Monthly Archives: November 2026

Testing an Algorithm for Assessing Internet User Activity in the North Caucasus Under Seismic Hazard Conditions in Spring 2026

Leave a reply

DOI: 10.31038/GEMS.20268104

Abstract

The results of an assessment of Internet user activity under seismic hazard conditions in the North Caucasus before a series of earthquakes in late February–early March 2026 are presented. The analysis used Google search engines and estimates of query volumes for terms associated primarily with biological earthquake precursors, together with software for automated screening of Google search data.

The frequency of Internet searches for the terms “insects, birds, air, ambulance, dolphins” increases in advance under seismic hazard conditions and is consistent with descriptive models of precursory seismogenic disturbances of geophysical fields.

Keywords: Earthquake, Precursors, Trend, Search engines, Internet, Population

Introduction

The anthropogenic component of the environment responds to changes in the natural environment. This response is manifested in changes in behavioral activity at conscious and unconscious levels and in the intensity of communication, leading to variations in collective consciousness. Behavioral activity in the form of psychomotor acts is recorded in communication media [1]. The Internet makes it possible to track changes in Internet users’ activity through the flow of electronic messages containing signals associated with natural processes and phenomena and the disturbances of geophysical fields that they induce. Internet users record their observations in messages, including observations of other populations of organisms, and seek confirmation of their perceptions. The self-organization of Internet users through collective behavior amplifies weak observational signals from individual users by increasing their number and the repetition of the terms being discussed. Accordingly, monitoring Internet users’ activity using a set of topic-specific queries can be used to assess collective consciousness during disturbances of geophysical fields and to identify the causes of these disturbances [2-4]. A set of topic-specific terms constitutes a working ontological model of the process or phenomenon being assessed. This also applies to seismic processes, for which, particularly when instrumental observations are scarce, special attention should probably be paid to terms associated with biological earthquake precursors. The set of terms adapted to a seismically hazardous region should be refined after each strong seismic event. Internet search engines [5] are tools for both information retrieval and the analysis of population activity and variations in collective consciousness. Internet search engines allow the boundaries of the study region to be specified.

Experimental Data

In February–March 2026, a deformation wave propagated across the Greater Caucasus and was associated with a series of minor earthquakes. These began in the Eastern Caucasus. However, the most distinctly felt earthquakes occurred near Sochi on March 3, 2026:

  1. 12: 09: 47, latitude 43.52°, longitude 39.54°, hypocentral depth 10 km, mb = 4.4, I0 = 4; and
  2. 13: 05: 38, latitude 43.52°, longitude 39.49°, depth 10 km, mb = 4.7, I0 = 4.5–5.0.

Another earthquake was recorded near Sochi on March 6 at 04: 29: 09, latitude 43.52, longitude 39.60, depth 10 km, mb = 4.5, I0 = 4–4.5. At 13: 44: 24, another deep-focus earthquake was recorded in the Eastern Caucasus (hypocentral depth 100 km, latitude 43.38, longitude 46.53, mb = 4.5).

These earthquakes are classified as felt events on the SIS-17 scale [6].

Table 1 lists the preceding earthquakes in the region, according to data from the Geophysical Survey of the Russian Academy of Sciences [7].

Table 1: Characteristics of earthquakes in the Caucasus.

No. Date, time GMT Latitude, ° Longitude, ° Depth, km mb* I0* Region
1 2026-02-23 08: 28 43.60 45.80 110 3.9 — Eastern Caucasus
2 2026-02-27 23: 38 40.58 48.75 10 4.3 3.5–4 Eastern Caucasus
3 2026-02-27 22: 08 44.59 37.45 10 3.5 — Western Caucasus
4 2026-02-27 22: 01 40.64 48.40 10 4.0 3 Eastern Caucasus
5 2026-02-28 02: 34 40.77 48.60 10 4.3 3.5–4 Eastern Caucasus
6 2026-03-01 21: 42 42.75 46.78 70 4.0 — Eastern Caucasus
7 2026-03-01 01: 01 42.83 47.72 10 3.7 2–2.5 Eastern Caucasus
8 2026-03-02 14: 52 38.27 38.44 5 4.5 5–5.5 Turkey
9 2026-03-02 12: 52 42.39 45.80 2 3.4 4 Eastern Caucasus

* mb — earthquake magnitude; I0 — shaking intensity.

The experiment used software for automated screening of data from the Google search engine using topic-specific queries.

Discussion of Experimental Results

Figure 1 shows changes (in %) in the normalized volumes of Google search queries submitted by Internet users in Krasnodar Krai for the terms “insects, birds, air, ambulance, dolphins” before the Sochi earthquakes of March 3 and 6, 2026.

Figure 1: Popularity of Internet search queries in Krasnodar Krai. Arrows indicate earthquakes (blue arrows mark those in the Eastern Caucasus).

Figure 1: Popularity of Internet search queries in Krasnodar Krai. Arrows indicate earthquakes (blue arrows mark those in the Eastern Caucasus).

The arrows in Figure 1 mark the earthquakes listed in Table 1. Earthquakes in the Eastern Caucasus are highlighted in color. The daily numbers of these queries were normalized to the highest query count (100%), which in this case corresponded to the term “dolphins.” No significant changes were found for domestic and wild animals or for the term “water.” The terms were selected empirically. Variations in the search query volumes of Internet users in Krasnodar Krai were consistent with descriptive models of precursory phenomena, with increased activity before earthquakes. The deformation wave in the Earth’s crust propagated across the Caucasus from east to west. The Eastern Caucasus became active first. The full moon occurred on March 3. The bombing of Iran should also be kept in mind. In the collected statistics, changes in air composition became an exceptionally popular search topic around the earthquakes of March 3 and 6, 2026. The query “private ambulance Sochi” was exceptionally popular; the number of queries for “blood pressure pills ambulance” increased by 200%, “ambulance phone number” by 120%, “ambulance Sochi” by 110%, “private ambulance Krasnodar” by 90%, and “ambulance Krasnodar” by 40%. People dreamed of dolphins and insects. Considerable interest in ornithology can be explained by observations of the sky and seasonal bird migration. The largest query volumes came from the major Internet populations of Krasnodar, Greater Sochi, Tuapse, and Anapa. The degree of order in the sum of normalized queries in Figure 1 can be characterized by the coefficient of variation of the query counts shown in Figure 2, without resorting to spectral or fractal analysis. Each value was assigned to the right boundary of a sliding one-week window.

The coefficient of variation in Figure 2 revealed a wave of increasing order in the sum of queries before the Sochi earthquakes of March 3 and 6, 2026. A strong geomagnetic storm (K = 7) was recorded on January 20–21, 2026. However, it is not reflected in Figure 2. Therefore, the coefficient of variation peaks on February 22, 2026, in Figure 2 are not associated with the weak geomagnetic storm with K = 4++. The earthquake of March 3, 2026, occurred at a minimum of geomagnetic activity (K = 1). Under the assumption that a deformation wave propagated toward Sochi from the Eastern Caucasus, the results for Internet users’ activity in Chechnya (Figure 3a) and Dagestan (Figure 3b) are useful.

