Received Date: April 08, 2017; Accepted Date: April 29, 2017; Published Date: May 04, 2017
Citation: Oechsle1 K, Ehlert J, Kodolitsch Y, Ullrich A, Bokemeyer C, et al. (2017) Self-assessment and Screening for Palliative Care Need in Patients with Chronic Heart Failure. J Palliat Care Med 7:304. doi:10.4172/2165-7386.1000304
Copyright: © 2017 Oechsle K, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Visit for more related articles at Journal of Palliative Care & Medicine
Objective: Patients with chronic heart failure (CHF) rarely receive palliative care today.
Methods: In a cross-sectional pilot study we evaluated feasibility of self-assessment and physician-directed screening for palliative care need in CHF patients. Patients answered German versions of the “Minimal Documentation system for patients in palliative care” (MIDOS), “Distress-Thermometer” (DT), and the “Patient Health Questionnaire” (PHQ4). A modified version of the “Five-Item Palliative Care Screening Tool” was used to screen for palliative care need by the treating cardiologists.
Results: 218 patients were asked to participate and 148 (67%) completed all questionnaires. Median symptom burden (MIDOS) was 7 out of 30 points (SD 4.868), overall wellbeing was two of 5 (SD 0.902), and median distress score was 6 of 10 (SD 2.318). Most frequent moderate or severe symptoms were dyspnea (52%), weakness (51%), and tiredness (49%). Significant distress (DT score>5) was indicated by 106 patients (72%). Exhaustion (79%), sleeping problems (75%), and problems with breathing (69%) were most frequent problems. Only 11% and 14% of patients presented with significant anxiety and depression, respectively (PHQ4-score>3).
Palliative care need was scored with mean 6.8 out of 12 points (SD 2.223). Using a cut-off value of >5, 113 patients (76%) would have needed palliative care. Objective parameters for palliative care need (symptom burden 43%, distress 68%, decision-making 67%) were indicated, but request for palliative care by the patients, their relatives (5%) or the cardiology team (5%) was rare. Symptom burden and palliative care screening showed significant, but not clinically relevant interaction.
Conclusion: This pilot study, feasibility of palliative care self-assessment and the modified screening instrument could be demonstrated in CHF patients. For comprehensive screening, both perspectives have to be included. First results indicate significant symptom burden, psychological distress and a high need for additional palliative care in CHF patients.
Palliative care addresses problems and needs of patients with advanced progressive or life-threatening diseases in general, but usually about 90% of patients receiving specialist palliative care suffer from advanced cancer [1,2]. However, several clinical studies indicate that patients with chronic heart failure exhibit comparable symptoms and problems, quality of life, recurrent hospitalizations and survival prognosis with advanced cancer patients [3-5]. These studies suggest even higher rates of severe dyspnea, psychological distress, and depression, as well as lower spiritual wellbeing [3,4]. Other studies report various unmet palliative care needs in patients with advanced heart failure [6,7]. Although some studies suggest beneficial effects of palliative care consultation [8-12], less than 10% of patients with heart failure receive palliative care today [13,14].
Various studies analyzed potential barriers responsible for this almost lacking integration of palliative care in patients with advanced cardiologic diseases [2,15-19]. Potential reasons may be that the estimation of prognosis and trajectory of disease is considered more difficult than in cancer patients [2,17,18]. In contrast to oncology clinics, there are usually no established interactions, cooperation or communication structures between cardiology and palliative care specialists [2,15,19]. A limited understanding of the other specialist discipline hampers a professional collaborative approach to the specific patients` needs [15-17]. In the context of cardiologic diseases, palliative care is usually considered to be restricted to end-of-life care and cardiologists, but also the patients themselves and their family caregivers usually have no or only little knowledge about the potential benefits of palliative care in chronic heart failure patients [15,16,20]. The most crucial reason seems to be the absence of criteria or trigger points defining the optimal time frame to initiate palliative care [2,16,17]. Until today, preliminary approaches using screening instruments for palliative care need in patients with chronic heart failure could not be established in daily practice [14,21-23].
