Physical Diagnosis: A Lost Art?
- Spotlight Case
Approach to Improving Safety
- Appreciate the decline in proficiency and reliance on physical examination skills among health care providers.
- List barriers to performance of comprehensive physical examinations.
- List interventions that may increase competence in physical examination skills.
Case & Commentary: Part 1
A 57-year-old male with T8 paraplegia from a
remote gunshot wound, hypertension, and diastolic dysfunction
presented from home with 1 week of intermittent fevers, chills,
increasing shortness of breath, and low back discomfort. Due to
neurogenic bladder, the patient performed self-catheterization
daily. Initial physical examination was recorded as
|General: Mild respiratory distress with
Vitals: Blood pressure 110/62, Pulse 106, Respiratory rate 18, Temp 37.8°, Room air saturation 100%
HEENT: Crusting periorbitally, no edema, extraocular movements intact, pupils equal round and reactive, dry tongue
Neck: No meningismus
Resp: Increased work of breathing with diffuse rhonchi, crackles bilaterally
Cardiovascular: Tachycardic without murmurs or gallops, normal apical impulse, 2+ pulses throughout, no lower extremity edema
Abdomen: Normal active bowel sounds with some tenderness to palpation over the suprapubic area
Skin: Stage I gluteal cleft decubiti
GU examination: Deferred
Neuro: Strength 5/5 upper extremities, 1/5 bilaterally lower extremities with flexion contractures; sensation absent below T8.
The history and physical examination remain the backbone of medical evaluation and assessment. However, the many advances in both laboratory and imaging technology and the pace of modern medicine have resulted in the physical examination being abbreviated and undervalued, and viewed (subconsciously, perhaps) as redundant.
Although few studies examine physical diagnosis skills over successive generations of physicians, skill and familiarity with certain bedside maneuvers and confidence in eliciting physical signs appear to have declined, with increased dependence on the aid of a radiologist or first-tier laboratory data. The new student on the wards soon finds that skills at the computer in getting data back and arranging for tests to be done are valued as much or more than learning to percuss well or hear a pericardial friction rub. At times, it almost seems as if the patient in the bed is an icon for the real patient who exists in the computer, and 'rounds' (a word that in this context connotes motion) are conducted with the participants immobile and seated in a room and with the patient represented either on an index card or a PDA (personal digital assistant) screen.
The reasons for this trend are complex. Physician reimbursement has become increasingly volume-driven, with little or no financial reward for one's ability to pick up subtle physical examination findings and with little time for that kind of detailed examination. One can get into the habit of reflexively ordering a series of tests to traverse assorted diagnostic algorithms. The many 'protocols' for various conditions are well intended, provided the physician has picked the right algorithm based on the history, physical, and initial laboratory tests. The higher sensitivity and specificity offered by laboratory and radiologic testing makes it more likely that a physician might be reluctant to make a clinical diagnosis that could be readily made at the bedside (of say splenic enlargement, aortic stenosis, or pleural effusion) until the echocardiogram or CT scan is reported. The result is that we see few people percuss the chest, and fewer still do it with any confidence or knowledge of the normal boundaries of chest resonance.
Case & Commentary: Part 2
Initial laboratories were significant for a white blood cell count of 20.9 with 82% segs, bicarbonate of 19, blood urea nitrogen of 27, and creatinine of 1.7. Urinalysis revealed 2+ protein, +nitrite, 3+ leukocyte esterase, 3+ blood, 10-25 white blood cells, and 10-25 red blood cells. Electrocardiogram revealed sinus tachycardia only. Chest radiograph revealed a bilateral interstitial edema pattern. The initial assessment by the female night float resident was sepsis from a urinary tract source and acutely decompensated diastolic heart failure. The patient was managed with afterload reduction, diuresis, oxygen, and intravenous antibiotics and was evaluated for myocardial infarction with serial enzymes and electrocardiograms.
Studies comparing the sensitivity and specificity of bedside diagnosis with that of laboratory and radiographic testing can be complicated and difficult to interpret, particularly when comparisons are made between, say, trainees and established physicians. One such study cites the inability of intensive care unit personnel to accurately determine the jugular venous pressure (JVP), calling instead for central venous access for determination of this parameter.(1) In that study, medical students performed best, with residents and attending physicians the least accurate at JVP assessment. This somewhat unexpected result was probably because the medical students involved in this study had participated in weekly cardiology rounds with examination of all observed patients' JVP for the duration of their rotation, whereas the residents and attending admitted to infrequent attempts to routinely assess JVP. Furthermore, the medical students in this study were unaware of the patients' clinical diagnoses, whereas residents and staff physicians had been given this information.(1) Such diagnostic information can bias a "retrospective" physical examination. The residents and staff perhaps subconsciously attempted to correlate their examination findings with what they expected to observe. The study illustrates the need for physicians to continually use and sharpen their examination skills and their use of specific maneuvers, lest these degrade over time.
