(PO-097) (PO-097) Follow-up and re-admission data as feasible quality measures for a C-L service

Significance: Outcome measures are increasingly important in healthcare, and feasible measures for consult-liaison psychiatry (CL) are lacking (Oldham 2008, Kroll 2020). Our goal was to analyze whether CL encounters predicted a differential likelihood of mental health (MH) follow-up or re-admission rates using a method requiring minimal manual inputting of data.

Methods: This retrospective review for quality improvement purposes was based on encounter diagnosis and active problem list codes in an electronic health record for patients admitted to medical floors at the San Diego VA Medical Center between January-June 2019. Data on patient sex, age, race, MH diagnosis, prior MH admissions or outpatient encounters, and CL encounters were analyzed with logistic and linear regression. MH diagnoses were categorized as serious mental illness (SMI), defined by psychotic or bipolar disorders; other mental health diagnoses, defined by PTSD, depressive disorders, and anxiety disorders; substance use disorders; and neurocognitive disorders.

Results: The sample included 2,681 unique admissions, of whom 92 received CL services during their hospital course. Mean age was 67; 92.7% identified as male, 71.5% white, and 84.7% non-Latinx. Nearly half (n=1,259) had at least one MH diagnosis. After accounting for race, sex, age, and recent MH treatment, those seen by CL were 1.95 times more likely to have MH follow up in the month after discharge (p=.011). CL visits predicted greater likelihood of psychiatric admission over the following year, and 80% lower likelihood for medical re-admission over the same time period. CL visits did not predict greater likelihood of MH follow-up for patients with SMI but did predict greater likelihood for those with other mental health diagnoses.


Discussion: Our data suggest CL contact increases rates of MH follow up across all patients and potentially identifies that patients are at heightened risk of psychiatric admission in the following year. Though this may be a reflection of the illness severity of patients for whom CL consultation was requested, it may also serve as a potential outcome measure which may lead to the development of increased post-discharge engagement, particularly with SMI patients.


Conclusions: Encounters with CL increased the likelihood of MH follow up overall, though our analysis also identified areas of improvement for patients with SMI. Data from this study were extracted with minimal manual data input or individual chart review, confirming the ability to repeat this analysis on a regular basis for performance measurement in the future, including the impact of resources, if applied.

References:
Kroll DS, Gopal A, Kimmel RJ, Mattson J, Beizai K, Danovitch I. Performance measurement tools for consultation-liaison psychiatry services must consider feasibility. Gen Hosp Psychiatry. 2020 May-Jun;64:46-49.

J.M. Oldham, W.E. Golden, B.M. Rosof. Quality improvement in psychiatry: why measures matter. Psychiatr Pract, 14 (Suppl. 2) (2008 May), pp. 8-17