Differentiation Between “Normal” and “Abnormal” in Ambulatory Electrocardiography

Abstract

The establishment of normal limits for electrocardiographic features is based on a statistical description of a healthy population sample, but neither “normal” nor “healthy” has been unambiguously defined. At the tenth Bethesda Conference on Optimal Electrocardiography in 1977 (Abildskov 1978) a healthy population sample was defined as a representative sample of ambulatory noninstitutionalized persons without medical complaints or evidence by history or examination of any disease or dysfunction that may impair health or affect the electrocardiogram. None of the studies concerning ambulatory ECG recording in healthy subjects have been performed on representative healthy population samples, but there are several good reasons why. Healthy subjects are not immediately accessible, but must be drawn from a general population sample. This means that a lot of subjects would have to be examined and excluded from the study because they did not fulfil the requirements for healthy. “Nonhealthy” subjects are prevalent especially in higher age groups (Hinkle et al. 1969; Camm et al. 1978). Another problem is the high number of refusals, with approximately every third subject unwilling to participate (Raftery and Cashman 1976; Camm et al. 1978; Goulding 1978), making the requirement of a representative sample illusory. Many reports in the literature concerning healthy subjects therefore deal with subjects drawn from selected populations with a low incidence of disease and a low number of refusals (medical students, hospital staff, company employees, etc.). In studies concerning electrocardiographic features unknown to the subject himself such an approach may in most cases be acceptable, but it does involve a certain risk of not being sufficiently representative.