Increasing Access to Acceptable and Affordable Gluten-Free Baked Goods
Introduction
The demand for acceptable gluten-free (GF) food products has been steadily increasing since the early 2000s (Bael, 2015). In the past few years, the demand for commercial GF products has increased by 41%, which suggests that this interest may not be a simple food trend, but rather a life-long commitment to GF eating (Bael, 2015). One of the reasons some consumers choose to limit their intake of gluten is that GF eating is thought to provide health benefits and enhance weight loss (Arslain et al., 2021; Bascuñán et al., 2017). Other individuals are encouraged to completely avoid gluten because of a diagnosis of celiac disease (Bael, 2015; Naqash et al., 2017). Celiac disease is an autoimmune disorder of the gastrointestinal tract (Gandini et al., 2021; Levinson-Castiel et al., 2019; Naqash et al., 2017). Ingestion of gluten by individuals with this disease can result in damage to the intestinal mucosa, which reduces the ability of the intestines to absorb nutrients. Total avoidance of the gluten protein is necessary to reduce immediate symptoms of malabsorption and to reduce long-term damage to intestinal tissue (Anton & Artfield, 2008; Gandini et al., 2021; Naqash et al., 2017). The increase in demand for GF products has spurred an interest in research focused on wheat-flour substitutes capable of producing acceptable baked goods without gluten.
The U.S. Food and Drug Administration (FDA) states that food acceptability is based on safety, nutrition, quality, and value as perceived by the consumer (Gardner, 1992). Often, commercially available GF products are considered unacceptable by consumers because of reduced food quality, such as undesirable flavor and texture. The value of the product is also lacking, as the cost for commercially produced GF products is much higher than for gluten-containing products (Capacci et al., 2018). The protein gluten is responsible for the texture, volume, and taste of bread, cake, and cookie products (Anton & Artfield, 2008; Bender & Schönlechner, 2020; Naqash et al., 2017). There are no other grains that contain a protein that acts similarly to the gluten protein found in wheat. This creates challenges when attempting to produce baked goods that are acceptable to consumers. Several common grains have been ground and studied as potential flour substitutes, namely rice, oats, and corn (Gómez & Martínez, 2016; Susman et al., 2020). While these grains can produce a variety of baked goods, the consumer acceptability of these goods is often poor. However, research has shown that the use of flour blends combining one or more of these flours, with other additives, can produce acceptable baked products (Anton & Artfield, 2008; Bender & Schönlechner, 2020; Naqash et al., 2017; Susman et al., 2020). A few additives and ingredients that help mimic the effect of gluten in baked goods include hydrocolloids (commonly known as gums), dry milk solids, evaporated milk, gelatin, an additional egg, oil, or honey. These additives increase water-holding capacity in starch, which helps improve texture and tenderness. These additives also improve the structure and mouthfeel of baked goods (Anton & Artfield, 2008; Salehi, 2019).
Because demand for GF baked products has increased, prices of commercial GF baked goods have also risen (Capacci et al., 2018). Often, the price for GF products is 2 to 2½ times higher than for non-GF products. This often results in a 29% increase in food costs for those regularly consuming GF products (Capacci et al., 2018). Because of the combination of cost, limited availability, and poor quality of GF products, many individuals who cannot eat gluten have poor diet quality (Capacci et al., 2018). This can increase their risk of nutrient deficiencies, as individuals with celiac disease tend to eat 23% less food than those without the disease because limited food products are acceptable to them (Capacci et al., 2018; Gorgitano & Sodano, 2019).
The objectives of this research study were to determine 1) the feasibility of using commercially available flour blends to bake GF products at home, 2) the effect that GF commercial flour blends would have on overall consumer acceptability of GF home-baked goods (i.e., taste, texture, flavor, appearance, and aroma), 3) recipes that could be used by the average consumer to make affordable GF baked goods at home instead of purchasing commercially baked products.
Methods
Preliminary Testing
Preliminary testing began with basic recipes for bread, pizza crust, drop cookies, and rolled cookies (see Appendix 1) obtained from an introductory food science course (Utah State University, Introduction to Food Science Course). These products were chosen for testing because of the demand for GF versions of these products. Demand was determined based on personal experience of the investigators, inquiries from consumers in their counties, and the availability of similar commercially baked products. Extension faculty in four Utah counties used the same standard recipe to bake GF products using commercial GF flour blends available in their local grocery stores. These flours were selected because they claimed to have multipurpose use and were readily available for purchase, enabling Extension faculty to make recommendations that would be beneficial to home consumers across the state. Each Extension faculty member purchased two or three flour blends available in their area and did preliminary testing in their home or Extension office kitchen. Researchers recorded processes and outcomes during the preliminary testing and shared these data in a collective discussion. Conclusions from this discussion were used to decide necessary recipe modifications and preferred flour blends and to develop procedures for the in-lab testing that was conducted at Utah State University in Logan, Utah. Products to be tested were chosen from two different gluten-function categories, yeast breads (white bread and pizza dough) and cookies (drop cookies and rolled/refrigerated cookies).
Four flour blends were chosen for further testing. The product specifications can be found below:
- Cup4Cup Multipurpose Flour (Cup4Cup, LLC, Yountville, CA) claims a 1:1 substitution of wheat flour for GF flour in recipes for baked goods including cakes, cookies, quick breads, pie crust, sauces, scones, muffins, etc. The descending order of ingredient predominance by weight is cornstarch, white rice flour, brown rice flour, milk powder, tapioca flour, potato starch, and xanthan gum. The product is certified GF and comes in a 3-lb bag.
- King Arthur Measure for Measure Flour (King Arthur Baking Company, Inc., Norwich, Vermont) claims a 1:1 substitution of wheat flour for GF flour in recipes for baked goods including bread, muffins, cookies, cakes, brownies, pancakes, and other non-yeast recipes. The descending order of ingredient predominance by weight is rice flour, whole grain brown rice flour, whole sorghum flour, tapioca starch, cellulose, xanthan gum, and vitamin and mineral blend. The product is allergy-free and certified GF and comes in 1-lb and 3-lb bags.
- Extra White Gold (Better Foods Jeyer, LLC, Alpharetta, GA) is an all-purpose flour that claims to replace wheat flour in a 1:1 cup ratio in baking pastries, cupcakes, cakes, and pancakes. The descending order of ingredient predominance by weight is rice flour, potato starch, modified tapioca starch, tapioca starch, corn starch, pea protein, cellulose fibers, xanthan gum, sodium acid pyrophosphate, sodium bicarbonate, mono diglyceride (emulsifier), and salt. The product is certified GF and is sold in 15.9-oz packages.
- Grandpa’s Kitchen Flour Blend (Grandpa’s Kitchen, Spanish Fork, UT) is certified GF and claims a 1:1 substitution of wheat flour for GF flour in making baked goods. The descending order of ingredient predominance by weight is white rice flour, potato starch, corn starch, tapioca flour, xanthan gum, brown rice flour, and sorghum flour. The product is sold in 2-lb bags.
In-Lab Testing
Product Preparation and Method Standardization
Using the optimized recipes from the preliminary testing, standard operating procedures (SOPs) were established to minimize errors by the researchers. The typical process for SOPs was used to develop the specific procedures for this experiment (Stup, 2023). All researchers (n = 4) testing food products were trained on the SOPs before testing began. The SOPs and training included information on measuring ingredients accurately, using volume-to-volume substitutions for the flour blends, proper use of the stand mixers (KitchenAid Professional 600, Whirlpool Co., Benton Harbor, MI), and oven settings. The SOPs included step-by-step instructions on when to add each specific ingredient, mixing ingredients for exact time increments, scraping mixing bowls at exact times in the process, and portioning completed product using scoops or weight. Instructions for how long to bake the product and actions after baking (e.g., remove cookies to cooling rack) were given. Each step in the procedure was documented by the researcher and observations were recorded to ensure quality control.