Therefore, this pilot study aimed to evaluate feasibility of selfassessment and screening instruments well established in daily clinical palliative care practice in advanced cancer patients or palliative care patients in general that were partly modified for use in patients with chronic heart failure. In addition, the different measurements were compared to identify overlapping information to perspectively create a minimal set of instruments. Finally, we retrieved some first explorative results on symptom burden, psychological distress, and palliative care need in patients with chronic heart failure.
Study design and inclusion criteria
This pilot study evaluated measurements for potential palliative care need in patients with chronic heart failure during cardiologic inpatient treatment at the University Heart Center of Hamburg-Eppendorf, Germany. Due to the explorative character, we used a cross-sectional design including patients at any time during their stay on an inpatient ward of the University Heart Center and at any time during the total course of their heart failure disease. Therefore, inclusion criteria were presence of a cardiologically diagnosed chronic heart failure, treatment on an inpatient ward of the University Heart Center, age>18 years, sufficient cognitive function, and adequate knowledge of German language to answer the questionnaire (estimated by the treating cardiologist). All patients had to give written informed consent prior to study inclusion. The local ethics committee of the General Medical Council of Hamburg approved the study protocol (PV 5010).
This study aimed to evaluate the practicability of measurements in patients with chronic heart failure that were primarily developed for and validated in cancer patients or in palliative care patients in general (with cancer patients being the major group). We used tools for subjective self-assessment of symptoms and psychological problems [24-26] as well as a screening tool for palliative care needs assessed by professional caregivers  to include the patients’ and the professionals’ views. Long-term aim realized by a following larger study will be the development and validation of a short screening instrument indicating palliative care need in patients with chronic heart failure in daily clinical routine.
Further, this pilot study revealed some first explorative results on symptom burden, psychosocial distress and palliative care need in patients with chronic heart failure including the patients’ and the cardiologists’ perspective.
Symptom burden and psychological distress: The patients’ selfassessment questionnaires included German versions of three standardized and validated instruments that were regarded as standard instrument for specialized palliative care assessment in daily practice in Germany. Symptom burden was assessed by the MIDOS symptom questionnaire (Minimal Documentation system for patients in palliative care) . Psychological distress and potential causing factors were measured by the Distress-Thermometer of the National Comprehensive Cancer Network (NCCN) [25,28]. For screening for anxiety and depression the PHQ4-questionnaire (Patient Health Questionnaire) was used [26,29].
The MIDOS questionnaire , a German version of the Edmonton Symptom Assessment Scale (ESAS) , includes 10 symptoms (pain, nausea, emesis, dyspnea, constipation, weakness, loss of appetite, tiredness, sadness, and anxiety). Patients had two rate these 10 symptoms on a four-point scale from “0=no” up to “3=severe”. In addition, the patients had to score their subjective overall wellbeing on a five-point scale from “0=very good” to “4=very bad”.
The Distress Thermometer (DT) [25,28] assesses the psychological burden within the last 7 days on an 11-point-Likert-scale range from “0=no distress” up to “10=extreme distress”. Scores of >5 indicate clinically relevant psychological distress. In addition, 36 problems from five categories (practical, family, emotional, spiritual, and physical problems) potentially causing distress could be attributed with “yes” and “no”.
The 4-item PHQ-4 questionnaire (Patient Health Questionnaire) is a combination of the GAD-2 (General Anxiety Disorders Scale) and PHQ-2 questionnaires [26,29]. In the 2-item PHQ-2, a short form of the PHQ-9, two items assess the frequency of depressive symptoms within the past week. They are scored on a four-point Likert scale rated from “0=not at all” to “3=nearly every day” with scores >3 indicating depression [26,31]. In the two-item GAD-2, a short form of the GAD-7 , two items assess anxiety symptoms within the past week. Scales are scored on a four-point Likert scale rated from “0=not at all” to “3=nearly every day”. A score of >3 indicates anxiety [26,29]. The PHQ-4 total score ranges from 0-12 and scores of 0-2 can be interpreted as normal, 3-5 as mild, 6-8 as moderate and 9-12 as severe [26,29].
