Clinical and Functional Outcomes in Day Hospitals in Times of COVID-19

Abstract

Objective

Mental health services have been clearly affected by the pandemic and its restrictions. In our day hospital units users attended on fewer days, group therapy was drastically reduced, the number of users in groups was limited and individual and group teletherapy was initiated. This study evaluates the impact of these changes on clinical and functional outcomes.

Methods

This is a descriptive study prospectively comparing two cohorts of users: those admitted and discharged from our day hospitals before or during the pandemic. All users received a multidisciplinary assessment at admission and discharge including clinician and user-rated scales on psychopathology, daily living skills, quality of life and caregiver burdens. Improvement in both groups was compared.

Results

Both cohorts were homogenous at admission. Length of admission was not different for the two groups, but users admitted during the pandemic attended the day hospital on significantly fewer days. Improvement observed in most scales was equivalent in both groups, suggesting a similar evolution of both cohorts.

Conclusions

Despite the reduction in assistance days and group therapeutic interventions no major changes were found in the outcomes of both groups. This study shows that careful adaptations and availability, even in critical situations, can result in equally effective treatments. Further research is essential to determine which of these adaptations should be maintained beyond the pandemic.

Resumen

Objetivo

Los servicios de salud mental se han visto claramente afectados por la pandemia y sus restricciones. En nuestras unidades de hospital de día los usuarios acudieron en menos días, la terapia de grupo se redujo drásticamente, el número de usuarios en grupos fue limitado y se iniciaron terapias individuales y grupales en formato online. Este estudio evalúa el impacto de estos cambios en los resultados clínicos y funcionales.

Métodos

Se trata de un estudio descriptivo que compara prospectivamente dos cohortes de usuarios: los ingresados y dados de alta de nuestros hospitales de día antes o durante la pandemia. Todos los usuarios recibieron una evaluación multidisciplinaria al ingreso y al alta, incluyendo escalas auto y heteroadministradas de psicopatología, habilidades de la vida diaria, calidad de vida y cargas del cuidador. Se comparó la mejoría en ambos grupos.

Resultados

Ambas cohortes eran homogéneas al ingreso. La duración del ingreso no fue diferente para los dos grupos, pero los usuarios ingresados durante la pandemia asistieron al hospital de día significativamente menos días. La mejoría observada en la mayoría de las escalas fue equivalente en ambos grupos, lo que sugiere una evolución similar de ambas cohortes.

Conclusiones

A pesar de la reducción de los días de asistencia y las intervenciones terapéuticas grupales, no se encontraron cambios importantes en los resultados de ambos grupos. Este estudio muestra que las adaptaciones cuidadosas y la disponibilidad continuada, incluso en situaciones críticas, pueden facilitar tratamientos igualmente efectivos. Es esencial continuar investigando para determinar cuáles de estas adaptaciones deberían mantenerse más allá de la pandemia.

Introduction

The first known cases of COVID-19 were detected in Wuhan, China, in December 2019. COVID-19 has rapidly spread worldwide being declared by the World Health Organization (WHO) a Public Health Emergency of International Concern on the 30th of January 2020 and a pandemic on the 11th of March 2020. In an attempt to control the pandemic, many countries imposed strict quarantine measures and even when the quarantine finished, important restrictions on mobility and social relations prevailed. The WHO has expressed its concern over the pandemic's mental health and psycho-social consequences,¹ speculating that the impact of restrictions and quarantine on usual activities, routines, and livelihoods of people may lead to an increase in loneliness, anxiety, depression, insomnia, substance use, and self-harm or suicidal behavior. People with a previous psychiatric history have been considered a vulnerable group of population regarding the effects of the pandemic.²

Mental health services have been clearly affected by the pandemic³: the number of psychiatric beds was initially reduced, several patients were discharged and many psychiatric wards were reconverted into COVID+ units. During lockdown, many rehabilitation and vocational units for psychiatric patients, including day hospitals, either closed^4,5 or drastically reduced and adapted their activity.^6–8 Telehealth was widely endorsed and many of the interventions were converted to an online format.^6,8–14 It has been suggested that besides the outbreak and the restrictions that may act as a source of stress, the discontinuities and changes in the health-care system may have resulted in an additional burden for people with mental disorders and might have increased in-patient admissions.^15–17 There is an urgent need to establish evidence on the impact of service adaptations due to COVID-19 on psychiatric patients.^3,11,18

