Abstract
Introduction:
Chronic disease management programs (CDMPs), especially those incorporating health coaching, play a crucial role in streamlining and improving diabetes care. This study investigated the effectiveness of a health coaching program, comparing face-to-face sessions with telephone follow-ups against telephone-only coaching over a 12-month period. The goal was to evaluate changes in diabetes knowledge, self-reported health status, diabetes distress, body mass index (BMI), and glycemic control among participants.
Methods:
In 2013, patients with diabetes at Royal North Shore Hospital in Sydney, Australia, enrolled in a health coaching program. Data was collected at baseline, 3, 6, and 12 months using questionnaires, comparing results to baseline measurements. Glycemic control (HbA1c) and BMI were assessed at baseline and 12 months.
Results:
A total of 238 patients participated, with 178 in the face-to-face CDMP group (with telephone follow-up) and 60 in the telephone-only coaching group. While BMI remained unchanged in both groups, significant improvements were observed in HbA1c levels for patients with elevated baseline levels (>7%), decreasing from 8.5% (SD, 1.0%) to 7.9% (SD, 1.0%) (P = .03). Patients reporting the poorest health status at the start of the program showed improvement, increasing their score from 4.4 (SD, 0.5) to 3.7 (SD, 0.9) (P = .001). Overall diabetes knowledge also improved across all participants (24.4 [SD, 2.4] to 25.2 [SD, 2.4]; P < .001), and diabetes distress decreased in those with high baseline distress (3.8 [SD, 0.6] to 3.2 [SD, 0.7] at 6 months, P = .04 and to 3.0 [SD, 0.4] at 12 months, P = .003).
Conclusion:
Diabetes health coaching programs are effective tools for enhancing glycemic control and alleviating diabetes distress, particularly for patients who present with higher initial levels of distress or poorer glycemic management. These findings highlight the potential of tailored programs in improving diabetes home care and overall disease management.
Introduction
Diabetes mellitus is a pervasive chronic condition that significantly contributes to morbidity and mortality worldwide (1). Living with diabetes often entails a diminished quality of life and considerable psychological distress (2,3). This “diabetes distress”—the emotional burden associated with managing the disease—along with reduced overall well-being and insufficient diabetes knowledge, can negatively impact self-management practices and blood sugar control. This, in turn, elevates the risk of developing serious complications (4,5). Although the impact of diabetes on a patient’s well-being is not always systematically evaluated in clinical settings, understanding these factors is crucial for developing personalized and effective diabetes management strategies.
Patient education is a cornerstone of effective diabetes care. Programs designed to empower individuals with self-management skills are proven to be effective in fostering patient engagement and improving health outcomes (6,7). Research has demonstrated that participation in structured diabetes education sessions can enhance disease-specific knowledge and problem-solving capabilities over time (8). Similarly, diabetes self-management education programs have been shown to improve quality of life and reduce diabetes-related distress in both type 1 and type 2 diabetes populations (9,10). Health coaching, which offers a personalized educational approach focusing on problem-solving and goal setting, emerges as a valuable complement to standard diabetes education. It holds promise in further improving glycemic control and enhancing self-reported health status (11,12), especially in the context of a diabetes home care disease management program.
In 2013, Royal North Shore Hospital (RNSH) in Sydney introduced a chronic disease management program (CDMP) that integrated health coaching. This innovative approach differed from traditional group education by prioritizing patient-defined goals and targets. This study aimed to assess the impact of this program on key patient outcomes over a 12-month period. The outcomes included changes in self-reported health status, diabetes distress, diabetes knowledge, BMI, and glycemic control. The study considered both face-to-face health coaching combined with telephone support and telephone health coaching alone. A key objective was to identify patient profiles that would benefit most significantly from such programs. The hypothesis was that understanding these profiles could lead to more targeted and effective enrollment strategies, ultimately optimizing program impact and resource allocation in diabetes home care disease management programs.
