Clinical Interviewing: An Essential but Neglected Method of Medicine Free

The effectiveness of a physician depends very much on skills in eliciting information from the patient by interview and observation as well as on the ability to interpret such clinical information [1]. Clinical interviewing has multiple purposes. First, it is the way a physician understands how a patient feels in general and what are the presenting complaints [2]. Second, it is the basic method to obtain a medical history [2]. In general medical practice, the majority of final diagnoses can be made on the basis of patient’s history alone [3]. Third, interviewing allows detection of personal attitudes and behavior related to health and disease, which shape crucial determinants of illness such as lifestyle, presentation of symptoms, access to patient care, and adherence to medical advice [4]. Fourth, it gives the patient information about diagnosis, further testing, prognosis, and treatment [5]. Finally, empathic listening establishes the crucial bond between patient and physician that is needed for a successful therapeutic relationship, with particular reference to shared decision making and patient’s self-management [6, 7].

Despite its importance and efficacy, interviewing as a basic skill is threatened by a variety of forces that result in skill erosion, if not outright neglect. These forces include the allure of technology, time pressures, use of shortcuts such as questionnaires and chatbots, and the hegemony of the biomedical model of disease as exemplified by “precision medicine” [8‒11]. Further, the level of details that are required in many health care systems to provide written documentation of the interview can burden the clinician and negatively affect the nature of the interaction with the patient. Yet, clinical interviewing occurs in all medical consultations and affects outcomes for doctors and patients.

The aim of this paper is to provide an overview of the effective use of clinical interviewing in relation to a number of issues: styles; flow of interview and hypothesis testing; clinical domains; modifications according to settings and goals; and teaching.

Methods of interviewing can be grouped into two main categories. One is based on open-ended type of questioning, which allows patients to speak freely about their condition, with questions such as “how are you?” or “how do you feel?” This opening question allows patients to explain what has led them to seek evaluation. The other method involves closed-ended questions such as “do you feel any pain?” or “do you feel depressed?” that often follow an interrogative “yes/no” questioning that is in line with electronic medical records [7]. Of note, however, a fixed sequence of questions does not allow exploration of implications and extension of individual responses [12]. The two strategies are generally mixed in clinical practice. An initial open-ended question may be followed by closed-ended questions based on the patient’s initial response. As Feinstein remarks, “the distinctive talent a clinician brings to medical history taking is his ability to obtain descriptive specifications of the patient’s sensations, to elicit important information not volunteered spontaneously, and to ascertain that the features a patient leaves unmentioned are indeed absent” [12], p. 318. For pursuing the characterization of symptomatic sensations, as well as the sequence and timing of clinical manifestations, the clinician should be an active participant, not just a passive recorder [12]. The time allotted to the interview is an important determinant of the opportunity given to patients for speaking freely of themselves. Toward the end of the interview, the physician may ask the patients whether there is anything that they missed and what other things would be important to know.

Clinical interviewing is not only based on verbal exchanges and patient’s answers. It is the physician’s task to observe the patient’s nonverbal communications (e.g., gestures, posture, facial expression, way of speaking). Such observations require close attention and engagement [2], and nonverbal information is unlikely to be captured by an interviewer who only watches the computer screen instead of the patient [7, 13]. The electronic medical record is here to stay. The challenge is for physicians to use it well, while retaining the empathy, wisdom, and attentiveness that are needed to forge satisfactory working relationships.

Some patients enter the clinical encounter in an apprehensive or disorganized state. The task is to organize the underlying problem in a way that both the physician and the patients can understand so that a diagnosis is made and effective treatment can be instituted [7]. The presenting complaints need details and specifications from the patient for a characterization of the clinical picture. Closed-ended questions are required, but open-ended strategies may allow the physician to obtain additional valuable information that expands the starting point of the investigation. In the dialogue that ensues, clarification techniques may be used, such as under which circumstances the symptom became manifest, which factors worsen or improve the symptomatology, and whether or not other symptoms occur [2]. Steering techniques may also be of value, such as continuation (e.g., “tell me more”), accompanied by nonverbal gestures such as a sympathetic nod, echoing (repeating the part of the patient’s answer which needs to be elaborated), redirecting (asking the patient to stop diverging from the main line of inquiry), and transitions (moving to topics that need to be covered) [14]. These techniques not only enhance a collection of information but also let the patient indirectly know that the clinician is paying attention to the patient’s answers.

Medical history is based on the collection of information by clinical interviewing [2]. Several factors may hinder such collection. Patients may use various forms and degrees of denial, or they may show resistance in the form of avoidance of certain topics [15]. Shifting the focus (approaching a problem from another direction) and going back to the issue at some later point in time during the interview, when a higher level of trust and communication may be established, are two techniques that may help [14]. The doctor’s idiosyncrasies (culture, training, communication) and attitudes during the interview (e.g., lack of adequate attention, psychomotor signs of impatience or boredom) may inhibit expression [16, 17].

