ABSTRACT
Aims
To identify different longitudinal trajectories of hypoglycaemia problem‐solving ability in patients with diabetes mellitus (DM) and explore their predictive factors. To examine the impact of these heterogeneous trajectories on quality of life.
Design
This study adopted a prospective longitudinal design.
Methods
A total of 272 patients who completed follow‐up were longitudinally assessed for their hypoglycaemia problem‐solving abilities over 6 months. Latent class growth modelling (LCGM) was used to identify heterogeneous trajectories of hypoglycaemia problem‐solving ability. Multiple logistic regression was conducted to determine predictors, while univariate ANOVA and multiple linear regression analysis were applied to explore the effects of heterogeneous trajectories on quality of life.
Results
The overall level of hypoglycaemia problem‐solving ability in DM patients increased from hospitalisation to 1 month after discharge and gradually decreased from 3 to 6 months after discharge. LCGM identified three heterogeneous trajectories of hypoglycaemia problem‐solving ability. Results of multinomial logistic regression analysis showed that employment status, monthly income, frequency of blood glucose monitoring, fear of hypoglycaemia, and social support were predictors of heterogeneous trajectories of hypoglycaemia problem‐solving ability in DM patients. In addition, hypoglycaemia problem‐solving ability positively predicts quality of life.
Conclusions
Our findings establish a critical theoretical foundation for designing and implementing effective interventions tailored to patients' distinct trajectories in diabetes management.
Implications for the Profession and/or Patient Care
This study explores the trajectories and predictors of hypoglycaemia problem‐solving abilities in DM patients, providing a theoretical basis for nurses to guide patients in diabetes management.
Impact
Research findings indicate that nurses should regularly assess the hypoglycaemia problem‐solving abilities in DM patients, and use trajectory subgroups to identify distinct patient characteristics in hypoglycaemia problem‐solving abilities in order to implement personalised interventions.
Reporting Method
This study was based on the STROBE guideline.
Patient or Public Contribution
No patient or public engagement.
Keywords: diabetes mellitus, hypoglycemia problem‐solving skill, latent class growth modelling, quality of life, trajectory
What Does This Paper Contribute to the Wider Global Clinical Community?
This study identified that the overall level of hypoglycaemia problem‐solving ability in DM patients shows an initial increase followed by a decline, with three heterogeneous trajectory groups: high‐level decline group, medium‐level fluctuation group and low‐level stable group.
This study identifies the influencing factors of hypoglycaemia problem‐solving ability and provides evidence‐based suggestions, serving as a reference for clinical nursing practice.
1. Introduction
Diabetes mellitus (DM) is a chronic metabolic disorder caused by the combined effects of genetic and environmental factors, primarily characterised by persistent hyperglycemia. Its core mechanism involves insufficient insulin secretion or impaired insulin action, leading to the body's inability to effectively utilise blood glucose and disturbance in carbohydrate, lipid and protein metabolism. DM has emerged as a critical global health challenge. According to the International Diabetes Federation (IDF), an estimated 537 million people worldwide had DM in 2021, with projections indicating a rise to 783 million by 2045 (Magliano et al. 2021), which represents an immense disease burden.
The cornerstone of diabetes management lies in maintaining glycaemic homeostasis. However, during intensive glycaemic control aimed at preventing complications, hypoglycaemia has emerged as a common and dangerous acute complication, posing a major obstacle in treatment (Ibrahim et al. 2020). Studies indicate that approximately one‐third of patients with DM have experienced severe hypoglycaemia (Ratzki‐Leewing et al. 2023). Severe hypoglycaemia events not only induce acute discomforts like palpitations, sweating and confusion (Mahendiran et al. 2024) but are also significantly associated with an increased risk of falls (Mattishent and Loke 2021), frequent cardiovascular events (Lee et al. 2018), impaired cognitive function (Racca et al. 2022) and even elevated all‐cause mortality (Amiel 2021; Felton et al. 2016). Therefore, the effective identification, prevention and management of hypoglycaemia constitute a critical priority for optimising diabetes management strategies and enhancing patient quality of life.
2. Background
Problem‐solving skills refer to an individual's ability to proactively identify problems, generate solutions, evaluate those solutions, and implement effective actions when faced with complex health situations (Toobert and Glasgow 1991). This ability can significantly improve treatment adherence in patients with diabetes, enabling them to cope with the complex and evolving challenges throughout the disease course. Hypoglycaemia problem‐solving ability denotes DM patients' capability to address unexpected hypoglycaemic episodes (Wu et al. 2016). For individuals with DM, this skill plays a pivotal role in the daily management of hypoglycaemia. Wang et al. (2012) found that informed decisions regarding diet, exercise and medication during hypoglycaemic episodes help patients achieve better symptom control (Wang et al. 2012). Existing studies have primarily focused on cross‐sectional investigations of factors influencing hypoglycaemia problem‐solving ability. For example, Naser et al. (2019) revealed that age, marital status, treatment regimen, and history of hypoglycaemia‐related hospitalisations in the past 6 months can affect patients' hypoglycaemia management ability (Naser et al. 2019). Nevertheless, hypoglycaemia problem solving is a dynamic cognitive and behavioural process, and skill levels exhibit fluctuations over time (Wu et al. 2016). There remains a lack of in‐depth exploration into its natural trajectory and health impacts; thus, longitudinal observations are particularly important.
Dynamic Biopsychosocial Model (DBPS; Engel 1977) emphasises the complex interactions among an individual's biological, psychological, and social factors, which collectively influence the development, presentation and prognosis of health and disease. Biological dynamics reflect the physical impacts of disease on the body, including disease‐related traits, physiological changes and clinical management behaviours that directly interact with the body's functions. Psychological dynamics involve a variety of interrelated psychological components, such as cognitive processes and emotional experiences. Social dynamics can be divided into interpersonal dynamics and macro‐system environmental dynamics: interpersonal dynamics include the effects of actual and perceived social contacts on dyadic and group processes; environmental dynamics encompass broad patterns of shared culture, norms, policies and values. Thus, DBPS provides a comprehensive framework for understanding disease progression. Fear of hypoglycaemia (FOH) is a negative emotional experience in which patients worry about hypoglycaemia and its adverse consequences (Bloomgarden 2017). Previous studies and our team's preliminary work have consistently demonstrated that the presence of FOH exerts unfavourable effects on DM management, further leading to a series of adverse health outcomes (Mangas et al. 2023). Social support provides not only spiritual comfort and encouragement but also material assistance and practical help, thus playing a crucial role in alleviating psychological stress and enhancing social adaptation.
Given that hypoglycaemia is a challenge in diabetes management, it remains unclear how patients' ability to address hypoglycaemic episodes evolves dynamically over the disease course, and how these different trajectories exert differential impacts on quality of life. To address this gap, this study adopts a prospective longitudinal design. Based on the DBPS theoretical model and preliminary investigation, we hypothesise the predictive factors of the development trajectory. The main hypotheses of this study are as follows: (1) Hypoglycaemia problem‐solving ability among patients with DM exhibits distinct developmental trajectories. (2) These trajectories can be predicted by biological, psychological and social factors, and further influence the quality of life of patients with DM. The findings will provide an evidence‐based foundation for healthcare providers to develop personalised interventions tailored to patients' varying trajectories of hypoglycaemia problem‐solving ability.
3. Methods
3.1. Sampling and Participants
A purposive sampling method was employed. Participants were recruited from DM patients at the endocrinology departments of two Grade A tertiary general hospitals in Jiangsu Province, China, from April 2024 to August 2024.