Methods
Diabetes Chronic Disease Management Program
The health coaching initiative at RNSH’s Department of Diabetes, Endocrinology and Metabolism was supported by the Northern Sydney Local Health District CDMP. This broader CDMP is designed to support individuals with specific chronic conditions—including diabetes, coronary artery disease, chronic obstructive pulmonary disease, congestive heart failure, and hypertension—who are at increased risk of hospital readmission (13). The program’s services encompass facilitating access to and coordination of healthcare appointments and providing extra support to enhance diabetes self-management, with the overarching goals of reducing complications, improving overall health, and preventing hospitalizations (14). All patients enrolled in the RNSH CDMP were invited to participate in the health coaching component, with slightly over half (53%) choosing to participate in 2013. Eligibility criteria for the health coaching program included being English-speaking, aged 16 years or older, and diagnosed with type 1 or type 2 diabetes.
Diabetes Health Coaching
The RNSH diabetes health coaching program offered two modalities: a face-to-face group education session, known as the “Empowerment Program,” followed by 12 months of telephone calls from a dedicated health coach (a diabetes nurse educator); or, for patients unable to attend in-person sessions, a telephone-only coaching option. The telephone-only option commenced with an initial educational call and continued with follow-up calls from the health coach for 12 months. The frequency of telephone contact was tailored to patient preference, ranging from weekly to three times per month, with monthly calls being the most common arrangement.
The diabetes conversation map served as a guiding tool for both the initial face-to-face and telephone sessions, primarily focusing on managing diabetes complications and risk factors. Patients received education and guidance on healthy eating habits, recommended physical activity levels, and strategies for preventing diabetes complications. Furthermore, the program addressed diabetes-specific health targets and recommended timelines for primary care provider visits to review progress and conduct blood glucose testing. Patients were actively encouraged to set personal goals related to their diabetes management, such as daily exercise duration, healthy eating goals, or medication adherence. The health coach provided support in developing actionable strategies to achieve these self-identified goals. Subsequent telephone calls involved patients reporting on their progress, evaluating the effectiveness of their strategies, and adjusting their plans as needed.
CDMP Health Coaching Patient Assessments
To assess participants’ general diabetes knowledge, the validated Diabetes, Hypertension and Hyperlipidemia (DHL) knowledge instrument was utilized (15). This tool, previously validated in a Malaysian diabetes population (15), assesses understanding of glucose, blood pressure, and lipid control in mitigating diabetes complication risks. The DHL questionnaire comprises 28 questions, with each correct answer scoring one point, for a maximum score of 28. Self-reported health status was evaluated using the first question of the Short-Form 36 Quality of Life Instrument (SF-1) (16,17). A score of 4 or higher (maximum 6) on the SF-1 indicates poorer health status (17). Diabetes distress was measured using the Diabetes Distress Scale (DDS) (18,19). The DDS generates an overall score (maximum 6) from 17 questions, each rated on a 1 to 6 scale; a score of 3 or higher signifies moderate to severe diabetes distress (20). The DHL, SF-1, and DDS questionnaires were administered at four points: baseline, 3, 6, and 12 months. These instruments are validated for use in diabetes populations (16–20).
Figure 1. Tracking Diabetes Knowledge Improvement Through Health Coaching: This graph illustrates the positive changes in general diabetes knowledge among participants in a health coaching program, as measured by the DHL knowledge instrument. It compares the scores over time for all patients and those who consistently completed assessments, highlighting the program’s effectiveness in enhancing understanding of diabetes management.
Glycemic Control, BMI, and Medical History
All patients entering the Diabetes CDMP at RNSH had their HbA1c measured at baseline (just before health coaching referral) and again at their routine 12-month diabetes complication screening appointment. Weight and height measurements were also taken at baseline and 12 months to calculate BMI (weight in kilograms divided by height in meters squared). Medical records provided data on dyslipidemia, hypertension, and diabetes complications (retinopathy, neuropathy, or chronic kidney disease, defined clinically as an estimated glomerular filtration rate [eGFR] <60 mL/min/1.73 m2).
