Development and psychometric evaluation of self‐management scale for pregnant woman with gestational diabetes mellitus in China

Abstract Aim To develop a self‐management scale and evaluate its validity for pregnant woman with GDM in China. Design A cross‐sectional survey design. Methods This study was conducted through three phases. The item pools of the scale were developed through literature review and expert interview. Content validity was assessed by an expert panel. Structure validity was evaluated through exploratory factor analysis. In the end, internal consistency reliability was tested. Results The self‐management scale includes four dimensions, including self‐management consciousness, pregnancy management, blood glucose management and resource utilization, with a total of 35 items. In the scale, the Cronbach's α was 0.95. The split‐half reliability of the overall scale is 0.79. And the test‐retest reliability was 0.91. The content validity was 0.94. Conclusions The scale is significantly valid and reliable, and it can be used to evaluate the self‐management ability of pregnant woman with GDM in China.


| INTRODUC TI ON
Diabetes mellitus, a non-communicable chronic disease, has become a major challenging health problem all over the world (Kamradt et al., 2014;Veeraswamy et al., 2012). It was reported by the International Diabetes Federation (IDF Diabetes Atlas) that 463 million adults suffered from diabetes in 2019 and 9.3% of adults aged 20-79 years are living with diabetes. IDF estimates that there will be 578 million adults with diabetes by 2030, and 700 million by 2045. Diabetes is one of the fastest growing health challenges of the 21st century, with the number of adults living with diabetes having more than tripled over the past 20 years (2019). Particularly, gestational diabetes mellitus (GDM) is one of the most common complications of pregnancy. Gestational Diabetes Mellitus refers to any degree of glucose intolerance with onset or first recognition during pregnancy. GDM currently affects approximately 2%-10% of pregnant women in the United States, 2%-6% in Europe, 5%-8% in Australia and 10%-15% in China (Carolan-Olah, 2016). Chinese women, compared to women of many other ethnic backgrounds, are at higher risk of developing GDM. Women with GDM are at high risk of serious health outcomes such as hypoglycaemia and respiratory difficulties, macrosomia or high infant weight, premature delivery, birth damage, labour dystocia, even hypertension and heart disease in later than mother without GDM though GDM generally resolves once the baby is born. The offspring of mothers with GDM are predisposed to childhood obesity, early onset of type 2 diabetes and cardiovascular disease in adult life (Mensah et al., 2020;Veeraswamy et al., 2012). Therefore, it is urgent to control blood glucose level and protect the health of the mother and the foetus once GDM is diagnosed.

| BACKG ROU N D
Self-management scale is an effective tool for pregnant women with GDM to prevent these complications and achieve optimal blood glucose level during pregnancy. Generally, self-management scale includes self-monitoring blood glucose, dietary modification and increasing physical exercise (Mensah et al., 2020). Selfmanagement is the individual's ability to manage the symptoms, treatment, physical and psychosocial consequences and life style changes inherent in living with a chronic condition (Barlow et al., 2002). Self-management skill is necessary to enable pregnant women managing their own GDM. Therefore, a valid and reliable tool, which assesses self-management behaviour in women with GDM, is needed. The Summary of Diabetes Self-Care Activities measure (SDSCA) is one of the most popular and frequently used tools in English-speaking regions. The questionnaire is an 11-item self-reporting tool assessing levels of self-care in adults with diabetes (Toobert et al., 2000). The Diabetes Management Self-Efficacy Scale (DMSES) is another widely used scale and also some countries such as Australia, United Kingdom and China had accepted the use of the scale as a best practiced model (Sturt et al., 2010).
In addition, educational and intervention programs, such as a Webbased intervention, are also useful to improve knowledge of GDM and GDM self-management principles (Carolan-Olah et al., 2015;Liu et al., 2021;Wan et al., 2019).
Generally, the self-management scale is used for patients with general chronic diabetes, not for pregnant woman with gestational diabetes mellitus (Gharaibeh et al., 2017;Schmitt et al., 2013;Sousa et al., 2010). Rossella Messina developed an Italian version of the diabetes management self-efficacy scale for type 2 diabetes (Messina et al., 2018). Martina Kamradt constructed a German version of the Summary of Diabetes Self-Care Activities measure and assessed self-management in patients with diabetes mellitus type 2 in Germany (Kamradt et al., 2014). Wah et al. (2018)

