Consultation behaviors and factors influencing hesitation to seek medical care among local dengue patients in Guangzhou, China: a retrospective analysis

Abstract

Section Background

Dengue fever remains a significant public health challenge in Guangzhou, China, where healthcare-seeking behavior plays a critical role in shaping disease outcomes. This study investigates patterns of consultation delays and their determinants among locally acquired dengue cases from 2015 to 2024, aiming to inform targeted public health interventions.

AbstractSection Methods

Surveillance data on locally acquired dengue fever cases in Guangzhou were extracted from Chinese National Notifiable Infectious Disease Reporting Information System for analysis. Categorical variables were summarized as frequencies and percentages (N, %), while continuous variables were presented as medians with interquartile ranges (IQR). Chi-square tests were used to examine differences in symptom onset and consultation timing across groups with varying demographic characteristics, and non-parametric Wilcoxon rank-sum tests assessed variations in consultation hesitation time (defined as the interval from symptom onset to medical consultation). To identify determinants of delayed care-seeking, a modified Poisson regression model with robust error variance adjustment was employed, incorporating key demographic characteristics as predictors.

AbstractSection Results

From 2015 to 2024, Guangzhou recorded 8,533 locally acquired dengue fever cases, with men accounting for 52.5% of cases (men-to-women ratio = 1.12:1). The overall median healthcare-seeking delay was 4 days (IQR 2–6). Notable spatiotemporal variation emerged, as residents of non-central areas presented earlier than those in urban centers (median 3 vs. 4 days), and while the median delay remained unchanged during high-incidence months, the interquartile range (IQR) narrowed from 3 to 6 days in low-incidence months to 2–6 days in high-incidence months. This contraction in IQR suggests more proactive healthcare-seeking behavior during epidemic periods (Z=-4.321, P < 0.001). Multivariable Poisson regression with robust standard errors confirmed significantly longer delays during low-incidence periods (IRR = 1.122, 95%CI:1.052–1.196). Weekend consultation rates decreased significantly during high-incidence months across most subgroups, though no significant weekday-weekend differences were observed among individuals in high-exposure occupations or during low-incidence periods (P > 0.05). Age-stratified analysis demonstrated significantly increased delay risk across all younger age groups compared to patients ≥ 65 years, with the greatest risk elevation in children and adolescents aged 0–18 years (IRR = 1.538, 95%CI:1.436–1.648), followed by adults aged 41–65 years (IRR = 1.093, 95%CI:1.053–1.134) and 19–40 years (IRR = 1.067, 95%CI:1.027–1.108), revealing a bimodal delay pattern. Distinct occupational gradients were equally apparent in the analysis. Workers in low exposure-risk occupations experienced the longest median delays (median 5 days; IQR 4–7) and highest adjusted risk (IRR = 1.429, 95%CI:1.341–1.523) compared with high-risk occupations, while moderate-risk occupations showed both the shortest delays (median 2 days, IQR 1–3) and significantly reduced risk (IRR = 0.652, 95%CI = 0.609–0.699). Similarly, special populations demonstrated reduced risk (IRR = 0.658, 95%CI:0.611–0.708).

AbstractSection Conclusions

This study highlights systematic disparities in dengue fever healthcare-seeking behavior, driven by occupational exposure risk, seasonal transmission dynamics, and demographic factors. To reduce delays, urgent implementation of targeted interventions is required. Interventions should incorporate health education initiatives in workplaces for high exposure risk occupational groups, while also focusing on low exposure risk occupational groups and urban residents experiencing delays in seeking care. Additionally, expanding weekend healthcare accessibility and developing age-specific education programs are essential to enhance epidemic response efficiency and reduce disease burden.

Background

Dengue fever, an acute infectious disease transmitted by Aedes mosquitoes, is prevalent in many tropical and subtropical regions and has emerged as a major global public health challenge. According to the World Health Organization (WHO), approximately half of the world’s population is now at risk of dengue, with an estimated 100–400 million infections occurring annually [1]. As of April 2024, dengue outbreaks have surged in most regions globally, with case numbers in some areas tripling compared to the same period in 2023 [2], underscoring the escalating severity of this health crisis. Concurrently​​, the annual economic burden of dengue is estimated to range from $610 million to $1.384 billion, with a per capita cost of $1.06 to $2.41, reflecting its substantial financial impact [3].

In China, dengue fever has similarly inflicted substantial harm and imposed a heavy economic burden [4]. Since the first reported case in 1978, the number of affected regions and cases has shown a continuous upward trend. In 2023 alone, mainland China recorded ​​19,627 dengue cases​​, primarily distributed in Yunnan, Guangxi, Guangdong, Fujian, and Zhejiang provinces. Among these, Guangdong Province, a key endemic area, reported 4,198 cases [5, 6]. Notably, Guangdong has recorded dengue cases annually since 1990, with increasing frequency of outbreaks in recent years. A major dengue epidemic centered in Guangdong in 2014 resulted in over 45,000 reported cases—more than three times the total number of cases recorded in the previous two decades—with 80% of these cases occurring in Guangzhou [7, 8].