Screening for palliative care need: A modified version of the “Five- Item Palliative Care Screening Tool” developed by Glare et al. was used to screen for potential palliative care need by the treating cardiologists [27,33]. The “Five-Item Palliative Care Screening Tool” was developed to screen advanced cancer patients for their potential palliative care need. To apply this screening instrument in patients with chronic heart failure, we replaced “metastatic or locally advanced cancer” in screening items 1 and 3 with “chronic heart failure” and we deleted typically cancer-associated complications from item 3. In item 2, we replaced the “Functional status score according to the Eastern Cooperative Oncology Group” with “functional status according to the NYHA-classification” (New York Heart Association). In item 4, we deleted the potential co-morbidity “chronic heart failure” and added “advanced malignant disease” instead. Item 5, including all sub-items of palliative care problems, were adopted without modifications. The modified “Five-Item Palliative Care Screening Tool”, which will be now called “Palliative Care Screening Tool for heart failure patients”, is presented in Table 1.
|1||Presence of chronic heart failure||1|
|2||Functional status according to the NYHA-classification||1 – 4|
|3||Presence of one or more serious complication of chronic heart failure usually associated with a prognosis of < 12 months||1|
|4||Presence of one or more serious co-morbid disease also associated with poor prognosis (e.g. moderate-severe COPD, advanced malignant disease, dementia, AIDS, end-stage renal failure, end stage liver cirrhosis).||1|
|5. Presence of palliative care problems||Points|
|a||Symptoms uncontrolled despite of cardiologic treatment||1|
|b||Moderate to severe distress in patient and/or family related to chronic heart failure diagnosis or therapy||1|
|c||Patient and/or family concern about course of disease and decision making||1|
|d||Patient and/or family request palliative care consult||1|
|e||Cardiologic team needs assistance with complex decision making or determining goals of care||1|
|NYHA: New York Heart Association|
Table 1: “Palliative Care Screening Tool for heart failure patients” (modified from Glare et al. [27,33]).
Statistical analysis: Descriptive statistics including mean, median, standard deviations, and frequencies were determined. Internal consistency of the “Palliative Care Screening Tool for heart failure patients “was assessed by Cronbach’s alpha coefficient and was considered satisfactory when α>0.70 . We conducted a series of correlation analyses between variables reflecting symptom burden and psychological distress and the sum score of “Palliative Care Screening Tool for heart failure patients “Depending on the data’s level of measurement, bivariate associations were calculated by either Pearson`s or Spearman`s tests to assess the relative magnitudes of the associations between variables. Following Cohen , a correlation of 0.2 was considered to be low, a correlation of 0.5 to be medium strength, and a correlation of 0.8 to be strong.
Five candidate variables were studied in a multivariate linear regression analysis (age, gender, MIDOS sum score, distress thermometer score, and PHQ-4 score) with the sum score of “Palliative Care Screening Tool for heart failure patients” being the dependent variable. Testing of assumptions for linear regression was conducted before analysis. Multicollinearity assessment showed intercorrelations of ≥ 0.5 between the MIDOS sum score and distress thermometer, resulting in the exclusion of the latter. Within the regression model, we applied a backwards variable selection procedure, and Analysis of Variance (ANOVA) of regression was used to demonstrate the significance of the whole model.
All significance tests were two-tailed using a significance level of α<0.05. All analyses were performed using SPSS Statistics software version 24.0 (IBM, USA).
Between 1st April 2015 and 31st March 2016 (interrupted for 8 weeks due to severe illness of study personal), a total of 218 patients with chronic heart failure were admitted to the Heart Failure Unit of the University Heart Center and were asked for study participation. Finally, 148 patients (67%) were willing to participate. Patients completed all questionnaires thoroughly, with only little proportions of missing data (<5% per item). Median age of these 148 patients was 67.5 years (SD 14.3; range: 18-90). For patient characteristics including details on chronic heart disease and clinical presentation (Table 2).
|Subtype of chronic heart failure||Ischemic Cardiomopathy (CMP)||79||53|
|Acute cardiac decompensation||7||5|
|Comorbid conditions||Chronic renal insufficiency||57||39|
|Diabetes Mellitus Type II||50||34|
|Chronic heart failure-associated co-mordidities||Arterial hypertension||75||51|
|Peripheral arterial occlusive disease||50||34|
|Previous cardiac surgery||41||28|
|Implanted cardiac peacemaker||31||21|
|Left ventricular assist device||11||7|
|Previous reanimation therapy||7||5|
|High urgency status for transplantation||6||4|
|Decompensation of prothrombine time||5||4|
Table 2: Patient characteristics: N pts=148.
