Impact of Drug Use Policy on the Appropriate Use of Direct Acting Antiviral Agents for Hepatitis C in Saudi Arabia.

BACKGROUND: Ministry of National Guard-Health Affairs in Saudi Arabia developed a new policy for the use of direct antiviral agents (DAAs) for hepatitis C. The present study was conducted to evaluate prescribers' compliance and the impact of the policy on DAAs appropriate use.
MATERIALS AND METHODS: This study was conducted at King Abdul Aziz Medical City in Jeddah, Saudi Arabia. The study compares patients' data during 1 year before and 1 year after policy initiation. The primary outcomes were compliance to monitoring parameters, appropriateness of treatment and treatment eligibility. Secondary outcomes included sustained virologic response at 12 weeks, documentation of potential drug-drug interactions and treatment costs. Independent samples t-test and Chi-square test were used when applicable. A P < 0.05 was considered statistically significant.
RESULTS: One hundred and three patients were included in analysis (46 before and 57 after policy). Prescriber compliance to baseline monitoring parameters was 67.4% before policy and 82.5% after-policy (P = 0.076). International normalized ratio (INR) was requested in 84.8% of cases before policy compared to 96.5% after-policy (P = 0.036). Treatment options offered to patients were appropriate in 52.2% of cases before policy and in 82.5% after-policy (P = 0.001).
CONCLUSION: There is a significant improvement in the baseline monitoring of INR. Treatment options offered after policy implementation were significantly more appropriate. Copyright:

INTRODUCTION

Worldwide, the prevalence of hepatitis C infection has increased over the past two decades with over 185 million infections.[1] The World Health Organization plans to eradicate hepatitis C virus (HCV) infection by 2030.[2] In Saudi Arabia, blood donor screening centers indicate infection rates of 0.4%–1.1%, whereas a recent systematic review provided an estimation of 1.0%–1.9%.[3]

Hepatitis C was traditionally treated by pegylated interferon alpha and ribavirin. These two medications were successful in patients with HCV genotypes 2 and 3 and less effective in patients with genotypes 1 and 4 where cure rate was around 40%–50%. Pegylated interferon and ribavirin have contraindications and multiple and severe side effects such as fever, headache, muscle ache, skin rash, depression, and anemia as well as limitation in their use in some HCV-induced liver diseases.[4]

When it comes to the worldwide distribution of genotypes, evidence indicated that genotypes 1 and 4 represent a common epidemiologic pattern and represent a significant risk in the Middle East and North Africa (MENA) and the European countries that border the Mediterranean Sea. A systematic review of HCV genotypes in nine selected European countries indicated that genotype 1 is the most common ranging from 80% in Croatia to 46% in Greece. Genotype 4 which is the third ranked shows an increase of 1.1% compared to 2014 data.[5] A recent report on the key associations for HCV genotypes in MENA indicated that genotype 1 was the dominant type across countries in the region and was more present in high-risk clinical populations than in the general population. Genotype 3 was much more prevalent in Afghanistan, Iran, and Pakistan than the rest of countries. Genotype 4 was largely distributed in Egypt in all populations.[6]

HCV infection is treated according to international guidelines issued by the infectious diseases society of America, the American Society for the Study of Liver Diseases, and The International Antiviral Society (USA). These guidelines prioritized patients' treatment into categories based on urgency of need, especially upon the emergence of the new direct antiviral agents (DAAs). The highest priority was given to patients with advanced fibrosis, compensated cirrhosis, extrahepatic hepatitis, and liver transplant patients.[7] After the approval of new DAAs by the United States Food and Drug Administration from May 2011 to June 2016, HCV has become a curable disease. This progress has modified the traditional trend of therapy and made treatment options more affordable for physicians. Treatment regimens are now of shorter duration, less difficult to tolerate, and more effective.[8] The introduction of these new treatment options can result in greater reduction of prevalence rates.[9]

DAAs are relatively safe and highly efficacious medications in eradicating HCV infection. However, the course of treatment for each patient with DAAs is highly expensive. Evidence shown that initiation of treatment protocols and policies and the adherence to them improves patient outcomes in a variety of health conditions.[101112] The Ministry of National Guard– Health Affairs in Saudi Arabia approved a policy for the use of all available hepatitis C medications on November 08, 2015.[13] The policy was developed to regulate and justify DAAs uses and to prioritize patients in receiving treatment. The policy identified treatment eligibility and priority along with recommendations for monitoring efficacy and toxicity. Before the policy, there was no previous approved local protocol or drug-use recommendations. Individual physicians had a liberty to prescribe HCV treatment regimens based on availability or preference. The NGHA-Western Region-King Abdul Aziz Medical City (KAMC), National Guard-a tertiary hospital in Jeddah has been treating hepatitis C patients before and after the approval of the policy. The policy has been introduced in the medical city since December 1, 2015, and according to it, patients are prioritized as follows:[13]