Guangzhou, a core node of the Maritime Silk Road with a permanent population of 18 million, has long faced risks of imported and locally transmitted dengue due to its geographical proximity to Southeast Asia, frequent trade and population movement, and warm, humid subtropical monsoon climate [9,10,11]. Historically a commercial gateway of China, Guangzhou has consistently borne a disproportionate burden of dengue. Between 1978 and 2011, dengue cases in Guangzhou accounted for approximately 70% of Guangdong Province’s total and 50% of the national total [12, 13]. Critically, Guangdong exhibits a ​​significantly higher average annual incidence of indigenous dengue​​ than other endemic regions like Yunnan Province (3.65 vs. 0.86 per 100,000 population), underscoring Guangzhou’s role as the ​​epicenter of local transmission​​ [14].

Addressing these challenges may require enhancing public awareness of dengue fever, strengthening diagnostic capacity, and implementing effective early intervention measures to mitigate its impact. Transmitted by infected mosquitoes, dengue fever is an acute infectious disease whose incubation period is characterized as 3–14 days, with a mean duration of 7 days [15]. It is followed by a clinical syndrome characterized by high fever, severe headache, muscle and joint pain, fatigue, and a distinctive rash [16]. While most infections remain mild, they can still progress to severe manifestations, specifically dengue hemorrhagic fever or dengue shock syndrome. In these severe forms, secondary infection with a different viral serotype triggers antibody-dependent enhancement, which poses significant mortality risks [17,18,19]. These distinct clinical attributes, including predictable incubation timelines, acute onset of debilitating symptoms, and potential for rapid progression to life-threatening complications, underscore the critical importance of seeking medical attention promptly after symptom onset.​ For effective management of infectious diseases like dengue, early diagnosis and intervention are paramount. Timely detection and treatment not only support patient recovery but also enhance outbreak control efficiency, reduce severe case morbidity and mortality, limit further viral transmission, and protect vulnerable populations within communities [20, 21]. Despite the importance of early intervention, research specifically addressing the consultation patterns and hesitation times of dengue patients remains limited. This knowledge gap may contribute to delays in seeking medical attention after symptom onset, increasing the risk of disease transmission and complicating control measures.

This study aims to comprehensively analyze the consultation behaviors and delay durations of local dengue patients in Guangzhou, identify determinants influencing their healthcare-seeking behaviors, and provide evidence-based insights for public health authorities in regions facing similar dengue challenges.

Methods

Data source

This study on healthcare-seeking awareness among dengue fever patients in Guangzhou was approved by the Human Research Ethics Committee of the Guangzhou Center for Disease Control and Prevention (Ethics Approval No.: GZCDC-ECHR-2024P0176) on February 5, 2024. Due to its retrospective design, the ethics committee waived the requirement for informed consent. Following ethical approval, patient data access commenced in April 2024. Critically, no personally identifiable information of participants was collected or retained during or after data acquisition.

Dengue fever case records in Guangzhou from January 1, 2015, to December 31, 2024, were obtained from the Chinese National Notifiable Infectious Disease Reporting Information System (CNNDS) managed by the Chinese Center for Disease Control and Prevention (China CDC). This system provides detailed epidemiological information, including gender, onset date, and consultation date, enabling a comprehensive analysis of consultation patterns and hesitation times among dengue fever patients. Moreover, the dataset comprehensively encompasses both community health service centers and hospitals, ensuring that the analysis captures a wide spectrum of healthcare - seeking behaviors across different levels of medical institutions. For cases diagnosed from 2015 to 2017, the Chinese Dengue Fever Diagnostic Criteria (WS 216–2008) [22] were applied. In contrast, cases from 2018 to 2024 were diagnosed according to the WS 216–2018 criteria [23]. A careful comparison between the two sets of criteria revealed that the core diagnostic elements for dengue fever, namely nucleic acid and antigen detection, remained consistent across both versions. The additional content introduced in the 2018 edition pertained mainly to definitions of severe dengue and disease management guidelines, which did not impact the identification and classification of cases within the scope of this study.

The division between “Central Urban Area” and “Non-Central Area” is primarily based on the Guangzhou Municipal Territorial Spatial Master Plan (2021–2035) [24]. Consultation Hesitation Time, treated as a continuous variable, is defined as the time interval from the onset of symptoms to the first clinical visit to a medical institution. refine the analysis of occupational exposure risks for dengue fever, occupations were categorized based on the Occupational Classification Dictionary of the People’s Republic of China and environmental transmission dynamics: workers frequently exposed to mosquito-prone outdoor environments (e.g., farmers, construction workers) were classified as high-risk occupations; those in semi-enclosed settings (e.g., food service staff) as moderate-risk; indoor professionals (e.g., office administrators) as low-risk groups; and populations with irregular activity patterns, such as students and unemployed individuals, were designated as a special category. Detailed occupational classification criteria are provided in Supplementary material 1.

Statistical analysis

Categorical variables were represented as numbers and percentages (N, %), whereas continuous variables were presented as median values along with interquartile ranges (Q1, Q3). The consultation hesitation period was defined as the duration from symptom onset to the date of medical consultation. Chi-square tests were utilized to investigate the differences in symptom onset and consultation days across participants with varying demographic characteristics.