Three complete replicates (n = 3) (where replicate represents a single, distinct batch) for each flour blend were completed for each product type. In total, there were twelve batches each of drop cookies, rolled cookies, bread, and pizza crust completed over a two-day period. The finished products were labeled and analyzed immediately or packaged in consumer-grade lowdensity polyethylene cling film (Glad Products Company, Oakland, CA) or zip-top bags (SC Johnson & Son, Inc., Racine, WI) designed for freezer storage. Products were held under frozen storage (10 °C (14 °F)) or room temperature (20 °C (68 °F)).
Bread. Each bread replicate was prepared following the established SOP. Dry yeast (12 g) (ACH Food Companies, Chicago, Il) and granulated sugar (38 g) (Walmart, Inc., Bentonville, AR) were added to 300 ml warm water (33°C (91°F)) to activate yeast. Flour blend (700 g), Canola oil (75 g) (Walmart, Inc., Bentonville, AR), eggs (150 g) (Walmart, Inc., Bentonville, AR), and salt (9 (Morton Salt. Chicago IL) were combined with the yeast mixture. Ingredients were mixed in a mixer (KitchenAid, Benton Harbor, MI) for 2 min at speed 5 using the paddle attachment to form a stiff batter. The batter was spooned into a 9- x 5-in loaf pan sprayed with cooking spray (Walmart Inc., Bentonville, AR) and placed in a proof box (Avantco Equipment, Meridian, ID) (80 °C (175 °F)) until doubled in size (20-25 min). Bread was baked in a preheated 190 °C (375 °F) oven to an internal temperature of 74 °C (165 °F). The bread was allowed to cool 5 min in the pan before being removed from the pan.
Bread was cooled on a cooling rack at room temperature (20 °C (68 °F)) for 1-2 hr before it was sliced into ½-in slices and analyzed or labeled and prepared for storage in zip-top freezer bags. Three slices from the center of each loaf were randomly assigned to one storage treatment: no storage (analyzed 2 hr after removing from oven); room temperature storage (20 °C (68 °F)) for 2 days before analysis; or frozen storage (-10 °C (14 °F)) for 1 month, then thawed at room temperature for 24 hr before analysis.
Pizza Crust. Each pizza crust replicate was prepared following the established SOP. Dry yeast (8 g) (ACH Food Companies. Chicago, IL) and granulated sugar (3 g) (Walmart, Inc., Bentonville, AR) were added to 60 ml warm water (33 °C (91 °F)) to activate yeast. Flour blend (360 g) and salt (5 g) (Morton Salt. Chicago, IL) were combined with the yeast mixture and mixed in a mixer (KitchenAid, Benton Harbor, MI) for 2 min at speed 5 using the paddle attachment. Olive oil (40 g) (Walmart, Inc., Bentonville, AR) and remaining water (15 ml) were added and mixed for 2 min at speed 10 to form a stiff batter. Batter was placed in a proof box (80 °C (176 °F)) until doubled in size (20-25 min). Raised dough was divided into two portions of equal weight to prepare two crusts from each batch. Each portion was placed between two pieces of parchment paper (Walmart, Inc., Bentonville, AR) and rolled to a ¼-in thickness. Rolling guides were used to ensure a uniform thickness across the entire crust. The top piece of parchment paper was removed, and the crust was transferred to a baking sheet that had been preheated in the oven for 5 min. The pizza crusts were baked in a preheated 205 °C (400 °F) oven. One crust per batch was parbaked for 7 min and removed from the oven. The second crust was baked for 15 min and removed from the oven.
Pizza crusts were cooled for 30-40 min at room temperature (20 °C (68 °F)) until completely cool to the touch. Fully baked crusts were analyzed immediately after cooling. Parbaked crusts were labeled and prepared for storage in consumer-grade, low-density polyethylene cling film and stored at freezer temperatures (-10 °C (14 °F)) for 6 weeks. After storage, the crusts were thawed at room temperature (20 °C (68 °F)) for 4 hr, cooked for an additional 7-8 min in a preheated 205 °C (400 °F) oven, and cooled as described above before analysis.
Drop Cookies. Each drop cookie replicate was prepared following the established SOP. Room-temperature salted butter (60 g) (Walmart, Inc., Bentonville, AR) and shortening (50 g) (The J.M. Smucker Company, Orrville, OH) were mixed using a mixer (KitchenAid, Benton Harbor, MI) at speed 5 using the paddle attachment to completely combine ingredients. Next, granulated sugar (100 g) (Walmart, Inc., Bentonville, AR) and brown sugar (110 g) (Walmart Stores, Inc., Bentonville, AR) were added and creamed for 1 min at speed 10. One egg (50 g) (Walmart, Inc., Bentonville, AR) was added and mixed at speed 5 just to incorporate. Most of the flour blend (240 g) (Walmart, Inc., Bentonville, AR), baking soda (3 g), (Walmart, Inc., Bentonville, AR), and salt (1.5 g) (Morton Salt, Chicago, IL) were whisked together in a separate bowl to combine, added to mixer, and mixed at speed 2 for 1 min. The remaining flour blend (120 g) and chocolate chips (200 g) were stirred in by hand to combine. The dough was scooped onto a parchment-lined (Walmart, Inc., Bentonville, AR), rimmed baking sheet using a #30 commercial scoop (1.22 oz). Cookies were baked in a preheated 177 °C (350 °F) oven for 10-11 min until cookies spread and had a slight amount of brown color on top.
Cookies were removed from the oven, and the pan was lightly tapped on the counter to deflate cookies. The cookies were allowed to cool on the pan for 5 min and were then removed to a cooling rack. They were left to cool on the cooling rack until they were cool to the touch. Four cookies from each replicate were reserved for immediate analysis or were labeled and prepared for room-temperature storage (20 °C (6 8°F)) in zip-top freezer bags.
Rolled Cookies. Each rolled cookie replicate was prepared following the established SOP. Butter (114 g) (Walmart, Inc., Bentonville, AR) and granulated sugar (100 g) (Walmart, Inc., Bentonville, AR) were creamed together using a mixer (KitchenAid, Benton Harbor, MI) for 1 min at speed 10 using the paddle attachment. One egg (50 g) (Walmart, Inc., Bentonville, AR) and vanilla (2 g) (Walmart, Inc., Bentonville, AR) were added and mixed at speed 5 just to incorporate. Half of the flour blend (240 g) (Walmart, Inc., Bentonville, AR), powdered sugar (20 g) (Walmart, Inc., Bentonville, AR), and baking powder (4 g) (Walmart, Inc., Bentonville, AR) were whisked together in a separate bowl to combine. This mixture was then incorporated into the creamed mixture at speed 2 for 1 min. The remaining flour blend (240 g) was stirred in by hand to combine, then the dough was gently kneaded by hand for 5 turns. Dough was placed between two sheets of parchment paper (Walmart, Inc., Bentonville, AR) and rolled to a ¼-in thickness using rolling guides. Round cookie cutters (2-in diameter) were used to cut cookies from the initial roll of dough and placed on parchment paper on a rimmed baking sheet. No rerolled dough was used. Cookies were baked in a preheated 163 °C (325 ° F) oven for 6-7 min.
Cookies were removed from the oven and allowed to cool on the pan for 5 min before being moved to a cooling rack. They were left to cool at room temperature until they were cool to the touch. Four cookies from each replicate were reserved for immediate analysis or were labeled and prepared for room-temperature storage (20 °C (68 °F)) in zip-top freezer bags.
Physicochemical Properties
Physicochemical properties were evaluated for each product as described below. Breads were analyzed for moisture, crust color, interior color, cross-sectional (slice) area, and air cell size. Pizza crusts were analyzed for moisture, color, and height. Drop cookies were analyzed for moisture, color, and spread. Rolled cookies were analyzed for moisture, color, height, and spread.
