Exploring a Need for a Cardiometabolic Disease Staging System as a Computerized Clinical Decision Support Tool: Qualitative Study

Background Although cardiometabolic diseases are leading causes of morbidity and mortality in the United States, computerized tools for risk assessment of cardiometabolic disease are rarely integral components of primary care practice. Embedding cardiometabolic disease staging systems (CMDS) into computerized clinical decision support systems (CDSS) may assist with identifying and treating patients at greatest risk for developing cardiometabolic disease. Objective This study aimed to explore the current approach to medical management of obesity and the need for CMDS designed to aid medical management of people living with obesity, at risk of being obese, or diabetic at the point of care. Methods Using a general inductive approach, this qualitative research study was guided by an interpretive epistemology. The method included semistructured, in-depth interviews with primary care providers (PCPs) from university-based community health clinics. The literature informed the interview protocol and included questions on PCPs’ experiences and the need for a tool to improve their ability to manage and prevent complications from overweight and obesity. Results PCPs (N=10) described their current approaches and emphasized behavioral treatments consisting of combined diet, physical activity, and behavior therapy as the first line of treatment for people who were overweight or obese. Results suggest that beneficial features of CDSS include (1) clinically relevant and customizable support, (2) provision of a comprehensive medical summary with trends, (3) availability of patient education materials and community resources, and (4) simplicity and ease of navigation. Conclusions Implementation of a CMDS via a CDSS could enable PCPs to conduct comprehensive cardiometabolic disease risk assessments, supporting clinical management of overweight, obesity, and diabetes. Results from this study provide unique insights to developers and researchers by identifying areas for design optimization, improved end user experience, and successful adoption of the CDSS.


Introduction
Cardiometabolic diseases are leading causes of morbidity and mortality in the United States, including a wide array of diseases, typically beginning with insulin resistance and progressing later into a cluster of conditions that increase the risk of type 2 diabetes, stroke, and cardiovascular disease [1,2]. Being overweight (BMI ≥25 kg/m 2 ) is associated with double the risk of developing cardiometabolic multimorbidity, while having mild and severe obesity (BMI ≥30 kg/m 2 ) increases the risk 4 and 10 times, respectively [3]. However, current diagnostic categories that are based on standard BMI ranges defining overweight and obesity have high specificity but low sensitivity for identifying insulin resistance and cardiometabolic disease [4]. For example, with the current diagnostic categories, some individuals with overweight and obesity might not have cardiometabolic risk factors and may exhibit low rates of future diabetes and cardiovascular-related mortality; alternatively, some individuals who do not meet criteria for either metabolic syndrome or prediabetes exhibit risk of future diabetes [4]. Thus, risk assessments for cardiometabolic disease with greater sensitivity should be an integral component of medical practice, with tools to evaluate preventive and therapeutic options in patients at greatest risk for developing disease. Currently, there is no stratification of the population by level of obesity-related disease and mortality risk [5].
An estimated 42.5% of US adults aged 20 years and older are living with obesity, including 9.0% with severe obesity, and another 31.1% are overweight [6]. Because this group is at high risk of developing diabetes and other obesity-related complications, there is a need for risk stratification approaches to identify early those at highest risk and identify weight loss programs with appropriate treatment intensity. To provide appropriate medical management of obesity and facilitate the diabetes risk assessment of people with excess adiposity, a comprehensive staging system that establishes 5 stages of cardiometabolic disease risk-the cardiometabolic disease staging system (CMDS)-was developed [7,8]. This validated staging system is based on Adult Treatment Panel III metabolic syndrome risk factors and includes waist circumference, systolic and diastolic blood pressures, fasting and 2-hour blood glucose levels, triglycerides, and high-density lipoprotein cholesterol (HDL-C; Table 1) [4]. The purpose of this system is to help clinicians select treatment modality and intensity in the management of cardiometabolic diseases while balancing benefit and risk. Evidence demonstrates the CMDS has higher predictive and discriminative ability compared with other systems and relies on data typically collected during primary care visits; thus, it is more feasible to integrate into busy workflows of primary care providers (PCPs) [5]. Vigilance in the management of modifiable risk factors is critical, given that people with overweight and obesity are at increased cardiovascular risk. Primary care settings, as familiar and accessible clinical venues for patients, are well positioned to screen people with overweight and obesity and recommend appropriate weight loss treatment plans to prevent complications and weight progression. Many studies found that the largest weight losses were achieved with high-intensity counseling by PCPs and referral of interested individuals to appropriate interventions [9][10][11][12][13]. However, a study of a nationally representative sample of adults aged 35 years and older found that, despite more adults reported being screened for obesity (78.6%) and of those screened, nearly 40% had a BMI of 30 kg/m 2 or higher (39.2%), only slightly more than one-half (53.5%) of obese adults screened reported receiving counseling about weight management [14]. Furthermore, BMI is the most preferred screening tool, though literature indicates it could be a poor indicator of cardiovascular disease and overall mortality risk [7,15]. Research finds BMI is not a good index of visceral fat, which is the basis of metabolic disorders associated with increased cardiovascular risk, whereas waist circumference might be superior as a risk assessment tool [16]. PCP-indicated practice improvements, helpful in treating and managing overweight and obesity, include better tools for early identification of risk and preventive treatment for those with multiple risk factors [11].
Providing CMDS to PCPs via computerized clinical decision support systems (CDSS) may assist in stratifying the population by obesity-related disease risk and targeting those patients who are at greater risk for obesity-related complications. To the authors' knowledge, this would be the first electronic health record-integrated CDSS that would incorporate CMDS. Despite literature indicating CDSS may have a positive impact on provider performance and patient outcomes [17], evidence also indicates that CDSS rarely reach their full potential [18]. As with any innovation, user acceptance and integration within the clinical workflow are critical for successful uptake and routine use [19].
System analysis and design involve the process of planning, analyzing, designing, developing, implementing, and maintaining systems. A user-centered approach focusing on the user experience necessitates coordinated relationships between the system specialists, designers, and developers and the nonspecialists and users with outcomes knowledge. The system development life cycle, when combined with the user experience life cycle, allows for that coordination to occur and has been shown to lead to better system adoption [20]. Figure 1 illustrates our conceptual model for system analysis and design of the CMDS. This paper reports on the first 3 phases of each cycle: (1) plan and define, (2) analyze and research, and (3) design. As such, with the aim of involving users at key milestone stages of system development, this study explored the current approach to management of overweight and obesity and a need for the CMDS system at the point of care to facilitate specificity in treatment modalities.

