Knowledge, attitudes, and practices of hemovigilance among doctors in a Sri Lankan tertiary care center

Abstract

BACKGROUND:

Hemovigilance systems are considered a core component of quality control in transfusion practice. Clinicians play a pivotal role in the detection, reporting, investigation, and management of transfusion incidents. It has been demonstrated that improving the health-care workers’ knowledge of hemovigilance invariably increases incident reporting rates. The objective of this study was to determine the knowledge, attitudes, and practice of hemovigilance among medical officers working at Colombo South Teaching Hospital.

MATERIALS AND METHODOLOGY:

A descriptive cross-sectional study was conducted using a self-administered prevalidated questionnaire with open- and closed-ended questions. All the doctors who gave informed written consent were included in the study.

RESULTS:

One hundred and eighty-five medical officers participated in the study, and the response rate was 82.2%. The knowledge of the participants was not satisfactory, especially with regard to the scope (11.9%) and the nonpunitive nature of the hemovigilance system (32.2%). Time constraints were the main factor that discouraged incident reporting (57.6%), and many (58.8%) stated that reporting should be made easy to increase the notification rates. Doctors who had previously been educated on transfusion adverse events were found to have better knowledge. A particularly lower number of respondents had read the national guidelines (2.8%) and annual hemovigilance reports (1.7%).

CONCLUSION:

Knowledge among the participants was unsatisfactory, especially concerning the scope and the nonpunitive voluntary framework of the hemovigilance system. Doctors should be made aware of the hemovigilance-related documentation and should have more access to continuous medical education programs.

Introduction

Blood transfusion is very safe and effective when used appropriately.[1] Nevertheless, even with a clear indication for transfusion, correct sample collection, testing methodology, and administration technique, blood transfusion procedures have an inherent risk of causing unfavorable events which are defined as transfusion reactions.[2] On the other hand, a near-miss event is a deviation from standard operating procedures or practices that is detected before the transfusion, which could have led to an inappropriate transfusion or a transfusion reaction in the recipient.[3]

A hemovigilance system is a continuous and uniform system intended for monitoring, recording, reporting, and evaluating the information on unforeseen and unwanted effects that occur due to the therapeutic use of blood and blood components with the intention of investigating and taking preventive actions to avert their occurrence and recurrence.[4] Hemovigilance systems are now considered a core component of quality control in the blood transfusion chain, and information gathered from a well-functioning hemovigilance system can reflect the success or failures of various measures taken to improve transfusion safety.[5,6]

The National Hemovigilance Unit, which is an integral part of the National Blood Transfusion Service, was established in May 2009 as a voluntary, nonpunitive hemovigilance program, to ensure the quality and safety of blood and blood products.[7] It collects data from all processes in the transfusion chain, including near-miss events. The total number of reported incidents has gradually increased since the conception of the national hemovigilance network. Despite that, the number of reported near-miss events has been consistently low over the years, although they could occur four times more frequently than actual adverse events.[8] This becomes evident when local data are compared with the data from the United Kingdom’s hemovigilance scheme, Serious Hazards of Transfusion[9,10,11] [Table 1].

Table 1.

Comparison of the number of reported near-miss events and transfusion incidents between Sri Lanka and the United Kingdom from 2015-2017

Year	Number of near-miss events		Total incidents reported		Near-miss events as a percentage of total incidents (%)
	UK	Sri Lanka	UK	Sri Lanka	UK	Sri Lanka
2015	1243	9	3288	2744	37.8	0.32
2016	1283	11	3091	2791	41.5	0.39
2017	1359	16	3230	3493	42.07	0.46

UK=United Kingdom

To minimize the risks associated with transfusions, it is important that all physicians who prescribe or administer blood or blood components must be able to detect and manage transfusion reactions, report the incidents, and take the necessary actions to institute protocols to prevent their occurrence and recurrence.[12] Therefore, this was the first ever study in Sri Lanka that was designed to assess the knowledge, attitudes, and practices of hemovigilance among doctors in a tertiary care setting.