Moisture. Moisture content was measured using a programmable moisture balance with a ceramic heating element (MA150 Sartorius Mechatronics, Bohemia, NY). The moisture balance heats samples individually in a tared pan until no additional loss in weight is detected during drying, at which point the program ends and results are displayed as the percentage of water in the sample. Measurements were taken the same day as baking after products had cooled completely. A section from the center of the product (bread, pizza crust, rolled cookie, or drop cookie) was cut into rough crumbs using a bread knife immediately before analysis, then a 2-3 g portion was added to the tared pan and heated at 110 °C (230 °F) until a constant weight was reached (Zhong et al., 2014).
Color. Color values were measured using a portable colorimeter (Hunter Lab Miniscan, Reston, VA) with a 5 mm diameter aperture, set to use illuminant D-65. The colorimeter was standardized through a single layer of low-density polyethylene film using both white and black standard tiles. CIELAB values were recorded, L* (lightness; L*=100 indicating white and L*=0 indicating black), a* (red/green; +a* indicating more red and -a* indicating more green), and b* (yellow/blue; +b* indicating more yellow and -b* indicating more blue). Chroma (color intensity) was calculated as [√ (a*2 + b*2)]. Hue angle (true color) was calculated as [arctangent (b*/a*)] and transformed into a four-quadrant (360°) system to facilitate statistical analysis (McLellan et al., 1995).
All color measurements were taken the same day that the products were prepared. For bread and pizza crust, measurements were taken at three random locations across the crust. Interior color was also taken for bread, across the face of a slice from the middle of each loaf. For dropped and rolled cookies, single measurements were taken on each of three randomly selected cookies from each batch.
Cross-Sectional Area and Air Cell Size. Cross-sectional area was measured by tracing a slice of bread taken from the center of the loaf onto standard graphing paper and counting the squares inside the tracing. The center slice was also photographed against a reference grid, with squares of exactly 4 square millimeters. The magnetic lasso tool in Photoshop CC 2015 (Adobe, Inc., San Jose, CA) was used to highlight a reference square from the grid and five individual cells from the slice within each individual photograph (Allen et al., 2007). Pixel counts within each of the highlighted cells were recorded and converted to square millimeters, based on the pixel count for the reference square.
Height. Rolled cookies (two per batch) and pizza crust were cut across the diameter for height measurements. Height was taken at the center of the cut surface using a digital caliper (Carrera Precision CP9807-TF, Max Tool LLC, La Verne, CA).
Spread. Three cookies from each batch (drop cookies and rolled cookies) were centered on a standard line spread template with concentric rings every 5 mm (Zhong et al., 2014). Spread was measured for each cookie as the average of four measurements, one from each of the quadrants of the line spread template.
Textural Properties
Textural properties were evaluated for each product as described below, using a TSM-Pro Texture Analyzer (Food Technology Corporation, Sterling, VA) equipped with a 50-Newton (N) load cell. Breads were analyzed by texture profile analysis (TPA) for hardness (peak force to compress, penetrate, or break a sample) and springiness (sample recovery or “bounce back” after an initial compression). Pizza crusts and cookies were analyzed for hardness (peak force).
Bread. TPA was conducted on three separate slices of bread taken from the center portion of each loaf. One slice was tested the same day as baking, one after 2 days of storage at room temperature (20 °C (68 °F)) in a zip-top freezer bag, and one after freezing for 1 month at -10 °C (14 °F) in a zip-top freezer bag and then thawing for 24 hr at room temperature (20 °C (68 °F)). A 7.5 cm diameter plate was used to compress bread cubes (2 x 2 x 2 cm) at a crosshead speed of 50 mm/min to 50% of the original height of the sample, then withdrawn to a fixed height to allow an approximate resting time of 25 s before a second compression cycle. Hardness was taken as the peak force in N in the first compression cycle (force required to compress the sample to half its original height). Springiness was calculated as the ratio (L2/L1) of the time from zero to max force for the second compression peak (L2) divided by the time from zero to max force for the first compression peak (L1) (the degree to which a sample recovers its original shape after an initial compression) (Armero & Collar, 1997).
Pizza Crust. Hardness was conducted on both crusts made from the same batch of dough. Crust 1 was fully baked, cooled, and tested on the same day. Crust 2 was parbaked, and frozen for 6 weeks at -10 °C (14 °F) and then thawed at room temperature (20 °C (68 °F)) for 2 hr, completed baking at 205 °C (400 °F), cooled, and tested. A ¼-in rounded-end probe was used to compress crusts at a crosshead speed of 50 mm/min until they reached 50% of their original height. Hardness was taken as the amount of force in N required to compress the crust to half its original height.
Cookies. Hardness was conducted for both drop cookies and rolled cookies on two cookies per batch, one on the same day as baking and one after 2 days of room temperature 20°C (68°F) storage in a zip-top freezer bag. A ¼-in rounded-end probe was used to compress cookies at a crosshead speed of 50 mm/min until they fractured or reached 50% of their original height, whichever occurred first. Hardness was taken as the peak force in N recorded for each cookie.
Statistical Analysis
Statistical analysis was performed using SAS version 9.4 (SAS Institute, Inc., Cary, NC). The effect of flour blend was evaluated by analysis of variance using the proc GLM function with the Tukey adjustment for multiple means comparison. For stored samples, additional statistical analysis was conducted to evaluate the change in texture during storage. Time series analysis was performed using the proc Mixed function with an autoregressive moving average (1,1), with flour blend and storage time as fixed effects. For all analyses, statistical significance was identified at the 95% confidence level (p < 0.05). Three complete replicates were performed for each flour blend for each product.
Sensory Evaluation
One additional batch of each product, using each flour blend, was made and reserved for sensory evaluation. Researchers (n = 6) completed sensory evaluation in the testing kitchen after baking was completed for each product. Researchers prepared their own samples and were in the same room during sampling but recorded observations on individual data sheets. Each researcher tasted a small portion of a drop cookie made from each flour blend and a rolled cookie made from each flour blend. The bread for each flour blend was evaluated both independently and as part of a meal (i.e., researchers made sandwiches with the bread, either a ham and cheese sandwich or peanut butter and jelly). The pizza crust for each flour blend was evaluated in a similar way, the crust by itself and the crust with standard pizza toppings (pizza sauce, cheese, and pepperoni). Researchers recorded their experience for each flour type and baked product in a chart that gathered information on flavor, texture, and appearance. Each attribute was rated using a 5-point hedonic scale (5 = like extremely, 3 = neither like nor dislike, and 1 = dislike extremely). Researchers provided qualitative feedback on the experience to better understand whether they preferred one specific product over another and, if so, why.
Consumer Testing
Approval for research was given by Utah State University IRB #12631. A consumer recruitment advertisement was posted on a popular Facebook group that focused on GF subjects. Members of this group volunteered (n = 153) to be part of the consumer testing by responding to the social media post with their email address. Information for all individuals who volunteered was placed in a spreadsheet and given a number. Funding available for this research allowed for twenty study participants (n = 20) to participate in the consumer testing portion of this study. An online random-number-generating program (www.random.org) was used to randomly assign 20 participants (plus 5 alternates) to the study. Participants were provided the informed consent form and other study information through their email address. Participants were asked to respond to the email indicating their agreement to the informed consent and to provide their mailing address in order to receive the food kit, recipes, and product evaluation data sheets.
Food kits were prepared (ingredients were weighed and packaged into individual zip-top bags) in a commercial kitchen lab at Utah State University and included all ingredients needed to complete the recipe except perishable ingredients (e.g., butter, shortening, eggs). Each participant was given ingredients to make all four recipes (bread, pizza crust, drop cookies, and rolled cookies). One flour blend per product was chosen by researchers based on scientific and preliminary sensory testing results (bread – Cup4Cup, pizza crust – King Arthur, drop cookies – King Arthur, rolled cookies – Extra White Gold). Two weeks after food kits were mailed, participants were emailed a link to a Qualtrics online survey to record the experimental data they collected while baking the products. Participants were also asked to conduct an informal sensory evaluation of products and provide data in answer to specific questions in the Qualtrics online survey. Consumer responses for the quality of baked goods were measured using a hedonic scale. The hedonic scale was defined as follows for a scale of 1-5: 5 = very appealing,4 = appealing, 3 = neither appealing nor unappealing, 2 = unappealing, and 1 = very unappealing.
