Rapidly growing patient interest in enhanced engagement in care processes has motivated health organizations to provide patient-centric care delivery both in clinical and homecare settings. With the goal of giving each patient a more proactive role in their care, we motivate and propose a formal process-driven framework for streamlining patient-centric care and improving patient-provider communication. It will lead to patients having better access to health services and taking more responsibility in their health management. At the same time the burden on healthcare professionals is reduced, while enabling greater efficiency, improved safety and higher quality.
Despite advances in life expectancy and quality of life, the current healthcare delivery
system faces significant challenges in terms of cost, accessibility and quality. One of
the goals established by the Institute of Medicine in 2001 is that healthcare delivery
should be patient-centric [
Fig. 1 shows a clinical workflow that delineates the path of a patient who interacts with healthcare teams such as clinics, labs, and pharmacies. In this care process, the patient is the only constant who is involved in all the steps and communications among the large number of participants in the healthcare ecosystem. For example, when a patient schedules an appointment, or is discharged from a hospital, the patient communicates with administrative staff. At other points of care the patient undergoes clinical activities such as detection and treatment, which involves various entities such as departments, staff, resources, etc. In this setting, it is important to consider a process-oriented perspective that coordinates and maintains the flow of information between the patient and other entities to ensure an optimal outcome.
Recent years have seen an increasing interest in IT-based systems that support care
delivery. Although many process-driven approaches have been proposed to support
clinical workflow, most are from the care providers’ point of view.
ComputerInterpretable Guidelines (CIGs) formalize medical guidelines that were originally in
the form of free-format text as computer executable languages, such as Asbru, EON,
GLIF, PROforma, and SAGE [
More recently, as the focus of care providers shifts towards patient-centric care, a
first step has been to develop applications that support patient access to their own
health data and facilitate patient communication with providers (e.g., schedule
appointments). A selection of web-based personal health record (PHR) systems, such as
WebMD, is reviewed in [
This paper proposes a process-driven approach to streamline patient-centric care. We formalize clinical pathways based on guidelines and propose a patient information model that incorporates patient needs and preferences. Thus, this framework aims to allow patients to: (1) access their health data and gain insights into the whole process; (2) express choices and take more responsibility; and (3) get a more personalized and coordinated continuum of care. Our approach also benefits the providers since it transfers patient communication workload from medical staff to the system, and tracks patient flows so that process improvement can occur. This paper is organized as follows. In Section 2, we propose a formal framework and describe the patient information model, followed by the decision making process in Section 3. Finally, we discuss future work and conclude the paper in Section 4. 2
A Process-driven Framework for Patient-centric Care
(Section 2.1) Rehabilitation Patient preference profile Preference 1 2 Treatment Medica- Surgery tion Exercise 3 Device therapy Salt- Alcoholrest. diet rest. diet Personal health records Item Parameter Symptoms Coughing Signs Blood pressure Lab tests Blood count Personal clinical pathways # Time … … Value YES 140 Normal Details …
(Section 3.2) bCuoinldteinxgt CVWoohmuagitthsiniyngmgptomHFsaedtaiogduyaeochuehaEvdeDe?ymsapnea (Se3c.2ti)on PDreevfiecreenthceertaopyth:e3trSeautrmgeernyt:m2ethods? Medication: 1 Preference to the living styles? Exercise: 1 salt-restricted diet: 2 Alcohol-restricted diet: 3
knowledge base (Section 2.2)
(Section 3.3) EHR
ECAxldipnmeicicsatselidopnaLtdOhawSte:a:y11:d1ha/e2ya8r/t2Af0at1itle1unrdeinDgPipsachtihyeasnritcgineaandm:ateeD::Ja_ov_hi_dn_S_Lm_e_eith Time Location Exam Test Treat. 8.15-9am rEoxoamm sVigitnasl N/A N/A 19.02.30-5am (RLMab01) N/A CXT-rsacya,n inhAibCiEtor 1-2.15pm Ward N/A N/A Diuretics 2.20 pm Discharge N/A N/A N/A
