If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
There is scarce information about sedation in nursing homes at the end of life. We aimed to assess (1) the use of sedatives generally and “sedatives with continuous effect,” based on objective operational criteria, within the last week of life in nursing homes and (2) factors associated with this treatment.
Design
Retrospective cohort study, using the nursing homes' medical records.
Setting and Participants
Residents who died in 4 German nursing homes from January 2015 to December 2017 and whose medical records were available (n = 512).
Methods
Sedatives analyzed were those recommended by guidelines for “palliative sedation”: benzodiazepines, levomepromazine, haloperidol (≥5 mg/d), and propofol. The definition of “sedatives with continuous effect” and doses judged as at least moderately sedating were consented by palliative care clinicians and pharmacists, based on the literature. Descriptive statistics and multivariate logistic regression analysis were performed (R version 3.6.1).
Results
Overall, 110/512 (21%) deceased residents received a sedative at least once during the last week of life, 46/512 (9%) “sedatives with continuous effect.” Oral lorazepam was used most frequently. Eleven of 512 (2%) residents received doses judged as at least moderately sedating. The term sedation was not used. Most frequent indications were agitation (58/110; 53%) and anxiety (35/110; 32%); no indication was noted for 36/110 (33%) residents. The resident's involvement in the decision for sedatives was documented in 3/110 (3%). Multivariate logistic regression analysis showed significant associations between use of sedatives and age (OR = 0.94, P < .001) as well as institution (P < .001).
Conclusions and Implications
Our data indicate a lower prevalence of sedation compared to international data and considerable differences regarding prevalence between institutions. These differences, potential setting-specific challenges, and need for support measures for consistent best practice of sedation in nursing homes should be further explored.
Board of the European Association for Palliative Care. European Association for Palliative Care (EAPC) recommended framework for the use of sedation in palliative care.
“Sedation in PC” or “palliative sedation” has been defined as “monitored use of medications intended to induce a state of decreased or absent awareness (unconsciousness) in order to relieve the burden of otherwise intractable suffering.”
Board of the European Association for Palliative Care. European Association for Palliative Care (EAPC) recommended framework for the use of sedation in palliative care.
Using continuous sedation until death for cancer patients: A qualitative interview study of physicians' and nurses' practice in three European countries.
Most research on sedation at the end of life focuses on specialist palliative care (SPC). However, many people die in nursing homes, and this number is predicted to increase, based on the demographic developments.
It has been shown that this specific type of sedation was used in 6% to 9% of Belgian nursing home residents and in 14% of dying persons treated by Dutch older adult care physicians.
Using continuous sedation until death for cancer patients: A qualitative interview study of physicians' and nurses' practice in three European countries.
Given the lack of research on the whole range of sedation practices (ie, practices other than continuous deep sedation) in nursing homes internationally and a scarcity of data from Germany, this study has the following aims
: (1) to assess the use of sedatives generally and of “sedatives with continuous effect” (based on objective operational criteria) within the last week of life in nursing homes and (2) to assess factors associated with this treatment.
Methods
Design, Setting, and Participants
We conducted a multicenter retrospective cohort study of residents who died in 4 German nursing homes between January 2015 and December 2017 (as part of a mixed-methods study). Nursing homes differing regarding number of residents, affiliation (municipal, Protestant, and Catholic), and location (urban and suburban) were selected by respective contact persons for municipal, Protestant, and Catholic nursing homes and asked for participation. The 4 participating nursing homes provide between 102 and 216 beds each, have Protestant (n = 2), Catholic (n = 1), and municipal (n = 1) affiliation, and are located in a large city (n = 3) and a suburb (n = 1). The study was approved by the relevant Research Ethics Committee.
Data Collection
Methods for data collection have been described for a previous single-center study.
Data were retrieved from the nursing homes' medical records, which contained drug sheets and nurses' notes. General practitioners and specialist PC teams kept separate records, which were not accessible for us. The data extraction tool was developed based on the literature, piloted using 15 residents' records and successfully used in the previous single-center study.