Participants who completed the survey received a complimentary electronic kitchen scale. See Appendix 2 for survey information. The survey was designed to gather data similar to the data collected during the in-lab testing. Due to the small number of external participants, surveys were not validated before external consumer testing.
Results
Bread
Though flour blend had a significant effect on the air-cell cross-sectional area (indicating expansion) of the bread slices (p = 0.02), no significant differences were seen between the flour blends (p > 0.05). Cup4Cup showed higher expansion, with approximately 20% larger surface area than the other flour blends except for White Gold (p > 0.05). Overall, the standard deviation shows that air cell size was highly variable within a single slice for all flour blends. See Table 1 for specific values.
Table 1. Effect of Gluten-free Flour Blends on Bread Air Cell Size
| Flour Blend | Air Cell Size (mm2) Mean ± Std Dev | ||
|---|---|---|---|
| Cup4Cup | 10.84 ± 7.88 | a | |
| White Gold | 6.20 ± 5.40 | a | b |
| Grandpa's Kitchen | 5.04 ± 2.20 | b | |
| King Arthur | 3.14 ± 1.23 | b |
Though air cell size is often related to bread hardness and the tendency to stale quickly, no correlation was seen between these parameters in this study. Also, no significant differences (p > 0.05) in moisture content were seen between loaves from different flour blends. All loaves were between 40% and 46% moisture. However, moisture content was inversely correlated (p = 0.04) with hardness, though the relationship was not linear (r = -0.59). Moisture content was positively correlated (p = 0.01) with springiness; however, the relationship was weakly linear (r = 0.69).
In Table 2, the type of flour blend did not have a significant effect on the hardness of loaves, regardless of when testing occurred (day of baking or after storage). However, Grandpa’s Kitchen tested higher for hardness, though not significantly different (p > 0.05), than the other flour blends. And as a general trend, hardness decreased for all flour blends after being frozen. The type of flour blend did have a significant effect (p = 0.02) on the amount of springiness observed, though the relationship was weakly linear (r = 0.69). The Cup4Cup flour blend showed higher springiness than the other flour blends on Day 1 and after freezing. Though springiness was not significantly correlated with the cross-sectional area (p > 0.05), this general observation is consistent between these two measurements. Overall, flour blend (p = 0.003) and storage time/technique (p = 0.007) had a significant effect on springiness, but there was no interaction effect.
Table 2. Effect of Gluten-free Flour Blend and Time/type of Storage on Bread Texture
| Day 1 | Day 3 (Room Temperature) | 1 month (Frozen) | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Hardness (N) | |||||||||
| Grandpa's Kitchen | 15.87 ± 4.76 | a | 15.96 ± 3.43 | a | 13.04 ± 0.09 | a | |||
| Cup4Cup | 9.29 ± 6.27 | a | 6.66 ± 3.90 | a | 5.08 ± 1.45 | a | |||
| White Gold | 9.20 ± 2.94 | a | 6.65 ± 1.07 | a | 5.65 ± 1.62 | a | |||
| King Arthur | 9.01 ± 2.37 | a | 11.16 ± 2.08 | a | 8.29 ± 0.59 | a | |||
| Springiness (%) | |||||||||
| Cup4Cup | 89.67 ± 5.01 | a | 77.79 ± 8.99 | a | 80.73 ± 11.25 | a | |||
| White Gold | 81.93 ± 1.99 | a | b | 59.36 ± 0.92 | a | b | 63.16 ± 9.02 | a | b |
| King Arthur | 76.34 ± 6.84 | b | 55.54 ± 9.70 | a | b | 52.80 ± 17.91 | b | ||
| Grandpa's Kitchen | 74.84 ± 2.63 | b | 50.61 ± 3.39 | b | 47.25 ± 0.24 | b | |||
The color of the bread varied significantly between flour blends, as seen in Table 3. However, these variations were difficult to detect with the naked eye. Test results suggest that crust color was most consistent in the Grandpa’s Kitchen flour blend and the White Gold flour blend. The interior color of each flour blend varied significantly from the other flour blends with no clear association between any two flour blends.
Table 3. Effect of Gluten-free Flour Blend on Bread Color
| Chroma (Color intensity) | Hue Angle (True color) | Lightness (L*) | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Crust Color | ||||||||||
| Grandpa's Kitchen | 30.17 ± 6.47 | a | 72.19 ± 2.12 | a | 57.05 ± 2.03 | a | ||||
| White Gold | 29.50 ± 5.64 | a | 67.22 ± 4.24 | a | b | 55.04 ± 5.00 | a | |||
| King Arthur | 27.29 ± 2.45 | a | 31.73 ± 1.07 | b | 47.59 ± 4.84 | a | b | |||
| Cup4Cup | 21.41 ± 12.59 | a | 53.56 ± 11.23 | b | 41.27 ± 7.55 | b | ||||
| Interior Color | ||||||||||
| White Gold | 17.50 ± 0.96 | a | 98.6 ± 2.7 | c | 65.99 ± 2.79 | a | ||||
| King Arthur | 14.84 ± 0.46 | a | b | 101.2 ± 1.5 | b | c | 67.89 ± 1.63 | a | ||
| Grandpa's Kitchen | 14.43 ± 1.78 | b | 103.9 ± 2.0 | a | b | 68.01 ± 2.00 | a | |||
| Cup4Cup | 12.40 ± 0.66 | b | 106.3 ± 1.0 | a | 60.95 ± 0.95 | b | ||||
Pizza Crust
No significant differences were seen between pizzas from different flour blends for moisture content (p > 0.05). All pizza crusts were between 31% and 35% moisture. Also, no significant differences in hardness were seen between flour type or how the crust was processed (p > 0.05) (e.g., the crusts fully baked on the same day gave similar texture readings as the crusts that were parbaked and stored frozen before being baked completely). Chroma (p = 0.01) and hue angle (p = 0.002) varied significantly between flour blends, but no significant effect was seen for lightness (p > 0.05). However, from a practical standpoint, this color variation was only slightly perceptible to the naked eye. Hue angle was positively correlated with moisture (p = 0.03, r = 0.43) and negatively correlated with hardness (p = 0.0009, r = -0.37).
Drop Cookies
No significant differences were seen for moisture content in cookies from different flour blends (p > 0.05). All cookies were between 5% and 7% moisture. Overall, the flour blend had a significant effect on hardness (p = 0.003): Grandpa’s Kitchen and King Arthur flour blends varied from the other flour blends on Day 1, and Grandpa’s Kitchen was significantly harder than all other flour blends at Day 3 (p < 0.05). However, no significant effect of storage was observed (p > 0.05) for fixed effect of storage time. Flour type also had a significant effect on the extent to which cookies spread, with King Arthur cookies spreading less (p < 0.05) (smaller diameter) than other flour blends. See Table 4 for specific measurements.