Fig. 2 presents an overall framework for process-driven, patient-centric care delivery. All steps and decisions are driven by medical guidelines, patient preferences, and needs and values as captured in the Patient Information Model (PIM). Contextbuilding is the process of obtaining patient information by conversing with patients in a structured way, e.g., when they are at home or waiting in a clinic. Then, the framework integrates guidelines, aggregate information from past executions stored in the clinical pathways repository, patient needs and preferences, and invokes the shared decision making module to suggest options to the patient. The patient can review the options to learn about the issues that pertain to the care process. Finally, the doctor reviews these options, and possibly others also under consideration, with the patient. An action is determined based on their discussion and agreement. All medical decisions, actions, and outcomes for each patient encounter are documented in a personal clinical pathway which is recorded in the PIM. The system detects deviations from medical guidelines and requests the doctor to enter reasons for any major deviations, which are logged. The depersonalized process logs collected by our system are analyzed to provide patients with insights about the care of other patients that have experienced similar situations. This framework helps to guide patient conversation by semi-structured process models and coordinates activities among various participants. It helps to reduce the work required by the patient to interact with the system (e.g., reduce duplicate data solicitation, suggests options for patient learning) and enables the patient’s participation in various tasks by letting them know what to expect of the care providers and what actions to take at all points of care. We describe the components of Fig. 2 further in the following subsections and in Section 3. 2.1
A patient information model (PIM) is comprised of three parts: personal health
record (PHR), personal preference profile (PPP), and personal clinical pathway (PCP).
PHR concerns a patient’s lifelong health information that she is allowed to access,
coordinate, and share with other parties [
The PPP captures an individual’s preferences pertaining to her current situation.
We use the rank-ordering method which is a popular comparative scaling technique to
evaluate users’ preference or liking [
The PCP documents the actual decisions, actions and outcome organized in chronological order pertaining to a specific episode of care. Deviations from best practice may be necessary to satisfy a patient’s needs.
A medical guideline is a document that guides decisions and criteria regarding
diagnosis, management and treatment in a specific medical discipline (e.g., heart disease).
This is naturally aligned with the way they are developed, i.e., by medical staff with
different expertise areas. A clinical pathway implements medical guidelines after they
are tailored to local and individual circumstances [
A personal clinical pathway (PCP) documents the actual execution for a specific patient. It may correspond to a clinical pathway such as Fig. 4. It keeps track of medical decisions (e.g., prescribe ACE inhibitor which is a pharmaceutical drug used primarily for treating hypertension and congestive heart failure), actions (e.g., dosage for ACE inhibitor), and patient outcomes in chronological order for each patient situation. Each task is associated with its time of occurrence. As noted above, deviations are allowed since humans control the actual execution of the process. A final outcome, e.g., the patient is cured, or ultimately passes away, indicates the end of a PCP. The PCP is a result of clinical decision making which is discussed further in Section 3. In this section, we describe the shared decision making process. Medical knowledge for decision points is formulated as rules that are used to derive recommendations based on best practice and patient preferences. We use rules to embody medical knowledge. The rules help to make complex decisions in clinical pathways through logical reasoning. For example in the clinical pathway of Fig. 4, N2 is a decision node that decides the next step, e.g., treatment or further evaluation, based on patient diagnosis results. A node can be associated with a number of medical rules. Integrating these rules and applying results from rule-based reasoning into a clinical pathway is critical for implementing evidence-based practice. In addition, each rule is associated with a strength of evidence (SOE) value to indicate its reliability. The three values for SOE are: A (good evidence), B (fair evidence), and C (expert opinion). They are based on a quality-rating system developed by AHRQ. For example, rule R1 is associated with task T10 and shows recommended medication based on “good evidence” (SOE equals A).
