Proactive psychological and psychiatric support of patients with chronic non-communicable diseases in a randomised trial: a Ukrainian experience

Introduction

Without question, the medical community recognises that many patients with organic diseases and other forms of bodily illness also suffer from psychiatric and psychological problems. Various models of medical care to address these needs have been proposed, with some showing greater efficacy than others, even as the models continue to evolve and new ones are developed. One model that has gained the attention of researchers worldwide is proactive psychosomatic medicine (PPM), a qualitative approach to providing psychological-psychiatric services to patients in non-psychiatric departments of general hospitals. PPM is based on the biopsychosocial theory and its clinical implementation—consultation-liaison psychiatry. The main principles include initiative, purposefulness, intensity and integration with general medical care.1 2

The psychological-psychiatric care system evolved in general medical practice because of the need for new mechanisms to manage the psychological and social aspects associated with diseases. If insufficiently addressed, inpatients’ unmet psychosocial needs interfere with their medical treatment effectiveness. Consequently, some patients may experience a more severe course of the disease and increase their length of hospitalisation while simultaneously placing a greater burden on the healthcare system.3 Meta-analyses findings on the concept of proactive psychiatric counselling, with confirmation by medical experts’ consensus,2–5 showed that from 2011 to 2018, 20% to 40% of patients in multidisciplinary hospitals also had a mental illness, significantly complicating the course of somatic pathology, the effectiveness of therapy and the prognosis, especially for non-communicable diseases.6–8

It is indisputable that patients’ mental illness in multidisciplinary hospitals often hinders timely discharge, leads to additional specialist consultations and increases total medical care costs. Based on the studies mentioned above, evidence supports psychological-psychiatric counselling as an important method for monitoring somatic patients. The main features of the proactive model of consultation-liaison psychiatry were identified: multidisciplinary care that includes joint supervision and close patient observation by a psychiatrist, primary care physician, psychologist, nurse and social worker. This team not only makes recommendations for treatment, risk reduction and crisis management—typical of the traditional care models—but also exerts efforts to prevent behavioural barriers to care, avoid crises and boost the synergy of patients.4

In 2019, HOME Study,9 a randomised controlled trial, compared the addition of proactive psychological medicine to usual care and then measured the amount of time older, acutely ill patients spent in the hospital. It also formulated the following specific recommendations for the organisation of inpatient psychiatric services and psychological and psychiatric interventions for patients of general medical practice, using the model of consultation-liaison psychiatry:

1. An early proactive biopsychosocial assessment of recently hospitalised patients to identify all problems, including mental illness.

2. The creation of a plan for comprehensive supervision and systematic management of the specific problems likely to cause potential obstacles to a rapid hospital discharge.

3. The implementation of a comprehensive treatment plan, including daily psychosomatic examinations to measure the patient’s progress.

4. Integrated work with the staff of various departments (doctors, nurses, other counsellors and social work specialists) and outpatient services to ensure the implementation of the comprehensive care plan.

In December 2020, a resource document on proactive consultation-liaison psychiatry, initiated by the Council of the American Psychiatric Association, was approved for publication. This document1 emphasises the implementation of a model of proactive consultation-liaison psychiatry that contains the following four elements:

1. Systematic screening focusing on current mental health problems in somatic patients (patients hospitalised in certain health facilities are systematically checked for signs of active mental health problems, especially those who may be at higher risk).

2. Early clinical intervention (proactive measures tailored to individual patients with a combination of interventions for somatic and mental disorders).

3. Providing care based on a multidisciplinary team approach (the mental health team is part of a multidisciplinary hospital and provides comprehensive mental healthcare directly in a general hospital).

4. Integration of care with primary teams and services (a proactive psychological-psychiatric team closely coordinates the work with primary services in real time between clinicians with relevant experience: doctor to doctor, nurse to doctor/nurse, social worker to social worker/rehabilitation specialist and vice versa).

Proactive models of psychiatric care are now increasingly recognised and widespread because of the effectiveness of their implementation in treating patients during the coronavirus disease 2019 (COVID-19) pandemic. For example, some hospitals created separate departments specifically for treating patients with COVID-19 with acute psychiatric needs10; others chose to develop psychological-psychiatric units or consultation-liaison psychiatry care within the established structure of multidisciplinary hospitals.11 12 The positive effects of reducing the treatment costs and length of hospital stay of patients with COVID-19 were emphasised.13–16 Based on the recent success of these approaches during the pandemic, proactive models of psychological-psychiatric care have been widely implemented in general practice as an essential link in managing somatic pathology. However, current scientific literature lacks information regarding their effectiveness in treating the anxiety associated with chronic non-communicable diseases. Therefore, this study aimed to evaluate the effectiveness of the proactive consultation-liaison psychiatry model, using the PPM conceptualisation, for the relief of subclinical symptoms of anxiety in patients with chronic non-communicable diseases.

