A descriptive-analytical cross-sectional study was conducted at Hiwa Hospital in Sulaymaniyah City, the Kurdistan Region of Iraq, to assess the clinical characteristics, severity and symptom burden of anaemia among adult patients with cancer receiving treatment. Hiwa Hospital is the largest public oncology and haematology referral centre in Iraq, receiving patients from all parts of the country. It provides full-spectrum cancer services, combining inpatient and outpatient oncology, chemotherapy and radiotherapy, laboratory and diagnostic imaging, and supportive care. Considering the status of the hospital as a central referral facility with routinely recorded data on the anaemia-related hospital visits of patients with cancer, the institution was identified as the most suitable venue for the study.

The intended demographic consisted of adult patients with both cancer and anaemia who were undergoing treatment at Hiwa Hospital between December 10, 2025 and March 10, 2026. The required sample size was calculated using the Sample Size Calculator found on Calculator.net. This calculator computes the minimum sample size based on the parameters selected, including confidence level, margin of error, population proportion and total population. As the expected population proportion was unknown, a value of 50% was used, as suggested, to obtain the maximum sample size estimate. A finite population correction was applied to the population size specification. With these assumptions, the minimum sample size for the study was 254 participants (https://www.calculator.net/sample-size-calculator.html). Non-probability purposive sampling was used as the study focused on a particular clinical subset of adults.

Eligible patients for the study had to be diagnosed with cancer, aged ≥18 years, and being treated with chemotherapy, radiotherapy, or both. They also had to participate in the study and have clinical and laboratory data in their records to be considered anaemic based on haemoglobin levels. Newly diagnosed cases in 2026 were excluded unless there were sufficient data, as the majority had just commenced their treatment and had received a minimal amount of chemotherapy. Patients were also excluded from participation in the study for having non-malignant haematological conditions that caused anaemia, cognitive or psychiatric disorders that were communicative, and also for receiving palliative care, or for not consenting to participate, or for having data that was not relevant to the study.

Using feedback from pertinent literature (3,11) and experts, a structured, interviewer-administered questionnaire was developed for the present study. The questionnaire consisted of three main sections: i) Socio-demographic characteristics: Age, sex, marital status, educational level, employment status and residence; ii) clinical and cancer-related history: Cancer type/site, stage of disease, duration of diagnosis, and current oncological treatment regimen (e.g., chemotherapy, radiotherapy, targeted therapy); iii) anaemia-related symptoms and assessment: Presence and severity of fatigue, dizziness, shortness of breath, weakness and history of blood transfusion or iron supplementation. Apart from the interviews, data were also collected from the medical records and laboratory reports of the patients covering the period from December 10, 2025, to March 10, 2026. The questionnaires received were on sociodemographic and clinical data, data on the clinical and laboratory indicators of the various levels of anaemia, and data on the symptom burden, which was measured through a physical examination. The haemoglobin concentration was used to classify the severity of anaemia according to the WHO criteria (2). Anaemia was classified according to sex-specific haemoglobin concentrations. In males, mild anaemia was defined as 11.0-12.9 g/dl, moderate anaemia as 8.0-10.9 g/dl and severe anaemia as <8.0 g/dl. In females, mild anaemia was defined as 11.0-11.9 g/dl, moderate anaemia as 8.0-10.9 g/dl and severe anaemia as <8.0 g/dl (2). Therefore, the individual symptom severity category and the overall percentage-based symptom category represent different measures. The Edmonton Symptom Assessment System (ESAS) was used to assess nine symptoms, each scored from 0 to 10(11). Individual symptom scores were categorised as none (score, 0), mild (score, 1-3), moderate (score, 4-6) and severe (score, 7-10). In addition, the total score for each symptom across the 254 participants was calculated as a percentage of the maximum possible score (254x10) and categorised as no/minimal (0-24%), mild (25-49%), moderate (50-74%), or severe (75-100%) (Data S1).

Data were collected from eligible participants during individual face-to-face interviews in a designated private space in the hospital. Each patient was provided with a study information sheet containing the objectives of the study, the procedures and approximate time commitment prior to study participation. Consent to participate was obtained in writing. The data collection questionnaire was read to the participants, and responses were documented by the researcher. The average time allocated for the interview was between 10 and 15 min. Sociodemographic information and symptoms were self-reported by the participants. Clinical and laboratory data were acquired by the researcher through a records and laboratory reports review.