Since dengue fever is an acute infectious disease with a disease course generally not exceeding two weeks, we excluded extreme values beyond this duration. Non-parametric Wilcoxon rank-sum tests were employed to assess differences in consultation hesitation time across demographic groups. To evaluate the impact of demographic characteristics on consultation hesitation time—modeled as a continuous variable—we incorporated ​​gender, area of residence, age groups, month, and occupation​​ into a modified Poisson regression model with robust error variance adjustment. This approach was chosen to circumvent convergence issues common in negative binomial regression, and multicollinearity was rigorously tested. Detailed model diagnostics (goodness-of-fit, multicollinearity tests) are provided in Supplementary material 2. Results are presented as ​​incidence rate ratios (IRR)​​ with 95% confidence intervals (CI).

Given the weekend closures of community health service centers, these facilities were excluded from our analysis in the weekend effect study (Comparison Between Symptom Onset and Consultation Timing). Only hospitals with confirmed weekend operating hours were retained to ensure methodological rigor. All statistical analyses were conducted using IBM SPSS Statistics (version 27, IBM SPSS Inc., Chicago, USA). A P-value of less than 0.05 was deemed statistically significant.

Results

Demographic characteristics of dengue fever cases

Between 2015 and 2024, Guangzhou documented a total of 8,533 locally acquired dengue fever cases. Among these, 4,482 cases (52.53%) were recorded in men, while 4,051 cases (47.47%) were noted in women, resulting in a men-to-women ratio of 1.12. The age distribution revealed that the highest proportion of cases, 41.04%, was among individuals aged 19 to 40 years. Local dengue outbreaks in Guangzhou were most prevalent from August to November, with 96.32% of all reported cases occurring during this timeframe. Regarding occupation, the Moderate Exposure Risk Occupations accounted for the largest share of cases (34.79%), whereas the High Exposure Risk Occupations comprised the smallest percentage, at a mere 5.68% of the total (Table 1).

[IMAGE OMITTED: SEE PDF]

Comparison between symptom onset and consultation timing

Statistical analysis revealed a marked divergence between symptom onset dates and medical consultation patterns. While incident distribution remained consistent across weekdays, a significant decline in healthcare-seeking behavior was observed during weekends (Saturdays and Sundays). This phenomenon demonstrated cross-population consistency across genders, age cohorts, and residential areas, though nuanced variations emerged when stratifying by Months and Occupations. The weekday-weekend consultation gap proved particularly pronounced during high-incidence months (August-November), with consultation rates decreasing by 25.12% (95%CI: 23.34–26.90%) on weekends compared to Symptom Onset. Conversely, no statistically significant temporal variation (P = 0.116) was detected in low-incidence periods. Notably, occupational subgroups with High exposure risks also exhibited distinct consultation patterns, maintaining stable weekend consultation rates (P = 0.438), whereas other occupational subgroups exhibited statistically significant weekday-weekend disparities characterized by a marked weekend reduction in consultation frequency (P < 0.001), especially for low exposure risk occupations with consultation rates decreasing by 34.16% (95%CI: 30.79–37.53%) on weekends compared to Symptom Onset (Fig. 1).

[IMAGE OMITTED: SEE PDF]

The figure illustrates the daily counts of symptom onset (orange bars) and clinical consultations (green bars) over a seven-day period, stratified by gender, residential areas, age groups, month of incidence, and occupational exposure risk. ​​Chi-square tests for differences​​ in daily counts of symptom onset and clinical consultations ​​within each subgroup​​ were performed, with corresponding ​​p-values​​ displayed in the upper-right corner of each bar.

Analysis of consultation hesitation time and influencing factors

The temporal patterns of healthcare-seeking delays among dengue fever patients and their associated determinants are detailed in Fig. 2. The overall median interval from symptom onset to first medical consultation was 4 days (Interquartile Range IQR 2–6), with no significant differences between men and women patients. However, marked heterogeneity emerged by area of residence, months, age groups, and occupational exposure risk. Residents of non‑central areas exhibited a shorter hesitation span (median 3 days, IQR 2–5) than those dwelling in the urban centers (median 4 days, IQR 2–6). Seasonally, although the median delay remained at 4 days during epidemic months (August–November), its dispersion narrowed (IQR 2–6) compared with non‑epidemic periods (median 4 days, IQR 3–6), reflecting more uniform care‑seeking under outbreak conditions. Agestratified analysis revealed a bimodal distribution, with the youngest (0–18 years) and middleaged (41–65 years) cohorts experiencing relatively prolonged delays, young adults (19–40 years) seeking care more promptly (median 4 days, IQR 2–5), and the elderly (≥ 66 years) recording the shortest median delay (3 days) coupled with the greatest variability (IQR 1–5.25). Occupational exposure risk similarly exhibited a clear gradient, as individuals in low exposure risk occupations were most hesitant (median 5 days, IQR 4–7), those in moderate exposure risk roles presented most rapidly (median 2 days, IQR 1–3), and both high exposure risk workers and special population patients displayed intermediate patterns (median 4 days, IQR 2–6, and median 3 days, IQR 1–4, respectively).