Symptom burden and distress
Patients rated their symptom burden measured by the MIDOS sum score with a median of 7 out of maximum 30 points (23%; mean 7.87; SD 4.86; range: 1-21). “Moderate” or “severe” intensity was scored in a median number of two out of 10 symptoms (SD (1.85; range: 0-7). The most frequent symptoms of “moderate” or “severe” intensity were dyspnea in 77 patients (52%), weakness in 75 (51%), and tiredness in 72 (49%).
Patients rated their overall wellbeing on the 5-point Likert scale with a median score of two (mean 1.99; SD 0.90; range: 0-4).
Median score of the Distress Thermometer was 6.0 (mean 5.83; SD 2.31; range: 0-10). Overall, 106 patients (72%) reported scores of >5 which indicates significant psychological distress in cancer patients . In the list of problems considered responsible for their psychological distress, patients indicated an average of 0.4 out of 5 practical problems (8%; SD 0.74; range: 0-3). Average number of family problems was 0.3 out of 2 (15%; SD 0.67; range: 0-2), emotional 2.0 out of 6 (30%; SD 2.22; range: 0-6), spiritual 0.15 out of 2 (8%; SD 0.47; range: 0-2), and 7.7 out of 19 physical problems (41%; SD 4.29; range: 0-19). Exhaustion (79%), sleeping problems (75%), problems with breathing (69%), and decreased mobility and activity (68%) were the most frequently complained problems. Table 3 presents the detailed results.
|N pts (%)||N pts (%)||N pts (%)||N pts (%)|
|Pain||79 (53%)||50 (34%)||14 (10%)||5 (3%)|
|Nausea||118 (80%)||26 (18%)||2 (1%)||2 (1%)|
|Emesis||134 (91%)||10 (7%)||3 (2%)||1 (1%)|
|Dyspnea||46 (31%)||25 (17%)||44 (30%)||33 (22%)|
|Constipation||89 (61%)||40 (27%)||12 (8%)||6 (4%)|
|Weakness||25 (17%)||48 (32%)||46 (31%)||29 (20%)|
|Loss of appetite||61 (41%)||50 (34%)||24 (16%)||13 (9%)|
|Tiredness||28 (19%)||48 (32%)||39 (27%)||33 (22%)|
|Sadness||109 (74%)||24 (16%)||12 (8%)||3 (2%)|
|Anxiety||94 (64%)||38 (26%)||13 (9%)||3 (2%)|
Table 3: Symptoms and problems (MIDOS ): N pts=148.
The total PHQ-4 score was low with 1.0 in median (mean 2.22; SD 2.48; range: 0-12). Patients presented without any evidence for anxiety/ depression in 61% as well as mild and moderate in 30% and 9%, respectively. None of these patients showed severe anxiety/depression in the PHQ-4 screening. Assuming a cut-off-value of >3 for both scales as giving evidence for anxiety or depression, only 11% and 14% of patients presented with anxiety and depression, respectively.
Palliative care screening
The mean score on the “Palliative Care Screening Tool for heart failure patients” indicated by the treating cardiologist was 6.8 out of a maximum of 12 possible points (median 7.00; SD 2.22; range: 0-12). Details are presented in Table 4.
|Problem present||N pts||%|
|Transportation / mobility||8||6|
|Employment / school||16||11|
|Relationship with partner||27||18|
|Relationship with children||19||13|
|Loss of interests on daily activities||42||29|
|Loss of trust||8||5|
|Spiritual aspects concerning god||15||10|
|Mobility / activity||100||68|
|Washing / clothing||66||45|
|Inflammation in the mouth||27||19|
|Problems with eating / diet||54||37|
|Problems with urinary||39||26|
|Dry or itching skin||60||41|
|Dry or stuffy nose||18||12|
|Tingling sensation in Fingers / toes||53||36|
|Cognitive problems / concentration||49||33|
|N pts: Number of Patients|
Table 4: Problem list of the Distress Thermometer : N pts=148.
