Abstract

Introduction:

Venous thromboembolism (VTE) is a significant complication in multiple myeloma (MM) patients, yet patient awareness and preventive behaviors in this population have received limited research attention.

Methods:

This cross-sectional survey, conducted at Bazhong Central Hospital (January-September 2024), assessed the knowledge, attitudes, and practices (KAP) of 504 MM patients regarding VTE.

Results:

Participants were predominantly male (65.1%), with 57.7% reporting prior VTE. Mean scores (SD) were suboptimal: knowledge (8.97 ± 2.92, range 0–13), attitude (29.59 ± 2.70, range 14–70), and practice (44.03 ± 4.07, range 10–50). Positive correlations were found between knowledge-attitude (r = 0.141, P = 0.002), knowledge-practice (r = 0.281, P < 0.001), and attitude-practice (r = 0.159, P < 0.001). Structural equation modeling revealed knowledge directly influenced attitude (β = 0.761, P < 0.001), attitude directly affected practice (β = 0.806, P < 0.001), and knowledge indirectly impacted practice via attitude (β = 0.613, P < 0.001).

Discussion:

Findings highlight gaps in VTE knowledge and negative attitudes among MM patients, despite proactive practices. Improving patient education on VTE may enhance attitudes and behaviors, potentially reducing VTE risk in this high-risk population. Targeted interventions are warranted to optimize VTE prevention strategies.

Introduction

Multiple myeloma (MM) is a malignancy of plasma cells, which are integral components of the immune system responsible for antibody production. In MM, plasma cells proliferate abnormally and cluster in the bone marrow, producing excessive antibodies that frequently lead to organ damage, including kidney injury, hypercalcemia, anemia, and bone lesions (1). Despite advancements in treatment options, the global incidence of MM has steadily increased over the past few decades, and it now accounts for approximately 17% of all hematologic malignancies (2).

Patients with MM face a significantly elevated risk of venous thromboembolism (VTE), estimated to be nine times higher than that of the general population (3, 4). This heightened risk is multifactorial, driven by the hypercoagulable state induced by MM, as well as treatment-related factors such as the use of immunomodulators and chemotherapy. Remarkably, even during the asymptomatic monoclonal gammopathy of undetermined significance (MGUS) stage, the risk of VTE is notably increased (5–8). Most VTE events in MM patients occur within the first six months of initiating treatment, underscoring the critical importance of early thrombotic risk assessment and prophylactic strategies for all patients beginning anti-myeloma therapy (3, 9, 10). Recommended prophylactic measures include low-molecular-weight heparin, warfarin, and direct oral anticoagulants (11). However, despite these established guidelines, patient adherence to VTE prevention strategies remains suboptimal across various cancer populations. Unlike many solid tumors, where thrombosis risk is often primarily associated with tumor burden or advanced disease stage, MM presents a uniquely treatment-driven thrombotic profile. The widespread use of immunomodulatory agents such as thalidomide and lenalidomide, particularly when combined with high-dose dexamethasone, substantially amplifies VTE risk and necessitates individualized prophylactic strategies. Moreover, MM is characterized by a chronic, relapsing disease trajectory, requiring prolonged therapy and repeated risk reassessment, which places greater demands on patient understanding and sustained adherence to preventive measures. Compared with other hematologic malignancies, the thrombotic risk in MM is more tightly interwoven with therapeutic regimens rather than disease biology alone, making patient awareness and engagement particularly critical. Therefore, MM patients represent a distinct and clinically meaningful population for investigating VTE-related knowledge, attitudes, and practices. Previous studies in other malignancies have demonstrated significant knowledge gaps and negative attitudes toward anticoagulation therapy, which directly impact preventive behaviors and clinical outcomes (12, 13). To date, limited research has specifically examined patient awareness and behaviors regarding VTE in the MM population, representing a critical gap in understanding how to optimize prevention strategies in this high-risk group.

The Knowledge, Attitude, and Practices (KAP) survey serves as a valuable diagnostic research tool to assess an individual's understanding, beliefs, and behaviors related to a specific topic. Within the context of health literacy, the KAP framework operates on the premise that knowledge positively shapes attitudes, which in turn influence behaviors (14, 15). In the case of MM patients, evaluating their KAP regarding VTE is particularly important because effective thromboprophylaxis requires not only the selection of appropriate anticoagulants but also an understanding of the patient's preferences, adherence likelihood, and comprehension of the disease. This ensures that the most minimally invasive and cost-effective strategies can be tailored to account for individual factors such as age, frailty, and economic considerations (16).

Addressing this gap is clinically significant, as inadequate knowledge or negative attitudes may hinder adherence to preventive measures, ultimately increasing morbidity and mortality. Understanding KAP gaps in this high-risk population allows for the development of targeted interventions to improve VTE prevention and management. Such efforts have the potential to reduce complications and enhance patient outcomes. Therefore, this study aimed to evaluate the KAP of MM patients regarding VTE to inform future individualized prevention strategies.