Methods

The study design was a randomised controlled trial with parallel groups. It adhered to the ethical and deontological norms in accordance with the principles set out in the Declaration of Helsinki. Initially, 209 subjects were selected from inpatients hospitalised at Kyiv Railway Hospital No 1 for non-communicable diseases from June 2021 to December 2021. Inclusion criteria for study participants were as follows: (1) all men and non-pregnant, non-breastfeeding women aged 18 to 65 years, (2) diagnosis of a non-communicable somatic disease and (3) a score of fewer than 24 points on the Hamilton Anxiety Scale (HAM-A). Exclusion criteria for study participants were: (1) participation in another study within 1 month of screening, (2) a diagnosis of mental disorder, (3) a history of traumatic brain injury or stroke, (4) abuse of psychoactive substances at the time of screening and (5) the use of antidepressants for 1 month or less before enrolment in the study. Before the study began, all participants reviewed the protocol and signed informed consent. One hundred and ninety-three subjects met the inclusion criteria, and they were then screened for signs of current mental health problems, especially patients deemed to be at risk.

Research protocol

After receiving information about the study, providing written informed consent and undergoing a screening procedure, on day 1 of the study (T1), participants were administered the CQLS and SSS-8 questionnaires in addition to a structured clinical interview with the HAM-A to assess their anxiety. The groups’ sociodemographic information is shown in table 1. Group participation was determined by block randomisation to obtain equivalent groups. Patients were assigned to the PPM model study group (SG) (n=104) or the control group (CG) (n=89) who received standard treatment for chronic non-communicable disease according to established clinical protocols. After 60 days, a reassessment was performed using the HAM-A, SSS-8 and CQLS. Results are shown in table 2. The study design is presented in figure 1.

Table 1

View inline|View popup

Sociodemographic variables of study participants

Table 2

View inline|View popup

Repeated-measures ANOVA results for HAM-A, SSS-8 and CQLS

Request permission

Figure 1

Flowchart of study design and distribution of respondents into groups. CQLS, Chaban Quality of Life Scale; HAM-A, Hamilton Anxiety Scale; SSS-8, Somatic Symptom Scale-8.

Results

Of the 209 inpatients with non-communicable somatic illnesses who were eligible for participation, 16 refused participation and 193 were enrolled for screening procedures after being provided information about the study. Only data from the 193 subjects were analysed in the final processing of the results. The mean age of patients in the study group (n=104) was 36.08 (10.37) years, and the mean age in the control group was 38.78 (9.87) years (t=1.843, p=0.066). Of the total, 115 subjects (59.6%) were women (59.6% in SG, 59.6% in CG), while 78 subjects (40.4%) were men (40.4% in SG, 40.4% in CG) (p=0.992). One hundred and forty-five of the subjects (75.1%) at the time of the study were married (76.0% in SG, 74.2% in CG), and a minority (24.9%, n=48) were without a permanent partner (24.0% in SG, 25.8% in CG) (χ²=0.014, p=0.903). Almost all participants (n=185, 95.9%) were employed (97.1% in SG, 94.4% in CG). Only a few participants (n=8, 4.1%) were temporarily out of work (2.9% in SG, 5.6% in CG) (χ²=0.345, p=0.557). Thus, before randomisation, the groups had no significant difference in sociodemographic indicators (table 1).

Regarding participants’ somatic diagnoses, 101 (52.3%) had hypertension (48.1% in SG, 57.3% in CG, χ²=1.287, p=0.256), 56 (29.0%) had bronchial asthma (29.8% in SG, 28.1% in CG, χ²=0.010, p=0.918), 45 (23.3%) had diabetes mellitus (25.0% in SG, 21.3% in CG, χ²=0.182, p=0.669) and 16 (8.3%) had chronic obstructive bronchitis (4.8% in SG, 12.4% in CG, χ²=2.672, p=0.102). The majority of participants—167 (86.5%)—had one chronic non-communicable disease, while 26 (13.5%) had two comorbid diseases. Therefore, no statistically significant difference in morbidity was found before the randomisation of the groups.

At the start of the study, all 193 participants had subclinical anxiety as rated by the HAM-A scale: the mean score of the study group was 19.21 (1.04) and of the control group was 19.28 (1.06) (t=0.454, p=0.650). The initial SSS-8 measurements of participants’ somatic manifestation levels were as follows: 16 (8.3%) minimum, 50 (25.9%) mild, 61 (31.6%) moderate, 60 (31.1%) severe and 6 (3.1%) very severe. The average SSS score of the study group was 9.02 (3.81) and of the control group was 9.84 (3.96) (t=1.451, p=0.148). The initial CQLS results measuring participants’ quality-of-life levels at the study’s onset were as follows: 63 (32.6%) very low, 55 (28.5%) low, 41 (21.2%) moderate, 15 (7.8%) high and 19 (9.8%) very high. The average quality-of-life score of patients in the study group was 63.99 (15.21) and in the control group was 60.85 (14.50) (t=−1.458, p=0.146).

