Abstract
Background
Eating a high-quality diet or adhering to a given dietary strategy may influence weight loss. However, these 2 factors have not been examined concurrently for those following macronutrient-limiting diets.
Objective
To determine whether improvement in dietary quality, change in dietary macronutrient composition, or the combination of these factors is associated with differential weight loss when following a healthy low-carbohydrate (HLC) or healthy low-fat (HLF) diet.
Design
Generally healthy adults were randomly assigned to HLC or HLF diets for 12 mo (n = 609) as part of a randomized controlled weight loss study. Participants with complete 24-h dietary recall data at baseline and 12-mo were included in this secondary analysis (total N = 448; N = 224 HLC, N = 224 HLF). Participants were divided into 4 subgroups according to 12-mo change in HEI-2010 score [above median = high quality (HQ) and below median = low quality (LQ)] and 12-mo change in macronutrient intake [below median = high adherence (HA) and above median = low adherence (LA) for net carbohydrate (g) or fat (g) for HLC and HLF, respectively]. Baseline to 12-mo changes in mean BMI were compared for those in HQ/HA, HQ/LA, LQ/HA subgroups with the LQ/LA subgroup within HLC and HLF.
Results
For HLC, changes (95 % confidence level [CI]) in mean BMI were -1.15 kg/m2 (-2.04, -0.26) for HQ/HA, -0.30 (-1.22, 0.61) for HQ/LA, and -0.80 (-1.74, 0.14) for LQ/HA compared with the LQ/LA subgroup. For HLF, changes (95% CI) in mean BMI were -1.11kg/m2 (-2.10, -0.11) for HQ/HA, -0.26 (-1.26, 0.75) for HQ/LA, and -0.66 (-1.74, 0.41) for LQ/HA compared with the LQ/LA subgroup.
Conclusion
Within both HLC and HLF diet arms, 12-mo decrease in BMI was significantly greater in HQ/HA subgroups relative to LQ/LA subgroups. Neither HQ nor HA alone were significantly different than LQ/LA subgroups. Results of this analysis support the combination of dietary adherence and high-quality diets for weight loss.
Clinical Trial Registry
clinicaltrials.gov (Identifier: NCT01826591).
Keywords: diet quality, low-carbohydrate, low-fat, weight loss, Healthy Eating Index
Introduction
According to the Centers for Disease Control National Center for Health Statistics, 42.4% of adults in the United States had obesity in 2017 and 2018, and it is predicted that by 2030, nearly 1 in 2 adults will have obesity [1, 2]. Obesity, dietary risks, hypertension, elevated fasting plasma glucose, and hyperlipidemia make up 5 of the top 6 risk factors for early mortality in the United States [3]. Dietary modification via a variety of healthy diet approaches has been shown to improve cardiometabolic risk factors and reduce weight [4, 5]. Therefore, it has long been a cornerstone of most successful weight loss strategies both in research and practice. Low-carbohydrate (LC) and low-fat diets (LF) are among the most studied approaches, but neither has been shown to be consistently superior for weight loss in the general population [[5], [6], [7], [8], [9], [10]]. Previous studies suggest that eating a high-quality diet or adhering to a given dietary strategy may influence weight loss; however, these 2 factors have not been examined concurrently for those following macronutrient-limiting diets.
Although evidence supports the use of LC or LF diets in weight loss efforts [5], the benefits of simultaneously focusing on achieving higher quality diets are unclear. To determine whether improvement in dietary quality, change in dietary macronutrient composition, or the combination of these factors associated with differential weight loss and change in cardiovascular disease risk factors, a post hoc secondary analysis was conducted using data from the Diet Intervention Examining The Factors Interacting with Treatment Success (DIETFITS) weight loss trial [11] in which randomly assigned participants follow a healthy low-carbohydrate (HLC) or healthy low-fat (HLF) diet for 12 mo. The dataset presented a unique opportunity because the intervention dually emphasized both reduction in macronutrient intake and improvement in dietary quality [12], whereas the majority of LC versus LF weight loss studies primarily focuses on macronutrient reduction [8, 10, [13], [14], [15], [16], [17], [18], [19]].
In this exploratory analysis, we aimed to determine whether those who most improved the quality of their diet, those who most decreased their net carbohydrate or fat intake, or both were more successful in their weight loss endeavors than those who did not improve their dietary quality nor reduce their assigned macronutrient intake to the same degree when assigned to follow HLC or HLF diets. Secondarily, we aimed to determine the impact of these factors on changes in cardiovascular disease risk factors, including blood pressure, fasting insulin, glucose, and lipids.
Methods
Study design
Our post hoc secondary analysis utilized data from the DIETFITS study, of which detailed methods have previously been published [12]. In brief, this was a single-site, parallel-group, randomized controlled weight loss trial of 609 free-living, generally healthy males and premenopausal females aged 18 to 50 y, with BMI from 28 to 40 kg/m2, who were randomly assigned to HLC or HLF diets for 12 mo to determine whether genetics (via 3-single nucleotide polymorphisms) or metabolic predispositions (insulin secretion) at baseline resulted in differential weight loss for those assigned to either diet. Randomization was performed using an allocation sequence determined by computerized random-number generation (Blockrand in R version 3.4.0, R Project for Statistical Computing) in block sizes of 8 (with 4 individuals going to each diet group) by a statistician not involved in intervention delivery or data collection. The trial enrollment was from January 29, 2013 through April 14, 2015, and the data of the final follow-up was from May 16, 2016. The study was conducted in the San Francisco Bay Area of California. Key exclusion criteria included—pregnancy or nursing, uncontrolled metabolic disease, diabetes, cancer, liver, kidney, or heart disease, and taking psychiatric medications or medications known to affect weight or energy expenditure, serum lipids, serum glucose, or blood pressure.
The DIETFITS trial was overseen by the Stanford University Institutional Review Board and was registered with clinicaltrials.gov (Identifier: NCT01826591). All study participants provided written informed consent. All data from the DIETFITS randomized trial were managed using the REDCap electronic data capture tool hosted at Stanford University [20].
After baseline data collection, participants’ random assignment to the HLC or HLF diet was revealed at their first intervention class. The intervention consisted of 22 sessions of class-based instruction conducted by registered dietitian nutritionists (RDNs) held over 12 mo within diet groups. Classes were conducted weekly for 8 wk, then every 2 wk for 2 mo, then every 3 wk until Month 6, and then monthly through Month 12. Dietary goals included reducing net carbohydrate or fat intake to 20 g or less for 8 wk, then slowly increasing over time until each participant found the level of intake they could sustain indefinitely, reducing intake of highly processed foods and added sugars, and maximizing intake of vegetables and other whole foods. There was no explicit energy restriction. Dietary data collection was conducted via 3 unannounced, 24-h dietary recalls using a standardized, multiple-pass approach [21] that included 2 weekdays and 1 weekend day at the data collection time points of baseline, 3, 6, and 12 mo.
For the current analysis, participants who provided complete 24-h dietary recall data at baseline and 12 mo were included (total N = 448; N = 224 HLC, N = 224 HLF) (Supplemental Figure 1). Diet quality scores were calculated from each dietary recall using the Healthy Eating Index-2010 (HEI-2010) [[22], [23], [24], [25]]. Then, scores from the 3 recalls from each time point for each participant were averaged to yield a single score per participant for each time point. Within each diet assignment arm, participants were divided into 4 subgroups according to 12-mo change in diet quality score [HEI-2010, above the median was defined as high quality (HQ); below the median was defined as low quality (LQ)] and 12-mo change in macronutrient intake [below the median was defined as high adherence (HA) and above the median was defined as low adherence (LA) for net carbohydrate (g) or fat (g) for HLC and HLF, respectively]. For the low-carbohydrate diet, the cutoff used for quality was 4.2 (median 12-mo change in HEI-2010), and the cutoff used for adherence was -103.2 (g) (median 12-mo change in carbohydrate). For the low-fat diet, the cutoff used for quality was 8.5 (median 12-mo HEI-2010), and the cutoff used for adherence was -29.7(g) (median 12-mo change in fat). Median values for 12-mo change in HEI-2010 and macronutrient intake by diet assignment are provided in Supplemental Tables 1 and 2. Therefore, the 4 subgroups for each diet were HQ/HA, HQ/LA, LQ/HA, and LQ/LA.
