Survey of knowledge, attitude and practice of healthcare professionals on dengue transmission, diagnosis and clinical classification

Dengue is a mosquito-borne viral disease that encompasses 129 countries around the world [1‐3]. Half of the global population stands at risk of dengue infection and the other half remains in dengue-endemic countries [4‐6]. The world witnesses more than five million dengue infection reported from World Health Organization (WHO) regions in 2019, which is about tenfold higher compared to that in 2000 [3]. All six regions of WHO were affected, however three most seriously affected regions are South-East Asia, Western Pacific Regions, and recently the Region of the Americas.

The infection of dengue starts with a bite of a female mosquito primarily of the species Aedes aegypti and, to a lesser extent, Aedes albopictus, Aedes polynesiensis, and Aedes scutellaris. There are four dengue serotypes: DEN-1, DEN-2, DEN-3, and DEN-4, and symptoms typically appear in four to seven days going through three phases: febrile, critical, and recovery [7]. Most infections by any of these serotypes result in a self-limiting fever; however, age, genetic predisposition, and background immunity can result in different clinical appearances, the most severe of which causes hemodynamic compromise due to plasma leakage [8, 9]. In 2009, WHO classified dengue into three categories: dengue without warning signs, dengue with warning signs and severe dengue to optimize clinical management since the strict application of the 1997 system missed the more severe manifestations [7]. Since a modest proportion of infected patient progress to the severe forms of dengue, it becomes more important than ever for healthcare professionals (HCPs) to recognize the warning signs in clinical practice.

Although the widely accepted WHO 2009 system was more sensitive to the severe manifestations of dengue compared to the 1997 system, it is lacking a clear definition of the warning signs [10]. Healthcare providers and researchers employed different criteria to define the warning signs, which led to diverse application of the guideline for the diagnosis and treatment of dengue patients. Therefore, there is an urgent need for the clear definitions of the warning signs and the severities in order to prevent unnecessary hospitalizations and to optimize triage and management [10‐13]. Few studies have investigated the definitions used by healthcare providers on the field and their subsequent outcomes. This study was conducted in eight hospitals in three different countries to investigate knowledge, attitudes, and practices of the healthcare professionals regarding dengue transmission, diagnosis and clinical classification using the warning signs of WHO 2009 classification.

In 2009, the WHO published a classification of dengue cases, which was attributed to several benefits shown in literature [18]. One of those values was the considerable sensitivity in capturing the severe case [10, 19, 20]. Moreover, a study proved that this classification was easy to apply with its user-friendliness to healthcare providers [10]. Nevertheless, the WHO 2009 classification listed seven warning signs of dengue without providing a precise definition of these signs. These lead to the different application of these signs for the diagnosis and treatment of dengue patients between healthcare personnel.

A previous systematic review showed the wide variation of the definitions and cutoff values used by researchers to classify dengue patients [11]. Likewise, the current study found a wide variation in defining the warning signs among physicians and nurses in clinical settings. Of these seven warning signs, the HCPs have not agreed on any signs, even with the definition of liver enlargement, which was predefined in WHO 2009 classification. This variation may be linked to the local modification of the guidelines, adoption of national guidelines, lack of proper training, and variation in clinical practice in different regions.

A multi-center study in 18 countries concluded that the variations in the definition for warning signs could be observed in different regions [10]. Therefore, guideline adaptation at the local level is needed. Some HCPs denounce the WHO 2009 classification-related symptoms for being vague and associated with another disease [10]. Consequently, if the warning sign does not have a precise definition, the hospitalization rates might rise significantly; therefore, the definition and usefulness of warning signs in different regions, countries and levels of healthcare should be prospectively investigated.

