Improvement method of type 2 diabetes mellitus miniature nutrition evaluation scale

Abstract

The invention discloses an improvement method of a type 2 diabetes mellitus miniature nutrition evaluation scale, and relates to the technical field of diabetes mellitus research. On the basis of exploring risk factors of malnutrition of T2DM patients, the micro nutrition evaluation scale for the type 2 diabetes mellitus, which is suitable for the clinical environment of China and has relatively good prediction performance, is constructed, comprises six entries, has a diagnosis boundary value of 3 points, has good reliability and effectiveness and prediction capability, and can be used for evaluating the type 2 diabetes mellitus in the clinical environment of China. The method is expected to provide reference for medical staff to early identify malnutrition high-risk people and formulate accurate predictive diagnosis and treatment plans.

Description

Technical Field

[0001] The invention relates to the technical field of diabetes research, and in particular to a method for improving a micro-nutritional evaluation scale for type 2 diabetes. Background Art

[0002] The World Health Organization (WHO) defines diabetes mellitus (DM) as a metabolic disease characterized by multiple etiologies, which is caused by defects in insulin secretion and / or insulin resistance, leading to chronic hyperglycemia, accompanied by disorders in carbohydrate, fat and protein metabolism. Among them, 90% of diabetic patients are type 2 diabetes mellitus (T2DM). T2DM is mainly caused by dysfunction of pancreatic β cells, resulting in impaired insulin production and secretion and peripheral insulin resistance.

[0003] The European Society of Clinical Nutrition and Metabolism (ESPEN) defines malnutrition as: due to insufficient intake or absorption of nutrients, resulting in a decrease in body fat and cell mass, which in turn causes a decline in physical and mental function and has an adverse effect on the clinical outcome of the disease. Experts such as Soeters further explained that malnutrition is a subacute or chronic nutritional state in which malnutrition and inflammation interact to jointly lead to changes in body composition and functional decline. Malnutrition has serious adverse consequences. It not only increases morbidity and infection rates, reduces immunity, but may also prolong the patient's hospitalization time, increase medical expenses, and even increase mortality. Malnutrition is prevalent in many diseases, especially in hospitalized patients, elderly patients, and patients with chronic diseases. Domestic and foreign studies have shown that the incidence of malnutrition / risk in patients with T2DM is 4.3%-56.03%. Research reports indicate that diabetes is an important risk factor for malnutrition and adverse clinical outcomes in patients. Although malnutrition is a common complication of diabetes, it is often neglected clinically because of its atypical symptoms or non-critical illness. Malnutrition in diabetic patients can lead to muscle loss, delayed wound healing, bone loss, immune dysfunction, and decreased systemic function. Compared with diabetic patients with good nutritional status, malnourished patients have a 69% increased risk of death and a 58% increased cost of care. Malnutrition is an independent factor affecting the survival of diabetic patients. Therefore, early attention to the nutritional status of T2DM patients and active preventive measures to correct malnutrition risk factors are of great significance to improve their adverse outcomes and quality of life.

[0004] In order to systematically deal with the widespread malnutrition problem in hospitals, patients should undergo a comprehensive nutritional risk assessment at the beginning of their admission. Nutritional risk screening not only helps medical staff accurately determine whether patients are currently at nutritional risk, but also provides an important basis for subsequent nutritional status assessment and nutritional intervention. According to a study by Schindler et al., in a questionnaire survey of 21,000 patients in 325 hospitals in 25 European countries, only 52% of hospitals routinely conducted nutritional screening. A study by Li Juan et al. pointed out that medical staff do not have a deep understanding of the knowledge related to nutritional screening and nutritional assessment, and there are differences in their knowledge levels, making it difficult to accurately assess the nutritional status of patients. Therefore, in order to increase the level of social attention to nutritional issues, the General Office of the State Council promulgated the National Nutrition Plan (2017-2030) in 2017, listing "Clinical Nutrition Action - Strengthening Patient Nutrition Diagnosis and Treatment, Improving Patient Nutritional Status" as a major action to be promoted. The "Guidelines for the Construction and Management of Clinical Nutrition Departments (Trial)" issued by the General Office of the National Health Commission in 2022 clearly stated that medical institutions with conditions should set up independent clinical nutrition departments to provide comprehensive nutrition screening and assessment, nutrition diagnosis, nutrition treatment, and nutrition education services for hospitalized patients. In addition, the European Society for Clinical Nutrition and Metabolism (ESPEN) and the American Society for Parenteral and Enteral Nutrition (ASPEN) have issued corresponding nutrition support guidelines, strongly recommending nutrition screening for hospitalized patients, further evaluation of patients at risk, and providing nutrition support when necessary.

[0005] Since the 1990s, domestic and foreign scholars have successively published many tools or indicators for nutrition assessment and diagnosis. Although traditional nutrition screening and diagnosis methods, such as anthropometric indicators and laboratory indicators, have the advantages of simple operation, their reliability is often difficult to guarantee in practical applications, and it is difficult to accurately assess the nutritional status of patients with T2DM in a short time. Currently, the widely used comprehensive nutrition risk screening scales, such as nutritional risk screening 2002 (NRS2002), malnutrition universal screening tool (MUST), patient-generated subjective global assessment (PG-SGA) and mini nutrition assessment scale (MNA-SF), have certain application value as universal nutrition screening tools, but lack specificity for specific diseases, especially in patients with T2DM, whose application effect has not been fully verified. At the same time, a small number of scholars in China have improved the original scale and developed screening tools suitable for specific populations. For example, Hua Hongxia et al. screened out items of poor quality in PG-SGA based on item response theory and formed a modified version of PG-SGA suitable for gastric cancer patients.

