Introduction
Obesity and overweight are major global health concerns, with substantial evidence linking them to increased risks of cardiovascular diseases, type 2 diabetes, and various forms of cancer. Despite the availability of numerous weight management strategies, many individuals struggle to achieve and maintain long-term weight loss. Recent advancements in digital health technologies, including mobile health (mHealth) applications, have emerged as promising tools for supporting weight management interventions by enabling self-monitoring and promoting behavioral changes.
Self-monitoring of dietary intake is recognized as a critical component of successful weight loss interventions. Traditional methods, such as food diaries, can be time-consuming and prone to inaccuracies. In contrast, food scanning mobile phone applications offer a more efficient and user-friendly alternative, allowing users to track their food intake by scanning barcodes or capturing images of meals. These features not only facilitate real-time dietary monitoring but also provide immediate feedback, which may enhance users’ adherence to dietary recommendations.
The effectiveness of food scanning mobile applications for weight loss has been evaluated in various interventional studies. For instance, Vaz et al. (2021) Qasrawi et al. (2025) demonstrated significant weight loss among participants using an interactive smartphone-based lifestyle intervention app over a six-month period. Similarly, Ferrante et al. (2020) highlighted improvements in weight, quality of life, and physical fitness outcomes among African American breast cancer survivors utilizing eHealth tools. However, other studies, such as those by Nezami et al. (2022), reported no significant differences in weight outcomes between intervention and control groups, emphasizing the need for further investigation into the factors influencing the effectiveness of these tools.
Given the growing popularity and potential benefits of food scanning apps, this systematic review aims to assess their impact on weight loss outcomes. By synthesizing findings from longitudinal interventional studies, this review seeks to provide evidence-based insights into the utility of these digital tools in weight management strategies. The findings of this review may inform healthcare providers and policymakers about the effectiveness of food scanning mobile phone applications and guide future interventions for combating obesity and overweight.
Methodology
Review question
The review aimed to assess the effects of using food scanning mobile phone apps on weight loss.
Search strategy
A comprehensive search was conducted to identify relevant studies published between 1, 2007, and November 1, 2024. The following electronic databases were searched: PubMed (MEDLINE), Embase, CINAHL, Cochrane, AMED, and SPORTDiscus. Search terms were tailored to meet the specific requirements of each database. Additionally, reference lists of included studies were hand-searched to identify other potentially relevant articles. The search was restricted to studies written in English and published in peer-reviewed journals. The systematic review protocol was registered through PROSPERO registry (registration number: CRD42024614971).
Keywords
To allow reproducibility of the search, the Medical Subject Headings (MeSH) was used. The keywords used were: S1 = “Food scanning” OR “calories in food” OR “food tracking” OR “food tracking” OR “food tracker” OR “food screening”; AND S2 = “mobile applications” OR “apps” OR “mobile apps” OR “mhealth” OR “ehealth” OR “smartphone applications” or “smartphone apps”; AND S3 = “weight loss” OR “weight reduction” OR “lose weight” OR “obesity” OR “overweight” OR “weight management”.
Inclusion and exclusion criteria
Articles were included if they assessed the effects of using food scanning mobile phone apps on weight loss and were longitudinal interventional studies. Articles were excluded if they: 1) did not include food scanning apps; 2) were not interventional trials; 3) were case reports; 4) were written in a language other than English; or were protocol papers or conference abstracts. Population, Intervention, Comparator, Outcomes, and Study design (PICOs) criteria were used throughout the inclusion and exclusion criteria as below:
Population: Adults aged 18 and above, regardless of baseline metabolic status, were included in the eligible studies.
Intervention: The use of mobile phone-based food scanning applications designed to facilitate dietary self-monitoring.
Comparator: Control conditions included standard care, waitlist, or non-intervention groups depending on the study design.
Outcomes: The primary outcome was weight loss; secondary outcomes included waist circumference, adherence, quality of life, and other metabolic markers where available.
Study Design: Only longitudinal interventional studies (randomized or non-randomized) were included.