It has long been said that physicians order an increasing number of tests because we are practicing 'defensive' medicine in an increasingly litigious environment. But, it is also likely that physicians order so many tests because we have lost confidence in our abilities to extract meaningful information from the physical examination. In particular, physicians seem to lack the confidence to say that an examination of a certain body part is normal, and no further testing is needed. In this regard, physicians in the United States differ from our colleagues in Canada and England, who tend to be frugal with their testing. The consequences of excessive reliance on diagnostic tests to convey information that should have been elicited on physical examination are twofold: first, there is time delay (often a day or two) in diagnosis as one awaits the test results; second, the patient is exposed to the risk and side effects of tests that may not be necessary. These risks include both the obvious unwarranted financial expense to the patient and the healthcare institution, but also the possibility for serendipitous discovery of "incidentalomas"—laboratory or radiographic abnormalities that are unrelated to the presenting complaint. The full impact of these incidental findings has yet to be defined, but the costs of follow-up imaging, additional laboratory testing, and increased patient concern of serious yet still undefined illness are obvious.(2)
Despite a growing body of literature questioning the value of the routine examination (3), this aspect of the physician-patient encounter is clearly valued by the patient. In one study, 90% of patients expected their blood pressure to be measured and their heart, lungs, abdomen, and reflexes be examined.(4) Even if routine examination may not be essential to actual patient care, we believe the skilled examination is critical to the development of the physician-patient relationship. Done well, it earns trust, patient confidence, and perhaps increasing patient compliance.
Case & Commentary: Part 3
The following morning, the patient was handed off to the daytime medical team. Genitourinary examination was again "deferred," and the treatment plan continued. Later that day, the attending physician examined the patient and found ecchymotic, edematous scrotal skin, a purulent perirectal fistulous tract, and perirectal crepitus. Urology and general surgery were contacted immediately. Shortly thereafter, the patient underwent surgical debridement for Fournier's gangrene, a life-threatening form of necrotizing fasciitis of the perineal area.
The practice of "handing off" patients necessitated by the 80-hour work week imposed on physicians in training might compound the risk of patients like this falling through the cracks. There is inevitably a tendency to rely heavily on the admitting physician's initial assessment and diagnosis, and the labels given the patient tend to stick. Bias creeps in. Rarely is the patient thoroughly reexamined by the physician completing the treatment plan. Frequent reassessment of the patient, rather than diligent follow-up of previously ordered laboratory tests, is more beneficial to correct clinical care. The old clinical saw, "There is no substitute for laying hands on your patient," remains true today, perhaps more than ever.
Physicians may defer examining parts of the body that seem unlikely to contribute to the presenting complaint. Patients may have impairments that make them unable to voice what is bothering them. Paraplegia or other conditions with sensory impairment represent a distinct class of comorbidities requiring diligent and thorough examination, similar to the common practice of examining a diabetic patient's feet regardless of the presence or absence of podiatric complaints. The chief complaint of these patients is often secondary to an underlying condition that resulted from sensory impairment. One of us (G.R.T.) has seen a patient with spinal cord injury who was transferred from an outside hospital for urinary tract infection. The patient voiced no complaints and was anxious to be discharged. The resident in charge of his care incidentally noted subcutaneous crepitus in the arm while measuring the blood pressure. The patient was subsequently found to have extensive necrotizing fasciitis of the shoulder and abdominal soft tissues.
Necrotizing fasciitis is a perfect example of a clinical condition in which one might make the diagnosis at the bedside and in which delay can be deadly. This is predominantly a clinical diagnosis with the direct visualization of the involved area being critical, along with recognizing the patient's apprehension and early signs of distress. What is not seen is not diagnosed. The common practice of "deferring" aspects of the physical examination viewed as non-essential was unfortunately responsible for this patient's initial incorrect diagnosis. Several factors can lead to deferment of a close examination of the genitals or areas in proximity to it, and one such factor is when physician and patients are of different genders. Factors that lead to deferral of the genital examination include a lack of confidence in performing an examination of a member of the opposite sex, fear that the patient will see this as an unnecessary examination, and difficulty in finding a chaperone.(5) Often, if one sees a patient had a rectal examination in the emergency department or by some previous examiner, and the stool guaiac has been done, there is a sense that one no longer needs to go near the genital area.
Medical education revolves around future doctors watching and learning from their faculty and seniors. The "apprentice" may only become as good as those under whom they train, and certain time honored and still valuable physical diagnosis skills may no longer get passed on with regularity. Attending physicians and others in educational roles must be sure to model these skills and be vigilant to ensure that what they are passing on is accurate and performed correctly.(6) The reassurance offered to those in training by watching their faculty member correctly make diagnoses at the bedside and forego more sophisticated testing would likely motivate them to put greater value on these skills.
