How Distressed Are Cancer Patients when They Get Discharged from Hospital? Patients’ Distress and its Associations with Health Care Professionals’ Communication Skills and Perceived Stress

Objective: For cancer patients, the period between hospital discharge and outpatient follow-up can result in “distressand” patients need to be prepared by health care professionals (HCP) for that while being in hospital. Adequate communication is important for that. HCPs have often high levels of stress which can have a negative impact on HCPs’ communication. This study examines patients’ distress, HCPs’ communication, HCPs’ perceived stress and the relationships between them. Methods: Sixty-two cancer patients answered questionnaires on their distress (anxiety, depression, symptom burden) two days before and two days after hospital discharge and evaluated HCPs’ communication. Thirty-eight HCPs’, in turn, evaluated their perceived stress. Fifty-three patient data sets and 38 HCP data sets were included and analyzed descriptively, and by linear regression. Results: Preand post-discharge anxiety and depression were in the normal range and symptom burden was low. However, approximately 10% had a substantial level of anxiety or depression pre-discharge and approximately 20% post-discharge. Correlations were found between HCPs’ perceived stress and a change in patients’ symptom burden. Conclusion: Many patients were prepared well enough for hospital discharge. Still, the percentage of anxious and depressed patients increased after returning home. HCPs’ perceived stress in hospitals is related to patients’ increase in distress. It needs to be taken seriously to ensure both HCPs’ and patients’ well-being. © 2020 Alexander Wuensch. Hosting by Science Repository. All rights reserved.


Introduction
The period between a hospital discharge and the follow-up as an outpatient is a critical time for cancer patients [1]. They still experience multiple symptoms at the end of the hospital stay [2]. Upon arriving home, constant care is no longer available, and daily life may be altered [3]. Continuing treatment and care need to be organized [4]. This new reality can be highly distressing. Studies conducted in the United States of America and in Europe have reported deficiencies in the preparation for this period including inappropriate or absent physician-patientcommunication [3,5,6]. In Germany, hospitalized patients are supposed to receive support through statutory hospital discharge management [1]. This includes the assessment of needs and preparation of further care and symptom management [4,7,8].

Design
A pre-post design was applied. The study was conducted in four departments at a German university medical center between May 2018 and July 2018. First, HCPs, both physicians and nurses were informed and gave their informed consent. They provided their sociodemographics. Next, suitable patients (discharge in two days, cancer diagnosis, knowledge of their diagnosis, an imminent hospital discharge home after an inpatient stay, sufficient knowledge of the German language and a Karnofsky index equal to or greater than 60) were informed about the study. After providing informed consent they answered a questionnaire on their socio-demographics and level of distress i.e. (anxiety, depression and symptom burden). A follow-up questionnaire in a prepaid reply envelope to be answered two days after their discharge was handed out to each participant. In the meantime, HCPs rated their current stress level at work on a visual analogue scale (VAS). The follow-up questionnaire for patients asked to evaluate the HCPs' communication that was part of the discharge procedure and assessed the patients' distress ( Figure 1). Full approval was given by the ethics committee of the Medical Center at the University of Freiburg (approval number 52/18, March 22, 2018). The study was registered under http://apps.who.int/trialsearch/Trial2.aspx?TrialID=DRKS00014055.

Sample
Ninety-six cancer patients were invited to take part in the study. A total of 62 cancer patients agreed to take part. Nine patients were excluded from the analysis due to vastly incomplete data. Patients mostly had malignant neoplasms of skin, eight out of 53 patients had metastases, and most of the patients were hospitalized for surgery. Detailed sociodemographics, diagnoses and treatments are reported in (Table 1). With regard to HCPs, 38 took part, including 21 physicians and 17 nurses or nursing students.

I Assessment
We assessed anxiety and depression through the Hospital Anxiety and Depression Scale (HADS), with modified time reference (two days instead of 14 days) [16,17]. Symptom burden was assessed using the Memory Symptom Assessment Scale (MSAS), with modified time reference (two instead of seven days) [15,18]. The item 'difficulties with sexuality' was discarded. The index TMSAS (Total MSAS) was used.
Communication between HCPs and patients during discharge management was evaluated using the Communication Assessment Tool (CAT), [19,20]. The modification pertained to communication during hospital discharge. HCPs' perceived stress level before discharge was assessed using a 10 cm visual analogue scale (VAS) referring to the current day [21]. Patients' socio-demographics and HCPs' sociodemographics were assessed using questionnaires.