[IMAGE OMITTED: SEE PDF]

The figure displays the median and interquartile range (IQR) of consultation hesitation time (defined as the time interval from symptom onset to the patient’s clinical visit to a medical institution), along with scatter points representing actual data distributions. These metrics are stratified by gender, area of residence, age group, months of disease incidence, and occupational exposure risk. Non-parametric statistical tests (Mann-Whitney U test for two-group comparisons and Kruskal-Wallis test for multi-group comparisons) were conducted to assess differences in hesitation time within each subgroup. The corresponding statistical test values (Z-score or H-statistic) and P-values are displayed in the top-right corner of each subplot.

Multivariable robust poisson regression analysis of consultation hesitation time determinants​

Multivariable robust Poisson regression analysis demonstrated significant heterogeneity in consultation hesitation duration. As detailed in Table 2, patients during low-incidence months exhibited significantly prolonged hesitation compared to high-incidence months (IRR = 1.122, 95%CI: 1.052–1.196). While central urban residency showed a non-significant trend toward increased hesitation versus non-central areas (IRR = 1.016, 95%CI: 0.989–1.043) and men demonstrated marginally longer hesitation than women (IRR = 1.017, 95%CI:0.990–1.044), neither reached statistical significance. Notably, all younger age cohorts exhibited significantly elevated hesitation relative to the elderly reference group (≥ 65 years), with the most pronounced effect observed in the 0–18 year group (IRR = 1.538, 95%CI: 1.436–1.648), followed by the 41–65 year group (IRR = 1.093, 95%CI: 1.053–1.134) and the 19–40 year group (IRR = 1.067, 95%CI: 1.027–1.108). Occupation strata revealed significantly longer hesitation durations for low exposure-risk occupations compared to the high-risk reference group (IRR = 1.429, 95%CI: 1.341–1.523), whereas both moderate-risk occupations (IRR = 0.652, 95%CI: 0.609–0.699) and special populations (IRR = 0.658, 95%CI: 0.611–0.708) demonstrated significantly shorter hesitation.

[IMAGE OMITTED: SEE PDF]

Discussion

Between 2015 and 2024, Guangzhou reported a total of 8,533 locally acquired dengue fever cases. The analysis revealed no statistically important differences in the distribution of the days on which symptoms began. However, consultation days exhibited marked variation, with most subgroups demonstrating a higher frequency of consultations on weekdays compared to weekends. This phenomenon, known as the “weekend effect,” has been observed not only in China but also in several other countries [25,26,27]. It is primarily attributed to restricted access to healthcare services during weekends, along with a range of individual factors. For consultation hesitation, the overall median hesitation time was 4 days, with statistically major variations observed among specific demographic subgroups.

Consultation patterns revealed notable variations in symptom onset-to-consultation timelines across genders, age groups, and residential areas, though stratification by months and occupations showed nuanced differences. ​​​​Gender-stratified and residential area–stratified analyses​​ demonstrated divergent healthcare-seeking behaviors. While weekend consultations were significantly fewer overall, ​​the absence of significant gender differences in hesitation time​​ suggests that gender minimally influences care-seeking decisions—a finding aligned with studies on infectious diseases like hand, foot, and mouth disease [28]. In contrast, Central Urban Area residents experienced longer consultation delays than Non-Central Area counterparts.​​ This divergence persists ​​despite Guangzhou’s healthcare resource gap narrowing, consistent with its first-tier city status​​, highlighting the complex interaction between occupational constraints and healthcare accessibility in urban settings. High-pressure professional environments in central zones may ​​compel working populations to prioritize occupational commitments over health needs​​, leading to symptom neglect until clinical deterioration [29]. Conversely, Non-Central Area residents exhibited ​​more proactive healthcare engagement​​, likely facilitated by ​​reduced occupational pressures and optimized primary care access. The observed urban-rural divergence extends to health literacy dynamics. ​​Metropolitan populations with higher education levels​​ frequently leverage digital health resources for symptom triage [30]. While enhancing health awareness, this engagement ​​may paradoxically delay consultation through self-reassurance about mild symptoms​​ [31, 32]. Urban traffic congestion​​ further compounds systemic barriers, particularly during peak hours, creating disincentives for timely clinical visits [33]. These structural barriers amplify individual-level hesitancy, especially among patients underestimating symptom severity. Emerging telemedicine solutions demonstrate potential for mitigating such accessibility challenges, with several nations implementing virtual care platforms to bypass physical infrastructure limitations [34,35,36]. Initiatives such as the “Internet Hospitals” framework represent progressive steps toward healthcare democratization, though operational challenges persist in service standardization and equitable resource distribution across diverse socioeconomic groups [37, 38]. The findings collectively underscore the necessity for spatially-tailored public health strategies that address both behavioral determinants and structural healthcare access barriers in metropolitan environments.