The objectives of this study were:

1. To determine the factors that encourage and discourage the reporting of transfusion incidents.

2. To identify the causes of poor near-miss event reporting.

3. To find methods to improve transfusion reaction reporting.

Materials and Methodology

This descriptive cross-sectional study was conducted from April to May 2019 at Colombo South Teaching Hospital, which is the second largest government hospital in Colombo district. Ethical clearance was obtained from the Institutional Ethics Review Committee of the Colombo South Teaching Hospital. A self-administered questionnaire with both open- and closed-ended questions was used as the study tool. The questionnaire consisted of six knowledge-related questions, four attitude-related questions, and ten practice-related questions. Many factors were considered when preparing the questionnaire, such as variation in clinical experience and the spectrum of clinical specialties of medical officers. Expert-driven pretesting of the study instrument was performed at the National Blood Center prior to the research to ensure its clarity and understandability. The sample size was calculated using the equation n = (z)² P (1–p)/d2 where the level of confidence was 95% (z = 1.96), and it was assumed that the proportion (p) was 50% with a margin of error (d) of 7%. All the doctors who regularly prescribed/issued blood and blood products were included in the study, and doctors who were working in units such as dermatology, pathology, and nutrition were excluded from the study. A random consecutive sampling method was used to select participants. Questionnaires were offered to physicians who worked during the day, and those who gave their informed written consent to the study were asked to complete the form immediately and return it. Some of the participants were unable to complete the questionnaire on the same day, and therefore, their questionnaires were collected the following day.

Results

Two hundred and fourteen questionnaires were administered to the doctors. One hundred and eighty-five questionnaires were returned, but only 177 were properly filled out, giving a response rate of 82.7%. There were 95 (53.7%) female respondents and 82 (46.3%) male respondents. Among the study group, there were 25 (14%) intern house officers, 20 (11%) released house officers, 95 (54%) senior house officers, 27 (15%) registrars, and 10 (6%) senior registrars. Due to their busy time schedules, none of the consultants participated in the study. Table 2 shows the distribution of the responders according to their specialty.

Table 2.

Distribution of the participants according to their departments

Department	n (%)
Anesthesiology	24 (13.6)
Blood bank	7 (4.0)
Dialysis	10 (5.6)
ETU	7 (4.0)
Obstetrics/gynecology	29 (16.4)
Hematology	9 (5.1)
General medicine	19 (10.7)
Medical ICU	7 (4.0)
Pediatrics	25 (14.1)
SCBU	8 (4.5)
Surgical ICU	7 (4.0)
Surgery	25 (14.1)

ETU=Ethylenethiourea, ICU=Intensive care unit, SCBU=Special care baby unit

Table 3 summarizes how the participants responded to the questions intended to assess their knowledge. All doctors who participated in the study were aware of the transfusion reaction reporting system, and most of them (95.5%) correctly responded that transfusion reactions should be reported to the hospital blood bank. Nevertheless, only 52 (29.4%) doctors were aware of what a near-miss event was. While the majority of the participants (57.1%) were of the opinion that all transfusion reactions should be reported, 19 (10.7%) and 13 (7.3%) believed that only serious reactions and only proven reactions should be reported, respectively. Only 44 (24.9%) participants knew that, according to the guidelines, all transfusion reactions including near-miss events should be notified.

Table 3.

Respondent’s knowledge of the hemovigilance system

Variable	n (%)
Know about the hemovigilance program
Yes	100 (100)
No	0
Know about near-miss events
Yes	52 (29.4)
No	125 (70.6)
Types of reactions that should be reported
Serious reactions only	19 (10.7)
All transfusion reactions	101 (57.1)
Only proven reactions	13 (7.3)
All reactions and near-miss events	44 (24.9)
Where to report transfusion reactions
Hematology department	0
National hemovigilance unit	7 (4.0)
Hospital blood bank	169 (95.5)
Do not know	1 (0.6)
The legal requirement of reporting transfusion reactions
Mandatory reporting	86 (48.6)
Voluntary reporting	21 (11.9)
Do not know	70 (39.5)
Taking legal actions against personnel who report reactions
Yes	24 (13.6)
No	57 (32.2)
Do not know	96 (54.2)

Although it is not mandated by Sri Lankan law to report transfusion reactions, merely 21 (11.9%) doctors were aware of it. Sri Lankan hemovigilance program, like most in the world, uses a nonpunitive and anonymized approach. Nonetheless, only 57 (32.2%) respondents were informed of this fact, and 24 (13.6%) of them thought the hemovigilance data could be used to take legal actions against the culprits. Another notable feature was that there was a statistically significant difference between the rate of correct responses between doctors who had previously trained in transfusion reactions and those who did not (χ² [1, n = 1062] = 37.5181, P < 0.05).

