Secondary Logo

Journal Logo

Critical care

Emotional sequelae among survivors of critical illness: a long-term retrospective study

Kowalczyk, Michał; Nestorowicz, Andrzej; Fijałkowska, Anna; Kwiatosz-Muc, Magdalena

Author Information
European Journal of Anaesthesiology: March 2013 - Volume 30 - Issue 3 - p 111-118
doi: 10.1097/EJA.0b013e32835dcc45


Rapid progress in the management of critically ill patients in ICUs during the past few decades has resulted in approximately 70 to 80% survival today. Old and chronically ill patients, with a variety of disabilities and life-threatening disorders, can survive several years postdischarge.1 Many of these patients suffer from physical problems that include lack of mobility, weakness, fatigue, poor appetite, shortness of breath, disturbances in sleep patterns and hair loss.2,3 Many of them are also affected by adverse psychological and cognitive sequelae. Psychological problems include posttraumatic stress disorder (PTSD), anxiety and depression, cognitive impairment associated with loss of memory, attention deficits and dysexecutive syndrome.4–7 Many studies report the negative impact that these have on the quality of life.8–12

Treatment in ICU can be stressful and psychologically traumatic for patients. It may result in many unpleasant ICU-related memories associated with the immediate threat of life, as well as with nursing and treatment procedures such as tracheal intubation, cannulation of central veins and arteries, defibrillation and others. Additionally, sepsis, organ dysfunction, narcotics and night and day disturbances can cause delusional memories, nightmares13 or even delirium, which affect up to 80% of ICU patients.14 Following discharge from ICU, there are persistent disabilities related to treatment, chronic diseases and even inability to work due to organ or system failure, such as respiratory,15 nervous system16 or neurocognitive dysfunction,17 which may take up to a year to recover. In turn, this can lead to frustration and emotional disturbance that can exacerbate many physical disorders. The result is a vicious circle in which physicians search for physical disease and use various diagnostic and treatment procedures, whereas the underlying problems are of a psychological nature.

The immediate psychological impact of critical illness has been well reported in numerous studies,18–20 but with less detail relating to delayed residual effects that can sometimes manifest themselves several months after discharge or even later.6,21 What is more, depression following intensive care has been well investigated in numerous studies,12 whereas anxiety has not.

According to the WHO, health is not only the absence of infirmity and disease, but also a state of physical, mental and social well-being. It is a broad concept involving different dimensions of functioning of individuals, their environment, as well as their social and economic status and culture. A variety of environmental factors may affect the health of individuals; some are not health-related. At present, the influence of social and economic status, the level of education, the place of residence, marital status, family support, material and housing conditions, and working status on the rates of anxiety and depression has not been examined.

The main purpose of the study was to assess the influence of social, economic and working status on long-term anxiety and depression among convalescents after ICU discharge. In order to explore risk factors connected with anxiety and depression after ICU stay, possible effects of sex, age, duration of ICU stay, APACHE II (Acute Physiology and Chronic Health Evaluation) scores on admission, health status, time after discharge and ICU-related memories were also evaluated.


This retrospective, cross-sectional survey was approved by the Ethics Committee (Ethics Committee of Medical University of Lublin, Al. Raclawickie 1, 20–059 Lublin, Poland; protocol number: KE-0254/135/2010; date of approval: 1 July 2010). Written informed consent was obtained from all participants.

All adults who survived a minimum of 24 h in the general ICU of two hospitals (the Teaching Hospital, Medical University of Lublin and the District Hospital in Lublin) between January 2005 and December 2009 were eligible for recruitment. Lublin is a Polish city of 350 000 inhabitants. Both units admit surgical and medical cases, except for heart surgery patients. Adult survivors who sustained severe head injury or stroke were excluded.

Questionnaires were sent by post to all ICU survivors in December 2010. Data collected included age, sex, main diagnosis on admission, severity of acute illness (APACHE II score) and length of ICU stay. Five groups were formed according to time passed between discharge and the questionnaire:

  1. group 1: 1 to 2 years
  2. group 2: 2 to 3 years
  3. group 3: 3 to 4 years
  4. group 4: 4 to 5 years
  5. group 5: 5 years or more.

All patients were sent comprehensive letters describing the study in detail, and seeking signed informed consent.

