Quality of Life and Family Support in Critically Ill Patients following ICU Discharge
, George Fotakopoulos and Demosthenes Makris *Round 1
Reviewer 1 Report
This is an interesting topic of how family support influences the quality of life of patients after being discharged from the ICU, but it should be reviewed:
Neither the abstract nor the objectives specify the lung function assessment performed. It should be assessed whether the inclusion of data on lung function is necessary to answer the objectives set out in the study.
As an inclusion criterion, it is stated that the patient must perform a 6MWT, this criterion seems to be difficult to achieve at the time of discharge from the ICU. It seems more logical to include patients and exclude them if they are unable to perform the 6MWT once they are released from the ICU and before they leave the hospital. Clarify when the 6MWT was done. Figure 1 expresses as an exclusion criterion not having inclusion criteria. First, there are inclusion criteria and then the patients are excluded according to the defined exclusion criteria.
Data collection is included that is not described in the tables later. Variables should be described (severity level (APACHE II), mechanical ventilation, type of patients). The reason for admission must be described with excluding variables.
Quality of life: How the SF-36 questionnaire is completed (in person, by phone, etc.) should be described. Previous quality of life is a very important factor in assessing the evolution of quality of life.
In the objectives paragraph: it is stated that the objective is to evaluate the impact of the family on the QoL in a period of one year. Not only is the family valued, but also in an environment of friends. Patients are classified (and this should be explained) according to the support received once they have left the ICU.
In the statistical analysis section: Better specify the methodology used in the analysis of repeated measures. It is stated that a correlation analysis will be used which is not shown in the results. The results of the continuous variables are expressed as mean (SD) even if they do not follow a normal distribution. It is better to express non-normal data as the median (IQR).
The SF-36 QoL results are expressed as a single measure. Clarify in methodology the calculation of this value. The SF-36 is usually expressed as two measures (physical and mental component).
The demographic and assessment results of the QoL should be differentiated according to the type of support (family or friends) to show if there are differences. For example, the age could be different if only family or friends give the support. Tables should be included where these differences could be observed, differentiating patients according to the type of support received.
Table 3 differentiates the patients according to the global value of SF-36. In methodology, it should be clarified how these categories are differentiated. Table 3 does not show an association between variables, it only shows whether there are differences in some variables according to the chosen partition criteria. The patients in the group with the highest QoL are different in some variables. No results are shown according to the different domains of the SF-36.
The conclusion of the work does not seem to correspond with the results obtained. It must be expressed by clarifying the concepts of association and groups of patients analysed in the results section.
Author Response
Dear Editor,
We were pleased to hear that the Healthcare Journal is interested in a revised version of our manuscript. We are grateful to the editors and the reviewers for the valuable suggestions and comments. After careful consideration, we have adjusted the manuscript accordingly. All comments have been addressed in the revised version. In more detail:
Below there is a point-by-point description of the changes that have been made to address the editors’ and reviewers' concerns.
EDITOR’S AND REVIEWERS’ COMMENTS (CAPITAL, BLACK)
Answer (lower case, blue)
- NEITHER THE ABSTRACT NOR THE OBJECTIVES SPECIFY THE LUNG FUNCTION ASSESSMENT PERFORMED. IT SHOULD BE ASSESSED WHETHER THE INCLUSION OF DATA ON LUNG FUNCTION IS NECESSARY TO ANSWER THE OBJECTIVES SET OUT IN THE STUDY.
Thank you for your comment. We agree that this point should be clarified better. As it was pointed out by previous publications (Sidiras 2019), ICU patients my present functional and pulmonary deterioration together with impairment of QoL following ICU admission. Patients included in the present study were critical care patients and in their majority patients who were treated with mechanical ventilation more than 48 hours. In this respect, we aimed to assess in addition to QoL, lung function and exercise performance in 6MWT. We specify in the abstract and in methods lung function assessment performed. In this respect, we have added in Abstract (page 1, lines 11-12 & 16-17), in Introduction (page 2, line 63-64) and Methods (page 2, lines 80-81) lung function assessment details and secondary objectives of the study.
- AS AN INCLUSION CRITERION, IT IS STATED THAT THE PATIENT MUST PERFORM A 6MWT, THIS CRITERION SEEMS TO BE DIFFICULT TO ACHIEVE AT THE TIME OF DISCHARGE FROM THE ICU. IT SEEMS MORE LOGICAL TO INCLUDE PATIENTS AND EXCLUDE THEM IF THEY ARE UNABLE TO PERFORM THE 6MWT ONCE THEY ARE RELEASED FROM THE ICU AND BEFORE THEY LEAVE THE HOSPITAL. CLARIFY WHEN THE 6MWT WAS DONE. FIGURE 1 EXPRESSES AS AN EXCLUSION CRITERION NOT HAVING INCLUSION CRITERIA. FIRST, THERE ARE INCLUSION CRITERIA AND THEN THE PATIENTS ARE EXCLUDED ACCORDING TO THE DEFINED EXCLUSION CRITERIA.