[0006] Given that the applicability of existing nutritional assessment tools in patients with T2DM remains to be discussed, constructing a simple and sensitive nutritional assessment tool is of great significance for reducing the workload of medical staff and providing early nutritional support for patients with T2DM. Summary of the invention

[0007] The purpose of the present invention is to provide an improved method for a micro-nutritional assessment scale for type 2 diabetes, by investigating the current status and risk factors of malnutrition in T2DM patients, providing a reference basis for medical staff to make medical decisions and conduct health education, and compiling a nutritional assessment scale for T2DM patients and conducting reliability and validity tests, providing a nutritional screening reference tool for medical staff to identify the nutritional status of T2DM patients at an early stage, so as to solve the technical problems raised in the background technology.

[0008] To achieve the above object, the present invention provides the following technical solution: a method for improving a mini nutritional assessment scale for type 2 diabetes, comprising at least the following steps:

[0009] S1: Construction of the item pool, through cross-sectional survey and statistical analysis of the independent risk factors of malnutrition in T2DM patients and screening of the items of the MNA-SF scale using item response theory;

[0010] S2: Construct a mini nutritional assessment scale for type 2 diabetes, invite experts from related fields to participate in two rounds of Delphi expert consultation, modify or add or delete scale items according to the item screening criteria and the opinions of experts, and form a clinical test version of the mini nutritional assessment scale for type 2 diabetes;

[0011] S3: Reliability and validity test, the reliability of the scale is evaluated through Cronbach's α coefficient, and the validity of the scale is described through content validity, structural validity and predictive validity.

[0012] Furthermore, the S1 at least includes the following steps:

[0013] Identify independent risk factors for malnutrition in type 2 diabetes;

[0014] Identify item response theory;

[0015] Result judgment includes general characteristics of T2DM patients and the occurrence of malnutrition, univariate analysis of malnutrition in T2DM patients, multivariate regression analysis of malnutrition / risk in T2DM patients and item response theory analysis results.

[0016] Furthermore, the determination of independent risk factors for malnutrition in type 2 diabetes mellitus at least comprises the following steps:

[0017] Conduct a preliminary investigation and determine the preliminary investigation results to determine whether the risk factor questionnaire design is reasonable, and determine the risk factor questionnaire for T2DM patients based on the preliminary investigation results;

[0018] Convenience sampling was used to determine the research subjects. It should be noted that the subjects needed to meet the inclusion criteria and were not excluded;

[0019] See formula The sample size was calculated, where P was the incidence of malnutrition in patients with T2DM, δ was the allowable error, and α was the test level;

[0020] The T2DM Patient Risk Factor Questionnaire, Global Leaders Malnutrition Initiative, NRS2002, MUST, MNA-SF, and PG-SGA were used as survey instruments;

[0021] Conduct quality control before, during and after the survey;

[0022] Data collection: According to the inclusion criteria, patients with T2DM hospitalized in the endocrinology department of the hospital were selected as research subjects. During the investigation, the researchers used face-to-face questioning to ensure that each patient could fully understand and answer the questions. Before the investigation, the purpose and significance of the investigation were explained to the patients in detail. Considering that some patients may have reading or writing difficulties, the researchers filled in the questionnaire on their behalf according to the patients' answers.

[0023] For statistical analysis, Excel was used for double data entry and SPSS25 was used for statistical analysis. Descriptive analysis and statistics showed that the measurement data that conformed to normal distribution were expressed as mean ± standard deviation (x ± s), and the count data were expressed as frequency and composition ratio or percentage. For the subsequent scale formulation, all data were classified and set, and the univariate comparison was performed using χ 2 Test or non-parametric test, multivariate logistic regression analysis was used; P < 0.05 was considered statistically significant.

[0024] Furthermore, the determining of the item response theory comprises at least the following steps:

[0025] The same research subjects were used to determine the independent risk factors of malnutrition for type 2 diabetes;

[0026] The items of the MNA-SF scale were analyzed using the Same Jima graded response model in the item response theory. The marginal maximum likelihood estimation method suitable for large samples was used to calculate the discrimination parameter and difficulty parameter of each item. The discrimination parameter was set as α, and the difficulty parameter was set as β. The items were screened according to the proposed parameter interval.

[0027] The discrimination parameter refers to the parameter that distinguishes different levels of subjects' abilities. The α value reflects the size of the item's discrimination, which is proportional to the amount of information provided by the subject's ability. The smaller the α value, the less information the item can provide for estimating the subject's ability. Therefore, when α<0.50, it indicates that the item needs to be deleted or modified.

[0028] The difficulty parameter is considered to be the ability value when the probability of the subject correctly answering the item is 0.5. The β value reflects the difficulty of the item. The smaller the β value, the simpler the item, and the larger the β value, the more difficult the item. Therefore, when β≦-3 and β≧3, it indicates that the item needs to be deleted or modified.

[0029] The same quality controls were used to identify independent risk factors for malnutrition in type 2 diabetes.

[0030] Furthermore, the S2 at least includes the following steps:

[0031] Generate a pool of scale items. Based on the results of S1, generate a pool of candidate items for the Type 2 Diabetes Mini Nutrition Assessment Scale and conduct expert consultation.