Inclusion Criteria: (1) Meeting the 2022 diagnostic criterion for DM (American Diabetes Association Professional Practice Committee 2022), including both type 1 and type 2 DM; (2) Duration of diabetes ≥ 1 year; (3) Age ≥ 18 years; (4) Experienced at least one hypoglycaemia episode within the past 6 months; (5) Possessing adequate communication and language comprehension abilities; (6) Voluntarily participating in the study and signed informed consent. Exclusion Criteria: (1) Pregnant or lactating women; (2) Individuals with diagnosed psychiatric or cognitive impairment disorders; (3) Patients who are critically ill and unable to cooperate with the investigation; (4) Concurrent participation in any other clinical trial.
3.2. Research Design and Data Collection
This longitudinal study systematically and comprehensively investigated patients' hypoglycaemia problem‐solving abilities at four predetermined time points: during hospitalisation (Time 0), 1 month after discharge (Time 1), 3 months after discharge (Time 2) and 6 months after discharge (Time 3), with a total follow‐up period of 6 months. The short‐term follow‐up period was determined based on a comprehensive literature review and expert panel discussions within the research group. The reasons are as follows: First, the 6 months after discharge is a critical window for evaluating the effect of transforming in‐hospital health education into practice among diabetic patients and monitoring the initial development of their self‐management ability (Xu et al. 2024). Monitoring this timeframe enables us to capture the early dynamic changes of hypoglycaemia problem‐solving ability, thus facilitating timely clinical intervention. Meanwhile, shorter‐term follow‐up aligns better with nurses' clinical workflow, facilitating the execution of follow‐up assessments and data collection, which in turn improves the feasibility and reproducibility of the study.
This study uses the Bayesian Information Criterion (BIC) and Entropy as the primary indicators for model selection; therefore, the sample size requirement is n ≥ 200 (An et al. 2024). Accounting for an anticipated 20% attrition rate, a minimum of 240 participants needed to be enrolled initially. To ensure the robustness of the research, a total of 321 diabetic patients were initially recruited for this study. During the baseline assessment (Time 0), 4 participants were excluded due to missing data. The subsequent number of participants lost to follow‐up was as follows: 3 participants at Time 1, 10 participants at Time 2 and 32 participants at Time 3. Consequently, 272 participants successfully completed the entire follow‐up protocol; 49 participants (15.26%) were lost to follow‐up throughout the study period. Participants were lost to follow‐up for the following reasons: incomplete data (n = 4), withdrawal from the study (n = 29) and non‐response (n = 16). The detailed participant flow is illustrated in Figure S1.
3.3. Research Variables and Tools
This study incorporates biological, psychological, interpersonal and environmental factors based on the DBPS model to explore the predictors of hypoglycaemia problem‐solving ability. Biological factors mainly include age, gender, BMI, educational level, medical insurance, monthly income, smoking, drinking, type of diabetes, course of the disease, complications, frequency of hypoglycaemia, hospitalisation due to hypoglycaemia, blood glucose monitoring and severe hypoglycaemia. Psychological factors mainly include the FOH; interpersonal factors include marital status, living circumstances and employment status; environmental factors mainly refer to social support.
3.3.1. General Information Questionnaire
A self‐designed questionnaire was developed based on a comprehensive literature review and in‐depth team discussions. It collected the following information: socio‐demographic characteristics, including age, gender, body mass index (BMI), education level, marital status, monthly income level and type of health insurance. Disease‐related information including duration of diabetes, presence of complications, current treatment regimen, frequency of hypoglycaemia and specifics of blood glucose monitoring practices.
3.3.2. Hypoglycaemia Problem‐Solving Scale
The hypoglycaemia problem‐solving ability of DM patients was evaluated using Hypoglycaemia Problem‐Solving Scale (HPSS) (Wu et al. 2016). This study adopted the revised version. With reference to the HPSS from the Taiwan region of China (Wu et al. 2018), all items and the structural framework of the Taiwanese version were fully retained, while the traditional Chinese characters were converted into simplified Chinese characters. This scale consists of 24 items across 7 dimensions, with items 1, 2, 3, 4, 23 and 24 being reverse‐scored. Responses are rated on a 5‐point Likert scale ranging from 0 (Never) to 4 (Always), yielding a total possible score between 0 and 96. A higher total score indicates a higher level of hypoglycaemia problem‐solving ability. This scale demonstrated excellent reliability and validity. Exploratory factor analysis extracted seven common factors, with a cumulative variance contribution rate of 77.902%. All item factor loadings were above 0.40, and no cross‐loadings were observed. Confirmatory factor analysis indicated good model fit (χ 2/df < 3.00, GFI > 0.85, RMSEA < 0.08, RMR < 0.05, NFI > 0.70, CFI > 0.70, TLI > 0.90). The overall Cronbach's α for the scale was 0.903. The split‐half reliability was 0.734, and the test–retest reliability was 0.834. In this study, the scale demonstrated excellent internal consistency with a Cronbach's α of 0.941.
3.3.3. Fear of Hypoglycaemia Scale‐15 (FH‐15)
The FH‐15 was developed by Anarte Ortiz et al. (2011) and validated in patients with type 1 DM to evaluate the degree of patients' FOH (Anarte Ortiz et al. 2011). Subsequently, the scholar Liu et al. (2018) Introduced and validated it in patients with type 2 DM (Liu et al. 2018). The FH‐15 comprises three factors: Fear, Interference and Avoidance. Each item is rated on a 5‐point Likert scale from 1 (Never) to 5 (Every day). The total score of the FH‐15 ranges from 15 to 75 points, with the optimal cut‐off value being 30.5 points. The Cronbach's α of this study was 0.915.
3.3.4. The Multidimensional Scale of Perceived Social Support (MSPSS)
The MSPSS developed by (Zimet et al. 1988; Dambi et al. 2018). This scale contains 12 items grouped into three dimensions: Family, Friends and Significant Other, each dimension comprising 4 items. Items are rated on a 7‐point scale ranging from 1 (very strongly disagree) to 7 (very strongly agree). The score range of this scale is from 12 to 84 points; higher scores indicate higher levels of perceived social support. This study's Cronbach's α of MPSS was 0.942.
3.3.5. Diabetes Specific Quality of Life Scale (DSQL)
The DSQL was used to assess the quality of life in patients with DM (Li et al. 2014). The DSQL consists of 27 items categorised into four dimensions: physiological, psychological, social and therapeutic. Responses are scored on a 5‐point Likert scale. Higher total scores indicate poorer diabetes specific quality of life. The total score of this scale ranges from 27 to 135 points, with 27 points representing the best health status and 135 points representing the worst health status. This study's Cronbach's α of DSQL was 0.896.
3.4. Statistical Analysis
All data underwent double data entry to ensure accuracy. Data analysis was performed using SPSS software 27.0. Quantitative data conforming to a normal distribution were presented as mean ± standard deviation (SD), and group comparisons were performed using Analysis of Variance (ANOVA). Data not conforming to a normal distribution were presented as median and interquartile, with group comparisons performed using non‐parametric tests. Qualitative data were presented as frequency and percentage (%). Group comparisons were performed using the Chi‐square test.
LCGM was conducted using Mplus software 8.0 to identify distinct trajectories of hypoglycaemia problem‐solving skills. The optimal model was selected based on the following criteria: Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC) and Sample‐Size Adjusted BIC (aBIC), where lower values indicate a better model fit. Entropy was used to assess the accuracy of class assignment (higher values indicate better classification, range 0–1). The Likelihood Ratio Test (LRT) and the Bootstrapped Likelihood Ratio Test (BLRT) were used to determine if a model with k classes provided a significantly better fit than a model with k‐1 classes (significant P‐value favours the k‐class model). The intercept and slope parameters of the trajectories were used to define and characterise each latent class. An unordered multinomial logistic regression model was employed to examine the influence of relevant variables on belonging to different hypoglycaemia problem‐solving skill trajectory classes. The dependent variable was the three‐category trajectory of changes in hypoglycaemia problem‐solving ability. Independent variables included those with p < 0.10 in univariate chi‐square tests, and covariates those with p < 0.10 in univariate ANOVA. The impact of different trajectory classes on quality of life was examined using one‐way ANOVA (for normally distributed outcomes) and multiple linear regression models. p‐values < 0.05 were considered statistically significant.
