3.2 - Clinical surveillance and patient safety

Summary

Introduction

Clinical surveillance is the routine collection of clinical data in order to detect and further analyse unusual health outcomes that may arise from a special cause. As in the closely related subject area of statistical surveillance, the aim is typically to isolate and understand special causes so that adverse outcomes may be prevented. Clinical surveillance is a way of providing appropriate and timely information to health decision-makers to guide their choice of resource allocation and hence improve the delivery of health care.

In order to detect unusual data points, first it is important to take account of the measurable factors that are known to affect the distribution and size of the data. Factors typically of key importance in clinical surveillance are discussed in the first section of this chapter. These include important aspects of clinical surveillance data that affect and govern analysis, including patient heterogeneity; the essential size of health-care facilities; and the dimensionality of the data. Given these essential factors, various statistical surveillance tools might be implemented. Statistical control chart options for surveillance are considered, keeping in mind the desirable characteristics of control charts – utility, simplicity, optimality and verity. A variety of such tools are discussed via example data, with an emphasis on graphical display and desirable characteristics. The graphs presented are based on data relating to cardiac surgery performed by a group of surgeons in a single cardiothoracic unit, and on data relating to the practice of Harold Shipman over the period 1987–1998.