Impact of carbohydrate quantity and quality on maternal and pregnancy outcomes in gestational diabetes mellitus: A systematic review and meta-analysis.

OBJECTIVE: To evaluate the impact of carbohydrate quantity and quality on maternal and pregnancy outcomes in gestational diabetes mellitus. METHODS: Using a pre-defined search strategy, two researchers systematically searched MEDLINE, CINAHL Plus, and PubMed for randomized controlled trials comparing low-carbohydrate, low-glycaemic index, or low-glycaemic load diets with usual care in gestational diabetes mellitus. Mean differences and risk ratios were extracted. RESULTS: Thirteen studies with 877 participants were included. Low-carbohydrate diet did not significantly differ from usual care for fasting blood glucose (3 studies; mean difference: 1.60 mmol/L; 95 % confidence interval: -1.95, 5.15), insulin requirement (2 studies; risk ratio: 1.01; 95 % confidence interval: 0.31, 3.05), birthweight (4 studies; mean difference: -0.23 kg; 95 % confidence interval: -1.90, 1.45), caesarean delivery (5 studies; risk ratio: 1.11; 95 % confidence interval: 0.66, 1.85), macrosomia (3 studies; risk ratio: 0.35; 95 % confidence interval: 0.00, 2130.64), large-for-gestational-age (2 studies; risk ratio: 0.46; 95 % confidence interval: 0.03, 7.20), and small-for-gestational-age infants (2 studies; risk ratio: 0.94; 95 % confidence interval: 0.00, 231.18). Low-glycaemic index diet did not significantly differ from usual care for the above outcomes either. However, low-glycaemic load diet reduced macrosomia risk (2 studies; risk ratio: 0.51; 95 % confidence interval: 0.43, 0.59). CONCLUSIONS: Low-carbohydrate and low-glycaemic index diets do not differ from usual care for most maternal and foetal outcomes in gestational diabetes mellitus. But low-glycaemic load diet may reduce macrosomia risk.

Dietary Strategies to Reduce Triglycerides in Women of Reproductive Age: A Simulation Modelling Study.

Many women of reproductive age have poor diet quality and are at higher risk of chronic diseases such as diabetes. Triglycerides are a critical risk factor for chronic diseases, and although they can be influenced by diet, there are minimal dietary intervention studies identifying key foods/food groups that reduce triglycerides. We performed data simulation modelling to estimate the potential reductions in fasting triglycerides that could be achieved by different dietary strategies in reproductive age women. The model was created using data from the 2011-2013 Australian Health Survey and incorporated various factors such as demographics, nutrient intake, and plasma biomarkers. Multiple linear regression analysis was conducted to estimate triglyceride levels, considering nutrient intake and pre-determined covariates. Dietary scenarios were developed, reducing the consumption of processed/ultra-processed foods, while increasing the intake of minimally processed foods like fruits, vegetables, fish, and nuts. A total of 606 women were included. Reducing processed foods by 50% plus increasing intakes of fruits (75-225 g/day), vegetables (75-225 g/day), or nuts (10-40 g/day) decreased triglycerides by up to 4.3%. Additionally, incorporating 80 g/day of omega 3 fish (>800 mg long-chain omega 3/100 g) decreased triglycerides by 8.2%. The clinical relevance of lowering triglycerides for cardiometabolic disease management should be tested in dietary intervention studies in women.

The effects of the ketogenic diet for the management of type 2 diabetes mellitus: A systematic review and meta-analysis of recent studies.

OBJECTIVE: To systematically review the effects of the ketogenic diet on glycaemic control, body weight, cardiovascular risk factors, and liver and kidney function in patients with type 2 diabetes. METHODS: PubMed, MEDLINE, Embase, Cochrane Library and CINAHL were searched for randomised controlled trials published between 2001 and 2021 that compared the ketogenic diet to a control diet for effects on glycaemic control, body weight, cardiovascular risk factors, liver and renal function markers in adults with type 2 diabetes for >14 days. Meta-analyses using fixed or random effects models were conducted. RESULTS: Nineteen reports from 11 randomised controlled trials were included. Compared to the control, the ketogenic diet showed no significant difference in changes in glycaemic control or body weight, but greater increases in HDL (standardised mean difference 0.19; 95%CI 0.02-0.37; I2 = 0 %; moderate-quality evidence) and greater reductions in triglycerides (standardised mean difference -0.41; 95%CI -0.64 to -0.18; I2 = 0 %; low-quality evidence). CONCLUSIONS: The ketogenic diet may improve lipid profiles but showed no additional benefits for glycaemic control or weight loss compared to control diets in type 2 diabetes patients over two years.