3.5. Validity and Reliability
This study was conducted and reported in accordance with the STROBE guidelines for cohort studies. Measurement tools with good reliability and validity were selected for this study, and the reliability of the tools was also analysed and reported.
3.6. Ethical Considerations
This study was approved by the Ethics Committee of Yangzhou University (No.: YZUHL2023077). Prior to participation, researchers clearly and accurately informed all participants of the study's purpose, specific procedures and scientific significance. All participants then signed a written informed consent form to confirm their voluntary involvement.
4. Result
4.1. General Characteristics of Patients With DM
In this study, baseline characteristics of 321 patients were analysed, including 272 patients who completed follow‐up and 49 patients who were lost to follow‐up. The mean age of the participants who completed the follow‐up was 60.69 ± 12.42 years, with 159 (58.46%) being male and 113 (41.54%) being female. Statistical analysis reveals no significant differences in demographic characteristics such as age and gender between the two groups (p > 0.05). However, significant inter‐group differences were observed in blood glucose monitoring frequency, FOH and social support (all p < 0.05). Compared with the completed follow‐up group, the lost to follow‐up group exhibited a significantly lower baseline frequency of blood glucose monitoring (p = 0.002), significantly higher fear of hypoglycaemia (p = 0.028) and significantly lower overall social support levels (p < 0.01). Refer Table S1 for details.
4.2. Changes in the Total Score and Dimension Scores of Hypoglycaemia Problem‐Solving Ability in DM Patients
Since the total score of hypoglycaemia problem‐solving ability and the scores of each dimension were all normally distributed, repeated measures ANOVA was used to statistically evaluate the dynamic changes in the overall level of hypoglycaemia problem‐solving ability and its specific dimensions among DM patients. As shown in Table S2, the total hypoglycaemia problem‐solving ability score is highest at Time 2 (47.49 ± 15.21) and lowest at Time 3 (45.30 ± 14.52). Statistically significant differences were found in the total score and all dimension scores of the hypoglycaemia problem‐solving ability among patients with DM (p < 0.05).
4.3. The Trajectory of Hypoglycaemia Problem‐Solving Ability in DM Patients
An analysis of the overall trajectory revealed an increase in hypoglycaemia problem‐solving ability at Time 1 among DM patients, followed by a decrease from Time 2 to Time 3. The results are shown in Figure 1.
FIGURE 1.
The general trajectory of the total score of hypoglycaemia problem‐solving ability in DM patients. [Colour figure can be viewed at wileyonlinelibrary.com]
We further applied LCGM to analyse the change trajectory of the scores across the dimensions of hypoglycaemia problem‐solving ability. As shown in Figure S2, the ‘problem perception’ dimension exhibited the most significant change, with a gradual downward trend. The ‘identifying attribute’ dimension also showed notable changes, while the remaining dimensions displayed relatively gentle trends.
4.4. Heterogeneous Trajectories of Change in Hypoglycaemia Problem‐Solving Ability in DM Patients
The hypoglycaemia problem‐solving ability scores of patients with DM at 4 time points were used as observational indices. The LCGM linear and quadratic models were applied to sequentially extract 1–4 latent classes for parametric goodness‐of‐fit analysis. Ultimately, the linear 3‐class model was selected as the optimal fitting model for the trajectory of changes in hypoglycaemia problem‐solving ability among patients with DM, with fit indices presented in Table S3.
In our study, two factors, intercept and slope, were selected from the LCGM (Table S 4) to reveal the trajectories of 3 classes of DM patients' level of hypoglycaemia problem‐solving ability over time. As shown in Figure 2: (1) Class 1 (n = 91, 33.45%) demonstrated initial stabilisation (Time 1) followed by a significant decline (Time 2), with intercept = 65.99, slope = −0.87, p < 0.01. Despite a declining trend, this group maintained a relatively high level of hypoglycaemia problem‐solving ability compared to the other two groups, and was thus termed the high‐level declining group. (2) Class 2 (n = 76, 27.94%) demonstrated an upward trend at Time 1, followed by a transient decline at Time 2 and then stabilisation (intercept = 46.89, slope = −0.55, p < 0.01). This subgroup was defined as the medium‐level fluctuation group since its hypoglycaemia management ability was consistently maintained at a medium level. (3) Class 3 (n = 105, 38.60%) demonstrated a significantly lower baseline value compared to other classes. During the 6‐month longitudinal observation, their indicators remained relatively constant and their hypoglycaemia management ability was consistently low (intercept = 32.09, slope = −0.54, p < 0.01). Based on these characteristics, this subgroup was defined as the low‐level stability group. These results indicate that there are differences in the trajectory of change in hypoglycaemia problem‐solving ability levels among patients with DM.
FIGURE 2.
Three classification trajectories of hypoglycaemia problem‐solving ability in DM patients. [Colour figure can be viewed at wileyonlinelibrary.com]
4.5. Differential Test of the Trajectory Group of Changes in Hypoglycaemia Problem‐Solving Ability
Incorporating variables based on the DBPS model, the chi‐square test results showed that gender, age, marital status, employment status, education level, residence, living style, primary caregiver, monthly income, smoking status, disease type, hypoglycaemia frequency, glucose monitoring frequency and FOH exhibit significant differences in distribution across the three trajectory categories (p < 0.10), whereas other variables were not statistically significant (p > 0.10). One‐way ANOVA results indicated a significant difference in the distribution of the trajectories of the social support among the three categories (p < 0.10). Details are presented in Table 1.
TABLE 1.
Differential test of the trajectory group of changes in hypoglycaemia problem‐solving ability.
