The global population with prediabetes is increasing annually. The International Diabetes Federation has stated that 318 million people worldwide were estimated to have prediabetes in 2015, and this number is expected to reach 482 million by 2040 [1]. Prediabetes is a condition in which an individual's blood glucose exceeds the normal range, but he or she has not been diagnosed with diabetes. The criteria for a prediabetes diagnosis are fasting blood glucose levels between 100 and 125 mg/dl (known as impaired fasting glucose), 2-h postprandial plasma glucose between 140 and 199 mg/dl (known as impaired glucose tolerance), or hemoglobin A1c (HbA1c) between 5.7% and 6.5% [1]. Prediabetes is associated with a high risk of developing diabetes. If no active preventive approach is adopted during the prediabetic stage, prediabetes usually progresses to type 2 diabetes within an average of 3.6 years [2]. In one study, it was shown that annually in Taiwan, 3.2% of prediabetic individuals (PDIs) develop type 2 diabetes [3]. Prediabetes is also associated with increased risk of cardiovascular disease (14.8%), early stage nephropathy (17.7%), and retinopathy (7.9%) [4, 5, 6, 7]. Diabetes has a negative impact on patients and requires complicated, lifelong medical care. Diabetes not only impacts the patients but also their families and their associations within society.

For PDIs, lifestyle modification is the easiest way to prevent or delay the onset of diabetes. Studies have shown that healthy eating habits, regular exercise, and weight control lower the incidence of developing diabetes among prediabetics [8, 9, 10, 11]. Specifically, engaging in regular exercise is the most effective intervention for preventing diabetes. One study showed that while adopting a healthy diet can reduce diabetes incidence by 31.0%, regular exercise further reduces the incidence by 46.0% [12]. Another study indicated that PDIs engaging in regular exercise can reduce the risk of developing diabetes by between 30.0 and 69.0% compared to those who do not exercise [9, 10, 12]. Therefore, increasing physical activity among individuals with prediabetes should be prioritized to prevent the onset of diabetes.

Physical activity is associated with significant reductions in fasting glucose and HbA1c [13]. Physical activity is noted for: 1) reduced blood pressure, 2) reduced overall body fat, 3) reduced low-density lipoprotein cholesterol, 4) increased insulin sensitivity, 5) improved muscle strength, and 6) increased high-density lipoprotein cholesterol [14, 15]. Unfortunately, exercise is suboptimal among individuals with prediabetes in Taiwan for cultural reasons. Fewer than one-third of PDIs exercise routinely, and most of those who exercise do not do so for 150 minutes per week, which is the goal for effective diabetes prevention [16]. Furthermore, most of those who begin exercising after they recognize the importance of physical activity maintain their exercise routines for no more than six weeks [17]. Therefore, it is a major challenge for individuals with prediabetes to maintain consistent, lifelong exercise habits.

Based on previous qualitative research using grounded theory, the psychological status of PDIs' willingness to engage in exercise was modeled [18]. The Kuo model contains four stages: 1) developing awareness (DA) stage, 2) creating the health blueprint (CHB) stage, 3) action cycle of internal struggle (ACIS) stage, and 4) spontaneous regular exercise (SRE) stage. Furthermore, the ACIS stage contains two opposing forces: driving force (DF) and resistive force (RF) (Figure 1). Correctly assessing these two conflicting forces is critical for determining whether PDIs will enter the SRE stage.

Figure 1
Kuo model of exercise-engagement process for pre-diabetic individuals [18].
H1=hypothesis 1; H2=hypothesis 2; H3=hypothesis 3; H4=hypothesis 4; H5=hypothesis 5.

• Download Figure

According to the Kuo model, the DA stage is the key point in time at which PDIs may be strongly influenced to begin exercising. After progressing through the DA stage, PDIs enter the CHB stage and focus on self-determination with regard to their exercise goals. However, in the process, all such individuals must progress through the ACIS stage. When the self-discipline, relationship supervision, and health feedback of the DF are more powerful than the RF, the exercise behavior is maintained. In contrast, if the internal and external resistance of the RF are stronger than the DF, exercise behavior is interrupted. The RF and DF work against each other until the individual breaks the action cycle to enter the SRE stage. In this model, the stages of PDIs' exercise-engagement process are closely interlinked and interrelated.