The Nutritional Quality of Gluten-Free versus Non-Gluten-Free Pre-Packaged Foods and Beverages Sold in Hong Kong.

INTRODUCTION: The consumption of gluten-free foods has continued to increase in recent years. Given their higher intake among individuals both with and without a medically diagnosed gluten allergy or sensitivity, it is important to understand how the nutritional quality of these foods compares against non-gluten-free foods. As such, we aimed to compare the nutritional quality of gluten-free and non-gluten-free pre-packaged foods sold in Hong Kong. METHODS: Data from 18,292 pre-packaged food and beverage items in the 2019 FoodSwitch Hong Kong database were used. These products were categorized as (1) "declared gluten-free"; (2) "gluten-free by ingredient or naturally gluten-free"; and (3) "non-gluten-free" according to information presented on the package. One-way ANOVA was used to compare the differences in the Australian Health Star Rating (HSR), energy, protein, fibre, total fat, saturated fat, trans-fat, carbohydrates, sugars, and sodium content between products in different gluten categories, overall and by major food category (e.g., bread and bakery products) and region of origin (e.g., America, Europe). RESULTS: Products declared gluten-free (mean ± SD: 2.9 ± 1.3; n = 7%) had statistically significantly higher HSR than those gluten-free by ingredient or naturally gluten-free (2.7 ± 1.4; n = 51.9%) and non-gluten-free (2.2 ± 1.4; n = 41.2%) (all pairwise comparisons p < 0.001). Overall, non-gluten-free products have higher energy, protein, saturated fat, trans-fat, free sugar and sodium, and less fibre compared with products in the other two gluten categories. Similar differences were observed across major food groups and by region of origin. CONCLUSIONS: Non-gluten-free products sold in Hong Kong were generally less healthy than gluten-free products (regardless of the presence of gluten-free declaration). Consumers should be better educated on how to identify gluten-free foods, given that many gluten-free foods do not declare this information on the label.

Sex Differences in Association Between Gut Microbiome and Essential Hypertension Based on Ambulatory Blood Pressure Monitoring.

BACKGROUND: Sex differences in the pathogenesis of hypertension exist. While gut microbiota (GM) has been associated with hypertension, it is unclear whether there are sex-linked differences in the association between GM and hypertension. METHODS: We conducted a cross-sectional study to investigate the sex differences in associations between GM characterized by shotgun sequencing, GM-derived short-chain fatty acids, and 24-hour ambulatory blood pressure in 241 Hong Kong Chinese (113 men and 128 women; mean age, 54±6 years). RESULTS: The hypertensive group was associated with GM alterations; however, significant differences in ß-diversity and GM composition in hypertensive versus normotensive groups were only observed in women and not in men under various statistical models adjusting for the following covariates: age, sex, body mass index, sodium intake estimated by spot urine analysis, blood glucose, triglycerides, low- and high-density lipoprotein cholesterol, smoking, menopause, and fatty liver status. Specifically, Ruminococcus gnavus, Clostridium bolteae, and Bacteroides ovatus were significantly more abundant in the hypertensive women, whereas Dorea formicigenerans was more abundant in the normotensive women. No bacterial species were found to be significantly associated with hypertension in men. Furthermore, total plasma short-chain fatty acids and propionic acid were independent predictors of systolic and diastolic blood pressure in women but not men. CONCLUSIONS: GM dysregulation was strongly associated with 24-hour ambulatory blood pressure in women but not men, which may be mediated through propionic acid. Our work suggests that sex differences may be an important consideration while assessing the role of GM in the development and treatment of hypertension.

Socio-economic difference in purchases of ultra-processed foods in Australia: an analysis of a nationally representative household grocery purchasing panel.