The questionnaire consisted of five parts. The first part, the Hospital Anxiety and Depression Scale (HADS), was used to assess psychological distress. This brief self-assessment questionnaire is developed for use in medical settings. It has two subscales, one for anxiety and one for depression. Each subscale consists of seven items scored from 0 (absent) to 3 (severe). The maximum score for both depression and anxiety is 21 and the minimum score is 0. Interpretation of scores is as follows: 0 to 7, normal; 8 to 10, mild; 11 to 14, moderate; 15 to 21, severe. A score greater than 10 (moderate, severe) was suggestive of severe psychological distress. The measure has good internal consistency and is considered a reliable and valid psychological tool for clinically ill patient groups.22 In this study, HADS administration and scoring were performed according to Snaith and Zigmond.23 The second part consisted of questions to define social and economic status, including education, place of residence, marital status, number of children, support of relatives and/or friends, income self-evaluation (three options: not enough to live on, enough to live on, more than enough to live on) and material and housing conditions (bad, medium, good). The third part concerned working status before and after ICU stay (not working, manual work, white-collar work), whereas the fourth one evaluated health status before ICU stay (entirely healthy, chronically sick but able to work and perform daily activities, or chronically sick and unable to work and perform daily routines). The second, third and fourth parts were tested in previous studies, have good internal consistency and are considered adequate for ill patients.24 The fifth part included questions about ICU-related memories and readmissions to ICU.

Statistical analysis was performed using STATISTICA software, with a significance level of 0.05. Mean ± SD or frequencies and percentages were used for personal, health, social and economic characteristics of the patients. To compare respondents (patients who returned the questionnaire) and non-respondents (patients who did not return the questionnaire) according to age and APACHE II scale, the parametric Student's t-test was used. To compare respondents and non-respondents according to sex, reason for admission and time from discharge, the non-parametric χ2 test was used. Risks and odds ratios of severe or moderate anxiety and depression disorders versus mild or normal ones were estimated. Anxiety and depression were also analysed in points from 0 to 21 – with a higher score corresponding with the exhibition of a more severe disorder. Pearson's correlation coefficients (r) between anxiety and depression, between age (in years) and anxiety or depression, and between APACHE II and anxiety or depression, were calculated. Spearman's rank correlation coefficients (rs) between anxiety or depression and ordinal data such as educational level, material conditions, housing conditions and health status before ICU treatments were also calculated. Multiple regression was also used as a means of performing multivariate statistical analysis. Two regression models were used; anxiety (in points) was the dependent variable in the first model, whereas depression (in points) was the dependent variable in the second one. Numerical data (age, APACHE II) and the categorical binomial variables that were significant in odds ratios analysis were used as independent variables. Finally, multiple testing using Scheffe's test was conducted in order to compare anxiety and depression (in points) among different groups of categorical polynomial variables.


Patients’ characteristics

A total of 533 questionnaires were sent, and 195 (36.59%) were returned. Of these, nine were incomplete, leaving 186 for analysis. The study sample included 107 men (57.5%) and 79 women (42.5%) whose mean age on ICU admission was 48.1 ± 18.7 years. The mean APACHE II score on admission was 14.9 ± 6.7, and mean duration of ICU stay was 18 ± 24 days. Selected characteristics of the patients, including the main diagnosis at admission, time from discharge, health status before ICU stay and memories of their ICU treatment, are summarised in Table 1. There were no significant differences between respondents and non-respondents with regard to age, APACHE II on admission, sex, main diagnosis at admission and time from discharge. The majority of patients would accept another ICU admission if required (85.3%); only 14.5% would decline.

Table 1
Table 1:
Patients’ characteristics

Social and economic characteristics

Social and economic status data such as education, place of residence, marital status, number of children, support of relatives and/or friends, income self-evaluation, material conditions, housing conditions and working status are listed in Table 2. The majority of patients – 83 (44.9%) – were not working before ICU admission (retired, disability payment, unemployment benefit, students), 49 (26.5%) were manual workers and 53 (28.7%) were white-collar workers. After their ICU stay, 59.2 and 45.3% of manual and white-collar workers, respectively, stopped working.