Figure 2: Changes in the weekly coefficient of variation of the sum of queries.

Figure 2: Changes in the weekly coefficient of variation of the sum of queries.

(a) Chechnya

Figure 3(a): Popularity of Internet search queries in Chechnya.

(b) Dagestan

Figure 3(b): Popularity of Internet search queries in Dagestan.

Figure 3: Popularity of Internet search queries in Chechnya (a) and Dagestan (b).

The results in Figure 3 do not contradict the hypothesis that the consciousness of Internet users in the Eastern Caucasus responds to a deformation wave in the Earth’s crust. To verify the results and select the search queries, findings from a similar experiment were used for the earthquake near Anapa on February 9, 2026, at 23: 21 GMT (February 10 at 02: 21 Moscow time), with magnitude M = 5.3 and a hypocentral depth of 10 km. The epicenter (φ = 45.01° N, λ = 37.51° W) was located 19 km east-northeast of Anapa and 36 km northwest of Novorossiysk. A distinctive feature of this earthquake was that Krasnodar, the largest city in the region (116 km from the epicenter), and its Internet users fell within the potential zone of precursory disturbances of geophysical fields (R = 100.43M km, where M is magnitude). Figure 4 shows the sum of queries for the terms “insects, ambulance, wild animals,” whose query counts were comparable.

The results in Figure 3 confirm increased activity of insect and wild animal populations and increased discussion of ambulance services 3–4 days before the earthquake, consistent with the lead time of precursors for earthquakes of this intensity. The new moon occurred on February 17. The overlap between the terms used in the experiments makes it possible to use the resulting set to assess variations in collective consciousness through the activity of the Internet population of Krasnodar Krai under seismic hazard conditions.

Figure 4: Sum of search queries for the terms “insects, ambulance, wild animals” before and after the earthquake of February 10, 2026 (black vertical arrow).

Figure 4: Sum of search queries for the terms “insects, ambulance, wild animals” before and after the earthquake of February 10, 2026 (black vertical arrow).

Conclusion

The Google search engine enables a spatial and temporal assessment of collective consciousness based on user activity within specified geographic boundaries and is a promising tool for geophysical monitoring. Using the felt earthquakes near Sochi on March 3 and 6, 2026, as a case study, a set of terms was developed to assess the activity of the Internet population of Krasnodar Krai under seismic hazard conditions, with software for automated screening of Google search data using topic-specific queries. The frequency of the terms “insects, birds, air, ambulance, dolphins” in Internet users’ behavior increases in advance under seismic hazard conditions and is consistent with descriptive models of precursory seismogenic disturbances of geophysical fields.

References

  1. Tertyshnikov AV (2020) The Internet-community response to the Mexican Earthquake of February 1, 2019. E3S Web of Conferences, 196, 03007. STRPEP 2020.
  2. Tertyshnikov AV (2023) The reaction of Internet Users to the Seismic Hazard. World Journal of Environmental Biosciences. 12: 14-17.
  3. Tertyshnikov AV (2022) The November 14, 2018 Sakurajima Volcano Eruption through the Lens of Internet-Based Seismic Hazard Semantic Criteria. Heliogeophysical Research [Solar-Terrestrial Physics Research]. 35: 12-20. [Original Russian publication].
  4. Tertyshnikov AV, Obelchenko DS (2025) Trend of the informational Precursor of macroseismicity near Santorini Island in early 2025 using ai resources / Proceedings of the International Scientific and Practical Conference “Actual problems of physics and technology” S.U. Umarov Physical-Technical Institute of the National Academy of Sciences of Tajikistan. Dushanbe: Donish Publishing House, 2025, 271 pages (60-66).
  5. Tertyshnikov AV, Obelchenko DS (2026) Activity of Turkey’s internet community before the Istanbul earthquake” with reference number // Global Journal of Ecology. 11: 001-004.
  6. State Standard 57546–2017. SIS-17 — Earthquakes. Seismic Intensity Scale. Moscow: Standartinform, 2017.
  7. Earthquake Data. Geophysical Survey of the Russian Academy of Sciences [Electronic resource].

KCNA1 in Glioblastoma: Biological Functions and Therapeutic Implications

Leave a reply

DOI: 10.31038/CST.20261122

Abstract

KCNA1 encodes Kv1.1, a voltage-gated potassium channel primarily recognized for its role in neuronal excitability. Evidence from glioma models has expanded its biological relevance to membrane potential regulation, tumor invasion, ferroptosis-associated responses, and mitochondrial homeostasis. Early electrophysiological experiments established functional Kv1.1 currents in glioma-derived cells, while subsequent transcriptomic studies identified heterogeneous KCNA1 expression across glioblastoma samples and invasive tumor regions. Pharmacological and genetic interventions further associated KCNA1 with malignant cellular behavior, and recent findings implicated SLC7A11-related antioxidant defense in its effects on ferroptosis susceptibility. Together, these observations suggest that KCNA1 participates in multiple aspects of glioblastoma cell biology. However, the relationship between its ion-conducting activity and tumor-associated phenotypes remains incompletely defined. Moreover, the physiological importance of Kv1.1 in neuronal function complicates therapeutic targeting. This mini-review summarizes current experimental evidence and examines the mechanistic and translational questions that must be addressed to determine the therapeutic relevance of KCNA1 in glioblastoma.

Keywords: KCNA1, Kv1.1, Glioblastoma, Invasion, Ferroptosis, Potassium channels

Introduction

Glioblastoma (GBM) is characterized by extensive infiltration into surrounding brain tissue and resistance to conventional treatment. Recent studies have demonstrated that glioblastoma cells integrate into local and long-range neuronal circuits, with neuronal activity and synaptic interactions influencing tumor invasion, progression, and treatment responses [1,2]. Electrophysiological investigations have further identified distinct electrical properties among tumor cells at the infiltrative brain–tumor interface, highlighting the functional heterogeneity of glioblastoma [3]. Together with growing evidence implicating potassium channels in glioblastoma biology [4], these findings provide a framework for examining the contribution of individual ion channels to malignant cellular behavior.

KCNA1 encodes Kv1.1, a member of the Shaker-related voltage-gated potassium channel family. Although Kv1.1 is best known for regulating neuronal excitability, its expression and electrical activity have also been documented in glioma-derived cells [5]. Subsequent studies have identified KCNA1-associated expression patterns in glioblastoma, including enrichment in invasive tumor populations, and have linked its genetic manipulation to changes in tumor growth, invasion, and ferroptosis-related responses [6-9].

These findings have expanded interest in KCNA1 beyond its established neurological functions. This mini-review examines the evidence connecting KCNA1 to glioblastoma biology, with particular emphasis on its functional expression, involvement in malignant cellular behavior, and potential relevance to therapeutic intervention.