BACKGROUND: Consumption of ultra-processed foods is associated with increased risk of obesity and non-communicable diseases. Little is known about current patterns of ultra-processed foods intake in Australia. The aim of this study was to examine the amount and type of ultra-processed foods purchased by Australian households in 2019 and determine whether purchases differed by socio-economic status (SES). We also assessed whether purchases of ultra-processed foods changed between 2015 and 2019.  METHODS: We used grocery purchase data from a nationally representative consumer panel in Australia to assess packaged and unpackaged grocery purchases that were brought home between 2015 to 2019. Ultra-processed foods were identified according to the NOVA system, which classifies foods according to the nature, extent and purpose of industrial food processing. Purchases of ultra-processed foods were calculated per capita, using two outcomes: grams/day and percent of total energy. The top food categories contributing to purchases of ultra-processed foods in 2019 were identified, and differences in ultra-processed food purchases by SES (Index of Relative Social Advantage and Disadvantage) were assessed using survey-weighted linear regression. Changes in purchases of ultra-processed foods between 2015 to 2019 were examined overall and by SES using mixed linear models. RESULTS: In 2019, the mean ± SD total grocery purchases made by Australian households was 881.1 ± 511.9 g/d per capita. Of this, 424.2 ± 319.0 g/d per capita was attributable to purchases of ultra-processed foods, which represented 56.4% of total energy purchased. The largest food categories contributing to total energy purchased included mass-produced, packaged breads (8.2% of total energy purchased), chocolate and sweets (5.7%), biscuits and crackers (5.7%) and ice-cream and edible ices (4.3%). In 2019, purchases of ultra-processed foods were significantly higher for the lowest SES households compared to all other SES quintiles (P < 0.001). There were no major changes in purchases of ultra-processed foods overall or by SES over the five-year period. CONCLUSIONS: Between 2015 and 2019, ultra-processed foods have consistently made up the majority of groceries purchased by Australians, particularly for the lowest SES households. Policies that reduce ultra-processed food consumption may reduce diet-related health inequalities.

Trans-fat labelling information on prepackaged foods and beverages sold in Hong Kong in 2019.

OBJECTIVE: To examine the labelling status of trans-fat of pre-packaged foods sold in Hong Kong. DESIGN: Data from 19 027 items in the 2019 FoodSwitch Hong Kong database were used. Ingredient lists were screened to identify specific (e.g. partially hydrogenated vegetable oil, PHVO) and non-specific trans-fat ingredient indicators (e.g. hydrogenated oil). Trans-fat content was obtained from the on-pack nutrition labels, which was converted into proportion of total fat (%total fat). Descriptive statistics were calculated for trans-fat content and the number of specific, non-specific and total trans-fat ingredients indicators found on the ingredients lists. Comparisons were made between regions using one-way ANOVA and χ2 for continuous and categorical variables, respectively. SETTING: Cross-sectional audit. PARTICIPANTS: Not applicable. RESULTS: A total of 729 items (3·8 % of all products) reported to contain industrially produced trans-fat, with a median of 0·4 g/100 g or 100 ml (interquartile range (IQR): 0·1-0·6) and 1·2 %totalfat (IQR: 0·6-2·9). 'Bread and bakery products' had the highest proportion of items with industrially produced trans-fat (18·9 %). 'Non-alcoholic beverages' had the highest proportion of products of 'false negatives' labelling (e.g. labelled as 0 trans-fat but contains PHVO; 59·3 %). The majority of products with trans-fat indicator originated from Asia (70 %). CONCLUSIONS: According to the labelling ∼4 % of pre-packaged food and beverages sold in Hong Kong in 2019 contained industrially produced trans-fat, and a third of these had trans-fat >2 %total fat. The ambiguous trans-fat labelling in Hong Kong may not effectively assist consumers in identifying products free from industrially produced trans-fat.

The longitudinal association between coffee and tea consumption and the risk of metabolic syndrome and its component conditions in an older adult population.

The present study aimed to assess the longitudinal associations of coffee and tea consumption with metabolic syndrome and its component conditions in a group of Australian older adults who participated in the Blue Mountains Eye Study (n 2554, mean age: 64 years, 43 % female). Participants' coffee and tea intake were measured using a validated food frequency questionnaire. Hazard ratios (HRs) over a 10-year period were estimated using Cox hazard regression models adjusting for lifestyle factors. Results showed that coffee consumption was not associated with the incidence of metabolic syndrome, high fasting glucose, high triglycerides, central obesity, high blood pressure and low HDL-cholesterol (HDL-C). Tea consumption was not associated with incidence of metabolic syndrome and the component conditions except for the risk of having low HDL-C, in which a nominally inverse association was observed (multivariate-adjusted HR at 2-3 cups/d: 0⋅48, 95 % CI 0⋅26, 0⋅87, P = 0⋅016; 4 cups/d or more: 0⋅50, 95 % CI 0⋅27, 0⋅93, P = 0⋅029). After stratifying for fruit consumption (P interaction between tea and fruit = 0⋅007), consuming four cups of tea per day was nominally associated with lower incidence of metabolic syndrome among those with high fruit consumption (multivariable-adjusted HR: 0⋅44, 95 % CI 0⋅20, 0⋅93, P = 0⋅033). Our results did not support a significant association between tea and coffee consumption and metabolic syndrome. Tea consumption may be associated with a lower risk of having low HDL-C, while high tea and fruit consumption together may be associated with a lower risk of developing metabolic syndrome.

