Characteristics and health problems of cancer patients admitted to palliative care service at the Oncology Hospital in Ho Chi Minh City, Vietnam: a cross-sectional study

A prospective cohort study of cancer patients admitted to the palliative care department began in July 2020 and ended in December 2021. This study describes a cross-sectional analysis of baseline data that was collected for the cohort study. The article was written based on the STROBE checklist for reports of cross-sectional study [11].

The Department of Palliative Care at Ho Chi Minh City Oncology Hospital was established in 2011 and specializes in the physical, mental, and treatment of terminal cancer patients, who are undergoing complicated developments in cancer treatment. The palliative care department accepts patients from the Emergency department or related departments based on medical records after the patient has been assessed for symptoms according to the Guidelines on Palliative care for Cancer and AIDS patients issued by the Ministry of Health in 2006 [12].

The study participants included adults aged 18 years and above, who had been hospitalized for a minimum of 3 days between July 2020 and January 2021 at the Oncology Hospital, a prominent cancer treatment center located in southern Vietnam.

The inclusion criteria specified that prospective participants were 18 years of age or older at the time of admission, recently admitted to the Palliative Care Department for less than 3 days, and accompanied by their primary caregivers. Exclusion criteria were cognitive impairment; unable to provide informed consent; unwell or too distressed to agree to participate in the study.

The study collected demographic information and clinical characteristics of patients including age, gender, ethnicity, marital status, living with relatives, cancer diagnosis.

In addition, the study also collected information about the patient’s health outcomes or health problems. Health problems were defined as physical symptoms, anxiety of the patient and their family, current health status, emotional expressions shared with the family, information received, and specific concerns within a 3-day timeframe, employing the 18-item APCA African POS questionnaire. This questionnaire was used to measure health outcomes and the severity of these problems, rated on a scale from 0 (most positive=no symptoms/problems) to 5 (most negative value=overwhelming symptom/problem burden). The total APCA African POS score ranges from 0 to 90, with higher values representing worse outcomes.

Prospective participants were identified from the palliative care department admission list at the Oncology Hospital. This list provides detailed information including age, sex, type of care, and diagnosis. he researcher approached potential participants in the hospital to explain the study objectives and invite them and their caregivers to participate in the study after obtaining permission from the treating physician. If the potential participant agreed, written consent was obtained from both the patient and the caregiver. Additionally, the researchers conducted a medical record review.

The investigator interviewed the patients using a prepared set of questions, including the APCA African POS, with a recall period of three days, considering cultural, family, and social characteristics of patients and caregivers. In addition, the Australian-modified Karnofsky Performance Scale (AKPS) patient’s severity and activity status questionnaire were assessed by the physician. This is an Australian revised questionnaire from the Karnofsky Performance Status gold standard scale to be more relevant to healthcare clinical settings [13].

The Palliative Care Outcome Scale (POS) measures are a group of tools developed to measure Palliative care needs of patients and their families. The Integrated Palliative care Outcome Scale (IPOS) is integrating the most important questions from POS, Palliative Care Outcome Scale – Symptoms (POS-S) and the APCA POS. POS collects demographic information or clinical characteristics of patients including age, sex, ethnicity, marital status, living with relatives, and cancer diagnosis [13],[14].

IPOS also collects information about the patient’s key concerns, common symptoms, patient and their family’s anxiety, current well-being, shared feelings with family or friends, information received and actual concerns, within a 3-day (for inpatient facility) or 7-day (for emergency patient) time frame [14]. The APCA African IPOS version, validated in Africa, consists of ten questions (18 items) assessing various aspects of physical, emotional, mental, and psychological well-being, information, and support for the patient and family. In addition, three sub-scores are calculated: POS Physical Symptoms Score (pain, shortness of breath, weakness or lack of energy, nausea, vomiting, poor appetite, constipation, sore or dry mouth, drowsiness, poor mobility) ranges from 0 to 50, POS Emotional Symptom Score (worry about illness or treatment, anxiety in family or friends, feelings of life’s worth, and the sense of peace) from 0 to 20 and POS Communication/Practice Score (Communication/practical issues: share feelings with family and friends, the acquisition of desired information, the availability of sufficient help and advice, and the confidence felt by the family) from 0 to 20 [14].