The attitudes of the respondents toward hemovigilance are shown in Table 4. The main factors that discouraged the notification of transfusion incidents included the considerable amount of time it takes to complete the notification forms (57.2%) and the fear of sending partially completed forms (37.8%). One hundred and four (58.8%) participants stated that reporting should be made easy to increase the notification rates. Almost a similar number of respondents agreed that adequate training in the hemovigilance system (57.1%) and continuous medical education (CME) programs (54.2%) would be beneficial in this regard. Of all participants, 151 (85.3%) agreed that reviewing transfusion incidents in a multidisciplinary meeting would help to reduce reaction rates. The attitude toward near-miss events was unsatisfactory; only 49 (27.7%) doctors agreed that near-miss event reporting should be included in the transfusion reaction reporting system.

Table 4.

Respondent’s attitudes toward hemovigilance

Variable	n (%)
Factors that discouraged the notification of adverse events (multiple responses)
Too busy to actively look for reactions	15 (8.5)
Fear of allegations	54 (30.5)
Concerns report may be wrong	49 (27.7)
Did not know how to report	29 (16.4)
Filling forms takes too much time	102 (57.6)
Apprehension of partially filled forms	67 (37.8)
Do not know how to detect reactions	13 (7.3)
Do not know where to report	0
Methods for improving reaction reporting (multiple responses)
Reporting made easy	104 (58.8)
Remuneration	16 (9.0)
CME	96 (54.2)
Availability of notification forms	73 (41.2)
Electronic forms	34 (19.2)
Provide adequate training	101 (57.1)
Reviewing incidents at multidisciplinary team meetings
Important	151 (85.3)
Not important	26 (14.7)
Importance of near-miss incident reporting
Important	71 (40.1)
Not important	7 (4)
Do not know	99 (55.9)

CME=Continuous medical education

A considerable number of participants had encountered and reported transfusion reactions while performing their duties (70.1%), and the febrile nonhemolytic transfusion reaction (51.4%) was the most common type of reaction they identified [Table 5]. When managing transfusion reactions, most of the respondents sought help from the blood bank (103%–58.2%) or senior members of the clinical team (62%–35%) such as consultants and senior registrars, while others referred to the literature (29%–16.4%) or took advice from their colleagues (28%–15.8%). Twenty (11.3%) respondents did not take any advice on the management of transfusion reactions. A total of 118 (66.7%) medical officers agreed that the definitions in the reaction reporting form were completely understandable, but 26 (14.7%) found it difficult to understand. Thirty-three (18.6%) stated that they do not know the comprehensibility of the definitions, as they had never used or read the reaction reporting form.

Table 5.

Types of reactions that were encountered by the respondents

Type of reaction (multiple responses)	n (%)
Acute hemolytic transfusion reaction	11 (6.2)
Allergy/anaphylaxis	71 (40.1)
Delayed hemolytic transfusion reaction	4 (2.3)
Anaphylaxis	6 (3.4)
Febrile nonhemolytic reaction	91 (51.4)
Hypotensive reactions	16 (9.0)
Transfusion-associated cardiac overload	11 (6.2)

Only five (2.8%) doctors had ever read the national hemovigilance guidelines. The majority of the respondents (104%–58.8%) were not aware whether printed guidelines were available in their departments while 68 (38.4%) reported that the guideline books were not available and only five medical officers (2.8%) stated that they were available. Similarly, a particularly lower number of participants (3%–1.7%) had read the annual hemovigilance report, and Table 6 shows the stated reasons why they had not read it. It is interesting to note that most of them (61%) were unaware of its existence.

Table 6.

Reason for not having read the annual report

Reason (multiple responses)	n (%)
Did not know it was important	17 (9.6)
Never knew about it	108 (61.0)
Not enough time to read	15 (8.5)
Not interested	8 (4.5)
Not available online	10 (5.6)
Not available in the department/ward	29 (16.4)

Discussion

It has been demonstrated that improving the knowledge, attitudes, and practices of health-care workers on transfusion incident reporting and hemovigilance invariably improves the reporting rates.[13] Several studies conducted in developing countries have confirmed that health-care professionals show poor knowledge and attitudes toward hemovigilance.[13,14,15] When considering the data from the current survey, it is apparent that the lack of insight into the importance of reporting could mainly be due to the inadequacy of knowledge among the participants. According to Shivgunde et al.,[13] only 38.88% of the participants were aware of the transfusion incident reporting system in India. However, in the current research, all participants were informed of the local transfusion-related adverse event notification system.