| Variable | Category | High level decline group (n = 91) | Medium level fluctuation group (n = 76) | Low level stable group (n = 105) | χ 2/F | p |
|---|---|---|---|---|---|---|
| Gender | Male | 62 (68.13) | 40 (52.63) | 57 (54.29) | 5.32 a | 0.070 |
| Female | 29 (31.87) | 36 (47.37) | 48 (45.71) | |||
| Age | 18–44 | 13 (14.29) | 9 (11.84) | 3 (2.86) | 23.97 a | < 0.001 |
| 45–59 | 44 (48.35) | 26 (34.21) | 28 (26.67) | |||
| ≥ 60 | 34 (37.36) | 41 (53.95) | 74 (70.48) | |||
| BMI (kg/m2) | < 18.5 | 4 (4.40) | 3 (3.95) | 4 (3.81) | — | 0.600 |
| 18.5–23.9 | 42 (46.15) | 36 (47.37) | 51 (48.57) | |||
| 24–27.9 | 35 (38.46) | 22 (28.95) | 39 (37.14) | |||
| > 28 | 10 (10.99) | 15 (19.74) | 11 (10.48) | |||
| Marital status | Unmarried | 3 (3.30) | 1 (1.32) | 2 (1.90) | — | 0.046 |
| Married | 87 (95.60) | 67 (88.16) | 94 (89.52) | |||
| Divorced or widowed | 1 (1.10) | 8 (10.53) | 9 (8.57) | |||
| Employment | On the job | 39 (42.86) | 20 (26.32) | 17 (16.19) | 40.22 a | < 0.001 |
| Retire | 36 (39.56) | 32 (42.11) | 47 (44.76) | |||
| Farmer | 5 (5.49) | 12 (15.79) | 36 (34.29) | |||
| Unemployment | 1 (1.10) | 1 (1.32) | 0 (0.00) | |||
| Other | 10 (10.99) | 11 (14.47) | 5 (4.76) | |||
| Education | Primary school | 10 (10.99) | 19 (25.00) | 44 (41.90) | — | < 0.001 |
| Middle school | 44 (48.35) | 43 (56.58) | 54 (51.43) | |||
| High school | 33 (36.26) | 12 (15.79) | 6 (5.71) | |||
| College and above | 4 (4.40) | 2 (2.63) | 1 (0.95) | |||
| Residence | City | 76 (83.52) | 48 (63.16) | 54 (51.43) | 22.44 a | < 0.001 |
| Village | 15 (16.48) | 28 (36.84) | 51 (48.57) | |||
| Living style | Living alone | 3 (3.30) | 10 (13.16) | 2 (1.90) | 12.01 a | 0.002 |
| Not living alone | 88 (96.70) | 66 (86.84) | 103 (98.10) | |||
| Medicare | No | 2 (2.20) | 4 (5.26) | 8 (7.62) | — | 0.250 |
| Yes | 89 (97.80) | 72 (94.74) | 97 (92.38) | |||
| Care giver | Self | 31 (34.07) | 35 (46.05) | 50 (47.62) | 21.06 a | < 0.001 |
| Spouse | 59 (64.84) | 31 (40.79) | 40 (38.10) | |||
| Children or others | 1 (1.10) | 10 (13.16) | 15 (14.29) | |||
| Monthly income | < 1000 | 2 (2.20) | 4 (5.26) | 10 (9.52) | 28.70 a | < 0.001 |
| 1000–3000 | 26 (28.57) | 37 (48.68) | 59 (56.19) | |||
| 3000–5000 | 46 (50.55) | 23 (30.26) | 31 (29.52) | |||
| > 5000 | 17 (18.68) | 12 (15.79) | 5 (4.76) | |||
| Smoking | No | 62 (68.13) | 63 (82.89) | 82 (78.10) | 5.34 a | 0.069 |
| Yes | 29 (31.87) | 13 (17.11) | 23 (21.90) | |||
| Drinking | No | 70 (76.92) | 67 (88.16) | 80 (76.19) | 4.61 a | 0.100 |
| Yes | 21 (23.08) | 9 (11.84) | 25 (23.81) | |||
| Disease type | 1 type | 12 (13.19) | 3 (3.95) | 0 (0.00) | 16.77 a | < 0.001 |
| 2 type | 79 (86.81) | 73 (96.05) | 105 (100.00) | |||
| Diabetes course | < 5 | 11 (12.09) | 16 (21.05) | 26 (24.76) | 8.72 a | 0.190 |
| 5–10 | 22 (24.18) | 10 (13.16) | 15 (14.29) | |||
| 10–15 | 22 (24.18) | 22 (28.95) | 24 (22.86) | |||
| > 15 | 36 (39.56) | 28 (36.84) | 40 (38.10) | |||
| Merged diseases | 1 | 22 (24.18) | 17 (22.37) | 18 (17.14) | 4.45 a | 0.348 |
| 2 | 33 (36.26) | 23 (30.26) | 30 (28.57) | |||
| ≥ 3 | 36 (39.56) | 36 (47.37) | 57 (54.29) | |||
| Diabetic complications | No | 41 (45.05) | 37 (48.68) | 51 (48.57) | 4.91 a | 0.297 |
| < 2 | 34 (37.36) | 31 (40.79) | 46 (43.81) | |||
| > 2 | 16 (17.58) | 8 (10.53) | 8 (7.62) | |||
| Treatment | No medication | 2 (2.20) | 1 (1.32) | 5 (4.76) | — | 0.290 |
| Oral drug | 25 (27.47) | 24 (31.58) | 39 (37.14) | |||
| Insulin | 30 (32.97) | 16 (21.05) | 23 (21.90) | |||
| Oral medication combined with insulin | 34 (37.36) | 35 (46.05) | 38 (36.19) | |||
| Admission due to hypoglycaemia | No | 76 (83.52) | 64 (84.21) | 87 (82.86) | 0.06 a | 0.971 |
| Yes | 15 (16.48) | 12 (15.79) | 18 (17.14) | |||
| Hypoglycaemia frequency | 1–3 | 33 (36.26) | 42 (55.26) | 60 (57.14) | 11.80 a | 0.019 |
| 3–6 | 52 (57.14) | 27 (35.53) | 38 (36.19) | |||
| > 6 | 6 (6.59) | 7 (9.21) | 7 (6.67) | |||
| Severe hypoglycaemia | No | 74 (81.32) | 59 (77.63) | 85 (80.95) | 0.42 a | 0.809 |
| Yes | 17 (18.68) | 17 (22.37) | 20 (19.05) | |||
| Blood glucose monitoring frequency | High | 33 (36.26) | 50 (65.79) | 93 (88.57) | 58.46 a | < 0.001 |
| Low | 58 (63.74) | 26 (34.21) | 12 (11.43) | |||
| FOH | < 0.5 | 71 (78.02) | 41 (53.95) | 49 (46.67) | 21.04 a | < 0.001 |
| ≥ 30.5 | 20 (21.98) | 35 (46.05) | 56 (53.33) | |||
| Social support | — | 62.75 ± 5.27 | 56.96 ± 9.11 | 51.67 ± 9.50 | 44.47 b | < 0.001 |
Abbreviations: BMI, body mass index; FOH, fear of hypoglycaemia.
χ 2 value.
F value.
4.6. Unordered Multinomial Logistic Regression Analysis of Heterogeneous Trajectories of Hypoglycaemia Problem‐Solving Ability in DM Patients
Unordered multinomial logistic regression analysis (Table 2) revealed that, compared with the low‐level stability group (Class 3), a higher frequency of blood glucose monitoring, absence of fear of hypoglycaemia, and higher social support were positive predictors in the high‐level decline group. For the medium‐level fluctuation group (Class 2), being a farmer and having a monthly income < CNY 1000 were negative predictors, while higher social support was a positive predictor.
TABLE 2.
Disordered multi classification logistic regression analysis of heterogeneous change track of hypoglycaemia problem‐solving ability in DM patients.
| Class | Variable | b | SE | Wald | p | OR | 95% CI |
|---|---|---|---|---|---|---|---|
| High‐level decline group (class 1) |
Blood glucose monitoring frequency (reference: low) |
||||||
| High | 2.057 | 0.522 | 15.500 | < 0.001 | 7.822 | 2.809–21.777 | |
| Fear of hypoglycaemia (reference: yes) | |||||||
| No | 1.434 | 0.496 | 8.368 | 0.004 | 4.195 | 1.588–11.085 | |
| Social support | 0.177 | 0.039 | 20.800 | < 0.001 | 1.193 | 1.106–1.287 | |
| Medium‐level fluctuation group (class 2) | Employment (reference: other) | ||||||
| On the job | −0.759 | 0.777 | 0.955 | 0.329 | 0.468 | 0.102–2.146 | |
| Retire | −1.269 | 0.701 | 3.280 | 0.070 | 0.281 | 0.071–1.110 | |
| Farmer | −1.814 | 0.911 | 3.959 | 0.047 | 0.163 | 0.027–0.973 | |
| Unemployment | 21.290 | — | 0.000 | 0.999 | — | — | |
| Monthly income (reference: 3000–5000) | |||||||
| < 1000 | −3.009 | 1.504 | 4.004 | 0.045 | 0.049 | 0.003–0.940 | |
| 1000–3000 | −0.226 | 0.872 | 0.067 | 0.795 | 0.797 | 0.144–4.403 | |
| > 5000 | −0.175 | 0.820 | 0.045 | 0.831 | 0.839 | 0.168–4.192 | |
| Social support | 0.057 | 0.027 | 4.627 | 0.031 | 1.059 | 1.005–1.115 |
Note: The low‐level stable group (class 3) is used as a reference; ‘—’ indicates a relatively large value, an abnormal OR value and no statistical significance. Bold values indicate statistically significant differences (p < 0.05).