Treatment is indicated for chronic hepatitis C treatment naive and experienced patients with compensated and decompensated liver disease

Treatment should be prioritized to patients with significant fibrosis (F3) or cirrhosis (F4), including decompensated cirrhosis, human immunodeficiency virus (HIV) coinfection, hepatitis B virus coinfection, patients with an indication for liver transplantation, HCV recurrence after liver or kidney transplantation, clinically significant extra-hepatic manifestations, and risk of transmitting HCV

Treatment is justified for patients with moderate fibrosis (F2) and patients seeking treatment to get employed or to get married

Treatment can be deferred to patients with no or mild disease (F0-F1)

Treatment is not recommended for patients with limited life expectancy (≤1 year) due to nonliver-related comorbidities and patients who are ≥75 years old with many comorbidities.

Aim of the study

The present study was conducted to evaluate the impact of new policy on the utilization of the new DAAs and prescribers' compliance to the policy recommendations, the value of the policy regarding achieving sustained virologic response at 12 weeks (SVR12), cost savings, as well as documentation of potential drug–drug interactions that patients might experience.

MATERIALS AND METHODS

Study design and setting

This is a retrospective study conducted at KAMC in Jeddah, Saudi Arabia. The study compares treatment courses and outcomes among patients treated at KAMC during 1 year before and 1 year after policy initiation.

Inclusion and exclusion criteria

Adult patients 18 years of age and older living with hepatitis C who were treated at KAMC in Jeddah from December 1, 2014 to November 30, 2016 were included. Patients were excluded if they received interferon-based regimen, or their data were incomplete. All consecutive patients who fulfilled the inclusion and exclusion criteria within the study defined timelines were included. Patients were categorized into two groups according to the year, in which they received treatment; either 1 year before or 1 year after the implementation of the drug use policy initiated on December 1, 2015.

Data collection and source

Pharmacy records and electronic files were used to collect data. Studied variables included demographics, comorbidities (liver cirrhosis, liver transplant, kidney transplant, diabetes mellitus (DM), hypertension, renal impairment, hepatitis B co-infection, HIV co-infection, cryoglobulinemia, thalassemia, and sickle cell anemia), laboratory values, monitoring indicators (viral load, Metavir Score, liver function tests [LFTs], renal profile, complete blood count [CBC], and international normalized ratio [INR]), medications history, and treatment cost.

The primary outcome

The primary outcome was compliance to the drug-use policy and the policy impact on the new DAAs appropriate use and this is defined by three determinants as follows:

Compliance to monitoring parameters which was subdivided into

Compliance to baseline monitoring parameters (baseline viral load, metavir score, LFTs, renal profile, CBC and INR)

Overall compliance to monitoring parameters (all viral load, all LFTs, all renal profile, all INR, all CBC, all metavir score)

Eligibility for treatment as stated by the policy for provision of DAAs to patients (prioritized, justified, could be deferred, not recommended).

Appropriateness of treatment

When a prescriber requests blood investigation (e.g. LFTs, serum creatinine [Scr], INR, CBC and Viral load) at baseline, then after 4 weeks, then at 12 or 24 weeks of therapy, then at week 12 or 24 after completion of therapy as well as baseline metavir score for all patients, the prescriber is considered compliant to the policy's monitoring recommendations. Metavir score is a system used to assess the extent of inflammation and fibrosis of liver. Viral load is the amount of viral genetic material in the blood. Treatment was considered appropriate if the medication selection is adherent to the guidelines used in the policy.

The secondary outcomes

Secondary outcomes included were:

SVR12

Documentation of potential drug–drug interactions

Treatment costs before and after policy implementation.

For drug–drug interactions, any C or D drug–drug interaction category was included. Decision on interactions was made by checking the concomitant documented patient medications using Lexi-comp drug reference. For example, ledipasvir + esomeprazole yields category D interaction. Hence, any combination that yields an interaction category was documented in patient chart or soft file, it was considered as a potential interaction. The rate of documented interactions was calculated from those potential interactions.

