A quantifiable alternative to double data entry

Recent articles in this journal have questioned the effectiveness of double data entry to enhance the quality of clinical trials data entered into the computer from case report forms. Although double data entry is widely used, no definitive agreement has been reached as to how to model and quantify the time/cost involved to perform double data entry and the exact gain in data quality derived from its use. In this paper, an alternative to the verification step of double data entry is described. This alternative procedure involves the verification of a sample of records from the data set created from the single entry of the data into the computer. The sampling procedure used for this alternative procedure is known in the statistical quality control literature as a continuous sampling plan. This type of secondary check of the data has the following advantages: (1) it is easily performed by data entry personnel, (2) the cost/time necessary to perform the check along with the gain in quality of the data set can be computed, and (3) the sampling plan can be constructed to attain data quality goals set by the data management staff. Examples of how to select the parameters for this type of data quality assurance procedure are given in the paper. Control Clin Trials 2000;21:94-102 (C) Elsevier Science Inc. 2000.