Physician Palliative Education Associated With High Use of Hospice Care Service.

BACKGROUND: Taiwan's National Health Insurance provides coverage for palliative and hospice care. The following 10 types of diseases have been added to the National Health Insurance reimbursement regulation: end-stage cancer, motor neuron disease, organic psychosis, brain degeneration, heart failure, chronic airway obstruction diseases, other lung diseases, chronic liver disease and cirrhosis, acute renal failure, and chronic renal failure.
OBJECTIVE: This study aimed to determine the association between physicians' palliative education and use of hospice care in hospitalized patients at the end of life. DESIGN AND
SETTING: A cross-sectional study in a Taipei community hospital. PARTICIPANTS: Patients who died between 2014 and 2019 were identified. The deceased had at least 1 of the 10 diseases covered by health insurance were included. Hospice care services included hospice ward care and hospice shared care. This study included 2,661 individuals. In total, 972 (36.5%) patients used hospice care services.
RESULTS: After adjusting for age, gender, and comorbidities, physicians' palliative education was found to significantly associated with the use of hospice care (OR: 14.38, 95% CI: 10.90-18.98).
CONCLUSIONS: Physicians' palliative education was found to be an independent factor associated with higher use of hospice care. The findings suggest increasing palliative and hospice education among physicians so that they can ensure that their patients have high-quality end-of-life medical care in an aging society.

Background

The population around the world is aging; in 2019, World Population Prospects revealed that 1 in 11 people will be seniors over 65 years old (about 9%) by 2020 and 1 in 6 people will be older than 65 (about 17%) by 2050. An increasing number of people are dying from cancer and chronic diseases. End-stage patients, including those with cancer, AIDS, heart disease, chronic obstructive pulmonary disease, and kidney disease, experience similar symptoms. In addition to pain, they experience fatigue, anorexia, cachexia, nausea, vomiting, constipation, delirium, and dyspnea.[2-4] Despite continuous advancements and innovations in medical technology, the pain of disease and, ultimately, death is inevitable. In addition to working hard to cure diseases, physicians must also relieve the suffering of patients. When a patient is fighting a terminal disease, palliative and hospice care are a treatment option that can improve the quality of life of both the patient and their family. Palliative care relieves physical, mental, and spiritual pain and can be combined with invasive, life-prolonging measures.[6,7] Palliative and hospice care are not universal, with only 35% of countries/regions including it as part of primary care and 37% reporting community or home palliative and hospice care. Known factors that affect people’s acceptance of palliative and hospice care are cancer, socioeconomic status,[8,9] race, the cognition and attitude of patients and family members,[9,11] and physicians’ cognitive factors. Physicians’ inability to predict the survival period of patients and their lack of knowledge concerning screening out patients who are suitable for palliative and hospice care are the main obstacles. Historically, medical education has rarely emphasized the importance of hospice care. Only if doctors receive continuing education in palliative and hospice care can their behavior produce lasting changes. Education for Physicians on End-of-Life Care is an effective tool for communicating palliative treatment to doctors. This kind of education has effectively improved doctors’ knowledge and practice of hospice medicine.[14,15] Taiwan’s National Health Insurance (NHI) provides coverage for palliative and hospice care including hospitalization, home care, and shared care (wherein patients remain at the original ward and accept joint care of original and hospice doctors). Among cancer patients who died in 2004 in Taiwan, the use rate of hospice care was approximately 15.4%.

Since September 2009, in addition to end-stage cancer and motor neuron disease, 8 other types of diseases have been added to the NHI reimbursement regulation, namely organic psychosis (such as dementia), brain degeneration (such as severe stroke, severe brain injury and other degenerative diseases in the terminal stage), heart failure, chronic airway obstruction diseases, other lung diseases, chronic liver disease and cirrhosis, acute renal failure, and chronic renal failure. In October 2015, Britain’s The Economist published the results of an evaluation of end-of-life care quality; Taiwan ranked sixth in the world and first in Asia. Taiwan’s NHI stipulates that all physicians who declare the various expenses of hospice care must have received palliative and hospice training and education. Since 2014, Taipei City Hospital has organized palliative and hospice education for all its staff. The 13-hour course includes the evaluation and control of end-stage disease symptoms and the psychosocial, spiritual, and religious needs of terminal patients; preparation for discharge; palliative home care; preparation for death; the ethics of decision-making about dying; grief counseling; and follow-up for family members. There is also a 4-hour practice course. All healthcare providers are asked to complete hospice and palliative training before providing hospice care. There was rare study that describes the relationship between doctors’ hospice education and their behaviors. Thus, the purpose of this study is to understand whether doctors’ hospice and palliative education will be related with the chances of end-of-life hospitalized patients to receive hospice care.

