Prevalence and factors associated with malnutrition on patients with cancer in Bangladesh: a cross-sectional study

Introduction

Cancer is a global public health concern and causes significant morbidity and mortality. According to Global Cancer Statistics 2020, there were 19.3 million newly diagnosed cancer cases and almost 10 million people died from cancer worldwide.1 2 These increasing trends in mortality from cancer accompany the demographic and epidemiological transitions around the globe, particularly in low-income and middle-income countries3 such as Bangladesh where cancer is the second-leading cause of death.4

According to the World Health Organization, ‘nutritional status’ can be defined as the condition of the body resulting from the intake, absorption and utilisation of various types of nutrients that influence a person’s physiological and psychological status.5 People’s socioeconomic and demographic characteristics are highly correlated with their health and nutritional status.6 Fundamental demographic factors such as age, gender and ethnicity can contribute in one form or another to nutritional status.7 Body mass index (BMI) is a widely used indicator of the nutritional status of a population. However, the Subjective Global Assessment, comprising an observational history focusing on gastrointestinal symptoms, weight loss and physical examination, is a reliable and widely used tool for nutritional screening.8

It is documented that nutrition is an important factor related to the prevention and/or development of chronic diseases and functional limitations in the general population.9 Disease-related malnutrition occurs frequently among patients with cancer, which is a major cause of mortality and morbidity among them.10 11 Malnutrition refers to deficiencies or excesses in nutrient intake, imbalance of essential nutrients or impaired nutrient utilisation.5 Cancer cachexia is a specific form of cancer-associated malnutrition, often occurring in patients with advanced disease.12 Malnutrition among patients with cancer increases the cost of healthcare and also increases the risk of infection.13 It also decreases the patient’s quality of life, affecting their treatment process and overall survival.14 15 There have been studies where the nutritional status of patients with cancer and their clinical outcomes have been assessed, such as deterioration of nutritional status and surgical outcome.16 17 Moreover, there are studies contributing to the association and correlation between nutritional status and sociodemographic status among the general population or any specific niche of population.18 19

In Bangladesh, there is a scarcity of research focusing on patients with cancer, though cancer is the second-leading cause of death.4 To our best knowledge, there is a lack of published study in Bangladesh that has empirically evaluated the association between the nutritional status of patients with cancer and their sociodemographic status and treatment measures. This study is, therefore, justifiably designed and aimed to identify the extent of malnutrition among patients with cancer and investigate its relationship with demographic status and clinical characteristics.

Methods and materials

Data collection

We used convenient sampling techniques to perform this study. Patients at two tertiary care cancer hospitals in Dhaka, Bangladesh, who were receiving treatment at the inpatient and outpatient departments were included in our study. Patients younger than 18 years and those who had severe illness or mental health illness were unable to understand or comprehend a research questionnaire and refused to give consent were excluded. Relevant queries from patients and their caregivers were answered to help them understand the study objectives and goals. Recruiting was conducted daily among eligible patients. The data collection process was maintained by experienced health professionals and physicians. A pilot study was conducted among 14 patients, after which the questionnaire was revised and finalised on feedback. It was made clear to the respondents that participation in the study was entirely voluntary, and the face-to-face interview took place one person at a time to ensure privacy. The selected information regarding demographic, cancer and treatment-related variables was collected by a Bengali version of the questionnaire from the respondents (online supplemental file 1).

Finally, a total of 275 data were selected for the analytical exploration. The confidentiality of patient data and anonymity was secured by using codes instead of personally identifiable information. Data selection, identification and inclusion flow diagram process are presented in figure 1.

Request permission

Figure 1

Distribution of study participants.

Results

A total of 275 patients with cancer were included in the study. The mean age of the study participants was 49.54±15.613 (SD) years with minimum 19 and maximum 74 years old. The average height of patients was 159.98±8.11 (SD) cm and highest weight was 92 kg with mean 56.59±10.37 (SD) kg. We also measured the BMI of patients where mean value was 22.12±3.67 (SD) with maximum 35.49 kg/m² (online supplemental file 2).