Data screening and extraction
The first reviewer (AAb) retrieved studies from database searches and imported them into EndNote software. Studies were sorted alphabetically by author surname, and duplicates were removed. Titles and abstracts were screened against predefined eligibility criteria by the first reviewer, with a second reviewer (MZD) independently conducting the same process. Disagreements were resolved through discussion between the reviewers.
For studies meeting the eligibility criteria, full-text reviews were performed. Data extraction included the following information: author and year of publication; sample size, study design and goals, mobile app used and its main specifications; outcome measures used; and Key findings. Extracted data were verified by the second reviewer (MZD).
Risk of bias assessment
The risk of bias in the included studies was assessed using the Cochrane Risk of Bias Tool 2 (CROB 2) by two independent reviewers (RB and AS). This assessment evaluated: 1) bias arising from randomization processes; 2) bias due to deviations from intended interventions; 3) bias from missing outcome data; 4) bias in the measurement of outcomes; and 5) bias in the selection of reported results. Each domain of the CROB 2 tool was judged as ‘low risk’, ‘some concerns’, or ‘high risk’. A detailed domain-level assessment for each included study is presented in Supplementary Table 1.
Table 1
Summary of included trials that investigated the effects of using food scanning mobile phone apps on wight loss.
| AUTHOR (YEAR) | LOCATION OF THE TRIAL | SAMPLE SIZE | AGE YEARS (MEAN ± SD) | WEIGHT KG (MEAN ± SD) | MOBILE PHONE APPS | APP PRESCRIPTION | PRIMARY OUTCOME | SECONDARY OUTCOME | KEY FINDINGS |
|---|---|---|---|---|---|---|---|---|---|
| Bender., et al (2017) | USA | Overall: 45M:F = 28:62 IG: 22M:F = 12:8 CG: 23M:F = 14:9 | Overall: 57.6 ± 9.8 IG: 57.4 ± 9.8 CG: 57.7 ± 10.0 | Overall: 75.8 ± 15.4 IG:72.6 ± 10.8 CG: 78.8 ± 18.6 | PilAm Go4Health | Intervention Duration: 6 months. IG: Phase 1 (from baseline to 3 months: participants were trained on using the Fitbit to self-monitor, PA steps and app with diary to self-report daily food/calorie intake. They joined private Facebook group for virtual social support, and coaching. Research office visits at 1, 2 and 3 months. Phase 2 ( from 3 months to 6 months): IG participants were removed from the Facebook group. They were asked to continue using their Fitbit and app with diary to track health behaviours.Research office visits at 4 and 6 months. CG: Phase 1 (from baseline to 3 months): participants received only the Fitbit accelerometer and training for daily wear. Research office visits at 1 and 3 months. Phase 2 (From 3 month to 6 month): participants transitioned to receive the PilAm Go4Health intervention. Research office visits at 4, 5 and 6 months. | Attendance and adherence | Percent weight change | - Both the IG and waitlist CG achieved near-perfect attendance at all 7 intervention office visits 95% for IG (21/22) and 100% for CG (23/23). - Both IG and CG participants (100%) completed the study at 6 months follow up. - During phase 1, about 18% (4/22) of the IG achieved a 5% weight loss, whereas 82% (18/22) of the remaining participants maintained or lost 2% – 5% of their weight. - During phase 2, 90% (20/22) of the IG continued to maintain or lose 2% – 5% more weight. - During phase 1, over 83% (19/23) of the CG maintained or gained 2% – 5% more weight. - During phase 2, 70% (16/23) of the CG receiving PilAm Go4Health having maintained or lost between 2% – 5% of their weight. |
| Ben Neriah., et al (2019) | USA | Nonuser: 113,916 M:F = 83,150: 72.99 Users: 9871 M:F = 7153: 72.46 | Nonuser: 36.1 ± 12.1 Users: 36.5 ± 11.9 | Nonuser: 93.6 ± 20.5 Users: 94.0 ± 20.6 | Lose it | Intervention Duration: 12 months The participants had to sign up in the app by entire information about height, gender, weight and age and they record their daily food intake by taking photo of the food. After that the users specifies the quantity consumed from the list and calculate the calories and macronutrients. | Weight loss | Number of logged days and duration | - Both users and nonusers had decrease their body wight with greater weight loss in users’ group (P < 0.001). - The users group showed that the number of days the photo feature was used was significantly associated with the %weight loss (P < 0.001). - The photo feature group used the app for 3.5 days more than nonusers (P < 0.001). - Photo users group logged 6.1 more days than nonusers (P < 0.001). |