[0032] Delphi expert inquiry, using the Delphi expert consultation method, through the form of questionnaires, anonymous consultation among expert group members, during this process, expert group members put forward opinions, the investigators sorted, summarized, and summarized, and then fed back to the experts anonymously. After multiple consultations, sorting, and feedback, until the experts' opinions converged, the Delphi expert consultation method was used to screen and determine the items in the process of compiling the scale item pool to ensure the scientificity and reliability of the scale;

[0033] The item screening criteria are carried out, and each item is modified, deleted or added according to the importance scores and modification opinions given by experts. The item screening criteria are that the dimensions and items with the importance value mean ≥ 4 and the coefficient of variation < 0.25 are retained, and the dimensions and items that do not meet the conditions are deleted;

[0034] Statistical analysis: The obtained data were entered into Excel by two persons and statistically analyzed using SPSS25.0 software. The basic information of the experts was described by frequency and composition ratio, the enthusiasm of the experts was expressed by the effective recovery rate of the questionnaire, the authority of the experts was expressed by the authority coefficient Cr, and the consistency of the expert opinions was expressed by the coefficient of variation and the coordination coefficient. P < 0.05 was considered statistically significant.

[0035] Finally, a clinical test version of the Mini Nutritional Assessment Scale for Type 2 Diabetes was obtained.

[0036] Furthermore, the Delphi expert inquiry at least includes the following steps:

[0037] Experts were selected for letter consultation. The inclusion criteria for experts included more than 10 years of experience in endocrinology clinical diagnosis and treatment, clinical nursing, clinical nutrition, etc.; bachelor degree or above; associate senior professional title or above and informed consent, and voluntary participation in this study; exclusion criteria: failure to return revision opinions on time;

[0038] The preparation of the inquiry form shall at least include the following contents:

[0039] Introduction: introduce the research background and purpose, explain the purpose, method and significance of this inquiry, extend a sincere invitation to experts, and specify the deadline for returning the inquiry questionnaire;

[0040] Expert consultation and review form for the Mini Nutritional Assessment Scale for Type 2 Diabetes Patients. Experts are requested to judge the importance of each item and put forward modification opinions. According to the Likert 5-level scoring method, the importance is divided into "very important, important, moderately important, unimportant, and very unimportant", with scores of 5, 4, 3, 2, and 1 respectively;

[0041] Expert basic information questionnaire, including at least name, age and working time;

[0042] The experts' familiarity with the content of the inquiry and the basis for their judgment were surveyed. The influence of the basis for judgment was divided into large, medium and small, and the familiarity with the content was very familiar, relatively familiar, generally familiar, unfamiliar and very unfamiliar.

[0043] Furthermore, the S3 at least includes the following steps:

[0044] Identify the research subjects;

[0045] Data collection and quality control: Based on the patient's data, the clinical test version of the Type 2 Diabetes Mini Nutritional Assessment Scale and the GLIM standard were used to score the research subjects. The assessment was conducted in the ward on the day the patient was admitted to the hospital. The questionnaire was distributed by two investigators, who were trained uniformly before the survey. The investigators must understand the purpose and content of this study and be familiar with the precautions for filling out the questionnaire. During the survey, the uniform guidance terms were used and the patient's doubts were answered promptly and accurately.

[0046] Statistical methods: Two people entered data into Excel software, and SPSS25.0 software was used for statistical analysis. Item analysis was used to further screen the scale items, factor analysis was used to determine the weights of each item in the scale, the receiver operating characteristic curve (ROC curve) was used to determine the diagnostic threshold of the scale, and the quartile method was used to determine the risk level of the scale. The reliability of the scale was evaluated by Cronbach's ɑ coefficient and inter-rater coefficient, and the validity of the scale was described by content validity, structural validity and predictive validity. P < 0.05 was considered statistically significant.

[0047] Compared with the prior art, the present invention has the following beneficial effects:

[0048] Based on the exploration of the risk factors of malnutrition in T2DM patients, the present invention constructs a type 2 diabetes micro-nutritional assessment scale suitable for my country's clinical environment and with good predictive performance. The constructed type 2 diabetes micro-nutritional assessment scale contains a total of 6 items with a diagnostic cut-off value of 3 points. The scale has good reliability, validity and predictive ability, in order to provide a reference for medical personnel to identify high-risk groups of malnutrition at an early stage and formulate accurate predictive diagnosis and treatment plans. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for describing the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0050] Figure 1 It is a schematic diagram of the process of the present invention;

[0051] Figure 2 This is a schematic diagram of the nutritional assessment of the present invention;

[0052] Figure 3 This is a schematic diagram of the predictive validity of the present invention. DETAILED DESCRIPTION

[0053] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments.

[0054] See also Figure 1 , a method for improving the micro-nutritional assessment scale for type 2 diabetes, comprising at least the following steps:

[0055] S1: Construction of the item pool, through cross-sectional survey and statistical analysis of the independent risk factors of malnutrition in T2DM patients and screening of the items of the MNA-SF scale using item response theory;

[0056] Objective: To analyze the independent risk factors of malnutrition in T2DM patients through cross-sectional investigation, screen items with item response theory, and jointly construct a mini-nutritional assessment scale for type 2 diabetes.

[0057] The general content is:

[0058] (1) To develop a questionnaire on risk factors for malnutrition in patients with T2DM and to identify independent risk factors for malnutrition in patients with T2DM.

[0059] (2) The item response theory was used to estimate the items of the MNA-SF scale and the items were screened according to the proposed parameter range.

[0060] S2: Construct a mini nutritional assessment scale for type 2 diabetes, invite experts from related fields to participate in two rounds of Delphi expert consultation, modify or add or delete scale items according to the item screening criteria and the opinions of experts, and form a clinical test version of the mini nutritional assessment scale for type 2 diabetes;

[0061] S3: Reliability and validity test, the reliability of the scale is evaluated through Cronbach's α coefficient, and the validity of the scale is described through content validity, structural validity and predictive validity.

