quarta-feira, 15 de novembro de 2017

Physician's words



The Iatrogenic Potential of the Physician’s Words

 
 
 
JAMA. Published online October 31, 2017.  doi:10.1001/jama.2017.16216
 
 
Some of the information that physicians convey to their patients can inadvertently amplify patients’ symptoms and become a source of heightened somatic distress, an effect that must be understood by physicians to ensure optimal management of patient care. This effect illustrates the iatrogenic potential of information, as opposed to the iatrogenic potential of drugs and procedures.

Somatic symptoms and underlying disease do not have a fixed, invariable, one-to-one equivalence. Symptoms can occur in the absence of demonstrable disease, “silent” disease occurs without symptoms, and there is substantial interindividual variability in the symptoms resulting from the same pathology or pathophysiology. One mediator of this variability between symptoms and disease is the patient’s thoughts, beliefs, and ideas. These cognitions can amplify symptoms and bodily distress. Although cognitions may not cause symptoms, they can amplify, perpetuate, and exacerbate them, making symptoms more salient, noxious, intrusive, and bothersome.

Several common clinical scenarios exemplify the iatrogenic potential of the physician’s words—for example, instituting a new medication regimen, reviewing an informed consent document, presenting ambiguous laboratory test information, and preparing patients for painful procedures.

Knowledge of the nonspecific, diffuse, ambiguous adverse effects of a drug (such as fatigue, difficulty concentrating, nausea, dizziness, headache) increases the frequency with which they are experienced and reported.1 Whether the information is imparted through discussion with the physician or when obtaining informed consent, patients who learn about the common, nonspecific adverse effects of β-blockers, statins, estrogen-containing oral contraceptives, and agents for obstructive urinary symptoms report more of these putative adverse effects than comparable patients not informed of them. For example, in a study of 76 patients who received β-blocker treatment for hypertension, erectile dysfunction occurred in 32% of the 38 patients explicitly informed of this adverse effect and in 13% of the 38 patients not specifically warned about it.2 Likewise, in a double-blind trial of statins, the incidence of muscle-related adverse effects increased from 1.00% per year to 1.26% per year when patients (n = 9899) were subsequently unblinded and given the active drug.3

The nocebo phenomenon (the development of adverse effects to placebo) strongly supports that patient knowledge of adverse effects influences the reported incidence of these symptoms. Thus, the frequency and profile of adverse effects manifested by patients randomized to receive placebo in controlled, double-blind clinical trials are similar to those they have been told may occur with the active comparator drug.

Providing test results of dubious clinical significance also can lead to increased symptoms. For example, in a randomized study of acute low back pain, one group (n = 210) underwent spine imaging, whereas the other group (n = 211) did not. A treatment plan of conservative medical management was implemented in both groups. At 3-month follow-up, the former group had significantly more pain, greater functional impairment, and more physician visits.4 The problems involved in conveying equivocal test results or anatomical abnormalities of unknown clinical significance (“incidentalomas”) are likely to increase in importance as the volume and resolution of diagnostic testing accelerate.

Pain is particularly sensitive to the beliefs, thoughts, and expectations of patients. The specific language used in describing and preparing patients for painful procedures can affect the pain experience. For example, in a randomized study of women receiving epidural anesthesia or spinal anesthesia (n = 140) for childbirth, those told that the intradermal injection of a local anesthetic would “feel like a bee sting: this is the worst part of the procedure” reported significantly more pain than did those told “the local anesthetic…will numb the area and you will be comfortable during the procedure.”5 The importance of cognitions in the experience of pain is particularly timely, given the current crisis in opiate treatment for chronic, nonmalignant pain.


The Mechanism of Viscerosomatic Amplification

Viscerosomatic amplification has been proposed as an explanatory mechanism whereby information can affect the perception of symptoms.6 The information conveyed by the physician does not cause somatic symptoms but rather amplifies symptoms—symptoms that may be due to the underlying medical condition or to normal physiology (eg, ectopy, orthostatic hypotension), common benign dysfunctions (hoarseness, bloating, cramps), transient and self-limited ailments (rashes, upper respiratory tract infections), stressful life events, lack of exercise, insufficient sleep, or dietary indiscretion.

New medical information can initiate a self-perpetuating and self-validating cycle of symptom amplification (eFigure in the Supplement). Learning that a symptom may be more noteworthy or medically significant amplifies it. Reattributing the symptom to a new and more serious and more concerning source then causes the patient to monitor and scrutinize the symptom more closely, and this heightened attentional focus amplifies the symptom, making it more intense and intrusive, more disturbing and distressing.6,7 The misattribution also launches a selective search for additional symptoms to corroborate the suspicion that something is wrong, resulting in a heightened awareness of other diffuse, transient, or ambiguous symptoms that were previously ignored, minimized, or dismissed as insignificant. The seeming emergence of these “new” symptoms (along with dismissal of observations that do not confirm the suspected cause) is taken as further evidence of seriousness. The cycle of amplification is also fueled by mounting anxiety: Increasing worry and concern about the symptom’s medical significance, and its apparent worsening, make it more threatening and ominous.

 
Modulating Symptom Amplification and Minimizing Undue Distress

Careful attention to what and how information is conveyed can minimize the inadvertent fostering of excessive, disproportionate, and unduly bothersome symptoms. The initial therapeutic step is an exploration of the patient’s ideas about the symptoms: What is the suspected cause of the symptoms and their putative significance, what is the future course expected to be, what is most worrisome and troubling about them? The answers to these questions can lead to more realistic and reassuring discussion of the patients’ specific concerns.