Palliative sedation in a specialized unit for acute palliative care in a cancer hospital: Comparing patients dying with and without palliative sedation.
Two researchers jointly extracted data for randomly selected 20% of all records, in accordance with guidelines for data collection from medical records.
As in the single-center study, we analyzed drugs recommended in guidelines for “palliative sedation”: benzodiazepines, levomepromazine, haloperidol ≥5 mg/d (as lower doses are unlikely to be sedating) and propofol.
Board of the European Association for Palliative Care. European Association for Palliative Care (EAPC) recommended framework for the use of sedation in palliative care.
For readability purposes, they are called “sedatives” in this article. We collected details on the use of sedatives including doses per day, indication, routes of administration and labeling of the treatment in the medical records. Additionally extracted demographic and clinical data included age, gender, length of stay, diagnoses, need for care, support by SPC services, artificial nutrition and hydration, prescription of opioids, and use of the words “palliative”, “sedation”, and “palliative sedation” in the medical records.
Analysis
The definition of “sedatives with continuous effect” of the single-center study was used: either continuous parenteral infusion for ≥0.5 hours or repeated application expected to result in sedation for ≥24 hours (see Table 1).
In addition, doses judged as probably at least moderately sedating in this older adult population were consented by SPC pharmacists and clinicians, based on the drugs' prescribing information and other available literature (see Table 1).
Board of the European Association for Palliative Care. European Association for Palliative Care (EAPC) recommended framework for the use of sedation in palliative care.
Consented between specialist palliative care (SPC) clinicians and pharmacists, based on the available data regarding the half-life and duration of action of the drugs in weak and/or older adult patients, as stated in the drugs' prescribing information as well as a widely used textbook for drug therapy in palliative care.34
Dose Judged as at Least Moderately Sedating in These Dying Patients (Oral Dose Equivalents, Except for Midazolam),
Consented between SPC clinicians and pharmacists, based on the drugs' prescribing information and other available literature.3,30,32,34,35 For the drugs that are licensed for anxiety and agitation, we chose the highest licensed dose for older adult and weak patients. For the drugs licensed for sleep disorders, we made a clinical-pharmaceutical judgement which total daily dose would probably result in at least moderate sedation over 24 hours, based on the doses licensed for sleep disorders. For midazolam and levomepromazine, the judgment was based on the licensed doses for sedation in anesthesia or acute agitation, respectively, as well as the lowest doses recommended or recorded for “sedation in palliative care.”3,32 We aimed for conservative judgements, in order to rather underestimate than overestimate the number of residents with moderately sedating doses. For comparison, in 2 previous studies, cutoff doses of midazolam 10 mg and levomepromazine 25 mg per 24 hours were used for defining a sedating dose.30,35
mg
Clonazepam
1×
Not judged, as no information available regarding sedating effect for certain doses.
Diazepam
1×
5
Flunitrazepam
2×
2
Lorazepam
2×
4
Midazolam
7×
24
Oxazepam
2×
30
Lormetazepam
3×
3
Haloperidol >5 mg/d
1×
Not judged because of large variability in individual sedating effect
Levomepromazine
1×
30
Propofol
—
Continuous administration judged as always used for at least moderate sedation
∗ Consented between specialist palliative care (SPC) clinicians and pharmacists, based on the available data regarding the half-life and duration of action of the drugs in weak and/or older adult patients, as stated in the drugs' prescribing information as well as a widely used textbook for drug therapy in palliative care.
Board of the European Association for Palliative Care. European Association for Palliative Care (EAPC) recommended framework for the use of sedation in palliative care.