[0062] Objective: To develop a mini nutritional assessment scale for type 2 diabetes mellitus based on cross-sectional surveys, item response theory screening items and expert consultation. The scale has certain scientificity and practicality, needs to be tested for reliability and validity in clinical practice, and items should be further screened according to indicators such as the correlation of each item to form the final mini nutritional assessment scale for type 2 diabetes mellitus.

[0063] The S1 at least comprises the following steps:

[0064] Identify independent risk factors for malnutrition in type 2 diabetes;

[0065] Identify item response theory;

[0066] Result judgment includes general characteristics of T2DM patients and the occurrence of malnutrition, univariate analysis of malnutrition in T2DM patients, multivariate regression analysis of malnutrition / risk in T2DM patients and item response theory analysis results.

[0067] Determining the independent risk factors for malnutrition in type 2 diabetes mellitus at least comprises the following steps:

[0068] Conduct a preliminary investigation and determine the preliminary investigation results to determine whether the risk factor questionnaire design is reasonable, and determine the risk factor questionnaire for T2DM patients based on the preliminary investigation results;

[0069] Convenience sampling was used to determine the research subjects. It should be noted that the subjects needed to meet the inclusion criteria and were not excluded;

[0070] Inclusion criteria:

[0071] The diagnostic criteria refer to the 2018 American Diabetes Association Diabetes Diagnosis and Treatment Guidelines;

[0072] Age ≥18 years;

[0073] Informed consent was obtained and participants were voluntarily enrolled in this study.

[0074] Exclusion criteria:

[0075] Patients with cognitive and consciousness disorders;

[0076] Combined with severe liver and kidney failure; tuberculosis; hyperthyroidism;

[0077] Patients who are unable to communicate normally due to aphasia, intellectual disability, hearing impairment, etc.

[0078] Critically ill patients.

[0079] See formula The sample size was calculated, where P was the incidence of malnutrition in patients with T2DM, δ was the allowable error, and α was the test level;

[0080] The T2DM Patient Risk Factor Questionnaire, Global Leaders Malnutrition Initiative, NRS2002, MUST, MNA-SF, and PG-SGA were used as survey instruments;

[0081] Conduct quality control before, during and after the survey;

[0082] (1) Before the survey

[0083] Before the formal survey, team members received training from senior clinical nutrition experts to clarify the purpose and significance of the survey, understand the content of the questionnaire and the precautions for filling it out, and determine the specific standards for collecting the questionnaire. Strictly abide by ethical principles, respect the patient's right to know and autonomy, and ensure that the questionnaire survey is carried out on the basis of the patient's full understanding and consent.

[0084] (2) Under investigation

[0085] The survey used a unified instruction, interviewed the patients face-to-face to collect their general information, and conducted a nutritional assessment within 48 hours of admission. After obtaining the consent of the hospital and the department, the patient's medical records were checked from the electronic medical record system to collect their general information, disease characteristics and laboratory indicators. A master's supervisor in endocrinology clinical nursing and an attending physician in the endocrinology department were responsible for quality control and supervision throughout the process.

[0086] (3) After the investigation

[0087] The data entry is double-checked to ensure the accuracy of data entry and to eliminate unqualified questionnaires.

[0088] Data collection: According to the inclusion criteria, patients with T2DM hospitalized in the endocrinology department of the hospital were selected as research subjects. During the investigation, the researchers used face-to-face questioning to ensure that each patient could fully understand and answer the questions. Before the investigation, the purpose and significance of the investigation were explained to the patients in detail. Considering that some patients may have reading or writing difficulties, the researchers filled in the questionnaire on their behalf according to the patients' answers.

[0089] For statistical analysis, Excel was used for double data entry and SPSS25 was used for statistical analysis. Descriptive analysis and statistics showed that the measurement data that conformed to normal distribution were expressed as mean ± standard deviation (x ± s), and the count data were expressed as frequency and composition ratio or percentage. For the subsequent scale formulation, all data were classified and set, and the univariate comparison was performed using χ 2 Test or non-parametric test, multivariate logistic regression analysis was used; P < 0.05 was considered statistically significant.

[0090] Determining the item response theory comprises at least the following steps:

[0091] The same research subjects were used to determine the independent risk factors of malnutrition for type 2 diabetes;

[0092] The Same Jima graded response model (GRM) in item response theory was used to analyze the items of the MNA-SF scale. The marginal maximum likelihood estimation method (MML) suitable for large samples was used to calculate the discrimination parameter and difficulty parameter of each item. The discrimination parameter was set as α, and the difficulty parameter was set as β. The items were screened according to the proposed parameter interval.

[0093] The discrimination parameter refers to the parameter that distinguishes different levels of subjects' abilities. The α value reflects the size of the item's discrimination, which is proportional to the amount of information provided by the subject's ability. The smaller the α value, the less information the item can provide for estimating the subject's ability. Therefore, when α<0.50, it indicates that the item needs to be deleted or modified.

[0094] The difficulty parameter is considered to be the ability value when the probability of the subject correctly answering the item is 0.5. The β value reflects the difficulty of the item. The smaller the β value, the simpler the item, and the larger the β value, the more difficult the item. Therefore, when β≦-3 and β≧3, it indicates that the item needs to be deleted or modified.

[0095] The same quality control as used in the identification of independent risk factors for malnutrition in type 2 diabetes was used;

[0096] Complete the overall determination of item response theory in accordance with ethical principles.

