Clinicians-related determinants of anticoagulation therapy and prophylaxis in Nigeria.

BACKGROUND: Thromboembolic and hypercoagulable diseases are common life-threatening but treatable problems in hospital practice. Fortunately, anticoagulation is an efficacious management practice indicated for arterial, venous, and intracardiac thromboembolism. Clinicians in developing countries may have gaps in their knowledge of anticoagulation therapy/prophylaxis which could affect their clinical decision.
OBJECTIVES: The study examined the knowledge and attitude of clinicians to anticoagulation therapy/prophylaxis in some tertiary hospitals in Nigeria.
METHODOLOGY: The study was a multicenter survey. A pretested questionnaire was administered to clinicians in six tertiary hospitals in Southeast Nigeria.
RESULTS: A total of 528 questionnaires were returned by 419 (79.4%) residents and 109 (20.6%) consultants. We observed significant abysmal knowledge and lack of awareness of direct oral anticoagulants (DOACs) among most respondents irrespective of their job grades (P = 0.02, odds ratio [OR] 0.59, 95% confidence interval [CI] 0.38-0.90). Their knowledge of anti-Xa assay as laboratory monitoring tool was also significantly inadequate (P = 0.001, OR 0.23, 95% CI 0.10-0.51). On statement analysis on their attitude to anticoagulation therapy/prophylaxis, "Do you think anticoagulation therapy/prophylaxis is clinically relevant" had the highest mean of 4.60, P = 0.01, and a high degree of agreement; while "Should hospital inpatient with > 3 days admission routinely receive anticoagulation/prophylaxis?" had the lowest mean of 2.27, P = 0.02, and a low degree of agreement.
CONCLUSION: There is the need to upscale knowledge of anticoagulation agents and an attitude change to anticoagulation therapy/prophylaxis, especially on the DOACs through continuing medical education activities in emerging countries such as Nigeria.

INTRODUCTION

Thromboembolic and hypercoagulable diseases are common life-threatening but treatable problems in hospital practice. Fortunately, anticoagulation is an effective management practice indicated for arterial thromboembolism, venous thromboembolism (VTE), and intracardiac thromboembolism. VTE is a common but preventable cause of a considerable morbidity, mortality, and costs among hospitalized patients.[1] Several clinical trials have demonstrated the effectiveness of thromboprophylaxis in reducing the rate of deep vein thrombosis (DVT) and fatal pulmonary embolism (PE) by >60%.[2] Following from these findings, guidelines have been published for the prevention of VTE.[34] Unfortunately, VTE risk continues to be high among hospitalized patients, and this to a large extent is attributable to underutilization of thromboprophylaxis according to data from the ENDORSE study.[5]

Several randomized clinical trials have shown that anticoagulant therapy decreases the incidence of thromboembolism by about 70% in patients with chronic atrial fibrillation (AF).[6789] This benefit was accompanied by a relatively low annual bleeding rate. Consequently, most evidence-based guidelines recommend anticoagulation therapy for chronic AF, with the provision that this must be individualized according to each patient's risk for thromboembolism.[1011] However, anticoagulation therapy tends to be underutilized in chronic AF patients even in Western countries.[12] Physicians’ knowledge and perceived risk of anticoagulants seem to be responsible for the reluctance of a good number of physicians to prescribe this therapy.

The most effective and economical approach to decreasing the burden of VTE is to prevent the development of DVT and PE in patients at high risk, including major orthopedic surgery,[13] moderate and high-risk general surgery patients,[14] and acutely ill hospitalized medical patients.[15] The implementation of international guidelines for the treatment and prevention of VTE remains suboptimal and results in patient morbidity and mortality, with a substantial economic burden.[16] Health-care providers in developing countries such as Nigeria may have significant gaps in their anticoagulation knowledge that could affect their decision to prescribe anticoagulation therapy.[17] Currently, there are no national guidelines on the use of anticoagulation in Nigeria.

The purpose of this present study was to examine the knowledge and attitude of clinicians on anticoagulation in some tertiary hospitals in Nigeria.