Study Design
A longitudinal audit of prospectively collected data was conducted to evaluate the impact of a diabetes health coaching program on patient outcomes over 12 months. Participants resided within the RNSH catchment area in the Northern Sydney Local Health District, a region characterized by high socioeconomic status (21). The study received ethical approval from the Northern Sydney Local Health District Human Research Ethics Committee (reference no. RESP/16/58).
Statistical Analyses
Changes in patient-reported diabetes knowledge, health status, and diabetes distress over time were analyzed using Student’s paired t-tests for parametric data and Wilcoxon matched-pairs signed-rank tests for non-parametric data. For repeated measures, one-way ANOVA was used for parametric data, and the Friedman test for non-parametric data. Group differences were assessed using Student’s unpaired t-tests or Mann–Whitney tests as appropriate. Changes in HbA1c and BMI were evaluated using paired t-tests. Statistical analysis was performed using GraphPad Prism Version 6, with statistical significance set at P < .05.
Results
Patient Demographics and Characteristics
Between January and December 2013, 178 patients engaged in face-to-face health coaching with telephone support, and 60 patients received telephone-only health coaching (Table 1). The majority (97.1%) of the total cohort had type 2 diabetes, and most participants were over 65 years old. Approximately two-thirds of participants in both groups were female.
Comorbidity rates were similar across both groups, with hypertension (69.3%) and dyslipidemia (91.2%) being the most prevalent. No significant differences in age or diabetes duration were observed between the groups (Table 1). However, the telephone-only health coaching group presented with higher baseline HbA1c (7.3%; P = .03) and BMI (31.1; P = .04) and were more likely to be using insulin (46.7%; P = .001) (Table 1).
Complete assessment data across all time points was available for 31.1% of participants (74 out of 238). Patients who did not complete all assessments were significantly younger (mean, 67.0 years; SD, 10.0 years) than those who did (71.3 years [SD, 8.7 years]; P = .003) and had a higher baseline BMI (30.2 [SD, 5.3]) compared to those who completed all assessments (28.8 [SD, 5.6]; P = .02). However, there were no significant differences in baseline diabetes knowledge, self-reported health status, or diabetes distress between these groups. Evaluable data (baseline and 12-month data) was available for 50.4% of the total patient cohort (120/238).
General Knowledge of Diabetes
The average baseline diabetes knowledge score for all participants completing the baseline assessment (n = 212) was 24.4 (SD, 2.4). Among participants who also completed the 12-month assessment, diabetes knowledge significantly improved from a mean score of 24.4 to 25.2 (SD, 2.2) (P < .001) (Figure 1A). Similar improvements were observed when considering only patients with complete assessments at all four time points (Figure 1B; P = .003).
The baseline diabetes knowledge scores were similar between the face-to-face and telephone-only coaching groups (mean score, 24.3 [SD, 2.5] vs 25.1 [SD, 1.8]; P = .10). However, patients in the face-to-face group showed significant improvement in diabetes knowledge at 12 months (mean score, 24.3 [SD, 2.5] vs 25.4 [SD, 2.4]; P < .001), while the telephone-only group did not (mean score, 25.1 [SD, 1.8] vs 24.9 [SD, 2.1]; P = .66).
Self-Reported Health Status
At baseline, 27.3% of patients reported poor health status (SF-1 score ≥4). Patients with diabetes complications (retinopathy, neuropathy, chronic kidney disease) were more likely to report lower health status at baseline compared to those without complications (3.2 [SD, 1.0] vs 2.9 [SD, 1.0]; P = .01). Among patients with poor baseline health status (n = 65) who completed at least two assessments (n = 51), a significant improvement in health status was observed for 36 patients, increasing from 4.4 [SD, 0.6] at baseline to 3.6 [SD, 1.1] at 6 months (P < .001) and 3.7 [SD, 0.9] at 12 months (P = .001) (Figure 2A). Baseline health status scores were similar for patients taking insulin and those on oral antihyperglycemic medications (3.2 [SD, 1.0] vs 3.0 [SD 1.0]). No significant differences in self-reported health status were found between the face-to-face and telephone-only coaching groups at any time point.