Day hospitals are units that provide diagnostic and treatment services for acutely and severely mentally ill patients and are aimed at improving the clinical condition, promoting functional and social recovery and facilitating community reintegration. Day hospitals are a less restrictive alternative to inpatient admission. A systematic review concluded that there is evidence that indicates that caring for people in acute day hospitals is as effective as inpatient care. Users reached the same levels of treatment satisfaction and quality of life, supporting reducing inpatient admissions whilst improving patient outcome amongst those suitable for day hospital care.¹⁹

There are very few articles on how day hospitals adapted to the pandemic. In Japan, a team used a self-reporting questionnaire to explore the worsening of symptoms related to the discontinuation of day care service.²⁰ Contrary to their expectations, most patients did not refer symptom worsening, and they highlight the importance of regular calls from clinicians to maintain regular routines and motivation. A French article describes how a day hospital adapted their service during the lockdown period, converting themselves into a virtual mental institution and creating a blog to share with their users.²¹ In New York, a day care team describes how they adapted all their activity to maintain full service via telehealth, both for individual and group therapies, describing no psychiatric decompensations, good rates of enrollment and positive qualitative feedback.⁸

In 2019, oblivious to the global health crisis that was yet to come, the two mental health day hospital units for adults of the health division Camp de Tarragona in North-Eastern Spain, started an observational research project to describe the profile of the users attended, study their evolution throughout their hospitalization and determine which factors were associated with more favorable outcomes. The recruitment period was established to start on the 1st of February 2019 and finish on the 31st of January 2021, coincidentally covering a year before the start of the pandemic and a year after.

The two day hospitals participating in the research project, located 10 miles apart, cover a population of roughly 600,000 inhabitants. The socio-demographic characteristics of the population of both centers are analogous. Both hospitals are under the same coordination, and have the same structure, resources, schedule and treatment format. The staff includes psychiatrists, clinical psychologists, occupational therapists, social workers, nurses and administrative assistants. Before the outbreak, users attended the service on a daily basis (Monday through Friday from 9.30 am until 3 pm). The treatment offered included diagnostic interviews, a multidisciplinary initial and final assessment, individual and family therapy, and group therapy sessions led by the different members of the staff regarding several areas (emotional regulation, social cognition, social skills, metacognitive training, cognitive remediation, daily living activities, habits and routines, motor and processing skills, leisure planning, health education, relaxation, physical activity and psychoeducation for caregivers). Most group activities were common for all users with groups of up to 20–23 people. Lunch was served for all users.

When the Spanish government declared the state of alarm on the 14th of March of 2020 and issued the quarantine order, these two day hospitals remained open but drastically reduced face-to-face visits. Group activity was ceased and new admissions were postponed. At the end of May 2020, as the pandemic improved, the two day hospitals started admitting new users. Face-to-face visits increased and small groups were set up on a reduced schedule. Groups had initially 5 participants and progressively increased up to a maximum of 11 users. Users started attending the day hospital two or three days per week. Group interventions were reduced, with an initial schedule of 4 h per week and gradually increasing up to 8 h. In addition to in-person groups, some group therapies were adapted to be conducted online. Users received training and support to be able to participate from their homes. A few didn’t have Internet access so a computer in a room was facilitated in the day hospital for the online sessions. Lunch service was not resumed. During the time the included participants received treatment in the day hospital, the schedule and group frequency changed according to the evolution of the pandemic and the government recommendations. During 4 months group therapy was only offered in an online format.

The clinical practice in the two day hospitals was dramatically affected by the pandemic, but the question remained as to whether these changes had affected the effectiveness of the treatment. The present study compares the clinical and functional outcome results of the cohort of users attended before versus during the pandemic.