4.7. Impact of Heterogeneous Trajectories in Hypoglycaemia Problem‐Solving Ability on Quality of Life in DM Patients
One‐way ANOVA revealed statistically significant differences in quality of life across hypoglycaemia problem‐solving ability trajectory groups among DM patients (p < 0.05). The analysis results indicated that the quality of life of the low‐level stable group was 51.370 ± 7.396. The medium‐level fluctuation group demonstrated the highest quality of life (54.070 ± 9.051), while the high‐level decline group showed the lowest values (46.890 ± 8.615). The details are shown in Table 3. In the multiple linear regression model (Table 4), after controlling for confounding factors, the differences in quality of life among the three trajectory groups remained statistically significant (p < 0.05). Moreover, there is a positive correlation between the hypoglycaemia problem‐solving ability trajectories in DM patients and their quality of life.
TABLE 3.
A univariate analysis of the impact of heterogeneous trajectories of change in hypoglycaemia problem‐solving ability on quality of life in DM patients.
| High level decline group | Medium level fluctuation group | Low level stable group | F | p | |
|---|---|---|---|---|---|
| Quality of life | 46.890 ± 8.615 | 54.070 ± 9.051 | 51.370 ± 7.396 | 16.185 | < 0.001 |
TABLE 4.
Multivariate linear regression analysis of the heterogeneous trajectory on quality of life in DM patients with hypoglycaemia problem‐solving ability.
| F | SE | t | p | |
|---|---|---|---|---|
| Constant | 50.401 | 6.974 | 7.227 | < 0.001 |
| Reference (high level decline group) | ||||
| Medium level fluctuation group | 6.735 | 1.534 | 4.391 | < 0.001 |
| Low level stable group | 6.068 | 1.649 | 3.680 | < 0.001 |
Note: The trajectory group is the independent variable, and the quality of life is the dependent variable. General information is used as a confounding factor.
5. Discussion
In this longitudinal study, we analysed the baseline characteristics of 321 enrolled patients, of whom 272 completed follow‐up and 49 were lost to follow‐up. Participants lost to follow‐up exhibited significantly lower baseline blood glucose monitoring frequency, higher FOH and lower overall social support levels compared to completers. These differences indicate that the observed variables are associated with follow‐up attrition. To adjust for potential attrition bias, we incorporated these influencing factors based on the DBPS model into our unordered multinomial logistic regression analysis. Notably, the direction of associations was consistent: lower monitoring frequency, higher FOH and lower social support were linked to both higher dropout risk and trajectory group change. This consistency strengthens the validity of these variables as key predictors of hypoglycaemia problem‐solving ability trajectories.
5.1. The Trajectory of Hypoglycaemia Problem‐Solving Ability in DM Patients
The hypoglycaemia problem‐solving ability of diabetic patients plays a crucial role in diabetes self‐management. This study identified a distinct ‘initial‐rise‐then‐decline’ trajectory in the overall trend of hypoglycaemia problem‐solving skills among DM patients: a transient increase at Time 1, followed by a progressive linear decline from Time 2 to Time 3. This fluctuation likely reflects the interplay of episodic support discontinuation and emerging self‐management fatigue. The initial rise at Time 1 can be reasonably attributed to two concurrent factors. On the one hand, experiencing hypoglycaemia during hospitalisation, coupled with structured health education from clinicians (Worku et al. 2025), may significantly bolster patients' hypoglycaemia awareness and problem‐solving abilities. On the other hand, patients typically exhibit heightened adherence to management protocols immediately following discharge (Brieger et al. 2018). However, the gradual fading of hypoglycaemia management knowledge acquired during hospitalisation over time, coupled with diminished vigilance and adherence due to waning attention to self‐health or life‐related factors (Lin et al. 2023), results in a decline in their ability to solve hypoglycaemic problems between the Time 2 to Time 3 time points. In addition, the chronicity of hypoglycaemia management imposes significant psychological stress, fostering anxiety, depression, or other mood disturbances (Hermanns et al. 2024; Matlock et al. 2022). Critically, these psychological factors emerge as key that impair sustained problem‐solving capacity. These findings emphasise the necessity of a continuous post‐discharge support system. It is recommended that clinical nurses implement structured longitudinal education and support programs after hospital discharge to reinforce hypoglycaemia problem‐solving skills and prevent skill depletion in patients with diabetes (Campbell et al. 2018).
A key contribution of this study is the first systematic identification, via LCGM, of distinct dynamic trajectories in hypoglycaemia problem‐solving abilities among patients with DM. LCGM revealed three distinct latent‐class trajectory groups: high‐level decline group (33.46%), medium‐level fluctuation group (27.94%) and low‐level stable group (38.60%). The high‐level decline group exhibited initially high hypoglycaemia problem‐solving abilities, but a significant decline after 3 months post‐discharge. In this study, most people in the high‐level decline group lived in urban areas. Their initial high skill level may be attributed to a predominance of urban residents exhibiting frequent glucose monitoring, absence of FOH, and established self‐management competencies. The subsequent decline likely stems from a lack of sustained education and support, hindering knowledge updates and effective problem‐solving for new challenges over time (Hansen and Bibby 2024). The medium‐level fluctuation group is characterised by a ‘rise‐decline‐stabilisation’ pattern, rising at Time 1, declining at Time 2, then stabilising by Time 3. The initial rise at Time 1 may reflect the application of hypoglycaemia management education received prior to discharge, which improved patients' short‐term problem‐solving abilities (Wu et al. 2018). The decline at Time 2 could result from waning adherence to dietary and exercise regimens, leading to glycaemic instability and reduced problem‐solving efficacy. Subsequent stabilisation at Time 3 may indicate patient adaptation to new routines or treatment regimens through self‐adjustment, and potentially facilitated by positive social support (Chen and Lin 2025). The low‐level stable group exhibited a consistently low trajectory of hypoglycaemia problem‐solving ability. In our study results, most patients in the low‐level hypoglycaemia group have a FOH, which appears to fundamentally weaken their ability to respond effectively to hypoglycaemia. Based on these findings, clinical nurses can develop and implement individualised hypoglycaemia prevention strategies tailored to patients in different trajectory groups. This should include enhanced structured training in hypoglycaemia problem‐solving abilities and efforts to improve patients' self‐efficacy, with the goal of achieving sustained self‐management.
5.2. Predictors of Heterogeneous Trajectories in Hypoglycaemia Problem‐Solving Ability
According to the results of unordered multinomial logistic regression, frequency of blood glucose monitoring, FOH and social support were significant predictors of high‐level decline group. Employment status, monthly income and social support were predictive factors for the medium‐level fluctuation group. Our findings revealed distinct fluctuations in hypoglycaemia problem‐solving abilities among patients engaged in farming and those with monthly incomes < CNY 1000. This aligns with prior evidence; Lu et al. (2024) research indicates that farmers often experience restricted access to health information resources post‐discharge (Lu et al. 2024), perhaps leading to suboptimal hypoglycaemia knowledge. This frequently manifests as inappropriate medication management, such as self‐adjusting dosages, switching regimens without consultation, contributing to fluctuations in problem‐solving efficacy. Meanwhile, we found that the efficiency of hypoglycaemia problem‐solving in low‐income patients is low, possibly due to economic constraints that limit their access to high‐quality care services (Tini et al. 2024).