Figure 2. Improvements in Well-being and Reduced Distress Through Diabetes Health Coaching: This figure demonstrates the positive impact of health coaching on self-reported health status and diabetes distress. Panel A shows the improvement in health status among patients with initially poor health, while Panel B illustrates the reduction in diabetes distress for those who started with moderate to high distress levels. These results underscore the program’s benefits for emotional and physical well-being.
Diabetes Distress
The majority of participants (91.6%) had low diabetes distress at baseline (DDS score < 3). Among the subset of patients with moderate to high diabetes distress at baseline (DDS score > 3) who completed at least two assessments (n = 17), a small but significant decrease in diabetes distress was observed at 6 months for 15 patients (DDS score, 3.2 [SD, 0.7] vs 3.8 [SD, 0.6]; P = .04) (Figure 2B) and at 12 months for 10 patients (3.0 [SD, 0.4] vs 3.8 [SD, 0.6]; P = .003) (Figure 2B) compared to baseline. Diabetes distress levels were similar across both coaching methods (face-to-face or telephone only) at all time points.
Glycemic Control and Body Mass Index
The average baseline HbA1c was 7.0% (SD, 1.1%), with 194 (81.5%) patients at or below the recommended target (≤7%). For the subgroup of patients with baseline HbA1c above target (>7%; n = 44), a significant improvement was observed at 12 months, with HbA1c decreasing from 8.5% (SD, 1.0%) to 7.9% (SD, 1.0%) (P = .03). Of these patients with poor baseline glycemic control, 20.05% achieved their target HbA1c level at 12 months. No significant difference in HbA1c improvement was found between the coaching methods (Table 1).
No significant changes in BMI were observed at 12 months compared to baseline in either the face-to-face or telephone-only coaching groups. Similarly, patients with significant obesity at baseline (BMI ≥35; n = 29) showed no BMI change at 12 months. The coaching method did not influence BMI changes over time (Table 1).
Discussion
This study’s findings highlight the patient profiles that derive the most benefit from a 12-month diabetes health coaching CDMP in terms of reduced diabetes distress and improved glycemic control. Patients who initially presented with high diabetes distress, poor self-reported health status, and limited diabetes knowledge showed the most substantial improvements in these areas at both 6 and 12 months post-CDMP initiation. Face-to-face health coaching within a CDMP framework led to significantly greater gains in general diabetes knowledge compared to telephone-only coaching. Furthermore, individuals with baseline HbA1c levels exceeding the target demonstrated significant improvements in glycemic control.
Identifying patients likely to benefit most from CDMP or health coaching participation is clinically vital. Programs can be strategically tailored to these specific populations, and pre-enrollment screening can be implemented to maximize outcomes and optimize healthcare resource utilization. This targeted approach directly benefits patients and ensures efficient healthcare service delivery. Moreover, interventions can be further customized to address the specific areas of distress or educational needs reported by patients. Therefore, this analysis focused on patients with low baseline health status, elevated diabetes distress, or above-target HbA1c levels. Health coaching represents a patient-centered approach to chronic disease management, empowering individuals to define their own health goals, contrasting with the traditional provider-directed goal-setting model. This patient-driven approach fosters a greater sense of control over their health, potentially contributing to the observed reductions in psychological stress, improved perceived health status, and decreased diabetes distress.