For the drugs that are licensed for anxiety and agitation, we chose the highest licensed dose for older adult and weak patients. For the drugs licensed for sleep disorders, we made a clinical-pharmaceutical judgement which total daily dose would probably result in at least moderate sedation over 24 hours, based on the doses licensed for sleep disorders. For midazolam and levomepromazine, the judgment was based on the licensed doses for sedation in anesthesia or acute agitation, respectively, as well as the lowest doses recommended or recorded for “sedation in palliative care.”
Board of the European Association for Palliative Care. European Association for Palliative Care (EAPC) recommended framework for the use of sedation in palliative care.
We aimed for conservative judgements, in order to rather underestimate than overestimate the number of residents with moderately sedating doses. For comparison, in 2 previous studies, cutoff doses of midazolam 10 mg and levomepromazine 25 mg per 24 hours were used for defining a sedating dose.
Opioids were not analyzed as “sedatives” for this article. However, as they may have additional sedating effects, we analyzed median opioid doses in our sample. We used standard equivalence factors to convert opioid doses to oral morphine equivalents (in mg) as follows: fentanyl transdermal (in mg/h) × 100 × 24 h, buprenorphine transdermal (in mg/h) × 75 × 24 h, hydromorphone oral (in mg) × 5, levomethadone oral (in mg) × 16, oxycodone oral (in mg) × 1.5, piritramide s.c. (in mg) × 0.7)/0.3, tapentadol oral (in mg) × 0.4, tilidine oral (in mg) × 0.1, Tramadol oral (in mg) × 0.1.
Leitlinienprogramm Onkologie (Deutsche Krebsgesellschaft, Deutsche Krebshilfe, AWMF) [Oncology programme guideline (German Cancer Society GCA, AWMF)]. S3 Leitlinie Palliativmedizin für Patienten mit einer nicht-heilbaren Krebserkrankung, Lang-version 2.0, AWMF-Registernummer: 128/001OL [S3 guideline for palliative care for patients with incurable cancer]. 2019.
We conducted descriptive statistics, bivariate analysis, and multivariate logistic regression analysis using R, version 3.6.1. Prevalence of use of sedatives, indications, routes of administration, and doses were analyzed as well as the labeling of the treatment in the medical records. For determining medians, interquartile ranges, and ranges of drug doses, values of 0 were excluded. We analyzed differences in prevalence of the use of sedatives between the different institutions. In addition, differences in sociodemographic and clinical characteristics between “use of sedatives” and “no use of sedatives” as well as between “use of sedatives with continuous effect” and “no use of sedatives with continuous effect” were analyzed. Relative frequencies are reported in valid percentages. We used the chi-square test or Fisher exact test (if frequencies <5) for categorical and the Mann-Whitney U test for continuous data.
We conducted multivariate logistic regression analysis to predict the use of sedatives. Variables were entered into the model as predictors when statistically significant group differences were detected in bivariate analyses. An exception was “care by SPC team,” which was not entered into the model because the number of cases was small. Additionally, gender was entered as a predictor. We ran no logistic regression model to predict “use of sedatives with continuous effect” because (1) relatively few patients received “sedatives with continuous effect”, and thus the possible number of predictors was reduced, and (2) cross tabulations with predictors showed very small cell sizes, which would have limited stability of estimations.
Alpha level was set at 0.05. Because of the study's exploratory nature, we did not adjust for multiple testing.
Results
From January 2015 to December 2017, 555 residents died in the 4 nursing homes. Medical records of 43 residents were missing for unknown reasons. Therefore, the data of 512 residents could be analyzed. Their median age was 89 years (range 55-107), and 70% were female. The majority had multiple diagnoses, the most frequent diagnosis was dementia (273/512, 54%). Sixteen of the 512 residents (3%) received support by an SPC team (see Table 2 for sociodemographic and clinical variables). The nursing homes provided additional information that 21% (range 15%-23%) of all deceased residents had died in hospital in 2017.