KCNA1 Expression and Function in Glioma

Functional Expression and Spatial Distribution

Early electrophysiological investigations demonstrated endogenous Kv1.1 expression in C6 glioma cells, where dendrotoxin-I-sensitive potassium currents contributed substantially to the resting membrane potential. Elevation of intracellular cyclic AMP accelerated KCNA1 mRNA degradation, followed by reduced Kv1.1 protein expression and potassium conductance [5]. These findings established a relationship between KCNA1 expression and the electrical properties of glioma-derived cells.

Subsequent transcriptomic investigations identified KCNA1 among genes associated with glioblastoma [7,9]. Integrated bioinformatic analyses included KCNA1 within a group of candidate genes related to neuronal signaling and glioblastoma-associated molecular alterations [9]. Another study reported reduced KCNA1 expression in glioblastoma tissues and cell lines compared with non-tumor controls, identifying it among several potential molecular markers of the disease [7].

Spatial analyses revealed additional heterogeneity in KCNA1 expression within glioblastoma. In six pediatric glioblastoma-derived orthotopic models, matched tumor-core and invasive-front cell populations were isolated and compared. KCNA1 mRNA expression was elevated in invasive populations in four models, while increased protein staining was observed in invasive regions in five models [6]. These findings demonstrated that KCNA1 expression can differ between spatially distinct tumor populations, adding a regional dimension to its expression profile in glioblastoma.

KCNA1-Associated Invasion and Tumor Progression

The identification of KCNA1 in invasive tumor populations was accompanied by functional evidence linking potassium channel activity to glioblastoma invasion. Treatment with ADWX-1, Agitoxin-2, or 4-aminopyridine (4-AP) reduced the invasive capacity of patient-derived glioblastoma cells. In an orthotopic model, systemic administration of 4-AP also suppressed intracranial tumor infiltration and prolonged animal survival [6].

Genetic manipulation provided complementary evidence for the involvement of KCNA1 in malignant cellular behavior. KCNA1 knockdown reduced glioblastoma cell proliferation, migration, and invasion, whereas its overexpression enhanced these phenotypes [8]. Intracranial experiments further demonstrated reduced tumor progression following KCNA1 depletion. These observations extended the functional association of KCNA1 from spatially invasive tumor populations to broader growth- and invasion-related phenotypes.

Together, electrophysiological, transcriptomic, and functional studies establish a foundation for investigating KCNA1 in glioma biology. They encompass several distinct dimensions of the gene, including its contribution to membrane electrical properties, heterogeneous expression within tumors, and association with malignant cellular behavior.

KCNA1, Ferroptosis, and Mitochondrial Dysfunction

Ferroptosis is a regulated form of cell death characterized by iron-dependent lipid peroxidation [10,11]. Its regulation involves the balance between oxidative damage to membrane phospholipids and cellular antioxidant defenses. The SLC7A11–glutathione (GSH)–GPX4 pathway is a major protective system, in which cystine uptake supports GSH synthesis and GPX4-dependent detoxification of lipid peroxides [12,13].

KCNA1 has been implicated in this antioxidant network in glioblastoma cells. KCNA1 expression was positively associated with SLC7A11, and KCNA1 knockdown decreased SLC7A11 and GPX4 protein expression, reduced intracellular GSH, and increased reactive oxygen species and lipid peroxidation [8]. Conversely, KCNA1 overexpression attenuated several ferroptosis-associated alterations following erastin exposure. These results linked KCNA1 expression to the cellular response to ferroptotic stress and suggested an association with SLC7A11-dependent antioxidant regulation.

Changes in mitochondrial function were also observed following KCNA1 manipulation. KCNA1 depletion was associated with reduced mitochondrial membrane potential and ATP production, increased mitochondrial reactive oxygen species, and alterations in mitochondrial ultrastructure [8]. Immunofluorescence analyses further demonstrated overlapping KCNA1 and TOMM20 signals, indicating a spatial association between KCNA1 immunoreactivity and mitochondria. Outside glioblastoma, KCNA1 silencing has also been associated with mitochondrial damage and suppression of malignant phenotypes in cervical cancer cells [14].

These observations extend the biological profile of KCNA1 from electrical regulation and tumor invasion to cellular redox balance and mitochondrial function. In glioblastoma, the reported changes in SLC7A11 expression, lipid peroxidation, and mitochondrial activity provide a basis for considering KCNA1-associated cellular stress responses alongside its established invasion-related phenotypes.

Mechanistic Questions and Therapeutic Implications

The available studies identify several KCNA1-associated phenotypes, but their mechanistic relationships remain to be resolved. A central distinction concerns KCNA1 expression and Kv1.1-mediated potassium conductance. Functional Kv1.1-associated currents have been demonstrated in C6 glioma cells [5], whereas the more recent evidence for glioblastoma progression comes largely from pharmacological inhibition and genetic manipulation [6,8]. Whether the effects of KCNA1 depletion on invasion, proliferation, and ferroptosis susceptibility require channel conductance remains an important question. Electrophysiological characterization combined with rescue experiments using channel-function-deficient KCNA1 variants would help establish the contribution of ion transport to these phenotypes.

The relationship between KCNA1-associated invasion and ferroptosis regulation also warrants closer examination. Although KCNA1 manipulation influences both invasive behavior and SLC7A11-related antioxidant responses, these observations do not establish that ferroptosis resistance mediates the invasive phenotype. The two processes may share upstream regulation or represent distinct consequences of KCNA1 activity. Similarly, mitochondrial dysfunction following KCNA1 depletion does not establish direct mitochondrial ion channel activity or identify mitochondria as the initiating site of ferroptotic signaling. Functional rescue, time-resolved analyses, and compartment-specific measurements could distinguish these possibilities.

Another consideration is the heterogeneous expression of KCNA1 across experimental settings. Reduced expression relative to non-tumor controls and enrichment within invasive tumor regions describe different biological comparisons [6,7]. Their interpretation requires attention to tumor cellular composition, anatomical location, and molecular subtype. Spatially resolved analyses combined with functional studies in molecularly characterized patient-derived models would help determine whether KCNA1 dependence is associated with particular glioblastoma cell populations.

From a therapeutic perspective, potassium channel inhibition offers an experimentally accessible means of modifying glioblastoma invasion. Among the compounds evaluated, 4-AP is of particular interest because its extended-release formulation, dalfampridine, is approved for improving walking performance in patients with multiple sclerosis [15]. Nevertheless, 4-AP blocks multiple voltage-gated potassium channel subtypes, and the peptide inhibitor ADWX-1 exhibits greater potency against Kv1.3 than Kv1.1 [16]. Accordingly, the anti-invasive effects observed with these agents cannot be assigned exclusively to Kv1.1 inhibition. The development or application of more selective pharmacological tools, supported by genetic target validation, would strengthen the rationale for KCNA1-directed intervention.

Neurological safety and drug delivery present additional challenges. Kv1.1 has an established role in neuronal excitability, and KCNA1 dysfunction is associated with neurological manifestations, including episodic ataxia [17]. Dalfampridine treatment also carries a recognized risk of seizures, particularly with increased systemic exposure [15]. These considerations are relevant to patients with intracranial tumors, in whom seizure susceptibility may already be elevated. In addition, effective targeting of infiltrative glioblastoma cells requires sufficient drug exposure in brain regions where the blood–brain barrier may remain relatively intact. Therapeutic development must therefore balance tumor activity, channel selectivity, brain penetration, and preservation of normal neuronal function.