These results align with Beverley et al.’s findings, which indicated that older patients (60–75 years) experienced positive changes in quality of life and diabetes distress following diabetes education in behavioral interventions (22). Similarly, a recent study demonstrated that health coaching for type 2 diabetes patients improved life satisfaction and reduced depressive symptoms (23). A meta-analysis further corroborated that self-management and education interventions for diabetes were most effective in patients with pre-existing depression symptoms or high stress levels (24), mirroring our study’s conclusions. Contrary to some prior studies (7,25), this research did not find overall glycemic control improvement across the entire cohort. A likely explanation is that a significant majority (81.5%) of participants had already achieved target HbA1c levels before program commencement and maintained this control throughout the 12-month follow-up, despite the anticipated natural decline in β-cell function over time. However, for patients with suboptimal baseline glycemic control, health coaching led to significant improvement, underscoring its substantial benefit for those with inadequately managed blood glucose levels.
Diabetes knowledge and attitudes are direct determinants of self-management behaviors, which, in turn, impact quality of life (26). Consistent with this, improvements in both diabetes knowledge and diabetes distress were observed, suggesting that enhanced understanding of diabetes can alleviate associated stress and facilitate easier disease management for patients. Furthermore, participant feedback regarding the program and the health coaching approach was overwhelmingly positive, highlighting its perceived benefits.
The study population exhibited relatively high levels of diabetes education upon program entry. Consequently, only a small proportion presented with high diabetes distress or poor glycemic control at baseline. This resulted in observed improvements primarily within a subset of the overall participant cohort. However, the distress reduction in this subgroup was significant and sustained at both 6 and 12 months, suggesting a lasting positive impact. Conversely, the observed improvement in self-reported health status over time may be influenced by response bias, as patients repeatedly completed the same questionnaire. Additionally, the patient-determined frequency of follow-up calls may have influenced outcomes, with more frequent contact potentially leading to greater improvements. However, this aspect could not be assessed within this study.
The study’s strengths include the large participant sample size and the longitudinal data collection integrated into the health coaching program’s standard patient assessments. Limitations include a notable attrition rate in questionnaire completion, potentially introducing response bias. Furthermore, the absence of a control group without health coaching and the non-randomized study design are limitations. Nonetheless, each participant served as their own control, with changes in questionnaire results, BMI, and glycemic control measured against their individual baselines. Potential sampling bias also exists, as participants may not fully represent the broader population of individuals with chronic conditions. However, consecutive recruitment of agreeable patients after CDMP enrollment and the representativeness of patient characteristics compared to other Australian diabetes cohorts (27,28) mitigate this concern.
In conclusion, patients entering health coaching with high diabetes distress or poor glycemic control were the primary beneficiaries of the program. Based on these findings, it is recommended that such patients be proactively offered additional health coaching or more specialized chronic disease management, rather than universally applying interventions that may not be equally relevant or effective for all patient populations. Personalized medicine is increasingly recognized as a crucial strategy for healthcare delivery, and identifying patient subgroups most responsive to specific educational approaches, like A Diabetes Home Care Disease Management Program, contributes to optimizing healthcare resource allocation and enhancing patient outcomes.
Acknowledgments
Grace Delaney and Neroli Newlyn contributed equally to this article. The authors declare no funding was received for this research and have no conflicts of interest.
Author Information
Corresponding Author: Rachel McGrath, Department of Diabetes, Endocrinology and Metabolism, Level 3, Acute Services Building, Royal North Shore Hospital, St Leonards, Sydney, NSW 2065, Australia. Telephone: +61 (2) 9463 1470. Email: [email protected].
Author Affiliations: 1Department of Diabetes, Endocrinology & Metabolism, Royal North Shore Hospital, St Leonards, Sydney, NSW 2065, Australia. 2University of Sydney, Northern Clinical School, Royal North Shore Hospital, St Leonards, NSW 2065, Australia. 3Charles Perkins Centre, University of Sydney, Australia. 4Kolling Institute of Medical Research, University of Sydney, Royal North Shore Hospital, St Leonards, NSW 2065, Australia.
References
[1] World Health Organization. Diabetes [Internet]. [cited 2016 Nov 17]. Available from: http://www.who.int/mediacentre/factsheets/fs312/en/.
[2] Peyrot M, Rubin RR, Kruger DF, Travis LB. Correlates of diabetes distress in people with type 1 and type 2 diabetes. J Diabete