Possible information bias in this study arose from patients self-answering the questionnaire due to their poor health status. This was addressed in the POS questionnaire by allowing designated caregivers or medical staff to respond on behalf of the patient.

Selection bias can also occur because the study may exclude very ill patients who cannot consent to participate in the study, leading to an underestimation of the palliative care needs of these patients.

This study was a part of an international project supported by the World Health Organization (WHO). Its primary objective was to set the foundation for a prospective cohort study evaluating changes in palliative care outcomes and related factors in cancer patients [8]. The project, a collaboration between the Cicely Saunders Institute of palliative care, Policy, and Rehabilitation, King’s College London, and the Ho Chi Minh City Oncology Hospital, aimed to provide data for assessing of the WHO’s palliative care program in the Western Pacific Region. Sample size calculations were based on changes in the baseline POS score and six follow-up assessments at monthly intervals, considering weak to moderate correlations between baseline and follow-up assessments [15]. A minimum sample size of 134 patients were needed to detect a small effect size of 0.25 in POS change score with 80% power and two-sided alpha level 0.05; the estimated number of patients who did not continue to participate during the study was 30% of the observational cohort [15],[16].

Sample size calculation formula:

with

In there:

v = 1: Number of initial evaluations

w = 6: Number of follow-up interview

α = 0.05; β = 0.2

ρ = 0.25: Change of mean POS score between follow-ups

φ = 2.4: SD of the POS score, the sample size calculated was 94, the estimated number of patients who did not continue to participate during the study was 30%, the sample size to be taken was 134.

After checking for the completion of the collected data, it was entered and cleaned using Microsoft Excel 2010 software. Subsequently, the data was analyzed using Stata 13.0 software (Stata Corporation, College Station, TX).

Frequencies and percentages were calculated for categorical variables such as patient and relative characteristics, and the patient symptom rates. Means and SD was estimated for continuous variables such as the five-point Likert scale and total POS score. For skewed distribution variables, median and interquartile range was reported instead. Univariate linear regression analysis was used to examine the association between patient and caregiver characteristics and APCA African POS scores.

All variables with a p-value of 0.25 in the simple linear regression analysis were used to construct the multivariate linear regression model. A two-sided p-value of 0.05 was considered statistically significant.

The study received medical ethics approval from the Ethical Council in Biomedical Research of the Oncology Hospital, Ho Chi Minh City (Reference Number: 427/BVUB-HDD) and King’s College London Research Ethics Committee (Reference Number: HR-18/19-10835).

Out of the total of 145 participants who agreed to take part in the study, six individuals (4.1%) passed away during their hospital stay, while an additional five participants (3.4%) could not be contacted after discharge. There were no significant differences between study and non-study participants in terms of age (p=0.35) and sex (p=0.27). Table 1 shows the demographic characteristics of the patients. In which, the proportion of female patients accounted for 44%, the average age of the patient was 58.1 years old, the youngest was 19 years old and the oldest was 91 years old. Married patients accounted for the highest rate of 81.3%. In terms of education, secondary and high school levels account for nearly equal proportions (34.3% and 35.1%), with above high school accounting for the lowest percentage (8.2%). Nearly half of the patients asked relatives or friends to complete the survey on their behalf, only 43.3% of the patients completed the questionnaires on their own. The proportion of patients experiencing financial hardship was twice that of patients not facing financial difficulties. The proportion of patients who were taking care of their relatives was 20.1%, and taking care of children accounts for a higher proportion.

Table 2 presents the characteristics of the caregivers. The majority of caregivers reported having good health accounting for nearly 50% and they rarely participated in social activities. The mean number of hours a day caring for patients was 19.9 hours, however, caregivers may count the time they have been with the patients. Currently, while taking care of patients, the main income of caregivers was from full-time employment, accounting for the highest percentage (30.2%); the second was the unemployment rate; housework and other activities had nearly equal distribution. Before taking care of patients, most of the relatives worked as full-time workers, the unemployment rate was low. Except for unemployed relatives and housewives, the monthly income of relatives under five million VND (USD210) had the highest rate.