The majority of participants were more than 60 years of old (n=75, 27%). More than half of the participants were males (n=164, 60%), had extended families (n=147, 54%) and were unemployed (n=147, 54%). Approximately 19% (n=52) of patients were underweight, while 66% (n=180) were normal weight. The majority of participants (n=84, 31%) had secondary level or more education. Approximately 67% of participants were married (n=183) and had monthly household income ranged from BDT40 000 to BDT80 000 (n=182) (online supplemental file 3).

Across cancer sites, approximately 29% (n=78) had lung cancer, 23% (n=63) had upper gastrointestinal tract cancer, 16% (n=44) had breast cancer and 16% (n=42) had cancer in reproductive system. Among participants, 20% (n=53) had metastasis. Approximately 32% (n=87) of patients were taking anticancer treatments for more than 12 months and half of patients (n=138, 50%) were following the instructions of a dietician. Nearly half (n=123, 45%) of patients were under chemotherapy, and 41% (n=112) had surgical treatment. Most participants were from inpatient departments (n=203, 78%), where 23% (n=62) were hospitalised for more than 15 days. Nearly half of (n=121, 44%) patients had poor PS (score: ≥2) (online supplemental file 4).

Nutritional status was classified into three categories (A=well nourished, B=suspected or moderate malnutrition and C=severe malnutrition). The overall prevalence of malnutrition was 67% (n=184); 35 (13%) patients had severe malnutrition and 149 (54%) patients had moderate malnutrition.

A χ² test was conducted to determine the association between dependent and independent variables. Among the considered sociodemographic characteristics, patients’ age (p<0.001), sex (p=0.042) were found significantly associated with nutritional status. However, no statistically significant association were observed for household members, patient’s occupation, education, marital status, smoking and monthly household income. Among the cancer-related variables, cancer site (p=0.002), follow dietician’s advice (p=0.008), duration of treatment (p=0.009), chemotherapy (p=0.001), surgical treatment (p=0.029), duration of hospitalisation (p<0.001) and PS (p=0.035) were significantly associated with nutritional status (table 1).

We further categorised moderate-severe malnutrition in the poor nutrition category and independent characteristics that showed significant association in the χ² test were subjected to multinomial logistic regression analysis. Patients between 40 and 60 years old were 2.96 times more likely to be malnourished (adjusted OR, AOR 2.96; 95% CI 1.40 to 6.24) compared with those under 40 years old. The regression model also showed that female patients had 7.74 times more risk of malnutrition compared with male patients (AOR 7.74; 95% CI 2.03 to 19.80). We found that patients who underwent surgical treatment (AOR 2.60; 95% CI 1.29 to 5.26) and underwent the treatment process for more than 12 months (AOR 4.06; 95% CI 1.82 to 9.08) were more likely to be malnourished (table 2).

Discussion

Our study aimed at exploring the prevalence of malnutrition and its association with sociodemographic status and treatment measures among Bangladeshi patients with cancer. We used a Bengali version of the questionnaire to ensure effective communication with patients, which in turn contributed to safeguarding the accuracy of the data. We obtained several key findings in our study that are pertinent to patients with cancer.

According to our data, 54% of patients with cancer had moderate malnutrition, and 13% had severe malnutrition. While a similar study conducted in India found that 32% of patients with cancer were moderately malnourished, lower than our findings and 53% were severely malnourished which is higher than our results.19 A study from China found that 32% of patients with cancer were malnourished using the same tool (PG-SGA tool),26 which is lower than our findings (67%). The difference might be due to several factors, study populations, area, ethnicity and sample size. This could also be because, despite healthy and nutritious food, the nutritional well-being of a significant portion of Bangladeshi population is still neglected.27 28 Malnutrition can negatively affect a cancer patient’s prognosis and outcome, and patients with cancer must maintain a healthy nutritional status to maximise their response to anticancer therapies. As a result, nutritional evaluation of patients with cancer should be established as an essential preventative measure to prevent malnutrition.