[0097] This study follows the principle of voluntariness. Patients sign informed consent forms and can stop the investigation at any time without delaying the patient's treatment or examination or violating the patient's privacy. The researchers guarantee that all personal information of the subjects will be kept strictly confidential and used only for this study. Without the permission of the unit and the individual, it will not be disclosed to people unrelated to this study. After the study is completed, all relevant information will be handed over to the hospital for storage. This study was approved by the ethics committee of the hospital where the research members are located, ethics number: 2022 Kelun Review No. (135).

[0098] General characteristics and malnutrition incidence in patients with T2DM

[0099] A total of 382 subjects were included, including 228 males (59.7%) and 154 females (40.3%), aged 19-93 (57.9±15.011) years old; 281 subjects (73.6%) used employee medical insurance as their main payment method, and mainly lived in a family setting (86.4%). Nearly half of the subjects had a family history of diabetes (48.7%); 13.4% of the subjects were diagnosed with malnutrition using GLIM as the diagnostic standard; 22% of the subjects were at nutritional risk using NRS2002 as the evaluation standard; 30.6% of the subjects were at nutritional risk using MUST as the evaluation standard; 33.8% of the subjects were at nutritional risk using MNA-SF as the evaluation standard; and 44.5% of the subjects were at nutritional risk using PG-SGA as the evaluation standard. Figure 2 .

[0100] Univariate analysis of malnutrition in patients with T2DM

[0101] To facilitate the subsequent scale compilation, all data were classified and the value assignment of all variables is shown in Table 1. Comparison of the data of patients with normal nutrition and malnutrition / risk patients showed that the BMI difference was statistically significant in all nutrition assessment tables (P<0.05). When GLIM was used as the evaluation standard, the differences in albumin, prealbumin, C-reactive protein, hemoglobin and white blood cell count were statistically significant (P<0.05); when NRS2002 was used as the standard, the differences in gender, age, self-care ability, living style, hemoglobin and white blood cell count were statistically significant (P<0.05); when MUST was used as the standard, the differences in gender, smoking, duration of diabetes, random blood glucose on admission, albumin and hemoglobin were statistically significant (P<0.05); when MNA-SF was used as the standard, the differences in gender, smoking, duration of diabetes and albumin were statistically significant (P<0.05); when PG-SGA was used as the standard, the differences in age, self-care ability, duration of diabetes, random blood glucose on admission, C-reactive protein and hemoglobin were statistically significant (P<0.05). See Table 2.

[0102] Table 1 Table of Variable assignment

[0103]

[0104] Note: *: Family includes living in a family form with children, spouse, blood relatives, lovers, etc.

[0105] Table 2 Univariate analysis of malnutrition / risk in patients with T2DM

[0106]

[0107]

[0108]

[0109] Item response theory analysis results

[0110] The scores of the MNA-SF scale of the above 382 T2DM patients were analyzed using R language, and the parameters of the three items were obtained, as shown in Table 3. According to the discussion results of this study, the items with α<0.5, β≦-3, and β≧3 were deleted or modified. The discrimination of item 6 was 0.42, and the difficulty was -6.05--2.72. However, considering that item 6 is an assessment of body mass index, it is crucial to the content and results of the nutrition assessment scale, and it was modified after discussion by the project team. Items 3.4.5 are activity assessments, mental problem assessments, and psychological problem assessments. Because their assessment results are basically normal and the scores are basically the same, the software cannot give clear discrimination and difficulty parameters, so they were deleted after discussion by the project team.

[0111] Table 4.Analysis results of items

[0112]

[0113] In summary, the item pool is summarized as follows: the original item 1 (degree of reduced food intake) and item 2 (degree of weight loss) remain unchanged, and item 6 (body mass index) is modified to <18.5 for those under 70 years old; <20 for those over 70 years old. Combined with the statistical results, age, living style, inflammatory status and blood sugar are added as new items. So far, the item pool contains a total of seven items. These items are established to more comprehensively and accurately assess the malnutrition status of diabetic patients.

[0114] The S2 at least comprises the following steps:

[0115] Generate a pool of scale items. Based on the results of S1, generate a pool of candidate items for the Type 2 Diabetes Mini Nutrition Assessment Scale and conduct expert consultation.

[0116] Delphi expert inquiry, using the Delphi expert consultation method, through the form of questionnaires, anonymous consultation among expert group members, during this process, expert group members put forward opinions, the investigators sorted, summarized, and summarized, and then fed back to the experts anonymously. After multiple consultations, sorting, and feedback, until the experts' opinions converged, the Delphi expert consultation method was used to screen and determine the items in the process of compiling the scale item pool to ensure the scientificity and reliability of the scale;

[0117] The item screening criteria are carried out, and each item is modified, deleted or added according to the importance scores and modification opinions given by experts. The item screening criteria are that the dimensions and items with the importance score mean ≧4 and the coefficient of variation (CV) <0.25 are retained, and the dimensions and items that do not meet the conditions are deleted;

[0118] Statistical analysis: The obtained data were entered into Excel by two people and statistically analyzed using SPSS25.0 software. The basic information of the experts was described by frequency and composition ratio, the enthusiasm of the experts was expressed by the effective recovery rate of the questionnaire, the authority of the experts was expressed by the authority coefficient Cr, and the consistency of the expert opinions was expressed by the coefficient of variation (CV) and the coordination coefficient (Kendall's W). P < 0.05 was considered statistically significant.

[0119] Finally, a clinical test version of the Mini Nutritional Assessment Scale for Type 2 Diabetes was obtained.