Material and methodsStudy design and participants

This cross-sectional study was conducted at Bazhong Central Hospital from January to September, 2024, focusing on patients with multiple myeloma (MM). A convenience sampling strategy was employed, recruiting consecutive MM patients who met the eligibility criteria during the study period. Ethical approval was obtained from the Bazhong Central Hospital Ethics Committee, and informed consent was secured from all participants.

Eligible participants were consecutive patients receiving treatment for multiple myeloma, including those with newly diagnosed and relapsed/refractory disease. For elderly patients who were unable to independently comprehend the questionnaire, responses were recorded with the assistance of family members. Patients were excluded if they had severe cognitive impairment that prevented questionnaire completion even with assistance, refused to provide informed consent, or submitted incomplete questionnaire responses (with >20% missing data).

Questionnaire Introduction

The questionnaire was developed based on guidelines and relevant literature (17–19), with additional insights drawn from more than two subsequent studies. Following its initial design, the questionnaire was reviewed and revised based on feedback from three hematology experts to ensure content validity. A small-scale pilot test involving 30 participants was conducted, resulting in a Cronbach's α coefficient of 0.861, which indicated good reliability. To further evaluate construct validity, confirmatory factor analysis (CFA) was performed to assess the three-factor structure corresponding to the knowledge, attitude, and practice dimensions. The model demonstrated good fit, with a root mean square error of approximation (RMSEA) of 0.046, standardized root mean square residual (SRMR) of 0.055, Tucker–Lewis index (TLI) of 0.855, and comparative fit index (CFI) of 0.865, all meeting commonly accepted criteria for adequate fit (Supplementary Table S1 and Supplementary Figure S1). These findings support the structural validity of the questionnaire. For patients reporting a history of VTE, medical records were reviewed to confirm the diagnosis and obtain clinical details. All VTE events were verified through review of imaging studies, discharge summaries, or physician documentation to ensure accuracy of self-reported data. In this study, prior VTE history included any documented venous thromboembolic event occurring before or after MM diagnosis, including deep vein thrombosis, pulmonary embolism, catheter-related thrombosis, and other clinically confirmed venous thrombotic events. Both recent and remote historical events were recorded if documented in medical records. The final version of the questionnaire, written in Chinese, consisted of four dimensions with a total of 55 items: 18 items for basic information, 13 for knowledge, 14 for attitudes, and 10 for practices. For statistical analysis, scores were assigned according to the nature of each item. In the knowledge dimension, correct answers were awarded 1 point, while incorrect or unclear answers received 0 points, with a total possible score ranging from 0 to 13. The attitude dimension included both positive and negative items; positive items (A1–A3, A5–A6, A12–A14) were scored from 5 (strongly agree) to 1 (strongly disagree), while negative items (A4, A7–A11) were reverse-scored, yielding a total score range of 14–70. The practice dimension consisted entirely of positive items, scored from 5 (always) to 1 (never), with a total score range of 10–50. To define adequate knowledge, positive attitudes, and proactive practices, a scoring threshold of >70% was established, based on prior research (20). This threshold was selected based on established benchmarks in health literacy research and has been validated in similar KAP studies across various chronic disease populations. The 70% cutoff represents a balance between achievable competency and meaningful health behavior change.

The questionnaire was distributed through both consultation rooms and WeChat groups, using a QR code generated through Wenjuanxing. Bazhong Central Hospital served as the central hub for survey distribution.

Sample size calculation

The sample size was calculated using the standard formula for cross-sectional studies: n = (Z₁−α/2/δ)² × p × (1−p), where α = 0.05 (Z₁−α/2 = 1.96), δ = 0.05, and p = 0.5 to maximize the sample size. The minimum required sample size was calculated as 384. Assuming a response rate of 80%, the final target sample size was set at 480 participants.

