Who knows how to use Liu Fengbin’s Traditional Chinese Medicine Health Scale?

1.1 Principles for the development of the scale

1.1.1 Purpose and conception ①The purpose is to follow the definition and connotation of international patient-reported outcome indicators, with the Chinese culture as the background , using the theory of traditional Chinese medicine as a guide to develop a traditional Chinese medicine health status scale that has Chinese cultural characteristics and reflects the health connotation of traditional Chinese medicine. ②The theoretical structure of the scale directly reflects the health connotation from eight aspects: energy, pain, stool, urination, sleep, emotion, diet, and physical fitness [4]. Every aspect should conform to the theory of traditional Chinese medicine without betraying the meaning of PRO. ③The scale should be concise and to the point, with strong practicality, scientificity and operability. As a universal scale, it is suitable for the general population and various chronic diseases. ④The scale is self-assessment, and each item has 4-level response options.

1.1.2 Requirements for items ① Develop items according to aspects. Each item only clearly reflects one issue. The language of the item description is concise and clear, avoiding ambiguity, and is clearly graded and easy to distinguish. ② The items should be easy to understand and easy to operate, and each respondent should be able to evaluate the items.

1.2 Establishment of the research team

Referring to the international scale development model and design method [5,6], a research team was established, consisting of experts in traditional Chinese medicine, statistics and quality of life** *Same composition, ***same organization and leadership for future scale development and assessment work.

1.3 Formation of the item pool

1.3.1 Establishment of the theoretical structure model The research team based on the conceptual connotation of PRO, combined with Chinese cultural characteristics and traditional Chinese medicine theory on health understanding and related research experience [7], based on a survey of experts, it was proposed that the TCM health status scale should include eight aspects: energy, pain, stool, urination, sleep, emotion, diet, and physical fitness.

1.3.2 Generation of item pool Items are mainly obtained from the following three aspects: ① Putting forward items based on the patient. Thirty patients with different types of diseases (chronic diseases in internal medicine) and 30 healthy people were selected. They proposed what aspects and items health should include based on their own experience and understanding of health. Then, members of the research team based on traditional Chinese medicine theory and clinical practice. Actually organize, modify, synthesize and split the proposed items to form items. ② Modified items based on SF-36. SF-36 is an internationally renowned health status scale. We have provided the SF-36 items and some background materials to relevant TCM experts, and asked TCM experts to propose items that the TCM health scale should have. What is missing from the SF-36 scale from a traditional Chinese medicine perspective? Ask TCM experts to add items that TCM considers important but not reflected in this scale. After collecting the expert opinion form, we discussed the opinions of each expert one by one. It was found that when the experts discussed health aspects and items, they gave items in full accordance with the aspects that the traditional Chinese medicine health theory should include. They believed that SF-36 could not reflect the health connotation of traditional Chinese medicine, thus discarding the previous items. SF-36 based modification entry. ③Propose items based on theoretical models. The research team made extensive reference to the connotation of health conditions in traditional Chinese medicine, carefully read relevant theoretical and clinical professional books on traditional Chinese medicine, proposed items based on the concept and composition of PRO, and carefully discussed, modified, deleted, and added items one by one. After sorting out the items from the above three aspects, we sent them to relevant TCM experts (of different ages and professional title structures) in Beijing, Yinchuan, Chengdu, Guangzhou, Zhengzhou and other places to collect opinions. 76 TCM experts rated the importance of the items and made suggestions. Comments for revision. Perform statistical analysis on the importance scoring results fed back by traditional Chinese medicine experts, calculate statistical indicators such as mean, standard deviation, median and interquartile range of score values, and classify items with lower mean or median scores into After elimination and modification, it became the first draft of the scale, with a total of 40 items. Combining the theory of traditional Chinese medicine and clinical reality, after expert discussion one by one, the answer options for each item were divided into 4 levels according to the different degrees of clinical manifestations. For example, items about energy are graded into four levels: full of energy, poor energy, tired, and extremely poor; items about dizziness are classified into no dizziness, slight dizziness, able to continue daily work and activities, dizziness, unable to walk, and dizziness , cannot stand 4 levels.

1.4 Determination of the survey scale

1.4.1 Small-scale testing (language debugging) and item reconstruction were conducted to select 50 healthy people and patients with moderate education levels ( Chronic diseases in internal medicine), the first draft of the scale was used for small-scale testing. After healthy people and patients filled in the scale, they were asked in detail about the filled-in content. The main assessment was whether healthy people or patients could answer the items? Do you understand the entry? How to understand the entry? Is its understanding consistent with what we designed? etc. Items that were difficult to understand or that differed widely among patients were deleted or modified.

1.4.2 Preliminary survey scale. After the language test of the first draft of the scale was completed, the items were further analyzed, compared, discussed, and modified, and compiled into a pre-survey scale. 35 items were retained as preliminary Survey scales include: energy, vision, reaction, physical strength, memory, fatigue, shortness of breath, dizziness, pain level, pain frequency, appetite, food intake, dry mouth, bitter mouth, light mouth, dry stool, frequency of stool, Loose stools, rotten stools, constipation, farting, yellow urine, frequent nocturia, difficulty urinating, difficulty falling asleep, restless sleep, dreaminess, prone to colds, fever in hands, feet, and soles, intolerance to cold, weight loss, upset, irritable, Restless and depressed.