After conducting a χ² test, we found the patient’s age and gender to be significantly associated with nutritional status. Previous studies reported that increasing age is independently associated with poor nutritional status and has a significant and independent effect on several key biochemical and anthropometric measure variables used in nutritional assessment.29 30 A study carried out among the Indian population found a significant association with age group and nutritional status.31 Although the aforementioned studies were carried out on healthy adults, further investigation should be conducted to determine whether age influences nutritional status in patients with cancer like healthy adults. Interestingly, a study carried out among advanced patients with lung cancer in northern China reported both age and sex to be associated with malnutrition,32 which is in congruence with our findings. Another study carried out in Turkey, reported that younger age was associated with inadequate nutritional status in hospitalised patients with cancer.33

Among cancer-related variables, we found several variables, including PS, to be significantly associated with nutritional status. A prior study concluded that malnutrition affects PS in patients with pancreatic cancer.34 Another article reported that PS is one of the most important variables that affect the prognosis of patients with cancer.35 The epidemiological estimates of future cancer cases suggest that older patients with cancer will continue to increase across the globe in the years to come.36 As such, using low-cost, practical nutritional risk assessment tools for older patients with cancer will allow specialised nutritional interventions and help patients’ quality of life.

In our study, multinomial logistic regression analysis yielded several predictors. The findings of this study are in agreement with those of previous studies. In the current study, we found that participants who were between 40 and 60 years old were 2.96 times more likely to be malnourished compared with those under 40 years old. A Brazilian study reported that older patients with cancer are at greater risk of being malnourished compared with younger patients with cancer.37 Another study from Australia carried out on patients with cancer that also used the PG-SGA to evaluate nutritional status, it was also observed that older patients with cancer were at higher risk of being malnourished compared with younger patients.38 It is evident that older adults are vulnerable to malnutrition due to age-related physiological decline, reduced access to nutritious food and comorbidities.39 A possible explanation for this could be that, with ageing, body composition changes, resulting in a reduction in lean body mass among older adults and these changes may alter muscular strength, functionality and independence in this population. These findings can make healthcare challenging in developing regions, such as Bangladesh, where there is a shortage of adequate resources and infrastructure for geriatric medicine. Therefore, early systematic nutritional status must be monitored to ensure favourable clinical outcomes in patients of all ages and with age progression.

Our study found that female patients with cancer had a 7.74-time higher risk of malnutrition than male patients. Opanga et al reported that more males than females were severely malnourished in their study (55% vs 45%). The samples were taken from Kenyan patients with cancer, and this was statistically significant (p<0.001).21 We also found that patients going through surgical treatment were 2.60 times more susceptible to malnutrition. Similarly, a more recent cohort study has reported a high incidence of severe malnutrition among patients undergoing surgery for gastrointestinal cancer.13 A Swedish study is consonant with our finding, where they reported malnutrition to be a problem occurring after oesophagectomy.40 Another study resonated with our finding, where they mentioned that patients undergoing pancreatic resection for malignant tumours are usually malnourished.41 It is important to point out that people with metastatic cancer have lower quality of life, which is linked to poorer treatment outcomes.42 Therefore, the authors emphasise that causality must be established through further causal inferential studies.

The study has several strengths. Out study’s main strength is that it collected data from two tertiary-level cancer care hospitals, and health professionals were involved in the data collection process. Furthermore, our results provide some insight into the status of malnutrition among patients with cancer in Bangladesh. Moreover, we maintained all appropriate guidelines during study period which is crucial for the study. The study also has limitations. First, our calculated study sample size was 316; however, we were able to collect 275 data from two cancer hospital due to lack of time and human resources. Such a small sample size can lead to fallacious conclusions and less accurate results. So, the rate of malnutrition may not be representative for each patient with cancer. However, longitudinal studies that include larger numbers of patients to better determine the results found in our research are still needed. Second, our study data were not homogeneous, which may have affected the results. Finally, there is a possibility that response-related biases may exist in our study.

In conclusion, our data revealed a relatively high prevalence of malnutrition among patients with cancer in Bangladesh. Furthermore, patients aged over 40, females, who underwent surgery and were hospitalised for more than 12 months had a higher risk of malnutrition. These results highlight the need for nutritional screening and assessment both for characteristics of malnutrition and for underlying risk factors soon before and after treatment/hospitalisation to enable early and multidisciplinary or interdisciplinary interventions for better treatment outcomes.