The objective of the present study was to develop a scale and validate the model for the process of exercise engagement in PDIs from a quantitative perspective, as seen in Figure 1 [18]. Currently, several exercise behavior assessment scales such as the Exercise Self-Efficacy Scale (ESES) [19], the Sport Motivation Scale (SMS) [20], and the Trans-theoretical Model (TTM) [21] are available. Using the mentioned scales and theories to validate the model of the process of exercise engagement for PDIs has several constraints: 1) none of these theories or scales conduct investigations from the PDIs' perspective; 2) none of the scales cover the complete exercise-engagement process; and 3) none of the scales evaluate PDIs' psychological or emotional status at each stage of the exercise-engagement process.

The TTM and the Kuo model are fundamentally similar, while differences remain to be understood within the context of the disease as the region of interest and control from the perspective of the PDI. Behavioral modification from the individual's perspective to benefit from making specific changes is similar but slightly different between the TTM and the Kuo model. The Kuo model is specific for understanding and developing a predictive methodology to assess PDIs and their capacity to alter long-term health outcomes. Presently, scales that are suitable for investigating PDIs' exercise-engagement process remain lacking. To effectively resolve this shortcoming, this study developed a scale consisting of five subscales based on the five subcategories of the Kuo model (Figure 1). The scale was subsequently used to collect responses from PDIs to evaluate the interrelationships between DA, CHB, DF, RF, and SRE based on the following hypotheses:

Hypothesis 1 (H1): A relationship exists between DA and CHB, where DA will have a positive effect on CHB.

Hypothesis 2 (H2): A relationship exists between CHB and DF, where CHB will have a positive effect on DF.

Hypothesis 3 (H3) and Hypothesis 4 (H4): An inner conflict-based relationship exists between DF and RF, where the magnitudes of DF and RF may positively or negatively affect each other.

Hypothesis 5 (H5): A direct relationship exists between ACIS and SRE.

This study had two objectives: 1) to develop a scale for assessing the process of exercise engagement (SPEE) for PDIs, and 2) to validate the Kuo model for the process of exercise engagement for PDIs.

The development of the SPEE for PDIs provides the medical profession with a convenient instrument to assess the psychological probability that a PDI will implement exercise. The SPEE is a 5-point scale containing 21 items, comprising DA (4 items), CHB (3 items), DF (6 items), RF (4 items), and SRE (4 items) (Appendix 3). This scale is the first exercise-engagement scale designed for PDIs and the first to assess the psychological milieu and behavioral outcome at each stage of the exercise-engagement process. The development process of the SPEE differs from other scales in several key manners: The hypoglycemia problem-solving scale developed by Wu et al. [31] for people with diabetes mellitus was constructed based on the original SPSI and its Chinese version, unlike the SPEE, which relied on the Kuo model foundation. Ebrahimi et al. [32] developed the Diabetes Adjustment Assessment Scale (DAAS) based on their qualitative study but without a conceptual model infrastructure. The DAAS is focused on patients diagnosed with type 2 diabetes and their adjustment to disease onset, not PDIs.

The DA and CHB subscales can be used to explore the levels of PDIs' self-awareness in terms of the importance of physical activity and psychological barriers to exercise. The concept is similar to Lin and Chi's SMS [20], which explores the factors that motivate an individual to engage in exercise. The main focus is interest in exercise, desire to learn a new sport, improved health, and sense of accomplishment. However, the SMS is designed for general public assessment and not for individuals with specific potential chronic health care needs. The DA subscale of the present study is designed to assess the perspective from the PDIs' point of view. Recognition of health risks is the cause or motivation, and the choice to engage in exercise is the effect, which leads to improved health and a reduction diabetes outcomes. Comparatively, the SMS adopted a broader perspective while covering more dimensions to determine the reasons that individuals engage in exercise.

The DF and RF subscales are similar to the ESES [19] in that both explore the factors that promote or hinder exercise habits among individuals. As with the SMS, the ESES focuses on the general population and not a specific at-risk subgroup. Both scales explore the counteracting effect between positive and negative factors that influence the choice to engage in exercise behavior. Further distinguishing the ESES from the SPEE, the DF and RF subscales are more effective at exploring PDIs' psychological and emotional status during the internal struggle stage. The psychological and emotional status in DF assesses willpower, self-blame, self-motivation, and consistency with regard to exercise. The SPEE is different from the ESES, where the focus explored positive motivations: 1) sense of happiness, 2) improved health, 3) enhanced muscle strength, and 4) bone density. However, the ESES failed to examine motivations to continue exercise over the long term. By comparison, the RF subscale explored: 1) denial, 2) difficulty, 3) sedentary passiveness, and 4) laziness, which are not assessed in the ESES.