Monthly analysis of healthcare-seeking patterns in Guangzhou’s dengue transmission revealed distinct differences between ​​high-transmission seasons (August–November)​​ and ​​low-transmission periods​​. During ​​epidemic months​​, healthcare facilities reported ​​significantly higher patient volumes on weekdays versus weekends​​, whereas off-peak months exhibited ​​balanced visit distributions without statistical significance​​. Further analysis indicated ​​prolonged intervals between symptom onset and medical consultation during low-incidence periods​​, reflecting ​​diminished public vigilance and care-seeking hesitancy under reduced risk perception​​—a trend contrasting sharply with ​​expedited healthcare engagement during outbreaks​​. This accelerated responsiveness during high-transmission periods is likely attributable to government-coordinated public health campaigns that implemented multichannel risk communication strategies, including community engagement, media dissemination, and digital alerts, to improve symptom awareness and encourage early care-seeking [39]. While health agencies in many countries, such as national Centers for Disease Control and Prevention, issue regular advisories to maintain vigilance against infectious diseases [40], and China’s CDC and other public health authorities similarly conduct awareness-raising campaigns [41], these efforts tend to focus on high-incidence or severe outbreak phases and may neglect the low-prevalence periods of the disease. Furthermore, during the pandemic, early case detection among high-risk populations primarily resulted from organized screening programs rather than voluntary health-seeking behavior, which might have led to a reduction in their consultation intervals. Additionally, the enhanced allocation of healthcare staff and increased accessibility of medical services during the pandemic period also played a significant role in shortening patients’ hesitation times [30].

Healthcare-seeking behaviors exhibit marked age-dependent heterogeneity, as evidenced by significant variations in symptom-to-consultation intervals across distinct life stages. Empirical analyses identify two cohorts with ​​prolonged care-seeking delays​​: the ​​youngest (0–18 years)​​ and ​​middle-aged groups (41–65 years)​​. Conversely, ​​young adults (19–40 years)​​ demonstrate ​​consistently timely healthcare engagement​​, while ​​elderly patients (≥ 66 years)​​ display the ​​shortest median delays​​ yet the ​​highest temporal variability​​ in consultation timing. These behavioral divergences stem from fundamentally distinct ​​socioeconomic role constraints​​ and ​​health prioritization mechanisms​​ governing each age group. Among young adults, public-facing occupational requirements, particularly employer-mandated infectious disease screening protocols, likely prompt timely medical consultations, supported by proactive health awareness that enhances early symptom detection [42]. Middle-aged individuals, burdened by caregiving responsibilities, frequently prioritize personal/family obligations over non-urgent care during weekends, deferring consultations to weekdays [43, 44]. This behavioral pattern potentially explains the observed surge in patient volumes during workdays when symptoms escalate or impair functionality [45, 46]. Delays in youngest care-seeking primarily arise from insufficient parental supervision, as children’s limited ability to articulate symptoms, combined with caregivers frequently overlooking early mild manifestations, leads to gaps in symptom detection [47, 48]. Conversely, accelerated elderly care-seeking arises from institutionalized health advocacy in assisted-living environments, where routine monitoring prioritizes early intervention. This trend is reinforced by expanding senior care infrastructure [49, 50] and specialized healthcare policies across nations, including China’s enhanced community health centers and family physician programs optimizing elderly care accessibility [51, 52]. However, similar to the dependency challenges seen in youngest care, homebound or socially isolated elderly individuals reliant on external assistance for medical access often exhibit prolonged hesitation in seeking care, resulting in significant temporal variability in the interval between symptom onset and clinical consultation. Equally pressing, persistent barriers in reducing middle-aged care delays remain critical, underscoring how life-stage-specific social roles and support systems fundamentally shape healthcare engagement timing. These findings collectively necessitate age-tailored public health interventions to address population-specific access barriers.

The study further reveals distinct mechanisms underlying healthcare-seeking behavior variations across occupational groups, with no significant weekend differences in case numbers versus hospital visits observed among high-exposure risk populations. When analyzing care-seeking delays, it was found that low-exposure risk occupational groups exhibited the longest hesitation time, followed by high-risk groups and special populations such as students and unemployed individuals, while moderate-risk groups demonstrated the shortest delay. From a risk perception perspective, low-risk groups often perceive non-acute symptoms as self-manageable health issues due to their safer work environments [53]. Additionally, their limited exposure to occupational risks also results in lower awareness of dengue prevention, as public health campaigns typically prioritize high-risk groups. Although targeted health education has resulted in high-risk occupational groups such as farmers and delivery workers not exhibiting weekend-related delays when seeking care, their prolonged hesitation persists due to physically demanding outdoor labor, limited health insurance coverage, and financial burdens from high out-of-pocket medical costs. These overlapping barriers may hinder timely healthcare access, prolonging both symptom evaluation and decision-making processes before initiating treatment [54, 55]. In contrast, moderate-risk groups operate within an “alert threshold” zone – benefiting from sufficient health campaign attention without livelihood-related care delays, thus maintaining optimal response efficiency. Notably, extended delays among students and unemployed populations may correlate with incomplete social security coverage and limited health literacy. Students’ care-seeking speed often depends on parental symptom recognition, while unemployed individuals face compounded barriers including inadequate medical insurance and financial constraints, both contributing to symptom neglect [56].