Table 2
Table 2:
Social and economic characteristics

Risks and odds ratios analysis of severe or moderate anxiety and depression according to personal, health, social and economic factors

HADS indicated that of 186 patients, 64 (34.4%) had an anxiety disorder, 38 moderate and 26 severe, and 51 (27.4%) were depressed, 32 moderate and 19 severe. The results of risk factor analysis of severe and moderate anxiety and depression are shown in Table 3.

Table 3
Table 3:
Risks and odds ratios for severe or moderate anxiety and depression

Anxiety and depression intensity (in points) and correlation with personal, health, social and economic factors

The detailed HADS data were also analysed; the mean anxiety score was 8.68 ± 4.79 and the mean depression score was 7.82 ± 4.82. There was a strong positive correlation between anxiety and depression (r = +0.726, P < 0.001). Age correlated with higher depression scores (r = +0.259, P < 0.001), but not with anxiety (r = +0.070, P = 0.344). Both anxiety and depression correlated negatively with education; higher education resulted in lower HADS scores (correlation coefficient between education level and anxiety was rs = −0.257; P < 0.001; correlation coefficient between education level and depression rs = −0.211; P = 0.004). Higher income, better material and housing conditions correlated with lower anxiety and depression rates. Correlation coefficients were as follows: between income and anxiety, rs = −0.324 P < 0.001; between income and depression, rs = −0.254 P < 0.001; between material conditions and anxiety, rs = −0.370 P < 0.001; between material conditions and depression, rs = −0.332 P < 0.001; between housing conditions and anxiety, rs = −0.260 P < 0.001; and between housing conditions and depression, rs = −0.229 P = 0.002.

APACHE II scores on admission correlated significantly with both anxiety and depression, but correlation was not very strong (anxiety r = +0.187 P = 0.011, depression r = +0.239 P = 0.001). A higher APACHE II score resulted in higher anxiety and depression scores. A negative correlation between health status before ICU admission and HADS scores was observed for anxiety: rs = −0.193 P = 0.008, and for depression: rs = −0.227 P = 0.002. Better health status before treatment resulted in less anxiety and depression disorders.

Multiple regression analysis of anxiety and depression intensity according to personal, health, social and economic factors

Estimated multiple regression models of anxiety and depression as point scores are presented in Table 4. Patients with negative ICU-related memories had higher anxiety (about 1.7 points on average) and depression scores (about 1.6 points on average) than patients with positive ICU-related memories. Furthermore, patients who evaluated their income as less than sufficient to live on had higher anxiety (about 1.7 points on average) and depression (about 1.9 on average) scores than those who thought they had enough. Patients with good housing conditions had lower depression scores by 1.6 points (on average) than did those with poor or moderate housing conditions. Depression score was found to be higher among older patients by 0.08 points on average for every year of age. Other variables that were significant in univariate analysis turned out to be not significant in multivariate regression analysis due to correlation between predictors.

Table 4
Table 4:
Multiple regression results of anxiety and depression

Multiple testing of anxiety and depression intensity by patients’ hospital data, health, social and economic status

Multiple testing also revealed that higher education, support from friends and family, higher income, better material and housing conditions, permanency of work, better health status before ICU stay and positive memories from ICU treatment resulted in lower anxiety and depression scores. Detailed data are presented in Tables 5 and 6 (Supplemental Digital Content 1,


According to published data, survivors of acute respiratory distress syndrome (ARDS) exhibit major psychological and cognitive dysfunction25 which may be longlasting.26 Similar impairment has also been described in general intensive care patients.27 Our findings add to this by demonstrating high levels of anxiety (34%) and depression (27%) in patients with ICU stays longer than 24 h.

HADS scores have been used before to assess anxiety and depression in critically ill patients. Sukantarat et al.20 found that 3 months after discharge, 16% had an anxiety disorder, increasing to 22% after 9 months. For depression, the number was 24% after 3 months, reaching 31% after 9 months. These findings are similar to our results, although our sample size was considerably larger. For comparison, other studies have found the incidence of anxiety disorder to be 43%, and 30% were depressed.5 Another study found the figures to be 45% and 27%, respectively.18 In both of these studies, however, a cut-off value of at least 8 on the HADS for anxiety and depression was used; if lower cut-off values were assumed, the incidence of anxiety among discharged patients would increase to 55%, and of depression, to 50%.