Table 4. Effect of Gluten-free Flour Blend and Time/type of Storage on Drop Cookie Hardness and Spread
| Flour Blend | Mean ± Std Dev | |||||
|---|---|---|---|---|---|---|
| Hardness (N) | ||||||
| Day 1 | Day 2 | |||||
| Grandpa's Kitchen | 28.1 ± 14.7 | a | 36.9 ± 4.3 | a | ||
| King Arthur | 17.5 ± 7.0 | a | b | 25.5 ± 8.0 | b | |
| Cup4Cup | 10.4 ± 3.7 | b | 18.0 ± 4.8 | b | ||
| White Gold | 8.6 ± 3.2 | b | 14.0 ± 5.7 | b | ||
| Spread of Cookies During Baking (mm) | ||||||
| White Gold | 5.49 ± 0.51 | a | ||||
| Grandpa's Kitchen | 5.39 ± 0.43 | a | ||||
| Cup4Cup | 4.71 ± 1.08 | a | ||||
| King Arthur | 3.58 ± 0.46 | b | ||||
In Table 5, chroma (p = 0.006) varied significantly between Cup4Cup and King Arthur flour blends and between White Gold and Grandpa’s Kitchen flour blends, but no significant effect was seen for hue angle and lightness (p > 0.05) between flour blends. From a practical standpoint, very little color variation was visible to the naked eye. Also, hue angle was positively correlated with hardness (p = 0.03, r = 0.43) and moisture content (p = 0.003, r = 0.56) and negatively correlated with spread (p = 0.0009, r = -0.37).
Table 5. Effect of Gluten-free Flour Blend on Drop Cookie Color
| Chroma (Color intensity) | Hue Angle (True color) | Lightness (L*) | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Cup4Cup | 17.20 ± 4.53 | a | 80.42 ± 5.88 | a | 57.31 ± 10.15 | a | ||||
| King Arthur | 17.07 ± 2.94 | a | 84.07 ± 1.43 | a | 58.90 ± 7.33 | a | ||||
| White Gold | 14.78 ± 3.10 | b | 81.63 ± 2.60 | a | 56.76 ± 6.29 | a | ||||
| Grandpa's Kitchen | 13.92 ± 2.08 | b | 84.08 ± 2.71 | a | 61.49 ± 6.71 | a |
Rolled Cookies
No significant difference in moisture content was noted between cookies made from the different flour blends (p > 0.05). All cookies were between 10% and 11% moisture. However, moisture was negatively correlated with height (p = 0.01, r = -0.50). Flour blend had a significant effect on hardness between the Cup4Cup and King Arthur flour blends and between the White Gold and Grandpa’s Kitchen flour blend at Day 1 (p < 0.05). And King Arthur was significantly different at Day 3 (p = 0.003) from all other blends, but no fixed effect of storage time was observed (p > 0.05). See Table 6.
Table 6. Effect of Gluten-free Flour Blend and Time/type of Storage on Rolled Cookie Hardness and Spread
| Flour Blend | Mean ± Std Dev | |||||||
|---|---|---|---|---|---|---|---|---|
| Hardness (N) | ||||||||
| Day 1 | Day 3 | |||||||
| King Arthur | 6.47 ± 2.42 | a | 7.16 ± 2.22 | a | ||||
| Cup4Cup | 5.91 ± 2.56 | a | 6.15 ± 2.35 | a | b | |||
| White Gold | 2.22 ± 1.15 | b | 2.44 ± 1.23 | b | ||||
| Grandpa's Kitchen | 2.06 ±0.41 | b | 2.42 ± 0.83 | b | ||||
| Spread of Cookie During Baking (mm) | ||||||||
| White Gold | 11.82 ± 0.66 | a | ||||||
| Grandpa's Kitchen | 10.90 ± 0.47 | b | ||||||
| Cup4Cup | 10.07 ± 0.14 | c | ||||||
| King Arthur | 9.65 ± 0.36 | d | ||||||
Flour type also had a significant effect (p = 0.001) on the extent to which cookies spread, with King Arthur cookies spreading less (smaller diameter). White Gold cookies had significantly more spread than the other flour blends (p < 0.05). See Table 6.
No differences were seen for height (p > 0.05). All cookies rose 4-5 times in height (from ¼-inthick dough to 1- to 1.2-in-thick cookies). Lightness (p = 0.0002) varied significantly between White Gold and all other flour blends. Chroma (p < 0.0001) varied significantly between White Gold and Grandpa’s Kitchen and King Arthur and Cup4Cup (p < 0.05). Also, a significant effect (p < 0.05) was seen for hue angle between Cup4Cup and Grandpa’s Kitchen and King Arthur and White Gold flour blends (p < 0.05). Despite these differences, very little color variation was visible to the naked eye. Chroma was positively correlated with moisture content (p = 0.003, r = 0.56) and height (p = 0.05, r = 0.41), but negatively correlated with spread (p = 0.004, r = -0.33). However, as mentioned, this has no practical implications, as very little color variation was visible to the naked eye. See Table 7.
Table 7. Effect of Gluten-free Flour Blend on Rolled Cookie Color
| Chroma (Color intensity) | Hue Angle (True Color) | Lightness (L*) | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| King Arthur | 15.87 ± 1.16 | a | 88.29 ± 0.64 | c | 74.22 ± 5.12 | a | ||||
| Cup4Cup | 14.97 ± 0.74 | a | b | 91.55 ± 0.43 | a | 73.20 ± 1.84 | a | |||
| White Gold | 14.35 ± 1.21 | b | 89.78 ± 0.55 | b | 67.36 ± 11.25 | b | ||||
| Grandpa's Kitchen | 14.13 ± 0.95 | b | 91.14 ± 0.65 | a | 77.19 ± 2.82 | a |
Consumer Testing
Sixteen out of twenty (n = 16, 80%) consumers who were sent a food kit completed the glutenfree baking and survey as part of the consumer testing stage of this study. All consumers were provided a baking kit through the mail. Four participants (n = 4) did not complete the survey within the allotted study time frame: these participants were emailed twice to determine their willingness to participate, with no response.
Survey results showed that participants’ total experience of making GF baked products at home ranged from less than 1 year (n = 4) to 2-3 years (n = 2) to 4-6 years (n = 4) to more than 6 years (n = 6) The frequency of making GF baked products at home ranged from daily (n = 2) to 2-3 times per week (n = 4) to 1 time per week (n = 4) to 2-3 times per month (n = 4) to once per month (n = 2). There was no correlation between the amount of experience a participant had and the frequency with which they made GF products.
Bread. The consumer acceptability of the finished bread loaf was highly correlated with flavor (r = 0.93, p < 0.0001), suggesting that the flavor of the finished bread was more important to consumers than the texture. No significant correlation (p > 0.05) was noted for this variable. The hedonic testing results suggested the bread was acceptable to most of the consumers based on the following mean values: appearance 5, taste 4.4, aroma 4.7, texture 3.8, and overall acceptance 4.7. Also, 10 of 16 (63%) consumers found the bread recipe instructions to be very clear. Overall, 13 of 16 (81%) participants indicated they would make the bread recipe again.
Pizza. The workability of the pizza crust dough (p = 0.01, r = 0.60) was just as important in determining whether participants would reuse the pizza crust recipe as the sensory characteristics of appearance (p = 0.02, r = 0.59), flavor (p = 0.0009, r = 0.75), and texture (p = 0.05, r = 0.50). Overall acceptability of the pizza crust was correlated with appearance (p = 0.04, r = 0.52), flavor (p = 0.0002, r = 0.80), and texture (p = 0.05, r = 0.50), but was not correlated with aroma. It is important to point out here that the aroma was rated lowest on the hedonic scale as 2.2, which is unappealing. The other hedonic values were taste 1.6, texture 1.0, and overall acceptability 1.3.
Of the consumers, 10 of 16 (63%), found the pizza crust recipe to be clear, but 11 of 16 (69%) indicated the baking time was too long. Participants (n = 7) commented that the recipe was unnecessarily complicated. Also, the dough quality was not as described in the recipe, as reported by participants (n = 7). Overall, 11 of 16 (69%) participants indicated they would not be likely to use this recipe in the future.