[0120] The Delphi expert inquiry includes at least the following steps:

[0121] Select an inquiry expert

[0122] Expert inclusion criteria: ① Engaged in clinical diagnosis and treatment, clinical nursing, clinical nutrition, etc. in endocrinology for more than 10 years; ② Bachelor degree or above; ③ Associate senior professional title or above; ④ Informed consent, voluntary participation in this study;

[0123] Exclusion criteria: Failure to return revision comments on time;

[0124] The preparation of the inquiry form shall at least include the following contents:

[0125] Introduction: introduce the research background and purpose, explain the purpose, method and significance of this inquiry, extend a sincere invitation to experts, and specify the deadline for returning the inquiry questionnaire;

[0126] Expert consultation and review form for the Mini Nutritional Assessment Scale for Type 2 Diabetes Patients. Experts are requested to judge the importance of each item and put forward modification opinions. According to the Likert 5-level scoring method, the importance is divided into "very important, important, moderately important, unimportant, and very unimportant", with scores of 5, 4, 3, 2, and 1 respectively;

[0127] Expert basic information questionnaire, including at least name, age and working time;

[0128] The experts' familiarity with the content of the inquiry and the basis for their judgment were surveyed. The influence of the basis for judgment was divided into large, medium and small, and the familiarity with the content was very familiar, relatively familiar, generally familiar, unfamiliar and very unfamiliar.

[0129] In summary, after rigorous screening, evaluation and improvement, the modified T2DM micro-nutritional assessment scale is comprehensive and scientific.

[0130] The S3 at least includes the following steps:

[0131] Identify the research subjects;

[0132] From December to February 2023, patients with type 2 diabetes who were hospitalized in the Endocrinology Department of a Class-A tertiary general hospital in Chongqing were selected as the survey subjects.

[0133] Inclusion and exclusion criteria: See Part I, Proposed Item Pool, Research Subjects 3.1.4 for details.

[0134] According to the requirement that the sample size is 5-10 times the number of items, the number of items in this study is 6. Considering that this part of the study needs to conduct factor analysis, the sample size must be no less than 200 cases, plus 10% sample loss, so the final sample size is 300 cases.

[0135] Data collection and quality control: Based on the patient's data, the clinical test version of the Type 2 Diabetes Mini Nutritional Assessment Scale and the GLIM standard were used to score the research subjects. The assessment was conducted in the ward on the day the patient was admitted to the hospital. The questionnaire was distributed by two investigators, who were trained uniformly before the survey. The investigators must understand the purpose and content of this study and be familiar with the precautions for filling out the questionnaire. During the survey, the uniform guidance terms were used and the patient's doubts were answered promptly and accurately.

[0136] Statistical methods: Two people entered data into Excel software, and SPSS25.0 software was used for statistical analysis. Item analysis was used to further screen the scale items, factor analysis was used to determine the weights of each item in the scale, the receiver operating characteristic curve (ROC curve) was used to determine the diagnostic threshold of the scale, and the quartile method was used to determine the risk level of the scale. The reliability of the scale was evaluated by Cronbach's ɑ coefficient and inter-rater coefficient, and the validity of the scale was described by content validity, structural validity and predictive validity. P < 0.05 was considered statistically significant.

[0137] The results for S3 are as follows:

[0138] Project Analysis

[0139] First, the scale items were screened by the critical odds ratio method, and the scores of the scale evaluation were sorted from high to low. The top 27% of the sample size was the high-scoring group, and the bottom 27% of the sample size was the low-scoring group. Whether there were differences in the items of the high-scoring group and the low-scoring group was compared.

[65] If there is no statistically significant difference between the items, it indicates that the identification ability of the item is poor and it is recommended to delete it. The results show that there are differences between the items. See Table 4 for details.

[0140] Table 1. Statistics of the critical ratios of each item of the Type 2 Diabetes Micro-Nutritional Assessment Scale (n = 300)

[0141]

[0142] Correlation analysis

[0143] The higher the correlation between an item and the total score of the scale, the higher the homogeneity of the item and the scale.

[66] If the correlation coefficient between the item and the total score is <0.4 or P>0.05, it indicates that the homogeneity between the item and the scale is low and it is recommended to delete it. The results show that there are differences between each item and the total score. See Table 5 for details.

[0144] Table 2. Correlation analysis between items and total score

[0145]

[0146] Factor analysis results

[0147] The results showed that the KMO value was 0.64 and Bartlett's sphericity test P < 0.01, indicating that the data were suitable for factor analysis. See Table 6.

[0148] Table 3. KMO and Bartlett's test

[0149]

[0150]

[0151] According to the SPSS output results, factors with eigenvalues ​​> 1 were extracted, and the cumulative contribution rate was 70.723%. See Table 7.

[0152] Table 4. Cumulative contribution rate of variance

[0153]

[0154] Extraction method: principal component analysis.

[0155] Table 5. Component Matrix

[0156] entry 1 2 Decreased food intake 0.795 -0.473 Weight loss 0.727 -0.562 age 0.512 0.779 Body Mass Index 0.466 0.08 Random blood sugar on admission 0.761 -0.101 Self-care assessment 0.605 0.704

[0157] Extraction method: principal component analysis.

[0158] a extracted 2 components

[0159] Table 6.Weight calculation table

[0160]

[0161] According to Table 8 and Table 9, the weight of each item was calculated, and the items were assigned values ​​again. The assignment standard was: 0-0.1 was 3 points, 0.1-0.2 was 4 points, and 0.3-0.4 was 5 points. The final type 2 diabetes micro-nutritional assessment scale was as follows:

[0162] Mini Nutritional Assessment Scale for Type 2 Diabetes

[0163]

[0164]

[0165] Reliability and validity analysis results:

[0166] Validity Analysis

[0167] (1) Content validity (CVI): Content validity refers to the extent to which the items can fully reflect the content of the measurement. The most commonly used indicators are the item-level Content Validity Index (I-CVI) and the scale-level Content Validity Index (S-CVI). It is generally believed that an I-CVI above 0.75 and an S-CVI above 0.90 indicate good content validity.