The reasons why critically ill patients should suffer from anxiety and depression are difficult to pinpoint. This is because anxiety and depression are non-specific disorders and their morbidity is likely to be increased by pain, sleep disturbances, elevated cytokines, hypoxaemia, drugs, organ dysfunction and cerebral atrophy, all of which can form part of the experience of the critically ill.28–31 One hypothesis attributes these disorders to the delusional memories triggered after an ICU stay. Ringdal et al.32 studied patients with delusional memories after intensive posttrauma treatment and found higher levels of anxiety and depression symptoms. Their health-related quality of life (HRQoL) was also affected. Similarly, in our study, the risk of severe or moderate anxiety and depression symptoms was higher among patients with negative ICU-related memories. It must be remembered that treatment in ICU is associated with many unpleasant nursing and therapeutic procedures that could harm the subconscious and promote psychological sequelae.

A second mechanism could be related to the trauma itself, as it is the life-threatening injury that precipitates the critical illness. Patients with trauma are usually healthy before the episode, so their situation changes immediately from being well to instant disability, and from an active lifestyle to requiring some form of long-lasting rehabilitation. They may also have the fear that they would not recover. Thus, patients with trauma suffer from lower HRQoL, demonstrate some adverse psychological reactions and show higher levels of anxiety and depression symptoms.5,32 Head injury is especially associated with higher levels of anxiety, and has been reported as a risk for psychological distress,33 but in our study, trauma patients constituted only 27% of the total sample, so this could not be the only reason for our findings.

Another issue is PTSD, which can complicate trauma, as Holbrook et al.34 found in adolescents after major trauma. In their report, the long-term PTSD rate was 27% and was significantly and strongly associated with low socioeconomic status. Even unpleasant nursing procedures can be recalled by patients as traumatic events and lead to the development of PTSD. Both Schelling et al.4 and Kapfhammer21 reported that patients after ARDS had significantly higher PTSD scores than did the control groups. After a stay in a general ICU, Scragg et al.5 found that a high rate of PTSD correlated with anxiety and depression symptoms.

Some explanation for anxiety and depression symptoms can be found in the presence of pre-existing disease among ICU patients, as this has been shown to have a significant impact on HRQoL.35 However, the frequency of pre-existing disease in patients discharged from an ICU in that study was 74%, whereas in our study, about half of patients reported themselves as being entirely healthy before their ICU stay. One form of pre-existing condition is psychiatric disorder and the prevalence in the general population needs to be considered. Bijl et al.36 have shown that the frequency of mood disorders in a Dutch population was 19% during their entire lifetime, but only 5.7% during the preceding 12-month period. With regard to anxiety disorders, the figures were 19.3% over a lifetime and only 8.3% during the preceding 12-month period.

Prompted by the WHO definition of health, we decided to evaluate the effect of social and economic status on HADS scores. We concluded that factors not directly associated with medical care, but with social and economic status, also influenced the emotional condition of the patients enrolled in our study. We found that patients with higher education, family support and better material and housing conditions were in an evidently better emotional state. In addition, family and/or friends’ support gave our patients the ability to overcome a fear of the future. Better economic status also predisposed patients to see positives (e.g. better rehabilitation and faster recovery) after severe illness. Moreover, our findings revealed that anxiety and depression were most common among patients without employment. This, we think, is because they lack the permanency of work and with it the drive to go back to society and their daily routine. Our findings are consistent with those of Bijl et al.36 who conducted a large study of 7076 people in Holland. This study revealed a higher prevalence of mood and anxiety disorders among people with lower income. When education and occupational status were taken into account, the picture was largely the same. People with highest levels of education had lower mood and anxiety morbidity rates. Furthermore, the occupational situation played a role: mental disorders were more prevalent among those unemployed and with occupational disabilities.

All the above-mentioned factors have the ability to affect the emotional condition of individuals after their ICU stay. We take the view that it would be beneficial to ascertain the social and economic condition of our patients when assessing them for further rehabilitation and care, and in selected cases, refer them to social services or non-governmental social assistance organisations. This might help reduce future emotional sequelae but further studies would be required to identify more precisely those social and economic factors that enable better medical care and a more suitable holistic approach.