Expert clinicians make a medical diagnosis through a process of hypothesis generation and verification [18], moving from questions such as “why are you here?” to relevant issues, and then to potential etiological explanations and differential diagnosis. Hypothesis generation is based on knowledge from past experience and information obtained by clinical interviewing [18]. Verification is based on further probing during the interview to confirm or reject each hypothesis, in addition to physical examination findings and laboratory test results. The interview allows evaluation of symptoms according to a number of variables, such as intensity, duration, and trajectory, and factors that may improve or worsen them. Symptoms and comorbidities that at first glance may appear to be unrelated and scattered among different organ systems may then reveal unifying features [19]. The expert interviewer will channel the course of the interview toward those directions that are most likely to verify initial hypotheses [20]. Intentionally abandoning issues at the beginning of the interview and reexploring them at a later point in time may become necessary when varying degrees of illness denial hinder the collection of information [2, 15, 20].

Feinstein [21] remarked that, when making a diagnosis, thoughtful clinicians seldom leap from a clinical manifestation to a diagnostic endpoint. Instead, the clinical reasoning goes through a series of “transfer stations,” where potential connections between presenting symptoms and pathophysiological processes are considered. These stations are a pause for verification or a change to another direction. Clinicians should think ahead, formulating the next question while still listening to the answer to the previous question, to keep the flow of the interview. Certain signs can only be discovered if they are suspected and searched for [22].

Collecting information regarding the general condition of the patient, presenting signs and symptoms and medical history are the universally recognized goals of the medical interview [23]. In addition, there is growing awareness of the importance of assessing psychosocial and cultural factors that potentially affect individual vulnerability to and patterns of medical illness [3, 23‒25]. Indeed, a strictly biological model of pathogenesis that overlooks patient’s life experiences and social context is insufficient to account for all the variables that make individuals susceptible to disease, resilient when disease arises, and variably responsive to treatment [25]. Extensive evidence suggests that even symptoms that can be attributed to a physical illness may not be explained fully by the disorder [3]. The objective severity of medical disease often correlates poorly with the degree of subjective impairment and quality of life [26]. For instance, exercise tolerance in patients with chronic obstructive pulmonary disease may correlate more closely with depression than with abnormalities in spirometric testing [27].

Illness denial may hinder a collection of symptoms that may be of crucial importance for understanding the clinical situation [2, 15]. In the presence of illness denial only skillful interviewing may disclose the presence of certain physical symptoms and increase the likelihood of early recognition of a life-threatening disease and its prompt treatment [2, 15].

Even though there are many questionnaires covering psychosocial aspects of medical illness [3, 4, 24], and patient-reported outcome measures appear to be particularly valuable [28], interviewing is the key instrument for collecting information on psychosocial aspects of medical disease. Individual perception of the stressfulness of an environment exceeding his/her resources (allostatic load), presence of depression, demoralization and anxiety, and health attitudes and behavior stand as particularly important in the clinical process [4, 29‒31].

The interview is the principal means of assessment in clinical psychiatry [32]. It is analogous to history taking and physical examination in medicine. The psychiatric interview includes the Mental Status Examination that encompasses appearance, attitude, affect, mood, behavior, speech, thought process and content, perception, cognition, insight, ability to abstract, and judgment [32]. The Mental Status Examination lends itself to the formulation of psychiatric diagnoses that are obtained with specific questions. It has been argued that exclusive reliance on diagnostic criteria in psychiatry has impoverished the clinical process and omits attention to other issues that may be as important, such as environmental stress (allostatic load), the psychosocial context, illness behavior, and psychological well-being [33].

The medical consultation (i.e., any clinician-patient interaction) is the most frequent task of physicians. Doctors have between 100,000 and 200,000 consultations with patients during their careers [34]. There are different modalities of clinical interviewing in relation to its goals and settings. The initial interview may have different characteristics as to whether it occurs in general practice/internal medicine or in a specialty setting. In the latter case, the initial interview may be restricted to the system that is of primary concern (e.g., cardiovascular) or expand to symptoms or diseases that pertain to other organ systems (e.g., gastrointestinal), as needed. It may be influenced by information obtained previously (e.g., referring clinician, family). Other sources of variation derive from the environment where the interview takes place and the priorities that ensue. For instance, emergency rooms and acute care units can be particularly difficult settings due to distractions caused by noise, lights, coming and going of people, and lack of confidentiality [20]. The primary question is whether the patient should be admitted to the hospital, or treated acutely in the emergency room, or discharged. Other challenges include the conduct of a clinical interview in patients with altered states such as delirium, dementia, withdrawal syndromes, or cognitive and affective alterations induced by medications such as analgesics and steroids [20].