Methods
To ensure we adhered to qualitative reporting standards, we followed the 32-time consolidated criteria for reporting qualitative studies (COREQ) checklist (Multimedia Appendix 1).

Sampling
Participant recruitment used convenience sampling where the research team coordinated with the medical director for primary care of a large academic medical center in the southeastern United States. Recruitment emails to potential candidates indicated the study purpose and invited participation. The number of participants was determined to be sufficient when saturation was reached (N=10) [21]. All participants were provided with a US $100 gift card. The present study is a foundation for ongoing research aimed at developing and implementing a CDSS based on the CMDS.

Data Collection
From August 2020 to January 2021, 10 semistructured interviews were conducted by 2 research team members, consisting of the principal investigator (TM; male) and a graduate research assistant (AK; female). Two senior female researchers (AH and SF) with training in qualitative interviewing provided guidance and supervision. The interviewers did not have prior relationships with the participants. Only the interviewers and participants were present during data collection. The duration of the interviews varied between 30 minutes and 45 minutes and were conducted via a collaborative, cloud-based videoconferencing service at a mutually agreed-upon time. Interviews were transcribed verbatim by a commercial transcription company. The interview guide was informed by the literature review and included questions designed to (1) understand how PCPs manage overweight and obesity and facilitate prevention and management of diabetes and cardiovascular disease risk during a standard primary care visit and (2) explore PCP needs for CMDS and preferences for a CDSS (Multimedia Appendix 2). Broad, open-ended questions along with permissive prompts were used to facilitate each semistructured interview. Prior to conducting interviews, the semistructured interview guide was pilot tested with several providers to ensure questions were clear, generated in-depth discussion, were acceptable to participants, and resulted in usable information. Feedback from pilot testing was used to modify the wording, content, and order of the interview questions.

Ethical Considerations
All investigations were conducted in conformity with ethical principles of research. Consent for participation and interview recording was obtained verbally before each interview. This study was determined to be exempt by the University of Alabama at Birmingham Institutional Review Board (IRB Protocol Number 300003559).

Data Analysis
Transcribed interviews were coded using an inductive thematic analysis approach with NVivo 12 Plus (QSR International, Melbourne, Australia). To increase reliability and reduce bias, all transcripts were coded by 2 team members (AK and JA) with expertise in thematic analysis [22,23]. The analysis consisted of 2 phases: codebook development and codebook refinement.
First, during open coding, coders examined an initial set of transcripts for categories (processes or events that share an attribute) of information related to our research questions. The second phase of our analysis focused on comparing and applying our initial codes to both existing and new data generated from subsequent interviews. This constant comparative analysis [24] across data sets allowed merging and clarifying codes. Following the initial coding process, research team members (AK and JA) discussed questions and discrepancies until 95% agreement was reached. Then, coders identified key points and recurring categories and themes that were central to the experience described by the participants. The process consisted of both coders dividing the text into semantic segments, labelling the segments with codes, together examining the codes for overlap and redundancy, and aggregating these codes into broader categories and themes [25].