METHODOLOGY

The present study is a multicenter cross-sectional survey of use of anticoagulants among clinicians in Southeast Nigeria. A pretested questionnaire was administered to clinicians in six tertiary hospitals in the Southeast of Nigeria; University of Nigeria Teaching Hospital, Enugu; Federal Medical Centre, Umuahia; Federal Teaching Hospital, Abakaliki; Abia State Teaching Hospital, Aba; Nnamdi Azikiwe University Teaching Hospital, Nnewi; Federal Medical Centre, Owerri; and Chukwuemeka Odumegwu Ojukwu Teaching Hospital, Amaku, Awka. The questionnaire was designed to assess the knowledge and attitude of clinicians to anticoagulation therapy. To assess clarity and consistency of questions, a pilot study was done before the final draft was prepared. The questionnaire was administered consecutively on medical doctors in the participating centers.

Respondents were required to fill in their age, gender, number of years since graduating, specialty, or grade and then answer questions on the knowledge and attitude to anticoagulation therapy in their center. The name of the respondents and hospitals was not requested. Ethical approval was obtained from the institutions’ ethics committee.

The Likert scale used was from Grades 1 to 5: 1 strongly disagree, 2 disagree, 3 neutral, 4 agree, and 5 strongly disagree. To determine the agreement degree, three levels were identified (high, medium, and low).

The low-level agreement degree has a mean < 3and P < 0.05%

The medium-level agreement degree has P > 0 when the mean < or > 3

The high-level agreement degree has a mean > 3 and P < 0.05% and this is a favorable advantage.[18]

Descriptive statistics with counts and percentages were used to illustrate the results. The data were also analyzed using Fisher's exact test to detect any association between demographic profiles and anticoagulation knowledge and attitude. A P < 0.05 was considered statistically significant. Statistical Package for the Social Science (SPSS) software, version 18 (SPSS Inc., Chicago, IL, USA) was used.

RESULTS

Demographic characteristics

Clinicians sampled were from the six teaching hospitals located in Southeast Nigeria. Of 528 respondents, there were more males 378 (71.6%) than females 150 (28.4%). The clinicians who practiced for < 5 years were in the majority, i.e., 189 (35.8%), and those with 15–20 years, i.e., 46 (8.7%), were in the minority. The residents were in the majority 419 (79.4%) and the consultants were in the minority 109 (20.6%). Full demographic characteristics of the respondents are presented in Tables 1 and 2.

Table 1

Years of practice and grade of the respondents

	Number (percentage of respondents)
Years of practice of respondents
<5	189 (35.8)
6-10	159 (30.1)
11-15	85 (16.1)
16-20	46 (8.7)
>20	109 (20.6)
Grade of clinicians
Residents	419 (79.4)
Consultants	109 (20.6)

Table 2

Summary of knowledge of anticoagulation agents

Grade	Heparin			Warfarin			Fondaparinux			LMWH			NOACs
	Yes	No	NR	Yes	No	NR	Yes	No	NR	Yes	No	NR	Yes	No	NR
Residents	135	2	-	133	4	-	72	65	-	106	31	-	70	67	-
Consultant	107	2	-	106	3	-	38	71	-	86	23	-	60	49	-
P	0.54			0.32			0.01			0.61			0.02*
OR	0.50			0.49			1.46			0.87			0.59
95% CI	0.96-1.01			0.14-1.65			0.94-2.27			0.52-1.44			0.38-0.90

*P value significant >0.05. NR=Not reported, LMWH=Low molecular weight heparin, OR=Odds ratio, CI=Confidence interval

Knowledge of anticoagulation

We have tabulated summary of the clinicians’ response on questions testing their knowledge of anticoagulation; types anticoagulant agents used, indications for anticoagulation, laboratory monitoring tools, and differences between the conventional anticoagulant agents and the direct oral anticoagulants (DOACs). Their responses were tabulated against job grades [Tables 3–6].

Table 3

Summary of the knowledge of indications of anticoagulation therapy

Indications for anticoagulation	VTE			Prolonged immobilization			Prolonged surgery			Malignancy			Stroke
	Yes	No	NR	Yes	No	NR	Yes	No	NR	Yes	No	NR	Yes	No	NR
Registrars	318	101	-	391	28	-	267	152	-	209	216	-	262	157	-
Consultant	82	27	-	106	3	-	75	34	-	58	51	-	69	40	-
P	0.90			0.16			0.37			0.52			0.91
OR	1.04			0.40			0.80			0.85			0.97
95% CI	0.64-1.70			0.12-1.32			0.51-1.25			0.56-1.23			0.62-1.50