Safety studies have demonstrated that the underlying causes of near-miss incidents are similar to transfusion incidents. Therefore, assessing near-miss events could generate valuable information for prevention and safety programs.[16] Research conducted by Singal et al. determined that although 94.3% of respondents were able to identify a near-miss event, 58.1% did not think near-miss events were reportable. Thus, a health-care worker who accurately recognized a near-miss incident was still more likely to not report the event.[17] By contrast, in the current study, 70.6% of the participants were not aware of what a near-miss event was and this could be one of the reasons for the considerably low number of reported near-misses.

An ideal hemovigilance program should cover all processes throughout the transfusion chain to identify, report, and investigate all incidents and near-miss events. Due to the differences in the spectrum of reporting, the scope of a hemovigilance system varies.[18] In this study, only 24.9% of the study subjects were aware of the correct reporting scope, likely reducing the intended scope of the Sri Lankan hemovigilance program. For the success of a hemovigilance system, standard definitions and terminology are very important, and clear and precise definitions are required to correctly classify and analyze transfusion-related incidents and near-miss events.[19] Twenty-six (14.7%) respondents who participated in the current research stated that they find it difficult to understand the definitions in the incident reporting form.

The main barriers to incident reporting are the fear of errors being used to evaluate the competency of the doctors, organizational or individual repercussions, and the possible legal discovery of error reports. Mandatory and punitive reporting systems may suffer from a lack of in-depth information and reporting rates as medical professionals could withhold some key information for their self-protection and fear of retribution. On the other hand, voluntary nonpunitive reporting systems provide more information on the causative factors of errors than mandatory systems and promote the reporting of near-miss events.[20] Despite the fact that the Sri Lankan hemovigilance system is based on voluntary reporting, a considerable percentage of medical practitioners (48.6%) believed that they were legally obliged to report transfusion reactions, and only 32.2% were informed of the nonpunitive approach. Furthermore, many respondents (57.6%) agreed that the main discouraging factor for reporting transfusion reactions was the considerable amount of time required to complete the notification forms. Similar findings were observed in the study conducted by Date et al.,[14] where 47% of the respondents stated that lack of time was a limiting factor in reaction reporting.

CME is a key component of the health education system, and sensitizing senior clinicians to transfusion reactions and reporting, and regular training sessions for junior physicians and nurses have been shown to improve adverse event reporting rates.[21] The teaching unit of the National Blood Transfusion Service (NBTS) conducts lectures, and other teaching activities related to hemovigilance for blood bank medical officers, nurses, and medical laboratory technologists. However, at the moment, NBTS does not conduct any CME programs for health-care professionals who are not in the field of transfusion medicine. This could probably be a reason why the majority (82.5%) of physicians stated that they had never attended lectures or training sessions on transfusion incident detection, reporting, and management.

Morbidity and mortality meetings provide a platform to identify errors and take necessary actions to prevent them in the future. Studies have shown that structured morbidity and mortality meetings at the departmental level could improve the quality of the services.[22] 85.5% of the participants agreed that discussing transfusion-related morbidity and mortality with the participation of different specialties would be helpful in improving transfusion safety.

In 2009, guidelines on the detection, reporting, and management of transfusion adverse events were first published, and distributed to all hospitals island-wide. The number of respondents who had read the national guidelines was a total disappointment (2.8%). 38.4% of the respondents stated that the handbook is not available in their respective departments, and 58.8% were doubtful about its availability. Similarly, the majority of the respondents (98.3%) were not aware of the annual hemovigilance report published by the National Hemovigilance Unit, which is primarily intended for end users of blood and blood products and serves as the communication medium between the hemovigilance unit and the clinicians to educate them about transfusion hazards and their prevention and to inform the national policy on transfusion safety.

Conclusion

The findings of this study suggest that the knowledge among doctors was unsatisfactory, especially concerning the scope and the nonpunitive voluntary framework of the hemovigilance system. There are multiple factors for the observed knowledge deficiency, including the lack of knowledge concerning documentation such as guidelines and annual hemovigilance reports, and an inadequate number of education programs. This study also revealed that there were considerably limited practices toward transfusion hazard reporting. Education could play a significant role in the resolution of the issue at hand. All clinicians should be made aware of the availability of documentation related to hemovigilance and the importance of reporting near-miss events. In addition, periodic CME sessions focusing on the detection, notification, management, follow-up, and prevention of transfusion reactions should help improve the knowledge deficit.

Conflicts of interest

There are no conflicts of interest.

Funding Statement

Nil.