Variation in Mental Health, Sleep Quality and Quality of Life Following COVID-19 Hospitalization

Abstract

COVID-19 causes instant and often prolonged effects with multisystemic and heterogeneous symptoms, significantly affecting the bio-psychosocial life of patients. In view of this, the present prospective cohort study aims to document the evolution of symptoms in hospitalized COVID-19 patients, monitoring variations in their mental and physical health for eighteen months after clinical recovery. A sample of 117 patients was examined at four distinct time points: at 2, 6, 12 and 18 months after discharge from the Pulmonology Clinic of the General University Hospital of Larissa (single-center approach). The data collection focused on psycho-physical symptoms, sleep disturbances and quality of life indicators. The study results revealed that over the 18-month period the participants gradually recovered both physically and psychologically, as well as regained pre-disease sleep patterns. However, substantial recovery was noted by month 6, followed by stabilization of their health status. However, a complete return to the pre-disease state of all assessed variables was not achieved, confirming and reinforcing the need for long-term follow-up and overall patient care.

1. Introduction

The ongoing global health crisis triggered by the pandemic of COVID-19, caused by the SARS-CoV-2 virus, has had a deep and complex impact at both individual and societal levels [1]. The wide spread of COVID-19, with hundreds of millions of confirmed cases worldwide [2], has highlighted the alarming tendency for persistence or de novo onset of symptoms in a significant proportion of survivors, which may persist for weeks or even months after the acute phase of infection has resolved [3]. This clinical entity, scientifically defined as long-term COVID, has been a point of attention and a major focus of research efforts in the global scientific community.

In contrast to a single clinical manifestation, long-term COVID constitutes a diverse spectrum of multisystemic symptoms that affect different organ systems, causing clinically significant functional deficits in sufferers [4,5,6]. Based on the National Institute for Health and Care Excellence (NICE) classification, the post-acute phase of COVID-19 is characterized by a two-phase onset of symptoms [7]. More specifically, the first phase is defined in terms of time from the fourth to the twelfth week after disease onset and is known as ongoing symptomatic COVID-19, while the second phase, referred to as l ong-term COVID syndrome, is defined by the continued presence of symptoms for more than twelve weeks.

In order to limit the spread and understand the pathogenesis of the new virus, the scientific community initially focused on addressing the acute phase of COVID-19. However, in the past years, a new need has emerged to systematically investigate the medium- and long-term consequences of the virus infection, as they cause significant impairment of daily functioning and degrade the health-related quality of life (HRQoL) of patients. In particular, persistent fatigue, respiratory distress, neurocognitive deficits, sleep disturbances, microvascular dysfunction, autoimmune disorders, myalgia, mood disordersand behavioral changes—regardless of the severity of the initial acute disease—are defined as the most frequently reported. Interestingly, two distinct symptom patterns have been described in patients with l ong-term COVID: one that focuses on fatigue, headacheand upper respiratory symptoms, and a multi-systemic syndrome with varying clinical manifestations with deterioration of overall well-being [3,4,5,6,8,9,10,11,12].

However, the ongoing evolution of SARS-CoV-2 mutation and the emergence of a wide range of strains, both new and re-emerging, has escalated concern in the scientific community, raising concerns about the increased ability to transmit the disease and the potential variation in the range of long-term health consequences for the patients [13,14,15]. In this perspective, it is worth mentioning the observation that the severity of acute COVID-19 is not necessarily a predictor of long-term complications. Consequently, even individuals with mild infection did not avoid the development of significant posterior pathological conditions. In this case, these covered a wide range of manifestations, from pulmonary fibrosis and cardiovascular thromboembolic events to inflammatory syndromes, neurological disorders and mood disorders, confining the patients to a context of restricted activities and social isolation [16,17,18,19,20].

Considering, therefore, the unpredictable nature and duration of disease progression, the need for a holistic approach, based on the systematic monitoring of patients and the adoption of individualized multidisciplinary treatment approaches, has become self-evident [3,21,22]. Therefore, the integrated approach to long-term COVID must also focus on its psychological implications. Symptoms such as delusions or paranoid ideations of being persecuted, fear of death, social isolationand perception of stigmatization often lead to the development of mood and behavioral disorders and increase the risk of long-term psychiatric morbidity [9,23,24,25], impairing the patients’ HRQoL [25,26].

Despite the extensive description of the clinical picture of patients with l ong-term COVID, the existing bibliography is significantly lacking in monitoring the temporal dynamics of symptoms and their correlation with mental health and quality of life indicators, especially in the Greek population.

In view of this, the present descriptive, prospective cohort study (single-center approach) investigates the symptom dynamics of mental well-being, sleep quality and HRQoL in patients after the acute phase of COVID-19 and their hospital discharge from the Pulmonology of the General University Hospital of Larisa. The main objective is to evaluate the longitudinal change in the severity of symptoms that make up the syndrome and recovery of patients’ pre-hospital quality of life over an eighteen-month period because the clinical phenotype and its pathogenesis remain unclear.

The in-depth analysis is expected to broaden the scientific understanding of l ong-term COVID syndrome and collect essential data to improve the quality of life of patients through targeted holistic treatment strategies, as little is known regarding post-COVID management. We sought to answer key questions first: how do people with long COVID experience the development, course and perhaps resolution of problems on their mental health, sleep quality and quality of life over time? Second, are there demographic variables (age, gender) and specific clinical characteristics (coexisting conditions, respiratory symptoms), indicating increased vulnerability in selected subgroups of patients? Third, what are their ideas for improving the management of their condition and the design and delivery of services?

2. Materials and Methods

2.1. Research Design

This prospective cohort study attempted to describe the time course and symptomatology of long-term COVID syndrome in patients with a history of hospitalization due to confirmed SARS-CoV-2 infection at an eighteen-month time horizon after cure from acute infection.