| DeLuca., et al (2020) | USA | HW: 12,378 M:F = 10,927: 1451 DPP: 2389 M:F = 2100: 289 | HW: 42.0 ± 38.0–47.0 DPP: 51.0 ± 44.0–58.0 | HW: 94.1 ± 20.4 DPP:94.4 ± 20.5 | Noom | Intervention Duration: 16 and 52 weeks. The participants had to sign up in the app. The users had tracking feature for food, exercise logging weighing, access to behaviour coach, and daily curriculum that includes diet-, exercise-, and psychology-based content. HW: Participants enrolled based on self-interest in weight loss. DPP: Participants encouraged to join following a prediabetes diagnosis from their health care provider. Both groups have access to the same features weight, healthy eating, and physical activity. DPP program includes specific diabetes prevention which is not included with HW group. | Wight loss | - HW group lost on average 4.74 kg at week 16 and 5.24 kg at week 52. - DPP group lost on average 5.61 kg at week 16 and 5.66 kg at week 52. - DPP participants showed greater weight loss compared with HW, losing 3.20 kg more at week 16 and 2.38 kg more at week 52. | |
| Ferrante., et al (2020) | USA | Overall: 35 IG: 18 CG: 17 | Overall: <60 = 14 ± 40.0 ≥60 = 21 ± 60.0 IG: <60 =7 ± 38.89 ≥60 = 11 ± 61.11 CG: <60 = 7 ± 41.18 ≥60 = 10 ± 58.82 | IG:91.98 ± 15.35 CG:104.06 ± 22.65 | Spark People website | Intervention Duration: 3, 6, and 12 months. All participants received a handout of goals for weight loss, calories intake and physical activity. wrist-worn physical activity tracker, Fitbit Charge were received for all. F: Each day I: Mild to moderate exercise T: 30 min per day T: 10,000 steps per day IG: Participants received one 30-minute session on the SparkPeople website. They instructed to self-monitor their diet at least weekly. CG: At 6 months, the participant receives SparkPeople treatment. | Adherence and acceptability | - Weight loss and BMI - QoL: in adult cancer survivors, scale range 1–7 (never-always; lower score is better quality). - 6MWT | - Participants logged into SparkPeople website more than once weekly throughout the study. - After 6 months of follow up, both IG and CG had significant decrease in weight loss (P = 0.006), (P = 0.002), respectively. - After 6 months of follow up, both IG and CG had significant decrease in BMI (P = 0.006), (P = 0.012), respectively. - After 6 months of follow up, IG (P = 0.031) had significant improvement in QoL compared to CG (P = 0.440). - After 6 months follow up, both IG and CG had significant improvements in the 6MWT (P = 0.006), (P < 0.001), respectively. - Participants found SparkPeople easy to use and somewhat to very useful, and many provided positive comments regarding the educational and inspirational articles, recipes, videos, incentives, and social groups. |
| Hu., et al (2021) | USA | Overall: 1,740 M:F = 1462:278 | Overall: 48 ± 11 | Overall:225 ± 40 | FoodSmart | Intervention Duration: 12, 24, and 36 months. The participants had to sign up in the app and fill a dietary questionnaire called “Nutriquiz” (53-item food frequency questionnaire adapted from the National Cancer Institute Diet History Questionnaire) and their information sex, age, weight, and usual frequency of dietary intake. Participants had to assign a score from 0 to 10 (10 being optimal) for each 7 component fruits, vegetables, protein ratio, carbohydrate ratio, fat ratio, sodium, and hydration. | Sustained weight loss: losing 5% of initial weight between first and second reported weights and additional weight loss or no change between the second and third reported weights. | - Among all participants, 22.4% sustained weight loss. - Among participant who were enrolled for greater than 12, 24, and 36 months, the percent of participants who sustained weight loss was 21.7%, 22.8%, and 23.8% respectively. - Participants with obesity class 2 and 3 had a significant greater change in diet quality and sustained weight loss (P < 0.001). | |