In addition to social and demographic factors, patients' self‐management, psychological factors and social factors also affect their hypoglycaemia problem‐solving ability in DM patients. Frequent blood glucose monitoring enables patients to promptly detect hypoglycaemia events (Li and Chen 2023), and guides timely medication adjustments based on glucose fluctuations (Tătaru et al. 2022), thereby mitigating the risk of severe hypoglycaemia and its potential serious consequences. Consistent with this, we discovered the influence of the frequency of blood glucose monitoring on the heterogeneous change trajectory of hypoglycaemia problem‐solving ability in DM patients. Additionally, patients without FOH demonstrated a greater propensity for flexible coping strategies, such as adjusting diet or activity in response to glucose readings. This ability contributed to more effective hypoglycaemia management, highlighting a critical link between psychological well‐being and problem‐solving abilities. In line with this, prior studies have shown that individualised psychological interventions can effectively reduce hypoglycaemia‐related fears, anxiety and emotional disturbances (Martyn‐Nemeth et al. 2024; Yan et al. 2024). Besides, we also found that patients with higher levels of social support had a strong association with better hypoglycaemia problem‐solving ability. This phenomenon has also been supported by previous studies. For instance, Ren et al. (2024) noted that hypoglycaemia can affect patients' work, study and family life, thereby causing psychological stress (Ren et al. 2024). Patients with strong social support networks benefit from enhanced understanding, care and practical assistance from family, friends and community members (van Husen et al. 2025). Such support facilitates psychological adaptation, improves stress‐coping abilities and bolsters the capacity for effective hypoglycaemia problem‐solving. In conclusion, we recommend regular screening for diabetes‐related concerns—such as FOH and depression—to promptly identify emotion‐related declines in hypoglycaemia problem‐solving ability. Concurrently, nurses should provide targeted psychological support focused on diabetes distress in their daily care to address the emotional factors contributing to impaired hypoglycaemia problem‐solving ability.
5.3. Impact of Hypoglycaemia Problem‐Solving Ability on Quality of Life
Notably, this study provides the important evidence that distinct trajectories of hypoglycaemia problem‐solving abilities significantly impact quality of life in DM patients. The sensitivity to and coping ability with hypoglycaemia vary among different patients. This study found that patients in the high‐level decline group exhibited the poorest quality of life at T3. In contrast, quality of life was significantly better in the medium‐level fluctuation group and the low‐level stable group than in the high‐level decline group. The results indicate that marked declines in hypoglycaemia problem‐solving ability exert a particularly strong negative impact on patients. A shift from high to low ability represents a relatively drastic change, which may leave patients unable to adapt effectively to hypoglycaemia‐related adverse symptoms. Such symptoms can directly increase physiological burdens, disrupt daily activities and consequently reduce overall quality of life (Palani et al. 2023). It is worth noting that significant inter‐group differences exist in living circumstances across the three trajectory classes, including marital status, employment, residential and living style. Specifically, the high‐level decline group mostly consisted of married individuals (95.6%), employed participants (42.86%), urban residents (83.52%) and those not living alone (96.70%). However, after controlling for relevant confounding factors, the influence of different trajectory groups on quality of life remained statistically significant. This suggests that quality of life in diabetic patients is not solely determined by factors related to their living environment, but may depend on the degree of consistency between the resources or support available in their living circumstances and their health expectations (Turnbull et al. 2023). This congruence influences the developmental trajectory of hypoglycaemia problem‐solving ability and ultimately affects patients' quality of life. Therefore, living conditions may serve as an indirect driving factor for quality of life among diabetic patients. Therefore, when developing interventions, healthcare providers should prioritise trajectory characteristics rather than merely improving abilities. This approach is crucial for improving disease prognosis and maintaining a higher quality of life.
5.4. Innovation and Limitation
This study may have some limitations due to human resources and uncontrollable factors. Firstly, the sample was drawn exclusively from DM patients in the endocrinology departments of two tertiary grade A general hospitals in Jiangsu Province, China, which may restrict the generalisability of findings to broader populations. Secondly, the follow‐up observation period was limited to 6 months post‐discharge, and outcomes relied on patient self‐reporting, so there might be potential information bias. Despite these limitations, this study still offers substantial contributions. This study breaks new ground by shifting the focus from cross‐sectional assessments to longitudinal tracking of hypoglycaemia problem‐solving abilities in DM patients, capturing their dynamic fluctuations. Meanwhile, it provides the first in‐depth exploration of predictors underlying heterogeneous trajectories of hypoglycaemia problem‐solving abilities, and the significant impact of these trajectories on quality of life. These findings establish a critical theoretical foundation for designing and implementing effective interventions targeting distinct trajectories of hypoglycaemia problem‐solving abilities in DM patients. Future studies should extend the follow‐up duration to capture longer‐term evolution, employ a multicenter design across diverse healthcare settings to enhance sample representativeness and statistical power.
6. Conclusion
In this prospective cohort study, we identified three distinct trajectories of hypoglycaemia problem‐ solving ability through LGGM, confirming significant population heterogeneity in their longitudinal trajectories. Furthermore, we identified key predictors of hypoglycaemia problem‐solving ability in DM patients, and clarified the substantial impact of heterogeneous trajectories of hypoglycaemia problem‐solving ability on quality of life. Based on this, nurses should leverage these trajectory subgroups to recognise distinct patient features in hypoglycaemia problem‐solving abilities, implement stratified, dynamically adjusted personalised interventions and ultimately achieve multifaceted health goals such as improving metabolic control, preventing complications and enhancing their quality of life.
Author Contributions
Qianqian Cheng is responsible for writing manuscripts, data curation and analysis. Jiahui Qiu is responsible for writing the original manuscript, data collection and analysis. Mengyao Han, Xiaoyu Wan, Qiuyu Zhang, Ziwei Xu, Qianqian Zheng and Beixi Shi are responsible for data collection and curation. Yu Zhang is responsible for research design, research guidance and thesis revision. All authors reviewed and approved the final draft of the manuscript.
Funding
This work was supported by the Postgraduate Research & Practice Innovation Program of Jiangsu Province (KYCX25_4112). The funding body plays no role in the design of the study and collection, analysis, interpretation of the data and in writing the manuscript.
Disclosure
Statistical analysis statement: The author(s) affirm that the methods used in the data analyses are suitably applied to their data within their study design and context, and the statistical findings have been implemented and interpreted correctly.
Ethics Statement
This study received approval from the Ethics Committee of the Nursing School at Yangzhou University (No.: YZUHL2023077) on October 9, 2023. All participants provided informed consent and participated voluntarily.
Consent
All participants provided informed consent and participated voluntarily.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Data S1: jocn70336‐sup‐0001‐supinfo01.docx.
Figure S1: Flowchart illustrating participant recruitment, follow‐up status and reasons for loss to follow‐up.
Figure S2: Dynamic change trajectories of scores for each dimension of hypoglycaemia problem‐ solving ability in patients with DM over the follow‐up period.
Table S1: Comparison of baseline characteristics between DM patients with complete follow‐up and those lost to follow‐up.
Table S2: Presents the changes in the total score and sub‐dimension scores of hypoglycaemia problem‐solving ability in DM patients at different time points.
Table S3: Fitting indices and goodness‐of‐fit evaluations of the LCGM for hypoglycaemia problem‐solving ability in DM patients.
Table S4: Parameter estimation results, statistical significance tests and covariance calibration of the LCGM for hypoglycaemia problem‐solving ability in DM patients.
Acknowledgements
The authors thank the researchers who participated for their contributions. We also thank all the patients who participated in this study.