Ethics approval

The study was approved by the hospital Institutional Review Board of KAMC in Jeddah (Memo Ref. No. IRBC/1218/17; E-CTS Ref. No. RYD-17-417780-143022).

Cost calculation

For calculation of cost, the total prices of antiviral agents used for all patients throughout their treatment courses before policy were calculated and compared with that after-policy. First, the total cost for all patients in each group was obtained by calculating the sum of the acquisition price of each patient DAAs, then the mean price (i.e., per patient) was obtained. The treatment costs were compared for patients whose treatment could be deferred but they received treatment, and differences were considered as potential cost savings.

Statistical analysis

Descriptive and inferential analyses were performed using Statistical Package for the Social Sciences by IBM V. 24 (IBM SPSS Statistics for Windows, Armonk, NY, US: IBM Corp). Descriptive analyses included calculation of mean and standard deviation, frequency and percentage and cross-tabulations. Statistical tests used for inferential analyses were independent samples t-test, and Chi-square test when applicable. A P < 0.05 was considered statistically significant.

RESULTS

Patients demographic and basic characteristics

During the 2 years, 112 patients received DAAs treatment and they were all eligible for inclusion. Patients who received interferon-based regimen were not included in the study. Fifty patients were treated before policy and sixty-two after the policy. Of those 112 patients, four patients were excluded from the before group and five from the after group due to incomplete data. One hundred and three was the final number included in analysis (46 [44.66%] before policy implementation and 57 [55.33%] after policy implementation).

Table 1 shows baseline characteristics of patients in the two groups. Before policy, patients' mean age was 56.4 years with a standard deviation of 12.7 years. Their mean weight was 76.6 kg and mean height was 161.3 cm. The majority of patients were females representing 67.4%. Genotype 4 and baseline fibrosis stage (Metavir score) F4 were the most common (52.2% and 45.7% respectively). After-policy, patients' mean age was 58.2 years with a standard deviation of 14.2 years. Their mean weight was 69.8 kg and mean height was 156.1 cm. The majority of patients (64.9%) were females. Genotype 4 and baseline fibrosis stage (Metavir score) F4 were the most common also (54.4% and 43.9% respectively).

Table 1

Baseline characteristics

	Before policy, n (%)	After policy, n (%)
Total number of patients	46	57
Age, mean±SD	56.4±12.7	58.2±14.2
Weight, mean±SD	76.6±17.4	69.8±19.4
Height, mean±SD	161.3±9.4	156.1±10.1
Male	15 (32.6)	20 (35.1)
Female	31 (67.4)	37 (64.9)
Genotype
1	0	0
1a	14 (30.4)	13 (22.8)
1b	7 (15.2)	10 (17.5)
2	0	0
3	1 (2.2)	2 (3.5)
3b	0	1 (1.8)
4	24 (52.2)	31 (54.4)
Baseline metavir score
F0	3 (6.5)	2 (3.5)
F0-F1	2 (4.3)	0
F1	3 (6.5)	3 (5.3)
F1-F2	1 (2.2)	3 (5.3)
F2	0	2 (3.5)
F2-3	1 (2.2)	2 (3.5)
F3	7 (15.2)	13 (22.8)
F3-4	0	1 (1.8)
F4	21 (45.7)	25 (43.9)
ND	8 (17.4)	6 (10.5)
Cirrhosis	31 (67.4)	30 (52.6)
Liver transplant	2 (4.3)	2 (3.5)
Kidney transplant	3 (6.5)	1 (1.8)
DM	20 (43.5)	23 (40.4)
HTN	15 (32.6)	28 (49.1)
Renal impairment	5 (10.9)	10 (17.5)
HBV	2 (4.3)	1 (1.8)
HIV	0	0
Cryoglobulinemia	1 (2.2)	0
Thalassemia	0	0
Sickle cell anemia	1 (2.2)	1 (1.8)

ND: Not documented, SD: Standard deviation, DM: Diabetes mellitus, HBV: Hepatitis B virus, HIV: Human immunodeficiency virus, HTN: Hypertension

The majority of patients before policy and after-policy were cirrhotic (67.4% and 52.6%, respectively). Substantial proportions of patients (43.5% before policy and 40.4% after-policy) reported having DM. Hypertension was reported slightly higher after-policy (49.1%) than before policy (32.6%). There was no widespread prevalence of liver and kidney transplant among both study samples and neither HIV nor thalassemia were reported [Table 1].