It seems unlikely that physical examination skills in North America will ever come back to their apogee, when this art form was practiced with great skill by the likes of Osler (Figure), Cabot, and so many others. At present, internists can become board certified without having their skills tested at the bedside by certifying examiners. This would be the equivalent of allowing commercial pilots to fly us around without anyone having demonstrated that they were capable of flying. Reliance on program directors to sign-off on the residents' clinical skills is putting too much faith in the system of residency training. There is nothing like a national clinical skills examination at the bedside to elevate the standard of bedside practice. The Fellowship examinations in England and Canada, although they can be quite subjective, nevertheless create a housestaff training culture that values physical diagnosis skills—at least as much as doing 'board review' questions. Skilled clinicians test the candidate at the bedside on real patients to see if they can sort out valvular heart disease or pick up all the physical signs that suggest the presence of an internal capsule thrombosis and stroke. There is at present a 'clinical skills exam' for medical students, which in our opinion, based on conversations with recent test takers, tests everything but the kind of true clinical skills that are tested in other countries; it does little to test the ability of the candidate to palpate an enlarged spleen or detect a pleural effusion by percussion or localize a lesion in the nervous system with a skilled neurological examination. Taylor and colleagues attempted to correlate USMLE exam scores, clinical skills exam scores, and undergraduate grade point averages (GPA) with intern performance measured by residency program director surveys. Rankings by program directors were most highly correlated with undergraduate GPA, followed by the interpersonal skills component of the clinical skills exam, USMLE step 2 scores, USMLE step 1 scores, and then step 2 clinical skills exam scoring.(7) Even if there had been a direct correlation between the exam and intern performance measured by residency program director, the kind of clinical skills we are discussing in this commentary are simply not tested at the student or resident level.
One of us (A.V.) has had the opportunity to see students and residents from the United States working in clinics in Africa or India. What is most gratifying is how quickly these young physicians pick up and see both the utility of the bedside examination and its limitations in a resource poor setting and how they come to see how valuable are the 'routine' laboratory and radiological tests that are rationed and not routine in these settings. More importantly, they discover that developing such skills is very rewarding and that they can translate these skills well to their residency programs when they return. The only way to bring rounds back to the bedside (where they belong) and to raise the level of physical diagnosis skills is for students and residents to see these attributes being modeled by attending physicians and senior residents. The generation of physicians who practiced in this fashion and were masterful at the bedside is beginning to retire. Without leadership in this area from regulatory organizations and specialty societies, this skill set will continue to disappear.
- Physical examination skills appear to be declining as reliance on laboratory and radiologic testing has increased.
- Revival of the art of physical diagnosis will require increased role modeling by faculty and senior residents of such skills and demonstration of competence for board certification.
- Proficiency in physical examination skills may lead to fewer missed diagnoses, improved physician-patient relations, and more economically sound medical care.
George R. Thompson III, MD
Department of Internal Medicine, Division of Infectious Diseases
University of Texas Health Sciences Center, San Antonio
Abraham Verghese, MD
Joaquin Cigarroa Chair and Marvin Forland Distinguished Professor
Department of Internal Medicine, Division of Infectious Diseases
Director, Center for Medical Humanities and Ethics
University of Texas Health Sciences Center, San Antonio
Faculty Disclosure: Drs. Thompson and
Verghese have declared that neither they, nor any immediate member
of their families, have a financial arrangement or other
relationship with the manufacturers of any commercial products
discussed in this continuing medical education activity. In
addition, their commentary does not include information regarding
investigational or off-label use of pharmaceutical products or
1. Cook DJ. Clinical assessment of central venous pressure in the critically ill. Am J Med Sci. 1990;299:175-178. [go to PubMed]
2. Stone JH. Incidentalomas--clinical correlation and translational science required. N Engl J Med. 2006;354:2748-2749. [go to PubMed]
3. US Preventive Services Task Force. Guide to Clinical Preventive Services: Report of the Preventive Services Task Force. 2nd ed. Washington, DC: US Department of Health and Human Services; 1996.
4. Oboler SK, Prochazka AV, Gonzales R, Xu S, Anderson RJ. Public expectations and attitudes for annual physical examinations and testing. Ann Intern Med. 2002;136:652-659. [go to PubMed]
5. Powell HS, Bridge J, Eskesen S, Estrada F, Laya M. Medical students' self-reported experiences performing pelvic, breast, and male genital examinations and the influence of student gender and physician supervision. Acad Med. 2006;81:286-289. [go to PubMed]
6. Economides E, Stevenson LW. The jugular veins: knowing enough to look. Am Heart J. 1998;136:6-9. [go to PubMed]
7. Taylor ML, Blue AV, Mainous AG 3rd, Geesey ME, Basco WT Jr. The relationship between the National Board of Medical Examiners' prototype of the Step 2 clinical skills exam and interns' performance. Acad Med. 2005;80:496-501. [go to PubMed]
Figure. Sir William Osler at a patient's
Photograph reprinted with permission of The Alan Mason Chesney Medical Archives of The Johns Hopkins Medical Institutions.