VTE=Venous thromboembolism, NR=Not reported, OR=Odds ratio, CI=Confidence interval

Table 6

Summary of the analysis of respondents’ attitude to anticoagulation therapy

Rank	Variables/statement	Mean	P	Agreement degree
1	Do you think anticoagulation therapy/prophylaxis is clinically important?	4.60	0.013	High
2	Early commencement of anticoagulation therapy in confirmed diagnosis of thrombotic stroke improve patients outcome?	4.05	0.007	High
3	Bleeding commonly complicates anticoagulation therapy?	3.49	0.072	Medium
4	Commencement of anticoagulation therapy should precede confirmation of diagnosis?	2.83	0.016	Low
5	Hospital inpatient with >3 days admission should routinely receive anticoagulation?	2.72	0.015	Low

We observed that most respondents irrespective of their job grades did not know about fondaparinux and the DOAC. There was a significant abysmal knowledge of the DOACs (P = 0.02, odds ratio [OR] 0.59, 95% confidence interval [CI] 0.38–0.90) [Table 3]. Most respondents knew of the different indications for anticoagulation therapy/prophylaxis; this was not statistically significant [Table 4].

Table 4

Summary of the analysis showing respondents knowledge on laboratory monitoring tools for anticoagulant agents

Laboratory monitoring tool	APTT			PT			Anti Xa assay			Dilute thrombin clotting time			Thrombin time
	Yes	No	NR	Yes	No	NR	Yes	No	NR	Yes	No	NR	Yes	No	NR
Registrars	342	77	-	408	37	-	65	354	-	48	261	-	173	246	-
Consultant	87	22	-	92	17	-	12	15	-	14	95	-	49	60	-
P	0.68			0.03*			0.001***			0.53			0.51
OR	1.12			2.04			0.23			1.25			0.86
95% CI	0.92-1.14			1.10-3.78			0.10-0.51			0.66-2.34			0.56-1.32

APTT=Activated partial thromboplastin time, PT=Prothrombin time, NR=Not reported, OR=Odds ratio, CI=Confidence interval

Majority of respondents had a significant knowledge of prothrombin test (P = 0.03*, OR 2.04, 95% CI 1.10–3.78). On the contrary, most also had a significant abysmal knowledge of anti-Xa assay, as a laboratory monitoring tool (P = 0.001***, OR 0.2, 95% CI 0.10–0.51) [Table 5].

Table 5

Summary of the analysis showing respondents knowledge on difference between warfarin and direct oral anticoagulants

Difference between warfarin and DOACs	Warfarin has slow onset of action			Many drug-drug interactions			Routine laboratory monitoring			Has specific antidote
	Yes	No	NR	Yes	No	NR	Yes	No	NR	Yes	No	NR
Registrar	79	207	174	67	227	163	91	180	200	89	112	228
Consultant	9	59		6	65		12	45		10	29
P	0.01*			0.005			0.08			0.03
OR	2.50			3.20			1.90			2.30
95% CI	1.18-5.29			1.32-7.70			0.96-3.76			1.07-4.98

DOACs=Direct oral anticoagulants, NR=Not reported, OR=Odds ratio, CI=Confidence interval

On their knowledge of the difference between warfarin and the DOACs, there was numerous none responses to the questions, but analysis of responses received showed was a significant knowledge of reversal agent for warfarin (P = 0.03, OR 2.30, 95% CI 1.07–4.98). On the other hand, there was abysmal significant knowledge onset of action and multiple drug reactions as major differences between the warfarin and the NOACs (P = 0.01, OR 2.50, 95% CI 1.18–5.29 and P = 0.005, OR 3.20, 95% CI 1.32–7.70, respectively) [Table 6].

Attitude on anticoagulation

Based on results of the statement analysis as summarized in Table 6, the variables were ranked according to the value of their mean. All, except one, variable had P < 0.05. The statement “Do you think anticoagulation therapy/prophylaxis is clinically important” had the highest mean of 4.60 and had a high degree of agreement. The statement “Should hospital inpatient with >3 days admission routinely receive anticoagulation?” had the lowest mean of 2.27 with P = 0.015 had a low degree of agreement.