This study was a post hoc secondary analysis. The predefined primary outcome was 12-mo change in BMI. Predefined secondary outcomes included 12-mo change in the following cardiometabolic measures: systolic blood pressure (SBP), diastolic blood pressure (DBP), fasting plasma glucose, insulin, triglycerides, HDL cholesterol, and LDL cholesterol concentrations. The reference subgroup was LQ/LA. All clinical measurements (body weight, height, blood pressure, and a fasting blood draw, which would be analyzed for insulin, glucose, lipids, and triglycerides concentrations) were collected by the Stanford Clinical Translational Research Unit at each data collection time point [12].
Statistical analysis
Descriptive statistics were used to display baseline demographic, clinical characteristics, and dietary components for each subgroup (Table 1 and Table 2). Descriptive statistics, including mean and standard deviation, were calculated for each of the primary and secondary outcomes and dietary components for each subgroup for all participants with data available (Table 3 and Table 4). Two multiple linear regression models were used to compare the primary and secondary outcomes for those in the HQ/HA, HQ/LA, LQ/HA subgroups versus the LQ/LA subgroup within HLC and HLF diet groups (Table 5). Model 1 was adjusted for age, sex, and baseline weight, whereas Model 2 additionally adjusted for either baseline net carbohydrate or fat intake for the HLC and HLF diet groups, respectively. Model 2 represents the main analyses for study primary and secondary outcomes. To ensure results were not related to baseline energy intake or HEI-2010 total score, we also added these factors to Model 2 (Supplemental Table 5). No formal between-diet statistical comparisons were conducted, given likely bias of diet quality indices toward low-fat versus low-carbohydrate diets (manuscript concurrently under review), although confidence intervals are provided. There was no correction for multiple testing as this was an exploratory, post hoc secondary analysis.
TABLE 1.
Baseline characteristics by dietary quality and adherence to diet for those assigned to a healthy low-carbohydrate diet (n = 304).1,2,3
| Variable | High quality, high adherence (HQ/HA) N = 67 | High quality, low adherence (HQ/LA) N = 45 | Low quality, high adherence (LQ/HA) N = 45 | Low quality, low adherence (LQ/LA) N = 67 | Total N = 224 | Missing 12-mo dietary data N = 80 |
|---|---|---|---|---|---|---|
| Sex (N, %) | ||||||
| Female | 42 (62.7) | 27 (60.0) | 24 (53.3) | 39 (58.2) | 132 (58.9) | 47 (58.8) |
| Male | 25 (37.3) | 18 (40.0) | 21 (46.7) | 28 (41.8) | 92 (41.1) | 33 (41.2) |
| Age (y) | 41.22 (6.15) | 40.05 (6.58) | 41.33 (7.04) | 39.98 (6.80) | 40.6 (6.6) | 38.98 (6.97) |
| Education (N, %) | ||||||
| Less than high school | 0 (0.0) | 2 (4.4) | 0 (0.0) | 1 (1.5) | 3 (1.3) | 1 (1.2) |
| High school graduate | 0 (0.0) | 1 (2.2) | 1 (2.2) | 3 (4.5) | 5 (2.2) | 6 (7.5) |
| Some college | 12 (17.9) | 7 (15.6) | 4 (8.9) | 18 (26.9) | 41 (18.3) | 26 (32.5) |
| College graduate | 33 (49.3) | 20 (44.4) | 22 (48.9) | 20 (29.9) | 95 (42.4) | 23 (28.7) |
| Postgraduate degree | 22 (32.8) | 15 (33.3) | 18 (40.0) | 25 (37.3) | 80 (35.7) | 23 (28.7) |
| Race and ethnicity (N, %)4 | ||||||
| White | 43 (64.2) | 24 (53.3) | 32 (71.1) | 39 (58.2) | 138 (61.6) | 44 (55.0) |
| Hispanic | 12 (17.9) | 9 (20.0) | 8 (17.8) | 13 (19.4) | 42 (18.8) | 19 (23.8) |
| Asian | 7 (10.4) | 8 (17.8) | 2 (4.4) | 7 (10.4) | 24 (10.7) | 6 (7.5) |
| African American | 0 (0.0) | 1 (2.2) | 3 (6.7) | 5 (7.5) | 9 (4.0) | 4 (5.0) |
| Other5 | 5 (7.5) | 3 (6.7) | 0 (0.0) | 3 (4.5) | 11 (4.9) | 7 (8.8) |
| Weight (kg) | ||||||
| Female | 88.5 (12.2) | 86.5 (12.4) | 89.2 (12.1) | 86.8 (12.2) | 87.7 (12.1) | 92.3 (13.0) |
| Male | 104.5 (12.8) | 110.7 (17.3) | 104.6 (9.4) | 104.6 (16.3) | 105.8 (14.3) | 109.5 (11.8) |
| BMI (kg/m2) | 33.22 (3.14) | 33.58 (3.55) | 33.19 (3.50) | 32.43 (3.36) | 33.1 (3.4) | 34.2 (3.5) |
| LDL (mg/dL) | 118.25 (26.77) | 114.80 (22.74) | 118.55 (28.13) | 114.91 (28.58) | 116.6 (26.7) | 108.9 (24.8) |
| HDL (mg/dL) | ||||||
| Female | 52.0 (11.2) | 52.1 (9.5) | 49.1 (10.7) | 50.2 (9.0) | 51.0 (10.1) | 51.1 (11.7) |
| Male | 40.6 (7.4) | 43.0 (9.0) | 46.3 (9.1) | 45.2 (9.7) | 43.8 (9.0) | 42.5 (10.0) |
| Triglycerides (mg/dL) | 142.40 (71.61) | 123.14 (49.94) | 150.23 (225.37) | 123.78 (56.11) | 134.5 (114.4) | 116.8 (53.8) |
| SBP (mmHg) | 122.97 (12.99) | 122.39 (12.63) | 122.32 (11.15) | 123.40 (12.80) | 122.9 (12.4) | 123.3 (12.5) |
| DBP (mmHg) | 81.07 (8.69) | 80.84 (8.18) | 81.13 (6.37) | 81.43 (8.19) | 81.2 (8.0) | 81.6 (7.7) |
| Fasting insulin (μU/mL) | 16.88 (6.51) | 15.40 (6.14) | 13.35 (4.81) | 13.71 (7.72) | 14.9 (6.7) | 16.9 (10.9) |
| Fasting glucose (mL/dL) | 100.54 (11.34) | 96.53 (7.45) | 96.27 (8.75) | 97.87 (10.77) | 98.1 (10.1) | 98.8 (8.8) |
| Waist circumference (cm) | ||||||
| Female | 104.1 (9.1) | 100.4 (13.4) | 102.1 (7.9) | 98.7 (9.9) | 101.4 (10.3) | 105.9 (10.6) |
| Male | 113.2 (8.1) | 114.8 (11.0) | 109.9 (6.7) | 110.9 (11.2) | 112.1 (9.5) | 114.4 (11.0) |
| Baseline Dietary Components | ||||||
| Total Energy (kcal) | 2490.05 (622.57) | 1843.27 (469.64) | 2451.06 (631.52) | 2024.14 (578.05) | 2212.93 (639.23) | 2251.52 (693.38) |
| Intake of Carbohydrate (g) | 263.85 (68.14) | 181.19 (53.89) | 246.98 (62.71) | 195.1 (63.98) | 223.29 (71.68) | 229.70 (80.80) |
| Intake of Fat (g) | 101.14 (32.32) | 78.05 (24.99) | 100.99 (36.24) | 86.34 (32.5) | 92.04 (33.08) | 94.09 (35.02) |
| HEI-2010 Score | 48.7 (9.9) | 51.22 (10.87) | 61.88 (11.99) | 61.26 (9.91) | 55.61 (12.06) | 53.10 (11.30) |
| 12-mo dietary components | --- | |||||
| Total energy (kcal) | 1581.38 (400.27) | 1745.44 (500.43) | 1589.74 (428.66) | 1852.35 (531.05) | 1697.07 (480.22 | --- |
| Intake of carbohydrate (g) | 90.25 (37.76) | 132.9 (53.67) | 84.46 (51.19) | 144.27 (69.49) | 113.81 (60.25) | --- |
| Intake of fat (g) | 85.47 (29.87) | 86.22 (29.02) | 86.04 (28.38) | 87.13 (32.2) | 86.23 (29.94) | --- |
| HEI-2010 Score | 63.86 (8.78) | 64.61 (10.8) | 56.54 (11.35) | 54.81 (9.48) | 59.83 (10.82) | --- |
Abbreviations: SBP, Systolic Blood Pressure; DBP, Diastolic Blood Pressure; HEI, Healthy Eating Index -2010.
Dietary quality is defined by 12-mo minus baseline change in HEI-2010 score (above the median is high, below the median is low).
Dietary adherence is defined by 12-mo minus baseline change in grams of carbohydrate (below the median is high, above the median is low).