Cheng, Q. , Qiu J., Han M., et al. 2026. “The Trajectory of Hypoglycaemia Problem‐Solving Ability in Patients With Diabetes and Its Impact on Quality of Life.” Journal of Clinical Nursing 35, no. 8: 3474–3485. 10.1111/jocn.70336.
Qianqian Cheng and Jiahui Qiu are regarded as joint first author.
Data Availability Statement
The data sets analysed in the current study are available from the corresponding authors upon reasonable request.
References
- American Diabetes Association Professional Practice Committee . 2022. “2. Classification and Diagnosis of Diabetes: Standards of Medical Care in Diabetes‐2022.” Diabetes Care 45, no. Suppl 1: S17–S38. 10.2337/dc22-S002. [DOI] [PubMed] [Google Scholar]
- Amiel, S. A. 2021. “The Consequences of Hypoglycaemia.” Diabetologia 64, no. 5: 963–970. 10.1007/s00125-020-05366-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- An, C. , Wei R., Yao W., et al. 2024. “Association of Serum AFP Trajectories and Hepatocellular Carcinoma Outcomes After Hepatic Arterial Infusion Chemotherapy: A Longitudinal, Multicenter Study.” Cancer Medicine 13, no. 11: e7319. 10.1002/cam4.7319. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Anarte Ortiz, M. T. , Caballero F. F., Ruiz de Adana M. S., et al. 2011. “Development of a New Fear of Hypoglycemia Scale: FH‐15.” Psychological Assessment 23, no. 2: 398–405. 10.1037/a0021927. [DOI] [PubMed] [Google Scholar]
- Bloomgarden, Z. 2017. “Fear of Hypoglycemia.” Journal of Diabetes 9, no. 2: 108–110. 10.1111/1753-0407.12491. [DOI] [PubMed] [Google Scholar]
- Brieger, D. , Chow C., Gullick J., et al. 2018. “Improving Patient Adherence to Secondary Prevention Medications 6 Months After an Acute Coronary Syndrome: Observational Cohort Study.” Internal Medicine Journal 48, no. 5: 541–549. 10.1111/imj.13736. [DOI] [PubMed] [Google Scholar]
- Campbell, F. , Lawton J., Rankin D., et al. 2018. “Follow‐Up Support for Effective Type 1 Diabetes Self‐Management (The FUSED Model): A Systematic Review and Meta‐Ethnography of the Barriers, Facilitators and Recommendations for Sustaining Self‐Management Skills After Attending a Structured Education Programme.” BMC Health Services Research 18, no. 1: 898. 10.1186/s12913-018-3655-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Chen, W.‐J. , and Lin L.‐Y.. 2025. “The Impact of Diabetes Self‐Care, Healthy Lifestyle, Social Support, and Demographic Variables on Outcomes HbA1c in Patients With Type 2 Diabetes.” Clinical Medicine Insights. Endocrinology and Diabetes 18: 11795514251331907. 10.1177/11795514251331907. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Dambi, J. M. , Corten L., Chiwaridzo M., Jack H., Mlambo T., and Jelsma J.. 2018. “A Systematic Review of the Psychometric Properties of the Cross‐Cultural Translations and Adaptations of the Multidimensional Perceived Social Support Scale (MSPSS).” Health and Quality of Life Outcomes 16, no. 1: 80. 10.1186/s12955-018-0912-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Engel, G. L. 1977. “The Need for a New Medical Model: A Challenge for Biomedicine.” Science 196, no. 4286: 129–136. 10.1126/science.847460. [DOI] [PubMed] [Google Scholar]
- Felton, A.‐M. , LaSalle J., McGill M., and Global Partnership for Effective Diabetes Management . 2016. “Treatment Urgency: The Importance of Getting People With Type 2 Diabetes to Target Promptly.” Diabetes Research and Clinical Practice 117: 100–103. 10.1016/j.diabres.2016.04.026. [DOI] [PubMed] [Google Scholar]
- Hansen, K. W. , and Bibby B. M.. 2024. “Rebound Hypoglycemia and Hyperglycemia in Type 1 Diabetes.” Journal of Diabetes Science and Technology 18, no. 6: 1392–1398. 10.1177/19322968231168379. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Hermanns, N. , Ehrmann D., Finke‐Gröne K., Roos T., Freckmann G., and Kulzer B.. 2024. “Evaluation of a Digital Health Tool for Titration of Basal Insulin in People With Type 2 Diabetes: Rationale and Design of a Randomized Controlled Trial.” Journal of Diabetes Science and Technology 18, no. 4: 946–955. 10.1177/19322968221148756. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ibrahim, M. , Baker J., Cahn A., et al. 2020. “Hypoglycaemia and Its Management in Primary Care Setting.” Diabetes/Metabolism Research and Reviews 36, no. 8: e3332. 10.1002/dmrr.3332. [DOI] [PubMed] [Google Scholar]
- Lee, A. K. , Warren B., Lee C. J., et al. 2018. “The Association of Severe Hypoglycemia With Incident Cardiovascular Events and Mortality in Adults With Type 2 Diabetes.” Diabetes Care 41, no. 1: 104–111. 10.2337/dc17-1669. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Li, M. , Li T., Shi B.‐Y., and Gao C.‐X.. 2014. “Impact of Motivational Interviewing on the Quality of Life and Its Related Factors in Type 2 Diabetes Mellitus Patients With Poor Long‐Term Glycemic Control.” International Journal of Nursing Sciences 1, no. 3: 250–254. 10.1016/j.ijnss.2014.05.022. [DOI] [Google Scholar]
- Li, Y. , and Chen Y.. 2023. “Review of Noninvasive Continuous Glucose Monitoring in Diabetics.” ACS Sensors 8, no. 10: 3659–3679. 10.1021/acssensors.3c01538. [DOI] [PubMed] [Google Scholar]
- Lin, Y. K. , Agni A., Chuisano S., et al. 2023. “‘You Have to Use Everything and Come to Some Equilibrium’: A Qualitative Study on Hypoglycemia Self‐Management in Users of Continuous Glucose Monitor With Diverse Hypoglycemia Experiences.” BMJ Open Diabetes Research & Care 11, no. 3: e003415. 10.1136/bmjdrc-2023-003415. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Liu, Y.‐Q. , Xiong S.‐Q., Sang M., et al. 2018. “Reliability and Validity of the Chinese Version of the New Fear of Hypoglycemia Scale: FH‐15.” International Journal of Nursing Sciences 5, no. 4: 343–351. 10.1016/j.ijnss.2018.09.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lu, Y. , Li C., Xu W., Lv S., and Li L.. 2024. “Blood Glucose Control Among Type 2 Diabetic Farmers in Chinese Resettlement Areas: A Mixed Methods Study.” BMC Endocrine Disorders 24, no. 1: 163. 10.1186/s12902-024-01686-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Magliano, D. J. , Boyko E. J., and IDF . 2021. IDF Diabetes Atlas. 10th ed. International Diabetes Federation. http://www.ncbi.nlm.nih.gov/books/NBK581934/. [Google Scholar]
- Mahendiran, K. , Moorthy S., Lakshmi M., et al. 2024. “Spontaneous Hypoglycemia in a Non‐Diabetic Patient: A Diagnostic and Therapeutic Conundrum.” Cureus 16, no. 7: e65467. 10.7759/cureus.65467. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Mangas, N. , Mateu‐Salat M., Martínez M. J., et al. 2023. “Hybrid Closed‐Loop Systems Can Help Patients With Extreme Fear of Hypoglycemia or Hyperglycemia.” Hormones 22, no. 3: 453–456. 10.1007/s42000-023-00451-9. [DOI] [PubMed] [Google Scholar]
