2.

FIGURE 2

Open in a new tab

Forest plot summary of the association between chronotype, energy intake, and cardiometabolic risk factors assessed in cross-sectional studies. MD, mean difference; P-het, probability of the null hypothesis that there is no heterogeneity between studies.

Chronotype and risk of diseases

Table 2 summarizes the characteristics of cross-sectional studies that investigated the association between chronotype and the risk of hypertension (n = 5), diabetes (n = 7), cancer (n = 3), depression (n = 2), cardiovascular disease (n = 1), myocardial infarction (n = 1), stroke (n = 1), and metabolic syndrome (n = 1). As depicted in Figure 3, meta-analytic pooling under a random-effects model showed an increased risk of diabetes (OR: 1.30; 95% CI: 1.20, 1.41; I2 = 39%; P = 0.13), cancer (OR: 1.18; 95% CI: 1.08, 1.30; I2 = 0%; P = 0.82), and depression (OR: 1.86; 95% CI: 1.20, 2.88; I2 = 62%; P = 0.11) among evening subjects, with nonsignificant (P > 0.05) heterogeneity between studies. In contrast, no association with hypertension emerged.

FIGURE 3.

FIGURE 3

Open in a new tab

Forest plot summary of the association between chronotype and disease risk assessed in cross-sectional studies. P-het, probability of the null hypothesis that there is no heterogeneity between studies.

Subgroup analyses and publication bias

To examine the potential sources of heterogeneity, subgroup analyses were conducted (Table 4). For fasting blood glucose and HbA1c, the heterogeneity disappeared (I2 = 6%; P = 0.37 and I2 = 0%; P = 0.99, respectively) when populations with a clinical diagnosis of disease, which were also the studies reporting the greatest difference between evening and morning subjects, were excluded. The opposite was observed for triglycerides, where the heterogeneity disappeared (I2 = 0%; P = 0.96) when only patients were considered. A significant difference according to geographical region was observed for LDL cholesterol and triglycerides, with no heterogeneity among studies conducted in Southern countries (I2 = 0%; P = 0.53 and I2 = 0%; P = 0.43, respectively). Finally, heterogeneity was reduced to 9% (P = 0.30) for blood glucose when only medium-quality studies were considered, and it disappeared for blood glucose and HbA1c when only high-quality studies were considered, suggesting that the quality of the original studies may influence the results.

TABLE 4.

Subgroup analyses1

Blood glucose HbA1c LDL cholesterol Triglycerides
n MD (95% CI) I 2 (P-het) n MD (95% CI) I 2 (P-het) n MD (95% CI) I 2 (P-het) n MD (95% CI) I 2 (P-het)
Geographical region
 Northern 4 2.25 (−0.39, 4.88) 66% (0.03) 4 8.37 (2.34, 14.40) 94% (<0.00001) 4 16.22 (8.19, 24.26) 88% (<0.00001) 5 7.54 (−3.37, 18.45) 99% (<0.00001)
 Southern 4 14.58 (2.14, 27.02) 97% (<0.00001) 3 6.47 (1.08, 11.86) 97% (<0.00001) 2 8.40 (1.44, 15.36) 0% (0.53) 3 26.99 (26.55, 27.43) 0% (0.43)
Study population
 Clinically healthy subjects 5 1.01 (−0.04, 2.05) 6% (0.37) 2 1.00 (0.63, 1.37) 0% (0.99) 4 12.80 (5.22, 20.37) 90% (<0.00001) 5 10.68 (−2.96, 24.33) 100% (<0.00001)
 Patients2 3 22.88 (9.85, 35.92) 87% (0.0004) 5 10.77 (4.87, 16.68) 94% (<0.00001) 2 16.40 (−9.70, 42.52) 75% (0.04) 3 18.02 (13.71, 22.33) 0% (0.96)
Study quality
 Low (≤3 points, NOS) 1 31.80 (20.58, 43.02) — 3 12.22 (3.46, 20.98) 96% (<0.00001) — —
 Medium (4–6 points, NOS) 5 7.91 (1.33, 14.49) 95% (<0.00001) 2 6.50 (3.25, 9.75) 9% (0.30) 5 14.84 (7.11, 22.56) 85% (<0.0001) 5 9.62 (−6.75, 26.00) 97% (<0.0001)
 High (7–9 points, NOS) 2 1.43 (−0.67, 3.53) 0% (0.43) 2 1.00 (0.63, 1.37) 0% (0.99) 1 9.20 (1.80, 16.60) — 3 16.57 (−2.08, 35.21) 100% (<0.00001)
Open in a new tab
1