The translational value of KCNA1 ultimately depends on establishing a reproducible functional dependence in glioblastoma cells and identifying a pharmacological approach capable of exploiting that dependence within an acceptable neurological safety margin.

Conclusions

Research on KCNA1 in glioma has progressed from the characterization of functional potassium currents to the identification of invasion-associated expression patterns and links with cellular antioxidant responses. These findings broaden the biological relevance of Kv1.1 in glioblastoma, while highlighting the complexity of interpreting ion channel expression in a heterogeneous tumor.

The next stage of investigation should move beyond expression-based associations toward defining the cellular contexts and molecular functions that make KCNA1 relevant to glioblastoma progression. Such evidence will determine whether KCNA1 remains primarily a marker of particular tumor phenotypes or develops into a therapeutically actionable vulnerability.

Declarations

Conflict of Interest

The authors declare that they have no competing interests.

Authors’ Contributions

Yutong Mei: Conceptualization, Writing – original draft.

Yi Jin: Writing – original draft.

Yuhan Liu: Literature review, Writing – original draft.

Siyuan Zhao: Literature review, Validation.

Runze Han: Literature review, Validation.

Yucheng Shen: Literature review, Validation.

Zongqi Wang: Conceptualization, Supervision, Writing – review & editing.

Jiangang Liu: Supervision, Funding acquisition, Writing – review & editing.

All authors reviewed and approved the final version of the manuscript.

Funding

This work was supported by the Soochow University Undergraduate Extracurricular Academic Research Fund (KY2023096A), and the Scientific Research Project of the Health Commission of Jiangsu Province (M2022050).

Artificial Intelligence (AI)-Assisted Technology

ChatGPT was used to assist with language editing and manuscript preparation. All authors reviewed the manuscript and take full responsibility for the accuracy and integrity of its content.

References

  1. Sun Y, et al. (2025) Brain-wide neuronal circuit connectome of human glioblastoma. Nature. [crossref]
  2. Tetzlaff SK, et al. (2025) Characterizing and targeting glioblastoma neuron-tumor networks with retrograde tracing. Cell. [crossref]
  3. Tong T, et al. (2026) Combined patch-clamp electrophysiology and single-cell genomic analysis reveal spiking tumor cells at the neocortical glioblastoma interface in humans. Neuro Oncol. [crossref]
  4. Maliszewska-Olejniczak K, et al. (2026) Potential therapeutic targeting of BK(Ca) channels in glioblastoma treatment. Mol Oncol. [crossref]
  5. Allen ML, Koh DS, Tempel BL (1998) Cyclic AMP regulates potassium channel expression in C6 glioma by destabilizing Kv1.1 mRNA. Proc Natl Acad Sci U S A. [crossref]
  6. Huang Y, et al. (2021) Spatial Dissection of Invasive Front from Tumor Mass Enables Discovery of Novel microRNA Drivers of Glioblastoma Invasion. Adv Sci (Weinh) [crossref]
  7. Huang Z, et al. (2024) Bioinformatics Analysis and Experimental Validation for Exploring Key Molecular Markers for Glioblastoma. Appl Biochem Biotechnol. [crossref]
  8. Wang W, et al. (2024) KCNA1 promotes the growth and invasion of glioblastoma cells through ferroptosis inhibition via upregulating SLC7A11. Cancer Cell Int. [crossref]
  9. Yang J, Yang Q (2020) Identification of Core Genes and Screening of Potential Targets in Glioblastoma Multiforme by Integrated Bioinformatic Analysis. Front Oncol. [crossref]
  10. Banu MA, et al. (2024) A cell state-specific metabolic vulnerability to GPX4-dependent ferroptosis in glioblastoma. EMBO J. [crossref]
  11. Wang Z, et al. (2024) AMPKalpha1-mediated ZDHHC8 phosphorylation promotes the palmitoylation of SLC7A11 to facilitate ferroptosis resistance in glioblastoma. Cancer Lett. [crossref]
  12. Mori M, et al. (2026) Dual inhibition of xCT and GGCT induces ferroptosis in glioblastoma cells by depleting cysteine and disrupting redox homeostasis. Cell Death Discov. [crossref]
  13. Tian Q, et al. (2025) IDO1 inhibits ferroptosis by regulating FTO-mediated m6A methylation and SLC7A11 mRNA stability during glioblastoma progression. Cell Death Discov. [crossref]
  14. Liu L, et al. (2019) Silencing of KCNA1 suppresses the cervical cancer development via mitochondria damage. Channels (Austin) [crossref]
  15. Cornblath DR, Bienen EJ, Blight AR (2012) The safety profile of dalfampridine extended release in multiple sclerosis clinical trials. Clin Ther. [crossref]
  16. Yin SJ, et al. (2008) Different residues in channel turret determining the selectivity of ADWX-1 inhibitor peptide between Kv1.1 and Kv1.3 channels. J Proteome Res. [crossref]
  17. D’Adamo MC, et al. (2020) Kv1.1 Channelopathies: Pathophysiological Mechanisms and Therapeutic Approaches. Int J Mol Sci. [crossref]

Content Validation of a Colour-Coded Glycemic Index Food Wheel for Diabetes Dietary Education: A Cross- Sectional Study of Physician and Dietitian Consensus

Leave a reply

DOI: 10.31038/EDMJ.20261042

Abstract

Aim and Background: Colour-coded, hand-portion visual tools are increasingly used to translate glycemic index (GI) science into bedside dietary counselling, yet few such tools developed for Indian food habits have undergone formal content validation. This study evaluated the content validity of a newly developed Glycemic Wheel, a circular colour-coded educational aid that classifies commonly consumed South Indian foods by GI category and expresses portions using household hand-based measures, among a multidisciplinary panel of physicians and dietitians.

Methods: In this cross-sectional expert-panel validation study, 161 experts (61 physicians holding an MD in General Medicine and 100 dietitians holding a B.Sc. or M.Sc. in Clinical Nutrition and Dietetics) independently rated the relevance of 27 items across six content domains (Food Selection, Glycemic Index Classification, Portion Size Information, Visual Organization, Clinical Utility, and Overall Assessment) on a four-point ordinal scale. Item-level (I-CVI) and scale-level (S-CVI/Ave and S-CVI/UA) content validity indices were computed following established psychometric conventions.

Results: All 27 items exceeded the recommended I-CVI acceptability threshold of 0.78, with values ranging from 0.851 to 0.932. Domain-wise I-CVI ranges were: Food Selection 0.870-0.913; Glycemic Index Classification 0.907-0.932; Portion Size Information 0.851-0.913; Visual Organization 0.851-0.925; Clinical Utility 0.876-0.913; and Overall Assessment 0.882-0.913. The overall Scale-Level Content Validity Index by the averaging method (S-CVI/Ave) was 0.897, and the Scale-Level Content Validity Index by universal agreement (S-CVI/UA) was 0, a pattern consistent with the size of the expert panel.