Table 2Comparison of Sociodemographic and Clinical Characteristics of Residents With and Without Use of Sedatives and of Residents With and Without Use of “Sedatives With Continuous Effect” Within the Last 7 Days of Life
“Pflegestufe”: until December 2015 official grading of the need for care (in activities of daily living) used by the insurance companies (1 = medium, 2 = high, 3 = very high need of care); “Pflegegrad”: from January 2016 official grading of the need for care (in activities of daily living) used by the insurance companies (1 = low, 2 = medium, 3 = high, 4 = very high need of care, 5 = very high need of care with special requirements for nursing care). Summarization of the 2 grading systems into: no or little need for care (no “Pflegestufe”/“Pflegegrad” or “Pflegegrad 1”), medium to high need for care (“Pflegestufe 1-2” or “Pflegegrad 2-3”), and very high need for care (“Pflegestufe 3” or “Pflegegrad 4-5”).
Fisher exact test was used to assess group differences (expected frequencies < 5).
Yes [parenteral/enteral]
10 [1/9] (2)
4 [1/3] (4)
6 [0/6] (2)
2 [0/2] (4)
8 [1/7] (2)
No
501 (98)
105 (96)
396 (99)
44 (96)
457 (98)
Missing, n
1
1
0
0
1
“Palliative situation” or “palliative treatment” documented,
Within the last 7 days of life.
n (%)
.22
.13
Yes
146 (29)
37 (34)
109 (27)
18 (39)
128 (28)
No
366 (71)
73 (66)
293 (73)
28 (61)
338 (73)
The figures are column numbers and percentages. Percentages are reported in “valid percent,” that is, based on the number of residents for whom data for the respective variable were available. Figures in bold denote statistically significant differences between residents with and without use of sedatives or between residents with and without use of “sedatives with continuous effect.”
∗ Test for difference judged as not clinically important.
† Including intracranial hemorrhage and stroke, excluding dementia.
‡ Not recorded were arterial hypertension and atrial fibrillation.
§ “Pflegestufe”: until December 2015 official grading of the need for care (in activities of daily living) used by the insurance companies (1 = medium, 2 = high, 3 = very high need of care); “Pflegegrad”: from January 2016 official grading of the need for care (in activities of daily living) used by the insurance companies (1 = low, 2 = medium, 3 = high, 4 = very high need of care, 5 = very high need of care with special requirements for nursing care). Summarization of the 2 grading systems into: no or little need for care (no “Pflegestufe”/“Pflegegrad” or “Pflegegrad 1”), medium to high need for care (“Pflegestufe 1-2” or “Pflegegrad 2-3”), and very high need for care (“Pflegestufe 3” or “Pflegegrad 4-5”).
‖ Fisher exact test was used to assess group differences (expected frequencies < 5).
Overall, 110 of the 512 residents (21%) received a sedative at least once during the last week of life. Within the 3 years, lorazepam was used for 98/512 residents (19%); lormetazepam for 7/512 (1%); oxazepam and haloperidol for 4/512 (0.8%) each; and diazepam, midazolam, and levomepromazine for 1/512 (0.2%) each (see Supplementary Table 1). Six of the 512 residents (1%) received 2 different sedatives, in combination or subsequently. The terms “sedation” or “palliative sedation” were used in none of the analyzed medical records.
Most frequently noted indications for sedatives were agitation (53%) and anxiety (32%). In single cases, sedatives were given because of pain or groaning. For 33% of residents receiving sedatives, no indication was recorded (see Table 3). Involvement of the patient or family members in the decision about sedatives was documented for 3/110 (3%) and 5/110 (5%) residents, respectively.
Table 3Documented Indications for the Use of Sedatives
Indication (Multiple Indications Possible; Documented in Any Part of the Residents' Medical Records, Including the Daily Nursing Records)
For Use of Sedatives, n (%) (n = 110)
For Use of “Sedatives With Continuous Effect,” n (%) (n = 46)
Calling out, n = 3; groaning, n = 3; spasticity, n = 2; cough, n = 1; lamenting, n = 1; weeping, n = 1; aggressiveness, n = 1; not feeling legs any more, n = 1.