All values presented as Mean (SD) unless specified.
Self-reported by participants from the following options (White, Hispanic, Black, Asian, American Indian, Alaska Native, Pacific Islander and Other).
American Indian/Alaska Native/Pacific Islander and Other.
TABLE 2.
Baseline characteristics by dietary quality and adherence to diet for those assigned to a healthy low-fat diet (n = 305)1,2,3.
| High quality, high adherence (HQ/HA) N = 64 | High quality, low adherence (HQ/LA) N = 48 | Low quality, high adherence (LQ/HA) N = 48 | Low quality, low adherence (LQ/LA) N = 64 | Total N = 224 | Missing 12-mo dietary data N = 81 | |
|---|---|---|---|---|---|---|
| Sex (N, %) | ||||||
| Female | 38 (59.4) | 33 (68.8) | 18 (37.5) | 37 (57.8) | 126 (56.2) | 41 (50.6) |
| Male | 26 (40.6) | 15 (31.3) | 30 (62.5) | 27 (42.2) | 98 (43.8) | 40 (49.4) |
| Age (y) | 39.9 (7.3) | 40.4 (6.0) | 40.3 (6.1) | 39.3 (6.7) | 39.94 (6.57) | 37.72 (7.17) |
| Education (N, %) | ||||||
| Less than college degree | 15 (23.5) | 12 (25.0) | 2 (4.2) | 17 (26.6) | 46 (20.6) | 24 (29.6) |
| College graduate | 24 (37.5) | 13 (27.1) | 30 (62.5) | 22 (34.4) | 89 (39.7) | 38 (46.9) |
| Postgraduate degree | 24 (37.5) | 23 (47.9) | 16 (33.3) | 25 (39.1) | 88 (39.3) | 19 (23.5) |
| Race and ethnicity (N, %)4 | ||||||
| White | 42 (65.6) | 24 (50.0) | 33 (68.8) | 35 (54.7) | 134 (59.8) | 42 (51.9) |
| Hispanic | 11 (17.2) | 16 (33.3) | 6 (12.5) | 12 (18.8) | 45 (20.1) | 22 (27.2) |
| Asian | 5 (7.8) | 4 (8.3) | 5 (10.4) | 7 (10.9) | 21 (9.4) | 9 (11.1) |
| African American | 3 (4.7) | 2 (4.2) | 0 (0.0) | 3 (4.7) | 8 (3.6) | 2 (2.5) |
| Other5 | 3 (4.7) | 2 (4.2) | 4 (8.3) | 7 (10.9) | 16 (7.1) | 6 (7.4) |
| Weight (kg) | ||||||
| Female | 93.2 (10.3) | 87.5 (9.2) | 91.6 (13.6) | 90.0 (13.4) | 90.6 (11.6) | 91.1 (11.1) |
| Male | 107.1 (17.1) | 105.2 (12.2) | 106.5 (10.9) | 101.9 (11.1) | 105.2 (13.0) | 107.1 (15.8) |
| BMI (kg/m2) | 33.9 (3.2) | 33.0 (3.1) | 33.9 (2.7) | 33.0 (3.6) | 33.45 (3.20) | 33.36 (3.41) |
| LDL (mg/dL) | 114.1 (30.7) | 106.6 (26.9) | 108.0 (33.1) | 117.7 (33.1) | 112.22 (31.28) | 114.19 (29.60) |
| HDL (mg/dL) | ||||||
| Female | 49.9 (9.2) | 50.5 (11.5) | 53.9 (18.4) | 50.4 (9.7) | 50.8 (11.6) | 50.7 (9.6) |
| Male | 43.3 (7.8) | 44.7 (9.0) | 43.0 (12.2) | 40.2 (7.5) | 42.6 (9.4) | 45.1 (7.4) |
| Triglycerides (mg/dL) | 138.3 (75.8) | 127.9 (101.4) | 131.5 (65.6) | 130.6 (71.0) | 132.4 (78.3) | 124.4 (55.8) |
| SBP (mmHg) | 124.1 (12.2) | 124.3 (14.5) | 127.5 (14.0) | 119.9 (12.4) | 123.7 (13.4) | 121.5 (11.1) |
| DBP (mmHg) | 81.8 (6.6) | 82.0 (8.4) | 84.4 (9.1) | 78.8 (6.5) | 81.5 (7.8) | 80.3 (6.6) |
| Fasting insulin (μU/mL) | 15.8 (8.0) | 15.1 (11.5) | 16.4 (7.6) | 17.4 (24.8) | 16.2 (15.3) | 14.8 (6.4) |
| Fasting glucose (mL/dL) | 99.3 (8.2) | 97.8 (9.6) | 98.6 (8.7) | 97.9 (9.0) | 98.4 (8.8) | 98.6 (8.1) |
| Waist circumference (cm) | ||||||
| Female | 104.5 (9.2) | 103.5 (13.3) | 104.9 (9.6) | 103.2 (10.8) | 103.9 (10.8) | 102.3 (9.2) |
| Male | 112.5 (11.6) | 111.6 (8.3) | 114.3 (8.4) | 109.0 (8.4) | 112.0 (9.4) | 111.4 (10.5) |
| Dietary Components | ||||||
| Total Energy (kcal) | 2555.5 (661.54) | 1686.41 (457.73) | 2560.23 (461) | 1950.03 (566.82) | 2197.29 (665.02) | 2010.24 (732.12) |
| Intake of Carbohydrate (g) | 264.24 (89.9) | 181.43 (68.05) | 249.76 (71.21) | 205.19 (62.85) | 226.52 (80.85) | 200.99 (79.61) |
| Intake of Fat (g) | 107.54 (31.38) | 62.51 (22.62) | 111.96 (26.54) | 73.9 (28.07) | 89.22 (34.43) | 80.88 (35.11) |
| HEI-2010 Score | 52.82 (10.92) | 53.51 (10.43) | 58.79 (9.17) | 61.23 (10.95) | 56.65 (11.02) | 52.40 (12.63) |
| 12-mo Dietary Components | — | |||||
| Total Energy (kcal) | 1621.24 (450.59) | 1554.29 (472.19) | 1756.26 (460.89) | 1914.07 (588.09) | 1719.49 (516.36) | — |
| Intake of Carbohydrate (g) | 189.44 (67.22) | 162.83 (67.04) | 199.6 (64.86) | 204.38 (72.38) | 190.19 (69.49) | — |
| Intake of Fat (g) | 47.95 (19.03) | 55.3 (24.56) | 55.54 (22.33) | 70.2 (27.81) | 57.51 (25.05) | — |
| HEI-2010 Score | 69.9 (10.76) | 71.15 (8.97) | 57.68 (8.88) | 57.31 (12.23) | 63.95 (12.29) | — |
Abbreviations: SBP, Systolic Blood Pressure; DBP, Diastolic Blood Pressure; HEI, Healthy Eating Index -2010.
Dietary quality is defined by 12-mo minus baseline change in HEI-2010 score (above the median is high, below the median is low).
Dietary adherence is defined by 12-mo minus baseline change in grams of fat (below the median is high, above the median is low).
All values presented as Mean (SD) unless specified.
Self-reported by participants from the following options (White, Hispanic, Black, Asian, American Indian, Alaska Native, Pacific Islander and Other).
American Indian/Alaska Native/Pacific Islander and Other.
TABLE 3.