Material and methods

A descriptive observational research project was set up to describe the profile of the users attended in the two day hospitals and observe their evolution throughout their hospitalization. During the recruitment period, users admitted to any of the two day hospitals who were 18 years old or more and voluntarily accepted participation and gave written informed consent were included. Users with a psychopathological or cognitive state that hindered reasonable understanding of the project or whose clinical condition discouraged the proposal were excluded. Of a total of 309 admissions, 160 participants accepted to participate in the study (51.8%). If a user returned to the day hospital after their discharge during the recruitment period, they were offered to participate again and were assessed as a new participant (9 users, 5.6%).

For the purpose of the present study we selected those users that entirely completed their admission in the day hospital before the lockdown period that started on the 14th of March 2020 (N = 72) and those admitted afterwards (N = 74). Participants that received treatment during the two time periods were discarded in the statistical analysis (N = 14).

Instruments

All participants completed an initial multidisciplinary assessment that included a wide range of self-administered and clinician-administered measures. Socio-demographic characteristics and biographic history was registered, including childhood adverse effects, age of onset of disorder and previous admissions in psychiatric wards. Suicidal risk was assessed with the clinician-rated Suicidal Scale of the Mini International Neuropsychiatric Interview (MINI).^22,23 The psychopathological assessment included a diagnostic interview, the clinician-rated Global Assessment Scale (GAS),²⁴ and clinical scales according to diagnosis such as the self-administered Beck Depression Inventory-II (BDI-II)^25,26 and the clinician-rated Positive and Negative Syndrome Scale (PANSS).^27,28 Functional assessment included the clinician-rated Basic Everyday Living Skills Scale (BELS)^29,30 and a register of the use of mental health and community resources (community rehabilitation centers, mental health associations, sheltered or ordinary work, education or leisure resources, volunteering). Users completed a self-administered measure of health and quality of life (SF-36)^31,32 and their main caregiver completed a questionnaire on their perception of burden (Zarit scale, non-abbreviated form).^33,34

At discharge, participants completed the same comprehensive assessment. In cases with unexpected discharge (such as abandonment or urgent admission in a psychiatric ward) some of the scales were not administered.

Ethical approval

This research was conducted with integrity and in line with generally accepted ethical principles and approved by the Pere Virgili Institute for Health Research Ethics Committee (S159/31.10.18) and the Clinical Research Commission of the Institut Pere Mata (PR_09-17-2018_02_Estrada).

Statistical analyses

Descriptive statistics were used to summarize the variables; mean and standard deviation were used for continuous variables, while frequency and percentage were used for categorical variables. Categorical and continuous variables were analyzed and compared between the two groups of participants. Normality of all the continuous variables was assessed with the Kolmogorov–Smirnov test. Inferential statistics, including independent sample t-test (for variables with a normal distribution), Mann–Whitney U test (for variables without a normal distribution) and Fisher's exact test, were used to examine if there were significant differences in the outcomes between the two specified groups of participants.

Statistical analyses were performed using SPSS for PC (version 17.0) software. The level of significance was set at p ≤ 0.05 (two-tailed).

Results

To determine whether the two groups were initially equivalent, statistical comparisons were carried out. No statistical differences were found regarding age, gender, severe mental disorder among 1st or 2nd degree relatives, years since the onset of the disorder, number of previous admissions in in-patient psychiatric wards, number of childhood adverse events, and main diagnostic (Tables 1 and 2).

Table 1.

Comparison of gender, diagnostic category and family history of severe mental disorder between users admitted before and during the Covid-19 pandemic.

	Before COVID		During COVID
	N	%	N	%
Gender
Male	27	35.7	33	44.6
Female	45	62.5	41	55.4
Severe mental health disorder in relatives
Not referred	38	52.8	38	51.4
1st grade relatives	21	29.2	27	36.5
2nd grade relatives	9	12.5	8	10.8
1st & 2nd grade	4	5.6	1	1.4
Main diagnosis
Schizophrenia	12	16.7	9	12.2
Schizoaffective disorder	7	9.7	7	9.5
Bipolar	12	16.7	13	17.6
Unipolar affective dis.	19	26.45	15	20.3
Borderline personality	10	13.9	11	14.9
Other personality dis.	4	5.6	9	12.2
Obsessive–compulsive d.	1	1.4	2	2.7
Eating disorder	4	5.6	2	2.7
Autism spectrum	1	1.4	3	4.1
Others	2	2.8	3	4.1

Table 2.