HbA1c, glycated hemoglobin; I2, magnitude of heterogeneity; MD, mean difference; n, number of studies; NOS, Newcastle-Ottawa Scale; P-het, probability of the null hypothesis that there is no heterogeneity between studies.

2

Subjects with a clinical diagnosis of disease (diabetes, inflammatory bowel disease, acute myocardial infarction).

To confirm that our findings were not driven by any single study, a leave-one-out sensitivity analysis was performed. In this case, little change in the quantitative summary measures of MD or OR with the 95% CI was reported, with no study influencing results for all outcomes. Publication bias was assessed for energy intake and BMI, the outcomes assessed in >10 studies, using the funnel plot (Supplemental Figure 1). For both outcomes, the shape of the funnel plot showed little asymmetry, suggesting little evidence of publication bias.

Discussion

To our knowledge, this is the first systematic review with meta-analysis including all available cross-sectional and prospective cohort studies estimating the association between chronotype, energy intake, and multiple health outcomes. The overall analysis included 39 studies with a total of 377,797 subjects. Evening subjects were found to be associated with a worse cardiometabolic profile. In fact, they showed significantly higher concentrations of fasting blood glucose, HbA1c, LDL cholesterol, and triglycerides than morning individuals, and a significantly higher risk of diabetes, cancer, and depression.

Because a wide range of physiological and metabolic functions are set and programmed by the time of day, the interaction between circadian rhythms, food intake, and health status has been increasingly studied in recent years. Circadian rhythms are cyclical endogenous processes that occur with a periodicity of ∼24 h and play an important role in regulating sleep/wake cycles, metabolism, hormonal secretions, immune function, and cell cycle control (47, 49). Despite the regulation of the master circadian clock, humans living in modern industrialized societies often engage in behaviors that are inappropriately timed relative to their endogenous circadian system, or chronotype (19). The 24-h access to light, irregular eating patterns, and social rhythms imposed by professional obligations and school schedules can result in a timing mismatch, defined “circadian misalignment” or “chrono-disruption” (50). As reported by a growing body of evidence, this misalignment can disrupt the natural oscillations of physiologic processes such as the regulation of blood pressure and glycemic and lipid metabolism, resulting in an increased risk of obesity and chronic degenerative diseases (6). Furthermore, it seems that the evening chronotype is associated with a higher risk of chrono-disruption and consequently with developing pathological conditions (40), probably because the circadian phase of such individuals is shifted by as much as 2–3 h (6).

In this meta-analysis, both daily energy intake and anthropometric parameters were evaluated. Previous studies have suggested that evening subjects consume more calories during the day (18, 21, 29, 51) and have a lower-quality diet (4, 22, 52) than morning subjects. One possible explanation for this behavior is that disruption of the circadian system affects appetite, energy expenditure, and several determinants of obesity (53). However, no significant differences in energy intake, body weight, BMI, and fat mass percentage were observed between the 2 chronotypes in our analysis, probably owing to the high heterogeneity in terms of study population and sample size of the studies included in the meta-analysis.