Baseline to 12-mo change in primary and secondary outcome variables by dietary quality and adherence for those assigned to a healthy low-carbohydrate diet1,2,3.
| N | High quality, high adherence N = 67 | N | High quality, low adherence N = 45 | N | Low quality, high adherence N = 45 | N | Low quality, low adherence N = 67 | N | Total N = 224 | N | Missing N = 80 | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Primary outcome | ||||||||||||
| BMI (kg/m2) | 62 | -2.49 (2.33) | 42 | -2.22 (2.84) | 44 | -2.25 (2.11) | 62 | -1.73 (2.26) | 210 | -2.16 (2.38) | 70 | -2.33 (1.74) |
| Secondary outcomes | ||||||||||||
| Fasting glucose (mL/dL) | 62 | -4.31 (9.94) | 42 | -1.96 (8.54) | 43 | -1.38 (9.79) | 62 | 0.26 (9.68) | 209 | -1.88 (9.66) | 70 | -4.60 (4.45) |
| Fasting insulin (μU/mL) | 62 | -3.87 (4.34) | 42 | -2.90 (5.69) | 43 | -2.05 (4.62) | 62 | 0.05 (5.86) | 210 | -2.14 (5.36) | 70 | -1.95 (2.53) |
| SBP (mmHg) | 62 | -5.41 (12.52) | 42 | -4.19 (7.24) | 44 | -5.25 (7.78) | 62 | -0.56 (8.16) | 210 | -3.70 (9.61) | 70 | 0.25 (5.27) |
| DBP (mmHg) | 62 | -3.57 (8.06) | 42 | -2.45 (5.06) | 44 | -3.61 (5.39) | 62 | -1.04 (5.11) | 209 | -2.61 (6.23) | 70 | -0.78 (3.75) |
| Triglycerides (mg/dL) | 62 | -40.77 (54.32) | 42 | -20.61 (48.50) | 43 | -52.73 (191.12) | 62 | -13.35 (36.78) | 208 | -31.05 (96.65) | 70 | -19.35 (37.21) |
| HDL-C (mg/dL) | 62 | 3.19 (6.25) | 42 | 0.92 (5.56) | 42 | 2.76 (6.96) | 62 | 3.81 (7.02) | 208 | 2.83 (6.54) | 70 | -0.45 (5.29) |
| LDL-C (mg/dL) | 62 | 5.39 (21.78) | 42 | 4.17 (18.31) | 42 | 2.34 (19.44) | 62 | 2.08 (20.79) | 224 | 3.54 (20.26) | 70 | -2.53 (19.43) |
| Dietary components | ||||||||||||
| Energy (kcal) | 67 | -908.67 (569.93) | 45 | -97.83 (381.11) | 45 | -861.32 (566.08) | 67 | -171.79 (496.66) | 224 | -515.86 (634.27) | - | - |
| Carbohydrate (g) | 67 | -173.60 (58.28) | 45 | -48.29 (36.51) | 45 | -162.53 (45.49) | 67 | -50.83 (43.27) | 224 | -109.48 (76.31) | - | - |
| Fat (g) | 67 | -15.66 (37.60) | 45 | 8.18 (28.18) | 45 | -14.95 (43.61) | 67 | 0.79 (37.00) | 224 | -5.81 (38.13) | - | - |
| HEI-2010 Score | 67 | 15.16 (7.11) | 45 | 13.39 (5.99) | 45 | -5.35 (8.78) | 67 | -6.45 (7.31) | 224 | 4.22 (12.60) | - | - |
Abbreviations: SBP, Systolic Blood Pressure; DBP, Diastolic Blood Pressure; HEI, Healthy Eating Index -2010.
Dietary quality is defined by 12-mo minus baseline change in HEI-2010 score (above the median is high, below the median is low).
Dietary adherence is defined by 12-mo minus baseline change in grams of carbohydrate (below the median is high, above the median is low).
All values presented as Mean (SD). N columns are the number of participants whose data was available to include in the analysis of each outcome.
TABLE 4.
Baseline to 12-mo change in primary and secondary outcome variables by dietary quality and adherence for those assigned to a healthy low-fat diet1,2,3.
| N | High quality, high adherence N = 64 | N | High quality, low adherence N = 48 | N | Low quality, high adherence N = 48 | N | Low quality, low adherence N = 64 | N | Total N = 224 | N | Missing N = 81 | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Primary outcome | ||||||||||||
| BMI (kg/m2) | 62 | -2.54 (3.07) | 41 | -1.72 (1.67) | 45 | -1.99 (2.32) | 59 | -1.36 (2.39) | 207 | -1.92 (2.51) | 73 | -1.57 (2.38) |
| Secondary outcomes | ||||||||||||
| Fasting glucose (mL/dL) | 62 | -4.95 (8.90) | 42 | -2.13 (8.72) | 45 | -4.55 (10.14) | 59 | -2.85 (7.45) | 208 | -3.70 (8.78) | 73 | -5.65 (4.93) |
| Fasting insulin (μU/mL) | 62 | -3.35 (5.96) | 42 | -2.39 (8.81) | 45 | -0.95 (6.00) | 59 | -4.17 (24.99) | 208 | -2.87 (14.49) | 73 | -3.58 (6.47) |
| SBP (mmHg) | 61 | -4.85 (9.05) | 41 | -5.88 (8.29) | 45 | -3.78 (9.22) | 59 | -0.33 (8.83) | 208 | -3.53 (9.07) | 73 | -2.43 (8.30) |
| DBP (mmHg) | 61 | -3.46 (5.52) | 41 | -3.08 (5.32) | 45 | -2.86 (5.98) | 59 | 0.27 (4.90) | 208 | -2.19 (5.60) | 73 | 1.07 (3.12) |
| Triglycerides (mg/dL) | 62 | -17.51 (62.03) | 42 | -18.86 (50.35) | 45 | -5.44 (43.37) | 59 | -3.36 (57.90) | 208 | -11.16 (54.99) | 73 | -11.06 (68.67) |
| HDL-C (mg/dL) | 62 | -1.03 (6.48) | 42 | 0.25 (5.49) | 45 | -0.19 (5.97) | 59 | 1.67 (5.48) | 208 | 0.18 (5.95) | 73 | 3.75 (5.11) |
| LDL-C (mg/dL) | 62 | -6.29 (23.31) | 42 | 2.34 (18.78) | 45 | -3.87 (18.52) | 59 | 0.42 (17.53) | 208 | -2.12 (20.02) | 73 | -8.04 (16.14) |
| Dietary components | ||||||||||||
| Energy (kcal) | 64 | -934.26 (578.64) | 48 | -132.12 (380.04) | 48 | -803.98 (449.38) | 64 | -35.97 (430.93) | 224 | -477.80 (620.08) | - | - |
| Carbohydrate (g) | 64 | -74.80 (79.79) | 48 | -18.60 (66.48) | 48 | -50.16 (69.52) | 64 | -0.81 (68.13) | 224 | -36.33 (77.18) | - | - |
| Fat (g) | 64 | -59.58 (26.84) | 48 | -7.21 (15.85) | 48 | -56.42 (23.57) | 64 | -3.70 (16.88) | 224 | -31.72 (34.10) | - | - |
| HEI-2010 Score | 64 | 17.08 (5.85) | 48 | 17.64 (7.22) | 48 | -1.11 (8.11) | 64 | -3.92 (8.28) | 224 | 7.30 (12.47) | - | - |
Abbreviations: SBP, Systolic Blood Pressure; DBP, Diastolic Blood Pressure; HEI, Healthy Eating Index -2010.
Dietary quality is defined by 12-mo minus baseline change in HEI-2010 score (above the median is high, below the median is low).
Dietary adherence is defined by 12-mo minus baseline change in grams of fat (below the median is high, above the median is low).
All values presented as Mean (SD). N columns are the number of participants whose data was available to include in the analysis of each outcome.
TABLE 5.
Baseline to 12-mo changes in primary and secondary outcomes for those with high dietary quality, high dietary adherence, or both versus those with low dietary quality and adherence (Change (95% Confidence Interval)).
