By S. Sanuyem. University of Notre Dame.
A study of admissions to dance rate in practice seems to be 5% in most British hospitals reported that 6% of the admitting diag- 25 discount 20mg benicar with visa blood pressure chart 19 year old,26 cases buy benicar 40 mg mastercard hypertension nclex questions. The emergency department requires Mammography has attracted the most attention in re- complex decision making in settings of above-average un- gard to diagnostic error in radiology. The rate of diagnostic error in this arena variability from one radiologist to another in the ability to 14,15 ranges from 0. A recent study of breast cancer found that the nostic error in clinical medicine was approximately 15%. In diagnosis was inappropriately delayed in 9%, and a third this section, we review data from a wide variety of sources 29 of these reﬂected misreading of the mammogram. Several studies have ex- frequently recommending biopsies for what turn out to be amined changes in diagnosis after a second opinion. Given the differences regarding insurance 17 coverage and the medical malpractice systems between and associates, using telemedicine consultations with spe- cialists in a variety of ﬁelds, found a 5% change in diagno- the United States and the United Kingdom, it is not sis. There is a wealth of information in the perceptual surprising that women in the United States are twice as specialties using second opinions to judge the rate of diag- likely as women in the United Kingdom to have a neg- 30 nostic error. It is important to emphasize that only a fraction of the 18,27,31–46 studies that have measured the rate of diagnos- discordance in these studies was found to cause harm. An unsettling consistency emerges: the frequency of diagnostic error is disappoint- Dermatology. For exam- tions and disorders where rapid and accurate diagnosis is ple, in a study of 5,136 biopsies, a major change in diag- essential, such as myocardial infarction, pulmonary em- nosis was encountered in 11% on second review. Of6 at ien t w ho died o fp ulm o n ar y em b o li m , he diagn o s i w as n o t us ect ed clin ically in L eder le et al( up ur ed ao r ic an eur ys m eview o fallcas es at a in gle m edicalcen t er o ver a yr er io d. Of2 cas es in vo lvin g ab do m in alan eur ys m s vo n o do li ch et al diagn o s i o fr up ur ed an eur ys m w as in iially m i ed in in at ien t es en t in g w ih ches ain , ( diagn o s i o fdi ect in g an eur ys m o ft he p o xim alao r a w as m i ed in o fcas es E dlo w Sub ar achn o id hem o r hage Up dat ed eview o fp ub li hed udies o n ub ar achn o id hem o r hage: ar e m i diagn o s ed o n in iialevaluat io n B ur o n et al( an cer det ect io n ut o p y s udy at a in gle ho s ial o ft he 2 m align an t n eo p las m s fo un d at aut o p y, w er e eiher m i diagn o s ed o r un diagn o s ed, an d in o ft he cas es he caus e o fdeat h w as judged o b e r elat ed o he can cer B eam et al( eas can cer accr edied cen t er agr eed o eview m am m o gr am s o f7 w o m en , o fw ho m had b r eas can cer he can cer w o uld have b een m i ed in M cG in n i et al( elan o m a Seco n d eview o f5 b io p y s am p les diagn o s i chan ged in fo m b en ign o m align an t fo m m align an t o b en ign , an d had a chan ge in um o r gr ade) Per li i o lar di o r der The in iialdiagn o s i w as w r o n g in o fp at ien t w ih b i o lar di o r der an d delays in es ab li hin g he co r ect diagn o s i w er e co m m o n G affet al( en dicii et o s ect ive s udy at ho s ial o fp at ien t w ih ab do m in alp ain an d o p er at io n s fo r ap en dicii Of1 p at ien t w ho had ur ger y, her e w as n o ap en dicii in o f9 at ien t w ih a ﬁn aldiagn o s i o f ap en dicii he diagn o s i w as m i ed o r w r o n g in R aab et al( an cer at ho lo gy The feq uen cy o fer o r in diagn o s in g can cer w as m eas ur ed at ho s ial o ver a yr er io d. The autopsy has been described as “the What Percentage of Adverse Events is 47 most powerful tool in the history of medicine” and the Attributable to Diagnostic Errors and What “gold standard” for detecting diagnostic errors. Richard Percentage of Diagnostic Errors Leads to Cabot correlated case records with autopsy ﬁndings in Adverse Events? In the Harvard Medical Practice Study of tween clinical and autopsy diagnoses were found in a 30,195 hospital records, diagnostic errors accounted for more recent study of geriatric patients in the Nether- 58,59 50 17% of adverse events. On average, 10% of autopsies revealed that the study of 15,000 records from Colorado and Utah reported clinical diagnosis was wrong, and 25% revealed a new that diagnostic errors contributed to 6. Using the same methodology, the Canadian a fraction of these discrepancies reﬂected incidental ﬁnd- Adverse Events Study found that 10. The Qual- crepancies that potentially could have changed the out- ity in Australian Health Care Study identiﬁed 2,351 ad- come were found in approximately 10% of all verse events related to hospitalization, of which 20% 32,51 autopsies. A large study in New Zealand examined 6,579 cause the diagnostic error rate is almost certainly lower inpatient medical records from admissions in 1998 and among patients with the condition who are still alive, found that diagnostic errors accounted for 8% of adverse 63 error rates measured solely from autopsy data may be events; 11. That is, clinicians are attempting to make the diagnosis among living patients before death, so the more Error Databases. Although of limited use in quantifying relevant statistic in this setting is the sensitivity of clin- the absolute incidence of diagnostic errors, voluntary error- ical diagnosis. For example, whereas autopsy studies reporting systems provide insight into the relative incidence suggest that fatal pulmonary embolism is misdiagnosed of diagnostic errors compared with medication errors, treat- approximately 55% of the time (see Table 1), the misdi- ment errors, and other major categories. Out of 805 volun- agnosis rate for all cases of pulmonary embolism is only tary reports of medical errors from 324 Australian physi- 32 cians, there were 275 diagnostic errors (34%) submitted 4%. Shojania and associates argue that a large discrep- 64 ancy also exists regarding the misdiagnosis rate for myo- over a 20-month period. Compared with medication and treatment errors, diagnostic errors were judged to have cardial infarction: although autopsy data suggest roughly caused the most harm, but were the least preventable. A 20% of these events are missed, data from the clinical smaller study reported a 14% relative incidence of diagnos- setting (patients presenting with chest pain or other rel- tic errors from Australian physicians and 12% from physi- evant symptoms) indicate that only 2% to 4% are missed. Mandatory error-reporting sys- tems that rely on self-reporting typically yield fewer error reports than are found using other methodologies. One method of test- ample, only 9 diagnostic errors were reported out of almost ing diagnostic accuracy is to control for variations in case 1 million ambulatory visits over a 5. One such Diagnostic errors are the most common adverse event approach is to incorporate what are termed standardized 67,68 reported by medical trainees.