Clinical characteristics of users admitted before and during the Covid-19 pandemic.

	Before COVID			During COVID
	N	Mean	S.D.	N	Mean	S.D.
Age	72	40.24	13.23	74	39.01	13.17
Years since the onset of the disorder	72	13.53	10.01	74	14.30	9.37
Number of adverse childhood events	71	1.89	2.23	72	2.01	2.16
Previous in-patient admissions	72	4.07	7.09	74	4.18	7.31
GAS at admission	72	40.79	6.13	74	41.77	5.67
Admission duration in natural days	72	102.2	30.73	74	105.42	35.77
Days of real assistance*	72	49.35	18.70	74	35.58	14.79

*p < 0.001

Regarding the initial psychometric assessment, no statistical differences were found in suicide risk, GAS, BDI2 or any of the subscales of either the BELS or the SF-36. Only the positive syndrome subscale of the PANSS (T-test p = 0.027) and the Zarit scale (T-test p = 0.049) showed statistical significance with worse scores in the pandemic group. The use of mental health and general community resources for the two groups of users were not different in the initial assessment. Taking all these analysis into account we conclude that the two groups of participants were homogeneous.

The length of the admission was similar in both groups, but the days of real assistance to our hospitals (either in person or virtually) was significantly different in both groups (T-test p < 0.001). This difference is due to the changes and adaptations of our hospitals in relation to the pandemic and its associated restrictions. Users admitted during the pandemic received assistance fewer days than the previous year. There were no statistical differences in the type of discharge, with the majority of discharges due to improvement and therefore referring users to the corresponding mental health center.

Regarding psychometric assessment, the improvement in all scales (Table 3), including user and clinician-rated scales, did not show statistical differences between the two groups, the only exception being the domestic skills section of the BELS (T-test p = 0.021), showing less improvement in the pandemic group. No differences were found in the other scales (PANSS, BDI2, GAS, SF-36, Zarit) or the other three sections of the BELS (self-care; community skills; activity and social relationships). Suicide risk at discharge and its improvement regarding initial assessment was not different in the two groups.

Table 3.

Improvement in self and clinician-rated psychometric scales at discharge.

Scales improvement	Before COVID			During COVID
	N	Mean	S. D.	N	Mean	S.D.
BELS	59			60
Self-care		0.41	0.50		0.34	0.51
Domestic skills*		0.76	0.90		0.40	0.80
Community skills		0.54	0.80		0.56	0.81
Social skills		0.60	0.84		0.63	0.72
SF-36	55			57
Limitations of activities		2.50	22.84		−2.54	20.16
Physical problems		24.10	50.51		8.33	50.52
Pain		7.78	27.41		−1.73	28.62
Social activities		22.50	38.52		29.16	34.21
Mental health		17.20	21.10		16.46	20.35
Emotions		31.30	31.30		27.44	27.44
Energy		12.00	12.00		14.58	14.58
General health		10.33	10.33		16.61	16.61
Changes		31.31	31.31		23.20	23.20
BDI 2	39	12.51	12.43	36	15.3	14.05
PANSS	9			13
Negative		4.4	8.83		0.0	5.20
Positive		4.31	3.31		2.84	4.88
General		10.0	11.44		7.77	9.97
GAS	71	8.08	5.53	74	7.35	7.44
MINI Suicidality Scale	60	0.43	0.98	74	0.58	1.12
ZARIT	47	3.30	13.33	31	7.58	9.84

*p < 0.021

In relation to the use of community or mental health resources (Table 4), there were no significant differences between the two groups, with the exception of structured community leisure activities (Fischer's test p = 0.003). Users from the pandemic group showed fewer enrollments in these activities.

Table 4.

Use of structured community and mental health resources among users attended before or during the Covid-19 pandemic.