Primary outcomes

Compliance to monitoring parameters

Compliance to baseline monitoring parameters (baseline viral load, metavir score, liver function tests, renal profile, complete blood count, and international normalized ratio)

Prescribers' compliance to baseline parameters' monitoring was 67.4% before policy and 82.5% after-policy [Table 2]. However, such differences were statistically not significant.

Table 2

Compliance to monitoring parameters

	Before policy, n (%)	After policy, n (%)	χ2, P
Baseline monitoring parameters
Compliant	31 (67.4)	47 (82.5)	3.143, 0.076
Noncompliant	15 (32.6)	10 (17.5)
Viral load
Compliant	45 (97.8)	57 (100.0)	1.251, 0.263
Noncompliant	1 (2.2)	0
Metavir score
Compliant	38 (82.6)	57 (89.5)	1.021, 0.312
Noncompliant	8 (17.4)	6 (10.5)
Renal profile (scr)
Compliant	46 (100.0)	56 (98.2)	0.815, 0.367
Noncompliant	0	1 (1.8)
CBC
Compliant	46 (100.0)	57 (100.0)	-
Noncompliant	0	0
LFTs (ALT, AST, ALP, and T. Bili)
Compliant	46 (100.0)	55 (96.5)	1.646, 0.200
Noncompliant	0	2 (3.5)
INR
Compliant	39 (84.8)	55 (96.5)	4.376, 0.036
Noncompliant	7 (15.2)	2 (3.5)
Monitoring of all parametersa
Compliant	3 (6.5)	8 (14.0)	1.506, 0.220
Noncompliant	43 (93.5)	49 (86.0)

aAt baseline+at 4 weeks+at 12 and/or 24 weeks+at week 12 or 24 after completion of therapy. CBC: Complete blood count, LFTs: Liver function test, INR: International normalized ratio, Scr: Serum creatinine, ALT: Alanine transaminase, AST: Aspartate transaminase, ALP: Alkaline phosphatase

All prescribers after-policy (100%) were complaint to baseline viral load monitoring compared to 97.8 before policy. Compliance to baseline Metavir score calculation was 89.5% after-policy compared to 82.6% before policy. Compliance to Scr monitoring was 100% before policy compared to 98.2% after-policy. Compliance to CBC request was 100% in both groups. Prescriber's compliance to requesting LFTs was 100% before policy compared to 96.5% after-policy, while requesting INR was 84.8% before policy compared to 96.5% after-policy [Table 2]. Only INR-related difference was statistically significant.

Overall compliance to monitoring parameters (all requests of viral load, liver function tests, renal profile, international normalized ratio, complete blood count, and metavir score)

The compliance to the monitoring of all parameters was seen in only 6.5% of patients before policy compared to 14% after-policy [Table 2]. However, such difference was statistically not significant.

Appropriateness of treatment

Table 3 shows that 52.2% of the treatment options offered to patients before policy were appropriate, compared to 82.5% after-policy. Those differences were statistically significant.

Table 3

Appropriateness of treatment and eligibility

Before policy, n (%)	After policy, n (%)	Significance (χ2, P)
Appropriateness of treatmenta
Appropriate	24 (52.2)	47 (82.5)	10.900, 0.001
Inappropriate	22 (47.8)	10 (17.5)
Eligibility
Prioritized	37 (80.4)	49 (86.0)	1.055, 0.590
Justified	3 (6.5)	4 (7.0)
Could be deferred	6 (13.0)	4 (7.0)
Not recommended	0	0

aTypical to the same guidelines used in the policy.

Eligibility

Table 3 shows that 80.4% of patients before policy were treated with priority compared to 86% after-policy. The treatment of 6.5% of patients before policy was justified compared to 7.0% after-policy, whereas treatment could be deferred for 13.0% of patients before policy compared to 7.0% after-policy. No patient in the two groups received nonrecommended treatment. Such differences were statistically not significant.

Secondary outcomes

Sustained virologic response at 12 weeks

Table 4 shows that 96.5% of patients after-policy achieved SVR12 compared to 91.3% before policy and only one patient (1.8%) achieved SVR at a time beyond 12 weeks after-policy. However, such differences were statistically not significant.