Clinical interviewing also depends on whether it takes place in the early presentation of a disorder, during its acute or subacute phase, in a chronic state, or in a stage of remission [7, 20]. In all phases of a disorder, the response to therapy needs to be evaluated. In the initial interview, history of previous treatments plays an important role in determining treatment responsiveness or resistance, the occurrence, nature and severity of side effects, and the presence of iatrogenic factors [35]. Treatment history appears to be particularly complex in view of the frequency of polypharmacy [36] and requires accurate exploration, a process that can be compromised if an insufficient amount of time is allotted [35].

Reassessments may yield important information, such as stability of symptoms, areas that were left untouched in the initial interview, emerging aspects that may have been neglected, and vulnerabilities triggered by treatment.

Discrepancies may occur between physicians who regard their patients as fully recovered on the basis of objective measures (e.g., laboratory values, imaging) and patients who still perceive a certain degree of impairment [4]. Despite the increasing emphasis on patient-reported outcome measures in medicine [28], sensitive clinical interviewing with full attention to the patient’s reports is the main tool for determining the subjective perception of recovery [4]. Further, clinical interviewing may shed some light on the social determinants of a patient’s biography that can illuminate treatment response [37].

Keeping the flow of interviewing while always staying focused on the person and his/her concerns is a complex and difficult task that requires adequate training and substantial experience [34]. Systematic reviews show that several different educational interventions are effective in teaching history taking skills to medical students [34, 38]. In particular, small-group teaching with role-playing, interviews with real and standardized patients, and structured feedback are widely employed [34]. Teaching interview skills is also essential for nurse practitioners and physician assistants, who are increasingly involved in obtaining clinical data.

Technological advances play an important role in changing communication patterns. Many patients come to the doctor with information they have found on internet. Targeted communication training for medical students that addresses specific issues in interviewing patients who have done internet searches on their conditions has already shown positive effects [39]. The role of artificial intelligence (AI) and chatbots is an upcoming challenge [40]. The health care system is undergoing a dramatic digital transformation, including the use of AI in medicine [9]. It is important for the student to understand, however, that AI is of limited help in situations where the interpretative skills of the professional are necessary, as often occurs in primary care [17], and that AI-generated information can be inaccurate and misleading. There is no substitute to clinical interviewing in establishing shared understanding about the nature of a problem [17].

Engel [41] identified the key characteristics of clinical science in its explicit attention to humanness, where observation (outer-viewing), introspection (inner-viewing), and dialogue (inter-viewing) are the basic methodologic triad for making patient data scientific. Unfortunately current health care trends do privilege expensive tests and procedures and tag the time devoted to patient interaction as lacking cost-effectiveness. On the contrary, the time spent to inquire about problems and life settings may be cost-effective, saving on further testing, procedures, and referrals [3]. Physicians should make any effort to obtain an adequate time for clinical interviewing of both public and private sectors. Indeed, in a reimbursement system, clinical interview should become the best paid procedure, since it is an extraordinary investigative technique that is “the most clinically sophisticated procedure of medicine” [12], p. 299.

Motivating people to make beneficial changes in their behavior (lifestyle medicine), sharing the decision process, and increasing the role of the patient in the therapeutic plan (self-management) are now regarded as major health care targets [6, 42‒44]. The dialogue between patient and physician is essential in this task [6, 44].

The quality of interviewing determines the quality of data that are collected [22], and, ultimately, of assessment and treatment. Thus, interviewing deserves more attention in educational training and more space in clinical encounters than it is currently receiving.

Dr. Concato reports that this article reflects the view of the authors and should not be construed to represent FDA’s views or policies. He is a senior advisor to McKinsey and also a member of the RWJF National Advisory Committee and of Scripps Translational Research Institute, Scientific Advisory Committee. Dr. Wright has an equity interest in MindStreet, Inc., is developer and distributor of computer programs for behavioral health, and is a consultant for Otsuka Pharmaceutical. He receives no royalties or other payments from sales of these programs. His conflict of interest is managed with an agreement with the University of Louisville. All other authors have no conflicts of interest to declare.

All authors have no funding to declare.

Giovanni A. Fava wrote the initial draft. Nicoletta Sonino, David C. Aron, Richard Balon, Carmen Berrocal Montiel, Jianxin Cao, John Concato, Ajandek Eory, Ralph I. Horwitz, Chiara Rafanelli, Ulrich Schnyder, Hongxing Wang, Thomas N. Wise, Jesse H. Wright, Stephan Zipfel, and Chiara Patierno revised and contributed to the final version of the manuscript.