| Vaz., et al (2021) | USA | Overall: 28 M:F = 24:4 IG: 13 M:F = 11;2 CG: 15M:F = 13:2 | Overall: 43.25 ± 2.48 IG:40.15 ± 3.72 CG: 45.93 ± 3.29 | Overall: 93.12 ± 2.78 IG:94.13 ± 3.40 CG: 92.25 ± 4.37 | The Smart Food Diary™ | Intervention Duration: 6 months. IG: Participants received two smart devices: wrist worn to monitor physical activity and smart scale. The participants had to logged onto three apps: the Fitbit™ app that links to the activity tracker, smartscale, and photo-sharing apps. Smart food diary was used by all the participants with professional coach to observe food intake and allowed the coach to provide positive reinforcement with “stars” on healthy meals. F: Daily tracking T: 40–50 min CG: The participants were wait-listed to receive the smart lifestyle intervention after 6 months. Both IG and CG received conventional outpatient weight-management visits at baseline and at 3 and 6 months including: diet plan, goals for physical activity and weight, and minimal amount of behavioural feedback. | Change of wight loss | - Waist circumference - HgbA1c - Systolic and diastolic blood pressure. - Plasma concentrations of triglycerides | - After 6 months of follow up, IG had significant decrease in weight loss compared to CG (P < 0.01), (P < 0.05), respectively. - After 6 months of follow up, IG had signification Waist circumference and HgbA1c compared to CG (P < 0.01), (P < 0.05), respectively. - After 6 months of follow up, no significant difference in the changes in systolic or diastolic blood pressure, or plasma triglyceride concentrations between both groups. (P = 0.51), (P = 0.3476), (P = 0.6450). respectively. |
| Li., et al (2022) | USA | Overall: 15 M:F = 6:8 | 61.87 ± 10.67 | 86.1 ± 25.9 | Led, mobile health (mHealth) | Intervention Duration: 12 weeks. All the participants received mHealth and one-on-one training from health workers. F: Daily sessions T: at least 10 min T: Diabetes Prevention Program Group Lifestyle Balance Program, self-monitoring goals for frequency of weight self-monitoring, food logging, physical activity tracking, and blood glucose self-monitoring. | Retention of adherence | - CSQ-8 - Weight loss | - 93% (14/15) of participants accessing all digital education modules and 53% (8/15) completed all courses. - All the participants were satisfied by the intervention. The mean score for CSQ-8 was (29.53 ± 3.04). - Participants had significant decrease in body weight of 3.5 kg from the bassline (P = .001). - Participants had a significant improvement in PTES score (P < 0.001) and eHealth literacy (P < 0.001). |
| Nezami., et al (2022) | USA | Overall: 72 Standard group:37 M:F = 35:2 Simplified group: 35 M:F = 33:2 | Standard group: 39.8 ± 4.7 Simplified group: 40.2 ± 4.7 | Standard group: 99.1 ± 21.6 Simplified group: 91.0 ± 15.9 | The PATH | Intervention Duration: 6 months. Standard group: Participants received a daily calorie goal based on their starting weight, and they were tracking their daily caloric intake from Fitbit app. Simplified group: Participants received a simplified version of dietary self-monitoring based on the Traffic Light Die, which classified the food into 3 categories green, yellow, or red (high calorie foods). Participants were asked to track only red foods and were given a daily red food limit based on their starting weight. They tracked their foods using PATH app. Both groups received digital automated behavioural weight loss intervention, activity tracker (Fitbit Inspire), and smart scale. T and T: attended 75-minute group kick-off sessions and move at least 10 minutes per day and progress to 60 minutes. Lessons weekly in 1 to 3 months and biweekly in 4 to 6 months regarding behavioural strategies for cognitive and skills to meet their dietary and activity goals. | Retention of adherence | Weight change | - 96% of participants accessing all digital education modules. All participants who did not complete the 6-month weight measurement were in the Standard group 3 participants, compared with no participants in Simplified group. - After 6 months of follow up, there is no significant differences between both group (P = 0.24). Standard group participants lost 5.9 kg at 6 months compared to Simplified group participants lost 3.5 kg. |
[i] IG: Intervention Group; CG: Control Group; PilAm Go4Health: Pilipino Americans Go4Health; PA: pedometers; HW: Healthy Weight; DPP: Diabetes Prevention Program; BMI: Body Mass Index; QoL: Quality of Life; 6MWT: 6 Minute Walk Test; CSQ-8: Customer Satisfaction Questionnaire short version; PTES: Perceived Therapeutic Efficacy Scale; HgbA1c: Haemoglobin A1c; The PATH: PArents Tracking for Health.