Current WHO recommendation to reduce free sugar intake from all sources to below 10% of daily energy intake for supporting overall health is not well supported by available evidence.

Sugar is widely consumed over the world. Although the mainstream view is that high added or free sugar consumption leads to obesity and related metabolic diseases, controversies exist. This narrative review aims to highlight important findings and identify major limitations and gaps in the current body of evidence in relation to the effect of high sugar intakes on health. Previous animal studies have shown that high sucrose or fructose consumption causes insulin resistance in the liver and skeletal muscle and consequent hyperglycemia, mainly because of fructose-induced de novo hepatic lipogenesis. However, evidence from human observational studies and clinical trials has been inconsistent, where most if not all studies linking high sugar intake to obesity focused on sugar-sweetened beverages (SSBs), and studies focusing on sugars from solid foods yielded null findings. In our opinion, the substantial limitations in the current body of evidence, such as short study durations, use of supraphysiological doses of sugar or fructose alone in animal studies, and a lack of direct comparisons of the effects of solid compared with liquid sugars on health outcomes, as well as the lack of appropriate controls, seriously curtail the translatability of the findings to real-world situations. It is quite possible that "high" sugar consumption at normal dietary doses (e.g., 25% daily energy intake) per se-that is, the unique effect of sugar, especially in the solid form-may indeed not pose a health risk for individuals apart from the potential to reduce the overall dietary nutrient density, although newer evidence suggests "low" sugar intake (<5% daily energy intake) is just as likely to be associated with nutrient dilution. We argue the current public health recommendations to encourage the reduction of both solid and liquid forms of free sugar intake (e.g., sugar reformulation programs) should be revised due to the overextrapolation of results from SSBs studies.

Effect of lifelong sucrose consumption at human-relevant levels on food intake and body composition of C57BL/6N mice.

Introduction: Controversies surround the issue if chronic consumption of a high-sugar diet is detrimental to health or not. This study investigates whether lifelong consumption of a higher sucrose diet will induce overeating, and obesity, and cause metabolic dysfunctions such as hyperglycemia and dyslipidaemia in C57BL/6N mice, compared to a lower sucrose diet. Methods: Male C57BL/6N mice at 3 weeks of age were randomized into consuming a diet with 25 or 10% kcal from sucrose for the rest of their lives. Body weight, food and water intake, fasting blood glucose, insulin, and lipid levels were measured at regular intervals. At the end of the study, organs and tissues were collected and gene expression was measured. Results: There was no discernible difference in the impact on food intake, body composition, glucose and lipid homeostasis, liver triglyceride content, life expectancy, as well as gene expression related to intermediary metabolism between mice fed a diet with 10 vs. 25% kcal as sucrose over their lifespan. We also showed that switching from a 25% kcal diet to a 10% kcal diet at different life stages, or vice versa, did not appear to affect these outcomes of interest. Discussion: The results from our study suggest that lifelong consumption of a higher sugar diet generally did not induce overeating and obesity, disrupt carbohydrate metabolism and lipid homeostasis, and reduce life expectancy compared with a lower sugar diet. Our unorthodox findings disagreed with the popular belief that higher sugar consumption is detrimental to health, which should be confirmed in future studies.

A priori dietary patterns and cardiovascular disease incidence in adult population-based studies: a review of recent evidence.