Our results could link the resulting inconsistent definition of the warning signs to the specialties of HCPs (physicians or nurses), country of HCPs, prior received training on dengue fever, and frequency of taking care of dengue patients. However, there are limitations to this study. Because this study is a self-administered survey with given answer choices, the knowledge gap may not be sufficiently addressed compared to a blank questionnaire. But, the current study seems to point out a knowledge gap between nurses and doctors concerning dengue warning signs in this study: for example, more nurses (16.5%) chose “do not know” compared to physicians (3.8%) in choosing a definition for abdominal pain. This holds for most warning signs except for lethargy, which showed more physicians (29.5%) choosing “do not know” versus 25.7% for nurses. Interpretation of lethargy is still many ambiguities, and an objective differentiation from tiredness is lacking might be a possible explanation. Another possibility could be nurses having more contact with patients [21]. Interestingly, nurses and physicians were similar in frequency in choosing “do not know” concerning the definitions for the decrease in platelet count warning sign: 21.1% for physicians versus 19.9% for nurses. One of the possible explanation is HCPs including doctors and nurses have received continuous education on dengue. However, training in health facilities are more likely concentrated on treatment rather than transmission, clinical classification and diagnosis. Doctors and nurses in Vietnam, for example, are supposed to understand how to diagnose and classify dengue cases, therefore continuous education health units will most likely update new points related to treatment. Experience in terms of years in treating dengue does not seem to play that big of a role for the knowledge gap as the average mean of years for nurses is 6.7 years versus 5.3 years for physicians, according to Table 1.

Additionally, the variation amongst the definitions of the warning choices may partly be due to random guessing. Finally, we cannot comment on the geographical differences playing a role as the sample size of doctors and nurses at hospitals in Pakistan and Egypt is too small versus Vietnam, of which this study has the majority (89.8% of total sample size). Furthermore, varying in current dengue situation, public health response and clinical classification system of dengue cases in three countries might have enormous impact on the results. To improve this study, a bigger sample size overall is needed especially in hospitals of Pakistan and Egypt, in which the study can then stratify and analyze data variations by geography or hospitals.

Despite having mostly vague warning criteria, the only specific WHO 2009 classification of “liver enlargement > 2 cm” is recognized by only 57.5% by physicians and 45.3% by nurses of this study. This points out that part of the knowledge gap is not due to just the “vague” WHO 2009 criteria, but also has to do with either the HCPs’ training or something else. Instead, nearly 65% of HCPs chose painful hepatomegaly as a warning sign, and 15% of HCPs says that it depends on the case.

For certain warning signs, it is evident that when HCPs use more definitions that it would increase the overall sensitivity of the WHO 2009 criteria. For example, for the plasma leakage warning sign, the HCPs must be wary for pleural effusion, ascites, edema (face and extremities), or the hemoconcentration. Thus, it is dependent on the healthcare provider’s ability to detect more physical exam signs or abnormal lab values, which means that training, experience, and a HCPs’ knowledge bank plays a significant role in the triaging of a patient. Thus, it would be recommended that hospitals in endemic areas to have more training of HCPs in the knowledge and recognition of dengue patients, as well as have a concise clinical protocol for them to follow.

Recognition of dengue fever might be complicated because not only clinical manifestation of dengue might vary from person to person according to age or gender, but also there may be different defining criteria that vary from country to country defined and agreed upon by different experts, and maybe redefined later due to regional events. For instance, a survey conducted on primary care providers (PCPs) in Singapore presented that 52.2% of PCPs set a platelet cutoff < 80.000 mm³ as the threshold for hospital referral. However, after the epidemic in 2013, there was a significant increase in the PCPs who use < 50.000 mm³ as the cutoff. This number could even be lower, as seen in PCPs from Pakistan, as a survey from 2011 showed that 76% of them use < 20.000 mm³ as the platelet cutoff [22]. In our study, 43.3% of physicians, and 55.5% of nurses used < 100.000 mm³ as the definition criteria for the warning sign of platelet decrease (Table 3). Warning signs were determined based on laboratory tests such as hemoconcentration are more likely to identify by doctors. This may be a reason for a significantly higher proportion of nurses who unknown this sign. It may be that having not a global definition of a warning sign may be beneficial because dengue might appear differently regionally. Also, scientific studies are done regionally, which may affect the current expert consensus of what the definition of a warning sign would be for that region.