Sample Characteristics and Suggestions
We recruited 10 PCPs (7 physicians and 3 certified registered nurse practitioners) with practice experience ranging from 3 years to 43 years, with a mean of 12.2 years. Out of 10 respondents, 4 were male, and 6 were female. The most common practice-based barriers included lack of time and knowledge of resources, including access to evidence-based medical models and affordable community options. Considering the results of this study, 4 factors emerged as important for consideration in the development of a CDSS for metabolic conditions: (1) clinically relevant and customizable information delivery, (2) provision of comprehensive medical summary with trends, (3) availability of patient education materials and community resources, and (4) simplicity and ease of navigation. Table 2 describes the key suggestions voiced by the PCPs for future design of the CDSS to be successfully adopted. Synthesis of available information "So, whatever you come up with has to be something that's integrated and uses the data that's there, and gives you immediate feedback. It can't be something that takes three minutes to enter the data." [Primary care physician, male] Fit in the workflow "So, ideally something self-contained, within the same page gives me kind of risk information and recommendations based of that, especially if it could be set up such that off of that page, I could directly order things. That would be amazing." [Primary care physician, female] User-friendly with minimalist design "I think you definitely need to maintain the ability to customize or edit because, again, these are just sort of recommendations and sort of a part of the picture that the risk calculator gives you, but, you know, as long as you know, you could sort of edit to customize and individualize to a patient." [Primary care physician, female] Flexibility "If there was something to standardize [management of] obesity and would give you a quantifiable number that puts them at a higher risk factor. So, if there was something that took in more either genetic versus biological markers that could be influential, I think that would be very useful and something that we would definitely want to implement and make it more of a standardization and not just an extra research tool." [Primary care physician, male]

Focus Not on Prevention But on Comorbidities
Almost all respondents reported that a significant portion of their patient populations was overweight, and they also noted that about 60% to 70% of patients had hypertension, diabetes, or other comorbidities. Even young populations presenting to primary care tended to have elevated BMIs or abnormal glucose levels. However, the respondents noted that they gave priority to management of the comorbidities rather than focusing on prevention and management of obesity. Respondents also noted they did not routinely use pharmacologic treatments for overweight or obesity but more to treat comorbidities, such as hypertension or elevated blood glucose levels.

BMI as a Main Diagnostic Measure
According to the respondents, BMI remains the primary tool for assessing obesity, as it is easy to access, is affordable to measure, and can conveniently be used to monitor weight changes. Additionally, participants responded that waist circumference measurement has not been integrated into routine practice. Patient risk factors associated with being identified or diagnosed as overweight or obese by their physician included higher BMI, family history, lifestyle, and habits. Respondents noted that they provided metabolic screening depending on patient's BMI, including blood glucose and blood lipids levels.

Reliance on Lifestyle Modifications
Most of the respondents' approaches to weight management were limited to assessing physical activity and assessing readiness for change, dietary habits, and expectations. The most common recommendations were to increase physical activity and dietary changes. Interestingly, half (5/10, 50%) of the respondents noted they did not have any formalized treatment plan to manage overweight or obesity and did not follow specific treatment guidelines. In addition, there was limited use of external sources of weight management support, with only few patients being referred to weight loss clinics, mainly due to limited coverage of services by health insurance companies. External resources frequently included a nutritionist and a commercial weight loss program (eg, Weight Watchers).

Lack of Knowledge About Referral Options in the Community
Respondents agreed that resources for intense lifestyle intervention and social support were important for the patients; however, respondents also noted the lack of knowledge about referral options in the community, including commercial-based programs. Because of the range in the socioeconomic status of their patient population, respondents expressed wanting point-of-care information about various affordable and convenient options that would be readily accessible and affordable for patient engagement.

Lack of Patient Education Literature
One of the challenges voiced by respondents was lack of appropriate, "meet them where they are" weight management educational materials accessible for use at the point of care or after consultation.

Need for a Risk Stratification Tool Embedded Into the CDSS
Almost all respondents (9/10, 90%) expressed an interest in having a CDSS that would incorporate diabetes and cardiovascular disease risk assessment and, based on the risks, outline a treatment plan. As respondents noted, the advantage of using a CDSS would be providing "legitimate justification" for a treatment plan with an assumption that patients understand their risk and the reason for the proposed treatment. To the authors' knowledge, there is no decision support system available to assist providers in evidence-based weight loss treatment intensification. There is, however, a diabetes management protocol that has been developed but is not part of the electronic health records at this institution.