Table 3 presents the clinical characteristics of the patients. There were 52.2% inpatients and 44.8% outpatients. Most patients in the study’s sample were admitted to hospital for less than three days. Cancer of digestive, respiratory, and intrathoracic organs were equal in distribution and accounting for the highest rate, whereas pancreatic cancer had the lowest rate. Regarding the disease stage, the highest rate of stabilization was 43.3%; had roughly equal distribution in the unstable and deteriorating stages. About 32.8% of patients achieved an overall performance state of 60%, which means that they can meet most of their own needs but need help occasionally. Functional status at 70% reached the second-highest rate of 30.6%, roughly equally distributed at 50% and 80%, with no patients dead or normal, no evidence of disease.

Fig. 1 depicts the patient’s physical, emotional, and communication problems based on the POS scale. Over 90% of patients have pain symptoms, with moderate to severe pain accounting for more than 50%. Clinical symptoms such as weakness/lack of energy, difficulty in movement, loss of appetite, pain/dry mouth, and shortness of breath, were prevalent, ranging from 58% to 75%. More than 97% of patients had two or more clinically relevant symptoms, especially nearly 5% of patients who reported having all ten physical symptoms. More than 43% of patients felt worried about their illness in the last three days. This rate is as high as the anxiety level of relatives (over 66%). Most patients have a habit of sharing their feelings with family members or friends (62.5%), feeling peaceful and feeling that life is worth living only on average (57.3% and 44.84%, respectively).

Fig. 1. APCA African POS scale physical, emotional symptoms and communication/practice issues. (a) APCA African Palliative care outcome scale (APCA POS) physical symptoms. (b) APCA POS emotional symptoms. (c) APCA POS communication/practice issues. APCA, African Palliative Care Association; POS, palliative outcomes scale.

Download Original Figure

More than half of patients received help or advice from their families to solve problems for the future. The patient’s family also received more information about their medical condition and felt confident in caring for the patient (71.3% and 89.7%, respectively).

Tables 4 and 5 present the average APCA African POS scores overall and by patient characteristics. The scores of all items were converted to a scale of 0=better status and 5=worse status. The mean total score of the patients was 32.2±10.8, ranging from 11 to 67 points.

The physical symptoms score is 15.1±7.6. Which, symptoms of pain, weakness/lack of energy, and difficulty in functioning were the most common symptoms and had the highest average score. Nausea, vomiting, and drowsiness are the least common symptoms.

The emotional symptoms of APCA African POS score have an average score of 9.7±3.7. Both the patients and their family felt worried a lot of the time about the patient’s situation with the highest score of 2.4 to 2.8 points.

The APCA African POS score in communication/practice has an average score of 7.3±3.7. Generally, the problems with sharing feelings or getting help from family or friends were occasionally. Additionally, it seems that the family was not confident about some issues when caring for the patient with the highest average score of 3.8±1.1 points.

Tables 6 and 7 present the results of univariable and multivariable linear regression analysis examining risk factors for APCA African POS score. Multivariable linear regression found that outpatients were less likely to have severe physical symptoms than inpatients (β=–3.5; 95% CI: –6.1 to –1.1). Conversely, home care or day care patients had more severe physical symptoms than inpatients (β=6.2; 95% CI: –1.1 to 13.4).

In addition, the greater the caregiver’s involvement in social activities, the less likely the patient was affected by physical problems (β=–4.8; 95% CI: –8.1 to –1.5).

The association was also found for caregivers with worse health status, the worse the patient’s emotional problems (β=3.9; 95% CI: –0.4 to 8.2).

Many studies around the world such as the United Kingdom, Germany and Brazil have shown that APCA Africa POS has high value and reliability [10],[21]. Harding et al. have shown that APCA African POS has acceptable value and reliability with Cronbach’s alpha=0.6, showing that the IPOS toolkit is reliable and there is consistency across items [10]. In addition, APCA African POS has good psychometric properties (Cronbach’s alpha=0.6), easy to understand and concise [10].