On the other hand, significantly higher concentrations of fasting blood glucose, HbA1c, LDL cholesterol, and triglycerides were reported in evening subjects than in morning subjects. From a chronobiological perspective, glucose metabolism in humans follows a circadian rhythm through diurnal variation in glucose tolerance, which typically peaks during the daytime hours, when food consumption usually occurs, and declines during the nighttime hours, when fasting usually occurs (54). As previously reported, evening subjects tend to eat later than morning ones (1, 18) and this may be associated with poorer glycemic control and increased risk of type 2 diabetes (1, 9). In addition, it has been reported that plasma triglyceride concentrations are elevated during the biological nighttime, and that the postprandial response after a night meal is amplified compared with the same meal consumed during the day (55). Our results reinforce this hypothesis, because both glycemic and lipid profiles were worse in evening subjects.

Consistent with a worse cardiometabolic risk profile, evening subjects also reported an increased risk of developing diseases such as diabetes, cancer, and depression. Regarding cancer, evidence suggests that the direct disruption of the functions of circadian clock genes that control cell proliferation, or disruptions of clock-controlled settings such as sleep disturbances, may increase the risk (56, 57). Another hypothesis, called light-at-night, places the hormone melatonin at the center of the cancer disease process. It argues that melatonin is a major scavenger of reactive oxygen species (1). Because melatonin is primarily produced at night and suppressed by light, oncogenesis becomes more likely when people are exposed to light at night (1). Regarding depression, our results are in line with a previous meta-analysis that investigated the relation between chronotype and mood disorders such as depression, bipolar, and seasonal disorders (2). Indeed, evening subjects seem to be potentially associated with major depressive disorder (1, 2, 24, 37, 58), likely owing to a change in the rhythmic activity of neurotransmitter systems involved in mood regulation, including dopamine and serotonin secretion (58).

There are some limitations that should be discussed. First, in the included studies, chronotypes were assessed through self-reported questionnaires (the MEQ, reduced MEQ, the 19th question of the MEQ, and MCTQ), which are generally accepted, but are susceptible to bias. To date, the most reliable circadian phase marker in humans is the dim light melatonin onset, but the cost of this test is relatively high to allow its use in epidemiologic studies. Other reliable and valid methods to detect the circadian rhythm in humans are polysomnography or actigraphy, but their use in epidemiologic studies also has limitations owing to high costs and the specific expertise required. To consolidate our findings and to better examine how chronotype may affect the association between chrono-nutrition and health, further research on the best methods to assess chronotype is required. Second, the overall analysis for several outcomes such as insulin, HOMA-IR, cancer, and diabetes was performed in a limited number of studies, thus reducing the statistical power of the analysis. Furthermore, the possibility of publication bias could only be assessed for the 2 outcomes with >10 available studies. In this regard, publication bias is known to be a major threat to the validity of meta-analyses. Indeed, a higher probability of including statistically significant positive results generally causes an increase in the false-positive rate. Therefore, further well-conducted studies are needed to confirm these results and to better elucidate the interplay between chronotype, nutrition, and health status. Third, most of the included studies were cross-sectional. Although useful for identifying associations, these studies cannot infer causality or analyze behavior over time. Additional longitudinal or experimental studies are needed to investigate a possible cause-and-effect relation. Fourth, most of the included studies did not account for confounders. Given that previous researchers have suggested that morning people tend to be more persistent, cooperative, conscientious, proactive, and less likely to procrastinate than evening people (59, 60), we cannot rule out that chronotype differences are due to subject characteristics and behaviors that lead to the morning instead of the evening state rather than a disruption of circadian rhythms. Finally, the high heterogeneity among the studies in terms of country, population, sample size, and chronotype assessment introduces a limitation in the interpretation of the results. Despite all these limitations, however, our study has several strengths, such as a rigorous search and selection strategy that identified all available cross-sectional and prospective cohort studies examining the relation between chronotype and health status, and the fact that most of the included studies were of good methodological quality.