Interestingly, the treating cardiologists recognized objective parameters for palliative care need, like symptom burden (43%), distress (68%) or decision-making (67%), frequently. In contrast, request for palliative care by the patient or relatives (5%) or the cardiology team (5%) was indicated very rarely.
Using the cut-off value of >5 of the original version of the “Five Item Palliative Care Screening Tool” for cancer patients by Glare et al. [14,20] as indicator for palliative care need, 113 patients (76%) would have needed additional palliative care (Table 5).
|Screening items||N pts||%|
|1||Presence of chronic heart failure||148||100|
|2||Functional statusNYHA 1||22||15|
|Functional statusNYHA 2||37||25|
|Functional statusNYHA 3||53||36|
|Functional statusNYHA 4||36||24|
|3||Presence of one or more serious complication of chronic heart failure usually associated with a prognosis of < 12 months||87||56|
|4||Presence of one or more serious co-morbid disease also associated with poor prognosis (e.g. moderate-severe COPD, advanced malignant disease, dementia, AIDS, end-stage renal failure, end stage liver cirrhosis).||120||81|
|5. Presence of palliative care problems|
|A||Symptoms uncontrolled despite of cardiologic treatment||63||43|
|b||Moderate to severe distress in patient and/or family related to chronic heart failure diagnosis or therapy||100||68|
|C||Patient and/or family concern about course of disease and decision making||99||67|
|D||Patient and/or family request palliative care consult||7||5|
|E||Cardiologic team needs assistance with complex decision making or determining goals of care||8||5|
|N pts: Number of Patients; NYHA: New York Heart Association|
Table 5: Results of the “Five Item Palliative Care Screening Tool for heart failure patients”.
The internal consistency of the “Palliative Care Screening Tool for heart failure patients” was Cronbach’s α=0.580, indicating multidimensionality of the screening tool.
Interaction between measurements
The “Palliative Care Screening score” significantly correlated with the symptom total score (MIDOS; r=0.235; p=0.004) and psychological distress (DT; r=0.189; p=0.022), but not with overall well-being (r=-0.118; p=0.155) or anxiety/depression (PHQ-4; r=0.117; p=0.157). It was also not correlated with age (r=0.027; p=0.749) or gender (r=0.104; p=0.206). In contrast, symptom burden (MIDOS) correlated significantly with all other parameters: overall well-being (r=0-.544; p<0.001), anxiety/depression (PHQ-4; r=0.383; p<0.001), distress (DT; r=0.555; p<0.001), age (r=0.251; p=0.002), and gender (r=0.255; p=0.002). Within the regression model, the symptom burden (MIDOS) was the only variable with significant effects on the “Palliative Care Screening score” based on 95% CI (unstandardized regression coefficient B=0.107, standard error=0.037, Beta=0.235, t(2,915) p=0.004, 95% CI=[0.035, 0.180]). Age, gender, and anxiety/depression (PHQ-4) were eliminated from the original regression model. R squared and adjusted R squared of the final regression results were 0.055 and 0.049, respectively. The selected model with symptom burden as single explanatory variable was statistically significant (F(1,146)=8.500, p=0.004), but not clinically relevant.
This cross-sectional study evaluated the feasibility of a set of selfassessment instruments well established in advanced cancer patients [24-26] as well as a modified screening tool for palliative care needs  in patients with chronic heart failure. Self-assessment of symptom burden and psychological distress as relevant indicators of palliative care need was applicable using the MIDOS and PHQ-4 questionnaires plus the distress thermometer in patients with advanced heart failure. Overall, 67% of patients who were invited to participate in this questionnaire analysis could be included. Questionnaires were completed in full by almost all patients. The corresponding cardiologists declared that the questionnaires appeared to them to be applicable in patients with chronic heart failure and they completed the “Palliative Care Screening Tool for heart failure patients” for all patients.