Based on the findings of the study, several strategies may be proposed to optimize medical services during dengue outbreaks. Firstly, the “Health Administration Department” should adopt a dynamic methodology for the detection and management of dengue cases, customized to align with the epidemic phase and the availability of medical resources. In the initial stages of an epidemic, healthcare facilities in regions identified as hotspots should implement a “test-upon-fever” strategy, whereas areas devoid of local cases may adhere to a “test-upon-suspicion” protocol. The prioritization of hospitalization and isolation for confirmed dengue patients is essential to enable early detection and mitigate the spread of infection. As the epidemic advances and the number of cases increases, the emphasis should transition towards optimizing resource allocation to avert overcrowding in healthcare facilities and to diminish the proportion of severe and fatal cases. Secondly, enhancing the accessibility of outpatient fever clinics during weekends in peak epidemic months is imperative to address the decline in consultation rates stemming from constrained medical resources. Additional provisions should be allocated to these clinics, and investments in telemedicine and digital health solutions should be prioritized to improve accessibility and minimize patient wait times. These measures may facilitate a more efficient healthcare delivery system and ensure that patients receive prompt medical care. Thirdly, public health education campaigns should be augmented to promote the early recognition of dengue symptoms and to raise awareness regarding associated risks. While these initiatives are particularly crucial during months of heightened incidence, sustaining year-round awareness campaigns is equally significant to maintain public vigilance. Utilizing various media platforms to disseminate targeted health messages can ensure that both the general populace and high-risk groups remain informed and proactive in seeking timely medical attention. Lastly, when addressing the diverse spectrum of occupational groups, it is imperative that our focus extends beyond merely targeting individuals with a high risk of dengue exposure. Equal consideration should be given to those with lower occupational exposure risks, as they represent a substantial segment of the population. Despite their relatively diminished risk, overlooking this group could inadvertently exacerbate delays in accessing timely medical care.

There also existed some limitations to this research. Firstly, our reliance on data from the National Surveillance Network for Infectious Disease Prevention introduces potential risks to data integrity; any inaccuracies or incomplete records within this database may compromise the validity of our results. For instance, heterogeneity in hospital service provision—specifically operational disparities between 24-hour fever clinics and general outpatient departments—may have confounded the findings. Due to constraints in the surveillance data, these variations across service tiers could not be differentiated in our analysis, highlighting the need for future data collection that incorporates granular stratification of operational differences. Secondly, the exclusive focus on locally transmitted dengue cases in Guangzhou limits the generalizability of our findings. While the insights may be applicable to regions with similar outbreak patterns driven by imported cases, validation across diverse geographic and epidemic contexts remains essential. Thirdly, consultation hesitation time was calculated based on self-reported dates of symptom onset, which are susceptible to recall bias. Although dengue’s acute presentation with prominent symptoms (e.g., high fever, rash) likely enhances the vividness of recall, thereby mitigating the impact of recall bias on the results to a certain extent. Finally, unmeasured cases involving initial self-medication or informal care may overestimate consultation delays and affect behavioral interpretations.

Conclusion

The findings of this study elucidate the intricate interplay of factors influencing consultation behavior regarding dengue fever in Guangzhou. Comprehending these dynamics is imperative for formulating effective public health strategies that facilitate timely healthcare access and enhance health outcomes. Sustained public health education, the balanced consideration of both high exposure risk and low exposure risk workplaces, along with the adequate provision and enhancement of healthcare infrastructure, are crucial components in reducing patients’ hesitation to seek medical care and addressing the identified disparities.

Data availability

The data that support the findings of this study are available from the Chinese National Notifiable Infectious Disease Reporting Information System (CNNDS) but restrictions apply to the availability of these data, which were used under license for the current study and managed by the Guangzhou Center for Disease Control and Prevention, and so are not publicly available. Data are however available from the corresponding author upon reasonable request and with permission of the Guangzhou Center for Disease Control and Prevention.

Abbreviations

WHO:

World Health Organization

CNNDS:

Chinese National Notifiable Infectious Disease Reporting Information System

China CDC:

Chinese Centre for Disease Control and Prevention

IQR:

Interquartile Range​​

IRR:

Incidence rate ratio

CI:

Confidence interval

References

World Health Organization. Dengue and severe dengue. https://www.who.int/news-room/fact-sheets/detail/dengue-and-severe-dengue. Accessed 25.09.2024.

World Health Organization. Dengue - Global situation. https://www.who.int/emergencies/disease-outbreak-news/item/2024-DON518. Accessed 25.09.2024.

Shah S, Abbas G, Riaz N, Anees UR, Hanif M, Rasool MF. Burden of communicable diseases and cost of illness: Asia Pacific region. Expert Rev Pharmacoecon Outcomes Res. 2020;20(4):343–54.

Xu M, Chang N, Tu T, Sun J, Jiang J, Xia Y, et al. Economic burden of dengue fever in china: A retrospective research study. PLoS Negl Trop Dis. 2022;16(5):e0010360.

China Centers for Disease Control and Prevention. https://www.chinacdc.cn/. Accessed 20 July 2025.

Health Commission of Guangdong Province. https://wsjkw.gd.gov.cn/. Accessed 20 July 2025.

Lin YP, Luo Y, Chen Y, Lamers MM, Zhou Q, Yang XH, et al. Clinical and epidemiological features of the 2014 large-scale dengue outbreak in Guangzhou city, China. BMC Infect Dis. 2016;16:102.

Wu T, Wu Z, Li Y. Dengue fever and dengue virus in the people’s Republic of China. Rev Med Virol. 2022;32(1): e2245.

