Objective: The study was conducted to assess the implementation of a psychiatric consultation-liaison service (C-L) from the perspective of cost-savings, staff satisfaction, patient satisfaction and to assess the general features of patients referred to the C-L service. Methodology: Cost-savings were evaluated using a large cohort of referrals to the hospital were identified using data derived from the Institute of Clinical Evaluative Sciences (N = 2246); these data were divided into pre and post periods with respect to imitation of the C-L service. To evaluate staff satisfaction, 170 nurses and physicians completed an online survey. Patient satisfaction was assessed through a survey assessing various aspects of their experiences with the C-L service that was completed by each patient (N = 40). Finally referrals to the C-L service (N = 445) were analyzed to discern indicators of the C-L service’s efficacy (i.e. reasons for referral, time to accommodate referral). Results: The data indicated: 1) a reduction in the number of re-admissions and length of stay after the initiation of the C-L service translating into significant cost-savings for the hospital, 2) that increased staff satisfaction was associated with providing confidence, support, and improved communication, and 3) that the C-L service accommodated approximately 90% of patients within 1 day. Conclusion: The results of this study support stakeholders’ decisions to implement C-L services and also indicate areas of improvement that may improve the quality of C-L services within other institutions.
Consultant-liaison (C-L) psychiatry refers to services provided to medically-ill patients within a general hospital setting. As summarized by Hamburg [
According to Gomez [
Strategies aimed at optimizing patient outcome while reducing cost and wait times are an ongoing priority at Health Sciences North and throughout the Province of Ontario. The C-L service at this hospital was initiated in March 2014 and is directed at supporting patients admitted for a primary medical concern that have co-morbid psychiatric illness, whether previously diagnosed or not. The service provides primary caregivers the support and psychiatric services necessary to optimize the patients’ mental health while they are in hospital receiving treatment for their primary health care issue. While the service is currently in its infancy, its growth will include nursing and allied health care support, which will further improve patient care. It is expected that long term costs will also be reduced, since providing this service will: alleviate wait times for the patient; reduce or prevent internal transfers to psychiatric wards; and provide collaborative support for the primary care physician with respect to medication interaction, therapy options and outpatient support. Further, studies find that patients are more likely to continue to seek care and respond to recovery if the therapy is initiated and organized prior to discharge from hospital [
Here we present a detailed examination of the clinical and financial outcomes of implementing a C-L service during the first three years of operation. In the series of studies that follow, we aimed to characterize 1) staff and patient satisfaction with regards to their perceptions of the C-L service, 2) demographics and descriptions of patients that have utilized the service, and 3) overall efficacy of the service.
To discern nurse/physician satisfaction with the consultation liaison service implemented at Health Sciences North, we invited 170 clinicians (N = 45) and nurses (N = 125) from various departments located on the hospital premises to complete a survey accessible online through Survey Monkey. The survey was made available and conducted within the time period of October 2015 and January 2016. The nurses/physicians came from a wide variety of clinical departments including family medicine, critical care, emergency, cardiology, and intensive care. In addition, a total of 40 patients who were referred to the C-L service completed a similar questionnaire.
We developed a questionnaire to assess the staff satisfaction with the C-L service and created two versions which represented the two general practices: nursing and physician. These items were uploaded to Survey Monkey and a link containing the questionnaire was e-mailed to the nurses and physicians separately. Client satisfaction was similarly assessed using a questionnaire which consisted of items which captured overall patient experiences, their willingness to utilize the C-L service in the future, as well as various aspects of their ongoing treatment after consultation with the C-L service. The responses to both staff and patient questionnaires were imported into Excel where frequencies were tabulated for each of the responses.
Clinical outcomes of the consultation psychiatric service were attempted by conducting a chart review. A total of 217 unique records covering 445 referrals (N =445) were available for the analysis. For each patient seen by the consultation liaison service, two broad streams of information were obtained from 1) the referring physician who disclosed demographic information, the reason for referral, and observed behaviours, and 2) the consultation liaison psychiatrist who collected information regarding when the referral was received/reviewed, diagnostic impressions, medication review, treatment, recommended interventions, patient care management, and follow up plan at discharge. The data was then consolidated at the Psychiatric Outpatient Clinic and entered into an Excel spreadsheet, coded, and entered into SPSS software for descriptive statistical analysis.
Participants were individuals between the ages of 18 to 59 years of age, admitted to floors of HSN other than psychiatry for a primary condition other than psychiatry or psychological but were discharged within the three years prior to and following the inception of the C-L service within the study period of April 1, 2011 to March 31 2016. Consequently, participants were categorized into two groups: 1) before C-L service initiation, and 2) after C-L service initiation. All data was extracted and summarized by the Institute for Clinical Evaluative Sciences (ICES). Participants were excluded if they were admitted to HSN floors with a primary psychiatric condition or possessed invalid health card numbers.