Statistical methods

Data were analyzed using R version 4.3.2 and Stata version 18.0 (StataCorp, College Station, TX, USA). Continuous variables were expressed as mean ± standard deviation (SD), and categorical variables were presented as number (percentage). Group comparisons of KAP scores across demographic and clinical characteristics were performed using independent-samples t-tests or one-way analysis of variance (ANOVA). Spearman's rank correlation analysis was conducted to assess the relationships among knowledge, attitude, and practice scores. For logistic regression analysis, KAP scores were dichotomized according to the median value of each dimension. Univariable logistic regression analyses were first conducted, and variables with P < 0.05 were subsequently entered into multivariable logistic regression models to identify independent factors associated with knowledge, attitude, and practice levels. Prior to multivariable logistic regression, multicollinearity was assessed using variance inflation factors (VIFs). For categorical variables, generalized VIFs (GVIFs) were calculated and adjusted as GVIF^[1/(2 × Df)]. All adjusted values were <2, indicating no significant multicollinearity. Therefore, all covariates were retained in the final models. Based on the KAP framework, structural equation modeling (SEM) was performed to examine the direct and indirect relationships among knowledge, attitude, and practice. SEM analysis was conducted to test the hypotheses that (H1) knowledge directly affects attitude, (H2) knowledge directly affects practice, and (H3) knowledge indirectly affects practice through attitude. Model fit was evaluated using the root mean square error of approximation (RMSEA), standardized root mean square residual (SRMR), Tucker–Lewis index (TLI), and comparative fit index (CFI). All statistical tests were two-sided, and P < 0.05 was considered statistically significant.

Results

Initially, 508 questionnaires were collected. The following were excluded: one case where age was erroneously recorded as >150 years, two cases with abnormal height and weight, and one case where the Padua score was marked as “unknown”. The final dataset included 504 valid responses. The internal consistency of the questionnaire was robust overall and across sections. The overall reliability (Cronbach's α) was 0.8784, with section-specific scores of 0.8123 for knowledge, 0.7679 for attitude, and 0.7363 for practice. The overall validity (Kaiser-Meyer-Olkin, KMO value) was 0.8982.

Demographic information on participants

Of the 504 participants, 328 (65.1%) were male, 214 (42.5%) were aged 70 years or older, and 367 (72.8%) had a BMI within the normal range. Additionally, 291 (57.7%) had experienced venous thromboembolism, including both recent and historical events occurring before or after MM diagnosis. 309 (61.3%) had IgG myeloma subtype, 288 (57.1%) were in stage II, 452 (89.7%) had a Padua score of 0–3 (low risk), 393 (78.0%) had a Khorana score of 1–2 (medium risk), and 300 (59.5%) had an IMPEDE Venous Thromboembolism Risk Assessment Model (IMPEDEVTE) score of 4–7 (medium risk). The mean (SD) scores for knowledge, attitude, and practice were 8.97 (2.92), 29.59 (2.70), and 44.03 (4.07), respectively. Knowledge scores differed significantly based on gender (P = 0.023), age (P < 0.013), residence (P < 0.001), education level (P < 0.001), employment status (P < 0.001), monthly income (P < 0.001), marital status (P = 0.009), health insurance type (P < 0.001), risk disclosure (P < 0.001), Padua score (P < 0.001), Khorana score (P = 0.001), and IMPEDEVTE score (P = 0.007). Attitude scores varied significantly by residence (P = 0.042), education level (P = 0.010), employment status (P < 0.001), monthly income (P = 0.027), risk disclosure (P = 0.003), venous thromboembolism history (P < 0.001), multiple myeloma type (P = 0.035), Padua score (P < 0.001), and Khorana score (P = 0.001). Practice scores showed significant variation by residence (P = 0.003), education level (P < 0.001), employment status (P < 0.001), monthly income (P < 0.001), health insurance type (P = 0.006), risk disclosure (P < 0.001), Padua score (P < 0.001), Khorana score (P = 0.001), and IMPEDEVTE score (P = 0.001) (Table 1).