Calling out, n = 3; groaning, n = 1; spasticity, n = 2; cough, n = 1; weeping, n = 1; aggressiveness, n = 1.
No indication reported
36 (33)
14 (30)
∗ Calling out, n = 3; groaning, n = 3; spasticity, n = 2; cough, n = 1; lamenting, n = 1; weeping, n = 1; aggressiveness, n = 1; not feeling legs any more, n = 1.
† Calling out, n = 3; groaning, n = 1; spasticity, n = 2; cough, n = 1; weeping, n = 1; aggressiveness, n = 1.
Lorazepam, lormetazepam, oxazepam, and levomepromazine were administered orally; midazolam subcutaneously; diazepam rectally; and haloperidol orally, intravenously, or subcutaneously. The median total daily dose of lorazepam within the last week of life was 1 mg [interquartile range (IQR) 1-2, range 0.5-6]. Eleven of the 110 residents receiving sedatives (10%, corresponding to 2% of all deceased residents) received total daily doses judged as at least moderately sedating (see Supplementary Table 1). Symptoms and consciousness level were not systematically recorded. Sixty-four of the 110 residents receiving sedatives (58%) were also prescribed opioids in the last week of life. The median total daily oral morphine equivalent was 28.8 mg/d (IQR 16.7-60.0, range 0.6-460.0).
The proportion of residents receiving sedatives at least once during their last week of life differed significantly between the nursing homes, ranging from 14% to 36% (P < .001). Residents who received sedatives and those who did not differed significantly regarding their age (P < .001), the diagnosis of dementia (P = .006) and support from an SPC team (P = .011). In the group receiving sedatives, residents were younger, dementia was less prevalent, and support from an SPC team was more frequent (Table 2). Multivariate logistic regression analysis showed a significant association between use of sedatives and age (OR = 0.94, P < .001) as well as institution (nursing home B: OR = 0.26, P < .001; nursing home C: OR = 0.28, P < .001) (Table 4).
Table 4Factors Associated With Use of Sedatives Estimated From a Multivariate Logistic Regression Model
OR (95% CI)
P Value
Age
0.94 (0.92-0.97)
<.001
Sex (female)
1.56 (0.94-2.66)
.09
Dementia (no)
1.56 (0.99-2.48)
.06
Institution
Nursing home A
Ref (Ref)
Nursing home B
0.26 (0.14-0.47)
<.001
Nursing home C
0.28 (0.15-0.54)
<.001
Nursing home D
0.52 (0.24-1.06)
.08
Ref, reference.
Although the frequency of “Care by specialist palliative care team” differed significantly between use of sedatives and no use of sedatives, it was not entered into the model because of the small number of cases. Figures in bold denote statistically significant associations.
Forty-six of the 110 residents who were prescribed sedatives (42%, corresponding to 9% of the total sample) received these “with continuous effect” on at least 1 day within the last week of life, mostly lorazepam (n = 41) (see Supplementary Table 1). Median number of days of this treatment was 3 (range 1-7). Twenty-five of the 46 residents (54%) received the “sedatives with continuous effect” until death. The number of residents receiving this treatment increased from 20 (6 days before death) to 25 on the day of death (see Supplementary Figure 1).
The most frequently noted indications for use of “sedatives with continuous effect” were agitation (54%) and anxiety (50%). For 14 of the 46 residents receiving “sedatives with continuous effect” (30%), no indication was documented in the records (see Table 3). The median total daily dose of lorazepam “with continuous effect” within the last 7 days of life was 2 mg (IQR 2-3.5, range 1-6). It varied between 3 mg (IQR 2-4, range 1-5) 6 days before death and 2 mg (IQR 2-3, range 1-6) on the day of death (see Figure 1). Nine residents (2% of all deceased residents) received “sedatives with continuous effect” in doses judged to be at least moderately sedating (see Supplementary Table 1). None of these 9 residents received support by an SPC team.