To capture the long-term dynamics of symptoms, systematic data collection was performed at four discrete time points (T1 = 2 months, T2 = 6 months, T3 = 12 months, T4 = 18 months post-hospitalization), allowing for the tracking of the longitudinal course of HRQoL in patients with COVID-19.

The manuscript addresses a highly relevant and important topic concerning the long-term sequelae of COVID-19, specifically focusing on mental health, sleep and quality of life. The prospective cohort design with multiple follow-up points over 18 months is a strength, allowing for the examination of temporal dynamics. The use of multiple standardized and validated questionnaires provides comprehensive data.

2.2. Study Sample and Data Collection Process

In the context of the prospective cohort study, a sample of patients who were hospitalized at the Pulmonology Clinic COVID-19 of the General University Hospital of Larissa (single-center approach) during the period July 2021 to December 2023 was examined. Patients were selected based on strict eligibility criteria, including laboratory-confirmed diagnosis of COVID-19, completion of the acute phase of the disease, discharge and evidence of sufficient cognitive competence to ensure proper implementation of the research design.

Similarly, the presence of serious underlying diseases or comorbidities was an exclusion factor, considering the possible existence of language restrictions to ensure the validity of the data.

Patients received written and verbal information about the nature of the study and their right to withdraw. Criteria for participation were as follows:

Inclusion criteria: (1) Age ≥ 18 years. (2) Previous infection with SARS-CoV-2 confirmed by a positive PCR test result or positive rapid self-test (two months after discharge from hospital). (3) Ability to give informed consent. (4) Availability during the intervention period. (5) Possess a residence suitable for attending the study site appointment.

Exclusion criteria: (1) Suffering from severe mental disorders according to ICD-10. (2) Current suicidal tendencies. (3) Currently receiving psychotherapy or previously receiving psychotherapy for long-term/post-pandemic times. (4) Insufficient knowledge of Greek language or Greek text.

All investigators were psychologists or physicians or nurses with appropriate psychotherapeutic training and they had received training in conducting the interventions according to the assay.

Following the application of these strict inclusion and exclusion criteria, the final study sample included 117 adults (aged at least 30 years), of which seventy-one were men and forty-six were women, with an average age of fifty-eight years. In this context, sampling was performed at four predefined time points (T1 = 2 months, T2 = 6 months, T3 = 12 months, T4 = 18 months) after hospitalization.

Postponing the start of the follow-up period for two months after discharge was a methodological necessity in order to examine and evaluate the ultimate consequences of the disease in a discrete manner, neglecting the possibility of interaction with data associated with the acute phase of infection.

2.3. Ethics and Morality

The present research was conducted in accordance with the principles of the Declaration of Helsinki and after approval by the Internal Ethics Committee of the University of Thessaly (No. 76/17-01-2022). Ensuring the protection of the participants, their written informed consent was obtained.

2.4. Data Collection

2.4.1. Measurements Obtained

In the prospective cohort study, patients who recovered from COVID-19 and discharged from the Pulmonology Clinic of the PGL participated in a long-term evaluation protocol at four predefined time points (2, 6, 12 and 18 months). To enhance the validity and reliability of the data, at each time point in the study, participants underwent a clinical examination and completed a series of standardized questionnaires.

2.4.2. Measuring Tools

To achieve the research objectives, the following data sets were collected: (a) demographic data of the participants, (b) sleep quality data through the Pittsburgh Sleep Quality Index (PSQI), (c) psychological symptomology data through the Symptom Catalogue 90 (SCL-90), (d) perceived health data through the SF-36 scale, (e) symptom severity data through the Edmonton Symptom Assessment System-Revised (ESAS-R) and (f) Post-traumatic stress disorder PTSD Checklist.

Demographic Data

The first section of the questionnaire was designed to create a comprehensive participant profile through the collection of basic demographic data (gender, age, marital status, education, occupation, residence, income), medical history and COVID-19 disease history, allowing the investigation of the contribution of the bio-psychosocial context to the clinical differentiation of the disease.

Sleep Quality Assessment (PSQI)

The sleep quality of patients with long-term COVID was assessed using the multi-parameter and valid Pittsburgh PSQI Sleep Quality Index [27], a common tool in relevant research, as suggested by contemporary research reports [28].

Psychological Profile (SCL-90-R)

The psychological aspect of long-term COVID syndrome was assessed using the psychometrically valid SCL-90-R scale. This scale, structured in nine subscales and three general indicators, provides a global assessment of psychological distress and overall psychopathological burden, given its international acceptance and Greek standardization [29].

Perceived Health Assessment (SF-36)

The assessment of respondents’ holistic health and well-being was reliably achieved by incorporating a wide range of bio-psychosocial variables into the SF-36, which analyzes multiple discrete components of quality of life over time. In particular, physical functioning, functional limitations due to psycho-physical problems, pain intensity, sense of vitality, self-assessed general health, social participation and mental well-being were assessed, and it was possible to generate two summary assessments of health, separating physical from mental health [30,31,32].

Symptom Assessment Through ESAS-R

In order to investigate the participants’ multi-systemic symptomatology, the psychometric weighted ESAS-R tool, the ESAS-R, was used. The systematic assessment of a wide range of symptoms (pain, fatigue, nausea, depression, anxiety, drowsiness, anorexia, well-being, dyspnoea) enabled the investigation of the association between symptomatology and perceived quality of life, contributing to the clinical profile of the disease, while providing critical information about patients’ needs [33,34].

Post-Traumatic Stress Disorder PTSD

The PTSD Checklist was chosen for this study because of its relevance to the profile of patients with long COVID syndrome. These individuals may experience PTSD due to the prolonged nature of the disease, the uncertainty about their healthand the psychosocial consequences of the pandemic. Using PTSD allows for early detection and intervention in such symptoms [35].