II Statistical Methods
Statistical analyses were carried out using SPSS 25.0. It was assumed that missing data were distributed completely at random, except for missing HCPs' stress data. Data missing completely at random was imputed using multiple imputations. Patients' datasets were excluded if follow-up questionnaires were missing completely. The patients' distress before and after discharge were analyzed descriptively. The average evaluation of physicians' communication and nurses' communication was calculated. Perceived stress was calculated as the mean of physicians' and nurses' stress on particular wards. Communication and perceived stress were analyzed descriptively.
Several linear regressions were employed to analyze associations between changes in patients' distress, HCPs' communication and HCP's perceived stress. In the first analysis (Analysis 1), communication was the criterion and perceived stress was the predictor. In the second (Analysis 2) and third analysis (Analysis 3), in which the change in patients' distress (anxiety, depression and symptom burden) was the focus, the measurements collected after the discharge from the hospital were the criteria and the measurements collected before the discharge from the hospital were the first predictors.
To analyze the association between change in patients' distress and HCPs' communication (Analysis 2) and the association between change in patients' distress and HCPs' perceived stress (Analysis 3), HCPs' communication and HCPs' perceived stress was included as the second predictor, respectively. Data were transformed (i.e., to inverses, square roots, logarithms, respectively) to meet the preconditions for linear regression and multiple imputation. For each analysis, preconditions for data were analyzed separately and data transformed to meet the conditions. Thus, used transformations differ between and within the analyses. For the linear regression analyses an a priori power analysis revealed that 55 patients were needed (moderate effect, power 80%, α level 5%). In auxiliary analyses, we carried out Mann Whitney U tests and exact Fisher tests to explore pre-discharge characteristics and sociodemographics that distinguished patients with post-discharge clinically or borderline anxiety or depression from other patients. Symptoms were analyzed descriptively.

I Patients' Distress Two Days before and Two Days after Hospital Discharge
Patients' distress was operationalized as anxiety, depression and symptom burden.

Analysis 2
The association between change in patients' distress and HCPs'
For depression, pre-discharge characteristics that distinguished between patients with post-discharge clinical or borderline depression and other patients were depression (above cut-off median = 36.21, below cut-off median = 24.86) and symptom burden (above cut-off median = 38.79, below cut-off median = 24.26).

V Symptoms Pre-and Post-discharge
Patients reported seven symptoms on average before (M = 6.77, SD = 5.01) and after discharge (M = 7.49, SD = 6.86). 30% reported lack of energy before and 51% after hospital discharge; 43% reported pain before and 55% after hospital discharge.

Discussion
In our sample, anxiety and depression were in a normal range. Two days before hospital discharge, approximately 10% of patients in each case showed clinically relevant or borderline anxiety and depression. Two days after discharge, however, approximately 20% of patients in each case showed clinically relevant or borderline anxiety and depression. Interestingly, reported symptom burden was low and patients rated HCPs' communication as generally good. Perceived stress of HCPs was at a medium level in our sample. While HCPs' perceived stress predicted individual patients' increase in symptom burden from pre to post, no other associations were found between (1) HCPs' perceived stress and HCPs' patient-rated communication, (2) HCP's patient-rated communication and patient anxiety, depression or symptom burden, as well as (3) HCPs' perceived stress and patient anxiety or depression. Patients with high anxiety post-discharge were distinguishable from other patients by their pre-discharge levels of anxiety, depression and symptom burden, also by not having children.

Implications
We found that hospital discharge works well for patients with a low distress level. They experience low levels of anxiety, depression and symptom burden before and after hospital discharge. Some patients experience clinically relevant anxiety or depression symptoms before discharge and these numbers increase after hospital discharge. These patients need more support. Risk patients for greater distress after discharge can be identified before discharge. These patients report more anxiety, depressive symptoms and a higher symptom burden before discharge. They are more likely to have no children, which might be an indicator for less social support. Increased support for these patients can be organized prior discharge of hospital. Thus, there is a need to assess the patients' distress not only in the beginning, but also toward the end of a hospital stay.
Furthermore, we advocate an emphasis on pain management to be included in the discharge process as patients frequently report pain and lack of energy after discharge. This study suggests that there is an association between HCPs' stress and patients' distress. Therefore, for healthcare providers and patients alike, it is a worthwhile endeavor to alleviate stress in the hospital, be it by tackling its structural, procedural or human causes.

Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. This work is the master thesis of the first author.