| High quality, High adherence | High quality, Low adherence | Low quality, High adherence | |
|---|---|---|---|
| Primary outcome | |||
| BMI (kg/m2) | |||
| Low-carb: M11 | -0.64 (-1.48, 0.20) | -0.43 (-1.36, 0.49) | -0.40 (-1.31, 0.52) |
| Low-carb: M22 | -1.15 (-2.04, -0.26) | -0.30 (-1.22, 0.61) | -0.80 (-1.74, 0.14) |
| Low-fat: M11 | -0.98 (-1.88, -0.08) | -0.30 (-1.29, 0.69) | -0.53 (-1.51, 0.44) |
| Low-fat: M23 | -1.11 (-2.10, -0.11) | -0.26 (-1.26, 0.75) | -0.66 (-1.74, 0.41) |
| Secondary outcomes | |||
| Fasting glucose (mL/dL) | |||
| Low-carb: M11 | -4.51 (-7.94, -1.08) | -2.12 (-5.92, 1.68) | -1.46 (-5.25, 2.32) |
| Low-carb: M22 | -5.40 (-9.12, -1.69) | -1.88 (-5.70, 1.93) | -2.12 (-6.04, 1.81) |
| Low-fat: M11 | -1.92 (-5.12, 1.28) | 0.72 (-2.79, 4.23) | -1.79 (-5.26, 1.68) |
| Low-fat: M23 | -2.13 (-5.66, 1.40) | 0.80 (-2.76, 4.36) | -2.01 (-5.84, 1.81) |
| Fasting insulin (μU/mL) | |||
| Low-carb: M11 | -3.93 (-5.75, -2.10) | -2.93 (-4.95, -0.91) | -1.84 (-3.85, 0.17) |
| Low-carb: M22 | -4.11 (-6.09, -2.13) | -2.88 (-4.92, -0.85) | -1.97 (-4.07, 0.12) |
| Low-fat: M11 | 1.21 (-4.04, 6.47) | 1.49 (-4.27, 7.26) | 4.38 (-1.32, 10.09) |
| Low-fat: M23 | -0.33 (-6.11, 5.44) | 2.05 (-3.78, 7.87) | 2.73 (-3.52, 8.99) |
| SBP (mmHg) | |||
| Low-carb: M11 | -4.66 (-8.01, -1.31) | -3.32 (-7.03, 0.39) | -4.66 (-8.33, -0.99) |
| Low-carb: M22 | -5.54 (-9.16, -1.92) | -3.09 (-6.81, 0.63) | -5.35 (-9.17, -1.53) |
| Low-fat: M11 | -4.00 (-7.23, -0.77) | -5.61 (-9.16, -2.06) | -2.87 (-6.35, 0.62) |
| Low-fat: M23 | -4.46 (-8.03, -0.90) | -5.45 (-9.04, -1.85) | -3.36 (-7.20, 0.48) |
| DBP (mmHg) | |||
| Low-carb: M11 | -2.40 (-4.59, -0.20) | -1.24 (-3.68, 1.20) | -2.44 (-4.85, -0.03) |
| Low-carb: M22 | -3.04 (-5.42, -0.67) | -1.07 (-3.52, 1.37) | -2.95 (-5.46, -0.44) |
| Low-fat: M11 | -3.67 (-5.66, -1.68) | -3.39 (-5.57, -1.20) | -2.94 (-5.09, -0.79) |
| Low-fat: M23 | -4.09 (-6.28, -1.89) | -3.24 (-5.45, -1.02) | -3.38 (-5.74, -1.01) |
| Triglycerides (mg/dL) | |||
| Low-carb: M11 | -26.88 (-61.34, 7.58) | -8.21 (-46.38, 29.96) | -39.53 (-77.58, -1.49) |
| Low-carb: M22 | -22.62 (-60.04, 14.81) | -9.32 (-47.74, 29.10) | -36.41 (-75.96, 3.14) |
| Low-fat: M11 | -16.77 (-36.80, 3.25) | -15.85 (-37.83, 6.12) | -2.99 (-24.72, 18.74) |
| Low-fat: M23 | -14.65 (-36.75, 7.45) | -16.62 (-38.88, 5.65) | -0.72 (-24.64, 23.20) |
| HDL-C (mg/dL) | |||
| Low-carb: M11 | -0.73 (-3.06, 1.60) | -2.82 (-5.39, -0.24) | -1.10 (-3.69, 1.48) |
| Low-carb: M22 | -1.07 (-3.59, 1.46) | -2.73 (-5.32, -0.14) | -1.34 (-4.02, 1.34) |
| Low-fat: M11 | -2.83 (-4.99, -0.66) | -1.42 (-3.79, 0.96) | -2.03 (-4.38, 0.31) |
| Low-fat: M23 | -2.35 (-4.73, 0.03) | -1.58 (-3.98, 0.81) | -1.53 (-4.11, 1.05) |
| LDL-C (mg/dL) | |||
| Low-carb: M11 | 3.81 (-3.44, 11.07) | 2.59 (-5.45, 10.63) | 0.34 (-7.73, 8.42) |
| Low-carb: M22 | 4.17 (-3.72, 12.06) | 2.49 (-5.60, 10.59) | 0.60 (-7.78, 8.98) |
| Low-fat: M11 | -6.83 (-14.11, 0.44) | 1.56 (-6.42, 9.55) | -3.90 (-11.80, 3.99) |
| Low-fat: M23 | -8.20 (-16.23, -0.18) | 2.05 (-6.03, 10.13) | -5.37 (-14.05, 3.32) |
Results based on linear regression models performed separately for each outcome and for each diet intervention (low-carbohydrate and low-fat), with the low quality, low adherence group as the reference group. Results display parameter estimates and 95% confidence intervals for all models.
Abbreviations: SBP, Systolic Blood Pressure; DBP, Diastolic Blood Pressure.
Model 1 adjusted for age (y), sex (male/female), and baseline weight (kg).
Model 2 for low carbohydrate adjusted for age (y), sex (male/female), baseline weight (kg), and baseline net carbohydrate intake (g).
Model 2 for low fat adjusted for age (y), sex (male/female), baseline weight (kg), and baseline total fat intake (g).
All statistical procedures were performed using SAS (SAS Institute, Cary, NC, version 9.4) and R (R Foundation for Statistical Computing, version 3.3.2), and all statistical tests were evaluated using an alpha cutoff of 0.05.
Results
Baseline characteristics by dietary quality and adherence to diet are provided in Table 1 for those assigned to an HLC diet and in Table 2 for those assigned to an HLF diet. Across both diet arms, study participants were 56.8% female, with the majority of participants having at least some college education. The racial and ethnic makeup was 58.8% non-Hispanic White, 21.0% Hispanic, 9.9% Asian, and 6.6% other race or ethnicity. Mean baseline weight was 106.2 kg for males and 89.8 kg for females. Mean BMI was 33.4 kg/m2, and mean waist circumference was 112.2 ±9.8 cm for males and 103 ±10.5 cm for females. Mean lipid values were in the normal ranges (LDL <130mg/dL; HDL ≥40mg/dL for males and HDL ≥50 for females; Triglycerides <150mg/dL), and mean fasting glucose concentration was 98.4 mg/dl. Mean HEI-2010 was 55.2 ± 11.7. Mean HEI components at baseline and 12 mo are provided in Supplemental Tables 3 and 4.
There were no missing data at baseline. At 12 mo, 366 participants (81.7%) included in this analysis provided all 3 24-h dietary recalls, 46 provided 2 recalls (10.3%), and 36 (8.0%) provided 1 recall. DIETFITS participants who did not provide dietary recalls at 12 mo were excluded from the analysis (N = 161, excluding 80 from HLC and 81 from HLF). Reasons for missing data were the majority (53%) unknown or not provided by participants; other reasons were diet assignment dissatisfaction, health, personal reasons, or a schedule change/move. The number of participants missing data for the clinical measures (i.e., BMI, SBP, DBP, glucose, insulin, triglycerides, HDL cholesterol, LDL cholesterol concentrations) within each quality/adherence subgroup ranged between 1 and 7 (Table 3 and Table 4).
Absolute 12-mo changes in BMI were greatest for HQ/HA and lowest for LQ/LA subgroups for both the HLC and HLF diets; values for HQ/LA and LQ/HA were intermediate between the 2 (Table 3 and Table 4). For the HLC diet group, fasting glucose, insulin concentrations, and SBP followed a similar pattern, as did fasting glucose concentration, DBP, and LDL cholesterol concentration for the HLF diet group. The remainder of clinical measures showed variability between group results for both diet groups (1).
The primary and secondary outcomes were analyzed via multiple linear regression using LQ/LA as the reference group (Table 5). For both the HLC and HLF diets, the HQ/HA subgroups had statistically significant decreases in 12-mo mean BMI [change (95% CI): -1.15 (-2.04, -0.26) for HLC; -1.11 (-2.10, -0.11) for HLF] (Figure 1A,B). The HQ/LA and LQ/HA subgroups also decreased mean BMIs more than the reference group, though these results were not statistically significant. For fasting glucose within HLC, all subgroups improved compared with the reference group, but this was statistically significant only for HQ/HA [-5.40 (-9.12, -1.69)]. For fasting glucose within HLF, only HQ/HA and LQ/HA improved compared with the reference group, though these results were not statistically significant. Fasting insulin values decreased significantly for the HQ/HA subgroup [-4.11 (-6.09, -2.13)] and the HQ/LA [-2.88 (-4.92, -0.85)] within the HLC diet group. Systolic blood pressure decreased significantly for both the HQ/HA [-5.54 (-9.16, -1.92)] and LQ/HA [-5.35 (-9.17, -1.53)] subgroups within the HLC diet group and for both the HQ/HA [-4.46 (-8.03, -0.90)] and HQ/LA [-5.45 (-9.04, -1.85)] subgroups with the HLF diet group. DBP decreased significantly for all subgroups except for HQ/LA within HLC; the greatest decreases were seen for the HQ/HA subgroups [-3.04 (-5.42, -0.67) for HLC; -4.09 (-6.28, -1.89) for HLF]. For triglycerides, values in all subgroups decreased compared with LQ/LA, but the results were not statistically significant. Concentrations of HDL cholesterol were worse for all subgroups compared with LQ/LA, but this was only statistically significant for the HQ/LA subgroup within HLC. Values of LDL cholesterol decreased in HQ/HA and LQ/HA within HLF; however, results were only statistically significant for the HQ/HA subgroup [-8.20 (-16.23, -0.18)]. For sensitivity analysis, adjustment for baseline caloric intake and baseline HEI 2010 score (Supplemental Table 5) did not meaningfully change the results. There were slight changes to lipid values such that those at the border of significance shifted slightly to cross the threshold in both directions.