Table 4

Secondary outcomes

Before policy, n (%)	After policy, n (%)	Significance (χ2, P)
SVR12
Achieved	42 (91.3)	55 (96.5)	3.406, 0.182
Not achieved	4 (8.7)	1 (1.8)
Achieved at a time beyond 12 weeks	0	1 (1.8)
Potential drug-drug interactions documentation
	Before policy, n (%)	After policy, n (%)	Significance (χ2, P)
Potential interaction	20 (43.5)	22 (38.6)
Documented	1 (5.0)	0	1.073, 0.300
Treatment opportunity cost
	Before policy, n (%)	After policy, n (%)	Significance (t, P)
Number of patients	46	57	-
Total cost (mean±SD)	SAR 11,099,622 (241,296.1±49,555.1)	SAR 16,455,491 (288,692.8±149,513.5)	−2.059, 0.042
Number of patients whose treatment could be deferred	6 (13.0)	4 (7.0%)	-
Potential cost saving (mean±SD)	SAR 1,372,896 (228,816±5,761.2)	SAR 1,359,624 (224,280±2,521.9)	1.461, 0.182
Potential cost saving difference	SAR1,372,896 - SAR 1,359,624=SAR 13,272		-

SD: Standard deviation, SVR12: Sustained virologic response at 12 weeks, SAR: Saudi arabia riyal

Documentation of potential drug-drug interactions

Table 4 shows that 43.5% of patients before policy had a potential drug–drug interaction but only 5.0% of them were documented. On the other hand, 38.6% of patients after-policy had a potential drug–drug interaction and none were documented. However, such differences were statistically not significant.

Treatment opportunity cost

The results showed that the total cost of treatment before policy (46 patients) was 11,099,622 SAR and the total cost of treatment after-policy (57 patients) was 16,455,491 SAR. The mean cost before policy was 241,296.13 SAR compared to SAR 288,692.8 after-policy. Such differences were statistically significant.

Only 13% of patients before the policy could have their treatment deferred and the total potential cost savings would be SAR 1,372,896 with a mean of 228,816 SAR. Moreover, 7.0% of patients after the policy could have their treatment deferred and the total potential cost savings would be SAR 1,359,624 with a mean of SAR 224,280. There was no significant statistical difference between both groups in terms of potential cost savings (P = 0.182) [Table 4]. This finding indicates a non-significant reduction in opportunity cost.

DISCUSSION

Hepatitis C infection became a curable disease after the introduction of DAAs. For example, a high cure rates were achieved with increasing access to DAAs in England.[14] DAAs even improve the outcomes associated with viral infection complications such as liver failure and liver transplantation. According to Axelrod et al., liver transplant recipients treated for HCV before the introduction of DAAs experienced higher rates of graft loss and death, whereas post-DAAs treatment was not associated with death or graft failure among recipients.[15] According to Xie et al., a universal DAA adoption would reduce the cases of incident decompensated cirrhosis, hepatocellular carcinoma, liver transplants, and liver-related deaths by 61%, 45%, 50%, and 61%, respectively.[16]

However, these agents are highly expensive which made some institutions unable to afford them or adopt restricting policies regulating their use. In Saudi Arabia, Ministry of National Guard-Health Affairs is among institutions decided to adopt DAAs drug-use policy. The policy is expected to have a good impact on treatment priority, keeping the rights of patients in treatment chances and treating them according to their clinical and social conditions by stated criteria clearly mentioned in the policy. According to Chehl, et al., HCV compliance and treatment success rates are higher with DAAs in structured HCV clinics compared to general hepatology clinics.[17]

The current study was not looking merely for the effectiveness of DAAs in treating HCV which has already been established elsewhere.[18] The study was assessing the impact of having an implemented well-enforced drug policy aimed at rationalizing and controlling the use of DAAs in an organized health facility. The study also aimed at assessing prescribers' compliance to the policy recommendations, the value of the policy regarding achieving SVR12, cost savings, as well as documentation of potential drug-drug interactions that patients might experience.

The present study indicated an overall adherence to the implemented DAAs policy in the current setting and preliminary positive outcomes. This is evident from the appropriateness of treatment options after policy implementation. Drug selection after policy implementation was according to the guidelines stated in the policy which reflects a good compliance of prescribers in this regard. Adherence to the treatment guidelines is generally known to improve the outcomes as shown in a retrospective cohort study among patients with chronic HCV infection at the Philadelphia Veterans Affairs Medical Center.[19] The study indicated that adherence to the traditional HCV therapy was positively impacting the HCV suppression and the viral load as well. Early SVR was reported in highly adherent patients as shown in a cohort study evaluating the relationship between adherence to HCV therapy and early and sustained virologic response.[20] DAAs are well known to have excellent efficacy in treating HCV infection, however failure to eradicate the virus was reported in some cases and poor adherence was one of the main causes of failure.[21] The lack of adherence to guidelines is well documented in the literature and warrants further continuous education, dissemination of guidelines, and feedback to prescribers.[22]