Drop Cookies. Product acceptability analysis of the drop cookies showed that appearance (p = 0.01, r = 0.60), flavor (p = 0.007, r = 0.64), and overall acceptability (p = 0.04, r = 0.52) of cookies were correlated with intent to reuse the recipe. Texture was highly positively correlated with flavor (p = 0.0008, r = 0.75), and both parameters were highly positively correlated with overall acceptability (texture: p = 0.0005, r = 0.77; flavor: p < 0.0001, r = 0.83), suggesting these characteristics (flavor and texture) were more important to participants than other sensory characteristics (appearance and aroma). Most consumers, 10 of 16 (63%), found the drop cookie recipe to be clear or very clear (instructions for when to add vanilla extract were inadvertently omitted from the recipe, 7 of 16 participants commented on this), though the baking time was too short. All sensory evaluation characteristics were found to have high averages on the hedonic scale: appearance 4.7, taste 5, texture 4.3, aroma 5, and overall acceptability 5. Of the consumers who responded, 14 out of 16 (88%) said they would be likely or very likely to use this recipe in the future.
Rolled Cookies. All sensory characteristics for rolled cookies were correlated with intent to reuse the recipe: appearance (p = 0.009, r = 0.63), flavor (p < 0.001, r = 0.90), texture (p = 0.002, r = 0.72), aroma (p = 0.003, r = 0.70), and overall acceptability (p = 0.04, r = 0.52). The appearance was inversely correlated with baking time (p = 0.04, r = -0.52), suggesting that cookies baked for less time were more appealing. Overall acceptability was highly correlated with all other sensory characteristics: appearance (p = 0.0006, r = 0.76), flavor (p < 0.0001, r = 0.83), texture (p = 0.0004, r = 0.78), and aroma (p = 0.0008, r =- 0.75). Consumers found the rolled cookie recipe to be very clear, though 12 of the 16 (75%) responses stated the cooking time was not long enough. Overall, 11 of the 16 (68%) participants indicated they would be only slightly likely to use this recipe in the future. The sensory evaluation characteristics were found to have medium averages on the hedonic scale: appearance 4.3, taste 3.2, texture 2.5, aroma 4.1, and overall acceptability 3.2.
Discussion
The results of this study provide valuable information to better inform consumers concerning preferred techniques for making acceptable GF baked goods. We will focus on product tenderness, texture, color, and flavor to provide understanding of overall acceptability for each baked good.
Bread
The Cup4Cup flour blend had significantly different springiness (p < 0.05) than the other flour blends (King Arthur, White Gold, and Grandpa’s Kitchen). One reason for this could be the higher amount of cornstarch by weight in the Cup4Cup flour blend. In gluten-containing bread, protein networks expand as bread is baked, then gelatinized and partially gelatinized starch molecules “set” the texture (this prevents further expansion as well as the collapse of the proteinbased foam network) (Horstmann et al., 2017; Khatkar et al., 1995). In GF bread, the gluten network would be replaced by gums (e.g., xanthan) and other protein substitutes (e.g., non-wheat flours, pea protein). During storage, water is lost as starches begin to stale (retrogradation). Though this may not result in a noticeable increase in hardness (at least during the preliminary stages), retrogradation may be preventing these protein foams from rebounding to their original size. These research results indicate that cornstarch molecules have the ability to gel, thicken, and stabilize gluten-free bread structures (Horstmann et al., 2017). This type of action produces a product more similar to gluten-containing breads and increases consumer acceptability of the product.
One result of note was the hardness results for bread (Table 2). These suggest that, regardless of the flour blend used, storage in zip-top freezer bags did not increase the hardness of the bread. In fact, this type of storage maintained or reduced the effects of storage, particularly when stored in the freezer. Based on these results, it is recommended that consumers bake the bread, cool it completely, and then slice it into uniform slices. These slices can then be packaged in freezer bags and stored in the freezer to preserve the optimal texture of the bread. Another item to note is that the springiness, or “bounce back,” results for the bread suggested that, despite the flour blends and time of storage, springiness decreased when compared to freshly baked bread. This relationship was not significant (p > 0.05), but the general trend of results suggests a general trend toward less springiness over time. This is consistent with preliminary starch retrogradation (staling), which may not be sufficient to result in a noticeable change in hardness but prevents the bread from “bouncing back.” The amount of starch in the flour blend may be one reason for the change in bounce back and staling over time. Starch also has the ability to prevent increased staling in these types of products, which would contribute to the increased hardness of bread over time (Horstmann et al., 2017).
The color of the bread also varied significantly between several of the flour blends (Table 3). The variations noted in each loaf are due to the ingredients included in the flour blends. Because the grains used in these flour blends tend to be a lighter color than wheat, they produce lighter baked goods than traditional wheat baked goods. Also, flour additives can change the color of the baked goods during baking (Alsaiqali et al., 2023; Gustafson, 2016; Vilmane & Straumite, 2014). Some of these additives include butter, gums, and milk powder. The Cup4Cup flour blend was the only blend studied that contained milk powder and registered a reddish-brown crust, which is consistent with this additive (Gustafson, 2016). This could suggest that milk powder will change the color of baked goods to be more consistent with wheat flour.
Pizza Crust
The results obtained for the pizza crust in this study showed that, regardless of the flour blend used, par-baking and freezing for future use did not change the texture of the pizza crust (p > 0.05). There were no differences seen between flour blends for height (p > 0.05). All pizza crust replicates rose roughly two times in height (from ½-in-thick dough to 1-in-thick crusts). Unfortunately, the taste of the pizza crust baked with King Arthur flour was ranked not acceptable, with a score of 2, by consumers in the sensory testing. And for all flour blends, consumers suggested an increase in the water content to improve crust texture and chew. One other suggestion recommended using a fairly wet dough and patting it into a pan vs. rolling it out like a traditional wheat pizza crust. More research is needed to determine a preferred way to bake GF pizza crust that meets consumer expectations.
Drop Cookies
One characteristic of particular interest regarding dropped cookies refers to the hardness of the cookies (Table 4), or the amount of force it takes to break or fracture the cookie. Low hardness values do not necessarily indicate desirable cookie texture, depending on the type of cookie being tested. Drop cookies, such as chocolate chip cookies, tend to be harder if they are crispy and less hard if they are chewy and soft. The hardness of the drop cookies tested was variable depending on the flour blend used; no significant differences were noted in the scientific testing. However, the flour blend that was rated the highest during the sensory testing done by researchers was the King Arthur blend, which consumers received to make these cookies. In the consumer testing, the cookies received a 5 on the hedonic scale (very appealing) from every consumer. The addition of tapioca starch and xanthan gum to this flour blend likely created a higher moisture content, which was more preferred by consumers (Xu et al., 2020). The most noted characteristic of acceptability for the cookies made with the King Arthur flour was the crispy outside and chewy center, which is directly influenced by moisture content in the cookie. Scientific testing results (Table 4) also suggest that, regardless of the flour blend used, storage in air-tight packaging for up to 2 days did not significantly change the texture of the cookies (p > 0.05). Based on these results, it is recommended that consumers store cookies in plastic zip-top bags or other airtight containers for up to 2 days or freeze them to preserve freshness and texture.
Rolled Cookies
One major conclusion noted from the testing results of the rolled cookies was that consumers who baked the cookies for less time (2 min less than indicated in original recipe) found the cookie more appealing. Consumers altered the time based on the degree of browning the cookies were acquiring, as sugar cookies are expected to be very light in color. These results could suggest that consumers prefer a more tender rolled cookie, which would be accomplished with less cooking time. And, depending on the variability of ovens used to bake the cookie, a time range with instructions on the state of browning would be useful in the recipe. Also, rolled cookies with a taller height had a lower moisture content. Because steam helps baked goods rise, it follows that, in cookies where more water was converted to steam (adding to the rise and eventually escaping), a taller cookie was the result.
Another notable conclusion is related to the texture, or hardness, of rolled cookies. Typically, these types of cookies need to be stable enough to handle frosting and still have a tender texture (less hardness, N = 2.22 ± 1.15) when eaten. Because hardness varied (Table 6) significantly in the GF rolled cookies depending on the flour blend used, the flour that created a taller cookie would produce a more preferred cookie, as seen with the White Gold flour. Again, the addition of starch and xanthan gum in all of the flour blends created the appropriate structure for the rolled cookies, while the White Gold flour also contained pea flour protein, which has been shown to help with tenderness in a cookie (Xu et al., 2020). Like dropped cookies, storage in airtight packaging did not significantly change the texture of the cookies (p > 0.05) across storage times, regardless of the flour blend used. The recommendations for storage of this type of cookie are the same as for rolled cookies.