Conclusion: The Glycemic Wheel demonstrated good-to-excellent item-level content validity and good scale-level content validity across a large, multidisciplinary panel of physicians and dietitians, supporting its scientific and educational adequacy as a diabetes dietary counselling aid before pilot implementation.

Clinical Significance: A content-validated, regionally tailored GI wheel offers clinicians and dietitians a standardized, low-cost adjunct for point-ofcare
dietary counselling, particularly valuable in high-volume outpatient settings where dietitian access is limited and rapid, visually anchored patient communication is required.

Keywords: Glycemic index, Diabetes mellitus, Type 2, Patient education, Content validity, Expert panel, Visual aids, Dietary counselling; India

Introduction

Diabetes has become one of the defining public health challenges in India. The 11th edition of the International Diabetes Federation (IDF) Diabetes Atlas estimates that India was home to roughly 89.8 million adults living with diabetes in 2024, second globally only to China, with age-standardised prevalence projected to rise further by 2050 [1]. Complementing this global estimate, the ICMR-INDIAB national cross-sectional study confirmed wide inter-state variation in diabetes prevalence across the country, with several southern states, including Tamil Nadu, already in an advanced stage of the epidemiological transition [2]. Nutrition therapy sits at the centre of managing this burden, and within nutrition therapy, the glycemic index (GI) has remained one of the most enduring, if imperfectly applied, tools available to clinicians and dietitians.

The Glycemic Index as a Clinical Nutrition Tool

The GI concept, first described by Jenkins and colleagues in 1981, ranks carbohydrate-containing foods according to their postprandial glycaemic response relative to a reference food [3]. Foods are conventionally grouped as low, moderate, or high GI, and this classification has since been incorporated into major nutrition therapy guidance for diabetes, including the American Diabetes Association (ADA) consensus report on medical nutrition therapy, which recognises glycemic index and glycemic load as adjuncts to overall carbohydrate quality assessment [4,5]. The World Health Organization and Food and Agriculture Organization joint expert consultation on diet and chronic disease prevention similarly identifies the quality, not merely the quantity, of dietary carbohydrate as relevant to metabolic
risk [6].

From Numbers to Practice: The Communication Gap

Despite this evidence base, GI values are typically published as long numerical tables that are poorly suited to a busy outpatient consultation. Two barriers compound the problem. First, health literacy and numeracy vary widely among patients with diabetes, and the literature consistently links limited health literacy to poorer self-management, higher hospitalisation, and worse glycaemic outcomes [7]. Second, even accurate portion instructions are of little use if patients cannot translate them into an actual plateful of food; household measuring tools are frequently unavailable in the home, whereas hands are always present. A validation study using finger width, fist, and palm measurements found that hand-based portion estimation performed comparably to conventional household measures such as cups and spoons across a range of solid and liquid foods, supporting hands as a practical, low-literacy-friendly portion cue [8].

Visual Communication in Diabetes Education

Pictures and colour-coded visual schema have a demonstrated advantage over text-only patient materials. A widely cited review by Houts and colleagues found that pictures closely linked to text improve attention, comprehension, and recall of health information, with the largest gains observed among patients with limited literacy [9]. This body of evidence has motivated the development of pictogram and colour-based diabetes education tools internationally, and underlies
the rationale for wheel- or plate-based visual formats that condense a complex nutrition message into a single glance.

Regional Food Specificity

GI values are not interchangeable across food cultures. South Indian meals are structured around rice, fermented cereal-legume preparations, and a distinct repertoire of vegetables, pulses, and snacks whose glycaemic behaviour differs from Western reference foods. Population-based work from Chennai has shown that dietary carbohydrate quality and glycemic load are independently associated with both HDL-cholesterol and incident type 2 diabetes among urban South Indian adults, reinforcing the clinical relevance of food-specific GI classification in this population [10,11]. The ICMR-National Institute of Nutrition Dietary Guidelines for Indians likewise emphasise portion-appropriate, regionally grounded dietary counselling as a cornerstone of non-communicable disease prevention
[12].

Development of the Glycemic Wheel

Against this background, our team developed the Glycemic Wheel, a circular, colour-coded educational aid intended for use at the point of care during diabetes dietary counselling and structured diabetes self-management education and support (DSME/S) sessions [13]. The Wheel organises commonly consumed South Indian foods into six food groups — grains and cereals, pulses and legumes, vegetables, fruits, milk and milk products, and snack foods — and assigns each item a Low, Moderate, or High GI classification using a green-amber-red colour scheme. Alongside each food, the Wheel depicts an appropriate serving size expressed in familiar Indian household handbased measures, such as a closed fist or a cupped palm, so that the visual portion cue and the glycaemic message are presented together rather than as separate pieces of information the patient must mentally combine. The tool is designed to be used interactively by a physician or dietitian during a brief counselling encounter, functioning as a shared reference point rather than a document to be read independently.

To our knowledge, no previously published South Indian food specific GI wheel has combined colour-coded GI classification with hand-based portion cues in a single point-of-care format, and the tool required systematic evaluation before wider deployment across our clinical network. Content validity, the extent to which the items of an instrument adequately represent the construct they are intended to measure, is the recommended first step in the validation of any new clinical or educational instrument, and is typically established through structured, quantified expert judgement rather than through statistical inference alone [14-16].

The aim of this study was therefore to establish the content validity of the Glycemic Wheel among a large, multidisciplinary panel of physicians and dietitians, and to quantify item-level and scale-level content validity indices across the instrument’s six content domains, together with the clinical implications of the findings for each domain.

Materials and Methods

Study Design

This was a cross-sectional, expert-panel content validation study conducted to establish the content validity of the Glycemic Wheel prior to pilot implementation. The validation followed established recommendations for the systematic, quantitative appraisal of instrument content by content experts [14-18].

Description of the Instrument and Content Domains

The Glycemic Wheel rating instrument comprised 27 items distributed across six pre-specified content domains reflecting the structural components of the tool: (A) Food Selection Appropriateness (6 items), evaluating whether the grain/cereal, pulse/legume, vegetable, fruit, milk product, and snack food items included represent commonly consumed South Indian foods; (B) Glycemic Index Classification Accuracy (4 items), evaluating the correctness and appropriateness of the Low/Moderate/High GI classification and cut-offs; (C) Portion Size Information (3 items), evaluating the appropriateness, comprehensibility, and clinical usefulness of the hand-based household portion cues; (D) Visual Organization (5 items), evaluating colour-coding, food grouping logic, wheel layout, and text/font readability; (E) Clinical Utility (5 items), evaluating suitability of the Wheel for routine counselling, patient food-choice improvement, and use in Type 2 diabetes and community education; and (F) Overall Assessment (4 items), evaluating global scientific content, educational value, design quality, and readiness for clinical implementation.