- Martyn‐Nemeth, P. , Duffecy J., Quinn L., et al. 2024. “FREE: A Randomized Controlled Feasibility Trial of a Cognitive Behavioral Therapy and Technology‐Assisted Intervention to Reduce Fear of Hypoglycemia in Young Adults With Type 1 Diabetes.” Journal of Psychosomatic Research 181: 111679. 10.1016/j.jpsychores.2024.111679. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Matlock, K. A. , Broadley M., Hendrieckx C., et al. 2022. “Changes in Quality of Life Following Hypoglycaemia in Adults With Type 2 Diabetes: A Systematic Review of Longitudinal Studies.” Diabetic Medicine: A Journal of the British Diabetic Association 39, no. 1: e14706. 10.1111/dme.14706. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Mattishent, K. , and Loke Y. K.. 2021. “Meta‐Analysis: Association Between Hypoglycemia and Serious Adverse Events in Older Patients Treated With Glucose‐Lowering Agents.” Frontiers in Endocrinology 12: 571568. 10.3389/fendo.2021.571568. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Naser, A. Y. , Wong I. C. K., Whittlesea C., et al. 2019. “Attitudes and Perceptions Towards Hypoglycaemia in Patients With Diabetes Mellitus: A Multinational Cross‐Sectional Study.” PLoS One 14, no. 10: e0222275. 10.1371/journal.pone.0222275. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Palani, G. , Stortz E., and Moheet A.. 2023. “Clinical Presentation and Diagnostic Approach to Hypoglycemia in Adults Without Diabetes Mellitus.” Endocrine Practice: Official Journal of the American College of Endocrinology and the American Association of Clinical Endocrinologists 29, no. 4: 286–294. 10.1016/j.eprac.2022.11.010. [DOI] [PubMed] [Google Scholar]
- Racca, C. , Bouman E. J., Van Beers C. A. J., Smits M. M., van Raalte D. H., and Serné E. H.. 2022. “Association Between Hypoglycaemic Glucose Variability and Autonomic Function in type1 Diabetes With Impaired Hypoglycaemia Awareness.” Diabetes Research and Clinical Practice 189: 109964. 10.1016/j.diabres.2022.109964. [DOI] [PubMed] [Google Scholar]
- Ratzki‐Leewing, A. , Black J. E., Kahkoska A. R., et al. 2023. “Severe (Level 3) Hypoglycaemia Occurrence in a Real‐World Cohort of Adults With Type 1 or 2 Diabetes Mellitus (iNPHORM, United States).” Diabetes, Obesity & Metabolism 25, no. 12: 3736–3747. 10.1111/dom.15268. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ren, X. , Tian Z., and Li H.. 2024. “Correlation Analysis of Hypoglycemic Fear and Psychological Well‐Being and Family Support Function in Older Adults Diabetic Patients.” Diabetes, Metabolic Syndrome and Obesity: Targets and Therapy 17: 4501–4509. 10.2147/DMSO.S486509. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Tătaru, I. , Dragostin O. M., Fulga I., et al. 2022. “The Modern Pharmacological Approach to Diabetes: Innovative Methods of Monitoring and Insulin Treatment.” Expert Review of Medical Devices 19, no. 7: 581–589. 10.1080/17434440.2022.2113387. [DOI] [PubMed] [Google Scholar]
- Tini, P. , Rubino G., Pastina P., et al. 2024. “Challenges and Opportunities in Accessing Surgery for Glioblastoma in Low‐Middle Income Countries: A Narrative Review.” Cancers 16, no. 16: 2870. 10.3390/cancers16162870. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Toobert, D. J. , and Glasgow R. E.. 1991. “Problem Solving and Diabetes Self‐Care.” Journal of Behavioral Medicine 14, no. 1: 71–86. 10.1007/BF00844769. [DOI] [PubMed] [Google Scholar]
- Turnbull, A. E. , Lee E. M., Dinglas V. D., et al. 2023. “Health Expectations and Quality of Life After Acute Respiratory Failure: A Multicenter Prospective Cohort Study.” Chest 164, no. 1: 114–123. 10.1016/j.chest.2023.01.016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- van Husen, G. , Burger T. J., de Koning M. B., de Wit M. A. S., Segeren M. W., and Beekman A. T. F.. 2025. “Needs of the Network: A Qualitative Study of the Needs of Family Members, Partners and Close Friends of People With a Severe Mental Illness (SMI).” BMC Psychiatry 25, no. 1: 220. 10.1186/s12888-025-06607-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wang, J. , Zgibor J., Matthews J. T., Charron‐Prochownik D., Sereika S. M., and Siminerio L.. 2012. “Self‐Monitoring of Blood Glucose Is Associated With Problem‐Solving Skills in Hyperglycemia and Hypoglycemia.” Diabetes Educator 38, no. 2: 207–218. 10.1177/0145721712440331. [DOI] [PubMed] [Google Scholar]
- Worku, Y. B. , Tekle M. T., Bekalu A. F., and Simegn M. B.. 2025. “Self‐Reported Hypoglycemia and Associated Factors Among Patients Living With T1D s at University of Gondar Comprehensive Specialized Hospital, Northwest, Ethiopia: A Cross‐Sectional Study.” Frontiers in Clinical Diabetes and Healthcare 6: 1320610. 10.3389/fcdhc.2025.1320610. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wu, F.‐L. , Juang J.‐H., and Lin C.‐H.. 2016. “Development and Validation of the Hypoglycaemia Problem‐Solving Scale for People With Diabetes Mellitus.” Journal of International Medical Research 44, no. 3: 592–604. 10.1177/0300060516636752. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Wu, F.‐L. , Wu E.‐C., Chang Y.‐C., Hu W.‐Y., Juang J.‐H., and Yeh M. C.. 2018. “Factors Affecting the Ability of People With Diabetes to Avoid Hypoglycemia.” Journal of Nursing Research: JNR 26, no. 1: 44–51. 10.1097/JNR.0000000000000198. [DOI] [PubMed] [Google Scholar]
- Xu, H.‐M. , Zhai Y.‐P., Zhu W.‐J., et al. 2024. “Application of Hospital‐Community‐Home Linkage Management Model in Patients With Type 2 Diabetic Nephropathy.” Journal of Health, Population, and Nutrition 43, no. 1: 40. 10.1186/s41043-024-00521-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Yan, M. , Yu Y., Li S., Zhang P., and Yu J.. 2024. “Effectiveness of King's Theory of Goal Attainment in Blood Glucose Management for Newly Diagnosed Patients With Type 2 Diabetes: Randomized Controlled Trial.” Journal of Medical Internet Research 26: e59142. 10.2196/59142. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Zimet, G. D. , Dahlem N. W., Zimet S. G., and Farley G. K.. 1988. “The Multidimensional Scale of Perceived Social Support.” Journal of Personality Assessment 52, no. 1: 30–41. 10.1207/s15327752jpa5201_2. [DOI] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data S1: jocn70336‐sup‐0001‐supinfo01.docx.
Figure S1: Flowchart illustrating participant recruitment, follow‐up status and reasons for loss to follow‐up.
Figure S2: Dynamic change trajectories of scores for each dimension of hypoglycaemia problem‐ solving ability in patients with DM over the follow‐up period.
Table S1: Comparison of baseline characteristics between DM patients with complete follow‐up and those lost to follow‐up.
Table S2: Presents the changes in the total score and sub‐dimension scores of hypoglycaemia problem‐solving ability in DM patients at different time points.
Table S3: Fitting indices and goodness‐of‐fit evaluations of the LCGM for hypoglycaemia problem‐solving ability in DM patients.
Table S4: Parameter estimation results, statistical significance tests and covariance calibration of the LCGM for hypoglycaemia problem‐solving ability in DM patients.
Data Availability Statement
The data sets analysed in the current study are available from the corresponding authors upon reasonable request.