Fig. 1Total daily dose of lorazepam “with continuous effect” within the last 7 days of life. Bottom of the box: first quartile; top of the box: third quartile; band inside the box: median; “whiskers” with maximum 1.5 × interquartile range. •, outliers: lying beyond 1.5 × interquartile range.
Residents receiving “sedatives with continuous effect” and those not receiving this treatment differed significantly regarding their age (P < .001) and the diagnosis of chronic kidney disease (P = .015). In the group receiving this treatment, residents were younger and chronic kidney disease was more prevalent (Table 2).
Discussion
Main Results
These are the first multicenter data on different sedation practices at the end of life in nursing homes, based on previously published objective operational criteria.
Within the last week of life, about a fifth of all residents received a sedative, about half of these “with continuous effect.” Median doses were low, but 2% of residents received “sedatives with continuous effect” in doses judged as at least moderately sedating. The terms “sedation” or “palliative sedation” were never identified in the nursing homes' records. For a third of residents receiving sedatives, no indication was noted, and involvement of the resident and/or family members in the decision was only recorded for 3% and 5%, respectively. The prevalence of use of sedatives differed significantly between the nursing homes, and multivariate analysis showed a significant association between use of sedatives and institution.
Comparison With Data From SPC
Two main differences between the use of sedatives in these nursing homes and SPC settings can be noted. While in the nursing homes, 21% of the dying residents received sedatives and 9% “sedatives with continuous effect,” 69% to 86% of patients dying in British SPC settings received sedatives, and 78% of patients dying on a German PC unit received “sedatives with continuous effect.”
To explore the relationship between the use of midazolam and cessation of oral intake in the terminal phase of hospice inpatients: A retrospective case note review: Does midazolam affect oral intake in the dying?.
This lower prevalence of sedative use at the end of life in nursing homes may be explained by the more complex needs of patients in SPC settings compared to nursing home residents. However, other factors such as the need to comply with specific legal regulations for use of sedatives in German nursing homes and associated uncertainties of staff may also contribute to the lower frequency. Less overall experience regarding the use of sedatives among general practitioners and nursing home staff compared to SPC professionals might be another contributing factor. The second difference relates to drugs and routes of administration. While parenteral midazolam was used most frequently on the German PC unit, oral lorazepam was the drug of choice in nursing homes.
Similarities in both settings were the low median doses, probably mainly used for anxiolysis and mild sedation, and the increasing number of persons receiving “sedatives with continuous effect” towards death.
To explore the relationship between the use of midazolam and cessation of oral intake in the terminal phase of hospice inpatients: A retrospective case note review: Does midazolam affect oral intake in the dying?.
Palliative care specialists in hospice and hospital/community teams predominantly use low doses of sedative medication at the end of life for patient comfort rather than sedation: Findings from focus groups and patient records for I-CAN-CARE.
Comparison With International Data on Continuous Deep Sedation in Nursing Homes
In our sample, 2% of residents received “sedatives with continuous effect” in doses judged as at least moderately sedating. The terms “sedation” or “palliative sedation” were never used in the analyzed records, that is, neither for the sedatives analyzed in this study nor for any other prescribed medication. This stands in contrast to Belgian and Dutch data that demonstrated the use of continuous deep sedation in 6% to 14% of nursing home residents or persons treated by older adult care physicians.
The comparison of these data to our data is problematic because the cited studies are based on clinicians' self-labeling of their practice as “continuous deep sedation.” Therefore, the practice analyzed in these studies covers different drugs, doses, and modes of administration.
