The demographic, psychosocial, and case-mix characteristics of a stratified random sample of 150 general hospital patients receiving varying amounts and types of consultation-liaison (C/L) psychiatry and C/L nursing services were examined in this retrospective study. Relationships between: (a) intensity and type of C/L services and nursing resource use, and (b) C/L intensity and utilization of C/L recommendations were also explored.

The analysis of the data indicated that there are several unique subpopulations of C/L service patients. Subjects in the high intensity group demonstrated significantly higher resource use in terms of: (a) length of hospitalization (LOS), (b) average standard nursing hours per day of hospitalization, (c) contacts by both a C/L psychiatrist and a C/L nurse, (d) both direct and indirect C/L contacts, and (e) discharge to another health care facility.

The low intensity group consisted of two subgroups: one subgroup of patients with psychiatric diagnoses or history upon admission to the hospital (concurrent psychiatric and physical disorders), and a second subgroup with a primary physical disorder and psychological symptoms (adjustment disorder or no psychiatric diagnosis).

The medium intensity group (representing the majority of C/L patients) was found to have a moderately positive correlation (r =.53, p $<$.01) between timing of the C/L referral during the hospitalization and LOS, suggesting that if C/L referrals were made earlier in the hospitalization, LOS may decrease. This group also had the largest amount of decrease in nursing resource utilization from before to after C/L initial contact.

Future studies should focus on specific subgroups of general hospital patients (both with and without C/L contact) in order to meaningfully determine the types, timing, and amount of C/L services that may be most effective in improving the quality and cost-effectiveness of health care.