You are right that this point needs also clarification. Patients were included if they were able to perform spirometry, 6MWT and to complete the study questionnaire at the day of hospital discharge based on treating physicians’ judgement. In light of your comment, we have modified the description (method, page 2, line 70-72) in inclusion criteria and we modified Figure 1 accordingly.
- DATA COLLECTION IS INCLUDED THAT IS NOT DESCRIBED IN THE TABLES LATER. VARIABLES SHOULD BE DESCRIBED (SEVERITY LEVEL (APACHE II), MECHANICAL VENTILATION, TYPE OF PATIENTS). THE REASON FOR ADMISSION MUST BE DESCRIBED WITH EXCLUDING VARIABLES.
Thank you for your comment. We have now added description for variables such as APACHE II in Methods (page 3, lines 86-88) and in Results (Table 2). We regret that we have no details for mechanical ventilation variables such as ventilator mode or settings of MV variables or mechanical properties of the respiratory system. This has been added in the Discussion section as a limitation (Page 13, line 329-332).
- QUALITY OF LIFE: HOW THE SF-36 QUESTIONNAIRE IS COMPLETED (IN PERSON, BY PHONE, ETC.) SHOULD BE DESCRIBED. PREVIOUS QUALITY OF LIFE IS A VERY IMPORTANT FACTOR IN ASSESSING THE EVOLUTION OF QUALITY OF LIFE.
The SF-36 questionnaire was completed in person by each participant. Participants were seen in outpatient clinics by study investigators and in cases where the presence of the participant to the hospital was impossible, questionnaires and lung function studies were performed at home. This information was added in methods (page 3, lines 110-112). We agree that previous status of QoL is important in assessing the evolution of the QoL. We regret we had no QoL data previous to this study to provide a more in-depth analysis.
- IN THE OBJECTIVES PARAGRAPH: IT IS STATED THAT THE OBJECTIVE IS TO EVALUATE THE IMPACT OF THE FAMILY ON THE QOL IN A PERIOD OF ONE YEAR. NOT ONLY IS THE FAMILY VALUED, BUT ALSO IN AN ENVIRONMENT OF FRIENDS. PATIENTS ARE CLASSIFIED (AND THIS SHOULD BE EXPLAINED) ACCORDING TO THE SUPPORT RECEIVED ONCE THEY HAVE LEFT THE ICU.
Thank you for your comment. Classification of support received by spouses, family, and friends was based on previous literature (Ashwin. et al, 2018; Davidson. et al, 2017). We provide more details for this classification in Methods (page 3, line 94-98).
- IN THE STATISTICAL ANALYSIS SECTION: BETTER SPECIFY THE METHODOLOGY USED IN THE ANALYSIS OF REPEATED MEASURES. IT IS STATED THAT A CORRELATION ANALYSIS WILL BE USED WHICH IS NOT SHOWN IN THE RESULTS. THE RESULTS OF THE CONTINUOUS VARIABLES ARE EXPRESSED AS MEAN (SD) EVEN IF THEY DO NOT FOLLOW A NORMAL DISTRIBUTION. IT IS BETTER TO EXPRESS NON-NORMAL DATA AS THE MEDIAN (IQR).
Thank you for your comment. We have now modified the statistical section aiming to present better our analysis. Based on your remark, we now present non-normally distributed data as Median (IQR).
- THE SF-36 QOL RESULTS ARE EXPRESSED AS A SINGLE MEASURE. CLARIFY IN METHODOLOGY THE CALCULATION OF THIS VALUE. THE SF-36 IS USUALLY EXPRESSED AS TWO MEASURES (PHYSICAL AND MENTAL COMPONENT).
You are right that SF 36 results might be better presented as the two different measures of the SF36 tool. In this respect we have modified our presentation and we present the two different components of the questionnaire. On this basis we have replace previous. Table 3 with two new tables, one for the mental component and one for the physical component of the questionnaire.
- THE DEMOGRAPHIC AND ASSESSMENT RESULTS OF THE QOL SHOULD BE DIFFERENTIATED ACCORDING TO THE TYPE OF SUPPORT (FAMILY OR FRIENDS) TO SHOW IF THERE ARE DIFFERENCES. FOR EXAMPLE, THE AGE COULD BE DIFFERENT IF ONLY FAMILY OR FRIENDS GIVE THE SUPPORT. TABLES SHOULD BE INCLUDED WHERE THESE DIFFERENCES COULD BE OBSERVED, DIFFERENTIATING PATIENTS ACCORDING TO THE TYPE OF SUPPORT RECEIVED.
