An Economic Evaluation of a Proactive Consultation-Liaison Psychiatry Pilot as Compared to Usual Psychiatric Consultation on Demand for Hospital Medicine.

BACKGROUND: A growing literature demonstrates the value of the proactive consultation-liaison psychiatry model for health care systems through reductions in hospital length of stay. Few studies include financial outcomes. OBJECTIVE: We evaluated the return on investment of a 1-year proactive consultation-liaison psychiatry pilot in a hospital medicine unit. METHODS: We used a pre-post method with an active comparison of three hospital medicine units with regular psychiatric consultation on demand. We staffed the pilot unit with one full-time psychiatrist who focused on early case finding, close communication with unit staff, frequent follow-up visits, and proactive discharge planning. We assessed the 1-year change in mean length of stay for all patients receiving psychiatric consultation, from which we estimated the direct contribution margin from bed-backfills and the program's return on investment. RESULTS: On the pilot unit, the percent of discharges that received psychiatric consultation increased from 7.34% to 13.79%, and the length of stay for patients who consulted declined by a mean of 1.82 days (P < 0.05), as compared to an increase of 0.15 days (P > 0.05) on the usual-care units. The pilot unit reduction in length of stay saved 451.71 days in total, allowing for 73.81 bed-backfills, a direct contribution margin of $419,343 (95% confidence interval, $50,754 to $787,933), a net direct contribution margin of $298,245 (-$70,344 to $666,835), and an return on investment of 132% (-31% to 295%). CONCLUSION: This study strengthens existing evidence for the relative cost-effectiveness of proactive consultation-liaison psychiatry for hospital medicine compared with usual psychiatric consultation on demand.

The Impact of Integrated Psychiatric Care on Hospital Medicine Length of Stay: A Pre-Post Intervention Design With a Simultaneous Usual Care Comparison.

BACKGROUND: Psychiatric comorbidity is highly prevalent in general medicine inpatient settings and is associated with increased duration and cost of hospitalization. OBJECTIVE: To evaluate the impact of integrated, proactive psychiatric care on hospital medicine length of stay (LOS), expanding upon methods from earlier studies. METHODS: A full-time psychiatrist was dedicated to a single hospital medicine unit to focus on early case finding and intensified treatment, interdisciplinary communication, and discharge planning. To a pre-post intervention design, we added a simultaneous usual care comparison. We also added adjustments for age, sex, insurance type, and whether the patient was discharged home or to a facility. We included a sensitivity analysis to remove outliers for whom LOS was ≤30 days. RESULTS: Statistically significant differences in LOS occurred on the pilot unit in the pre-post analysis (-1.66 d, P = 0.04) and on the pilot versus control units in the intervention year (-1.91 d, P = 0.003). The differential pre-post change in LOS on the pilot versus control units revealed a positive trend but was not statistically significant (-1.59 d, P = 0.14). This more rigorous test approached statistical significance when patients with LOS >30 days were excluded (-1.15 d, P = 0.07). CONCLUSION: This analysis strengthens existing evidence that dedicated, proactive psychiatric services integrated into hospital medicine units lower LOS more than does usual psychiatric consultation upon request, particularly in patients with an LOS ≤30 days.

Evaluation of the appropriate use of a CIWA-Ar alcohol withdrawal protocol in the general hospital setting.

BACKGROUND: The Clinical Institute Withdrawal Assessment-Alcohol, Revised (CIWA-Ar) is an assessment tool used to quantify alcohol withdrawal syndrome (AWS) severity and inform benzodiazepine treatment for alcohol withdrawal. OBJECTIVES: To evaluate the prescribing patterns and appropriate use of the CIWA-Ar protocol in a general hospital setting, as determined by the presence or absence of documented AWS risk factors, patients' ability to communicate, and provider awareness of the CIWA-Ar order. METHODS: This retrospective chart review included 118 encounters of hospitalized patients placed on a CIWA-Ar protocol during one year. The following data were collected for each encounter: patient demographics, admitting diagnosis, ability to communicate, and admission blood alcohol level; and medical specialty of the clinician ordering CIWA-Ar, documentation of the presence or absence of established AWS risk factors, specific parameters of the protocol ordered, service admitted to, provider documentation of awareness of the active protocol within 48 h of initial order, total benzodiazepine dose equivalents administered and associated adverse events. RESULTS: 57% of patients who started on a CIWA-Ar protocol had either zero or one documented risk factor for AWS (19% and 38% respectively). 20% had no documentation of recent alcohol use. 14% were unable to communicate. 19% of medical records lacked documentation of provider awareness of the ordered protocol. Benzodiazepine associated adverse events were documented in 15% of encounters. CONCLUSIONS: The judicious use of CIWA-Ar protocols in general hospitals requires mechanisms to ensure assessment of validated alcohol withdrawal risk factors, exclusion of patients who cannot communicate, and continuity of care during transitions.