2.5. Statistical Analysis

The data were statistically analyzed using SPSS (version 26.0). Initially, descriptive statistical analysis was performed for the demographics and the categorical variables. Continuous variables (PSQI, SCL-90, SF-36, ESAS-R, PTSD) at four time points (T1, T2, T3, T4) were analyzed by ANOVA (repeated measures ANOVA) to determine statistically significant variations. For the effect size of time, the partial eta squared index was calculated. This index is considered negligible when it is below 0.01, small when it ranges between 0.01 and 0.06, moderate when it falls between 0.06 and 0.14and large when it is 0.14 or higher. Variables were presented as means and standard deviations, followed by post-hoc Bonferroni comparisons (p < 0.05) at significance. For easier interpretation, results were illustrated by means of tables.

3. Results

The sample consisted of 117 participants, 60.7% of whom were men. The mean age was 58.5 years (SD = 12.1 years). The majority were married (82.1%,), were living with their spouse and children (42.7%) or only with their spouse (31.6%). A total of 59.8% had two children, 14.5% had one child, while only 7.7% had no children. The total number of people living in the household was four or more for 37.6% and two for 30.8%. Regarding employment status, 38.6% were retired, 17.9% were public sector employees and 15.4% were private sector employees. In terms of education, 40.2% were secondary school/high school/vocational institute graduates and 30.8% were university or technological educational institute graduates. Additionally, 70.1% resided in an urban area, 48.7% had an income of 1346 or more and 40.2% were dealing with two or more medical conditions. Sample’s characteristics are presented in

Table 1. Sample’s characteristics.

		N	%
Gender	Male	71	60.7
	Female	46	39.3
Age, Mean (SD)		58.5 (12.1)
Married	No	6	5.1
	Yes	96	82.1
Living	Alone	14	12.0
	With the spouse	37	31.6
	With the children	12	10.3
	With spouse and children	50	42.7
	Other	4	3.4
Children	0	9	7.7
	1	17	14.5
	2	70	59.8
	3	14	12.0
	≥4	7	6.0
Number of people living in the same house	0	1	0.9
	1	13	11.1
	2	36	30.8
	3	23	19.7
	≥4	44	37.6
Occupation	Unemployed	5	4.3
	Private employee	18	15.4
	Civil servant	21	17.9
	Household	13	11.1
	Pensioner	43	36.8
	Other	17	14.5
Education	Primary school graduate	33	28.2
	Secondary School/High School/Vocational Institute Graduate	47	40.2
	University/Technological Educational Institute (TEI) Graduate	36	30.8
	Master’s degree/Ph.D.	1	0.9
Residential Area	Urban area	82	70.1
	Semi-urban area	16	13.7
	Rural area	19	16.2
Monthly family income	Up to 860 euros	24	20.5
	861–1345 euros	36	30.8
	1346 and above	57	48.7
Are you experiencing two or more medical conditions?	No	70	59.8
	Yes	47	40.2

.

The changes in dimensions of SF-36 scale are presented in

Table 2. Changes in SF-36 dimensions during the follow-up period.

	2nd Month	6th Month	12th Month	18th Month	Mean Change (SD)	P12	P23	P34	P14
	Mean (SD)	Mean (SD)	Mean (SD)	Mean (SD)
Physical functioning	78.5 (20.1)	87.0 (16)	88.3 (17.4)	90.3 (17.7)	11.8 (20.2)	<0.001	>0.999	0.957	<0.001
	Partial eta squared = 0.143
Physical role	39.7 (43.4)	80.8 (31.6)	75.9 (34.6)	87.2 (23.8)	47.5 (46.2)	<0.001	>0.999	0.004	<0.001
	Partial eta squared = 0.341
Physical pain	78.8 (17.5)	78.4 (17.2)	84.6 (16.7)	87.1 (23.7)	8.3 (24.7)	>0.999	0.012	>0.999	0.002
	Partial eta squared = 0.067
General health	71.2 (16.8)	73.7 (15.6)	78.3 (14.6)	77 (18.5)	5.8 (19.4)	>0.999	0.014	0.242	0.009
	Partial eta squared = 0.068
Vitality	73.4 (19)	78.1 (13.4)	74.1 (11.3)	75.6 (13)	2.1 (18.6)	0.028	0.035	>0.999	>0.999
	Partial eta squared = 0.031
Social functioning	67.1 (40.1)	90.3 (17.1)	85.1 (16.7)	86.5 (16.9)	19.4 (42.6)	<0.001	0.063	>0.999	<0.001
	Partial eta squared = 0.173
Emotional role	77.8 (37.7)	92 (22.6)	92 (20.8)	94.9 (17.3)	17.1 (40.5)	0.001	>0.999	0.999	<0.001
	Partial eta squared = 0.095
Mental health	85.0 (9.4)	87.1 (9.9)	87.6 (7.0)	87.6 (12.1)	2.6 (11.5)	0.007	>0.999	>0.999	0.086
	Partial eta squared = 0.096
PCS	45.6 (8.7)	48.5 (8.4)	50.3 (6.9)	50.1 (7.6)	4.5 (8.8)	<0.001	0.108	0.104	<0.001
	Partial eta squared = 0.121
MCS	54.0 (8.4)	56.5 (5.8)	56.3 (4.1)	55.6 (5.1)	1.6 (8.7)	0.025	>0.999	0.796	0.299
	Partial eta squared = 0.044

P¹²: p-value for difference between 1st measurement (2 months) and 2nd measurement (6 months); P²³: p-value for difference between 2nd measurement (6 months) and 3rd measurement (12 months); P³⁴: p-value for difference between 3rd measurement (12 months) and 4th measurement (18 months); P¹⁴: p-value for difference between 1st measurement (2 months) and 4th measurement (18 months).