The main limitation of the study was its cross-sectional type and that there was only a 37% survey return rate with potential for response bias. However, although the rate of response was low, patients withholding the questionnaire did not differ significantly in terms of their general characteristics from those returning it, so bias should not be very strong. Moreover, in our study, assessment of anxiety and depression a long time after ICU treatment might be subject to influence from additional health and environmental factors that were not taken into account and would require further multilateral studies. The other limitation was our inability to assess the rate of anxiety and depression prior to ICU admission, as the need to assess these factors could never be anticipated. The next limitation was that the assessment of social and economic status was completed at the same time as the assessment of symptoms of anxiety and depression, so its objectivity might be influenced by the patients’ mental status. Assessment of health status prior to critical illness was also subjective and could bias this study, as anxiety and depression might affect the patients’ evaluation of their well-being, even retrospectively. What is more, HADS, being a screening test, should not be used to diagnose anxiety or depression, and should be considered merely a suitable tool to flag up a group of patients requiring specialist attention and further treatment. The final limitation was that multiple testing is desirable but would require a larger sample size to make statistical calculations valid.


In conclusion, our data demonstrate that more than one third of patients after their ICU discharge had abnormal levels of anxiety persisting for several years, and about one quarter had abnormal levels of depression. Furthermore, adverse social and economic status was found to be associated with higher rates of anxiety and depression disorders. Factors connected with severity of illness on admission (such as those revealed in the patients’ APACHE II scores), and with preadmission health status, have an effect on long-term anxiety and depression rates.


Assistance with the study: none declared.

Financial support and sponsorship: none declared.

Conflicts of interest: none declared.