Expert Panel Selection

Content validation was performed through structured expert review, consistent with recommended practice for selecting and using content experts in instrument development [19]. The panel comprised two professional categories: physicians holding an MD in General Medicine with clinical experience in diabetes care, and dietitians holding a B.Sc. or M.Sc. in Clinical Nutrition and Dietetics. Experts were drawn from clinical and nutrition practice settings across TamilNadu, South India, to ensure ratings reflected real-world familiarity with the regional foods depicted on the Wheel.

Validation Materials and Process

Each expert received the complete Glycemic Wheel, a structured 27 item rating sheet organised by the six content domains, and written instructions describing the objective of the exercise and the item rating procedure. Experts rated each item independently to minimise anchoring or social-desirability bias between raters.

Item Rating Scale

Item relevance was rated on a four-point ordinal scale (1 = not relevant, 2 = somewhat relevant, 3 = relevant, 4 = highly relevant), consistent with standard content validity index methodology [14,16].

Content Validity Index Calculation

The Item-Level Content Validity Index (I-CVI) was calculated for each item as the proportion of experts who rated the item as either relevant or highly relevant (a rating of 3 or 4), divided by the total number of experts [14,16]. Consistent with the widely cited recommendation of Polit, Beck, and Owen, an I-CVI of 0.78 or higher, in a panel of six or more experts, was taken as evidence of acceptable item-level content validity [17,18].

The Scale-Level Content Validity Index by the averaging method (S-CVI/Ave) was calculated as the mean of all I-CVI values across the 27 items, with a value of 0.90 or higher conventionally interpreted as excellent scale-level content validity and 0.80 or higher as the minimum acceptable threshold [14,17,18]. The Scale-Level Content Validity Index by universal agreement (S-CVI/UA) was calculated as the proportion of items rated 3 or 4 by every expert on the panel; Polit and Beck have cautioned that the universal-agreement method becomes progressively more conservative, and less informative, as panel size increases, and recommend the averaging method as the more meaningful metric for larger panels [17].

Statistical Analysis

Content validity analysis was descriptive. Item-level and domainlevel I-CVI values, together with the overall S-CVI/Ave and S-CVI/UA, were calculated from the raw expert ratings using standard spreadsheet-based computation and are presented in tabular form. Panel demographic characteristics are summarised as counts, percentages, means, and ranges.

Ethical Considerations

As this phase of the study involved the professional judgement of expert reviewers rather than patient-level data, it did not require formal ethics committee approval under standard institutional practice for instrument content validation. Expert participation was voluntary, and all individual responses were anonymised prior to analysis.

Results

Expert Panel Composition

A total of 161 experts completed item-level ratings for the Glycemic Wheel: 61 physicians (37.9%) holding an MD in General Medicine, and 100 dietitians (62.1%), of whom 56 held an M.Sc. in Clinical Nutrition and Dietetics and 44 held a B.Sc. in Nutrition and Dietetics. Physicians had a mean of 17.7 years of professional experience (range 5-30 years), and dietitians had a mean of 9.1 years (range 1-20 years); across the full panel, mean experience was 12.4 years (range 1-30 years) (Table 1).

Table 1: Composition of the expert panel

Category N (%) Qualification Mean Years Experience (Range)
Physicians 61 (37.9%) MD (General Medicine) 17.7 (5–30)
Dietitians 100 (62.1%) M.Sc. Clinical Nutrition & Dietetics (n=56); B.Sc. Nutrition & Dietetics (n=44) 9.1 (1–20)
Overall Panel 161 (100%) — 12.4 (1–30)

Abbreviations: MD, Doctor of Medicine; M.Sc., Master of Science; B.Sc., Bachelor of Science

Item-Level Content Validity by Domain

All 27 items exceeded the recommended I-CVI acceptability threshold of 0.78 [17,18], with item-level values ranging from 0.851 to 0.932 across the full instrument. Domain-wise results, together with their clinical importance, are summarised in Table 2 and described below.

Table 2: Item-level content validity index by domain

Domain Items I-CVI Range Domain Mean I-CVI Interpretation
A. Food Selection Appropriateness 6 0.870–0.913 0.892 All items acceptable
B. Glycemic Index Classification Accuracy 4 0.907–0.932 0.921 Highest domain; excellent relevance
C. Portion Size Information 3 0.851–0.913 0.890 Acceptable; lowest-scoring domain
D. Visual Organization 5 0.851–0.925 0.891 Acceptable; readability flagged
E. Clinical Utility 5 0.876–0.913 0.897 Excellent item relevance
F. Overall Assessment 4 0.882–0.913 0.896 Excellent item relevance
Overall Scale (27 items) 27 0.851–0.932 S-CVI/Ave = 0.897 Good scale-level content validity

Abbreviations: I-CVI, Item-Level Content Validity Index; S-CVI/Ave, Scale-Level Content Validity Index (averaging method). Interpretation follows the threshold of I-CVI ≥ 0.78 for acceptable item-level content validity in panels of six or more experts [17,18].

Domain A — Food Selection Appropriateness (I-CVI range 0.870-0.913)

All six items evaluating the grain/cereal, pulse/legume, vegetable, fruit, milk product, and snack food selections met the acceptability threshold, with milk and milk product selection recording the domain’s lowest value (0.870, Q5) and pulse/legume selection its highest (0.913, Q2). Clinically, this domain determines whether patients recognise their own everyday foods on the Wheel; strong agreement here supports the ecological validity of the tool’s food repertoire for South Indian dietary patterns.

Domain B — Glycemic Index Classification Accuracy (I-CVI range 0.907-0.932)

This domain recorded the highest agreement of any domain in the instrument, with the correctness of the Low/Moderate/High classification (Q7, I-CVI 0.932) and the appropriateness of the GI cutoff ranges used (Q9, I-CVI 0.932) both among the two highest-rated items overall. Because this domain represents the scientific backbone of the Wheel, this result provides reassurance that the classification scheme is unlikely to mislead counselling messages.

Domain C — Portion Size Information (I-CVI range 0.851-0.913)

Appropriateness of serving sizes for Indian adults (Q11) and comprehensibility of household measures (Q12) both scored 0.913, whereas clinical usefulness of portion sizes during counselling (Q13) scored 0.851, the lowest value in this domain and one of the three lowest item scores in the entire instrument. Portion communication is the step at which GI knowledge is translated into an actual plateful of food, so this comparatively lower, though still acceptable, agreement flags portion cueing as the component most likely to benefit from refinement before wider rollout.

Domain D — Visual Organization (I-CVI range 0.851-0.925)

Colour-coding usability (Q14, I-CVI 0.925) rated strongly, while the item assessing whether the wheel format facilitates quick food selection (Q17) and the item assessing text/font readability (Q18) both scored 0.851. Because the Wheel is intended for use within a brief outpatient encounter, layout speed and legibility directly affect whether the tool is actually usable at the point of care, making this the second domain, alongside portion size, meriting design attention.

Domain E — Clinical Utility (I-CVI range 0.876-0.913)

Items assessing suitability for community diabetes education (Q23) and improvement of patients’ food choices (Q21) scored 0.913 and 0.907 respectively, while suitability for use during routine diabetes counselling (Q19) scored 0.876. This domain bridges the Wheel’s design to its intended real-world function within DSME/S delivery, and its consistently acceptable scores support the tool’s readiness for counselling-based use.