. Physical functionality increased at the 6th month compared to the 2nd month (p < 0.001). It then remained at similar levels in the third measurement (12th month; p > 0.999) and the final measurement (18th month; p = 0.957 compared to 12 months). Physical functionality was significantly better in the last measurement compared to the second month (p < 0.001). The score in the physical role dimension showed a significant increase at the 6th month compared to the second month (p < 0.001) and in the final measurement (18th month), compared to the 12th month (p = 0.004). The physical role was also significantly better in the last measurement compared to the first (p < 0.001). Similarly, physical pain showed a significant improvement at the 12th month compared to the 6th month (p = 0.012). Pain in the last measurement was significantly better than in the first (p = 0.002). General health also showed a significant improvement at the 12th month compared to the 6th month (p = 0.014) and was better in the final measurement compared to the first (p = 0.009). Vitality increased significantly at the 6th month compared to the 2nd month (p = 0.028) and then decreased at the 18th month (p = 0.035), remaining at levels similar to the initial ones (p > 0.999 between last and first measurement). Both the social and emotional roles improved at the 6th month compared to the 2nd month (p < 0.001, p = 0.001 respectively) and then remained at similar levels, which, in the last measurement, were significantly better than in the first (p < 0.001 for both dimensions). The dimension and the summary scale of mental health showed improvement at the 6th month compared to the second month (p = 0.007, p = 0.025 respectively) and remained at similar levels at the 12th and 18th month. The scores in the last measurement were significantly higher than in the 2nd month, but no statistically significant difference emerged (p = 0.086, p = 0.299 respectively). In contrast, physical health, which also improved at the 6th month compared to the 2nd month (p < 0.001), remained at similar levels afterward, was significantly better in the final measurement compared to the initial one (p < 0.001). The changes in PCS and MCS are given in Figure 1 and Figure 2, respectively.

Post-traumatic stress was assessed at the 2nd, 6th and 18th months (Figure 3). Its reduction was significant at the 6th month compared to the 2nd month (p < 0.001). In the final assessment, it had decreased further compared to the 6th month, although no statistically significant change was observed (p > 0.999). A significant reduction was noted in the final measurement compared to the first at the second month (p < 0.001). Changes on PTSD score are given in

Table 3. Changes in PTSD score during the follow-up period.

	2nd Month	6th Month	18th Month	Mean Change (SD)	P12	P24	P14
	Mean (SD)	Mean (SD)	Mean (SD)
PTSD	5.8 (6.7)	3.2 (4.3)	2.6 (2.7)	−3.2 (6.3)	<0.001	>0.999	<0.001
Partial eta squared = 0.114

P¹²: p-value for difference between 1st measurement (2 months) and 2nd measurement (6 months); P²⁴: p-value for difference between 2nd measurement (6 months) and 4th measurement (18 months); P¹⁴: p-value for difference between 1st measurement (2 months) and 4th measurement (18 months).

.

The changes in the participants’ sleep quality are illustrated in

Table 4. Changes in PSQI during the follow-up period.

	2nd Month	6th Month	12th Month	18th Month	P12	P23	P34	P14
	Ν (%)	Ν (%)	Ν (%)	Ν (%)
Good sleep quality	52 (44.4)	78 (66.7)	45 (38.5)	75 (64.1)	<0.001+	<0.001+	<0.001+	0.001+
Poor sleep quality	65 (55.6)	39 (33.3)	72 (61.5)	42 (35.9)

+McNemar’s test. P¹²: p-value for the comparison between 1st measurement (2 months) and 2nd measurement (6 months); P²³: p-value for the comparison between 2nd measurement (6 months) and 3rd measurement (12 months); P³⁴: p-value for the comparison between 3rd measurement (12 months) and 4th measurement (18 months); P¹⁴: p-value for the comparison between 1st measurement (2 months) and 4th measurement (18 months).

. The percentage of individuals reporting adequate sleep quality increased significantly at the 6-month mark (66.7%) compared to the 2nd month (44.4%). However, by the 12-month, a significant decrease was observed again (38.5%). In the final assessment (18 months), the percentage increased significantly once more (64.1%), with the proportion of individuals experiencing good sleep being notably higher than in the 2nd month.

The respiratory symptoms were assessed using a score, with higher values corresponding to more intense symptoms. This score showed a significant decrease at the 6-month mark compared to the 2nd month (p = 0.030), and at the 12-month mark compared to the 6th month (p = 0.004). In the final assessment at the 18-month mark, the symptoms had significantly improved compared to the first measurement at the 2nd month (p < 0.001). These results, as well as the results for ESAS-R scale, are presented in

Table 5. Changes in respiratory symptoms and in ESA-R during the follow-up period.