Jiang L, Liu Y, Su W, Liu W, Dong Z, Long Y, et al. Epidemiological and genomic analysis of dengue cases in guangzhou, china, from 2010 to 2019. Sci Rep. 2023;13(1):2161.

Cheng Q, Jing Q, Spear RC, Marshall JM, Yang Z, Gong P. Climate and the timing of imported cases as determinants of the dengue outbreak in guangzhou, 2014: evidence from a mathematical model. PLoS Negl Trop Dis. 2016;10(2):e0004417.

Jing Q, Li Y, Liu J, Jiang L, Chen Z, Su W, et al. Dengue underestimation in Guangzhou, China: evidence of seroprevalence in communities with no reported cases before a large outbreak in 2014. Open Forum Infect Dis. 2019;6(7): ofz256.

Lai S, Huang Z, Zhou H, Anders KL, Perkins TA, Yin W, et al. The changing epidemiology of dengue in china, 1990–2014: a descriptive analysis of 25 years of nationwide surveillance data. BMC Med. 2015;13:100.

Xiong Y, Chen Q. [Epidemiology of dengue fever in China since 1978]. Nan Fang Yi Ke Da Xue Xue Bao. 2014;34(12):1822–5.

Yue Y, Liu Q, Liu X, Wu H, Xu M. Comparative analyses on epidemiological characteristics of dengue fever in Guangdong and yunnan, china, 2004–2018. BMC Public Health. 2021;21(1):1389.

Kularatne SAM. Dengue fever. BMJ. 2015;351:h4661.

Paz-Bailey G, Adams LE, Deen J, Anderson KB, Katzelnick LC. Dengue. Lancet. 2024;403(10427):667–82.

Waggoner JJ, Gresh L, Vargas MJ, Ballesteros G, Tellez Y, Soda KJ, et al. Viremia and clinical presentation in Nicaraguan patients infected with Zika virus, Chikungunya virus, and dengue virus. Clin Infect Dis. 2016;63(12):1584–90.

Waggoner JJ, Balmaseda A, Gresh L, Sahoo MK, Montoya M, Wang C, et al. Homotypic dengue virus reinfections in Nicaraguan children. J Infect Dis. 2016;214(7):986–93.

van der Schaar HM, Wilschut JC, Smit JM. Role of antibodies in controlling dengue virus infection. Immunobiology. 2009;214(7):613–29.

Kok BH, Lim HT, Lim CP, Lai NS, Leow CY, Leow CH. Dengue virus infection - a review of pathogenesis, vaccines, diagnosis and therapy. Virus Res. 2023;324: 199018.

Raafat N, Blacksell SD, Maude RJ. A review of dengue diagnostics and implications for surveillance and control. Trans R Soc Trop Med Hyg. 2019;113(11):653–60.

National Health Commission of the People’s Republic of China. Diagnosis of Dengue Fever (Replacing WS 216–2001). http://www.nhc.gov.cn/wjw/s9491/200802/38819.shtml. Accessed 25 September 2024.

(National Health Commission of the People’s Republic of China. Diagnosis of Dengue Fever, Replacing. WS 216–2008). http://www.nhc.gov.cn/wjw/s9491/201803/d524df26df28453eada8371dc3565818.shtml. Accessed 20 September 2024.

Guangzhou Municipal Territorial Spatial Master Plan. (2021–2035). https://www.gz.gov.cn/zwgk/fggw/szfwj/content/post_9960352.html. Accessed 20 July 2025.

Meacock R, Anselmi L, Kristensen SR, Doran T, Sutton M. Higher mortality rates amongst emergency patients admitted to hospital at weekends reflect a lower probability of admission. J Health Serv Res Policy. 2017;22(1):12–9.

Honeyford K, Cecil E, Lo M, Bottle A, Aylin P. The weekend effect: does hospital mortality differ by day of the week? A systematic review and meta-analysis. BMC Health Serv Res. 2018;18(1):870.

Mitchell D, O’Brien L, Bardoel A, Haines T. Challenges, uncertainties and perceived benefits of providing weekend allied health services-a managers’ perspective. BMC Health Serv Res. 2017;17(1):118.

Wei J, Zhu Z, Qi Q, Zeng L. Patient Delay in Hospital Visiting and the Weekend Effect of Surveillance Report on Hand-Foot-and-Mouth Disease and Epidemic Parotitis in Hanzhong City, China. Can J Infect Dis Med Microbiol. 2020.https://doi.org/10.1155/2020/7081219

Hulten AM, Holmgren K, Bjerkeli P. Work-related stress, reason for consultation and diagnosis-specific sick leave: how do they add up? PLoS One. 2023;18(7): e0288751.

Chen X, Orom H, Hay JL, Waters EA, Schofield E, Li Y, et al. Differences in rural and urban health information access and use. J Rural Health. 2019;35(3):405–17.

Li X, Yang H, Wang H, Liu X. Effect of health education on healthcare-seeking behavior of migrant workers in China. Int J Environ Res Public Health. 2020. https://doi.org/10.3390/ijerph17072344.

