Bollettino epidemiologico nazionale

Hospital discharge records: Are they a useful tool for defining hospital service needs?

Enrico de Campora, Renato Pizzuti, Sergio Lodato and Noberto Cau

Regional Health Agency of Campania and Office of the Assessore, Naples

In 1999 in the Campania Region, an estimated 9,200 residents of the region (population 5.7 million) with cardiovascular diseases were hospitalized in other regions. Of these, 37% underwent cardiac or vascular surgery

In Italy, the region in which a patient resides is required to reimburse the region where they are hospitalized.  In 1999, admissions for cardiovascular diseasesoutside the region accounted for 20% of the extra-regional expenditures. In 2000, there was an additional 3% increase in emigration for cardiovascular diseases, and the amount paid out to other regions for these conditions was 46 million Euro (89 billion Lira), approximately 17% of total disbursements for health care outside the region. Slightly less than 70% of this sum was related to cardiovascular surgery.

On the basis of this information, in September 2001 the Regional Council of Campania developed a series of initial recommendations to the local health and hospital authorities for gradually reducing the trends in emigration (1). The first of the two objectives presented in this document was to evaluate the current operative capacity of existing cardiovascular surgery units and promote their optimal use. The present study presents data concerning this objective.

The analysis was conducted using the archives of the nationally standardized and computerized hospital discharge records for the Campania region (3). Unless otherwise specified, the five-year period 1996-2000 was used to examine data on Campania residents admitted to hospitals within and outside the region. Information on beds and indicators of the activity of the cardiovascular surgery units in the region are based on 1999 data. The analysis represents the data available as of November 26, 2001.

The numbers of hospital admissions in Italy were obtained from the Ministry of Health’s website for the years 1996-1999; 2000 data were not yet available (4). For this reason, comparisons between Campania and Italy were limited to 1996-1999. The population used to calculate rates were those published by ISTAT (the Italian National Statistics Institute) for 1998, the midpoint for the five-year interval under study.

In order to define an “ideal” case mix for cardiovascular surgery consisting of procedures that should be done in these units and to be able therefore to evaluate the function of the existing units in Campania, we identified a series of cardiovascular surgery-specific Diagnosis-Related Groups (DRGs). These DRGs were defined as those for which at least 50% of the cases during the five-year period had been hospitalised in a cardiovascular surgery unit.

Seven DRGs were thus defined as cardiovascular surgery-specific: 103-108 and 115 (heart transplant, coronary artery bypass, with or without cardiac catheterisation, cardiac valve procedures, with or without catherization, other cardiac surgical procedures, cardiac pacemaker placement with acute myocardial infarction, cardiac insufficiency, or shock.  For the first six DRGs, 85% or more of the patients with these diagnoses were indeed hospitalised in cardiovascular surgery units; for the seventh, it was 59%.

For the 7 DRGS included in the ideal case mix, the trends in hospitalization rates per 1000 residents were compared for Campania and for the remaining 20 regions of Italy. Because cardiac bypass (DRGs 106-107) accounted for more than half of the discharge diagnoses in the case mix, trends for these procedures were examined separately. There has been a steady increase in admission for the cardiovascular surgery-specific DRGs in Campania, (from 0.54 to 0.79 per mille residents), similar to the trend noted for the rest of Italy, where rates increased from 0.63 to 0.86 per 1000 residents). In Campania, the rate for bypass increased from 0.31/1000 in 1996 to 0.44/000 in 1999, while for the rest of Italy, the corresponding figures were 0.35 and 0.47/1000. (Figure).

To determine whether the hospitalisations outside the region could have been due to lack of beds, the occupancy rates were calculated for the various catchment areas in the region using 1999 data. As shown in the Table, the occupancy rates for cardiovascular surgery beds is consistently in excess of 70%, with a mean value for the region of 98%, near the limit of the total capacity. When the rates calculated for the seven cardiovascular surgery-specific DRGs, the values obviously decline, even though there are substantial differences between the various areas. The declines are particularly evident in the areas where most of the beds are in private facilities under contract with the public system rather than in public hospitals.

It is interesting to note the existence of a negative correlation (r=-0.8) between the rate of extra-regional hospitalization for bypass procedures and the occupancy rates based on the cardiovascular surgery-specific DRGs. For example, in the Salerno catchment area, where the DRG-specific occupancy rates are high, the rate of extra-regional hospitalisation is lower than for Campania as a whole, while in the Avellino, Benevento, and Caserta areas, the situation is the opposite, with low rates of bed occupancy but also high rates of extra-regional admissions. In other words, this may imply that where the number of surgical procedures is high, extra-regional hospitalisation rates decrease. Based on occupancy rates and duration of stay, these findings also suggest that the cardiac surgery beds in the Caserta catchment area are not being optimally utilized.

The cardiovascular surgery-related DRGS also can be used to estimate the overall needs in terms of admissions or hospital days for residents of the catchment area, wherever they are hospitalised. The calculated needs can then be compared against the actual situation to estimate the percentage of needs currently being met within the region. In this case, the percentage of needs being met for the 7 DRGs varies from 45.4% in the Caserta area to 75% in the Naples area.

Although it is possible to identify the areas of relative need, it is not possible to determine whether the unmet need is due to the lack of beds or to the low number of surgeries performed. For this reason, it is useful to examine (assuming that the services being delivered are appropriate) how the beds are being used. The specificity of the activity of the cardiovascular surgery units, measured as hospital days for the cardiovascular sugery-related-DRGs in the unit divided by the total hospital days in the same unit, varies from 43.1% in the Caserta area to 87.4% in the Salerno area.

Comment

Donato Greco, Director of the Laboratory of Epidemiology and Biostatistics, ISS

The article by de Campora is an example of operations research on health services designed to make the health planning process transparent, explicit, and reproducible, or in other words, open to discussion and evaluation based on the identification of actual health needs. The work done in Campania again raises the important issue of emigration to other regions to obtain health care and confirms the relationship between volume of surgical procedures and migration.

This article raises a number of interesting methodologic issues. One of these is the development of an ideal case mix. In addition to the 7 conditions selected for analysis in this study, other diagnoses may sometimes require treatment on a cardiovascular surgery ward and it is thus necessary to take into account such other conditions in calculating health needs.

A second methodologic question is the difficulty of comparing units because of possible differences in the severity of the illnesses they treat. Because of such problems, it is necessary to combine data on DRG with other systems that take into consideration disease severity.

A third concept regards catchment area and the minimum number of procedures that are required to maintain technical competence (4), a topic of considerable interest in other countries. Finally, none of the measures examined in this study address quality and efficacy of the services delivered; these are important elements to take into consideration when evaluating the distribution of utilization of health services.

These problems require more in depth discussions with the experts in the sector. Furthermore, the evaluation of utilization of services remains incomplete if not accompanied by the evaluation of the efficacy of such services. The ISS is pursuing this topic through a variety of initiatives designed to study the outcomes of health services, such the recently-launched pilot study on the outcomes of coronary bypass that is being done at the urging of the Ministry of Health.

Bibliography

1.      D.G.R della Campania n° 4183 del 14.09.01, B.U.R.C. n° 52 del 8.10.01 pag. 294 e seguenti;

2.      Archivio SDO, Agenzia Regionale Sanitaria, Regione Campania;

3.      http://www.sanita.it/sdo/dati/datiind.htm;

4.      http://www.hcfa.gov/research/mpqsdem.htm