N = 504N(%)KnowledgePAttitudePPracticePmean (SD)mean (SD)mean (SD)Total score (ranges)504 (100.0)8.97 (2.92) [0–13]29.59 (2.70) [14–70]44.03 (4.07) [10–50]Gender0.0230.0940.093Male328 (65.1)9.19 (2.79)29.72 (2.59)44.30 (3.78)Female176 (34.9)8.56 (3.11)29.35 (2.88)43.53 (4.53)Age0.0130.1330.070Under 60 years old109 (21.6)9.27 (3.32)29.73 (2.84)44.70 (4.18)60–69 years old181 (35.9)8.86 (3.16)29.85 (2.64)43.66 (3.95)70 years old and above214 (42.5)8.91 (2.46)29.30 (2.65)44.01 (4.08)BMI0.1060.0710.134<18.529 (5.8)8.21 (3.35)30.07 (3.45)42.48 (4.72)18.49–23.99367 (72.8)9.05 (2.59)29.46 (2.71)44.17 (3.99)> = 24.0108 (21.4)8.88 (3.74)29.93 (2.40)43.98 (4.11)Ethnicity0.2500.4390.124Han489 (97.0)8.94 (2.94)29.60 (2.66)44.06 (4.11)Minority15 (3.0)9.87 (2.29)29.40 (3.92)43.07 (2.40)Residence<0.0010.0420.003Rural/suburban280 (55.6)8.47 (2.97)29.38 (2.73)43.50 (4.24)Urban224 (44.4)9.58 (2.74)29.86 (2.63)44.70 (3.74)Education<0.0010.010<0.001Middle school or below220 (43.7)8.23 (3.29)29.50 (2.92)43.00 (4.69)High school/technical school234 (46.4)9.41 (2.21)29.44 (2.39)44.70 (3.13)Associate degree/bachelor's degree50 (9.9)10.12 (3.32)30.68 (2.83)45.44 (3.98)Employment status<0.001<0.001<0.001Full-time109 (21.6)9.89 (2.20)30.04 (2.47)45.15 (3.25)Part-time/self-employed/freelancer208 (41.3)9.46 (2.14)29.31 (2.37)44.72 (3.29)Unemployed/laid off48 (9.5)5.02 (4.49)31.65 (3.66)38.50 (4.65)Full-time homemaker89 (17.7)9.07 (2.14)28.52 (2.32)44.17 (3.98)Retired50 (9.9)8.50 (3.41)29.74 (2.83)43.84 (4.11)Monthly income<0.0010.027<0.001<2,00032 (6.3)5.59 (4.05)29.47 (3.62)39.81 (4.54)2,000–5,000432 (85.7)9.07 (2.67)29.49 (2.57)44.29 (3.84)>5,00040 (7.9)10.50 (2.51)30.75 (2.96)44.65 (4.28)Marital status0.0090.3680.237Married471 (93.5)9.04 (2.90)29.61 (2.66)44.10 (4.04)Divorced33 (6.5)7.85 (3.06)29.27 (3.16)43.12 (4.46)Type of health insurance<0.0010.3540.006Social medical insurance only425 (84.3)8.83 (2.85)29.54 (2.72)43.88 (3.98)Both social and commercial medical insurance64 (12.7)10.09 (3.25)29.98 (2.52)45.39 (4.29)No insurance15 (3.0)8.13 (2.59)29.27 (2.71)42.67 (4.43)Whether the doctor informed you about the risk of venous thromboembolism<0.0010.003<0.001No24 (4.8)1.50 (3.44)31.42 (3.37)36.75 (3.31)Yes480 (95.2)9.34 (2.34)29.50 (2.63)44.40 (3.75)Venous thromboembolism0.435<0.0010.530No213 (42.3)8.43 (3.77)30.25 (2.76)43.86 (4.32)Yes291 (57.7)9.36 (2.00)29.11 (2.55)44.16 (3.87)Type of multiple myeloma0.9330.0350.510IgA myeloma128 (25.4)8.87 (3.12)29.52 (2.63)43.91 (4.22)IgG myeloma309 (61.3)9.01 (2.74)29.49 (2.64)44.10 (4.01)Light chain myeloma55 (10.9)9.04 (3.15)29.71 (2.58)44.33 (3.90)Other (b/d/e/g/h)12 (2.4)8.50 (4.19)32.58 (3.85)42.25 (4.61)Multiple Myeloma Staging (R-ISS)0.7120.1530.989Stage I84 (16.7)9.18 (2.38)30.04 (2.68)44.29 (3.56)Stage II288 (57.1)8.91 (2.89)29.41 (2.61)44.11 (3.79)Stage III132 (26.2)8.95 (3.29)29.70 (2.88)43.70 (4.89)Padua Score<0.001<0.001<0.001Low risk = 0–3 points452 (89.7)9.44 (2.29)29.30 (2.44)44.67 (3.54)High risk ≥ 4 points52 (10.3)4.88 (4.34)32.10 (3.47)38.52 (4.21)Khorana Score0.001<0.001<0.001Low risk: total score 0 point95 (18.8)7.25 (4.23)31.00 (3.26)41.89 (4.70)Medium risk: total socre 1–2 points393 (78.0)9.35 (2.37)29.30 (2.45)44.58 (3.72)High risk: total score > 3 points16 (3.2)9.69 (2.15)28.38 (2.00)43.44 (4.08)IMPEDEVTE Score0.0070.1970.001Low risk: total score ≤ 3 points191 (37.9)8.24 (3.66)29.86 (2.72)43.35 (4.20)Medium risk: total score 4–7 points300 (59.5)9.45 (2.24)29.44 (2.65)44.62 (3.70)High risk: total score ≥ 8 points13 (2.6)8.38 (2.53)29.15 (3.31)40.62 (6.65)

Baseline characteristics and KAP score comparisons across demographic and clinical variables (N = 504).

KAP, knowledge–attitude–practice; BMI, body mass index; R-ISS, revised international staging system; IMPEDE VTE, IMPEDE venous thromboembolism risk assessment model.

Distribution of responses to knowledge, attitude, and practice

The distribution of knowledge dimensions showed that the questions with the highest number of participants choosing the “Not sure” option were “Multiple myeloma patients only need to prevent venous thromboembolism within the first six months of treatment”.