Materials and Methods

Study Design, Participants, and Setting

Electronic medical records in a branch of Taipei City Hospital served as the data source. Underlying diseases were coded according to the World Health Organization’s International Classification of Diseases Ninth Revision Clinical Modification (ICD-9-CM) or Tenth Revision, Clinical Modification (ICD-10-CM).

From January 2014 to December 2019, patients who died in hospital were assigned the following diagnostic codes: malignancy (ICD-9-CM 140-208, ICD-10-CM C00-C97), amyotrophic lateral sclerosis (ALS; ICD-9-CM 335.20, ICD-10-CM G12.21), dementia (ICD-9-CM 290, 294.1, 294.9, 331.0, 331.2, 331.7, 331.89, ICD-10-CM F01-F03, G30, G31.09, G31.83), cerebrovascular disease (ICD-9-CM 430-437, ICD-10-CM, I60-I67, I69), Parkinson’s disease (ICD-9-CM 332.0-332.1, ICD-10-CM G20, G21.4, G21.1, G21.11, G21.19, G21.2, G21.3, G21.8, G21.9), heart failure (ICD-9-CM428.0,428.1,428.9, ICD-10-CM I50-I50.9), chronic obstructive pulmonary disease (ICD-9-CM 491.20, 491.21, 492.0, 492.8, ICD-10-CM J43-J44), chronic liver disease and cirrhosis (ICD-9-CM 571.2, 571.5, 571.6, 572.2, 573.4, ICD-10-CM K70, K71.10, K71.11, K72-K74, K76 .3), acute renal failure (ICD-9-CM 584.5-584.9, ICD-10-CM N17.0-N17.2, N17.8-N17.9), and chronic renal failure (ICD-9-CM 585-587, ICD-10-CM N18.4-N19, N26.1, N26.9).

The outcome variable was the use of hospice care services, including hospice ward and hospice shared care. The main explanatory variable was physicians’ palliative education. The month in which the physician obtained the certificate of hospice and palliative training was regarded as the time of training completion. Based on their physicians’ training, the patients were divided into 2 groups: those whose physicians received palliative education—the PE group—and those whose physicians did not receive palliative education—the PNE group.

The control variables included age, gender, and comorbidities. In addition to the inclusion diagnoses, the presence of pneumonia, urinary tract infection, sepsis, hypoglycemia, hyponatremia, cellulitis, as well as diabetes with hyperosmolarity and coma were recorded. Their diagnostic codes are as follows: pneumonia (ICD-9-CM 480.0-487.0, ICD-10-CM J09.X1, J10, J11, J12-J18.9), urinary tract infection (ICD-9-CM 599.0, ICD-10-CM N30.9), sepsis (ICD-9-CM 38.0-38.9, ICD-10-CM A40.0, A41.9, A42.7), hypoglycemia (ICD-9-CM 251.0, 251.1, 251.2, ICD-10-CM E11.641, E16.0, E16.1, E16.2), hyponatremia (ICD-9-CM 276.1, ICD-10-CM E87.1), cellulitis (ICD-9-CM 681.9-682.9, ICD-10-CM L03.1-L03.9), and diabetes with hyperosmolarity and coma (ICD-9-CM 250.20, 250.22, 250.30, 250.32, ICD-10-CM E11.01, E13.01).

Statistical Analysis

Data were reported as mean ± standard deviation (SD) or n (%), as appropriate. Differences in proportions were tested using the chi-square test, and 2-sample t-tests were used for numerical variables. A 2-tailed statistical test was adopted, and a p-value < 0.05 was regarded as statistically significant. A stepwise multivariate logistic regression was used to estimate the association between physicians’ palliative education and the use of hospice care. Statistical analysis was conducted using SAS version 9.4 (SAS Institute, Inc., Cary, NC).