Cardiovascular disease (CVD) is the leading cause of death worldwide. Studies using the a priori dietary pattern approach have been criticized for the heterogeneity of their scoring methods. This review aimed to evaluate the evidence on the association between a priori dietary patterns and CVD incidence in recent adult population-based studies. Similar to the conclusions of previous systematic reviews and meta-analyses, our review found that the majority of recent studies suggested that Mediterranean diet (6 out of 10 studies), Dietary Approaches to Stop Hypertension diet (4 out of 6 studies), Dietary Guidelines-based (11 out of 13 studies) patterns, and other emerging dietary patterns, including carbohydrate quality, Dietary Inflammatory Index, Plant-based Diet Index and Healthy Nordic Food Index, were cardioprotective in the general population; however, there was substantial heterogeneity among the studies, possibly due to differences in scoring methods and analytical approaches used and inclusion of different confounders, as well as other methodological drawbacks, such as low numbers of cases and short follow-up periods. Future studies should simultaneously examine and compare multiple a priori dietary patterns in a specific population using a uniform statistical approach. A consensus on the scoring methods for each a priori dietary pattern is also necessary.

Changes in the Presence of Nonnutritive Sweeteners, Sugar Alcohols, and Free Sugars in Australian Foods.

BACKGROUND: In parallel with growing consumer interest in reducing sugar intake, manufacturers have increased availability of food and beverage products containing nonnutritive sweeteners (NNSs). However, emerging evidence indicates that specific NNS types have differential effects on cardiometabolic health. OBJECTIVE: This study examined overall changes in the presence of NNSs, sugar alcohols, and free sugars in the Australian food supply and the use of specific NNS types. PARTICIPANTS/SETTING: Data for 21,051 products in 2015 and 21,366 products in 2019 were extracted from The George Institute's FoodSwitch database. MAIN OUTCOME MEASURES: The proportion of products containing NNSs, sugar alcohols, free sugars, and a combination of these, as well as proportion of products containing specific NNS types. STATISTICAL ANALYSES PERFORMED: Changes between 2015 and 2019 were examined using Pearson χ2 tests. RESULTS: Between 2015 and 2019, there was a significant increase in the proportion of food and beverage products containing NNSs (from 3.8% to 4.3%; P < .001) and a significant decrease in products containing free sugars (from 62.7% to 59.9%; P < .001),) driven primarily by nondairy beverages. There were changes in the use of specific NNS types between 2015 and 2019, with a large increase in the use of steviol glycosides (from 33.7% to 50.2%) and a large decrease in the use of sucralose (from 42.4% to 30.5%), aspartame (from 21.0% to 14.4%), and acesulfame K (from 57.4% to 27.7%) (P < .05 for all). CONCLUSIONS: These findings on the use of different NNS, sugar alcohol, and free sugar ingredients and combinations provide important research insights and will be useful in informing government policies that address sugars and other sweeteners in Australian foods.

A Machine Learning Approach to Predict the Added-Sugar Content of Packaged Foods.

BACKGROUND: Dietary guidelines recommend limiting the intake of added sugars. However, despite the public health importance, most countries have not mandated the labeling of added-sugar content on packaged foods and beverages, making it difficult for consumers to avoid products with added sugar, and limiting the ability of policymakers to identify priority products for intervention. OBJECTIVE: The aim was to develop a machine learning approach for the prediction of added-sugar content in packaged products using available nutrient, ingredient, and food category information. METHODS: The added-sugar prediction algorithm was developed using k-nearest neighbors (KNN) and packaged food information from the US Label Insight dataset (n = 70,522). A synthetic dataset of Australian packaged products (n = 500) was used to assess validity and generalization. Performance metrics included the coefficient of determination (R2), mean absolute error (MAE), and Spearman rank correlation (ρ). To benchmark the KNN approach, the KNN approach was compared with an existing added-sugar prediction approach that relies on a series of manual steps. RESULTS: Compared with the existing added-sugar prediction approach, the KNN approach was similarly apt at explaining variation in added-sugar content (R2 = 0.96 vs. 0.97, respectively) and ranking products from highest to lowest in added-sugar content (ρ = 0.91 vs. 0.93, respectively), while less apt at minimizing absolute deviations between predicted and true values (MAE = 1.68 g vs. 1.26 g per 100 g or 100 mL, respectively). CONCLUSIONS: KNN can be used to predict added-sugar content in packaged products with a high degree of validity. Being automated, KNN can easily be applied to large datasets. Such predicted added-sugar levels can be used to monitor the food supply and inform interventions aimed at reducing added-sugar intake.