	2nd Month	6th Month	12th Month	18th Month	Mean Change (SD)	P12	P23	P34	P14
	Mean (SD)	Mean (SD)	Mean (SD)	Mean (SD)
Respiratory symptoms	1.40 (0.4)	1.27 (0.3)	1.17 (0.2)	1.23 (0.3)	−0.17 (0.4)	0.030	0.004	0.182	<0.001
	Partial eta squared = 0.136
Pain	2.3 (1.4)	2.4 (1.4)	1.8 (1.5)	2 (2)	−0.3 (1.9)	>0.999	<0.001	>0.999	0.001
	Partial eta squared = 0.101
Fatigue	3.6 (2.4)	3 (1.6)	3.6 (1.7)	3.2 (1.2)	−0.4 (2.3)	0.982	0.012	0.534	0.021
	Partial eta squared = 0.030
Drowsiness	1.8 (1.7)	1.3 (1.1)	1.4 (0.9)	1.4 (0.7)	−0.4 (1.9)	0.030	>0.999	>0.999	0.207
	Partial eta squared = 0.034
Nausea	1.1 (0.5)	1 (0.1)	1 (0.1)	1 (0.2)	−0.1 (0.5)	0.348	>0.999	0.500	>0.999
	Partial eta squared = 0.023
Loss of appetite	1.2 (1.2)	1 (0.2)	1 (0.2)	1 (0.1)	−0.2 (1.2)	0.377	>0.999	>0.999	0.168
	Partial eta squared = 0.029
Difficulty breathing	1.6 (1.4)	1.3 (0.9)	1.2 (0.9)	1.2 (0.7)	−0.4 (1.5)	0.019	>0.999	>0.999	0.113
	Partial eta squared = 0.038
Depression	1.4 (1.4)	1.2 (0.8)	1.3 (0.8)	1.3 (0.9)	−0.1 (1.5)	0.977	>0.999	>0.999	>0.999
	Partial eta squared = 0.008
Anxiety	3.6 (1.5)	3.8 (2)	2.9 (1.7)	3.1 (2.7)	−0.5 (2.4)	>0.999	<0.001	>0.999	<0.001
	Partial eta squared = 0.130
Well-being	3.3 (1.9)	3.2 (1.6)	3.2 (1.2)	3.3 (1.1)	0.0 (2.0)	>0.999	>0.999	0.134	0.833
	Partial eta squared = 0.013
Total ESAS-R	2.2 (0.8)	2.0 (0.5)	1.9 (0.5)	2 (0.7)	−0.2 (0.8)	0.006	0.084	>0.999	<0.001
	Partial eta squared = 0.098

P¹²: p-value for difference between 1st measurement (2 months) and 2nd measurement (6 months); P²³: p-value for difference between 2nd measurement (6 months) and 3rd measurement (12 months); P³⁴: p-value for difference between 3rd measurement (12 months) and 4th measurement (18 months); P¹⁴: p-value for difference between 1st measurement (2 months) and 4th measurement (18 months).

. Concerning the ESAS-R scale, the dimensions of nausea, loss of appetite, depression and well-being did not show significant changes over time. The dimension related to pain showed a significant reduction at the 12th month compared to the 6th month (p < 0.001), remaining at similar levels at the 18-month mark (compared to the 12th month; p > 0.999). The final pain score was significantly lower than the initial score (p < 0.001). Fatigue showed a significant increase at the 12-month mark compared to the 6th month (p = 0.012). It then decreased, but not significantly, at the 18-month mark compared to the 12th month, and this final score was significantly lower than the initial one (p = 0.021). Drowsiness decreased at the 6-month mark compared to the 2nd month (p = 0.030) and then remained at similar levels. The difference between the last and the first measurement was not significant (p = 0.207). The same trend was observed with difficulty breathing (p = 0.019 for the decrease at the 6-month mark compared to the 2nd month; p = 0.113 between last and first measurement). Anxiety (feeling of nervousness) showed a significant reduction at the 12th month compared to the 6th month (p < 0.001), remaining at similar levels thereafter. The final score for this dimension was significantly lower than the initial score (p < 0.001). Overall, the total score showed a significant reduction at the 6-month mark compared to the second month (p = 0.006), and then remained at similar levels. The final score was lower (thus better) than the initial one (p < 0.001).

The results from the analysis of the changes in the dimensions and indicators of the SCL-90 scale are given in

Table 6. Changes in SCL-90 dimensions during the follow-up period.

	2nd Month	6th Month	12th Month	18th Month	Mean Change (SD)	P12	P23	P34	P14
	Mean (SD)	Mean (SD)	Mean (SD)	Mean (SD)
Global Severity Index (GSI)	0.32 (0.2)	0.25 (0.2)	0.31 (0.2)	0.27 (0.1)	−0.05 (0.2)	0.014	0.002	0.007	0.270
	Partial eta squared = 0.055
Positive Symptom Total (PST)	15.2(10.9)	15.4(9.1)	20.3(10.1)	18.9 (7.2)	3.63 (10.1)	>0.999	<0.001	0.004	>0.999
	Partial eta squared = 0.162
Positive Symptom Distress Index (PSDI)	1.85 (0.5)	1.41 (0.4)	1.36 (0.2)	1.26 (0.2)	−0.59 (0.1)	<0.001	<0.001	<0.001	<0.001
	Partial eta squared = 0.393
Somatization	4.7 (5)	4.4 (3.5)	5.6 (3.7)	4.6 (2.9)	−0.1 (4.7)	>0.999	0.014	0.010	0.079
	Partial eta squared = 0.062
Compulsive	3.9 (3.7)	3.5 (3.1)	4.8 (3.9)	4.1 (3.2)	0.2 (4.1)	>0.999	0.019	0.024	0.971
	Partial eta squared = 0.033
Interpersonal Sensitivity	3.2 (3.4)	2.3 (2.1)	2.6 (2.2)	2.1 (1.7)	−1.1 (2.9)	0.088	>0.999	0.009	0.001
	Partial eta squared = 0.034
Depression	6.2 (5.7)	4.3 (4)	4.6 (4.2)	4.4 (3.5)	−1.8 (4.7)	0.013	>0.999	0.018	0.004
	Partial eta squared = 0.049
Anxiety	2.8 (2.9)	2.1 (2.3)	1.9 (2.5)	1.4 (1.6)	−1.4 (2.6)	0.475	0.290	0.300	<0.001
	Partial eta squared = 0.092
Hostility	1.6 (2.1)	1.4 (1.9)	1.3 (1.4)	1.4 (1.6)	−0.2 (2.3)	>0.999	>0.999	>0.999	>0.999
	Partial eta squared = 0.002
Phobic Anxiety	0.9 (2)	0.4 (1.3)	0.7 (1.6)	0.2 (0.5)	−0.7 (1.9)	0.030	0.130	<0.001	<0.001
	Partial eta squared = 0.066
Paranoid Ideation	1.5 (2.4)	2.3 (2.2)	4.3 (2.2)	3.8 (1.9)	2.3 (2.8)	<0.001	<0.001	0.379	0.031
	Partial eta squared = 0.462
Psychoticism	1 (1.6)	0.5 (1)	0.9 (1.2)	1 (1.2)	0.0 (1.8)	0.003	0.030	0.627	>0.999
	Partial eta squared = 0.062