Results

A total of 2,661 patients were included in the study, of whom 972 (36.5%) received hospice care. Their basic information is shown in Table 1. The average age of the sample was 78.2 ± 13.5 years, and there were 1,528 men (57.4%). There was a greater proportion of patients who received palliative and hospice medical care in the PE group than in the PNE group (54.9% and 7.4%, respectively, p < 0.01). In terms of personal disease history, there was a greater incidence of the following diseases among the PE group than among the PNE group: cancer (45.7% and 16.6%, p < 0.01), dementia (14.4% and 11.7%, p < 0.01), chronic obstructive pulmonary disease (15.0% and 6.9%, p < 0.01), chronic liver disease and cirrhosis (11.1% and 7.7%, p < 0.01), and pneumonia (19.6% and 8.5%, p < 0.01), respectively. However, there were significantly more instances of cerebrovascular disease, heart failure, acute renal failure, chronic renal failure, ALS, urinary tract infection, cellulitis, and diabetes with hyperosmolarity and coma in patients of the PNE group. Table 2 presents the factors related to receiving hospice care.

Table 1.

Characteristics of Deceased Patients Based on the Palliative Care Education of Physicians.

	Total (N = 2661)	No. (%) of subjects
		PE (n = 1632)	PNE (n = 1029)	p-value
Sociodemographics
Age in years
Mean ± SD	78.2 ± 13.5	78.3 ± 13.2	78.0 ± 14.0	0.63
<=39	27 (1.0)	14 (0.9)	13 (1.3)	0.22
40–49	70 (2.6)	35 (2.1)	35 (3.4)
50–59	184 (6.9)	121 (7.4)	63 (6.1)
60–69	341 (12.8)	204 (12.5)	137 (13.3)
70–79	559 (21.0)	350 (21.5)	209 (20.3)
>=80	1480 (55.6)	908 (55.6)	572 (55.6)
Gender
Male	1528 (57.4)	953 (58.4)	575 (55.9)	0.2
Female	1133 (42.6)	679 (41.6)	454 (44.1)
Comorbidity
Malignancy	917 (34.5)	746 (45.7)	171 (16.6)	<.01
Dementia	355 (13.3)	235 (14.4)	120 (11.7)	0.04
Cerebrovascular disease	718 (27.0)	366 (22.4)	352 (34.2)	<.01
Parkinson’s disease	119 (4.5)	79 (4.8)	40 (3.9)	0.26
Heart failure	751 (28.2)	341 (20.9)	410 (39.8)	<.01
Chronic airway obstruction	316 (11.9)	245 (15.0)	71 (6.9)	<.01
Chronic liver disease and cirrhosis	260 (9.8)	181 (11.1)	79 (7.7)	<.01
Acute renal failure	611 (23.0)	330 (20.2)	281 (27.3)	<.01
Chronic renal failure	672 (25.3)	329 (20.2)	343 (33.3)	<.01
Amyotrophic lateral sclerosis	33 (1.2)	14 (0.9)	19 (1.9)	0.02
Pneumonia	407 (15.3)	320 (19.6)	87 (8.5)	<.01
Urinary tract infection	52 (2.0)	19 (1.2)	33 (3.2)	<.01
Sepsis	159 (6.0)	65 (3.4)	94 (9.1)	0.8
Hypoglycemia	36 (1.4)	17 (1.0)	19 (1.9)	0.08
Hyponatremia	486 (18.3)	294 (18.0)	192 (18.7)	0.68
Cellulitis	96 (3.6)	38 (2.3)	58 (5.6)	<.01
Diabetes with hyperosmolarity coma	27 (1.0)	9 (0.6)	18 (1.8)	<.01

PE: Patients’ physicians with palliative care education; PNE:Patients’ physicians with no palliative care education.

Table 2.

Univariate and Multivariate Analysis of Factors Associated With Utilization of Hospice.