In this study, the overall mean health outcome score was the sum of 18 items and ranged from 0 to 85 on the APCA African POS questionnaire. This is one of the first studies in Vietnam to evaluate the symptoms of patients at the beginning of palliative care. The results of the study show that the average POS score of cancer patients at the time of admission is higher than that of studies in other countries [21],[25]. The higher the score, the worse the patient’s health and vice versa [26].

The mean scores of physical, emotional and communication symptoms are also higher than studies in other countries [17],[21],[23],[25]. Pain, difficulty in movement and weakness/lack of energy are the three most severe symptoms.

Pain was a typical symptom for cancer patients, there were many causes of pain including tumor pressing on bones, nerves, and other organs or pain after cancer treatment such as surgery, radiation treatment or as a long-term side effects of chemotherapy [25]. A study by author Richard Harding and colleagues performed in palliative care facilities in South Africa and Uganda also found that 82.6% of patients reported pain [27]. Too much rest and bed rest leads to loss of body function, muscle weakness, and reduced range of motion as well as difficulty functioning [28],[29]. Besides the direct effects of cancer and its treatment, people with cancer also often experience other things like surgery, stress, changes in activity levels, changes in blood counts, and blood pressure levels, hormone levels… can interact with each other to make the patient feel weak and lack energy [29],[30]. Nausea, vomiting, and drowsiness were the least common patient symptoms. Some patients also complain of constipation, shortness of breath, and loss of appetite. Some other symptoms reported were headache, insomnia, dizziness, numbness in the limbs, leg pain, cough, inability to move the legs due to loss of sensation, abdominal pain, diarrhea, edema on the backs of hands/feet, pruritus, blurred vision, ulcers, axillary lymph nodes, jaundice, abdominal distention, urinary incontinence… The above results suggested that pain and physical symptoms such as weakness/lack of energy and difficulty in functioning are serious problems in newly admitted palliative care patients.

The results of this study also show that patients are less likely to feel peaceful and have a worthwhile life after their illness. Many newly hospitalized patients were found to have cancer and were transferred to the intensive care unit, leaving them sad, no longer hoping to live, and worried about their families and children after their death. Studies in the UK and Germany also showed similar results, the three most frequently reported psychological problems were patient anxiety, anxiety for family/friends, and feelings of depression [26],[27].

Results found that inpatients had worse physical symptoms than outpatients. In contrast, patients choosing home care or day care service had higher physical symptoms severity than inpatients. Typically, home care patients who cannot go to the hospital regularly because of their poor health.

Our results were like those of Tanya Pidgeon’s study in Australia, which showed that inpatients had more severe physical symptoms than outpatients [31].

For caregivers of patients, if they have a higher income this is also associated with an improvement in the physical health of the patient. Besides, the results of multivariable linear regression analysis also showed that caregivers’ health was also related to patients’ negative emotions. If the person taking care of the patient was in good or very good health status, the patient will have better and more positive emotions. Inpatient or home care patients had more serious physical problems than outpatients. Meanwhile, patients were less likely to have serious physical problems if their caregivers were in better health or participated in more social activities. These findings demonstrate the great importance and dependence of patients on their relatives.

There were several limitations to our study. First, the APCA African POS questionnaire was used to assess the improvement of palliative care outcomes in patients. However, the cross-sectional study only evaluated the mean APCA African POS score at the time of admission, so it did not evaluate the change or improvement in the patient’s health over time. Second, this study only found factors related to patients’ QoL including inpatient and the health status of caregivers, although this study did not prove a causal-and-effect relationship between these factors. Third, the proportion of patients who self-reported the APCA African POS questionnaire was low, more half of their caregivers responded to the questionnaire instead, which could lead to bias in reporting patient symptoms. Fourth, the study may exclude patients with very poor health who cannot consent to participate in the study, resulting in underestimating the patient’s health problems requiring palliative care.