To ensure the face and content validity of the questionnaire, it was sent to a panel of 15 specialists, including experts in oncology, nursing, research methodology, and biostatistics. Experts commented on relevance, clarity, coverage and fit to study objectives for appropriate adjustments to be made. The comments were used to rephrase the questionnaire and improve its validity and precision. A pilot study was performed with 10 patients with cancer who were included in the criteria, but were excluded from the primary sampling study. This pilot study tested the clarity and feasibility of the questions, the interview time, and potential data-collection issues. A small number of adjustments were made based on the pilot results. Instrument reliability was measured with a Cronbach's alpha score of 0.849, representing strong internal consistency, which is an effective reliability score.

Ethical approval was granted by the Research and Ethics Committee of the College of Nursing, University of Sulaimani, with approval no. 2/1 dated November 17, 2025. Prior to commencing data collection, additional permission was requested from the administration of Hiwa Hospital. Participation was voluntary, and written informed consent was obtained from all participants. Their confidentiality was protected through the use of identification numbers. Participants were informed of their rights to withdraw from the study at any time without any repercussions for their care or treatment. All procedures were carried out in compliance with the Declaration of Helsinki.

Data coding, entry and statistical analyses were performed using the Statistical Package for the Social Sciences (SPSS), version 27 (IBM Corp.). Descriptive statistics included frequencies, percentages, means, standard deviations (SDs), minimums and maximums. The normality of continuous variables was assessed using the Kolmogorov-Smirnov and Shapiro-Wilk tests prior to statistical analysis. The assessed continuous variables were normally distributed (P>0.05) and were therefore presented as the mean ± SD. For inferential analysis, Pearson's Chi-square test was used to assess the associations between the severity of anaemia and categorical clinical and symptom-related variables when the expected cell frequencies met the assumptions of the test. Fisher's exact test was used when expected cell frequencies were very low, where applicable. Mean haemoglobin levels across symptom burden categories were compared using one-way analysis of variance (ANOVA) followed by Tukey's post-hoc test. A P-value of ≤0.05 was considered to indicate a statistically significant difference. As the anaemia severity outcome was highly unbalanced, with only 2 patients classified as having mild anaemia compared with 222 patients with moderate anaemia and 30 patients with severe anaemia, multivariable modelling of the three-category severity outcome was considered likely to produce unstable estimates and limited interpretability. Therefore, categorical analyses were retained as the primary analytical approach.

As demonstrated in Table I, the socio-demographic profile of the 254 patients involved in the study provides numerous patterns. Females comprised a larger percentage of the sample than males (57.09 vs. 42.91%). The majority of the participants were aged 55-66 years (46.06%), followed by 43-54 years (35.83%), with an average age of 51.19±10.79 years, indicating a predominantly middle-to-late-adult cohort. The majority were married (89.76%), while being single, divorced, or widowed was less common. As regards education, many were illiterate (22.44%), some graduated from elementary school (40.55%), and some had graduated from college or higher (13.78%), with no one having graduated from postgraduate school. As regards occupation, almost half were homemakers (48.82%), followed by manual labourers (27.56%) and employees (17.32%). Regarding the place of residence, the majority resided in suburban communities (53.54%), urban communities (41.34%), and, lastly, rural areas (5.12%). With respect to monthly income, the vast majority reported that their income was barely adequate (64.57%), while only 13.39% reported having an adequate monthly income. Finally, a large percentage of the participants had a normal body mass index (42.13%); yet 32.28% were overweight and 19.69% were obese, indicating an average body mass index of 25.58±5.39, suggesting that this sample has moderate risks of developing health-related issues due to body weight.

The findings presented in Table II demonstrated that the mean red blood cell (RBC) count was 3.21±0.62x10⁶/µl (range, 1.40-5.58), consistent with anaemia in numerous patients. The mean haemoglobin level was 9.14 g/dl (range, 3.8-11.8 g/dl), and the mean haematocrit was 27.46% (±3.6%), indicating that patients in this group were, for the most part, anaemic. The mean corpuscular volume (MCV) and mean corpuscular haemoglobin (MCH) were within moderate ranges (85.41±10.34 fl and 28.66±3.85 pg, respectively), while the red cell distribution width (RDW) was elevated (15.21±2.25%), suggesting variability in red cell size. The mean platelet count was 205.69±152.23x10³/µl, with a wide range of 6 to 1,421x10³/µl, indicating that some patients had thrombocytopenia or reactive thrombocytosis. The average serum iron level was low (33.15±13.28 µg/dl). By contrast, the mean ferritin level (311.77±102.76 ng/ml) suggested that iron was being sequestered due to inflammation or that iron-restricted erythropoiesis was occurring. The mean C-reactive protein (CRP) level (61.25±30.33 mg/l; range, 2.07-243.90 mg/l) indicates an ongoing inflammatory or disease process in the majority of patients in this group.