Any disagreements were resolved through discussion. The results of the bias assessments were considered in the interpretation of findings, particularly when synthesizing evidence from studies rated as having “some concerns” or methodological limitations. This approach helped ensure that conclusions drawn from the review were grounded in the methodological quality of the included evidence.
Data synthesis and analysis
For quantitative data, risk ratios (for categorical outcomes) and standardized mean differences (for continuous outcomes) were calculated with 95% confidence intervals where applicable. Comparable data from eligible studies were intended to be pooled using a random-effects model in OpenMetaAnalyst software. However, due to substantial heterogeneity in study design, intervention types, and outcome measures, meta-analysis was not feasible, and a narrative synthesis was conducted instead. . Although no new statistical power analysis was conducted for this review, we considered the reported sample sizes and effect sizes within each included study when interpreting the strength and robustness of their findings.
Results
Study characteristics
The systematic search identified 42 citations from the databases. Two records were duplicates. Consequently, 40 citations were screened from titles, and 15 were considered not to be relevant after reading the titles and abstracts, and a total of 25 studies were screened by reading the full-text. Out of the 25 studies, 17 studies were excluded; 11 studies were not longitudinal trials, and 6 studies did not use mobile apps or food scanning intervention. Consequently, eight studies were included in the review. Figure 1 represents the PRISMA flowchart for the search records. A meta-analysis was not feasible because of the heterogeneity of the included trials.
Figure 1
PRISMA Flow-Chart of the search records.
Quality assessment
The quality of the trials was independently assessed by RB and AS using and the CROB2 tool. Overall, five studies were ranked as low risk and three studies were ranked as some concerns, mainly due to difficulties in randomization and blinding. Figure 2 represents a summary of the CROB2 results. Table 1 is the data extraction table for the eight included studies.
Figure 2
CROB2 results of the included studies.
It is important to note that the study by Nezami et al. (2022) reported no statistically significant differences between intervention and control groups, underscoring the need for cautious interpretation of the overall positive trends.
Screening apps and websites used
Various screening food apps and application websites were used in the eight studies for weight loss including PilAm Go4Health, Lose it, Noom, SparkPeople website, FoodSmart. The Smart Food Diary™, mHealth, and The PATH. PilAm Go4Health is an mHealth culturally adapted weight loss lifestyle intervention promoting physical activity and healthy eating for Filipino Americans with obesity and type 2 diabetes to reduce incidence of cardiovascular risks.
Lose it, is a free weight loss mobile app which allows users to record their daily food intake. Users enter their self-reported weight, height, gender, and age when they sign up. They choose their goal weight and weight-loss rate. The app allows users to record food intake and exercise, calculates calories consumed, and estimates calories expended by screening food items and the user request to confirm their meal items from a list of potential matching foods and to add estimated portion sizes.
Noom, is an mHealth behaviour-change lifestyle intervention that provides users with tracking features for food and exercise logging and weighing-in as well as access to a virtual behaviour change coach, support group, and daily curriculum that includes diet, exercise, and psychology-based content.