Rule R1 (Node: T10-medication): SOE=A If a patient’s systolic blood pressure < 90 mmHg and
there is a higher risk of complications
Then prescribe ACE inhibitors managed by an experienced physician 3.2
A medical decision is context-dependent, where context is patient specific. Our system can facilitate the process of learning about context by asking questions we expect of patients prior to their interaction with the care provider, and recording their responses. For example, in Fig. 4, context that is used at node T3 (detection and treatment) can be collected prior to that point, e.g., at node T1 (pre-admission) or T2 (patient admission). Then, depending on the patient answers, the subsequent questions need to be adjusted. We propose a patient conversation model (PCM) that describes the key questions asked at various points of care for a specific clinical pathway.
Fig. 5 shows an example PCM for heart failure, represented as a decision tree. The top part is derived from medical guidelines and the other two parts are developed based on practical experience and patient needs. These questions are available for patients to answer any time prior to T3 in the clinical pathway of Fig. 4. For example, a patient can enter her answers at home or while waiting for examination. PCM is process-aware since context becomes increasingly available as the care process proceeds. Via this model, we also give an opportunity to the patient to access information, e.g., the details of each treatment option (e.g., general success rate, relative cost, and side effects). Thus care providers spend less time on explanation. 3.3
Medical decision making should follow best practice through medical rules and take into account patient information obtained during context-building. The decision algorithm works, briefly, like this: when a decision node D is reached, we retrieve the rule set RS associated with D, run them against PHR and get evidence-based results. Other options not triggered by rules may still be presented to patients who know that no guideline supports the options but that may better meet patient preferences. Fig. 6 shows an example of decision making at node N2 (Diagnosis). During initial evaluation, this patient underwent a physical exam and diagnostic testing. Her signs indicate that she might have had heart failure. Her systolic blood pressure is 85 mmHg, and there is a high risk of complications. As a result, medical rules (R1-R6) are triggered at different points of care and produce the results shown in Fig.6. ACE inhibitor (SOE = A) and Diuretics (SOE = C) are recommended based on best practice. Nevertheless, patients and doctors decide which option is chosen.
CABG (T8)
PTCA (T9) Prcd. Cost SOE … 1.5k R4 (A) … … 0.8k 1k
R6 (C) N/A
Pref 2 In general, care providers should promote consistency and uniformity in care delivery through implementation of evidence-based practice. The paradox is that, on the one hand, it is desirable to reduce variation by standardizing workflows to conform to best practice; on the other, clinical pathways should be designed to allow flexibility to meet specific needs of patients and resource constraints of a health system. Thus, a formal and radically new approach is required for streamlined communication beContext-building
Patient information
model Patients with signs
of heart failure R4, R5, R6 …
Medication therapy (T2)
… R4,R5…
ACE inhibitor (T5)
Beta blocker (T6)
R6… Diuretics
(T7) Options Pros Side effect/Risk ACE Reduce mortality in severe heart Decrease in blood pressure, inhibitor failure… increase in potassium … Diuretics Treat various conditions, e.g., high BP Nausea, dizziness, fatigue...
Beta- Reduce risk of recurrent heart Side effects similar to heart blocker attacks… attack itself
R1, R2, R3…
No signs of heart failure Patient and family consulting (T3) Heart failure patients with angina, or history of MI
Surgery treatment (T4)
tween patients and providers to deliver evidence-based, yet personalized, care where patients can play a more proactive role in their health care matters.
In this paper, we describe the blueprint for such an approach. We propose a formal process-driven framework to streamline the communication between patients and care providers. Specifically, we introduce a patient conversation model (PCM) that informs the patient and the care provider, and a patient preference profile that informs the care provider. Introducing this information within care processes in a systematic way contributes to patient-centric delivery of care. This approach can benefit patients by allowing them to express their preferences and needs, and play a more active role in their own care. It also transfers a lot of the workload of handling patient communication from the medical staff to the system.
In future, we plan to extend and refine the structure of PCM models based on
inputs from health professionals and patients. We also intend to automate the
construction of the conversation model. Further, we expect to develop a patient portal based
on existing open source tools as an engagement platform for patients and use HL7
messaging protocol [