Thank you for your comment. In light of your recommendation, we analysed the relationship between demographic variables such as age, sex, marital status, educational level, professional status and the type of support (family or friends). Statistically significant results are now presented in Results (page 9, line 203-209).
- TABLE 3 DIFFERENTIATES THE PATIENTS ACCORDING TO THE GLOBAL VALUE OF SF-36. IN METHODOLOGY, IT SHOULD BE CLARIFIED HOW THESE CATEGORIES ARE DIFFERENTIATED. TABLE 3 DOES NOT SHOW AN ASSOCIATION BETWEEN VARIABLES, IT ONLY SHOWS WHETHER THERE ARE DIFFERENCES IN SOME VARIABLES ACCORDING TO THE CHOSEN PARTITION CRITERIA. THE PATIENTS IN THE GROUP WITH THE HIGHEST QOL ARE DIFFERENT IN SOME VARIABLES. NO RESULTS ARE SHOWN ACCORDING TO THE DIFFERENT DOMAINS OF THE SF-36.
Thank you for your valuable comment. In light of this remark and of your previous one, we now present the two different components of SF36 score (physical and mental) (Tables 3 and 4). In addition, in Methods section (page 3 lines 105-108) we describe that we use the median score to differentiate participants in those with higher scores (≥ median) or not (<median). We acknowledge that this criterion was arbitrary selected and this is stated in Methods. We agree that Tables 3 and 4 does not show any association but the differences in variables according to the chosen partition criteria, i.e. SF36 score over or under the median score. In light of your remark, we modified results (page 8 lines 203-217). Ιn light of your third remark, we now present differences in those variables according to the two different domains of the SF36 (Tables 3 & 4).
- THE CONCLUSION OF THE WORK DOES NOT SEEM TO CORRESPOND WITH THE RESULTS OBTAINED. IT MUST BE EXPRESSED BY CLARIFYING THE CONCEPTS OF ASSOCIATION AND GROUPS OF PATIENTS ANALYSED IN THE RESULTS SECTION.
Thank you for your comment. The conclusion was modified based on your remark (page 12, line 337-342).
Kind regards
DEMOSTHENES MAKRIS
Professor in Critical Care Medicine
Author Response File: 
Author Response.doc
Reviewer 2 Report
please see the attachment
Comments for author File: 
Comments.pdf
Author Response
Dear Editor,
We were pleased to hear that the Healthcare Journal is interested in a revised version of our manuscript. We are grateful to the editors and the reviewers for the valuable suggestions and comments. After careful consideration, we have adjusted the manuscript accordingly. All comments have been addressed in the revised version. In more detail:
Below there is a point-by-point description of the changes that have been made to address the editors’ and reviewers' concerns.
EDITOR’S AND REVIEWERS’ COMMENTS (CAPITAL, BLACK)
Answer (lower case, blue)
REVIEWER 2
IN THIS PROSPECTIVE STUDY, KONSTANTINA ET AL. ATTEMPT TO DEMONSTRATE HOW THE QUALITY OF LIFE (QOL) OF PATIENTS UP TO 12 MONTHS AFTER ICU DISCHARGE CAN BE IMPROVED BY SUPPORTIVE FAMILY/FRIENDS. THE AUTHORS USED A DEDICATED QOL QUESTIONNAIRE AND ASSESSED SPIROMETRY VALUES AND SIX-MINUTE WALKING TESTS OVER THE SAME TIME PERIOD. THIS STUDY WAS CONDUCTED IN ONE HOSPITAL IN GREECE BETWEEN 2020-2021.
THIS RAISED AN OVERALL RELEVANT POINT OF THE HOLISTIC APPROACH OF CARING FOR SOMEONE WHO WAS ADMITTED TO THE ICU. I HAVE SEEN IN MY PRACTICE THE LONG-LASTING IMPACT OF ICU INDUCED PTSD AND HOW IMPAIRED THESE PATIENTS CAN BE AFTER DISCHARGE AND UNDERSTANDING HOW MUCH SUPPORT THESE PATIENTS HAVE AT HOME MAY PREVENT FURTHER COMPLICATIONS SURROUNDING MENTAL HEALTH.