SparkPeople website, is eHealth/mHealth weight loss program, which is has the following features: educational and motivational articles, fitness, wellness, weight tracking tools, integration with many popular physical activity trackers, daily meal plans, social support communities, and exercise videos from certified personal trainers and fitness instructors.
FoodSmart, is a digital nutrition platform that encourages behaviour change through personalization of nutrition and meal/recipe recommendations. It has two main services: FoodSmart and FoodsMart, which both use behaviour change theory to facilitate access and engagement with affordable, tasty, and healthy food.
The Smart Food Diary™, is an app which is assisted with behaviour modification by enabling the individual to observe food intake in a visually appealing, and organized manner. It also allows the nutritionist to provide positive reinforcement marked with “stars” on healthy meals after the users share their photographs of their meals.
mHealth-based DSMES (Diabetes Self-Management Education and Support) program, is an app which is associated behaviour change to improve health outcomes, self-monitoring to behavioural goals, physical activity and diet monitoring and community health workers provides personalized DSMES service.
The PATH (PArents Tracking for Health), is an app which includes lessons page for behavioural strategies with specific instructions for cognitive and behavioural skills, dietary and activity goal, and social support. Lessons included topics such as setting limits, snacking, physical activity, sedentary behaviour, problem solving, stress management, and relapse prevention.
Effect of using screening app and website on weight loss
The eight trials investigated the effect of using screening apps and website on weight loss. Bender., et al (2017) reported that after six months of follow up, both the intervention group (IG) and the control group (CG) lost weight ranging from 2% to 5% of their weight.
Ben Neriah., et al (2019) reported that both IG and CG had decrease their body wight with greater weight loss in the IG (P = 0 < 0.001) after one year follow up. Also, DeLuca., et al (2020) stated that both IG and CG participants had a decrease on body weight. The IG lost on average 4.74 kg at week 16 and 5.24 kg at week 52 follow-up. The CG lost an average of 5.61 kg at week 16 and 5.66 kg at week 52.
Ferrante., et al (2020) stated that both IG and CG had a significant weight loss after six months of follow up (P = 0.006, P = 0.002, respectively). Furthermore, Hu., et al (2021) reported that all participants had sustained weight loss of 23.8% after 36 months of follow up.
Vaz., et al (2021) reported that IG had a significant weight loss compared to CG (P < 0.01, P < 0.05, respectively) after six months of follow up. Moreover, Li., et al (2022) reported that participants had significant decrease in body weight of 3.5 kg from the bassline (P = 0.001). The last trial, Nezami., et al (2022) reported that there were no significant differences between IG and CG (P = 0.24) after six months of follow up.
One trial investigated the effect of using screening food app on weigh loss. Ben Neriah., et al (2019) reported that the IG showed that the number of days the photo feature was used was significantly associated with the percentage of weight loss (P = 0 < .001) compared with no significant changes in the control group.
Effect of using screening app and website on waist circumference
One trial investigated the effect of using screening apps and websites on waist circumference. Vaz., et al (2021) reported that after six months of follow up, IG had significant decrease in waist circumference compared to the CG (P < 0.05).
Effect of using screening app and website on attendance/adherence
Five trials investigated the effect of using screening apps and website on participants attendance and adherence. Bender., et al (2017) reported that both IG and CG achieved near-perfect attendance at all seven intervention office visits (95% for IG [21/22] and 100% for CG [23/23]). Also, Ben Neriah., et al (2019) stated that the IG attended 6.1 more days than the CG (P = 0 < .001) during the study.
Ferrante., et al (2020) reported that Participants logged into SparkPeople website more than once weekly throughout the study. Moreover, Li., et al (2022) reported that 93% (14/15) of participants accessing all digital education modules and 53% (8/15) completed all courses. Nezami., et al (2022) reported that 96% of participants accessing all digital education modules. All participants who did not complete the six months were three participants in the CG, compared with no participants in IG.
Effect of using screening app and website on quality of life (qol)
One trial investigated the effect of using screening apps and website on QOL. Ferrante., et al (2020) reported that after six months of follow up, the IG (P = 0.031) had significant improvement in QoL compared to CG (P = 0.440).
Effect of using screening app and website on 6mwt (6 minute walk test)
One trial investigated the effect of using screening apps and website on 6MWT. Ferrante., et al (2020) reported that after 6 months follow up, both IG and CG had significant improvements in the 6MWT (P = 0.006, P = 0 < .001, respectively).
Effect of using screening app and website on hgba1c, systolic and diastolic blood pressure and plasma concentrations of triglycerides
Vaz., et al (2021) investigated the effect of using screening apps and websites on waist circumference. They reported that after 6 months of follow up, IG had significant decrease in HgbA1c compared to CG (P < 0.01). Furthermore, there were no significant differences in the changes in systolic or diastolic blood pressure, or plasma triglyceride concentrations between both groups after six months of follow up, (P = 0.51, P = 0.3476, P = 0.6450, respectively).
Discussion
The prevalence of obesity among children, youth, and adults is increasing at an alarming rate in many parts of the word with its associated health issues such as heart disease, high blood pressure, cancer, and in some circumstances, low self- esteem. Obesity management strategies face a number of obstacles due to environmental, biological and behaviour factor. However, the emergence of smart mobile devices and health application has introduced new opportunities for obesity management by providing accessible, user-friendly tools for tracking dietary intake, physical activity and overall health indicators.
The purpose of this paper was to systematically review longitudinal intervention studies to assess the effectiveness of various food applications and websites for weight loss. A total of eight studies that met the inclusion criteria for using mobile applications or food scanning interventions were included in this review.
Overall, the findings suggest that mobile applications are effective in promoting weight loss. All of the studies included in this review reported that the use of food mobile applications had a beneficial impact on weight loss in both IG and CG through behaviour modification, self-monitoring and real-time feedback.
For instance, Ben Neriah et al. (2019) investigated the effect of a food item recognition feature within a mobile application for tracking food intake on weight loss. The results showed a decrease in body weight in both the intervention group (IG) and control group (CG), with greater weight loss observed in the group using the photo based feature. This method might be more time efficient and motivating than typing food items. In addition, it prompts users to acknowledge the nutritional content of the food item before consuming it, which may influence their food choices and portion sizes. Similar findings regarding weight loss have also been reported in other studies utilizing different food scanning applications.
Nezami et al. (2022) evaluated the efficacy of two mobile-delivered weight loss interventions. The results demonstrated clinically meaningful weight loss at six months, with no significant differences in dietary tracking adherence or dietary intake between the two groups. The study suggested that lower-burden, simplified monitoring of high-calorie foods could be a promising alternative to standard calorie tracking.
Beyond weight loss, food scanning apps also showed positive effects on other health outcomes. Several studies reported a significant reduction in waist circumference in the IG compared to the CG (P < 0.05), which is an important indicator of abdominal fat and an independent risk factor for dyslipidemia, hypertension, CVD, and type 2 diabetes. Additionally, the use of various mobile applications has been associated with improvements in HbA1c levels compared to the control group (CG) (P < 0.01), indicating that these mobile applications may be effective tools for managing blood glucose levels in diabetic patients.
The impact of these applications can be attributed to behaviour changes that support better health outcomes. Mobile applications, through their features such as food tracking, physical activity monitoring, and personalized feedback, may facilitate the adoption of these healthy behaviour, thereby contributing to improved health outcomes such as better glucose control, weight management and overall well-being.
Furthermore, Ferrante (2020) reported that after six months of follow up, the IG showed significant improvements in quality of life (QoL) (P = 0.031) compared to CG. These findings suggest that food scanning apps may not only support weight loss but also contribute to overall health improvements and mental well-being.
An important consideration in interpreting these findings is the risk of bias present in several of the included studies. Although five studies were rated as having low risk, three studies were rated as having “some concerns” according to the Cochrane Risk of Bias Tool 2. Common issues included unclear or inadequately described randomization procedures, which raise the possibility of selection bias. Additionally, due to the nature of mHealth interventions, blinding of participants and personnel was generally not feasible, increasing the risk of performance bias. Most studies relied on self-reported weight and dietary intake, which can introduce reporting bias and affect the accuracy of outcome measures. Furthermore, selective reporting of outcomes was a concern in some studies, especially where full protocols were not available. Acknowledging these limitations is essential to appropriately framing the strength of the evidence and underscores the need for more rigorously designed trials in this domain.
This review also emphasized the impact of using food scanning applications and websites on participants’ adherence, which plays an important role in the success of these interventions. The findings from five trials highlight the positive impact of these digital tools on participant attendance and engagement, leading to effective weight management.
Ben Neriah et al. (2019) found that users of the photo feature logged more days and participated in the program for a longer duration compared to non-users. Similarly, Ferrante et al. (2020) observed that participants logged into the SparkPeople website more than once per week throughout the study. Participants reported that the application was user friendly and useful, with many finding the food tracking feature, particularly its ability to scan food labels, both beneficial and easy to use.
Moreover, one study reported that adherence to using mobile technology was excellent for wearing the Fitbit to track PA and logging foods to monitor calorie intake. However, adherence to self-monitoring weight was markedly lower. The study suggested that negative psychological effects associated with self-weighing, such as feelings of anxiety, depression, and stress, could discourage individuals with overweight or obesity from regularly monitoring their weight. Future research is needed to identify the barriers and facilitators to weight tracking, in order to enhance intervention strategies aimed at promoting weight loss.
Key sources of heterogeneity included the diversity in app functionalities (e.g., barcode scanning, photo logging, behavioural coaching), wide variation in intervention durations (ranging from 12 weeks to 36 months), and the inclusion of distinct populations such as diabetic patients, cancer survivors, and generally healthy adults. Despite the overall positive trends, inconsistencies in weight loss outcomes across studies were observed. Several contextual factors may have contributed to these variations. First, participant characteristics differed significantly between trials, with some including metabolically healthy individuals while others involved populations with comorbidities such as type 2 diabetes or cancer. These differences in baseline health status may influence the physiological responsiveness to dietary interventions. Second, app usage frequency and adherence varied considerably, with some studies reporting near-daily engagement, while others lacked detailed usage metrics. Higher frequency of self-monitoring is known to correlate with greater weight loss, which may explain some of the disparities. Additionally, the duration of intervention and follow-up ranged from 12 weeks to over 36 months, further affecting the comparability of outcomes. Finally, cultural adaptations, such as those seen in the PilAm Go4Health app tailored to Filipino Americans, may have enhanced intervention effectiveness in specific populations. These factors underscore the importance of considering both individual and contextual variables when evaluating digital health interventions.
On the other hand, no significant difference in dietary tracking adherence was found between simplified and standard calorie dietary self-monitoring interventions. However, future studies with a 12- month follow-up period would be useful to determine whether the type of dietary self-monitoring affects long-term adherence and weight change.
Additionally, as no multiple comparison corrections were applied due to the heterogeneity of study designs and outcome measures, the findings should be interpreted with caution. This limitation underscores the need for future studies to employ standardized outcomes and designs that allow for pooled analyses and adjusted statistical comparisons. Furthermore, the absence of statistical correction for multiple comparisons across the synthesized outcomes may increase the likelihood of Type I errors. This limitation warrants caution when interpreting the aggregated results and underscores the need for standardized outcome reporting in future research.
A major strength of this review is the inclusion of eight longitudinal intervention studies that utilized different types of mobile applications and websites, allowing the assessment of participants over an extended period of time. Furthermore this review highlights on the effects of these applications on various aspects, including weight reduction, adherence, quality of life and other health outcomes. Future studies should explore long-term adherence patterns, real-world clinical applications, and cost-effectiveness of these tools across diverse demographic and health contexts to fully realize their potential in weight management strategies.”
Acknowledgements
The authors would like to thank Middle East University for covering the publication fees of this study.
Competing Interests
The authors have no competing interests to declare.