P¹²: p-value for difference between 1st measurement (2 months) and 2nd measurement (6 months); P²³: p-value for difference between 2nd measurement (6 months) and 3rd measurement (12 months); P³⁴: p-value for difference between 3rd measurement (12 months) and 4th measurement (18 months); P¹⁴: p-value for difference between 1st measurement (2 months) and 4th measurement (18 months).

. The general symptom index decreased from the 2nd month to the 6th month (p = 0.014), increased from the 6th to the 12th month (p = 0.002) and then decreased again from the 12th to the 18th month (p = 0.007). At the final measurement, it was reduced compared to the initial measurement, but the difference was not significant (p = 0.270). Similarly, the Global Severity Index (GSI) increased from the 6th to the 12th month (p < 0.001), but decreased during the final measurement at the 18th month (compared to the 12th month; p = 0.004). The final score was at similar levels to the first measurement (p > 0.999). The Positive Symptom Distress Index (PSDI) showed a significant decrease at the 6th month compared to the 2nd month (p < 0.001), at the 12th month compared to the 6th (p < 0.001) and at the final assessment at the 18th month compared to the 12th month (p < 0.001) (Figure 4). At the final measurement, the index was significantly reduced compared to the initial measurement (p < 0.001). Scores for the dimensions of somatization and obsessive-compulsiveness increased from the 6th to the 12th month (p = 0.014 and p = 0.019, respectively) but decreased at the 18th month compared to the 12th month (p = 0.010 and p = 0.024, respectively). At the final measurement, the scores were at similar levels to the first measurement (p = 0.079 and p = 0.971, respectively). Interpersonal sensitivity remained at similar levels from the 2nd to the 12th month (p = 0.088 for the comparison between 2nd and 6th month; p > 0.999 for the comparison between 6th and 12th month). It decreased significantly at the 18th month compared to the 12th (p = 0.009) month, and the final measurement was significantly lower than the initial score at the 2nd month (p = 0.001). Scores for depression and phobic anxiety showed significant reductions, indicating improvement at the 6th month compared to the 2nd month (p = 0.013 and p = 0.030, respectively), and again at the final assessment compared to the 12th month (p = 0.018 and p < 0.001, respectively). Ultimately, they were significantly improved compared to the initial measurement (p = 0.004 and p < 0.001, respectively). Anxiety showed a continuous decrease, which was not statistically significant between consecutive measurements. However, the reduction was significant when comparing the last measurement to the first (p < 0.001). Hostility showed no significant change over time. Paranoid ideation increased both from the 2nd to the 6th month (p < 0.001) and from the 6th to the 12th month (p < 0.001). It then remained at similar levels from the 12th to the 18th month (p = 0.379). The final measurement was significantly higher compared to the initial one (p < 0.001). The score for psychoticism decreased at the 6th month compared to the 2nd month (p = 0.003), increased at the 12th month compared to the 6th (p = 0.030) and remained at similar levels in the final measurement (p = 0.627 for the comparison between the 12th month and the 18th one), which did not differ from the first measurement (p > 0.999).

4. Discussion

The present study delved into the evolution of long-term COVID syndrome symptomatology in a cohort of 117 patients with a history of hospitalization at the General University Hospital of Larissa through consecutive assessments at four predefined time intervals after discharge (at 2, 6, 12 and 18 months). The research process demonstrated ongoing distress at the physical, psychological and social levels. For in-depth assessment, a series of well-established psychometric tools (PSQI, PTSD, SF-36, ESAS-R, SCL-90) were applied, in addition to the collection and analysis of demographic data, in order to identify the factors shaping the path to full recovery. In particular, the analysis focused on longitudinal changes in symptoms (improvement or persistence), mutual correlations between psychometric tools used and the determinants that potentially increase the vulnerability of specific patient subpopulations, including gender, age, potential co-morbidities and socio-economic level. Subsequently, an attempt was made to interpret the clinical implications of the findings, making clear the need for targeted interventions and a holistic approach to the support and rehabilitation of patients with long-term COVID, aimed at enhancing their HRQoL and well-being in general.

4.1. Presentation and Interpretation of the Results

4.1.1. Demographic Analysis

Demographic data analysis of 117 patients with long-term COVID revealed essential determinants of symptomatic course and quality of life.

More specifically, advanced age and female gender emerged as significant risk factors for greater psychological distress and physical restrictions in patients with long-term COVID. The increased psychological distress and reduced physical functioning observed in older patients may be attributed to the combined effect of reduced biological reserve and the presence of multiple, co-existing chronic diseases or medical conditions [36]. The elderly’s increased susceptibility to the development of the syndrome is strongly supported in the existing literature, as demonstrated by the study by Daitch et al. [37], which revealed a statistically significantly higher incidence of long COVID symptoms in the age group over 65 years, and is further supported by a multitude of studies [38,39,40].

Furthermore, the statistically significant association of female gender with increased vulnerability to mood disorders and lower self-reported physical health is confirmed by the findings of multiple studies [40,41,42,43], supporting the hypothesis of an increased female predisposition to the occurrence and severity of long-term COVID symptoms.

On the other hand, the presence of a strong family supportive framework appeared to have a beneficial effect on the rehabilitation process of the patients, while social isolation was inversely correlated, negatively and significantly affecting their mental well-being [44,45].