[67] According to the corresponding calculation formula, the I-CVI of this scale is 0.88-1, and S-CVI / Ave=0.94.

[0168] (2) Structural validity: In the process of developing the scale, this study used factor analysis to determine the weights of each item in the scale. The factor analysis results showed that KMO = 0.64 > 0.5, χ 2 =777.60, P<0.01, and 2 common factors were extracted, with a cumulative contribution rate of 70.723%. It is generally believed that when the cumulative contribution rate of common factors reaches 50%, it can be said that the structural validity of the scale is good. Therefore, this scale has good structural validity.

[0169] (3) Predictive validity: The area under the ROC curve of the scale was 0.78, indicating that the scale had moderate predictive value. The maximum value of the Youden index (0.485) was used as the critical value to divide the scale into high-risk and low-risk groups. The sensitivity and specificity of the scale were 100% and 48.5%, respectively, indicating good predictive validity. Figure 3 .

[0170] Test outcome variable: total score

[0171]

[0172] The total score of the test outcome variable has at least one tie value between the positive actual status group and the negative actual status group. The statistics may be biased.

[0173] a According to the nonparametric assumption

[0174] b Null hypothesis: True area = 0.5

[0175] (4) Determine the risk level of the scale

[0176] With 3 points as the diagnostic threshold, 30 patients diagnosed with malnutrition by GLIM were screened, and the scores of the malnourished patients were ranked. The risk of malnutrition in type 2 diabetes was graded according to the quartile method, with P0-25 as low risk, P25-75 as medium risk, and P75-P100 as high risk. The following conclusions were drawn: See Appendix

[0177] (1) 3-5 points, low risk;

[0178] (2) 6-12 points for medium risk;

[0179] (3) A score greater than 12 indicates high risk.

[0180] Reliability Analysis

[0181] Cronbach's α coefficient is used to evaluate the consistency of the scale. Generally speaking, a Cronbach's α coefficient greater than 0.9 means that the internal consistency of the scale is very high. A Cronbach's α coefficient between 0.7 and 0.9 means that the internal consistency of the scale is good. A Cronbach's α coefficient less than 0.7 indicates that the consistency between the items of the scale is not high. The Cronbach's α coefficient of this scale is 0.711, indicating that the consistency of this scale is good. Considering that the subjects are all patients with type 2 diabetes and their hospitalization period is usually maintained within 7 to 10 days, the nutritional assessment is intended to accurately reflect the recent nutritional status of the subjects. Given that the difference in the results of two nutritional assessments of the same group of subjects before and after in a short period of time may not be significant, this study does not conduct test-retest reliability measurement.

[0182] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A method for improving the micro-nutritional assessment scale for type 2 diabetes, characterized in that: At least the following steps are included: S1: Construction of the item pool, through cross-sectional survey and statistical analysis of the independent risk factors of malnutrition in T2DM patients and screening of the items of the MNA-SF scale using item response theory; S2: Construct a mini nutritional assessment scale for type 2 diabetes, invite experts from related fields to participate in two rounds of Delphi expert consultation, modify or add or delete scale items according to the item screening criteria and the opinions of experts, and form a clinical test version of the mini nutritional assessment scale for type 2 diabetes; S3: Reliability and validity test, the reliability of the scale is evaluated through Cronbach's α coefficient, and the validity of the scale is described through content validity, structural validity and predictive validity.

2. The method for improving the mini nutritional assessment scale for type 2 diabetes according to claim 1, characterized in that: The S1 at least comprises the following steps: Identify independent risk factors for malnutrition in type 2 diabetes; Identify item response theory; Result judgment includes general characteristics of T2DM patients and the occurrence of malnutrition, univariate analysis of malnutrition in T2DM patients, multivariate regression analysis of malnutrition / risk in T2DM patients and item response theory analysis results.

3. The method for improving the mini nutritional assessment scale for type 2 diabetes according to claim 2, characterized in that: Determining the independent risk factors for malnutrition in type 2 diabetes mellitus at least comprises the following steps: Conduct a preliminary investigation and determine the preliminary investigation results to determine whether the risk factor questionnaire design is reasonable, and determine the risk factor questionnaire for T2DM patients based on the preliminary investigation results; Convenience sampling was used to determine the research subjects. It should be noted that the subjects needed to meet the inclusion criteria and were not excluded; See formula The sample size was calculated, where P was the incidence of malnutrition in patients with T2DM, δ was the allowable error, and α was the test level; The T2DM Patient Risk Factor Questionnaire, Global Leaders Malnutrition Initiative, NRS2002, MUST, MNA-SF, and PG-SGA were used as survey instruments; Conduct quality control before, during and after the survey; Data collection: According to the inclusion criteria, patients with T2DM hospitalized in the endocrinology department of the hospital were selected as research subjects. During the investigation, the researchers used face-to-face questioning to ensure that each patient could fully understand and answer the questions. Before the investigation, the purpose and significance of the investigation were explained to the patients in detail. Considering that some patients may have reading or writing difficulties, the researchers filled in the questionnaire on their behalf according to the patients' answers. For statistical analysis, Excel was used for double data entry and SPSS25 was used for statistical analysis. Descriptive analysis and statistics showed that the measurement data that conformed to normal distribution were expressed as mean ± standard deviation (x ± s), and the count data were expressed as frequency and composition ratio or percentage. For the subsequent scale formulation, all data were classified and set, and the univariate comparison was performed using χ 2 Test or non-parametric test, multivariate logistic regression analysis was used; P < 0.05 was considered statistically significant.

