In this study, we found that overall knowledge regarding hyperkalemia management among Yemeni hemodialysis patients was moderate, with a mean correct score of 52.6%. Attitudes were generally positive (72.2% of the maximum score), whereas self-reported practices were suboptimal (68.5% of the maximum). Younger age and higher education were associated with better scores across all domains. Given the cross-sectional design, all observed relationships should be interpreted as associations rather than causal effects.
Hyperkalemia can precipitate life-threatening cardiac arrhythmias by altering myocardial membrane potential (
17). Therefore, effective self-management is essential for patient safety in hemodialysis populations (
18-
20).
In our study, the mean knowledge score was 7.37 out of 14 (52.6%). This finding is similar to findings from Jazan, Saudi Arabia, where only 20.2% of hemodialysis patients demonstrated adequate knowledge about potassium-rich diets (equivalent to approximately 55% correct on comparable items) (
6). In Jordan, Khalil et al. reported that 61% of patients had poor dietary knowledge (
21), and in Kuwait, expert consensus highlighted knowledge deficits in 68% of patients (
22). In contrast, a study from Shenzhen, China, a higher-resource setting, reported a mean knowledge score of 12.9 out of 14 (92.1%) (
5). These differences likely reflect disparities in healthcare infrastructure, access to dietitians, and structured patient education.
Examination of item-level knowledge responses identified important gaps in three key areas. First, pharmacological knowledge was limited: only 21.5% of patients were aware that some diuretics may help reduce potassium levels. Similarly low awareness (18%) was reported for medication-related items in the Jazan study (
6).
Second, recognition of high-potassium foods was suboptimal. Only 37.0% of participants correctly identified dried fruits (e.g., dates, figs, raisins) as high in potassium. This finding is notable given the frequent consumption of these foods in the local diet. A comparable Middle Eastern study reported a similar pattern, with 41% correctly identifying dates as high-potassium foods (
6,
23).
Third, knowledge of practical dietary management strategies was limited. Only 30.9% of participants knew that boiling vegetables may reduce potassium content, and 27.2% understood the role of potassium binders. In contrast, a study from Shenzhen reported that more than 90% of patients answered similar items correctly (
5).
These findings indicate specific knowledge deficits, particularly in pharmacological understanding, food identification, and practical dietary techniques. Clinically, these gaps are relevant in settings where dietary counseling may be limited and laboratory monitoring is not consistently accessible (
12,
13). However, these observations reflect associations within the study population and should not be interpreted as causal determinants of behavior.
In this study, attitudes were generally positive, with most participants acknowledging the importance of potassium control and expressing trust in healthcare providers. However, self-efficacy remained limited, as only some patients felt confident in following a low-potassium diet. This mismatch between general attitudes and behavioral confidence has been widely reported in chronic disease populations (
6,
24). Such discrepancies may reflect environmental constraints, limited dietary options, and low self-efficacy rather than a lack of awareness.
From a behavioral science perspective, this pattern aligns with established health behavior theories, such as the Health Belief Model and the Theory of Planned Behavior, which suggest that attitudes alone are insufficient to drive behavior without perceived control and enabling conditions (
10,
11). In this context, attitudes may act as a key mediating construct between knowledge and practice rather than as a direct determinant of behavior.
In this study, practice levels were suboptimal, particularly for dietary self-management behaviors such as food labeling, dietary recording, and meal preparation. In contrast, medication adherence was relatively high. This pattern of high medication adherence but poor dietary self-management mirrors findings from South India, where 78% adhered to medications but only 34% followed dietary advice (
8). These findings suggest that behavioral interventions should extend beyond knowledge provision and incorporate structured skill-building and reinforcement strategies.
Multivariable analysis showed that younger age and higher educational level were associated with better knowledge and attitudes. This finding is consistent with prior studies reporting a strong relationship between educational attainment and health literacy in chronic kidney disease populations (
6,
21-
23,
25). These associations likely reflect differences in comprehension, access to information, and ability to apply dietary recommendations.
Previous studies have identified a significant association between knowledge, attitudes, practices, and income (
25). However, this relationship was not examined in the current study, which was conducted in a governmental setting where all participants were economically disadvantaged.
The negative association between regular medication use and practice scores may suggest that patients who rely more on pharmacological management may perceive less need for dietary control. However, this interpretation remains speculative. Given the cross-sectional design, this finding should be considered an association only and requires confirmation in longitudinal studies.
Although we did not directly measure socioeconomic status, the broader humanitarian context in Yemen, characterized by poverty, food insecurity, and disrupted healthcare systems, likely contributes to difficulties in dietary self-management. These contextual constraints may partially explain the suboptimal practices observed in this study.
Structural equation modeling indicated that attitudes were strongly associated with practices, whereas knowledge influenced practices indirectly through attitudes. This finding is consistent with previous findings from Shenzhen, where attitudes demonstrated a strong association with behavioral outcomes (
5). These results support the role of attitudes as a key mediating construct in self-management behavior, in line with behavioral theory frameworks.
The model also showed a negative direct association between knowledge and practices. This finding should be interpreted cautiously and not as evidence of a causal inverse relationship. Several noncausal explanations may account for this observation. First, statistical suppression may occur when the indirect pathway (from knowledge through attitudes to practices) obscures the direct effect. Second, measurement differences between declarative knowledge (“knowing that”) and procedural behavior (“knowing how”) may contribute to inconsistent associations. Third, reverse causality may be possible, whereby patients experiencing poor control may seek more knowledge after clinical deterioration. Fourth, unmeasured confounding factors, such as health literacy, food insecurity, psychological distress, and cultural dietary practices, may influence both knowledge and behavior in opposite directions. Therefore, this result reflects a modeled statistical association rather than a true causal effect.
Regarding model performance, the SEM fit indices for CFI and TLI were slightly below conventional thresholds. However, the RMSEA and SRMR values were within acceptable ranges. Given the exploratory nature of this analysis and the complex humanitarian context, model fit was considered acceptable but not confirmatory. These findings should therefore be interpreted as hypothesis-generating.
Overall, the model explained a modest proportion of the variance in practice behavior, consistent with previous studies in hemodialysis populations. Dietary self-management is influenced by multiple interacting factors, including cultural habits, economic constraints, food availability, and psychosocial conditions (
26-
29), many of which were not captured in this model.
5.1. Implications
These findings suggest that improving knowledge alone is unlikely to meaningfully improve self-management behavior. Interventions should instead focus on strengthening attitudes, self-efficacy, and practical dietary skills. Given the associational nature of this study, these recommendations should be tested in future longitudinal or interventional research before implementation at scale (
18,
19,
30).
5.2. Study Limitations
This study has several limitations. First, the single-center design limits generalizability. Second, self-reported practices may be subject to recall and social desirability bias. Third, important variables, including income, food security, health literacy, and psychological stress, were not measured. Fourth, the cross-sectional design precludes causal inference; therefore, all findings should be interpreted as associations only. Fifth, SEM fit indices were slightly below conventional thresholds, indicating that the results should be interpreted cautiously. Sixth, the modest explained variance suggests that additional determinants of dietary behavior remain unmeasured. Future multicenter, longitudinal, and interventional studies are needed to clarify causal pathways.
5.3. Conclusions
Hemodialysis patients demonstrated moderate knowledge, relatively positive attitudes, and suboptimal practices regarding hyperkalemia management. Age and educational level were associated with knowledge, attitudes, and practices (KAP) scores, and attitudes showed a strong association with practices.
These findings indicate the need for targeted and practical educational interventions to improve self-management among hemodialysis patients.