We correlated the different measurements to identify overlapping information aiming to define a minimal set of instruments revealing the highest gain of information with lowest effort for the patients and their treating cardiologist. Within the regression analysis, age, gender, and PHQ-4 score were eliminated as they were not of additional information. The final regression model demonstrated a statistically significant, but clinically not sufficient variance explanation of 5.5% for the “Palliative Care Screening Tool” with the MIDOS score being the only predictor variable. Therefore, a combination of self-assessment using the MIDOS questionnaire and external screening using the modified “Palliative Care Screening Tool” seems to be the most practicable set of measurements that will be prospectively validated in an ongoing study. It also demonstrates that in cardiologic patients both perspectives - of the patients themselves and their professional caregivers - have to be taken into account. A Canadian study using the original ESAS symptom questionnaire and the externally assessed Palliative Care Performance Scale  also suggests a combination of symptom burden self-assessment (ESAS/MIDOS) plus a palliative care screening tool as most suitable daily clinical practice.
Secondly, this pilot study aimed to reveal some first explorative results on symptom burden, psychological distress and potential palliative care need in patients with chronic heart failure. Self-assessed symptom burden was moderate with median scores of 7 out of 30 points (23%). Patients classified about 20% of symptoms as of “moderate” or “severe” intensity. The most frequent symptoms of moderate or severe intensity were dyspnea, weakness, and tiredness. Previous studies using palliative care measurements in patients with chronic heart failure have described fatigue and anorexia/loss of appetite as the two most frequent distressing symptoms [4,36]. Dyspnea was less frequent in previous studies [4,36], but this might be explained by the fact that in our study, patients were included in a timely manner after admission to inpatient cardiologic care due to decompensation of their heart failure. Similarly, one of these studies described impaired well-being in only 20% of patients with severe heart failure (NYHA III/IV) , while in our study the patients` overall wellbeing was moderate in general.
Psychological distress was high with a median score of 6. Assuming a cut-off-value of >5 indicating significant psychological distress in cancer patients , 72% of patients with chronic heart failure presented with significant distress in this study. Patients considered physical and emotional problems most frequently to be responsible for their psychological distress. Exhaustion, sleeping problems, problems with breathing, and increased mobility and activity were the most distressing problems. In contrast, only 11% and 14% of patients presented with indicators for anxiety or depression, respectively.
Palliative care need assessed by external screening using the “Palliative Care Screening Tool for heart failure patients” was scored with 6.8 out of a maximum of 12 points. Using the cut-off value of >5 of the original version of the “Five Item Palliative Care Screening Tool” by Glare et al. validated in cancer patients as indicator for palliative care need [27,33], almost 80% of patients would have needed palliative care. Interestingly, the treating cardiologists recognized objective parameters for palliative care needs, like symptom burden, distress or decision-making quite frequently. In contrast, the request for palliative care by the patient, relatives or the cardiology team was indicated rarely (5% each). This is in correspondence to previous studies demonstrating that cardiologists and their patients restrict palliative care to end-of-life care and have no specific ideas about integrating palliative care in earlier phases of the disease trajectories [15,16,20].
This pilot study was designed as a feasibility and explorative study, which exhibits methodical limitations. The explorative data on symptom burden, distress, and palliative care need can only be interpreted as hypotheses generating, especially as the final validation of the “Palliative Care Screening Tool for heart failure patients” is pending. Prospective studies with larger patient cohorts have to evaluate these parameters and further randomized trials have to examine beneficial effects of palliative care in correspondence to their indicated need. Further, it has to be noticed that only few sociodemographic or disease-related factors were assessed in this pilot study. Therefore, the explanatory power and validity of regression analysis is limited and following studies evaluating this set of measurements have to respect further potentially influencing factors.
In conclusion, screening for palliative care need using selfassessment and the modified external screening instrument seems to be feasible in patients with chronic heart failure. Both perspectives, of the patients and their professional caregivers, have to be respected to reveal a comprehensive view the patients` palliative care need. The minimal set of self-assessment using the MIDOS questionnaire plus the modified Palliative Care Screening Tool will be validated in a further study including potential influencing factors. First results indicate significant symptom burden, psychological distress, and palliative care need in patients with chronic heart failure, which will be prospectively evaluated in a larger cohort of patients.
We thank all patients and cardiologists for their effort in participating in the present study.