Study Limitations
One major limitation of this study was our inability to test all the gluten-free flour blends available commercially. This limits the general recommendations that can be made from this study. Also, all the consumers who participated in the study lived in a similar geographic area, with high elevation and low humidity. Without conducting a study in areas with varied climates and other conditions, it is difficult to know if these recipes and flour blends would perform in a similar way in other geographic areas. Lastly, the recipe for the pizza crust was highly variable among participants of the consumer study. The in-lab testing showed very different results from the consumer testing. More study is needed on the pizza crust recipe instructions to produce more consistent results.
Extension Application
One focus of Extension work is to educate consumers in areas that will improve their life situations. This research was conducted to improve consumer understanding of the difficulties and costs associated with gluten-free eating. A consumer cost/benefit analysis was conducted in connection to this research to determine whether baking gluten-free products at home would be more economical than purchasing commercial products. Table 8 details the results of this analysis. Products available in Utah stores were used to inform the analysis. Results suggest that, on average, the consumer saves between $0.41 to $3.77 when making and consuming GF baked goods in the home as opposed to buying them pre-made from stores.
Extension faculty across Utah were provided with study results, recipes, consumer analysis, and recommendations for decreasing the difficulties and costs related to gluten-free eating. Classes and presentations were also conducted in several counties that included baking demonstrations, to further educate participants on preferred baking techniques learned through this research. Lastly, this information was shared through social media posts, news spots, and website articles. Participants in the above-mentioned classes reported high satisfaction with the information provided: 76% said they were very satisfied with the classes, and 82% of participants indicated that they learned many new concepts from the information provided. Lastly, 82% of participants said they were likely to use the information from the classes to bake gluten-free products at home.
Table 8. Cost//Benefit Analysis of Home Baked vs Commercially Purchased
| Baked Good Item | Recommended Flour Blend | Amount of Flour Per Bag | Price Per Ounce | Flour Cost for Home-Prepared Baked Goods | Commercially Prepared Cost for Baked Goods |
|---|---|---|---|---|---|
| Rolled Cookies | Extra White Gold all-purpose flour blend | 1.1 lb | $0.33 | $5.28 per recipe made | $4.59-$5.69 per package |
| Drop Cookies | King Arthur measure-for-measure flour blend | 3 lb | $0.19 | $2.32 per recipe made | $5.89-$6.09 per package |
| Pizza Crust | King Arthur measure-for-measure flour blend | 3 lb | $0.19 | $2.32 per pizza crust | $3.49-$4.19 per pizza crust |
| Loaf Bread | Cup4Cup multipurpose flour blend | 2.9 lb | $0.42 | $5.26-$7.46 per loaf | $6.29-$8.69 per loaf |
References
Allen, K. E., Carpenter, C. E., & Walsh, M. K. (2007). Influence of protein level and starch type on an extrusion-expanded whey product. International Journal of Food Science & Technology, 42(8), 953–960. https://doi.org/10.1111/j.1365-2621.2006.01316.x
Alsaiqali, A., Dizlek, H., & Ozer, M. S. (2023). Effects of separated and combined amaranth, quinoa, and chia flours on the characteristics of gluten-free bread with different concentrations of hydrocolloids. Chemical Papers, 77, 5275–5291. https://doi.org/10.1007/s11696-023-02861-w
Anton, A. A., & Artfield, S. D. (2008). Hydrocolloids in gluten-free breads: A review. International Journal of Food Sciences and Nutrition, 59(1), 11–23. https://doi.org/10.1080/09637480701625630
Armero, E., & Collar, C. (1997). Texture properties of formulated wheat doughs: Relationships with dough and bread technological quality. European Food Research and Technology, 204, 136–145. https://doi.org/10.1007/S002170050050
Arslain, K., Gustafson, C. R., Baishya, P., & Rose, D. J. (2021). Determinants of gluten-free diet adoption among individuals without celiac disease or non-celiac gluten sensitivity. Appetite, 156, Article 104958. https://doi.org/10.1016/j.appet.2020.104958
Bael, K. (2015). Better baking gluten-free. Better Baking Gluten-Free | 2015-09-11 | Prepared Foods
Bascuñán, K. A., Vespa, M. C., & Araya, M. (2017). Celiac disease: Understanding the gluten-free diet. European Journal of Nutrition, 56, 449–459. https://doi.org/10.1007/s00394016-1238-5
Bender, D., & Schönlechner, R. (2020). Innovative approaches towards improved gluten-free bread properties. Journal of Cereal Science, 91, Article 102904. https://doi.org/10.1016/j.jcs.2019.102904
Capacci, S., Leucci, A. C., & Mazzocchi, M. (2018). There is no such thing as a (gluten-)free lunch: Higher food prices and the cost for coeliac consumers. Economics & Human Biology, 30, 84–91. https://doi.org/10.1016/j.ehb.2018.06.001
Gandini, A., Gededzha, M. P., De Maayer, T., Barrow, P., & Mayne, E. (2021). Diagnosing coeliac disease: A literature review. Human Immunology, 82(12), 930–936. https://doi.org/10.1016/j.humimm.2021.07.015
Gardner, S. (1992). Consumers and food safety: A food industry perspective. FAO. https://www.fao.org/3/v2890t/v2890t05.htm
Gómez, M., & Martínez, M. M. (2016). Changing flour functionality through physical treatments for the production of gluten-free baking goods. Journal of Cereal Science, 67, 68–74. https://doi.org/10.1016/j.jcs.2015.07.009
Gorgitano, M. T., & Sodano, V. (2019). Gluten-free products: From dietary necessity to premium price extraction tool. Nutrients, 11(9), 1–12. https://doi.org/10.3390/nu11091997
Gustafson, K. L. (2016). Impact of ingredients on quality and sensory characteristics of gluten-free baked goods. Kansas State University.