DISCUSSION

Paucity of data on what the anticoagulation practices are in Nigeria negates any meaningful intervention being implemented. Our study showed that most clinicians knew of and were aware of the conventional anticoagulation agents; unfractionated heparin, warfarin, low molecular weight heparin, and fondaparinux. Our results were not consistent with a Brazilian survey done which showed that most emergency room doctors’ knowledge of oral anticoagulants was low;[19] unfortunately, these groups of doctors were not reflected in our survey. Our findings regarding clinicians’ knowledge and awareness with DOACs are consistent with other studies which had demonstrated similar incomplete knowledge among primary care physicians of guidelines for cardiovascular disease prevention in general.[202122]

With the growing popularity of the DOACs among clinicians as standard of care in different categories of patient needing anticoagulation,[23] our study highlights this shocking evidence why most patients are not receiving these agents simply because some clinicians are not aware they exist or are available/accessible for use in the country. Lack of hospital policies on coagulation or any dedicated anticoagulation clinic at the different centers of the respondents may contribute to this low level of knowledge.

The absence of which encourages the plethora of clinician-specific treatment protocols some of which are still in the era when conventional anticoagulants (heparins and Vitamin K antagonists) were considered the drug treatment of choice for anticoagulation episodes. We observed a dearth of knowledge on the various clinical indications for anticoagulation practice. While majority of respondents agreed that conditions such as VTE, prolonged surgery, immobilization, and stroke required either anticoagulation treatment or prophylaxis, approximately more than half would not use anticoagulation therapy or prophylaxis in patients with malignancies and postoperative periods though this was statistically insignificant. Arguably though some malignancies[24] and postoperative conditions are more thrombogenic than others, again there is a low incidence of thromboembolic events in AF patients.[25] The absence of policies, hospital guidelines, or doctor utilization of available guidelines[2627] may play a major contributory role.

Inability to monitor the use of anticoagulation agents has been documented to be one of the reasons of poor use of anticoagulation agents in resource-poor nations like Nigeria.[28] Majority of the respondents were able to recognize prothrombin time (PT) and activated partial thromboplastin time (APTT) as some laboratory monitoring tools for anticoagulation therapy. This agreed with their response of heparin and warfarin as the highest known group anticoagulation agents among respondents.

On the other hand, most respondents did not know anti-Xa assay and dilute thrombin time are tests that could also be used to monitor anticoagulation use. It could not be determined if the few who knew about these agents knew the exact anticoagulation agents that these tests are used for it was not covered by the questionnaire. Another could be that these tests were not routinely ordered for like PT and APTT. Most treatment protocols and guidelines[29] updated regularly serve are clinicians aid in using the current evidence-based treatment protocol for anticoagulant therapy.

We found a high degree of agreement on the importance of commencement of anticoagulant therapy/prophylaxis as well as the positive impact of early commencement of anticoagulant therapy in stroke patients among clinicians studied. This would suggest that clinicians will be favorably disposed to these variables as policies or guidelines in their various institutions. On the contrary, our studies also showed that they had a low level of agreement on routinely placing hospital inpatients on > 3 days admission on anticoagulation prophylaxis and commencement of anticoagulation therapy preceding diagnosis of VTE. These could be responsible for the undocumented increase in VTE among hospital inpatients as well PE-related mortality resulting from late or no commencement of anticoagulation therapy in hospital in patients. A sharp contrast from studies of hospitals in the United Kingdom which showed a marked reduction in morbidity and mortality resulting from VTE of hospital inpatients.[27]

CONCLUSION

Clearly, the study has shown that there is need for upscale knowledge attitude and practice of the use anticoagulation agents, especially the DOACs through well-articulated continuing medical education educational activities. This could be made more robust to involve other health-care personnel involved in anticoagulation practice. Similar efforts have been used recently to successfully improve adherence to cardiovascular guidelines in primary care.[30] A limitation of this study is the relatively small number of study participants and some subspecialties that were not reflected in this survey. However, the diverse perspectives of respondents provided a rich data set from which we highlighted the deficiencies in our anticoagulation practice. Social desirability bias is a potential limitation given that physicians may not be inclined to discuss their lack of familiarity with new medications.

We attempted to limit this risk by asking open-ended questions. In addition, it is ambiguous whether our findings can be applied across board to other subspecialties not reflected or other geopolitical zones of the country. To address this, a multicenter survey among physicians in our health system would help generate the necessary national data to that would bring out required changes.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.