Domain F — Overall Assessment (I-CVI range 0.882-0.913)

Global judgements of overall design supporting effective dietary decision-making (Q26) and overall educational value (Q25) scored 0.913 and 0.907 respectively, while scientific content appropriateness (Q24) and implementation readiness (Q27) both scored 0.882. As a composite judgement domain, these results indicate that experts, having rated individual components, converged on a positive overall appraisal of the instrument (Table 2).

Scale-Level Content Validity

The overall Scale-Level Content Validity Index by the averaging method (S-CVI/Ave) across all 27 items was 0.897. This exceeds the minimum acceptable benchmark of 0.80 proposed by Davis, and approaches, though falls marginally short of, the more stringent 0.90 threshold that Polit and Beck describe as indicative of excellent scale-level content validity [14,17,18]. The Scale-Level Content Validity Index by universal agreement (S-CVI/UA) was 0, meaning that no single item achieved a rating of 3 or 4 from all 161 experts simultaneously. This finding is consistent with the well-documented behaviour of the universal-agreement method in large panels: as the number of raters increases arithmetically, the probability that every rater will independently select the top two response categories on every item falls sharply, which is why Polit and Beck caution against relying on S-CVI/UA once panels extend beyond a small handful of experts, and instead recommend the averaging method as the more interpretable index in panels of this size [17].

Discussion

This study establishes that the Glycemic Wheel, a colour-coded, hand-portion GI education tool developed for South Indian dietary patterns, possesses good-to-excellent item-level content validity and good scale-level content validity when appraised by a large, multidisciplinary panel of 161 physicians and dietitians. Every item cleared the 0.78 I-CVI acceptability threshold, and the overall S-CVI/Ave of 0.897 sits comfortably above the minimum acceptable benchmark for scale-level content validity, even though it falls just short of the more conservative 0.90 “excellent” cut-off [14,17,18].

Comparison with other recently validated diabetes education instruments helps contextualise this result. Liu and colleagues validated a simplified Chinese adaptation of the U.S. CDC Clear Communication Index using diabetes health-education materials and reported a substantially higher S-CVI/Ave of 0.976 [20]. However, their content-validity exercise was rated by only twelve multidisciplinary raters, compared with the 161 experts in the present study. This contrast illustrates a point already noted in the CVI methodology literature: content validity indices calculated by the averaging method tend to be numerically more forgiving with smaller panels, because a handful of concordant raters can drive the average close to 1.0, whereas a panel of over 150 independent raters is statistically far less likely to produce uniformly perfect agreement even when the underlying content is genuinely sound [17]. Viewed this way, an S-CVI/Ave of 0.897 across 161 raters may represent a comparably, or even more,
rigorous demonstration of content adequacy than a nominally higher index computed from a much smaller panel.

A similar pattern is evident when the Glycemic Wheel is set against tools validated through different methodological paradigms altogether. Doucette and colleagues validated a pictogram-based diabetes counselling tool against the American National Standards Institute benchmark of 85% correct comprehension rather than a relevance-based CVI, and reported that individual pictogram sets varied considerably in how well they met this bar, with several requiring revision after the first round of testing [21]. This underlines that content validity, as measured here, and comprehension validity, as measured by pictogram studies, are related but distinct properties; a tool can be judged scientifically and structurally sound by expert reviewers while still requiring separate testing of whether patients themselves correctly interpret it. The present study establishes the former; it does not, and was not designed to, establish the latter.

The Interactive Nutrition-Specific Physical Exam Competency Tool (INSPECT), validated by Zechariah and colleagues through a two round Delphi process with 17 registered dietitian experts, achieved excellent internal consistency (α = 0.97 and 0.96 across the two rounds) [22]. The iterative Delphi design allowed items to be revised and rerated after the first round, a refinement opportunity that the present single-round rating exercise did not include. This represents a genuine methodological difference between the two studies rather than a weakness unique to either: a single-round design of the kind used here is more feasible at the scale of 161 raters, but it forfeits the chance to resolve borderline items, such as those in the Portion Size and Visual Organization domains, through a second round of expert feedback.

Two domains in the present study, Portion Size Information and Visual Organization, recorded the lowest item-level scores in the instrument (I-CVI as low as 0.851), even though both remained comfortably above the acceptability threshold. This pattern is plausible on clinical grounds. Hand-based portion estimation has been validated for a range of solid and liquid foods internationally [8], but South Indian meals frequently combine several food groups
within a single dish, for example a mixed rice preparation or a stuffed flatbread, and translating a hand-portion cue to composite dishes of this kind is inherently more demanding than applying it to a single discrete food item. Likewise, because the Wheel compresses six food groups, three GI categories, and portion cues into one circular graphic, legibility and layout speed are under greater design pressure than in a linear, page-based education material; the marginally lower scores for text readability and rapid food selection are therefore a reasonable signal for iterative design refinement rather than a threat to the tool’s underlying scientific content.

The multidisciplinary composition of the panel is a particular strength of this validation exercise. Many previously published diabetes education tools have been validated by single-discipline panels, whether physicians alone or dietitians alone; the pictogram tool described above, for instance, was developed and reviewed primarily within a pharmacy and diabetes-educator team [21], and even the INSPECT tool, despite its methodological rigour, drew its
expert panel exclusively from registered dietitians [22]. By contrast, the present panel combined 61 physicians and 100 dietitians, and the domain-wise I-CVI values did not diverge in any way that would suggest systematic disagreement between the two professional groups, supporting the ecological validity of the Wheel across the two clinician types most likely to use it in practice.

This study has several limitations that should temper interpretation. First, the single-round rating design, discussed above, did not allow items to be revised and re-rated, unlike Delphi-based validation approaches [22]. Second, the expert panel was drawn from Tamil Nadu, and while this ensured strong familiarity with the specific South Indian foods depicted on the Wheel, it also means the findings may not generalise directly to regions of India with different staple foods and culinary practices without further region-specific validation. Third, and most importantly, content validity establishes only that expert reviewers consider the instrument’s items relevant and representative; it does not, by itself, establish construct validity, test-retest reliability, or, critically, whether patients themselves can correctly interpret and act on the Wheel during a real counselling encounter. Patient facing comprehension testing, analogous to the ANSI-based recall testing used in pictogram validation [21], together with pilot testing of glycaemic and dietary outcomes, represents the necessary next step before the Wheel can be recommended for large-scale clinical deployment.

Notwithstanding these limitations, the clinical and public-health rationale for a content-validated tool of this kind remains strong. Given the scale of India’s diabetes burden [1,2] and the well-documented benefit of closely text-linked visual aids for patient comprehension and recall, particularly among patients with limited health literacy [7,9], a scientifically vetted, regionally grounded GI wheel offers a scalable, low-cost adjunct to structured diabetes self-management education and support [13]. Its value may be greatest in outpatient settings where dietitian consultation time is constrained and a rapid, visually anchored counselling aid can help standardise the core dietary message delivered by different members of the care team.