Based on the inclusion and exclusion criteria, the last discharge date of the first episode of care was kept for each person. Using the Registered Persons Database (RPDB) data, baseline characteristics (e.g., age, sex, income quintile, etc.) of the cohort were determined. Using the Discharge Abstract Database (DAD), hospitalization data such as average length of stay and readmissions were pulled for the cohort. To determine Emergency Department visits, the National Ambulatory Care Reporting System (NACRS) data was utilized. When extracting data from NACRS, comorbid psychiatric or psychological illness was defined as having a psychiatric or psychological illness in “other” dx10codes (2-10) whereas primary psychiatric or psychological medical conerns were defined as having psychiatric or psychological illness as the “main” dx10code (1). Using the cohort, the Ontario Health Insurance Plan (OHIP) billing was extracted to determine the frequency of outpatient billing/visits during the pre and post time frames. To identify patient transfers from hospitalization to mental health beds the last date of each patient’s hospitalization record was kept. The key number of the last hospitalization per patient was then linked to Ontario Mental Health Reporting System (OMHRS) to identify if a matching record with the same patient information and subsequent admission date existed. Similarly, to identify patient transfers from mental health beds to hospitalization, OMHRS data was pulled for the cohort. The key number from the last mental health bed date per person was then linked to DAD to identify if a patient transfer existed.
All data was pooled and separated into the two groups representing the pre and post-periods of the initiation of the CL service at HSN. Data was then summarized using descriptive statistics (frequency, mean, standard deviation) for the following measures: 1) hospital length of stay, hospital readmissions for both the total period as well as 6 months following the last discharge, patient transfers, emergency department visits, number of outpatient visits with general practitioners/family physicians as well as psychiatrists. These data, particularly the length of stay, formed the basis for calculating an estimated cost-savings of implementing the C-L service.
All ethical clearance for each study was approved by the Health Sciences North Research Ethics Board.
There were a total of 170 respondents to the satisfaction questionnaire which included 125 registered and practical nurses as well as 45 physicians. Preliminary one-way analyses of variance on individual items assessing the quality and effectiveness of the CL service indicated that there were no mean differences in how nurses and physicians responded to the items (p > 0.05). Therefore all data was pooled into a single database. A summary of responses to the set of core items assessing the quality and effectiveness of the survey as rated by both the physicians and nursing staff is presented in
An exploratory stepwise multiple regression analysis was performed to predict overall satisfaction of the CL service from the discrete survey items assessing various indices of CL service performance as viewed by the physician and nursing staff. Consequently all data was translated along a ratio scale by dividing the responses to the items by the total number of options for each item minus 1; for example the response “Quite dissatisfied” on item 15 would be 0 and very satisfied would be 3/3, or 1. The analysis demonstrated a significant relationship (Multiple R = 0.904, Adjusted R2 = 0.808) between overall satisfaction which could be predicted by three general constructs listed in decreasing importance as inferred by the beta standardized coefficient: 1) the ability for the CL service to help physicians and nurses treat their patients more effectively, 2) the ability for the CL service to endow confidence with addressing mental health, and 3) the verbal quality of the CL service. The B and standardized beta coefficients are listed in
A total of 40 patients completed the satisfaction questionnaire at the end of their consultation with the C-L service. A summary of their responses of selected items within the questionnaire are shown in
| Item | Option | Number of Endorsed Responses (%) | N |
|---|---|---|---|
| How quickly did a psychiatrist see your patient? | Within 24 hours | 35.1 | 77 |
| Within 48 hours | 33.8 | ||
| Within 72 hours | 13.0 | ||
| It took longer than 72 hours | 18.2 | ||
| Did the CL Service psychiatrist understand the core question being asked? | Always | 23.1 | 78 |
| Usually | 52.6 | ||
| Sometimes | 23.1 | ||
| Never | 1.3 | ||
| How would you rate the quality of verbal communication with the medical team in addition to charting patient record? | Excellent | 12.2 | 74 |
| Good | 28.4 | ||
| Satisfactory | 21.6 | ||
| Fair | 25.7 | ||
| Poor | 12.2 | ||
| Did the CL service psychiatrist offer practical and insightful medication choices for your patient? | Always | 13.5 | 74 |
| Usually | 37.8 | ||
| Sometimes | 39.2 | ||
| Never | 4.1 | ||
| Medications were not ordered | 5.4 | ||
| Did the CL service psychiatrist quickly manage the patient’s behavioural problems? | Always | 5.4 | 74 |
| Usually | 32.4 | ||
| Sometimes | 51.4 | ||
| Never | 5.4 | ||
| Not applicable | 5.4 | ||
| Did the CL service psychiatrist provide outpatient follow-up or referral to outpatient counseling management services? | Always | 13.5 | 74 |
| Usually | 31.1 | ||
| Sometimes | 43.2 | ||
| Not applicable | 12.2 | ||
| Did the CL service psychiatrist improve your confidence in addressing mental health issues with your patient? | Yes, they helped a great deal | 21.1 | 76 |
| Yes, they helped somewhat | 43.4 | ||
| No, they didn’t really help | 32.9 | ||
| No, they seemed to make things worse | 2.6 |
| Has the CL service helped you treat your patients more effectively? | Always | 15.6 | 77 |
|---|---|---|---|
| Usually | 31.2 | ||
| Sometimes | 44.2 | ||
| Never | 9.1 | ||
| Are you more confident in treating patients with psychiatric co-morbidities knowing the CL service is available to support you? | Always | 23.7 | 76 |
| Usually | 21.1 | ||
| Sometimes | 38.2 | ||
| Never | 17.1 | ||
| Overall, how satisfied are you with the CL service? | Very satisfied | 22.4 | 76 |
| Somewhat satisfied | 30.3 | ||
| Indifferent or mildly satisfied | 25.0 | ||
| Quite dissatisfied | 22.4 |