In conclusion, we report, to our knowledge for the first time in a systematic review with meta-analysis, the possible association between evening chronotype, worse cardiometabolic risk profile, and increased risk of cardiovascular diseases, diabetes, cancer, and depression. The limitations of the available literature and the methods used to define individual chronotype, which need improving, reduce the applicability of these results to the general population. In addition, the underlying biological mechanisms that explain the link between chronotype, dietary habits, and health status need to be better understood and future experimental designs capable of drawing causal inferences are needed. Further research is also needed to better understand how to effectively apply the concept of chrono-nutrition in communities and in clinical practice. Nonetheless, these results provide a greater understanding of the relation between chronotype, diet, and health, and contribute to developing chronobiological approaches for the prevention and treatment of cardiometabolic and chronic disorders.

Supplementary Material

nmab115_Supplemental_File
Click here for additional data file. (33.4KB, zip)

ACKNOWLEDGEMENTS

The authors’ responsibilities were as follows—MD and FS: designed the research; MD and SL: conducted the systematic literature search, performed the quality assessment and data extraction, and wrote the paper; SL: performed the statistical analysis; FS: critically reviewed the manuscript and had primary responsibility for the final content; and all authors: contributed to writing and reviewing the manuscript and read and approved the final manuscript.

Notes

The authors reported no funding received for this study.

Author disclosures: The authors report no conflicts of interest.

Supplemental Figure 1 and Supplemental Tables 1–4 are available from the “Supplementary data” link in the online posting of the article and from the same link in the online table of contents at https://academic.oup.com/advances/.

Abbreviations used: HbA1c, glycated hemoglobin; MCTQ, Munich Chronotype Questionnaire; MD, mean difference; MEQ, Morningness-Eveningness Questionnaire; NOS, Newcastle-Ottawa Scale; PRISMA, Preferred Reporting Items for Systematic Reviews and Meta-Analyses.

Contributor Information

Sofia Lotti, Department of Experimental and Clinical Medicine, University of Florence, Florence, Italy.

Giuditta Pagliai, Department of Experimental and Clinical Medicine, University of Florence, Florence, Italy.

Barbara Colombini, Department of Experimental and Clinical Medicine, University of Florence, Florence, Italy.

Francesco Sofi, Department of Experimental and Clinical Medicine, University of Florence, Florence, Italy; Unit of Clinical Nutrition, Careggi University Hospital, Florence, Italy.

Monica Dinu, Department of Experimental and Clinical Medicine, University of Florence, Florence, Italy.