Total and Free Sugar Levels and Main Types of Sugars Used in 18,784 Local and Imported Pre-Packaged Foods and Beverages Sold in Hong Kong.

There is limited information regarding the free sugar content of pre-packaged foods in Hong Kong. This study aims to assess the free sugar content and identify the most frequently used free sugar ingredients (FSI) in pre-packaged foods in Hong Kong. Data from 18,784 products from the 2019 FoodSwitch Hong Kong database were used in this analysis. Ingredient lists were screened to identify FSI. Total sugar content was derived from nutrition labels on packaging. Free sugar content was estimated based on adaptation of a previously established systematic methodology. Descriptive statistics of the total sugar and free sugar content, as well as the mean ± SD contribution of free sugar to total sugar of the audited products were calculated, stratified by food groups. Almost two-thirds (64.5%) of the pre-packaged foods contained at least one FSI. 'Sugar (sucrose)' was the most popular FSI that was found in more than half (54.7%) of the products. 'Fruit and vegetable juices' (median 10.0; IQR 8.3-11.5 g/100 mL) were found to have a higher median free sugar content than 'Soft drinks' (8.0; 6.0-10.6 g/100 mL). Mean ± SD contribution of free sugar to the total sugar content was 65.8 ± 43.4%, with 8 out of 14 food groups having >70% total sugar as free sugar. To conclude, free sugar, especially sucrose, was extensively used in a wide variety of pre-packaged products sold in Hong Kong. Further studies are needed to assess the population intake of free sugar in Hong Kong to inform public health policy on free sugar reduction.

Estimating the potential impact of the Australian government's reformulation targets on household sugar purchases.

BACKGROUND: Countries around the world are putting in place sugar reformulation targets for packaged foods to reduce excess sugar consumption. The Australian government released its voluntary sugar reformulation targets for nine food categories in 2020. We estimated the potential impact of these targets on household sugar purchases and examined differences by income. For comparison, we also modelled the potential impact of the UK sugar reduction targets on per capita sugar purchases as the UK has one of the most comprehensive sugar reduction strategies in the world. METHODS: Grocery purchase data from a nationally representative consumer panel (n=7,188) in Australia was linked with a large database (FoodSwitch) with product-specific sugar content information for packaged foods (n=25,261); both datasets were collected in 2018. Potential reductions in per capita sugar purchases were calculated overall and by food category. Differences in sugar reduction across income level were assessed by analysis of variance. RESULTS: In 2018, the total sugar acquired from packaged food and beverage purchases consumed at-home was 56.1 g/day per capita. Australia's voluntary reformulation targets for sugar covered 2,471/25,261 (9.8%) unique products in the FoodSwitch dataset. Under the scenario that all food companies adhered to the voluntary targets, sugar purchases were estimated to be reduced by 0.9 g/day per capita, which represents a 1.5% reduction in sugar purchased from packaged foods. However, if Australia adopted the UK targets, over twice as many products would be covered (n=4,667), and this would result in a more than four times greater reduction in sugar purchases (4.1 g/day per capita). It was also estimated that if all food companies complied with Australia's voluntary sugar targets, reductions to sugar would be slightly greater in low-income households compared with high-income households by 0.3 g/day (95%CI 0.2 - 0.4 g/day, p<0.001). CONCLUSIONS: Sugar-reduction policies have the potential to substantially reduce population sugar consumption and may help to reduce health inequalities related to excess sugar consumption. However, the current reformulation targets in Australia are estimated to achieve only a small reduction to sugar intakes, particularly in comparison to the UK's sugar reduction program.

An Innovative Machine Learning Approach to Predict the Dietary Fiber Content of Packaged Foods.

Underconsumption of dietary fiber is prevalent worldwide and is associated with multiple adverse health conditions. Despite the importance of fiber, the labeling of fiber content on packaged foods and beverages is voluntary in most countries, making it challenging for consumers and policy makers to monitor fiber consumption. Here, we developed a machine learning approach for automated and systematic prediction of fiber content using nutrient information commonly available on packaged products. An Australian packaged food dataset with known fiber content information was divided into training (n = 8986) and test datasets (n = 2455). Utilization of a k-nearest neighbors machine learning algorithm explained a greater proportion of variance in fiber content than an existing manual fiber prediction approach (R2 = 0.84 vs. R2 = 0.68). Our findings highlight the opportunity to use machine learning to efficiently predict the fiber content of packaged products on a large scale.