The validation of the SPEE is consistent with the Kuo model of the process of exercise engagement in PDIs (Figure 1). The DA stage and its impact on PDIs is of greatest importance and strongly effects the CHB stage. Data showed a direct effect of the DA stage on the CHB stage. This is highly predictive of PDIs' potential to achieve exercise goals that will mediate or eliminate the onset of diabetes, as PDIs may choose to further educate themselves on the dangers of diabetes. The CHB stage motivates the SRE stage not only directly but also indirectly through the DF and RF stages. Furthermore, the effect of the SRE stage on the DF stage is positive, while it is negative on the RF stage. PDIs are able to empower themselves through education via the hospital or other means, which explains the effects on the DF and RF. The DF and RF stages are negatively and significantly correlated. For PDIs, the more acute the DA stage, the greater their motivation is for engaging in exercise and the smoother their entrance is into the CHB stage. DA stage acuteness does not influence the DF; as such, it may be proposed that during the DA stage, the type of education is extremely important. Once PDIs have crossed the CHB threshold and engaged in exercise, they enter the internal struggle cycle between DF and RF. This influences their exercise behaviors, which enhance or exacerbate their probability for developing diabetes. During the exercise-engagement process, DF and RF counteract each other; however, when a higher DF is present, PDIs more readily enter the SRE stage. In essence, PDIs must first reduce their RF to successfully enter the SRE stage. Therefore, although the RF is a mediating variable in this model, it is critical for determining whether PDIs will develop long-lasting exercise habits.

The findings of this study suggest that whether PDIs initially engage in exercise depends on the impact of the DA stage and, subsequently, whether exercise habits are established. Attaining the SRE stage is dependent to the extent that the RF is reduced, as increasing the DF is not useful. Providing PDIs with physical activity-related health education with the purpose of motivating regular exercise behaviors is crucial. The priority is to increase their awareness of the importance of exercise, followed with solutions and supporting strategies to eliminate factors that prevent regular exercise. In doing so, it is possible to reduce the effect of the RF.

The single difference between the study results and the Kuo model (Figure 1) is the direct and significant (β=.60) effect of the CHB stage on the SRE stage. As PDIs enter the CHB stage, they may enter the SRE stage directly, bypassing the DF and RF stages. The findings of this study strongly support the Kuo model and are consistent with Self-Determination Theory. This theory states that human behaviors are generated from intrinsic motivation, and the expression of these behaviors is subject to the influence of extrinsic factors [33]. Therefore, when providing relevant clinical health education, medical staff will best serve PDIs by developing awareness, specifically by helping PDIs to plan health blueprints. If PDIs perceive a positive outcome for behavioral change, the exercise intervention may be more effective.

Several factors may influence the transferability of the SPEE and the findings of this study. Due to time constraints, this study could not assess the long-term benefits of exercise and the predictive capacity of the SPEE. As a result of limited research budget and time, the research was confined to one geographic location. Another confounding factor is that the level of participants' education was consistently high, as the government of Taiwan has an enforced state education policy that has been in existence for the last 50 years. It is known that level of education influences behavioral modification.

In order to enhance the transferability of the SPEE, further studies should focus on: 1) psychological status of exercise for PDIs, 2) behavioral expression at various stages of the exercise-engagement process, and 3) contributing factors to the formation of long-term exercise habits. This research would have benefited greatly with the inclusion of a quantitative approach to assess changes in glucose levels as participants became engaged in exercise to reduce the long-term liability of developing diabetes.

Although the sample size of 110 in EFA of the current study met the minimum requirement suggested by Gorsuch [23] and Thompson [24]. However, the sample size of 235 based on the rule of five subjects for each item is the more preferable guideline for scale development [22]. Lack of enough power due to less sample size might be a limitation of the current study. Therefore, a further study with larger sample is suggested to conduct for providing stronger evidence to validate the developed scale and model.

This study is the first to develop a scale that includes the five subscales of DA, CHB, DF, RF, and SRE for assessing the psychological transition and behavioral expression at each stage of the process of exercise engagement for PDIs. The Kuo model for the process of exercise engagement of PDIs is valid and has a goodness of fit. Higher scores on the DA, CHB, DF, and SRE for people with prediabetes provides deeper insights into the behavioral change stages required to initiate extended long-term health care outcomes in order to avoid developing diabetes. These insights are significant as they allow for patient-specific mapping of their DF and RF and the factors affecting their transition to the SRE stage. The use of SPEE within the clinical setting may prove useful in developing tailored exercise strategies that may lead to a reduction in the rate of diabetes onset in PDIs.