References

  • 1. Roenneberg  T, Merrow  M. The circadian clock and human health. Curr Biol. 2016;26(10):R432–43. [DOI] [PubMed] [Google Scholar]
  • 2. Adan  A, Archer  SN, Hidalgo  MP, Di Milia  L, Natale  V, Randler  C. Circadian typology: a comprehensive review. Chronobiol Int. 2012;29(9):1153–75. [DOI] [PubMed] [Google Scholar]
  • 3. Horne  JA, Östberg  O. A self-assessment questionnaire to determine morningness-eveningness in human circadian rhythms. Int J Chronobiol. 1976;4(2):97–110. [PubMed] [Google Scholar]
  • 4. Kanerva  N, Kronholm  E, Partonen  T, Ovaskainen  ML, Kaartinen  NE, Konttinen  H, Broms  U, Männistö  S. Tendency toward eveningness is associated with unhealthy dietary habits. Chronobiol Int. 2012;29(7):920–7. [DOI] [PubMed] [Google Scholar]
  • 5. Yu  JH, Yun  C-H, Ahn  JH, Suh  S, Cho  HJ, Lee  SK, Yoo  HJ, Seo  JA, Kim  SG, Choi  KM  et al.  Evening chronotype is associated with metabolic disorders and body composition in middle-aged adults. J Clin Endocrinol Metab. 2015;100(4):1494–502. [DOI] [PubMed] [Google Scholar]
  • 6. Almoosawi  S, Vingeliene  S, Gachon  F, Voortman  T, Palla  L, Johnston  JD, Van Dam  RM, Darimont  C, Karagounis  LG. Chronotype: implications for epidemiologic studies on chrono-nutrition and cardiometabolic health. Adv Nutr. 2019;10(1):30–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7. Vera  B, Dashti  HS, Gómez-Abellán  P, Hernández-Martínez  AM, Esteban  A, Scheer  F, Saxena  R, Garaulet  M. Modifiable lifestyle behaviors, but not a genetic risk score, associate with metabolic syndrome in evening chronotypes. Sci Rep. 2018;8(1):945. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8. Kwon  Y-J, Chung  T-H, Lee  HS, Park  J, Chung  J-Y, Lee  B-K, Lee  J-W. Association between circadian preference and blood lipid levels using a 1:1:1 propensity score matching analysis. J Clin Lipidol. 2019;13(4):645–53.e2. [DOI] [PubMed] [Google Scholar]
  • 9. Merikanto  I, Lahti  T, Puolijoki  H, Vanhala  M, Peltonen  M, Laatikainen  T, Vartiainen  E, Salomaa  V, Kronholm  E, Partonen  T. Associations of chronotype and sleep with cardiovascular diseases and type 2 diabetes. Chronobiol Int. 2013;30(4):470–7. [DOI] [PubMed] [Google Scholar]
  • 10. Knutson  KL, Schantz  M. Associations between chronotype, morbidity and mortality in the UK Biobank cohort. Chronobiol Int. 2018;35:1–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11. Drennan  MD, Klauber  MR, Kripke  DF, Goyette  LM. The effects of depression and age on the Horne-Ostberg morningness-eveningness score. J Affect Disord. 1991;23(2):93–8. [DOI] [PubMed] [Google Scholar]
  • 12. Page  MJ, McKenzie  JE, Bossuyt  PM, Boutron  I, Hoffmann  TC, Mulrow  CD, Shamseer  L, Tetzlaff  JM, Akl  EA, Brennan  SE  et al.  The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021;372:n71. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13. Wells  GA, Shea  B, O'Connell  D, Peterson  J, Welch  V, Losos  M, Tugwell  P. The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. [Internet]. Ottawa, Ontario: Ottawa Hospital Research Institute. Available from: http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp [Accessed 22 April 2021]. [Google Scholar]
  • 14. Higgins  JP, Thompson  SG, Deeks  JJ, Altman  DG. Measuring inconsistency in meta-analyses. BMJ. 2003;327(7414):557–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15. Higgins  JPT, Green  S. Cochrane handbook for systematic reviews of interventions version 5.1.0[Internet]. Oxford, United Kingdom: The Cochrane Collaboration; 2011. Available from: http://handbook-5-1.cochrane.org/ [Accessed 26 April 2021]. [Google Scholar]
  • 16. Brandt  W. North-South: a programme for survival. Report of the Independent Commission on International Development Issues. London, United Kingdom: Pan Books; 1980. [Google Scholar]