| Design
This cross-sectional survey study was conducted through three phrases to develop and evaluate psychometrical properties of selfmanagement scale for pregnant women with GDM. In Phase 1, the item pools of the scale were developed through literature review and expert interview with 9 experts from our hospital. In Phase 2, a Delphi survey was conducted to evaluate the authority and coordination among 28 experts from 20 hospitals in different cities of China. In Phase 3, the validity and reliability of the scale were tested among 190 participants of the pregnant women with GDM in our hospital. an initial self-management scale for pregnant women with GDM was determined, mainly including self-management, pregnancy management, blood glucose management, family and social support, with a total of 38 items. Expert consultation form, that is questionnaire, was composed by three parts, survey introduction, each item scoring table and experts' personal information. The introduction mainly describes the research purpose and background. The importance of items was valued using the Likert five-point scale. An item with 4-5 point means the expert agree with the item. The inclusion criteria for items: 80% experts agree with the item (importance score ≥ 4 points); average score for item importance > 3.5; variation coefficient < .20. The items can be added, removed or modified based on the expert' opinions. The experts' personal data include age, gender, position and title.

| Expert consultation based on Delphi survey
The Delphi survey was conducted by a two-round questionnaire enquiry. The questionnaire for the first round was formed based the literature search. After the first round of enquiry by email or social software of wechat, the research group added or deleted some items of the scale based on the opinions of the experts. After the second-round survey, the final self-management scale was formed with a total of 35 items and was used to test the reliability and validity by the pregnant women with GDM.

| Participants
A total of 190 participants were included in the study. Participants were recruited from pregnant women with GDM who seek for antenatal clinic service or were hospitalized in obstetrics department between June and December 2020 in the affiliated hospital of Qingdao University. The participants were explained the purpose of the study and were provided the inclusion and exclusion criteria. Inclusion criteria for pregnant women: (1) meet GDM diagnostic criteria, Diagnosis Guidelines for gestational diabetes mellitus, issued in 2004. That is, oral glucose tolerance test was performed for pregnant women after taking 75 g glucose. The blood glucose level was monitored at the moment of empty stomach, 1 hr after taking glucose, or 2 hr after taking glucose. The pregnant women can be diagnosed as GMD if the blood glucose level reaches or exceeds 5.1, 10.0, and 8.5 mmol/L, respectively. It is important to point out that if one of the three items reaches the standard, GMD can be diagnosed; (2) participants with the normal communication and understanding ability and can understand the research and communicate with the research group members freely; (3) be informed the purpose of the research and volunteer to participate in this study. Exclusion criteria: (1) pregnant women with chronic diabetes mellitus complication; (2) with serious diseases, such as severe heart failure and high blood pressure; (3) consciousness disorder, and unable to complete the investigation independently.

| Data analysis
The data were processed and analyzed by IBM SPSS Statistics for Windows, version 19 (IBM Corp., Armonk, N.Y., USA). Descriptive analysis is expressed by mean, standard deviation and coefficient of variation. The coefficient of variation is used to represent the dispersion degree of expert opinions. The questionnaire recovery rate is used to show the enthusiasm of experts. The degree of expert authority is represented by the expert authority coefficient. The degree of coordination of expert opinions is reflected by the coefficient of variation and the coefficient of coordination. The smaller value of coefficient of variation means the better coordination of expert's opinions. All the items were scored using a 5-point Likerttype scoring method. A standard score was used for the items in the scale. A standard sore is equal to raw score/ theoretical maximum score (Schmitt et al., 2013).

| Content validity
The content validity of the items was evaluated by 7 experts.
Reliability is a key facet of measurement quality, and split-half reliability is a statistical method used to measure the consistency of the scores of a test, which is a convenient alternative to other forms of reliability, including test-retest reliability. Test-retest reliability F I G U R E 1 Prisma flow diagram for searching literature and identifying initial items of self-management scale measures repeatability, that is, the degree to which test results are consistent overtime; (Mccrae et al., 2011;Vas et al., 2013). The content validity index (CVI) refers to the ratio of items graded as very or quite relevant by all of the raters involved. The acceptable CVI of items was equal or more than 0.8 (Denise, 2012

| Structural validity
The structure validity of the scale can be evaluated through exploratory factor analysis (EFA). The Cronbach's α coefficient was used to measure the internal consistency reliability (Conway et al., 2021).

| Participants' demographic characteristics
In this study, a total of 190 questionnaires were distributed and 178 were recovered with effective recovery of 93.68%. The age of the participant was 20-38 (30.50 ± 4.50) years; One hundred and fifty-five (87.08%) women have an education of college or above.
One hundred and thirty-three (74.72%) women are for first birth.
One hundred and forty-eight (83.15%) women are urban residents.
One hundred and twenty (67.42%) women use insulin. The detailed demographic characteristics of the participants are presented in Table 1.