Horstmann, S., Lynch, K., & Arendt, E. (2017). Starch characteristics linked to gluten-free products. Foods, 6(4), Article 29. https://doi.org/10.3390/foods6040029
Khatkar, B. S., Bell, A. E., & Schofield, J. D. (1995). The dynamic rheological properties of glutens and gluten sub-fractions from wheats of good and poor bread making quality. Journal of Cereal Science, 22(1), 29–44. https://doi.org/10.1016/S0733-5210(05)80005-0
Levinson-Castiel, R., Eliakim, R., Shinar, E., Perets, T.-T., Layfer, O., Levhar, N., Schvimer, M., Marderfeld, L., Ben-Horin, S., & Shamir, R. (2019). Rising prevalence of celiac disease is not universal and repeated testing is needed for population screening. United European Gastroenterology Journal, 7(3), 412–418. https://doi.org/10.1177/2050640618818227
McLellan, M. R., Lind, L. R., & Kime, R. W. (1995). Hue angle determinations and statistical analysis for multiquadrant hunter L,a,b data. Journal of Food Quality (USA), 18, 235240. https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1745-4557.1995.tb00377.x
Naqash, F., Gani, A., Gani, A., & Masoodi, F. A. (2017). Gluten-free baking: Combating the challenges - A review. Trends in Food Science & Technology, 66, 98–107. https://doi.org/10.1016/j.tifs.2017.06.004
Salehi, F. (2019). Improvement of gluten‐free bread and cake properties using natural hydrocolloids: A review. Food Science & Nutrition, 7(11), 3391–3402. https://doi.org/10.1002/fsn3.1245
Stup, R. (2023, February 24). Standard operating procedures: A writing guide. Penn State Extension. https://extension.psu.edu/standard-operating-procedures-a-writing-guide
Susman, I.-E., Culețu, A., Apostol, L., & Popa, M. E. (2020). Trends and challenges in glutenfree baking products ingredients: A review. Scientific Bulletin Series F. Biotechnologies, 24(1), 76–83. https://www.researchgate.net/profile/Mona-Popa/publication/345417978_TRENDS_AND_CHALLENGES_IN_GLUTENFREE_BAKING_PRODUCTS_INGREDIENTS_A_REVIEW/links/5fa657cea6fdcc06241cd5e7/TRENDS-AND-CHALLENGES-IN-GLUTEN-FREE-BAKING-PRODUCTSINGREDIENTS-A-REVIEW.pdf
Vilmane, L., & Straumite, E. (2014). The use of soy flour in yellow maize-amaranth gluten-free bread production. Proceedings of the Latvia University of Agriculture, 31(1), 1–11. https://doi.org/10.2478/plua-2014-0001
Xu, J., Zhang, Y., Wang, W., & Li, Y. (2020). Advanced properties of gluten-free cookies, cakes, and crackers: A review. Trends in Food Science & Technology, 103, 200–213. https://doi.org/10.1016/j.tifs.2020.07.017
Zhong, H., Allen, K., & Martini, S. (2014). Effect of lipid physical characteristics on the quality of baked products. Food Research International, 55, 239–246. https://doi.org/10.1016/j.foodres.2013.11.010
Appendix 1 - Recipes
Bread
Ingredients
- 3 cups gluten free flour (we use Cup4Cup GF flour blend)
- 1 tablespoon active dry yeast
- 3 tablespoon sugar
- 1 ¼ cup warm water (about 95°F)
- ⅓ cup oil
- 3 eggs
- 1 ½ teaspoon salt
Steps
Combine flour and salt in an electric mixer bowl, set aside. In a separate bowl, dissolve sugar and yeast in warm water and let stand until the yeast becomes foamy. Add yeast mixture, eggs, and oil to dry ingredients.
Mix ingredients on medium speed using the paddle attachment until fully incorporated (about 1 minute). Turn mixer to high and mix for 1-2 minutes until dough is smooth and stretchy. Scrape bowl as necessary to thoroughly blend all ingredients.
Grease a 9x5” loaf pan. Spoon dough into the pan and smooth the top with the back of a wet spoon. Using a butter knife cut a shallow line down the middle of the dough.
- Note. If you are using an 8.5x4.5” loaf pan, only add about ¾ of the dough.
Cover dough with greased plastic wrap and place dough in a warm room to rise for 20 minutes.
- Note. Do not let dough rise longer than 20-25 minutes, it will form large holes in the dough if it raises too long.
Bake in preheated 375F oven until golden brown. Check the bread after 30 minutes for doneness (you can use a toothpick, skewer, or thermometer to at least 165 degrees). If the top looks too brown, but the interior is not done, turn the oven down to 350F. Let the bread continue cooking, checking for doneness as appropriate.
- Note. You might want to put a cookie sheet or cake pan on the bottom rack of your oven in case the bread overflows the pan.
Pizza Crust
Ingredients
- 1 ½ cups gluten free flour blend (we use King Arthur Cup for Cup GF Blend)
- 1 ½ teaspoon instant yeast (or 2 tsp active dry)
- ¾ teaspoon sugar
- ¾ teaspoon salt
- ½ cup plus
- 1 tablespoon water
- 3 tablespoon olive oil
Steps
In the bowl of your stand mixer fitted with the paddle attachment place the flour, yeast, and sugar, and mix slowly to combine. Add the salt and mix slowly again to combine. Add the water and olive oil and mix on medium speed until the dough begins to come together. Turn the mixer on high-speed mix until dough begins to appear whipped.
Transfer the dough to an oiled container with a tight-fitting lid or a greased bowl, spray lightly with cooking oil spray, and cover tightly. Place in a warm, draft-free area to rise until it’s about 150% of its original volume (about an hour).
When you’re ready to make the pizza, place a pizza stone or overturned rimmed baking sheet in the oven and preheat it to 400°F. Place the tightly sealed dough in the refrigerator to chill for at least 15 minutes before working with it, as it’s easiest to work with when it’s chilled.
To make pizza, place the dough on a lightly floured surface and sprinkle the top lightly with a bit more flour. Knead the dough a bit until it’s smoother and roll out on the floured surface with a rolling pin, moving the dough frequently to prevent sticking. Sprinkle very lightly with additional flour as necessary. Create a smooth edge around the perimeter of the dough by pressing the edges with one hand toward the palm of your other.
Transfer the dough to a large piece of unbleached parchment paper and brush the top of the dough generously with olive oil. Using a pizza peel or other flat surface like a cutting board, transfer the dough to the pizza peel or baking sheet in the preheated oven and bake it plain for 5 to 7 minutes, or until the crust has begun to crisp on the underside.
Remove the crust from the oven. At this point, the parbaked crust can be cooled completely, wrapped tightly, and frozen for at least one month.
To use the crust, simply defrost at room temperature, and then continue with the recipe as written. To continue preparing the dough, add your favorite toppings to the parbaked crust, and return the pizza to the hot oven until any cheese is melted and the edges have browned and puffed (another 5 to 7 minutes). Allow to sit for 5 min.
Drop Cookies
Ingredients
- ¼ cup butter
- ¼ cup shortening
- ½ cup granulated sugar
- ½ cup brown sugar
- 1½ cups flour (we use King Arthur Cup for Cup GF blend)
- 1 cup chocolate chips
- 1 egg
- ½ tsp baking soda
- ½ tsp vanilla pinch of salt
Steps
Combine room temperature butter, shortening, and sugars (white and brown) in the bowl of an electric mixer. Beat on high until mixture is smooth. Add egg and beat on high until mixture lightens and becomes fluffy. Scrape bowl and add soda, vanilla, and salt. Mix well.
Add flour to mixture and mix on low speed to combine. Mix on high for 1-2 minutes until dough starts to form together, if it is still really sticky add more flour in ¼ cup measurements. Dough should form a ball and not stick to the sides of the mixing bowl. Stir in chocolate chips.
Scoop cookie dough (use #50 scoop, which is about 1 ½ Tbsp of dough) into hand and gently form into a ball. Place on parchment lined baking sheet.
Bake at 350F for 10-12 minutes until barely golden brown. Smack pan against countertop to flatten cookies. Let sit 1-2 minutes on pan then transfer to a cooling rack to cool completely.
Rolled Cookies
Ingredients
- ½ cup butter
- ½ cup granulated sugar
- ¾ tsp baking powder
- 1 egg
- ½ teaspoon vanilla
- 3 tablespoon powdered sugar
- 2 cups GF flour blend (we use White Gold GF baking blend)
Steps
Add room temperature butter and sugar to electric mixer bowl, mix well using paddle attachment. Add egg and then beat until butter mixture starts to lighten and become fluffy. Scrape bowl and add baking powder, salt, and vanilla. Add flour and powdered sugar all at once. Pulse mixer to incorporate flour/sugar slowly into other ingredients. Once flour is mostly mixed in, scrape bowl and turn mixer to high and beat for 1-2 minutes until it becomes smooth and elastic.
Test dough, it should not be sticky when touched, it should also be thick and hold its shape when pinched. Add more flour in ¼ cup increments until it is a thick, rollable dough.
Place dough onto a sheet of parchment paper and gently form into a ball. Place another piece of parchment on top of dough and roll out to ¼” thickness. Cut cookies into circles or shapes. Carefully transfer to parchment lined baking sheet. Combine scraps and re-roll, cut as before.
Bake at 325F for 8-9 minutes until set, but not browned. Let sit 1-2 min on pan then transfer to a cooling rack to cool completely.
September 2024
Journal of Human Sciences and Extension
Authors
April Litchford, Karin Allen, Cindy Jenkins, Eva Timothy, and Paige Wray
Related Research