- I WONDER HOW DIFFERENT THE RESULTS WOULD HAVE BEEN NOW OR BEFORE THE COVID-19 PANDEMIC. I AM ASSUMING THAT SOME OF THESE PATIENTS WOULD HAVE BEEN ADMITTED WELL WITHIN THE FIRST WAVE OF COVID AND THEREFORE THE VISITING RIGHTS TO THESE PATIENTS MAY HAVE BEEN REMOVED/SIGNIFICANTLY REDUCED AS IT WAS IN THE HOSPITAL, I WORKED IN. COULD THE AUTHORS CLARIFY.
Thank you for your comment. You are right that the answer in such a question could be useful. Unfortunately, we cannot answer to it because patients were recruited at late 2020 and 2021 and there was but few patients (recruited and followed) before the pandemic.
- CAN THE AUTHORS CLARIFY IF THE PATIENTS WITH STROKES WERE ADMITTED TO THE ICU? IF NOT, WHY ARE THEY INCLUDED?
Thank you for your comment. Causes of admission are depicted in Table 2. Among critical care patients with neurological problems there were also patients with cardiovascular events (i.e. stroke). Criteria for admission in the ICU were decided by treating ICU physicians. In light of your comment a relevant clarification was added in methods (page 2, lines 89-90).
- AS LATER DISCUSSED, STROKES HAVE A HIGHER CHANCE OF CAUSING DEBILITATING CHANGES TO A PATIENT QOL. IT WOULD HAVE BEEN USEFUL TO KNOW IF THESE PATIENTS WERE DISCHARGED WITH A PACKAGE OF CARE TO SUPPORT THEM OR THEIR FAMILY AT HOME.
Thank you for your comment. This was an observational study and, in this respect, there was no study intervention following patients’ discharge. However, patients with stroke in our institution are given specific bundles of care following their discharge. Nevertheless, this care may be highly variable. For example, in our study some patients were discharged home and almost half of them (48.9%) were transferred in private rehabilitation centers. We acknowledge that the support given by public and private health services to patients following their discharge is not standard; we thus cannot evaluate the impact of this variable in the QoL of patients. In light of your comment, we discuss this limitation in the relevant section (page 12-13, lines 327-329).
- CAN THE AUTHORS CLARIFY THE STANDARD POPULATIONS SF-36 SCORE. DID THE AUTHORS CONSIDER MEASURING THEIR VARIABLES FOR LONGER THAN A YEAR? TO SEE IF THEIR SCORE EVER RETURNS TO A NUMBER SIMILAR TO THE STANDARD POPULATION.
You are right that knowledge of SF score in a standard population may be useful to draw better conclusions from our results. Unfortunately, there is no data for a similar population in Greece. However, the mean SF score in a population of patients of Sweden with heart disease “CHD” (Nilsson et al, 2020) was 70 ± 21. In this respect one might argue that the mean SF score in the population of our study was respectively lower. In light of your comment a relevant comment was added in discussion (page 11, lines 249-252).
- THE AUTHORS DO NOT HAVE A SECTION ON LIMITATIONS OF THEIR STUDY. OBVIOUS ONES INCLUDE, ONLY USING ONE HOSPITAL ICU SO PATIENTS WILL BE LIMITED TO THE DEMOGRAPHIC OF THAT AREA IN GREECE AND IS THEREFORE NOT REFLECTIVE OF THE OVERALL GENERAL POPULATION. A RELATIVELY SMALL SAMPLE SIZE SO ARE UNABLE TO EVALUATE SUBGROUPS E.G. COMORBIDITIES, YOUNGER VS. OLDER PATIENTS. DID THE AUTHORS CONSIDER TRYING TO COLLECT INFORMATION (POSSIBLY FROM SPOUSE) ABOUT PRE- HOSPITAL STAY BASELINE? WERE SOME PATIENTS VERY ACTIVE? WERE SOME PATIENTS ALREADY ON ANTIDEPRESSANTS? WERE PATIENTS REVIEWED BY PSYCHIATRY WHEN STAYING IN ICU. OR DID PATIENTS SEEK OUT MEDICAL TREATMENT FOR MENTAL HEALTH AFTER DISCHARGE? DID THE AUTHORS CONSIDER COMPARING SURGICAL VS MEDICAL ICU PATIENTS? AS THEY CAN BE VERY DIFFERENT.
You are right that the manuscript could be a lot more improved with the addition of a paragraph stating the limitations of the study which they should be underlined when interpreting its findings. First, this is a one center study and represents data from a specific area in central Greece. On the other hand, this tertiary center provides services to a relatively large population (600.00 to one million people) and the results of the study could be useful in implementing strategies at a local level. We acknowledge that the sample size of the population studied is relatively small to evaluate subgroups. Moreover, we acknowledge that the questionnaire used in the study did not included questions regarding details of patients’ prehospital stay or for specific mental support following their discharge from hospital. These details may have provided more insight in the evolution of patient’s health. Based on your remark we discuss those limitations in page 12, lines 319-324.