		Utilization of hospice care	Univariate analysis	Multivariate analysis
Variables	Number of patients	n (%)	OR (95% CI)	AOR (95% CI)
Completion of hospice care education among physicians
No	1029	76 (7.4)	1	1
Yes	1632	896 (54.9)	15.27 (11.85-19.66)*	14.38 (10.90-18.98)*
Age in years
<=39	27	8 (29.6)	1
40–49	70	21 (30)	1.02 (0.39-2.69)
50–59	184	80 (43.5)	1.83 (0.76-4.39)
60–69	341	132 (38.7)	1.50 (0.34-3.52)
70–79	559	206 (36.9)	1.39 (0.60-3.22)
>=80	1480	525 (35.5)	1.21 (0.27-3.00)
Male	1528	530 (34.7)	0.83 (0.71-0.97)*	0.80 (0.65-0.98)*
Malignancy	917	614 (67.0)	7.85 (6.55-9.39)	7.17 (5.64-9.12)*
Dementia	355	159 (44.8)	1.49 (1.19-1.87)	2.46 (1.84-3.30)*
Cerebrovascular disease	718	197 (27.4)	0.83 (0.69-1.01)	1.84 (1.40-2.37)*
Parkinson’s disease	119	40 (33.6)	0.88 (0.59-1.29)
Heart failure	751	192 (25.6)	0.50 (0.41-0.60)	1.36 (1.06-1.75)*
Chronic airway obstruction	316	102 (32.3)	0.59 (0.49-0.72)
Chronic liver disease and cirrhosis	260	107 (41.2)	1.24 (0.96-1.61)
Acute renal failure	611	147 (24.1)	0.47 (0.38-0.58)*	0.76 (0.59-0.99)*
Chronic renal failure	672	187 (27.8)	0.59 (0.49-0.72)*
Amyotrophic lateral sclerosis	33	9 (27.3)	0.65 (0.30-1.40)	2.83 (1.11-7.25)*
Pneumonia	407	104 (25.6)	0.55 (0.43-0.70)*	0.48 (0.36-0.63)*
Urinary tract infection	52	3 (5.8)	0.10 (0.03-0.33)*	0.12 (0.03-0.42)*
Sepsis	159	28 (17.6)	0.35 (0.23-0.54)*
Diabetes with hyperosmolarity coma	27	4 (14.8)	0.30 (0.10-0.87)*
Hypoglycemia	36	4 (11.1)	0.21 (0.08-0.61)*	0.27 (0.08-0.90)*
Hyponatremia	486	168 (34.6)	0.90 (0.72-1.12)	1.34 (1.03-1.74)*
Cellulitis	96	25 (26.0)	0.60 (0.38-0.96)*

Care services among deceased patients. OR: Odds ratio; AOR: Adjusted odds ratio; CI: Confidence interval.

A stepwise multivariate logistic regression analysis revealed that patients whose attending physicians underwent palliative education had a higher chance of receiving hospice care (OR 14.38, 95% CI 10.90-18.98). Other factors that were associated with a higher chance of receiving hospice care included malignancy (OR 7.17, 95%CI 5.64-9.12), dementia (OR 2.46, 95% CI 1.84-3.30), cerebrovascular disease (OR 1.84, 95% CI 1.40-2.37), heart failure (OR 1.36, 95% CI 1.06-1.75), ALS (OR 2.83, 95% CI 1.11-7.25), and hyponatremia (OR 1.34, 95% CI 1.03-1.74).

Factors that were associated with a lower chance of receiving hospice care included being male (OR 0.80, 95% CI 0.65-0.98) and having acute renal failure (OR 0.76, 95% CI 0.59-0.99), pneumonia (OR 0.48, 95% CI 0.36-0.63), urinary tract infection (OR 0.12, 95% CI 0.03-0.42), and hypoglycemia (OR 0.27, 95% CI 0.08-0.90). When stratified by chronic airway obstruction, patients without chronic airway obstruction who suffered from pneumonia had a reduced chance of receiving hospice care (OR 0.45, 95% CI 0.32-0.62). When stratified by chronic renal failure, patients without chronic renal failure who had acute renal failure showed a reduced chance of receiving hospice care (OR 0.59, 95%CI 0.43-0.82) (Supplementary Table 1). Table 3 shows the sensitivity analysis based on gender. Among men as well as women, patients of the PE group had a higher chance of receiving hospice care (OR 12.77, 95% CI 8.76-18.62 and OR 14.81, 95% CI 9.96-22.02, respectively).

Table 3.

Sensitivity Analysis of the Associations Between Physicians’ Hospice Education and the Utilization of Hospice Care Services, Adjusting for Patient Characteristics.

Study subgroups	Utilization of hospice care services AOR (95% CI)
All patients (n = 2,661)
Physicians with hospice education	14.38 (10.90-18.98)
Physicians without hospice education	1
Male patients (n = 1,528)
Physicians with hospice education	12.77 (8.76-18.62)
Physicians without hospice education	1
Female patients (n = 1,133)
Physicians with hospice education	14.81 (9.96-22.02)
Physicians without hospice education	1

AOR: Adjusted odds ratio; CI: Confidence interval.