The results below show a reassessment of the presence and severity of anxiety, somatic symptoms and quality of life made at time point T2, 60 days after the initiation of proactive psychological and psychiatric support for the study group patients.

Anxiety

The median HAM-A value differed between groups by 5.22 points, with significantly lower results observed in the study group (14.34 (3.28), t=12.032, p<0.001) (figure 2), indicating a lower degree of anxiety in this group compared with the control group (19.56 (2.11)). The T2 HAM-A score for the study group also showed a significant decrease compared with the initial overall score (19.21 (1.04)). In addition, the results of repeated-measures ANOVA also revealed that the study group had better results over time than the control group (table 2).

Request permission

Figure 2

HAM-A values and IQR in control and study groups. HAM-A, Hamilton Anxiety Scale; IQR, inter-quartile range.

There was also a difference in the qualitative indicators of anxiety. At T1, all participants (n=193) had severe and very severe anxiety. At T2, of the 104 study group participants, 60 (57.7%) showed mild anxiety, while 30 (28.8%) showed severe and very severe anxiety. Data were missing from 14 (13.5%) participants. At the same time, a spontaneous but less common decrease in anxiety was found in the 89 control group participants, with 13 (14.6%) showing mild anxiety and 65 (73.0%) showing severe and very severe anxiety. Data were missing from 11 (12.4 %) participants. (figure 3). This difference was also statistically significant (χ²=50.90, p<0.001).

Request permission

Figure 3

Comparison of qualitative HAM-A values between study and control groups on the 1st and 60th day. HAM-A, Hamilton Anxiety Scale.

Somatic symptoms

In addition to the reduction in anxiety manifestations, the occurrence and severity of somatic symptoms were also significantly reduced in study group patients; somatic complaints and manifestations decreased significantly (p<0.001) within the group (table 2). Somatic manifestation levels of the 104 participants in the study group at T2 were as follows: 69(66.3%) minimum, 9(8.7%) mild, 8(7.7%) moderate and 4(3.8%) severe. Data were missing from 14(13.5%) participants. Somatic manifestation levels of the 89 participants in the control group participants at T2 were as follows: 10 (11.2%) minimum, 12 (13.5%) low, 33 (37.1%) moderate, 20(22.5%) severe and 3(3.4%) very severe. Data were missing from 11 (12.4%) participants (figure 4A).

Request permission

Figure 4

Comparison of qualitative SSS-8 values (A) and qualitative CQLS values (B) in all study participants on the 1st and 60th day. CQLS, Chaban Quality of Life Scale; SSS-8, Somatic Symptom Scale-8.

Quality of life

Improvement in quality of life was also recorded within the study group, with the difference between the two groups being statistically significant (p<0.05; table 2). At T2, levels of quality of life for the 104 study group participants were as follows: 18 (17.3%) very low, 28 (26.9%) low, 11 (10.6%) medium, 14 (13.5%) high and 19 (18.3%) very high. Data were missing from 14(13.5%) study group participants. In comparison, at T2, levels of quality of life for the 89 control group participants were as follows: 24 (27.0%) very low, 21 (23.6%) low, 20 (22.5%) medium, 5 (5.6%) high and 8 (9.0%) very high. Data were missing from 11 (12.4%) control group paticipants. A visual illustration of the data is presented in figure 4B.

The graph in figure 5 shows that after 60 days of proactive psychological and psychiatric support, anxiety symptoms decreased by 25.35% (a drop in median value by 4.87 points). The SSS-8 scale scores decreased even more by 63.08% (a decrease of the median value by 5.69 points). Moreover, CQLS quality-of-life indicators showed a 10.05% tendency toward improvement (an increase of the average value by 6.43 points).

Request permission

Figure 5

HAM-A, SSS-8 and CQLS score dynamics of the intervention and control groups on the 1st and 60th day. CQLS, Chaban Quality of Life Scale; HAM-A, Hamilton Anxiety Scale; SSS-8, Somatic Symptom Scale-8.

Discussion

Ethics approval

This study involves human participants and was approved by the Ethics Commission at the Kyiv Clinical Hospital on Railway Transport No 1 of the 'Health Care Center' Branch of the Joint-Stock Company 'Ukrainian Railway', protocol No 5 dated 20 May 2021. Participants gave informed consent to participate in the study before taking part.