Explaining the process of viscerosomatic amplification can be beneficial. Understanding that the interpretation of medical information can exacerbate and perpetuate symptoms, and learning about the processes of misattribution, selective attention, increased bodily scrutiny, and secondary anxiety, can have a palliative effect by providing patients with a more benign and reassuring explanation for their discomfort. The reassurance that the symptoms, however bothersome, are not medically harmful makes them less intrusive and more tolerable. The nocebo phenomenon provides a vivid, helpful, and nonpejorative illustration of the power of beliefs to amplify symptoms. This discussion also helps by emphasizing and encouraging adaptive coping to increase the tolerance of discomfort. It may be useful to prospectively identify patients at increased risk of disproportionate or undue nonspecific medication adverse effects, so that the amplification process can be explained to them in advance. This can be done with the Perceived Sensitivity to Medicines scale,8 a 5-item self-report questionnaire with demonstrated validity and reliability.

In addition to exploring the patient’s ideas and explaining the process of amplification, physicians need to use care in their choice of words. For example, in discussing potentially painful procedures, physicians can emphasize what will be done to alleviate the pain (such as simple relaxation techniques), use neutral language to describe the experience, and give the patient as much choice and control over the analgesic regimen as is medically feasible. Likewise, language is important in discussing nonspecific drug adverse effects, for example, focusing on the proportion of patients who do not have the adverse effects being enumerated, rather than on the proportion who do, and closely coupling information about benefits with information about adverse effects.

When obtaining informed consent, physicians must absolutely provide complete and truthful information and must avoid fostering a “paternalistic” patient-physician relationship. But balancing the requirement for full and complete disclosure with the iatrogenic potential of some information is problematic. “Contextualized informed consent” has been proposed as an ethical way of balancing these competing imperatives.9 When prescribing a medication, all serious and medically significant adverse effects are of course described, and the patient is instructed to report all adverse effects; however, if the patient agrees, benign, nonspecific symptoms are not enumerated in advance because it is explained that doing so makes them more likely.

Unusually distressing symptoms may point to difficulties in the patient-physician relationship, because symptoms can be a nonconfrontational and less explicit way of expressing doubts or concerns that patients feel and are unable or too embarrassed to voice openly. Symptoms can then become a covert, nonverbal communication of anxiety about pain, misgivings about medications, concerns about the meaning of a diagnostic test result, or concerns about the physician’s care or expertise.1

Conclusions
Information is an important mediator of the variability in the relationship between disease and symptoms. Some nonspecific drug adverse effects, undue pain from procedures, and symptoms exacerbated by learning about test results of unclear medical significance can be understood to share similar pathogenic mechanisms and respond to similar strategies of medical management.

Article Information 
 
Corresponding Author: Arthur J. Barsky, MD, Department of Psychiatry, Brigham and Women’s Hospital, 60 Fenwood Rd, Boston, MA 02115 (abarsky@bwh.harvard.edu).
Published Online: October 31, 2017. doi:10.1001/jama.2017.16216
Conflict of Interest Disclosures: The author has completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest and none were reported.
 
 
References
1.
Barsky  AJ, Saintfort  R, Rogers  MP, Borus  JF.  Nonspecific medication side effects and the nocebo phenomenon.  JAMA. 2002;287(5):622-627.PubMedGoogle ScholarCrossref
2.
Cocco  G.  Erectile dysfunction after therapy with metoprolol: the Hawthorne effect.  Cardiology. 2009;112(3):174-177.PubMedGoogle ScholarCrossref
3.
Gupta  A, Thompson  D, Whitehouse  A,  et al; ASCOT Investigators.  Adverse events associated with unblinded, but not with blinded, statin therapy in the Anglo-Scandinavian Cardiac Outcomes Trial-Lipid-Lowering Arm (ASCOT-LLA): a randomised double-blind placebo-controlled trial and its non-randomised non-blind extension phase.  Lancet. 2017;389(10088):2473-2481.PubMedGoogle ScholarCrossref
4.
Kendrick  D, Fielding  K, Bentley  E, Kerslake  R, Miller  P, Pringle  M.  Radiography of the lumbar spine in primary care patients with low back pain: randomised controlled trial.  BMJ. 2001;322(7283):400-405.PubMedGoogle ScholarCrossref
5.
Varelmann  D, Pancaro  C, Cappiello  EC, Camann  WR.  Nocebo-induced hyperalgesia during local anesthetic injection.  Anesth Analg. 2010;110(3):868-870.PubMedGoogle ScholarCrossref
6.
Brown  RJ.  Psychological mechanisms of medically unexplained symptoms: an integrative conceptual model.  Psychol Bull. 2004;130(5):793-812.PubMedGoogle ScholarCrossref
7.
Pennebaker  JW, Epstein  D.  Implicit psychophysiology: effects of common beliefs and idiosyncratic physiological responses on symptom reporting.  J Pers. 1983;51:468-496. PubMedGoogle ScholarCrossref
8.
Horne  R, Faasse  K, Cooper  V,  et al.  The perceived sensitivity to medicines (PSM) scale: an evaluation of validity and reliability.  Br J Health Psychol. 2013;18(1):18-30.PubMedGoogle ScholarCrossref
9.
Wells  RE, Kaptchuk  TJ.  To tell the truth, the whole truth, may do patients harm: the problem of the nocebo effect for informed consent.  Am J Bioeth. 2012;12(3):22-29.PubMedGoogle ScholarCrossref

Nenhum comentário: