Quality and Data Base Management

Lagoe, R.J., et. al. (13 Aug 1999) "Analyzing hospital readmissions using statewide discharge databases." Nursing Care Quarterly 13(6): 57-67.

Why did the authors select hospital readmissions as a way to evaluate the quality of care for these DRGs?

In contrast to simply measuring the patient's lengths of stay, the swiftness or slowness of patient discharge rates, lengths of patient occupancy, or patient mortality, measuring hospital readmissions rates were through to be an effective means of evaluating quality of care. The readmissions rate specifically measures treatment efficiency, namely that readmission often occurs when a condition that was not likely to have been treated during the patient's first stay of occupancy. In other words, length of stay by definition is affected by the patient's severity of illness -- the fact that a hospital might treat more children with broken legs (necessitating overnight stay) than chicken pox (not necessitating overnight stay) is not a determinative factor of quality of care.

Valuing discharge rates might mean a hospital that repeatedly kept patients too long for minor complaints, or discharged patients with serious complaints might receive praise, depending on the valuation method of length of swiftness. Occupancy might mean a patient was moved to another hospital, but still received inappropriate treatment or did not receive timely treatment. Lastly, some ailments have a higher mortality rate than others, regardless of the quality of care that the patients received.

The goal of the study was to find some factor with extensive and wide-ranging data, that was available on individual hospital and community databases, but which would not be biased against a hospital if, for example it had a larger amount of cardiac or critical care patients in its treatment units or if a particular community had a large amount of patients requiring long-term stays and care. But patients who are repeatedly discharged and must return are often not being treated effectively, that is clear, regardless of the severity of their aliment and the length of their stay. (58) The purpose of the study was also to encourage helpful sharing of data about patients in the most inexpensive and efficient fashions possible, as determined by the structure of comparing data over the course of the study, through the use of less expensive and smaller frameworks of recorded data. (66)

In terms of methodology in data analysis, patients with illnesses requiring readmission and examination could be flagged in the database such as stroke patients. For the study, urban areas were usually selected because they provided the widest and most extensive range of data, and patient confidentiality was protected by through the use of algorithms attached to social security numbers. (62; 60)

2.Define and describe the two components which the authors selected to measure the quality of care in the eleven DRGs.

The uniformity of the 30-day interval to define readmission was one component selected in the setup. Thus the same definition could be applied across many databases. Also, a list of conditions that automatically required a return examination, such as strokes, was determined. (61) Then, the main two components were selected, that of the rate of readmissions of individual hospitals involved in the study and the determination of 'communities.' Rates of readmission were determined on an institution-by-institution basis, and then on a community-by-community basis.

3. Define the seven outcomes measures they selected to assess each component. An outcome measure is always expressed as a numerical value such as a number or a ratio or a percentage or a statistical relationship.

The first of the seven variables was the readmission risk levels by individual hospital -- in other words, depending on the hospital, when would a patient be more at risk of being readmitted, depending solely on the hospital they attended? Next, the second variable was determined by a patient's designated presence in a community -- depending on one's community location, when would a patient be more at risk of being readmitted, regardless of the hospital one attended, depending upon the community he or she was located in? Risks of readmission, for example, were higher on the West Coast than in the Northeast section of the nation. (64)