Clinical reasoning is the process by which the information gathered from the history and physical examination is merged with clinical knowledge, experience, and the current best evidence to formulate the next steps in patient care—development of the diagnostic and management plans. Critical reflection involves thinking through the reasoning for these decisions. This reflection can help you progress from rote decision making to clinical reasoning.
The Clinical Examination
Thus far, we have been concerned with the initial patient interaction, establishment of respectful rapport, and information-gathering processes (i.e., history and physical examination). Organizing, integrating, and analyzing information represent the next step in caring for patients. This process, called clinical reasoning, leads to the development of potential diagnoses, workup priorities, and management plans ( Box 4.1 ). Formulating management plans should occur in partnership with patients and their families. Working with the patient to determine the best course of action is often called “shared decision making” or “patient-centered care.”
Clinical Reasoning
Assessment, Judgment, and Evidence
Clinical reasoning involves bringing knowledge of the patient, clinical experience, and decision-making skills together with the current best evidence regarding the issues involved. The best available evidence must be carefully obtained from the volumes of information available in books, journals, and online. Use of evidence in decision making is called “evidence-based practice” ( Box 4.2 ).
Evidence-based practice (EBP) incorporates best evidence, along with individual experience and patient preference, into making medical decisions ( Sackett et al, 1996 ). EBP involves questioning physical examination findings, the effects of therapy, the utility of diagnostic tests, disease prognosis, and/or the etiology of disorders ( Sackett et al, 1996 ).
The first step in clinical reasoning is to assess what has been learned from the patient and to determine its value and significance. Priorities are then assigned to information that might influence clinical judgment as management plans and clinical impressions are created.
Further assessment depends on healthcare provider, patient, and family preferences, which may be influenced by feelings, attitudes, and values. By using clinical reasoning, you can think through your own beliefs, values, cultural practices, and attitudes for potential bias affecting the decision-making process. Clinical impressions help direct further assessment, which must be balanced by benefits and risk to the patient, cost considerations, and available resources. Determining the leading diagnoses, potential diagnostic and management plans can be formulated using shared decision making.
Problem Identification
A problem may be defined as anything that will need further evaluation and/or attention. It may be related to one or more of the following:
- •
An uncertain diagnosis
- •
New symptoms or physical examination findings related to a previous diagnosis
- •
New symptoms or findings of unknown etiology
- •
Unusual findings revealed in the clinical examination or by diagnostic tests
- •
Personal, social, or emotional difficulties
Formulate problems as specifically as possible. Identify and list the signs and symptoms associated with each of the patient’s concerns as well as abnormalities discovered during the physical examination. This listing of findings, the problem list, is key to developing a complete understanding of a patient’s concern. The problem list is the foundation for clinical reasoning and is used to form hypotheses based on the available information ( Box 4.3 ). Review the problem list and note the absence of findings that you might expect in support of your hypotheses. Think through the patient’s verbal and nonverbal communication and determine whether there are questions you may have neglected to ask or information you did not fully understand. Gather additional information from the patient to fill in these gaps. Beware of “red herrings,” the bits of information that are distracting and draw your thinking away from central issues ( Box 4.4 ). Critically evaluate unexpected or unusual findings but do not let them distract you from full consideration of all you have learned.
There are several ways to make a diagnosis:
- •
Recognizing patterns (on the assumption that “If it walks like a duck, swims like a duck, and quacks like a duck—it is most likely a duck”)
- •
Sampling the universe (on the assumption that including everything precludes missing anything)
- •
Using algorithms (on the assumption that a rigidly defined thought process precludes error)
Although each of these may have limited use in specific situations, consideration of all your findings should most often result in the development of one or more hypotheses needing an evidence-based approach to solutions.
Guidelines to a sound decision-making process include the following:
- •
Always derive possibilities that are consistent with the chief concern, your findings, and known psychosocial and pathophysiologic mechanisms.
- •
Remember that common problems occur commonly, and rare ones do not.
- •
Common problems can have unusual presentations, and rare ones may have a seemingly common concern.
- •
Rare problems that have an available treatment should be considered.
- •
Do not rush to a diagnosis with no available treatment, and do not pursue a line of reasoning that will not alter your course of action. For example, if a patient cannot tolerate cancer treatment, do not peruse the screening and diagnostic evaluation.
- •
Do not undertake procedures that are not related to your hypotheses.
- •
Always consider potential harm and cost as well as benefit when determining the need for a test or action.
- •
Consider whether the risk is worth the potential gain in information, invoking the ethical principle of nonmaleficence: Primum non nocere —“First, do no harm.”
- •
Remain open in your thinking and be ready to discard or modify your hypotheses when necessary. Recognize that your leading hypotheses may not be valid, and avoid the tendency to discount information that may invalidate your favorite ideas.
- •
Try to have a single process explain all or most of your data, but do not be rigid in this regard. After all, a patient with many concerns and problems may have more than one disease; two common diseases may occur simultaneously more often than one rare disease alone.
- •
Probability and utility should always be your guides to sequencing your actions unless a life-threatening situation exists. A conscientious estimate of probability is the best way to define the limits of uncertainty and the best way to establish priorities.
A patient with swollen cervical nodes and fever owns a cat, which had claws, but the patient did not have cat scratch disease. The too obvious apparent source of the problem, a “red herring,” delayed the ultimate diagnosis of non-Hodgkin lymphoma.
After a match between the data (both subjective and objective) and a presumed diagnosis is made, consider the appropriate laboratory, imaging studies, or specialty consultation needed to confirm the diagnosis. When determining next steps, it is important to consider primacy of patient welfare, patient autonomy, and need for resource allocation. We are after all strong advocates for our patients and, at times, their only voice. As such we are responsible for appropriate resources allocation and scrupulous avoidance of superfluous tests and procedures ( Cassel and Guest, 2012 ).
Valid Hypotheses
Clinical reasoning allows you to consider and discard the variety of possible diagnoses—from the common to the rare—before settling on the best match between the patient’s signs and symptoms and a specific disorder. It has been said ( Kopp, 1997 ) that there are at least three diagnoses for every disease: the one that unifies what you have learned, the one you cannot afford to miss, and the one that it actually is. Sometimes they are the same one, but usually not. Do not have tunnel vision and let initial thoughts narrow the focus of questions during the interview. In other words, do not jump to conclusions.
One of the clichés of clinical practice is that all findings should be unified into one diagnosis, Occam’s razor or lex parsimoniae (“law of parsimony” or “law of succinctness”) ( Thorburn, 1918 ). Although you should strive to look for the fewest possible causes that will account for all the symptoms, this may not always be possible. More than one disease process can exist at one time in the same person, an acute illness can occur in the context of a chronic one, and a chronic disease can cycle through remission and relapse. Carefully chosen laboratory or imaging studies can often help validate your observations and confirm your clinical impressions ( Box 4.5 ).