FIGURE 1.
Baseline to 12-mo change in BMI by dietary quality and adherence for those assigned to a healthy low-carbohydrate diet (A) or healthy low-fat diet (B). Dietary quality is defined by 12-mo minus baseline change in HEI-2010 score (above the median is high, below the median is low); dietary adherence is defined by 12-mo minus baseline change in grams of net carbohydrates (below the median is high, above the median is low). Bars display mean (SD) 12-mo BMI change from baseline by diet quality/diet adherence groups. Numbers below the bars represent the changes in mean (95% CI) BMI compared with LQ/LA group, determined from linear regression modeling of 12-mo BMI change on diet quality and adherence groups adjusted for age, sex, baseline weight, and baseline net carbohydrate or total fat intake. Abbreviations: HQ, high quality; HA, high adherence; LQ, low quality; LA, low adherence.
Discussion
In this secondary analysis of a randomized controlled weight loss trial for generally healthy adults with overweight and obesity, the participants most successful in reducing their BMIs were those with both higher level of adherence to their assigned diets and higher quality diets. Additionally, this HQ/HA combination was associated with clinically and statistically significant reductions in systolic and diastolic blood pressures for those assigned to follow both HLC and HLF; fasting glucose and fasting insulin concentrations improved only for those assigned to the HLC diet. For the HLC group, HA seemed to drive the associations for both systolic and diastolic blood pressures. For the HLF group, HQ seemed to drive this association for systolic blood pressure; for DBP, either HA or HQ was sufficient to see this significant difference. Also, for the HLF group, HQ was associated with improvements in triglycerides values, and HQ and adherence were associated with slight worsening of HDL cholesterol values.
Within the primary analysis of the DIETFITS trial [11], weight change at 12 mo was −5.3 kg for the HLF diet compared with −6.0 kg for the HLC diet (mean between group difference, 0.7 kg [95% CI, −0.2 to 1.6 kg]). The findings of this post hoc analysis support prior studies showing the importance of adherence to a given dietary strategy for weight loss [[26], [27], [28]]. However, they also highlight the potential added benefits of choosing a high-quality diet within varying dietary strategies—such as low-carbohydrate or low-fat diets—for improved weight loss and blood pressure [[29], [30], [31]]. Much of the research on weight loss diet interventions has focused on a wide variety of dietary changes that directly or indirectly result in reduction of caloric intake [5]. Within the DIETFITS trial [11], calorie restriction was intentionally not part of the intervention but rather a change in the assigned macronutrient (carbohydrate or fat), and improvements in diet quality were the foci. However, in the AHA/ACC/TOS Guideline for Management of Obesity and Overweight in Adults, no explicit mention is made of aiming to improve dietary quality as part of a weight loss diet [5]. Clearly, this is an area that warrants further investigation. There are several scientifically plausible reasons why a higher quality diet may influence weight loss, including higher fiber content, lower calorie density, lower glycemic index, higher water content, and higher nutrient density, among others [[32], [33], [34], [35], [36]]. On a population level in observational studies, those with higher quality diets, as measured by the Healthy Eating Index, also have lower BMIs [[37], [38], [39]].
Moreover, this analysis showed associations between high-quality diets and improvements in metabolic measures such as fasting insulin concentration and systolic and diastolic blood pressures for those following low-fat diets. Additionally, improvements in fasting glucose and insulin values were seen in the low-carbohydrate group only for those with both HQ/HA diet. Improvements in insulin values were seen in the low-carbohydrate group only for both HQ/HA and HQ/LA. Due to the exploratory nature of this analysis, it is possible that these findings were due to chance. However, given the increased risk for metabolic disease among those with overweight and obesity, adding a focus on dietary quality in weight loss dietary interventions in order to potentially reduce risk of, or improve, related conditions independent of weight loss alone is intriguing and supported by the literature [38, [40], [41], [42], [43]].
Patterns of associations between dietary adherence/quality and changes in lipids among those assigned to HLC and HLF diets were less clear. Compared with the LA/LQ subgroup, none of the other adherence/quality groups had significant changes in their LDL cholesterol values. Triglyceride concentrations improved among only the HQ/LA subgroup assigned to the low-fat diet only. HDL cholesterol concentrations worsened very slightly among the HQ/HA subgroup assigned to the low-fat diet. These findings could have been due to a number of reasons, including chance, nondietary factors, or little room to change, given baseline mean lipid values were at or very near goal levels according to current guidelines [44]. Bias toward the null is also a possibility, given the method used to divide participants into dietary adherence/quality subgroups. Changes in dietary quality needed to impact lipids concentrations may also be different than dietary changes made that affected weight loss in this study. When considering lipids values and the dietary adherence/quality relationship, more research is needed.
The DIETFITS trial and this secondary analysis have several strengths. The DIETFITS trial had a large sample size (N = 609) that included females and males, had good retention (retention at 12 mo for the DIETFITS main outcomes, which was defined as participants who provided any data at 12 mo, was 79% for both groups), and a sufficient duration (12 mo) to evaluate long-term dietary adherence, dietary quality, and the impacts of these on primary and secondary outcomes. The gold standard for self-reported dietary data collection (i.e., the multipass 24-h recall) was used [45]. Study interventions were created from evidence-based frameworks of behavior change [12, 46, 47]. Medications that can influence weight and metabolism were part of the exclusion criteria; this is important because some of these same medications can also influence food choices. Additionally, baseline characteristics between participants included in the analysis and those with missing data were comparable (Table 1). Because study participants were generally healthy adults with overweight and obesity, it is potentially generalizable to a large segment of the US population. Statistical analysis controlled for baseline fat or carbohydrate intake (Table 5), which would have otherwise influenced the results.
Due to the inherent limitations of a secondary analysis, participants in this study were not randomly assigned to dietary quality/dietary adherence groups, making this essentially an observational study. Another limitation is a likely bias toward the null for all outcomes. This is because the method used to divide participants into diet quality/adherence groups was splitting at the median for 12-mo change in macronutrient (i.e., fat or carbohydrate) intake and median diet quality score (i.e., HEI). Many participants clustered around the median (Supplemental Figures 2A and 2B). To the contrary, a strength of this study is that the statistically significant findings are more likely to be real rather than found due to chance because of this bias. Participants were relatively well-educated compared with the general population and primarily non-Hispanic White, potentially limiting generalizability. A quarter of DIETFITS participants were excluded from this analysis due to missing 12-mo dietary recalls; however, those with missing data were comparable to those included in the analysis (TABLE 1, TABLE 2). Although high dietary quality is associated with better health in the literature [38, [40], [41], [42], [43]], there is no established cutoff above which it is considered high compared with low dietary quality for the outcomes included in the study. Similarly, although greater adherence to assigned diets within a dietary weight loss intervention is associated with greater weight loss [[26], [27], [28]], there is no clear cutoff for determining adherence compared with nonadherence within dietary interventions [48]. This is especially true for DIETFITS as instructions given in the study were to initially reduce to 20 grams of fat or net carbohydrate for 8 wk and then increase intake until a level was reached that participants could maintain long term. This maintenance level varied by participant, and thus, median was chosen as a rational scientific cutoff.
Further research is needed to assess the impact of dietary quality in dietary interventions aimed at weight loss and improvement in cardiovascular disease risk factors. The results of this secondary analysis suggest that rather than focusing solely on changes in macronutrient intake, such as low-carbohydrate and low-fat diets, additional benefits may be gained through emphasizing eating a high-quality diet, regardless of other dietary specifics. However, it should be noted that the quality of the diet is also influenced by macronutrient intake. To date, no studies have shown a negative impact on health of eating a high-quality diet, and countless have shown numerous health benefits. Therefore, in a clinical setting, risk-benefit calculation in deciding whether to recommend high-quality diet regardless of other dietary features is in favor of making this recommendation.
Acknowledgments
The authors would like to thank the Gardner research team members. Dr. Jennifer Robinson and Antonella Dewell served as study coordinators. Health educators included Rise Cherin, Susan Kirkpatrick, Jae Berman, Dalia Perelman, and Mandy Murphy Carroll. The diet assessment team included Sarah Farzinkhou, Valerie Alaimo, Margaret Shimer, and Diane Demis. Various other important study roles were played by Josephine Lin, Erin Avery, Alexandra Rossi, Katherine Dotter, and Dr. Sarah Mummah. Additional postdoctoral research fellows and visiting scholars who were involved in various phases of the study included Drs. Lisa Offringa, Kenji Nagao, Marily Oppezzo, Ben Chrisinger, and Michael Stanton.
