Referral from secondary care and to aftercare in a tertiary care university hospital in Japan

Abstract

Background: In Japan, all citizens are covered by the national insurance system in which universal free access to healthcare services is promised to everybody. There are no general physicians or gatekeepers in the Japanese healthcare system.

Methods: We studied the pattern of referral of inpatients from secondary care hospitals to a tertiary care university hospital and the reverse referral under the situations using a geographic information system (GIS), taking paediatric inpatients as an example.

Results: The results showed that 61.2% of the patients were directly admitted to the hospital without referral from other hospitals or clinics and that 82.8% of the inpatients were referred to the outpatient department of the hospital to which they had been admitted. GIS analysis for the inpatients service area showed the hospital functions as both a secondary care hospital and tertiary care hospital. Patients who lived near the hospital tended to be admitted directly to the hospital, and patients who lived far from the hospital tended to utilize the hospital as a tertiary care provider. There were territorial disputes with other secondary care hospitals. To estimate spatial differences in referral to aftercare, we analyzed the spatial distribution of inpatients with focus on their length of hospital stay (LOS). GIS analysis revealed apparent foci of patients with long LOS and those with low LOS.

Conclusion: These results suggest that the function of university hospital in Japan is unspecialized and that the referral route from the university hospital to aftercare is also unequipped.

Figure 1

Distribution of residence of inpatients. (A) All inpatients (n = 8,177) admitted to Niigata University Hospital during the period from April 2003 to March 2005 were geocoded to a GIS map of Niigata Prefecture. (B) All inpatients under 15 years of age (n = 1,323) were also plotted on a map.

Figure 2

Medical service area for inpatients. (A) Empirical Bayesian estimates for the rate of inpatients per childhood population at risk were plotted on a map with 2 km mesh. (B) Magnification of (A). (C) The relationship between travel time by car from Niigata University Hospital (closed circle) or five other hospitals (open circles) in Niigata City and the number of inpatients covered within each area.

Figure 3

Referral from clinics and other hospitals. (A) Kernel density plot of patients referred from clinics and other hospitals. Dark regions represent clustering of these patients. The cross on the map shows the location of Niigata University Hospital. (B) Kernel density plot of the patients who had more complex conditions, i.e., patients who received surgical treatment and patients who had severe complications. (C) Superposition of (A) and (B). The contour represents density shown in (A). (D) Kernel density plot of patients admitted through emergency (filled contour) and patients who were taken to the hospital by ambulance car (contour). Dark contour represents the clusters of patients.

Figure 4

Referral to aftercare. (A) Kernel density plot of patients who were referred to the Outpatient Department of Niigata University Hospital. (B) Kernel density plot of patients who were referred to outpatient departments of other hospitals. Dots represent the locations of hospitals with paediatricians.

Figure 5

Clusters of the patients with high LOSSDS. (A) Contour map of LOSSDS of the patients was produced by interpolation using the ordinary Kringing method. Dark contours represent the clustering of patients with high LOSSDS. Cross and arrow represent the locations of Niigata University Hospital and a focus of clustering of patients with high LOSSDS, respectively. (B) The arterial roads around the focus shown in (A) and their 1,000 meter buffer are shown. (C) Hospitals around the focus shown in (A). The focus is located in an area 30 min by car from the two hospitals a and b. The locations of the two hospitals and their 30-min areas are shown as bold crosses and hatch patterns. The small crosses represent other hospitals.