1.5 Clinical Investigation

1.5.1 Implementation of Clinical Investigation Between 2004 and 2006, a preliminary survey scale was used to randomly select 300 people (100 people) in Guangdong Province and Ningxia Autonomous Region. The survey was carried out on healthy people, 100 outpatients and 100 inpatients. The diseases were mainly chronic diseases; the form of self-measurement scale was used; the cultural level required the survey subjects to be able to complete the scale independently; investigators were recruited , requiring a serious and honest character; drafting an investigator guide and implementation manual, and conducting investigator training; the investigator appears as a doctor, interviews the subjects one by one, gives a brief explanation, and asks them to fill in the scale one by one. WHOQOL-100, SF-36, and CH-QOL were distributed at the same time, and 273 questionnaires were finally collected, with a recovery rate of 91%.

1.5.2 General situation of the respondents Among the 273 survey respondents, the age range is 18 to 65 years old (average 39.6 years old, S = 14.1), divided into 18 to 39 years old and 40 years old. ~ Two groups of 65 years old. There are 101 cases in Ningxia and 172 cases in Guangdong; 137 male cases and 136 female cases; 80 healthy people, 91 outpatients and 102 inpatients; 123 people with junior high school education and 150 people with technical secondary education or above; 43 people with rural household registration There were 66 cases in counties and towns and 169 cases in cities. There were 76 completely healthy patients, 56 patients who were sick and stable, and 139 patients who were sick and under treatment.

1.6 Statistical methods

Statistical analysis was completed with the help of SPSS 11.0, SAS 8.1 and EQS 6.1 (structural equation analysis) software packages. Items were screened using statistical analysis methods such as discrete trend method, variance analysis, confirmatory factor analysis, correlation coefficient, and Cronbach's coefficient method. The scale was assessed using confirmatory factor analysis, Cronbach's coefficient method and analysis of variance.

2 Results

2.1 Item Analysis

According to a set of procedures and methods for scale development, the items of the pre-survey scale were screened. Including item difficulty, response analysis, etc., the results show that patients can basically understand and complete the scale alone. It has a relatively appropriate level of difficulty and good response, and there is no ceiling effect (all given the highest score) or floor effect (all given the lowest score).

2.2 Filtering entry results

Use different statistical analysis methods to further screen entries. ① Discrete trend method (coefficient of variation method). This method selects indicators from the perspective of sensitivity. In the measurement of quality of life, the standard deviation can be directly used to reflect the discrete trend. However, if the score values ??of each item are not normally distributed, the variables should be transformed first to make them into a normal distribution. Calculate the standard deviation of each item (the dimensions of each item are the same, and the standard deviation is directly used to compare the degree of variation), and select the three indicators with the largest standard deviation. As a result, 22 items are selected. ②Variance analysis. This method also selects indicators from a sensitivity perspective.

Based on P < 0.05, 27 items were finally selected. ③ Confirmatory factor analysis. Screen the items from the structural perspective of the scale. First calculate the CFI of the aspect, and then compare it with the CFI after removing any item in the aspect. If the CFI increases significantly after removing an item, it means that the existence of the item has reduced the CFI. If there is a negative impact on the aspect, it should be removed, otherwise it should be retained. Finally, 22 entries are retained. ④The correlation coefficient between the aspect total score and the items. By calculating the total score of the aspect, the correlation coefficient between each item and the total score is calculated, with 0.5 as the boundary. If the correlation coefficient is higher than 0.5, it indicates that the contribution of the item is relatively large, otherwise it will be deleted. Finally, 23 entries were retained. ⑤Cronbach’s coefficient method. Items were screened from the perspective of internal consistency. First, calculate the overall Cronbach's α coefficient of a certain aspect, and then compare it with the Cronbach's α coefficient after removing any one of the items. If there is a large increase in the Cronbach's α coefficient after removing an item, it means that the item's If it has the effect of reducing the internal consistency of this aspect, it should be removed, otherwise it should be retained. Finally, 22 items were retained. From the statistical analysis results of the above item screening, it can be seen that the items screened out by 5 different methods are not the same. The items screened out by more than 4 methods include 2, 3, 21, 24, 31 and other 5 items.

2.3 Assessment of the scale

The results of the confirmatory factor analysis of the retained 30-item scale structure showed that CFI=0.919, proving that it has good Construct validity. The Cronbach coefficients of the eight aspects of energy, pain, diet, stool, urination, sleep, emotion, and physical fitness were 0.8102, 0.8298, 0.7885, 0.6331, 0.5253, 0.8161, and 0. 8701, 0.5638. It can be seen that except for the slightly worse Cronbach's coefficient in bowel, urine and physical fitness, the other six aspects all show good reliability. Analysis of variance in eight aspects between healthy people and outpatients and inpatients showed that P values ??were all <0.01, indicating that the scale has the ability to distinguish between healthy people and outpatients and inpatients.

2.4 Formal survey scale

Based on the results of statistical analysis and scale assessment, the structural model of the scale was finally established as: Aspects reflect health status, that is, the structure of this scale Thirty items are reserved for TCM health conditions that are directly reflected in eight aspects: energy, pain, stool, urination, sleep, emotion, diet, and physical fitness. Among them, there are 6 items for energy, 2 items for pain, 5 items for diet, 5 items for stool, 2 items for urine, 3 items for sleep, 3 items for physical constitution, and 4 items for emotion, forming the official version of the Traditional Chinese Medicine Health Status Scale