Author contributions
The authors’ responsibilities were as follows—M.E.H. and J.C.H. formulated the research question and wrote the initial and final drafts of the manuscript, C.D.G designed the DIETFITS research project and contributed to writing both the initial and final drafts of the manuscript, M.J.L. contributed to writing and editing the final drafts, and J.C.H., L.D.G, C.W.S., F.Q., J.R. provided feedback and critical revisions of the manuscript. All authors have read and approved the final manuscript.
Conflicts of interest
All authors have completed the ICMJE uniform disclosure form and declare: no support from any organization for the submitted work; no financial relationships with any organizations that might have an interest in the submitted work in the previous 3 years; no other relationships or activities that could appear to have influenced the submitted work.
Funding
National Institute of Diabetes and Digestive and Kidney Diseases, NIH 1R01DK091831; Nutrition Science Initiative (NuSI); National Heart, Lung, and Blood InstituteNIH T32HL007034; NIH 1 K12 GM088033; Stanford Clinical and Translational Science Award (CTSA). The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH or other funders.
The study funders had no role in the study design, in the collection, analysis, and interpretation of data, in the writing of the report, or in the decision to submit the article for publication. All authors, external and internal, had full access to all the data (including statistical reports and tables) in the study and can take responsibility for the integrity of the data, and the accuracy of the data analysis is also required.
Data availability
The data that support the findings of this study are available from the corresponding author, C.D.G., upon reasonable request.
Footnotes
Supplementary data to this article can be found online at https://doi.org/10.1016/j.ajcnut.2023.10.028.
Appendix A. Supplementary data
The following are the Supplementary data to this article:
References
- 1.Hales C.M., Carroll M.D., Fryar C.D., Ogden C.L. Prevalence of obesity and severe obesity among adults: United States, 2017–2018. NCHS Data Brief. 2020;360:1–8. [PubMed] [Google Scholar]
- 2.Ward Z.J., Bleich S.N., Cradock A.L., Barrett J.L., Giles C.M., Flax C., et al. Projected US state-level prevalence of adult obesity and severe obesity. N. Engl. J. Med. 2019;381(25):2440–2450. doi: 10.1056/NEJMsa1909301. [DOI] [PubMed] [Google Scholar]
- 3.U. S. Burden of Disease Collaborators. Mokdad A.H., Ballestros K., Echko M., Glenn S., Olsen H.E., et al. The state of US health, 1990-2016: burden of diseases, injuries, and risk factors among US states. JAMA. 2018;319(14):1444–1472. doi: 10.1001/jama.2018.0158. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Dietary Guidelines Advisory Committee . U.S. Department of Agriculture, Agricultural Research Service; Washington, DC: 2020. Scientific Report of the 2020 Dietary Guidelines Advisory Committee: Advisory Report to the Secretary of Agriculture and the Secretary of Health and Human Services.https://www.dietaryguidelines.gov/2020-advisory-committee-report [Google Scholar]
- 5.Jensen M.D., Ryan D.H., Apovian C.M., Ard J.D., Comuzzie A.G., Donato K.A., et al. AHA/ACC/TOS guideline for the management of overweight and obesity in adults: a report of the American College of Cardiology/American Heart Association task force on practice guidelines and the obesity society. J. Am. Coll. Cardiol. 2014;63(25 Pt B):2985–3023. doi: 10.1016/j.jacc.2013.11.004. 2013. [DOI] [PubMed] [Google Scholar]
- 6.Johnston B.C., Kanters S., Bandayrel K., Wu P., Naji F., Siemieniuk R.A., et al. Comparison of weight loss among named diet programs in overweight and obese adults: a meta-analysis. JAMA. 2014;312(9):923–933. doi: 10.1001/jama.2014.10397. [DOI] [PubMed] [Google Scholar]
- 7.Tobias D.K., Chen M., Manson J.E., Ludwig D.S., Willett W., Hu F.B. Effect of low-fat diet interventions versus other diet interventions on long-term weight change in adults: a systematic review and meta-analysis. Lancet Diabetes Endocrinol. 2015;3(12):968–979. doi: 10.1016/s2213-8587(15)00367-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Sacks F.M., Bray G.A., Carey V.J., Smith S.R., Ryan D.H., Anton S.D., et al. Comparison of weight-loss diets with different compositions of fat, protein, and carbohydrates. N. Engl. J. Med. 2009;360:859–873. doi: 10.1056/NEJMoa0804748. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Shan Z., Guo Y., Hu F.B., Liu L., Qi Q. Association of low-carbohydrate and low-fat diets with mortality among US adults. JAMA Intern. Med. 2020;180:513–523. doi: 10.1001/jamainternmed.2019.6980. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Landry M.J., Crimarco A., Gardner C.D. Benefits of low carbohydrate diets: a settled question or still controversial? Curr. Obes. Rep. 2021;10(3):409–422. doi: 10.1007/s13679-021-00451-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Gardner C.D., Trepanowski J.F., Del Gobbo L.C., Hauser M.E., Rigdon J., Ioannidis J.P.A., et al. Effect of low-fat vs low-carbohydrate diet on 12-month weight loss in overweight adults and the association with genotype pattern or insulin secretion: the DIETFITS randomized clinical trial. JAMA. 2018;319(7):667–679. doi: 10.1001/jama.2018.0245. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Stanton M.V., Robinson J.L., Kirkpatrick S.M., Farzinkhou S., Avery E.C., Rigdon J., et al. DIETFITS study (Diet Intervention Examining The Factors Interacting with Treatment Success) - study design and methods. Contemp. Clin. Trials. 2017;53:151–161. doi: 10.1016/j.cct.2016.12.021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Chawla S., Tessarolo Silva F., Amaral Medeiros S., Mekary R.A., Radenkovic D. The effect of low-fat and low-carbohydrate diets on weight loss and lipid levels: a systematic review and meta-analysis. Nutrients. 2020;12(12):3774. doi: 10.3390/nu12123774. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Stern L., Iqbal N., Seshadri P., Chicano K.L., Daily D.A., McGrory J., et al. The effects of low-carbohydrate versus conventional weight loss diets in severely obese adults: one-year follow-up of a randomized trial. Ann. Intern. Med. 2004;140(10):778–785. doi: 10.7326/0003-4819-140-10-200405180-00007. [DOI] [PubMed] [Google Scholar]
- 15.Foster G.D., Wyatt H.R., Hill J.O., Makris A.P., Rosenbaum D.L., Brill C., et al. Weight and metabolic outcomes after 2 years on a low-carbohydrate versus low-fat diet: a randomized trial. Ann. Intern. Med. 2010;153(3):147–157. doi: 10.1059/0003-4819-153-3-201008030-00005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Yancy W.S., Jr., Olsen M.K., Guyton J.R., Bakst R.P., Westman E.C. A low-carbohydrate, ketogenic diet versus a low-fat diet to treat obesity and hyperlipidemia: a randomized, controlled trial. Ann. Intern. Med. 2004;140(10):769–777. doi: 10.7326/0003-4819-140-10-200405180-00006. [DOI] [PubMed] [Google Scholar]
- 17.Dansinger M.L., Gleason J.A., Griffith J.L., Selker H.P., Schaefer E.J. Comparison of the Atkins, Ornish, Weight Watchers, and Zone diets for weight loss and heart disease risk reduction: a randomized trial. JAMA. 2005;293(1):43–53. doi: 10.1001/jama.293.1.43. [DOI] [PubMed] [Google Scholar]
- 18.Gardner C.D., Kiazand A., Alhassan S., Kim S., Stafford R.S., Balise R.R., et al. Comparison of the Atkins, Zone, Ornish, and LEARN diets for change in weight and related risk factors among overweight premenopausal women: the A TO Z weight loss study: a randomized trial. JAMA. 2007;297(9):969–977. doi: 10.1001/jama.297.9.969. [DOI] [PubMed] [Google Scholar]
- 19.Shai I., Schwarzfuchs D., Henkin Y., Shahar D.R., Witkow S., Greenberg I., et al. Weight loss with a low-carbohydrate, Mediterranean, or low-fat diet. N Engl J Med. 2008;359(3):229–241. doi: 10.1056/NEJMoa0708681. [DOI] [PubMed] [Google Scholar]