SOME OF THIS PAPER WAS HARD TO READ DUE TO STRUCTURE OF SENTENCES. A LIST OF SOME ISSUES ARE AS FOLLOWS:
LINE 8: FOLLOWING DISCHARGE FROM THE INTENSIVE CARE UNIT (ICU), CRITICALLY ILL .
Donne as recomended
LINE 85- NEXT OF KIN NOT NEXT OF KEEN.
Donne as recomended
LINE 106. SENTENCE NEEDS A , AFTER FINISHED
Donne as recomended
TABLE 1 PAGE 5 — TIME TO RETURN TO DAILY ROUTINE AFTER ICU? RATHER THAN TIME TO RETURN AT ...
Donne as recomended
LINE 175- DON'T NEED "AT HOSPITAL" AFTER "HOSPITALIZED AT HOSPITAL"
Donne as recomended
LINE 162- PRESENTED WITH INCREASED LUNG FUNCTION
Donne as recomended
LINE 181 NEEDS SOME REWORDING AND ABBREVIATION OF QUALITY OF LIFE — EVIDENCE SHOWS THAT EVEN AFTER YEARS FROM THEIR ICU ADMISSION, PATIENTS QOL HAS A SIGNIFICANT DECREASE COMPARED TO THE GENERAL HEALTHY POPULATION.
Donne as recomended
191 — IS PROVIDED BY 90% OF THEIR FAMILY.
Donne as recomended
LINE 192-194. REMOVE PARENTHESES AT THE END. NEEDS REPHRASING.
Donne as recomended
205- FEMALE NEEDS TO BE PLEURAL
Donne as recomended
- REMOVE PARENTHESES AT END OF SENTENCE
Donne as recomended
- REMOVE PARENTHESES AT END OF SENTENCE
Donne as recomended
- REMOVE PARENTHESES AT END OF SENTENCE
Donne as recomended
- REMOVE PARENTHESES IN THE MIDDLE OF THE SENTENCE
Donne as recomended
- REMOVE PARENTHESES
Donne as recomended
136- SPELT 142 AND THEN IMMEDIATELY USED 143 AS A NUMBER. NEEDS CONSISTENCY.
Donne as recomended
FIGURE 4 IS MISSING RIGHT Y AXIS LABEL.
Donne as recomended
Kind regards
DEMOSTHENES MAKRIS
Professor in Critical Care Medicine
Author Response File: Author Response.doc
Reviewer 3 Report
This is an interesting study.
However there are several potential confounding variables which affect the conclusions. The improved quality of life while correlated with the received care by more than 2 members of their families, it is possible that the patients who were older (who also had lower QoL scores) may have had no spouses/family members due to their advanced age (i.e. spouse passed away). We know that older persons have less social networks and therefore the difference in QoL may be due to age and associated comorbid conditions than the presence of social support itself.
Author Response
Dear Editor,
We were pleased to hear that the Healthcare Journal is interested in a revised version of our manuscript. We are grateful to the editors and the reviewers for the valuable suggestions and comments. After careful consideration, we have adjusted the manuscript accordingly. All comments have been addressed in the revised version. In more detail:
Below there is a point-by-point description of the changes that have been made to address the editors’ and reviewers' concerns.
EDITOR’S AND REVIEWERS’ COMMENTS (CAPITAL, BLACK)
Answer (lower case, blue)
REVIEWER 3
THE IMPROVED QUALITY OF LIFE WHILE CORRELATED WITH THE RECEIVED CARE BY MORE THAN 2 MEMBERS OF THEIR FAMILIES, IT IS POSSIBLE THAT THE PATIENTS WHO WERE OLDER (WHO ALSO HAD LOWER QOL SCORES) MAY HAVE HAD NO SPOUSES/FAMILY MEMBERS DUE TO THEIR ADVANCED AGE (I.E. SPOUSE PASSED AWAY). WE KNOW THAT OLDER PERSONS HAVE LESS SOCIAL NETWORKS AND THEREFORE THE DIFFERENCE IN QOL MAY BE DUE TO AGE AND ASSOCIATED COMORBID CONDITIONS THAN THE PRESENCE OF SOCIAL SUPPORT ITSELF.
Thank you for your comments. Based on your remark we have added a relevant comment in Discussion.
Kind regards
DEMOSTHENES MAKRIS
Professor in Critical Care Medicine
Author Response File: Author Response.doc