The symptom burden of the 254 surveyed patients across various domains is presented in Table III. Almost half of the patients reported moderate pain (49.21%), while 18.50% of patients classified their pain as severe. The measured pain score across the respondents was 1,148 (45.20%), indicating a moderate symptom burden. Other symptoms resulted in similar levels of burden; for example, 62.99% of the respondents reported being moderately fatigued or tired, and 24.80% reported being severely fatigued or tired, yielding an overall score of 1,336 (52.60%), indicating a moderate burden from fatigue or tiredness. In addition, 55.91% reported moderate drowsiness, while 25.59% reported severe drowsiness. The total drowsiness score among respondents was 1,323 (52.09), indicating a moderate impact.

Although 45.67% and 43.70% of the respondents, respectively, reported mild nausea and a lack of appetite, the total scores were still relatively low, at 934 (36.77%) and 1,023 (40.28%), respectively. There was also a low frequency of shortness of breath, with 70.08% reporting it mildly, for a total score of 652 (25.67%). Psychological symptoms were very numerous; depression was reported as severe by 77.17% of patients and accounted for a total score of 1,785 (70.28%), while anxiety was reported as severe by 79.13% of patients and contributed to a total score of 1,832 (72.13%). Overall well-being was rated moderate by 51.57% of patients, while 34.25% rated it severe, with a total score of 1446 (56.93%).

The association between the anaemia status (mild, moderate, severe) and clinical characteristics of the patients, including cancer type, time since cancer diagnosis, stage at diagnosis and current treatment is reported in Table IV. The majority of the patients had moderate anaemia, while mild and severe anaemia were less frequent, indicating that moderate anaemia was the predominant condition among these patients with cancer.

The severity of anaemia and cancer type were significantly associated, based on frequency distributions (Fr%; P-value=0.007; <0.05), indicating differences in the severity of anaemia across the cancer types. By comparison, there were no statistically significant differences in the severity of anaemia by time since cancer diagnosis (P-value=0.158; >0.05), the stage of cancer when diagnosed (P-value=0.345; >0.05), or current cancer treatment (P-value=0.573; >0.05), suggesting that these variables are not significantly associated with the severity of anaemia (Table IV).

The association between anaemia status (mild, moderate, severe) and symptom burden, including pain, tiredness, drowsiness, nausea, lack of appetite, shortness of breath, depression, anxiety and overall well-being is presented in Table V. Moderate anaemia was the most frequent severity, while mild and severe anaemia were less common, indicating that moderate anaemia predominates across all symptom categories.

The P-values for frequency distribution (Fr%) revealed no statistically significant associations for the majority of symptoms, as follows: Pain (P-value=0.554; >0.05), tiredness (P-value=0.887; >0.05), drowsiness (P-value=0.27; >0.05), nausea (P-value=0.37; >0.05), lack of appetite (P-value=0.72; >0.05), shortness of breath (P-value=0.35; >0.05), depression (P-value=0.408; >0.05), anxiety (P-value=0.344; >0.05) and overall well-being (P-value=0.733; >0.05) (Table V). A statistically significant difference was observed in the mean haemoglobin levels across symptom burden categories (P-value=0.003) (Table VI), indicating a statistically significant association between lower haemoglobin concentrations and greater symptom severity.

Indicating a statistically significant difference. Patients with more severe symptom burden had lower haemoglobin levels compared to those with milder or no symptoms.

In the present study, the predominant age group of the patients with cancer was middle-aged and the older segments of the population, with the average age being 51.19±10.79 years, and a slightly larger proportion being females. This is generally similar to previous studies (4,7,12), which have noted that cancer-related anaemia is more prevalent in the middle and later ages of the adult population. However, the sex distribution is different depending on the predominant cancer types in the respective populations (6,7,12). The high percentage of patients with low educational achievements, housewives or manual workers, and with income barely sufficient to cover basic costs reflects socio-economic vulnerability and probably affects diet quality, health literacy and access to cancer support services. Such factors relate to the risk of developing anaemia, as socio-economic disadvantage impacts nutrition and the ongoing care and timely supportive interventions available to patients with cancer (13-17). The majority of the patients resided in suburban or urban locations, possibly indicating that they have simpler access to specialised oncology services, whereas patients residing in rural areas appear to experience some barriers in terms of referral and treatment access (18-20). Furthermore, a considerable number of the participants were overweight or obese, which is critical for the consideration of inflammation-related disturbances of iron metabolism, even if many participants had a normal BMI (19,20). Anaemia in the present study was significant and likely multifactorial according to the laboratory results. Anaemia was confirmed by a low average RBC count, haemoglobin and haematocrit, while near-normal MCV and MCH pointed to a normocytic, normochromic pattern that is cancer-related. With low serum iron, high ferritin and high CRP levels, the case is made for functional iron deficiency due to inflammation-related iron sequestration and not isolated absolute iron deficiency. The increased RDW and changes in count for both the platelet and white blood cell count combine to suggest several factors at work, among them chronic inflammation, nutritional disturbances, marrow suppression, impaired erythropoiesis and the effects of treatments (3,21-22).