4.1.2. Quality of Life (SF-36)

The results of the SF-36 scale outline a non-linear and differentiated progression in the individual dimensions that make up quality of life.

More specifically, physical function and physical role showed significant recovery, marking the gradual recovery of patients’ ability to engage in their usual activities [35], which at the onset of the disease had been significantly limited, mainly explained by respiratory symptoms and fatigue [35,42]. This improvement appears to be causally associated with the remission of pain and respiratory symptoms. At the same time, there was a positive change in physical pain and general health, despite the presence of variations in the improvement of general health, with a more likely explanation being the general decrease in symptoms [35].

In contrast, vitality showed no significant change, remaining at low levels throughout the study. Persistent fatigue, a distinct and debilitating symptom of long-term COVID [46,47], appears to continue to undermine patients’ daily functioning and is negatively associated with sleep quality [48], exacerbating the already low vitality.

In the same context, despite the reported improvement in the social and emotional role dimensions reflecting enhanced social participation and emotional regulation [49], mental health did not show a significant deviation from baseline levels. This finding, consistent with previous studies [35], highlights the persistence of psychological symptomatology, with anxiety and emotional irritability as prominent features, with the feminine gender being over-represented.

In conclusion, the longitudinal follow-up revealed improvement in the physical health summary scale in contrast to the mental health summary scale, which did not display statistically significant variations, a result that is in line with a finding of another research study [50].

Noteworthy, in the cohort examined, is the significant impact of demographic and social variables on patients with long-term COVID quality of life. Advanced age, burden of co-morbidities and low economic level were associated with lower perceived quality of life [35,51], while in contrast, adequate family support appeared to exert a beneficial effect [38,39,40]. Furthermore, increased number of reported symptoms, reduced physical functioning, poor sleep quality and subjective emotional distress were associated as predictors of poor quality of life [50].

4.1.3. Sleep Quality (PSQI)

The sleep quality dynamics in long-term COVID patients, at a 18-month horizon, revealed an interesting nonlinear evolution. Despite the primary improvement detected by the sixth month, possibly attributed to a reduction in overall symptom severity [52], a relapse phase was observed in the twelfth month, to gradually recover sleep quality by the eighteenth month. It is therefore concluded that sleep quality contributes critically to the holistic health of individuals experiencing long-term COVID. Impaired sleep quality was emphatically associated with increased psychological distress, manifested as anxiety, depressive symptoms and panic ideation, as well as manifestations of post-traumatic stress and a decrease in physical functional abilities and vitality, accompanied by an increased sense of fatigue. Following on from the strong associations, sleep disruption appears to further exacerbate the clinical condition of patients, as it exacerbates both physical and mental symptoms, impairing recovery and undermining patients’ well-being [53,54]. In support of this, the established correlation of impaired sleep quality with the chronicity of syndrome symptoms raises the hypothesis of a pathogenetic factor potentially causing a slowed recovery [55].

4.1.4. Post-Traumatic Stress Disorder (PSTD)

The longitudinal course of post-traumatic stress disorder (PTSD) symptoms in patients with long-term COVID revealed a progressive decrease from the peak values detected in the second month to the complete disappearance recorded in the eighteenth month. The aforementioned decreasing trend can be interpreted in the light of the generalized recovery of physical health, the remission of the overall symptomatology, as well as with the promotion of mental well-being.

In addition to longitudinal changes, statistical analysis revealed significant correlations between the presence of PTSD symptoms and a wide range of factors related to the patients’ health and social life. In particular, an inverse relation with vitality was observed, leading to increased fatigue and dysregulation of mental and emotional balance [56,57,58], while contributing to social isolation, especially among people living without a supportive environment.

In addition, a positive correlation was also found between PTSD symptoms and psychological distress, impaired sleep quality (which reciprocally enhances PTSD symptoms) [59] and age, with the elderly and those with a history of severe acute COVID-19 displaying an increased predisposition [60].

4.1.5. Symptom Intensity (ESAS-R)

The longitudinal evaluation of symptomatic burden in patients with long-term COVID for 18 months demonstrated a general trend towards regression, a phenomenon that correlates with the gradual recovery of patient’s initial acute manifestations of infection. However, analysis of individual symptoms revealed a complex and multidimensional dynamic.

In particular, fatigue was identified as the most aggravating and predominant symptom, exerting broad negative effects on physical functioning, vitality, sleep quality and all daily activities of the patients, leading to an increased feeling of intense disorientation and a deterioration of well-being [61,62]. Similarly, recorded levels of anxiety, often manifested as nervousness arising from patients’ concern about the course of symptoms [63], despite its apparent regression, remained clinically notable and potentially adversely affected patients’ mental health [54]. Similarly, although there was a decrease in pain, which was positively associated with improved physical health and well-being [46,64], persistent pain remained a significant factor in disturbing patients’ sleep. In the same context, dyspnoea, despite the recorded improvement, remained a clinically significant presence, suggesting the possible incomplete restoration of lung parenchymal functionality [65,66,67]. Finally, the maintenance of relatively stable levels of depression and general well-being highlight the urgent and ongoing need for psychological support and care [68,69]. In conclusion, despite the progress reported in the overall symptom burden decline, a significant proportion of patients still report persistent symptomatology associated with a reduced quality of life [70,71].

4.1.6. Psychological Burden (SCL-90)

The 18-month study of the psychological effects of long-term COVID in a-patients using the SCL-90 scale indicated a heterogeneous evolution of psychological burden as reflected in changes in the General Symptom Index. Despite the primary trend towards improvement, the final value of the General Symptom Index outlines persistent psychological symptomatology of clinical severity, such as depression, anxiety, physicalization and sleep disturbances [72] in a significant proportion of patients.