Diagnostic-Supported CDSS
Respondents noted a need for an CDSS that would consider diagnostics, such as relevant patient data and lab results. In addition, respondents indicated it would be useful to have access to clinically meaningful trends and track risk scores for complications. A majority of respondents manually calculate various risk scores, such as the 10-year Atherosclerotic Cardiovascular Disease (ASCVD) risk; therefore, embedding such a calculator in the CDSS could increase efficiency and reduce error.

Incorporating Evidence-Based Practice
Respondents thought that having a CDSS that incorporates evidence-based clinical guidelines for management of obesity or overweight, both medical and behavioral, and that provides intervention recommendations would standardize and streamline the care provided and interventions suggested to their patients. The general idea was that such a system might help assist with managing patients with required tests, follow-up appointments, and preventive care.

Ability to Have Resources to Make Referrals and Educate Patients
Respondents expressed a strong interest in information about accessible and affordable resources in the local community for patients struggling with overweight or obesity. They believed that it would enhance patient engagement and motivate patients toward behavior change. In addition, several respondents suggested that, if the CDSS had the means to efficiently provide appropriate educational materials to patients, it could improve the patient's participation in their own care: .

Principal Findings
In this study, we sought to understand and capture user requirements for a system that evaluates the stage and severity of cardiometabolic disease that would be incorporated into a CDSS. The idea to involve intended users early in the design process is well supported in the literature [26] and results in aligning user expectations with the resulting functionality [27]. Currently, the PCPs' approach to management of overweight and obesity largely focuses on treating comorbidities and counseling lifestyle modifications such as diet and exercise. There is limited use of medications to combat overweight and obesity. Although there are existing guidelines for obesity treatment [28] and related cardiometabolic conditions such as diabetes [29] and hypertension [30], our respondents were not consistently using them for diagnosis or treatment purposes.

Comparison With Prior Work
Findings from this study are consistent with a growing body of literature on how PCPs manage overweight and obesity, as well as on what CDSS features increase likelihood of its uptake. Turner et al [31] found, among a nationally representative sample of active health care providers, that (1) knowledge of physical activity and dietary guidelines was limited and (2) understanding of the appropriate initiation, intensity, and duration of pharmacotherapy was often inconsistent with evidence-based guidelines. Another study found that PCPs were least likely to say they would prescribe medication or refer a patient to counseling [32]. Regarding CDSS features, a systematic review by Groenhof et al [33] found that design and usability were important drivers behind the success, noting that information should be displayed all at once and at one glance. In addition, the lack of insight into the automated computation and source of information decreased user satisfaction. Further, the most recent systematic review by Kouri et al [34] identified important CDSS features that significantly predict uptake, such as averting the need for provider data entry by mining patient data from within electronic health record systems to inform CDSS.

Implication for Practice
An important consideration would be providing easy access to the latest evidence-based clinical standards and protocols by embedding them in the CDSS. As a first step, CDSS could include measurement tools to perform a diagnostic evaluation based on evidence-based guidelines. If currently only BMI is considered, a more comprehensive evaluation must include additional measures such as measurement of waist circumference. Further, for the treatment, the physicians should be able to obtain clinical decision support by using CDSS to analyze pertinent information about the patient's current clinical condition, including information about medication, lab results, and treatment compliance. Given support for the CDSS among our sample, we propose a design of a CDSS that provides suggestions for treating the primary and augmenting medications with explanations. For the purposes of follow-up, the CDSS should have reminders to ensure the important considerations are not overlooked. Moreover, it could also recommend and display when the patient should return for a visit. All entries should be automatically stored, providing electronic documentation and record keeping, thus providing access to complete patient information. Overall, information about a patient's demographic characteristics and other clinical records should be accessible by a single click.

Implications for Development
Results from this study were used to better understand user requirements within a parallel system analysis and design framework (see Figure 1), the importance of which was to ensure the voice of the user was adequately and accurately represented [20]. In this phase, we present the conceptual framework with the findings applied as high-level categories (see Figure 2). These categories and the details behind them as presented throughout this study will be used to inform the evaluation.

Limitations
There are several limitations of this study. First, the 10 participants, predominantly non-Hispanic White (9/10, 90%), represented PCPs at a large academic center in the southeast. A small homogenous sample size could potentially limit the generalizability of our findings, and we recognize the need to add breadth and depth to this participant sample as development ensues. In addition, the strength of the study was our consideration of assuring intercoder reliability. Thus, we feel confident that we are correctly representing the voices of our participants.

Conclusion
Implementation of a CMDS system in the form of a CDSS could be used as a risk assessment tool that also provides risk-based and evidence-based treatment or program recommendations to better manage overweight and obesity and prevent diabetes. Results from this study provide unique insight to developers and researchers to identify areas for design optimization for improved end user experience to ensure successful adoption of the CDSS.