Primary care physicians (PCPs) often misdiagnose patients with chest pain related to coronary heart disease (CHD). The physicians face the dilemma of deciding whether they should act immediately to treat a severe disease, or whether they should adopt a “wait and watch” strategy. While many patients are referred to a cardiologist for further evaluation and an ECG test to rule out cardiac complications, many other cases of CHD go unrecognized. This study was conducted to evaluate the incidence of misdiagnosis by primary care doctors and the reporting of false negatives.
Chest pain is among the most common reasons to visit a physician. However, the diagnosis can be tricky. Doctors attempt to check for CHD or acute coronary syndrome (ACS) first, simply because these condition can be lethal. Such cases, however, constitute only eight to 15 percent of patients complaining of chest pain in a primary care setting. It is common practice to continue monitoring the patient and to take ECG readings and/or get the opinion of a cardiologist. Very few studies have investigated the accuracy of initial diagnosis of chest pain in patients, especially in cases of CHD. This study assessed the initial diagnosis of CHD patients. It also analyzed follow-up data on the development of CHD in patients who initially were thought not to have the disease.
* The study recruited 1,249 patients, aged more than 35 years, complaining of chest pain to 74 different physicians.
* Thorough details of past medical history and on-visit examination were documented, including diagnosis and course of action. The PCP rated the probability of CHD.
* Follow-up calls were made at six weeks and later at six months, after which a committee consisting of a PCP, a cardiologist, and an internist analyzed the data.
* The accuracy of diagnosis was evaluated by comparing suspected vs. reference diagnosis.
* The PCPs estimated zero to 20 percent likelihood of CHD in 820 patients. Retrospective analysis by the panel diagnosed CHD as the cause of chest pain in 180 patients, of which 123 matched the PCPs’ diagnosis while the remaining 57 cases had been misdiagnosed as chest wall syndrome, myocarditis, cardiac insufficiency, cardiac arrhythmias, or other disorders.
* Of these patients, 42 were advised ECG, six were referred to a cardiologist initially, and seven were referred during follow-up.
* Two fatalities were reported even though the PCPs knew the patients’ history and were treating them appropriately.
* The study further assessed whether lowering the diagnostic threshold would be feasible. It concluded that the accuracy of diagnosis may only improve slightly while the list of unwanted tests would prove inconvenient and expensive for the patient.
Voluntary participation from the PCPs and hence high motivation levels could have yeilded biased results. The diagnostic skill of the participating investigators may have been overestimated. The study committee did not intervene with the functioning of the PCP and hence, only limited data was presented for some patients to the reference committee.
While most of the PCPs misdiagnosed CHD in some some patients, the fact that they recommended ECG for 72 percent of the patients and sent 37 percent of the cases to a cardiologist shows that they still considered the possibility of CHD. Increasing the number of tests at the primary level of diagnosis in patients with chest pain would result in unnecessary investigations and higher costs, especially when the patient is not suffering from CHD. This dilemma can be overcome not by increasing the sensitivity of diagnosis by PCPs, but by integrating the data from the patient’s history with the findings of the patient’s physical examination.
For More Information:
The Diagnosis of Coronary Heart Disease in a Low-Prevalence Setting
Publication Journal: Deutsches Ärzteblatt International, 2011
By Stefan Bösner, Jörg Haasenritter, et al.; The Universität Marburg, Germany