Conclusion

The Glycemic Wheel demonstrated good-to-excellent item-level content validity, with all 27 items exceeding the 0.78 acceptability threshold, and good scale-level content validity (S-CVI/Ave 0.897) when evaluated by a large, multidisciplinary panel of 161 physicians and dietitians. These findings support the scientific and structural adequacy of the instrument’s food selection, glycemic classification, portion cueing, visual design, and clinical utility, and justify its advancement to pilot testing. Portion-size communication and visual layout emerged as the two domains most likely to benefit from targeted refinement. Future work should prioritise patient-facing comprehension testing, construct validity assessment, and evaluation of the Wheel’s real-world impact on dietary choices and glycaemic outcomes across diverse Indian regions and care settings.

Clinical Significance

A content-validated Glycemic Wheel gives physicians and dietitians a standardized, scientifically vetted visual tool for communicating glycemic index and portion guidance within the time constraints of a routine outpatient encounter. By pairing colour-coded GI classification with intuitive hand-based portion cues, the tool is designed to reduce reliance on numerical GI tables that many patients find difficult to apply, and to help standardise the core dietary message delivered across different members of a diabetes care team. Once patient-facing validation is complete, the Wheel may serve as a practical adjunct within diabetes self-management education and support programmes, particularly in high patient-volume Indian clinical settings where dedicated dietitian time is limited.

References

  1. Genitsaridi I, Salpea P, Salim A, Sajjadi SF, Tomic D, James S, Thirunavukkarasu S, Issaka A, Chen L, Basit A, Luk AOY, Ma RCW, Mbanya JC, Ramachandran A, Wild SH, Duncan BB, Boyko EJ, Magliano DJ (2026) 11th edition of the IDF Diabetes Atlas: global, regional, and national diabetes prevalence estimates for 2024 and projections for 2050. Lancet Diabetes Endocrinol. [crossref]
  2. Anjana RM, Unnikrishnan R, Deepa M, Pradeepa R, Tandon N, Das AK, Joshi S, Bajaj S, Jabbar PK, Das HK, Kumar A, Dhandhania VK, Bhansali A, Rao PV, Desai A, Kalra S, Gupta A, Lakshmy R, Madhu SV, Elangovan N, Chowdhury S, Venkatesan U, Subashini R, Kaur T, Dhaliwal RS, Mohan V; ICMR-INDIAB Collaborative Study Group (2023) Metabolic non-communicable disease health report of India: the ICMR-INDIAB national cross-sectional study (ICMR-INDIAB-17) Lancet Diabetes Endocrinol. [crossref]
  3. Jenkins DJ, Wolever TM, Taylor RH, Barker H, Fielden H, Baldwin JM, Bowling AC, Newman HC, Jenkins AL, Goff DV (1981) Glycemic index of foods: a physiological basis for carbohydrate exchange. Am J Clin Nutr. [crossref]
  4. Evert AB, Dennison M, Gardner CD, Garvey WT, Lau KHK, MacLeod J, Mitri J, Pereira RF, Rawlings K, Robinson S, Saslow L, Uelmen S, Urbanski PB, Yancy WS Jr (2019) Nutrition therapy for adults with diabetes or prediabetes: a consensus report. Diabetes Care. [crossref]
  5. American Diabetes Association Professional Practice Committee (2024) Summary of revisions: standards of care in diabetes—2024. Diabetes Care. [crossref]
  6. World Health Organization, Food and Agriculture Organization of the United Nations (2003) Diet, Nutrition and the Prevention of Chronic Diseases: Report of a Joint WHO/FAO Expert Consultation. WHO Technical Report Series No. 916. Geneva: World Health Organization. [crossref]
  7. Berkman ND, Sheridan SL, Donahue KE, Halpern DJ, Crotty K (2011) Low health literacy and health outcomes: an updated systematic review. Ann Intern Med. [crossref]
  8. Gibson AA, Hsu MSH, Rangan AM, Seimon RV, Lee CMY, Das A, Finch CH, Sainsbury A (2016) Accuracy of hands v. household measures as portion size estimation aids. J Nutr Sci. [crossref]
  9. Houts PS, Doak CC, Doak LG, Loscalzo MJ (2006) The role of pictures in improving health communication: a review of research on attention, comprehension, recall, and adherence. Patient Educ Couns. [crossref]
  10. Radhika G, Ganesan A, Sathya RM, Sudha V, Mohan V (2009) Dietary carbohydrates, glycemic load and serum high-density lipoprotein cholesterol concentrations among South Indian adults. Eur J Clin Nutr. [crossref]
  11. Mohan V, Radhika G, Sathya RM, Tamil SR, Ganesan A, Sudha V (2009) Dietary carbohydrates, glycaemic load, food groups and newly detected type 2 diabetes among urban Asian Indian population in Chennai, India (Chennai Urban Rural Epidemiology Study 59) Br J Nutr. [crossref]
  12. ICMR-National Institute of Nutrition (2024) Dietary Guidelines for Indians. Hyderabad: ICMR-National Institute of Nutrition
  13. Powers MA, Bardsley J, Cypress M, Duker P, Funnell MM, Fischl AH, Maryniuk MD, Siminerio L, Vivian E (2015) Diabetes self-management education and support in type 2 diabetes: a joint position statement of the American Diabetes Association, the American Association of Diabetes Educators, and the Academy of Nutrition and Dietetics. Diabetes Care. [crossref]
  14. Lynn MR (1986) Determination and quantification of content validity. Nurs Res. [crossref]
  15. Davis LL (1992) Instrument review: getting the most from a panel of experts. Appl Nurs Res.
  16. Polit DF, Beck CT (2006) The content validity index: are you sure you know what’s being reported? Critique and recommendations. Res Nurs Health. [crossref]
  17. Polit DF, Beck CT, Owen SV (2007) Is the CVI an acceptable indicator of content validity? Appraisal and recommendations. Res Nurs Health. [crossref]
  18. Yusoff MSB (2019) ABC of content validation and content validity index calculation. Educ Med J
  19. Grant JS, Davis LL (1997) Selection and use of content experts for instrument development. Res Nurs Health. [crossref]
  20. Liu L, Chen Y, Li X, Du H, Gui R, Jin H, Chen F, Lin Y, Huang Q, Tong Y (2026) Validation of the simplified Chinese Clear Communication Index using diabetes education materials: instrument adaptation and validation study. J Med Internet Res. [crossref]
  21. Doucette D, Vaillancourt R, Berthenet M, Li LS, Pouliot A (2014) Validation of a pictogram-based diabetes education tool in counselling patients with type 2 diabetes. Can Pharm J (Ott) [crossref]
  22. Zechariah S, Waller JL, De Leo G, Stallings J, Gess AJ, Lehman L (2021) Content and face validation of a novel, interactive nutrition specific physical exam competency tool (INSPECT) to evaluate registered dietitians’ competence: a Delphi consensus from the United States. Healthcare (Basel) [crossref]
  23. Almanasreh E, Moles R, Chen TF (2019) Evaluation of methods used for estimating content validity. Res Social Adm Pharm. [crossref]