Further, the correlation of the General Symptom Index with other clinical variables revealed significant associations. Higher psychological burden was negatively associated with perceived physical and mental health, suggesting limited physical mobility, low levels of vitality and reduced emotional resilience. In contrast to its negative association with well-being, psychological distress was positively associated with the intensity of symptoms such as shortness of breath, fatigue and lack of vitality, supporting the existence of a possible two-way mechanism that exacerbates psychological burden. In addition, susceptibility to the development of psychological burden appeared to be determined by demographic variables (age, gender) and specific clinical characteristics (coexisting conditions, respiratory symptoms), indicating increased vulnerability in selected subgroups of patients, namely patients of advanced age, the female gender and those with the aforementioned clinical characteristics [73,74].

In the same context, the complex evolution of long-term COVID symptomatology is also highlighted through the analysis of Descriptive Symptom Indicators, such as PSDI and PST. In particular, a significant decrease in the Positive Symptom Severity Index (PSDI) was recorded, positively associated with the observed improvement in vitality [75], in contrast to the increase in the Total Symptom Score (PST), a finding interpreted on the basis of the multifactorial nature of long COVID and the possible absence of targeted rehabilitation interventions [76].

Similarly, despite the observed decrease in depression and anxiety levels [77], particularly in patients with adequate sleep quality and social support [78], physicalization did not reveal a significant change [79]. This stability is probably due to the persistence of symptoms such as cognitive impairment [76], pain [80], fatigue [81] and respiratory symptoms.

In conclusion, compulsivity and paranoid ideation demonstrated an increasing trend [60,82] suggesting a possible increase in focusing on negative expectations of the future [83]. Conversely, phobic anxiety and interpersonal sensitivity decreased, possibly reflecting increased social adaptability [79,84], while, similarly, the anger/aggression and psychoticism variables did not reveal significant fluctuations, maintaining consistently low values throughout the follow-up period.

4.2. Interpretation of Temporal Changes in Long-Term COVID

This prospective cohort study investigated the long-term impact of COVID-19 hospitalization on mental health, sleep quality and quality of life in 117 patients over an 18-month period post-discharge. Data was collected at 2, 6, 12 and 18 months using standardized questionnaires (PSQI, SCL-90, SF-36, ESAS-R, PTSD Checklist) and demographic information. The study found a general trend of physical and psychological recovery, particularly by 6 months, but a complete return to pre-disease status was not achieved across all variables. Key findings include the persistence of psychological distress, low vitality and the influence of demographic factors and social support on recovery. The study highlights the complex, non-linear nature of recovery and the need for long-term, integrated care for long-term COVID patients.

In line with current research reports, the present study confirms that in patients with long-term COVID, the progression of their symptomatology and quality of life is neither linear nor similar for all. During recovery, patients’ physical and psychological status evolves dynamically, with initial remission of some symptoms, (e.g., long-term physical functioning and respiratory symptoms), with possible simultaneous remission and/or transient deterioration in others (e.g., mental health and fatigue).

The improvement, initially recorded on specific dimensions, probably reflects the intrinsic tendency to recovery after the acute phase of the disease. Nevertheless, the persistence or the worsening of symptoms suggests the involvement of ongoing pathophysiological mechanisms and demonstrates the need for targeted treatment protocols that address the multisystemic nature of the syndrome, taking into account the unique needs of each patient with long-term COVID.

In this perspective, the understanding of which is fund a mental for the development to effective interventions that will support the total restoration of their health, will provide effective treatment. This involves the development of individualized treatment plans and rehabilitation programs, customized to the extent and severity of symptoms, co-existing diseases and the psychological and social needs of each patient, while the coordinated action and collaboration of doctors of various specialities, physiotherapists, occupational therapists, psychologists and other health professionals is a non-negotiable condition for a more optimistic future for the patients’ recovery [85].

5. Conclusions

In the present study, the multisystemic nature of the long-term COVID syndrome and its extensive clinical effects on patients were revealed, the understanding of which is fundamental for the development of effective interventions that will support the total restoration of their health.

As mentioned above the psychological burden appeared to be determined by demographic variables (age, gender) indicating increased vulnerability in selected subgroups of patients, such as patients of advanced age and the female gender. The concept of sex differences is based on the assumption that depression is the same disorder in men and women, although there may be subtle phenotypic differences. Some authors have suggested that many men express depression in ways that are not captured in existing interviews and self-reports. This possibility is supported by the fact that some studies have shown that existing measures of psychological distress have lower concurrent validity in men than in women and are therefore less accurate in detecting depression in men. This does not rule out the possibility that sex differences in symptom patterns exist.

Indeed, there is a growing literature on how the social construction of masculinity varies across race, ethnicity and social class. Furthermore, gender differences in depression rates appear to be smaller in cultures where tolerance for antisocial and risk-taking behavior is lower. Therefore, it should be remembered that each work should be examined in its own cultural context and, if necessary, can be analyzed comparatively later.

There are currently no etiological treatments for the long/post-COVID phenomenon, and specific treatments that address the psychosocial needs of these patients are urgently needed. Understanding these factors is critical for future research, public policy and educational initiatives that can ultimately increase public awareness and improve clinical management and patient care.

Therefore, further, targeted research activities remain urgently needed. The research agenda requires in-depth analysis of the pathophysiological mechanisms of long COVID syndrome, identification of reliable biomarkers for early diagnosis and monitoring of disease progression, and development and clinical validation of personalized treatment and rehabilitation measures. Overall, there is a clear need for the development of targeted treatment options that take into account both the multisystem nature of the syndrome and the individual needs of each long COVID patient.

Creation of protocols for addressing long COVID syndrome is considered essential for the rapid reintegration of patients into their pre-COVID daily lives.