1. Griffiths RD, Jones C. Recovery from intensive care. BMJ 1999; 319:427–429.
2. Brooks R, Kerridge R, Hillman K, et al. Quality of life outcomes after intensive care. Comparison with a community group. Intensive Care Med 1997; 23:581–586.
3. Eddleston JM, White P, Guthrie E. Survival, morbidity, and quality of life after discharge from intensive care. Crit Care Med 2000; 28:2293–2299.
4. Schelling G, Stoll C, Haller M, et al. Health-related quality of life and posttraumatic stress disorder in survivors of the acute respiratory distress syndrome. Crit Care Med 1998; 26:651–659.
5. Scragg P, Jones A, Fauvel N. Psychological problems following ICU treatment. Anaesthesia 2001; 56:9–14.
6. Hopkins RO, Weaver LK, Collingridge D, et al. Two-year cognitive, emotional, and quality-of-life outcomes in acute respiratory distress syndrome. Am J Respir Crit Care Med 2005; 171:340–347.
7. Sukantarat KT, Burgess PW, Williamson RCN, Brett SJ. Prolonged cognitive dysfunction in survivors of critical illness. Anaesthesia 2005; 60:847–853.
8. Cuthbertson BH, Scott J, Strachan M, et al. Quality of life before and after intensive care. Anaesthesia 2005; 60:332–339.
9. Kowalczyk M, Wrońska-Sewruk A, Nestorowicz A. Quality of life after intensive care. Anaesthesiol Intensive Ther 2008; 40:162–168.
10. Jackson JC, Mitchell N, Hopkins RO. Cognitive functioning, mental health, and quality of life in ICU survivors: an overview. Crit Care Clin 2009; 25:615–628.
11. Tøien K, Myhren H, Bredal IS, et al. Psychological distress after severe trauma: a prospective 1-year follow-up study of a trauma intensive care unit population. J Trauma 2010; 69:1552–1559.
12. Davydow DS, Gifford JM, Desai SV, et al. Depression in general intensive care unit survivors: a systematic review. Intensive Care Med 2009; 35:796–809.
13. Kiekkas P, Theodorakopoulou G, Spyratos F, Baltopoulos GI. Psychological distress and delusional memories after critical care: a literature review. Int Nurs Rev 2010; 57:288–296.
14. Truman B, Ely EW. Monitoring delirium in critically ill patients. Using the confusion assessment method for the intensive care unit. Crit Care Nurse 2003; 23:25–36.
15. Herridge MS, Cheung AM, Tansey CM, et al. One-year outcomes in survivors of the acute respiratory distress syndrome. N Engl J Med 2003; 348:683–693.
16. Zifko UA. Long-term outcome of critical illness polyneuropathy. Muscle Nerve 2000; 23:S49–52.
17. Newman MF, Kirchner JL, Phillips-Bute B, et al. Longitudinal assessment of neurocognitive function after coronary-artery bypass surgery. N Engl J Med 2001; 344:395–402.
18. Rattray JE, Johnston M, Wildsmith JAW. Predictors of emotional outcomes of intensive care. Anaesthesia 2005; 60:1085–1092.
19. Sukantarat K, Greer S, Brett S, Williamson R. Physical and psychological sequelae of critical illness. Br J Health Psychol 2007; 12:65–74.
20. Sukantarat KT, Williamson RCN, Brett SJ. Psychological assessment of ICU survivors: a comparison between the Hospital Anxiety and Depression scale and the Depression, Anxiety and Stress scale. Anaesthesia 2007; 62:239–243.
21. Kapfhammer HP, Rothenhäusler HB, Krauseneck T, et al. Posttraumatic stress disorder and health-related quality of life in long-term survivors of acute respiratory distress syndrome. Am J Psychiatry 2004; 161:45–52.
22. Zigmond AS, Snaith RP. The Hospital Anxiety and Depression Scale. Acta Psychiatr Scand 1983; 67:361–370.
23. Snaith RP, Zigmond AS. The Hospital Anxiety and Depression Scale with the Irritability-Depression-Anxiety Scale and the Leeds situational Anxiety Scale. Manual. London, UK: NferNelson Publishing Company; 1994. pp. 1–15.
24. Wójcik M. Ambulatoryjna opieka kardiologiczna w opinii pacjentów [dissertation]. Lublin, Poland: Medical University of Lublin; 2003. p. 286.
25. Hopkins RO, Weaver LK, Pope D, et al. Neuropsychological sequelae and impaired health status in survivors of severe acute respiratory distress syndrome. Am J Respir Crit Care Med 1999; 160:50–56.
26. Rothenhäusler HB, Ehrentraut S, Stoll C, et al. The relationship between cognitive performance and employment and health status in long-term survivors of the acute respiratory distress syndrome: results of an exploratory study. Gen Hosp Psychiatry 2001; 23:90–96.
27. Jackson JC, Hart RP, Gordon SM, et al. Six-month neuropsychological outcome of medical intensive care unit patients. Crit Care Med 2003; 31:1226–1234.
28. Katz IR. Is there a hypoxic affective syndrome? Psychosomatics 1982; 23:846–853.
29. Katz IR. On the inseparability of mental and physical health in aged persons: lessons from depression and medical comorbidity. Am J Geriatr Psychiatry 1996; 4:1–16.
30. Nelson BJ, Weinert CR, Bury CL, et al. Intensive care unit drug use and subsequent quality of life in acute lung injury patients. Crit Care Med 2000; 28:3626–3630.
31. Pearlson GD, Rabins PV, Kim WS, et al. Structural brain CT changes and cognitive deficits in elderly depressives with and without reversible dementia (’pseudodementia’). Psychol Med 1989; 19:573–584.
32. Ringdal M, Plos K, Lundberg D, et al. Outcome after injury: memories, health-related quality of life, anxiety, and symptoms of depression after intensive care. J Trauma 2009; 66:1226–1233.
33. Jackson JC, Obremskey W, Bauer R, et al. Long term cognitive, emotional, and functional outcomes in trauma intensive care unit survivors without intracranial hemorrhage. J Trauma 2007; 62:80–88.
34. Holbrook TL, Hoyt DB, Coimbra R, et al. Long-term posttraumatic stress disorder persists after major trauma in adolescents: new data on risk factors and functional outcome. J Trauma 2005; 58:764–771.
35. Orwelius L, Nordlund A, Edéll-Gustavsson U, et al. Role of preexisting disease in patients’ perceptions of health-related quality of life after intensive care. Crit Care Med 2005; 33:1557–1564.
36. Bijl RV, Ravelli A, van Zessen G. Prevalence of psychiatric disorder in the general population: results of the Netherlands Mental Health Survey and Incidence Study (NEMESIS). Soc Psychiatry Psychiatr Epidemiol 1998; 33:587–595.
© 2013 European Society of Anaesthesiology