Each of these different sites of care within the same organization maintains a different medical record of its encounters with same patient order benicar 10 mg mastercard blood pressure medications list. These separate systems were primarily built to bill for each department’s services generic 20mg benicar fast delivery heart attack 90 percent blockage, not to guide patient care. There is also a nearly impermeable barrier between the hospital’s records and those of the physicians who direct the care. In the typical community hospital, it is impossible for the doctor or any other care worker to access the doctor’s ofﬁce records from any site other than that doctor’s ofﬁce because more than 80 percent of those ofﬁce records are still in paper form. Furthermore, most doctors in private medical practice have been unwilling to support shared digital record-keeping systems with their hospitals because of a profound lack of trust and poor communication with hospital management. Even where it is possible to link all of these fragments of a pa- tient’s history and medical situation electronically, a considerable feat of software engineering is required to move this information around quickly enough that it can actually be used by the physician in making important care decisions. When information reaches a digital dead end, it is printed out and piled up in various in-boxes or paper ﬁling systems. Thus, vital information remains locked up in paper, or in people’s short-term memories, and cannot ﬂow through wire or ﬁber or the air to where it is needed to make timely and accurate medical decisions. As long as the source documents detailing patient care remain in paper form, the only way to determine whether particular clinical decisions contributed to a positive health outcome is to hire squads of graduate students or nurses to cull the records by hand months later and tabulate the results. The fact that we know so little about 4 Digital Medicine what actually works in medical treatment can be attributed in large part to the prison of paper we have constructed around the care process. Public research in- vestments through the National Institutes of Health and private equity investment, including research and development expendi- tures by the nation’s pharmaceutical and biotechnology ﬁrms, are creating new medical knowledge at a stunning pace. In 2001, nearly $51 billion was invested in creating new knowl- edge in medicine just in the United States. The logistics of medical practice itself have become so dauntingly complex that physicians barely have time for their families, let alone time to keep pace with the exciting advances in their own ﬁelds. A monthly continuing medical education session at the hospital or local medical society and periodic visits from pharmaceutical salespeople are the principal conduits of new knowledge to most practicing physicians. A visit to a physician’s personal ofﬁce typically reveals piles of unread medical journals, pink telephone message slips, and scattered samples of new drugs from the last pharmaceutical sales representa- tive to visit the ofﬁce. The computer in the ofﬁce is probably turned off, is likely at least three years old, and is surely not conﬁgured to The Information Quagmire 5 reach or retain data about current medical practice. Physicians know they are not keeping up, and this both frustrates and frightens them. Professional training and culture in medicine conditioned physi- cians to rely principally on direct peer contact and what they can carry around in their memories to support their advice to patients. The channels by which new information reaches physicians in prac- tice are dangerously narrow and lack the bandwidth and intelligence to organize and transmit the ﬂood of new medical knowledge in a way that it can be absorbed and used in practice. Although major journals have digitized their articles and made them available online, medical knowledge is still largely paper driven. Unless physicians have a good relationship with a medical librarian or, as do a lucky few senior physicians in teaching hospitals, have residents and fellows to research issues for them, the large number of important questions about a patient’s health that occur to a physician during a typical practice day never get answered. How- ever, the reality is that the lack of timely and accurate information at the point of care is a major contributor to patient deaths and injuries, as well as resulting in a waste of time and money. As a vehicle for applying medical knowledge to solving problems, the healthcare system has become increasingly cumbersome, user- unfriendly, and expensive. When the Internet opened up new channels for consumers to access medical knowledge directly, it was rapidly ﬂooded with users. According to a recent Harris poll, roughly 110 million Americans used the Internet to seek health information in 2002. According to Peter Drucker (see Note 1), large healthcare institutions, like urban academic health centers, may be the most complex organizations in human history. Not only do the medical problems presented at the point of service vary tremendously, but no inventory exists; health services are, for the most part, custom manufactured for individual patients on a “just in time” basis. For most healthcare, there is no template on which physicians can rely to make decisions about health. This is because professional consensus on what best practice is or ought to be is only now emerging. Perhaps most signiﬁcantly, more complex, highly trained health professionals collide at the point of care than in any other business in our economy. Each profession has its own unique view of the patient’s needs, its own language for describing those needs, and an intensely territorial view of its involvement in care. Collaboration The Information Quagmire 7 among professionals is vital to effective care, yet professions compete for resources and control over patients.