  • 17. Iwasaki  M, Hirose  T, Mita  T, Sato  F, Ito  C, Yamamoto  R, Someya  Y, Yoshihara  T, Tamura  Y, Kanazawa  A  et al.  Morningness-eveningness questionnaire score correlates with glycated hemoglobin in middle-aged male workers with type 2 diabetes mellitus. J Diabetes Investig. 2013;4(4):376–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Lucassen  EA, Zhao  X, Rother  KI, Mattingly  MS, Courville  AB, De Jonge  L, Csako  G, Cizza  G, for the Sleep Extension Study Group . Evening chronotype is associated with changes in eating behavior, more sleep apnea, and increased stress hormones in short sleeping obese individuals. PLoS One. 2013;8(3):e56519. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19. Reutrakul  S, Hood  MM, Crowley  SJ, Morgan  MK, Teodori  M, Knutson  KL, Van Cauter  E. Chronotype is independently associated with glycemic control in type 2 diabetes. Diabetes Care. 2013;36(9):2523–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. Osonoi  Y, Mita  T, Osonoi  T, Saito  M, Tamasawa  A, Nakayama  S, Someya  Y, Ishida  H, Kanazawa  A, Gosho  M  et al.  Morningness-eveningness questionnaire score and metabolic parameters in patients with type 2 diabetes mellitus. Chronobiol Int. 2014;31(9):1017–23. [DOI] [PubMed] [Google Scholar]
  • 21. Maukonen  M, Kanerva  N, Partonen  T, Kronholm  E, Konttinen  H, Wennman  H, Männistö  S. The associations between chronotype, a healthy diet and obesity. Chronobiol Int. 2016;33(8):972–81. [DOI] [PubMed] [Google Scholar]
  • 22. Munoz  JSG, Cañavate  R, Hernández  CM, Cara-Salmerón  V, Morante  JJH. The association among chronotype, timing of food intake and food preferences depends on body mass status. Eur J Clin Nutr. 2017;71(6):736–42. [DOI] [PubMed] [Google Scholar]
  • 23. Ruiz-Lozano  T, Vidal  J, de Hollanda  A, Canteras  M, Garaulet  M, Izquierdo-Pulido  M. Evening chronotype associates with obesity in severely obese subjects: interaction with CLOCK 3111T/C. Int J Obes. 2016;40(10):1550–7. [DOI] [PubMed] [Google Scholar]
  • 24. Basnet  S, Merikanto  I, Lahti  T, Männistö  S, Laatikainen  T, Vartiainen  E, Partonen  T. Associations of common noncommunicable medical conditions and chronic diseases with chronotype in a population-based health examination study. Chronobiol Int. 2017;34(4):462–70. [DOI] [PubMed] [Google Scholar]
  • 25. Maukonen  M, Kanerva  N, Partonen  T, Kronholm  E, Tapanainen  H, Kontto  J, Männistö  S. Chronotype differences in timing of energy and macronutrient intakes: a population-based study in adults. Obesity. 2017;25(3):608–15. [DOI] [PubMed] [Google Scholar]
  • 26. Teixeira  GP, Mota  MC, Crispim  CA. Eveningness is associated with skipping breakfast and poor nutritional intake in Brazilian undergraduate students. Chronobiol Int. 2018;35(3):358–67. [DOI] [PubMed] [Google Scholar]
  • 27. Yoshizaki  T, Komatsu  T, Tada  Y, Hida  A, Kawano  Y, Togo  F. Association of habitual dietary intake with morningness-eveningness and rotating shift work in Japanese female nurses. Chronobiol Int. 2018;35(3):392–404. [DOI] [PubMed] [Google Scholar]
  • 28. Yazdinezhad  A, Askarpour  M, Aboushamsia  MM, Asadi  M, Mansoori  A. Evaluating the effect of chronotype on meal timing and obesity in Iranian housewives: a cross-sectional study. J Adv Med Biomed Res. 2019;27(124):31–6. [Google Scholar]
  • 29. Kayacan  AG, Tokay  A. Evaluation of the relationship between chronotype and biochemical findings, nutrition and gastrointestinal symptoms in inflammatory bowel patients. Sleep Med. 2021;81:358–64. [DOI] [PubMed] [Google Scholar]
  • 30. Gaspar-Barba  E, Calati  R, Cruz-Fuentes  CS, Ontiveros-Uribe  MP, Natale  V, De Ronchi  D, Serretti  A. Depressive symptomatology is influenced by chronotypes. J Affect Disord. 2009;119(1–3):100–6. [DOI] [PubMed] [Google Scholar]
  • 31. Selvi  Y, Smolensky  MH, Boysan  M, Aydin  A, Besiroglu  L, Atli  A, Gumrukcuoglu  HA. Role of patient chronotype on circadian pattern of myocardial infarction: a pilot study. Chronobiol Int. 2011;28(4):371–7. [DOI] [PubMed] [Google Scholar]
  • 32. Meule  A, Roeser  K, Randler  C, Kübler  A. Skipping breakfast: morningness-eveningness preference is differentially related to state and trait food cravings. Eat Weight Disord. 2012;17(4):e304–8. [DOI] [PubMed] [Google Scholar]
  • 33. Roeser  K, Obergfell  F, Meule  A, Vögele  C, Schlarb  AA, Kübler  A. Of larks and hearts—morningness/eveningness, heart rate variability and cardiovascular stress response at different times of day. Physiol Behav. 2012;106(2):151–7. [DOI] [PubMed] [Google Scholar]
  • 34. Kim  LJ, Coelho  FM, Hirotsu  C, Bittencourt  L, Tufik  S, Andersen  ML. Is the chronotype associated with obstructive sleep apnea?. Sleep Breath. 2015;19(2):645–51. [DOI] [PubMed] [Google Scholar]
  • 35. Merikanto  I, Kronholm  E, Peltonen  M, Laatikainen  T, Vartiainen  E, Partonen  T. Circadian preference links to depression in general adult population. J Affect Disord. 2015;188:143–8. [DOI] [PubMed] [Google Scholar]
  • 36. Vetter  C, Devore  EE, Ramin  CA, Speizer  FE, Willett  WC, Schernhammer  ES. Mismatch of sleep and work timing and risk of type 2 diabetes. Diabetes Care. 2015;38(9):1707–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37. Antypa  N, Vogelzangs  N, Meesters  Y, Schoevers  R, Penninx  BW. Chronotype associations with depression and anxiety disorders in a large cohort study. Depress Anxiety. 2016;33(1):75–83. [DOI] [PubMed] [Google Scholar]
  • 38. Dickerman  BA, Markt  SC, Koskenvuo  M, Hublin  C, Pukkala  E, Mucci  LA, Kaprio  J. Sleep disruption, chronotype, shift work, and prostate cancer risk and mortality: a 30-year prospective cohort study of Finnish twins. Cancer Causes Control. 2016;27(11):1361–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39. Suh  S, Yang  H-C, Kim  N, Yu  JH, Choi  S, Yun  C-H, Shin  C. Chronotype differences in health behaviors and health-related quality of life: a population-based study among aged and older adults. Behav Sleep Med. 2017;15(5):361–76. [DOI] [PubMed] [Google Scholar]
  • 40. Sun  X, Gustat  J, Bertisch  SM, Redline  S, Bazzano  L. The association between sleep chronotype and obesity among black and white participants of the Bogalusa Heart Study. Chronobiol Int. 2020;37(1):123–34. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41. Tomizawa  A, Nogawa  K, Watanabe  Y, Oishi  M, Tanaka  K, Sakata  K, Suwazono  Y. Effect of circadian rhythm type on serum lipid levels in shift workers: a 5-year cohort study. Chronobiol Int. 2019;36(6):751–7. [DOI] [PubMed] [Google Scholar]
  • 42. De Amicis  R, Galasso  L, Leone  A, Vignati  L, De Carlo  G, Foppiani  A, Montaruli  A, Roveda  E, Cè  E, Esposito  F  et al.  Is abdominal fat distribution associated with chronotype in adults independently of lifestyle factors?. Nutrients. 2020;12(3):592. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43. Hashemipour  S, Yazdi  Z, Mahabad  N. Association of evening chronotype with poor control of type 2 diabetes: roles of sleep duration and insomnia level. Int J Endocrinol Metab. 2020;18(3):e99701. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44. Henson  J, Rowlands  AV, Baldry  E, Brady  EM, Davies  MJ, Edwardson  CL, Yates  T, Hall  AP, CODEC Investigators . Physical behaviors and chronotype in people with type 2 diabetes. BMJ Open Diabetes Res Care. 2020;8(1):e001375. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45. Muscogiuri  G, Barrea  L, Aprano  S, Framondi  L, Di Matteo  R, Laudisio  D, Pugliese  G, Savastano  S, Colao  A. On Behalf Of The Opera Prevention Project. Chronotype and Adherence to the Mediterranean Diet in Obesity: Results from the Opera Prevention Project. Nutrients. 2020;12(5):1354. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46. Thapa  N, Kim  B, Yang  J-G, Park  H-J, Jang  M, Son  H-E, Kim  G-M, Park  H. The relationship between chronotype, physical activity and the estimated risk of dementia in community-dwelling older adults. Int J Environ Res Public Health. 2020;17(10):3701. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47. Ramin  C, Devore  EE, Pierre-Paul  J, Duffy  JF, Hankinson  SE, Schernhammer  ES. Chronotype and breast cancer risk in a cohort of US nurses. Chronobiol Int. 2013;30(9):1181–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48. Hurley  S, Goldberg  D, Von Behren  J, DeHart  JC, Wang  S, Reynolds  P. Chronotype and postmenopausal breast cancer risk among women in the California Teachers Study. Chronobiol Int. 2019;36(11):1504–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49. Tan  X, Ciuculete  DM, Schiöth  HB, Benedict  C. Associations between chronotype, MTNR1B genotype and risk of type 2 diabetes in UK Biobank. J Intern Med. 2020;287(2):189–96. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50. Erren  TC, Reiter  RJ. Defining chronodisruption. J Pineal Res. 2009;46(3):245–7. [DOI] [PubMed] [Google Scholar]
  • 51. Mota  MC, Waterhouse  J, De-Souza  DA, Rossato  LT, Silva  CM, Araújo  MBJ, Tufik  S, De Mello  MC, Crispim  CA. Association between chronotype, food intake and physical activity in medical residents. Chronobiol Int. 2016;33(6):730–9. [DOI] [PubMed] [Google Scholar]
  • 52. Patterson  F, Malone  SK, Lozano  A, Grandner  MA, Hanlon  AL. Smoking, screen-based sedentary behavior, and diet associated with habitual sleep duration and chronotype: data from the UK Biobank. Ann Behav Med. 2016;50(5):715–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53. Wittmann  M, Dinich  J, Merrow  M, Roenneberg  T. Social jetlag: misalignment of biological and social time. Chronobiol Int. 2006;23(1–2):497–509. [DOI] [PubMed] [Google Scholar]
  • 54. Henry  CJ, Kaur  B, Quek  RYC. Chrononutrition in the management of diabetes. Nutr Diabetes. 2020;10(1):6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 55. Morgan  L, Hampton  S, Gibbs  M, Arendt  J. Circadian aspects of postprandial metabolism. Chronobiol Int. 2003;20(5):795–808. [DOI] [PubMed] [Google Scholar]
  • 56. Kolstad  HA. Nightshift work and risk of breast cancer and other cancers—a critical review of the epidemiologic evidence. Scand J Work Environ Health. 2008;34(1):5–22. [DOI] [PubMed] [Google Scholar]
  • 57. Fritschi  L, Glass  DC, Heyworth  JS, Aronson  K, Girschik  J, Boyle  T, Grundy  A, Erren  TC. Hypotheses for mechanisms linking shiftwork and cancer. Med Hypotheses. 2011;77(3):430–6. [DOI] [PubMed] [Google Scholar]
  • 58. Taillard  J, Sagaspe  P, Philip  P, Bioulac  S. Sleep timing, chronotype and social jetlag: impact on cognitive abilities and psychiatric disorders. Biochem Pharmacol. 2021;191:114438. [DOI] [PubMed] [Google Scholar]
  • 59. Ottoni  GL, Antoniolli  E, Lara  DR. Circadian preference is associated with emotional and affective temperaments. Chronobiol Int. 2012;29(6):786–93. [DOI] [PubMed] [Google Scholar]
  • 60. Antúnez  JM. Circadian typology is related to emotion regulation, metacognitive beliefs and assertiveness in healthy adults. PLoS One. 2020;15(3):e0230169. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 61. Fabbian  F, Zucchi  B, De Giorgi  A, Tiseo  R, Boari  B, Salmi  R, Cappadona  R, Bassi  E, Signani  F, Raparelli  V  et al.  Chronotype, gender and general health. Chronobiol Int. 2016;33(863):82. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

nmab115_Supplemental_File
Click here for additional data file. (33.4KB, zip)

Articles from Advances in Nutrition are provided here courtesy of American Society for Nutrition

RESOURCES