Zhang Q, Feng S, Wong I, Ip D, Cowling BJ, Lau E. A population-based study on healthcare-seeking behaviour of persons with symptoms of respiratory and gastrointestinal-related infections in Hong Kong. BMC Public Health. 2020;20(1):402.

Du F, Wang J, Liu Y, Zhou Z, Jin H. Equity in health-seeking behavior of groups using different transportations. Int J Environ Res Public Health. 2022. https://doi.org/10.3390/ijerph19052765.

Esposito S, Rosafio C, Antodaro F, Argentiero A, Bassi M, Becherucci P, et al. Use of telemedicine healthcare systems in pediatric assistance at territorial level: consensus document of the Italian society of telemedicine (SIT), of the Italian society of preventive and social pediatrics (SIPPS), of the Italian society of pediatric primary care (SICuPP), of the Italian federation of pediatric Doctors (FIMP) and of the syndicate of family pediatrician Doctors (SIMPeF). J Pers Med. 2023. https://doi.org/10.3390/jpm13020198.

Cheng C, Humphreys H, Kane B. Transition to telehealth: engaging medical students in telemedicine healthcare delivery. Ir J Med Sci. 2022;191(5):2405–22.

Ellis MJ, Normand H, Gargaro J, Bayley M, Russell K. Current use and future considerations for concussion telemedicine healthcare in Canada. Can J Neurol Sci. 2023;50(2):257–61.

Xie X, Zhou W, Lin L, Fan S, Lin F, Wang L, et al. Internet hospitals in china: Cross-Sectional survey. J Med Internet Res. 2017;19(7):e239.

Xu X, Cai Y, Wu S, Guo J, Yang L, Lan J, et al. Assessment of internet hospitals in China during the COVID-19 pandemic: National Cross-Sectional data analysis study. J Med Internet Res. 2021;23(1):e21825.

Kinung’Hi SM, Mashauri F, Mwanga JR, Nnko SE, Kaatano GM, Malima R, et al. Knowledge, attitudes and practices about malaria among communities: comparing epidemic and non-epidemic prone communities of Muleba district, North-western Tanzania. BMC Public Health. 2010;10:395.

U.S. Centers for Disease Control and Prevention. Gateway to Health Communication. https://www.cdc.gov/healthcommunication/index.html. Accessed 27.09.2024.

China Centers for Disease Control and Prevention. https://www.chinacdc.cn/. Accessed 27.09.2024.

Park HE, Song HY. Factors affecting emerging infectious disease prevention behaviors among young workers based on ecological modeling. Sci Rep. 2025;15(1):9416.

Ali M, Qadir F, Javed S, Khan ZN, Asad S, Hanif B. Factors affecting outpatient cardiac rehabilitation attendance after acute myocardial infarction and coronary revascularization–a local experience. J Pak Med Assoc. 2012;62(4):347–51.

Thorogood M, Coulter A, Jones L, Yudkin P, Muir J, Mant D. Factors affecting response to an invitation to attend for a health check. J Epidemiol Community Health. 1993;47(3):224–8.

Anderson KE, Mcginty EE, Presskreischer R, Barry CL. Reports of forgone medical care among US adults during the initial phase of the COVID-19 pandemic. JAMA Netw Open. 2021;4(1): e2034882.

Rubin R. COVID-19’s crushing effects on medical practices, some of which might not survive. JAMA. 2020;324(4):321–3.

Alexander KE, Brijnath B, Mazza D. Parents’ decision making and access to preventive healthcare for young children: applying andersen’s model. Health Expect. 2015;18(5):1256–69.

Dawkins B, Renwick C, Ensor T, Shinkins B, Jayne D, Meads D. What factors affect patients’ ability to access healthcare? An overview of systematic reviews. Trop Med Int Health. 2021;26(10):1177–88.

Rowe EK, Leo YS, Wong JG, Thein TL, Gan VC, Lee LK, et al. Challenges in dengue fever in the elderly: atypical presentation and risk of severe dengue and hospital-acquired infection [corrected]. PLoS Negl Trop Dis. 2014;8(4):e2777.

Verhagen LM, de Groot R. Dengue in children. J Infect. 2014;69(Suppl 1):S77–86.

Li Y, Luo L, Dong H. Delivering integrated community care for the elderly: a qualitative case study in Southern China. Int J Environ Res Public Health. 2024;21(6): 680.

Wu J, Chen D, Li C, Wang Y. Effect of community-based public health service on health-related quality of life among middle-aged and older adults with chronic diseases in China. BMC Public Health. 2024;24(1):2039.

Ferrer R, Klein WM. Risk perceptions and health behavior. Curr Opin Psychol. 2015;5:85–9.

Ratnapradipa KL, Jadhav S, Kabayundo J, Wang H, Smith LC. Factors associated with delaying medical care: cross-sectional study of Nebraska adults. BMC Health Serv Res. 2023;23(1):118.

Kang L, Li C, Du H. Predictors of medical care delay or avoidance among Chinese adults during the COVID-19 pandemic. Patient Prefer Adherence. 2023;17:3067–80.

Korenman S, Hyson RT. Health insurance, labor market shocks, and mental health during the first year of the COVID-19 crisis. Econ Hum Biol. 2023;51: 101298.