- 20.Harris P.A., Taylor R., Thielke R., Payne J., Gonzalez N., Conde J.G. Research Electronic Data Capture (REDCap)--a metadata-driven methodology and workflow process for providing translational research informatics support. J. Biomed. Inform. 2009;42(2):377–381. doi: 10.1016/j.jbi.2008.08.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Johnson R.K., Driscoll P., Goran M.I. Comparison of multiple-pass 24-hour recall estimates of energy intake with total energy expenditure determined by the doubly-labeled water method in young children. J. Am. Diet Assoc. 1996;96(11):1140–1144. doi: 10.1016/S0002-8223(96)00293-3. [DOI] [PubMed] [Google Scholar]
- 22.Guenther P.M., Casavale K.O., Reedy J., Kirkpatrick S.I., Hiza H.A., Kuczynski K.J., et al. Update of the Healthy Eating Index: HEI-2010. J. Acad. Nutr. Diet. 2013;113(4):569–580. doi: 10.1016/j.jand.2012.12.016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Guenther P.M., Kirkpatrick S.I., Reedy J., Krebs-Smith S.M., Buckman D.W., Dodd K.W., et al. The healthy eating index-2010 is a valid and reliable measure of diet quality according to the 2010 dietary guidelines for Americans. J. Nutr. 2014;144(3):399–407. doi: 10.3945/jn.113.183079. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Landy D.C., Kurtz J.M., Miller T.L., Ludwig D.A. Statistical program to automate the creation of healthy eating index scores using nutrition data system for research output. J. Acad. Nutr. Diet. 2012;112(9):A14. [Google Scholar]
- 25.Ludwig D.A., Landy D.C., Kurtz J.M., Miller T.L. SAS Global Forum; San Francisco, CA: 2013. Using SAS to expand the application of standard measures and guide statistical exploration: creating healthy eating index scores using nutrition data system for research output.http://support.sas.com/resources/papers/proceedings13/216P-2013.pdf [Google Scholar]
- 26.Alhassan S., Kim S., Bersamin A., King A., Gardner C. Dietary adherence and weight loss success among overweight women: results from the A TO Z weight loss study. Int. J. Obes. (Lond). 2008;32(6):985–991. doi: 10.1038/ijo.2008.8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Gibson A.A., Sainsbury A. Strategies to improve adherence to dietary weight loss interventions in research and real-world settings. Behav. Sci. (Basel). 2017;7(3):44. doi: 10.3390/bs7030044. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Del Corral P., Bryan D.R., Garvey W.T., Gower B.A., Hunter G.R. Dietary adherence during weight loss predicts weight regain. Obesity (Silver Spring) 2011;19(6):1177–1181. doi: 10.1038/oby.2010.298. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Landry M.J., Crimarco A., Perelman D., Durand L.R., Petlura C., Aronica L., et al. Adherence to Ketogenic and Mediterranean study diets in a crossover trial: the Keto–Med randomized trial. Nutrients. 2021;13(3):967. doi: 10.3390/nu13030967. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Allaire B.T., Tjaden A.H., Venditti E.M., Apolzan J.W., Dabelea D., Delahanty L.M., et al. Diet quality, weight loss, and diabetes incidence in the Diabetes Prevention Program (DPP) BMC Nutr. 2020;6(1):1–11. doi: 10.1186/s40795-020-00400-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Pascual R.W., Phelan S., La Frano M.R., Pilolla K.D., Griffiths Z., Foster G.D. Diet quality and micronutrient intake among long-term weight loss maintainers. Nutrients. 2019;11(12):3046. doi: 10.3390/nu11123046. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Shan Z., Rehm C.D., Rogers G., Ruan M., Wang D.D., Hu F.B., et al. Trends in dietary carbohydrate, protein, and fat intake and diet quality among US adults, 1999-2016. JAMA. 2019;322(12):117–187. doi: 10.1001/jama.2019.13771. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Reynolds A., Mann J., Cummings J., Winter N., Mete E., Te Morenga L. Carbohydrate quality and human health: a series of systematic reviews and meta-analyses. Lancet. 2019;393(10170):434–445. doi: 10.1016/S0140-6736(18)31809-9. [DOI] [PubMed] [Google Scholar]
- 34.Ludwig D.S., Hu F.B., Tappy L., Brand-Miller J. Dietary carbohydrates: role of quality and quantity in chronic disease. BMJ. 2018;361 doi: 10.1136/bmj.k2340. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Ludwig D.S., Willett W.C., Volek J.S., Neuhouser M.L. Dietary fat: from foe to friend? Science. 2018;362(6416):764–770. doi: 10.1126/science.aau2096. [DOI] [PubMed] [Google Scholar]
- 36.Drewnowski A. Nutrient density: addressing the challenge of obesity. Br. J. Nutr. 2018;120(s1):S8–S14. doi: 10.1017/S0007114517002240. [DOI] [PubMed] [Google Scholar]
- 37.Guo X., Warden B., Paeratakul S., Bray G. Healthy eating index and obesity. Eur. J. Clin. Nutr. 2004;58(12):1580–1586. doi: 10.1038/sj.ejcn.1601989. [DOI] [PubMed] [Google Scholar]
- 38.Schwingshackl L., Bogensberger B., Hoffmann G. Diet quality as assessed by the healthy eating index, alternate healthy eating index, dietary approaches to stop hypertension score, and health outcomes: an updated systematic review and meta-analysis of cohort studies. J. Acad. Nutr. Diet. 2018;118(1):74–100.e11. doi: 10.1016/j.jand.2017.08.024. [DOI] [PubMed] [Google Scholar]
- 39.Asghari G., Mirmiran P., Yuzbashian E., Azizi F. A systematic review of diet quality indices in relation to obesity. Br. J. Nutr. 2017;117(8):1055–1065. doi: 10.1017/S0007114517000915. [DOI] [PubMed] [Google Scholar]
- 40.George S.M., Ballard-Barbash R., Manson J.E., Reedy J., Shikany J.M., Subar A.F., et al. Comparing indices of diet quality with chronic disease mortality risk in postmenopausal women in the Women's Health Initiative Observational Study: evidence to inform national dietary guidance. Am. J. Epidemiol. 2014;180(6):616–625. doi: 10.1093/aje/kwu173. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Xu Z., Steffen L.M., Selvin E., Rebholz C.M. Diet quality, change in diet quality and risk of incident CVD and diabetes. Public Health Nutr. 2020;23(2):329–338. doi: 10.1017/S136898001900212X. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Reedy J., Lerman J.L., Krebs-Smith S.M., Kirkpatrick S.I., Pannucci T.E., Wilson M.M., et al. Evaluation of the healthy eating index-2015. J. Acad. Nutr. Diet. 2018;118(9):1622–1633. doi: 10.1016/j.jand.2018.05.019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Sotos-Prieto M., Bhupathiraju S.N., Mattei J., Fung T.T., Li Y., Pan A., et al. Association of changes in diet quality with total and cause-specific mortality. N. Engl. J. Med. 2017;377(2):143–153. doi: 10.1056/NEJMoa1613502. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Arnett D.K., Blumenthal R.S., Albert M.A., Buroker A.B., Goldberger Z.D., Hahn E.J., et al. ACC/AHA guideline on the primary prevention of cardiovascular disease: a report of the American College of Cardiology/American Heart Association task force on clinical practice guidelines. J. Am. Coll. Cardiol. 2019;74(10):e177–e232. doi: 10.1016/j.jacc.2019.03.010. 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Thompson F.E., Subar A.F. In: Nutrition in the Prevention and Treatment of Disease. Fourth Edition. Edtion, editor. Elsevier; 2017. Dietary assessment methodology; pp. 5–48. [Google Scholar]
- 46.King A.C., Frey-Hewitt B., Dreon D.M., Wood P.D. Diet vs exercise in weight maintenance: the effects of minimal intervention strategies on long-term outcomes in men. Arch. Intern. Med. 1989;149(12):2741–2746. doi: 10.1001/archinte.149.12.2741. [DOI] [PubMed] [Google Scholar]
- 47.Brownell K.D., Cohen L.R. Adherence to dietary regimens 2: components of effective interventions. Behav. Med. 1995;20(4):155–164. doi: 10.1080/08964289.1995.9933732. [DOI] [PubMed] [Google Scholar]
- 48.Gardner C.D., Crimarco A., Landry M.J., Fielding-Singh P. Nutrition study design issues—important issues for interpretation. Am. J. Health Promot. 2020;34(8):951–954. doi: 10.1177/0890117120960580d. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data that support the findings of this study are available from the corresponding author, C.D.G., upon reasonable request.