4. The method for improving the mini nutritional assessment scale for type 2 diabetes according to claim 3, characterized in that: Determining the item response theory comprises at least the following steps: The same research subjects were used to determine the independent risk factors of malnutrition for type 2 diabetes; The items of the MNA-SF scale were analyzed using the Same Jima graded response model in the item response theory. The marginal maximum likelihood estimation method suitable for large samples was used to calculate the discrimination parameter and difficulty parameter of each item. The discrimination parameter was set as α, and the difficulty parameter was set as β. The items were screened according to the proposed parameter interval. The discrimination parameter refers to the parameter that distinguishes different levels of subjects' abilities. The α value reflects the size of the item's discrimination, which is proportional to the amount of information provided by the subject's ability. The smaller the α value, the less information the item can provide for estimating the subject's ability. Therefore, when α<0.50, it indicates that the item needs to be deleted or modified. The difficulty parameter is considered to be the ability value when the probability of the subject correctly answering the item is 0.

5. The β value reflects the difficulty of the item. The smaller the β value, the simpler the item, and the larger the β value, the more difficult the item. Therefore, when β≦-3 and β≧3, it indicates that the item needs to be deleted or modified. The same quality controls were used to identify independent risk factors for malnutrition in type 2 diabetes.

5. The method for improving the mini nutritional assessment scale for type 2 diabetes according to claim 1, characterized in that: The S2 at least comprises the following steps: Generate a pool of scale items. Based on the results of S1, generate a pool of candidate items for the Type 2 Diabetes Mini Nutrition Assessment Scale and conduct expert consultation. Delphi expert inquiry, using the Delphi expert consultation method, through the form of questionnaires, anonymous consultation among expert group members, during this process, expert group members put forward opinions, the investigators sorted, summarized, and summarized, and then fed back to the experts anonymously. After multiple consultations, sorting, and feedback, until the experts' opinions converged, the Delphi expert consultation method was used to screen and determine the items in the process of compiling the scale item pool to ensure the scientificity and reliability of the scale; The item screening criteria are carried out, and each item is modified, deleted or added according to the importance scores and modification opinions given by experts. The item screening criteria are that the dimensions and items with the importance value mean ≥ 4 and the coefficient of variation < 0.25 are retained, and the dimensions and items that do not meet the conditions are deleted; Statistical analysis: The obtained data were entered into Excel by two persons and statistically analyzed using SPSS25.0 software. The basic information of the experts was described by frequency and composition ratio, the enthusiasm of the experts was expressed by the effective recovery rate of the questionnaire, the authority of the experts was expressed by the authority coefficient Cr, and the consistency of the expert opinions was expressed by the coefficient of variation and the coordination coefficient. P < 0.05 was considered statistically significant. Finally, a clinical test version of the Mini Nutritional Assessment Scale for Type 2 Diabetes was obtained.

6. The method for improving the mini nutritional assessment scale for type 2 diabetes according to claim 1, characterized in that: The Delphi expert inquiry includes at least the following steps: Experts were selected for letter consultation. The inclusion criteria for experts included more than 10 years of experience in endocrinology clinical diagnosis and treatment, clinical nursing, clinical nutrition, etc.; bachelor degree or above; associate senior professional title or above and informed consent, and voluntary participation in this study; exclusion criteria: failure to return revision opinions on time; The preparation of the inquiry form shall at least include the following contents: Introduction: introduce the research background and purpose, explain the purpose, method and significance of this inquiry, extend a sincere invitation to experts, and specify the deadline for returning the inquiry questionnaire; Expert consultation and review form for the Mini Nutrition Assessment Scale for Type 2 Diabetes Patients. Experts are requested to judge the importance of each item and put forward modification opinions. According to the Likert 5-level scoring method, the importance is divided into "very important, important, moderately important, unimportant, and very unimportant", with scores of 5, 4, 3, 2, and 1 respectively; Expert basic information questionnaire, including at least name, age and working time; The experts' familiarity with the content of the inquiry and the basis for their judgment were surveyed. The influence of the basis for judgment was divided into large, medium and small, and the familiarity with the content was very familiar, relatively familiar, generally familiar, unfamiliar and very unfamiliar.

7. The method for improving the mini nutritional assessment scale for type 2 diabetes according to claim 1, characterized in that: The S3 at least includes the following steps: Identify the research subjects; Data collection and quality control: Based on the patient's data, the clinical test version of the Type 2 Diabetes Mini Nutritional Assessment Scale and the GLIM standard were used to score the research subjects. The assessment was conducted in the ward on the day the patient was admitted to the hospital. The questionnaire was distributed by two investigators, who were trained uniformly before the survey. The investigators must understand the purpose and content of this study and be familiar with the precautions for filling out the questionnaire. During the survey, the uniform guidance terms were used and the patient's doubts were answered promptly and accurately. Statistical methods: Two people entered data into Excel software, and SPSS25.0 software was used for statistical analysis. Item analysis was used to further screen the scale items, factor analysis was used to determine the weights of each item in the scale, the receiver operating characteristic curve (ROC curve) was used to determine the diagnostic threshold of the scale, and the quartile method was used to determine the risk level of the scale. The reliability of the scale was evaluated by Cronbach's ɑ coefficient and inter-rater coefficient, and the validity of the scale was described by content validity, structural validity and predictive validity. P < 0.05 was considered statistically significant.

Images