| Authority and coordination of experts in Delphi survey
Two rounds of expert questionnaire survey were completed. In the first round, twenty-eight questionnaires were distributed and 26 valid questionnaires were collected with an effective recovery rate of 92.90%. In the second round, twenty-six copies of questionnaire were distributed, and 25 copies were collected with an effective recovery rate of 96.20%. The authority coefficient of two rounds is 0.88 and 0.91, respectively, which indicates that the experts involved in this study have high authority. In the two-round survey, the Kendall's W harmony index was 0.12 and 0.11, respectively. It indicates that all experts have a high opinion consistency on the items.
Based on the experts' opinion, 3 items were deleted and 4 items were modified. The final self-management scale includes 4 dimensions and 35 items.

| Content validity
The content validity index (CVI) of each item is higher than 0.80, the content validity of each dimension is in range of 0.88-0.96, and the scale content validity is 0.94. All the content validity (including item, dimension and scale) met the criteria, indicating the scale with satisfied content validity (Polit et al., 2007).

| Structural validity
The χ2 value of Bartlett's spherical test was 4 352.36 (p < .001) and the Kaiser-Meyer-Olkin (KMO) value was 0.91, indicating that factor analysis was suitable for this study. Principal component analysis (PCA) and variance maximizing orthogonal rotation were used in the two-round survey. The first round of exploratory factor analysis (EFA) showed that there were 6 common factors whose eigenvalues > 1 and cumulative variance contribution rate was 61.45%. It is found that the slope gradually getting smaller in the scree graph after the fourth common factor. Therefore, four common factors were extracted for further analysis. In the second round of exploratory factor analysis, the cumulative variance contribution rate of the four common factors was 57.68%, and the factor loadings of each item on the corresponding factor was >0.40. The factor loadings TA B L E 1 Demographic characteristics of participants (n = 178)

| Reliability test of self-management scale
The  Table 3).

| Scoring method for items in selfmanagement scale
This research used various methods, such as qualitative interview, expert consultation, and reliability and validity test. The formal GDM self-management scale for pregnant women included four dimensions, self-management awareness, pregnancy management, blood glucose management, and resource utilization, with a total of 35 items. Likert 5-level scoring method was used for all items. The items were graded by a standard score, with standard score = (actual score/highest possible score) × 100. It was defined that 60 was poor self-management ability, 60 ~ 80 was moderate self-management ability, 80 or higher was good self-management ability.  (Conway et al., 2021;Polit et al., 2007). In this study, the content validity of each item ranged from 0.75 to 1.00, and the content validity of the total scale was 0.94. The results indicated that the content validity of the scale were high. Because no high-quality selfmanagement scale for pregnant women with GDM has been found, criterion-related validity has not been conducted in this study. In exploratory factor analysis, in general, if the common factors extracted from the scale can explain more than 50% of the variation, and each item loaded on the corresponding factor > 0.40, it is considered that the scale has good structural validity. In this study, 4 common factors were extracted, and the cumulative variance contribution rate was 57.68%. The loads of each item on the corresponding factors  It is useful to provide basis and evidence for follow-up intervention research and clinical health education.

| DISCUSS ION
It is very interesting and relevant due to the importance of the topic and the implications for the health of pregnant women.
Therefore, it is necessary to discuss the relevance or contributions for nursing or health professionals by the fact that women perform self-management actions. The self-management scale can provide the pregnant women with GDM more effective diabetes education to ensure clear understanding of self-management principles. It can assist pregnant women with GDM to better self-manage their condition and to plan appropriate interventions that can be effective in improving glycaemic control and delaying or preventing diabetesrelated complications for both mother and child, which in turn would decrease the costs of managing the disease. It also provides guidance for nurses-midwives on maternal and postpartum follow-up care for women at risk or diagnosed with gestational diabetes mellitus in clinical practice. However, due to region and time limitations, the sample size and representativeness of this study are limited. Subsequent studies should further increase the sample size, improve the representativeness of samples, and validate and optimize the scale using confirmatory factor analysis.

ACK N OWLED G EM ENTS
The authors thank all the experts who participated in this investigation.

CO N FLI C T O F I NTE R E S T
No conflict of interest has been declared by the authors.

E TH I C A L CO N S I D ER ATI O N
The study protocol was approved by the Ethics Committee of the affiliated hospital of Qingdao University. The institutional review board has approved the study and waived the need for individual informed consent by formulating a declaration of no objection.

DATA AVA I L A B I L I T Y S TAT E M E N T
The authors confirm that the data supporting the findings of this study are available within the article.