Although clinical criteria can be different for different regions, the WHO 2009 guidelines recognize the use of the NS1 strip test for early recognition and diagnosis of acute infection of dengue, and the IgM ELISA test for being specific to diagnose dengue infection. In our study, only 50.9% of HCPs were able to recognize the NS1 strip test, and 42.9% knew about the IgM ELISA as serology specific test to diagnose dengue infection. The majority of physicians preferred dengue serology test to polymerase chain reaction or NS1 antigen test [23]. The availability and affordability of serology tests may explain limited usage of the NS1 strip test. A study in Singapore reported only 11.1% of PCPs used NS1 antigen, but HCPs in private settings utilized NS1 antigen nearly ten times higher than polyclinic counterparts [24]. Our survey was conducted in public settings, where the NS1 antigen might not be affordable, which can result in a knowledge gap among HCPs. Neglecting the NS1 test may hinder the early diagnosis of dengue, as the dengue serology tests have been shown to have low specificity and may be falsely negative during the early febrile phase of illness [25, 26]. Hence, improving knowledge of physician regarding value of dengue diagnostic test is an area of potential improvement based on our findings.

Basic knowledge about dengue (the mosquito, the route of transmission) may not, in the beginning, help with the triage or the clinical management of patients but may be valuable in terms of preventive care for patients and families to avoid primary and secondary infection. In our study, 86% of HCPs chose correctly that the Aedes aegypti mosquito is the primary vector for the dengue infection; however, only 31.6% of HCPs chose correctly that dengue mosquitos are more likely to bite during the daytime. Also, few HCPs were able to identify the different routes of transmission: only 10.4% of HCPs identified mother to fetus or neonatal transmission, 7.2% for blood transfusion and organ transplantation, 3.2% for needle stick, and only 0.6% for sexual intercourse. Thus, there needs to be an improvement of HCPs’ preventive care knowledge in order to reduce dengue infection as well as the HCPs’ attitude in educating the patients and their families. In the current study, 90.2% of HCPs “agree” or “strongly agree” with their responsibility to talk to the patients’ family about dengue virus transmission and prevention. However, only 75.9% of HCPs during outpatient regularly recommend measures to prevent mosquito bites at home, and 73.1% advise on measures to prevent dengue fever for other people living at home with the patient. Hence, increasing HCPs’ knowledge of Aedes mosquito, the different routes of dengue transmission, and knowledge of the biting time of mosquitos is another area for improvement based on our survey.

Our survey has been conducted in three countries: Vietnam, Egypt, and Pakistan. In Vietnam, the dengue threat increased gradually from killing 32.5 people out of every 100.000 in 2000 to kill about 120 people out of every 100.000 in 2009. This study is valuable because Vietnam carries the highest burden of dengue infection in the Asia sub-region, especially the south of Vietnam [27, 28]. Besides, the majority of HCPs of this survey was in Vietnam, which makes this survey especially valuable. Egypt had a recent outbreak in November 2015 in the Dayrout District of Assiut Governorate [29, 30]. Pakistan experienced an epidemic of dengue fever since 2010, which so far infected 16.580 and killed 257 people in Lahore [29, 31]. Therefore, the HCPs working in those three countries should represent a sample that is the potential to deal with dengue patients, which urge the investigation of their knowledge, attitude, and practice.

Again, this study may be limited by the variation in sample size among the three countries, where we conducted our survey, with the vast majority of respondents were in Vietnam. Therefore, the results of this current study should be interpreted in the context of response bias. Lack of sample size calculation is another limitation of our study. Besides, our results should be considered the gap of demographic cofactors and the state of the dengue epidemic in each country. To improve the study later, we are increasing more investigations with a larger sample size to identify the error precisely and fix it by the healthcare authorities and organizations.

The current study highlighted the wide variation in the application of all warning signs listed in WHO 2009 classification and significant gaps in knowledge about dengue diagnosis test, Aedes aegypti and dengue rout of transmission among HCPs. These results provide a valuable opportunity to identify areas of dengue management that need improvement. Interventions should focus on increasing knowledge of HCPs regarding dengue transmission, clinical classification and diagnosis.