Discussion

Main Findings

This study included 2,661 patients, of whom 36.5% received hospice care. After adjusting for age, gender, and comorbidities, physicians who underwent hospice and palliative education were found to increase the chances that hospitalized patients would opt for hospice care at the end of life. Studies have found that the most common obstacle to hospice treatment is the unwillingness of physicians to refer patients to hospice specialists, followed by physicians’ lack of knowledge about which types of patients are suitable for hospice treatment and their lack of interest, knowledge, and skills in hospice care. In the past few years, the fields of hospice and palliative medicine have developed rapidly.[22,23] Hospice education continues to expand. Medical students, clerks, interns, residents, and attending physicians are required to undergo training courses. The Taipei City Hospital organizes many hospice and palliative education training programs every year. Some studies have found that the implementation of education and training programs will significantly improve the level of medical knowledge of hospice and palliative care. In this study, cancer patients constituted the highest proportion of those using palliative care (67.0%), followed by those with dementia (44.8%), and those with chronic liver disease (41.2%). Taiwan’s NHI covers hospitalization, home care, and shared care for hospice and palliative care. However, concerning non-cancer end-stage patients, several physicians and families of patients generally believe in the concept of rescue, even if the patients are in intensive care units; do-not-resuscitate orders are not signed until all therapeutic options have been exhausted.[26,27] According to data from the National Health Insurance Agency in 2014, about 96% of Taiwan’s hospice care services were provided to cancer patients. Considering the needs of the large aging population, non-cancer patients should also receive high-quality end-of-life care.

What This Study Adds

In this study, those with a lower chance of receiving hospice medical care included men and patients with acute renal failure, pneumonia, urinary tract infections, and hypoglycemia. The reason men had a lower chance than women of receiving hospice care can be attributed to the fact that men were most likely the breadwinners of the family and may have still been receiving monthly pensions. According to the stratification analysis, among patients without chronic airway obstruction, doctors and family members preferred active treatment for those who suffered from pneumonia because they had a better chance of being cured. Thus, they had a reduced chance of receiving hospice care (OR 0.45, 95% CI 0.32-0.62). Moreover, because the NHI pays for hemodialysis for patients without chronic renal failure, doctors and family members preferred active treatment for those who suffered acute renal failure (OR 0.59, 95%CI 0.43-0.82) (Supplementary Table 1).

Elderly people with chronic diseases are often hospitalized repeatedly in the years before dying, for reasons such as fever and weakness. Pneumonia, urinary tract infection, and hypoglycemia are also common causes of hospitalization. Usually, when patients are hospitalized due to an acute condition, antibiotics and other medication can be used until the cure fails. Therefore, the chance of receiving hospice care is reduced. To improve this situation, it is necessary to enhance the awareness of medical staff and the public so that timely recommendations for providing hospice and palliative treatment can be made to ensure proper care for patients at the end of life. The implementation of education and training programs among physicians will significantly improve the level of medical knowledge of hospice and palliative care, thereby leading to better quality of care for end-of-life patients.

Strengths and Limitations

Our study had several strengths, including providing insights into a novel issue (i.e., doctors’ hospice education and their behaviors), considering a significant number of comorbidities to adjust for multiple confounders, and having studied patients for a period of 6 years. However, failing to consider religion, educational level, and cognition of patients and their family members may be one of the limitations to this study. Moreover, the subjects of a hospital are not representative of all patients, thus limiting the generalizability of our results to all terminal patients.

Implications for Current Practice

The implementation of education and training programs among physicians will significantly improve the level of medical knowledge of hospice and palliative care, thereby leading to better quality of care for end-of-life patients.

Conclusion

From 2014 to 2019, the use of palliative care among patients with end-stage diseases who were admitted to the Taipei City Hospital was approximately 36.5%. This study found an association between physicians’ palliative education and use of hospice care services. Physician palliative education was found to be an independent factor associated with greater use of hospice care services. This finding could help inform hospital policies and encourage doctors and other medical staff to undergo palliative care training so that they can guide their patients holistically, especially toward the end of their lives. Thus, it is recommended that physicians receive palliative and hospice education so that they can provide high-quality end-of-life medical care in an aging society. Further research is needed to test the effects of religion and cognition of patients and their family members on hospice care use.

Click here for additional data file.

Supplemental Material, sj-pdf-1-ajh-10.1177_10499091211014160 for Physician Palliative Education Associated With High Use of Hospice Care Service by Mei-Hsing Chuang, Fang-Niarn Lee, Yih-Tsong Shiau, Hsiu-Yi Shen, Chih-Ching Lee, Saint Shiou-Sheng Chen and Sheng-Jean Huang in American Journal of Hospice and Palliative Medicine®