Resources at discharge	Before COVID		During COVID
	N	%	N	%
Mental health resources
Community rehabilitation center	12	17.1	13	17.8
Sheltered work	2	2.9	4	5.5
Case Management	7	10.0	6	8.2
Mental Health Association	6	8.6	9	12.3
Other mental health resources	6	8.6	4	5.5
Ordinary resources
Education	13	18.8	6	8.5
Leisure activities*	25	35.7	10	13.7
Volunteering	14	20.0	6	8.2
Ordinary work	11	15.7	9	12.3
Other ordinary resources	6	8.6	1	1.4

*p = 0.003

Discussion

In line with other day care centers around the world, clinical practice in our day hospitals was greatly affected by the restrictions and recommendations related to COVID-19. Through common consensus within the multidisciplinary team, our day hospitals had to set up careful and revisable adaptations to maintain adequate care and effective treatment while lowering the risk of infection for both users and staff. As part of our clinical practice we offered information on the infection and promoted awareness of WHO's recommendations regarding mental health care during the pandemic.¹

The coincidental fact that the recruitment of this study started one year before the declaration of the pandemic and continued for a year after has enabled a precise comparison of the results of our intervention before and during the pandemic. Despite the drastic reduction in therapeutic interventions, and especially group interventions, and the reduction in assistance days for our users, our study suggests that effectiveness was maintained. Globally all measures, including clinician and user rated scales, showed an equivalent improvement and similar treatment outcomes. The only exceptions were the significantly fewer enrollments in community leisure activities and the lower improvement in the domestic skills section of the daily living activities assessment in the group admitted during the pandemic. The first observed difference could be easily explained by the reduction of these types of activities in the community due to the pandemic and its restrictions. The second observed difference could indicate that the changes and adaptations in the type of intervention have resulted in a lower improvement in this area, and therefore imply that more attention should be paid to this specific area. However, there were no significant differences at either admission or discharge in this subscale in the comparison of both groups. The observed significant difference might also be a statistical artifact, as the analysis of both groups involved 30 variables.

Several factors might have played an important role in sustaining effectiveness. The effort to maintain contact with users, online or in person, and our availability, might have contributed to the prevention of relapses. Careful examination of the contents of group interventions led to a more specific and clinically relevant selection. The reduction of the number of participants in groups might have been favorable. The time previously assigned to group interventions by the staff was used for individual interventions, many of which were conducted online either by telephone or videoconference. Adapting to telehealth was challenging both for users and staff, and we had never been so aware of the digital divide in our population. When we started the virtual groups all users were trained on how to use the applications on their devices, but continuous support was needed for some people. However, the improvement on the use of new technologies might have had a positive impact on the life of our users, especially during the pandemic.

The present study has some limitations. During the pandemic there were several adaptations and changes: virtual groups were present during some months but not others, in person assistance fluctuated according to infection levels and government recommendations, number of users per group varied. These changes were not recorded separately and were included comprehensively in the pandemic group. Therefore we are unable to determine which changes contributed to the continuity of outcome results. Furthermore, due to the fact that the present study compares a cohort that received treatment before the pandemic with another admitted during the pandemic, we cannot separate the effect of the service adaptations from the impact of the pandemic itself.

The results of this study have enabled us to reflect on our clinical practice as well as the adaptations carried out in order to maintain adequate and efficient care. Further research is needed to determine which of these adaptations constitute an improvement and should prevail beyond the pandemic.

Day hospitals outcomes in times of COVID-19

The present study did not receive external funding.

The preliminary results of this work have been presented as an abstract on the XXIV Congreso Nacional de Psiquiatría, in Spain, that was held on the 28–30th of October 2021. None of the tables have been presented.

All of the authors report no financial relationships with commercial interests.

Conflicts of interest

The authors have no conflicts of interest to declare.

Acknowledgments

We would like to thank all the members of staff of the day hospital in Tarragona and Reus, as well as all the users that generously decided to participate in this study. We are grateful for the valuable contributions regarding statistical analysis, translation review and general revision by Gerard Muntané, Barbara Ternon and Elisabet Vilella.