The findings revealed that the symptom burden was primarily on tiredness and drowsiness, along with noted pain and overall poor well-being, whereas the milder symptoms were related to nausea, loss of appetite, or shortness of breath. This is comparable to the ESAS-based oncology literature (23), where fatigue and pain were the most well-documented and physically burdensome symptoms, while the presence of nausea and dyspnoea remained less noteworthy. Most notably, the prominence of tiredness and poor well-being is of particular concern in the context of cancer-related anaemia, as fatigue is closely associated with anaemia, along with advanced disease and the presence of a comorbidity during the course of cancer treatment. Moreover, the exceptionally high levels of anxiety and depression suggest that the symptom burden in this cohort was multiform, physically as well as psychologically, highlighting the need for integrated supportive care in anaemic patients with cancer (23-26).

In the present study, the only clinical feature linked to the degree of anaemia was cancer type, while the absence of significant associations for time since diagnosis, stage at diagnosis and current treatment suggests that the type of tumour was more important than the other clinical features in explaining the variance of anaemia within this cohort. This assessment is consistent with previous studies noting tumour type and disease status as key variables that determine the degree of anaemia associated with cancer (8-27). Reviews of the European Cancer Anaemia Survey have documented substantial differences in the levels of anaemia that accompany different types of cancer, with more advanced disease carrying a greater anaemia burden and variances related to treatment. Moreover, haematological malignancies are particularly associated with severe anaemia as a result of direct marrow involvement, impaired erythropoiesis and inflammatory mechanisms, compounded by the cumulative effects of therapy (27-29). The dataset of the present study indicates that the majority of cases of anaemia occur in patients with haematological malignancies, as well as in some cases with breast and other solid tumours, which makes clinical sense. Since blood cancers cause the bone marrow to stop producing blood cells, this often leads to anaemia. Additionally, solid tumours can cause anaemia through chronic inflammation and blood loss, poor nutrition, treatment-induced bone marrow suppression, or some combination of these factors. Similar research performed in Saudi Arabia demonstrated that the burden of anaemia varied with cancer types during active treatment and was particularly high in the genitourinary and gastrointestinal cancers (6,7). Thus, the findings from the present study demonstrate that the type of cancer was the primary clinical factor related to the severity of anaemia in the population examined herein (7,28,30).

The present study did not reveal any statistically significant association between categorical anaemia severity and overall symptom-burden categories (P=0.150). However, when the haemoglobin concentration was analysed as a continuous variable, mean haemoglobin levels differed significantly across symptom-burden categories (P=0.003), with lower haemoglobin levels observed among patients with a greater symptom burden. This finding suggests that the association between anaemia and symptom burden may be more evident when the haemoglobin concentration is considered as a continuous measure rather than grouped into severity categories. The predominance of moderate anaemia across the symptom categories may partly reflect the highly unbalanced distribution of the severity of anaemia in the present study, particularly the very limited number of patients with mild anaemia (n=2). These findings are partly consistent with those of previous studies using the ESAS, which have reported associations between anaemia, fatigue, dyspnoea, pain, appetite loss and other aspects of quality of life (32-34). Kink et al (32) and Barca-Hernando et al (33) also reported that lower haemoglobin levels were associated with a poorer quality of life, greater fatigue and depression among patients with solid tumours. The pathophysiological mechanisms underlying these associations may be related to reduced oxygen-carrying capacity and impaired tissue oxygen delivery, which can contribute to fatigue and reduced functional capacity in patients with cancer-related anaemia (32-34).

In conclusion, the present study assessed the severity and symptom burden of anaemia among adult patients with cancer receiving treatment at Hiwa Hospital. The majority of patients had moderate anaemia, with laboratory findings suggesting an inflammatory component. Anaemia severity was significantly associated with cancer type (P=0.007). Although categorical anaemia severity was not significantly associated with symptom burden (P=0.150), lower haemoglobin levels were associated with a greater symptom burden (P=0.003). These findings highlight the importance of regular haemoglobin monitoring and supportive care. The cross-sectional and single-centre design of the present study limits causal inference and generalisability. Further studies are warranted to confirm these findings.