In contrast, we extracted data on specific sedatives, which are recommended in guidelines for sedation, from the nursing homes' records and applied objective criteria for “sedatives with continuous effect” and “at least moderately sedating effect.” Acknowledging these different approaches, the main difference to the cited studies is the lack of documentation of “sedation”, “palliative sedation”, or equivalent terms in the medical records of our sample. In addition, our data seem to indicate a lower prevalence of moderate or deep continuous sedation, as defined by the criteria mentioned above. Reasons for this lower prevalence in our sample might include transfers of residents with more complex symptoms to hospitals, different perceptions regarding the need for sedation, and uncertainties regarding its use. However, our definition of “sedatives,” based on the guidelines' recommendations, may not cover the whole spectrum of drugs actually used in practice for the purpose of sedation. In particular, it has been demonstrated that opioids were used as a sole drug for sedation in Belgian nursing homes.
Opioids were not analyzed as “sedatives” in the present study, neither were antipsychotics other than haloperidol and levomepromazine. Our analysis may therefore underestimate the prevalence of—not accordingly labeled—moderate or deep sedation in our sample.
Factors Associated With the Use of Sedatives
Our data demonstrate a significant difference in the prevalence of use of sedatives between different nursing homes, ranging between 14% and 36%. Multivariate logistic regression analysis showed a significant association between institution and use of sedatives, controlled for age, gender, and diagnosis of dementia. To our knowledge, this is the first study demonstrating such differences between individual nursing homes. It has been shown that differences in sedation practice generally may be associated with differences in specialties, training in PC as well as religious beliefs of the responsible physicians.
Bivariate analyses as well as multivariate logistic regression analysis found a significant association between use of sedatives and age. This corresponds to findings for continuous deep sedation from SPC settings as well as studies covering all health care settings, which also showed higher sedation rates in younger patients.
Patient-related determinants of the administration of continuous palliative sedation in hospices and palliative care units: A prospective, multicenter, observational study.
Although the observed difference in age between residents receiving sedatives (median 87.5 years) and those not receiving sedatives (median 90 years) may not be of clinical importance, we regard the respective difference between residents receiving “sedatives with continuous effect” (median 83 years) and those not receiving this treatment (median 90 years) as potentially clinically relevant. To our knowledge, reasons for the association of sedation rates with age are unclear and should be further investigated. They might include more aggressive disease trajectories in younger patients and a higher rate of spontaneously reduced consciousness at the end of life with increasing age, resulting in less apparent need for sedation.
Besides, a higher rate of severe dementia with increasing age might make the assessment of suffering and consequently the identification of the need for sedation more difficult.
According to bivariate analysis, care by an SPC team was more frequent among residents receiving sedatives. However, the 9 residents who were prescribed “sedatives with continuous effect” in doses judged as at least moderately sedating were not seen by an SPC team. This is not in line with current guidelines for sedation in PC, which recommend involvement of PC experts.
Board of the European Association for Palliative Care. European Association for Palliative Care (EAPC) recommended framework for the use of sedation in palliative care.
Some other deviations from best practice standards were identified in this study. First, the involvement of the residents and family members in the decision regarding sedatives was mostly not documented in the analyzed records. Second, in about a third of cases, documentation of the indication for sedative use was lacking. Third, single indications found in the medical records were not in line with current guidance on sedation in PC. For example, pain without documentation of refractoriness does not justify use of sedatives.
Board of the European Association for Palliative Care. European Association for Palliative Care (EAPC) recommended framework for the use of sedation in palliative care.
Finally, treatment that probably resulted in at least moderate sedation was not recorded as “sedation” in the records. If this is not only a problem of documentation but of not perceiving the treatment as sedation, this may be an important reason for nonadherence to relevant guidance.
Strengths and Limitations
The main strength of this study is data collection on use of sedatives and different types of sedation, not only continuous deep sedation, in nursing homes. Objective criteria, established in an earlier study, were further complemented for this study to assess the use of sedatives at the end of life independent of its labeling in the medical records.
Additionally, measures were taken to maximize consistency and minimize errors in data extraction, as described in the Methods section.
A main limitation is the study's retrospective design and associated limitations such as unrecorded data, for example, regarding symptoms and level of sedation. As also discussed in our single-center study, this is aggravated by the fact that we only had access to the nursing homes' medical records, not to separate records kept by general practitioners and SPC teams.
However, we chose this design to gain a realistic picture of everyday practice. A prospective study could have influenced the decisions of the health care team and the documentation of their practice.
Another limitation is data collection from centers in one region, making generalizations difficult. To gain maximum variation, we included nursing homes differing in number of residents, affiliation, and location. The definitions of “continuous effect” and at least moderately sedating doses ultimately depended on judgment and are therefore partly subjective. However, we based these judgements on the available data and consensus by experienced SPC pharmacists and clinicians to gain criteria that are as objective as possible. As this was an exploratory study, we did not correct P values for multiple testing. This has to be considered in the interpretation of significant results.
Conclusions and Implications
Compared with international data, our data indicate a lower prevalence of sedation, without the label “sedation,” and considerable differences regarding prevalence between institutions. Further mixed-methods research should focus on gaining more detailed information regarding (1) sedation practices and their documentation in nursing homes, including the use of drugs that are not recommended for sedation in relevant guidelines; (2) reasons for differences between nursing homes and between countries; and (3) potential setting-specific challenges and staff needs for support. Results can inform possible adaptations of guidance to the nursing home setting and development of other support measures for sedation at the end of life in nursing homes.
Acknowledgments
We thank the nursing homes for providing their data and the logistics for data extraction. In addition, we thank AS Licher, Department of Palliative Medicine, LMU Klinikum, Munich, Germany, for checking and cleaning part of the data, and our Patient and Public Involvement group for their comments on the results of this study which helped refining the discussion.
Appendix
Supplementary Fig. 1Number of residents receiving “sedatives with continuous effect” per day.
Some residents received more than 1 sedative. Therefore, the sum of the numbers in the rows above may exceed the total number of residents receiving sedatives displayed in this row.
110 (21)
11 (2)
46 (9)
9 (2)
IQR, interquartile range; NA, not applicable.
∗ Oral dose equivalents except for midazolam.
† Combinations of different sedatives were not considered.
‡ Some residents received more than 1 sedative. Therefore, the sum of the numbers in the rows above may exceed the total number of residents receiving sedatives displayed in this row.
Board of the European Association for Palliative Care. European Association for Palliative Care (EAPC) recommended framework for the use of sedation in palliative care.
Using continuous sedation until death for cancer patients: A qualitative interview study of physicians' and nurses' practice in three European countries.
Palliative sedation in a specialized unit for acute palliative care in a cancer hospital: Comparing patients dying with and without palliative sedation.
Leitlinienprogramm Onkologie (Deutsche Krebsgesellschaft, Deutsche Krebshilfe, AWMF) [Oncology programme guideline (German Cancer Society GCA, AWMF)]. S3 Leitlinie Palliativmedizin für Patienten mit einer nicht-heilbaren Krebserkrankung, Lang-version 2.0, AWMF-Registernummer: 128/001OL [S3 guideline for palliative care for patients with incurable cancer]. 2019.
To explore the relationship between the use of midazolam and cessation of oral intake in the terminal phase of hospice inpatients: A retrospective case note review: Does midazolam affect oral intake in the dying?.
Palliative care specialists in hospice and hospital/community teams predominantly use low doses of sedative medication at the end of life for patient comfort rather than sedation: Findings from focus groups and patient records for I-CAN-CARE.
Patient-related determinants of the administration of continuous palliative sedation in hospices and palliative care units: A prospective, multicenter, observational study.
This work was supported by the Federal Ministry of Education & Research (Grant number 01GY1712, SedEol). The funding body had no role in the design of the study, data collection and analysis, and preparation of the paper.
30710-6&id=gr1.jpg){kind=link}
30710-6&id=fx1.jpg){kind=link}