In the current study, there were slight nonsignificant improvements in the compliance to baseline monitoring parameters, eligibility for treatment, and in the SVR attainment. Compliance to in-treatment monitoring was low before and after policy implementation but improved slightly with a non-significant difference after policy. Although increases were seen overall as a general trend, most of the monitored parameters before and after policy were not significantly different. This is likely because the sample size was not large enough to detect such differences. Given that, in most cases, compliance rates of monitoring were already high before policy, it is difficult to see significantly large effect difference (i.e., effect size) due to policy. Examples of this are compliance to baseline monitoring of viral load, Metavir score, renal profile, and CBC where compliance rates were already high before policy. However, there was a significant improvement in the baseline monitoring of INR. In most cases, differences were in favor of the policy (including monitoring of all parameters) which indicate an overall preliminary improvement that could have been statistically significant if a larger sample size was taken.

Hashim et al. evaluated the effectiveness of DAAs for HCV-infected patients in Saudi Arabia using data from private health-care sector.[18] They reported an overall HCV cure rate of 97% defined as achieving a sustained viral response (SVR) rate at least 12 weeks following completion of treatment. The patients in our study achieved an overall SVR of 91.3% before policy implementation and 96.5% after policy.

In the current study, poor documentation of potential DDIs was reported both before and after policy which warrants promoting pharmacovigilance practice and awareness activities. According to Schulte et al., drug-drug interactions during antiviral treatment with DAAs still affect about 40% of HCV patients.[23] The authors concluded that the lower DDI potential of modern DAA regimens is partly counteracted by changing patient characteristics. Therefore, DDIs should not be underestimated.

In the present study, the mean treatment cost has increased significantly after policy a thing might be due to the selection of medications with higher prices. The higher cost might be due to the big variation in the prices of DAAs. For example, the acquisition price of one tablet Dasabuvir was 104 SR, while one tablet Sofosbuvir was 2750 SR. This variation explains also the extremely high SD values for the cost which was almost 50% of the mean. However, such finding does not indicate a non-cost-effectiveness of implemented policy. There is a need to conduct a cost-effectiveness analysis taking into consideration all types of cost spent or saved including medical costs other than that of medications as well as nonmedical costs. In our study, despite an overall increase in the cost, finding indicates a nonsignificant reduction in opportunity cost after policy implementation as there was a nonsignificant reduction in the potential cost savings. Health economists aim at reducing opportunity cost a thing indicates economic efficiency in the drugs selections. Ruggeri and Romano evaluated the cost-effectiveness of health policies based on the second-generation DAA-interferon free regimens in an Italian population using real world data and showed cost-effectiveness of the policies.[24] Heffernan and associates in China found that switching to DAAs is cost effective 94% of the time.[25] In India, Chaillon et al. found that treating all diagnosed individuals with moderate-sever liver disease remained cost-saving compared to no treatment with all reinfection rates evaluated.[26] According to Martin et al. treating people who inject drugs with moderate or mild HCV with interferon-free DAAs is cost-effective compared to delay until cirrhosis, except when chronic HCV prevalence and reinfection risk is very high.[27]

Strengths of the study

The study was conducted at a well-organized facility which is stricter to policies and procedures, incomplete data were excluded, and both electronic and nonelectronic patient profiles were used in retrieving data.

Limitations

The present study has some limitations. This includes the relatively small sample size which prevented the detection of significant differences in the observations between the year before and that after policy implementation. The retrospective nature of the study is another limitation which resulted in some missing information. This has caused the investigators to exclude nine records (4 from before and 5 from after the policy). Data was collected primarily from the institution electronic system, but some data was taken from patient's charts. Introduction of the new electronic system did not include all patient information included in the charts. Problems with appointment booking vacancy may have contributed to inability of both prescribers and patients to comply strictly to the policy.

CONCLUSION

The policy has improved the prescriber's compliance to monitoring HCV patients' parameters although most increases were not significant. There was a significant improvement in the baseline monitoring of INR and the treatment options offered after policy implementation were significantly more appropriate. Further efforts need to be applied to increase compliance with the policy. In general, the policy has improved the proper utilization of DAAs and has had a good impact on their use at our institution.

Financial support and sponsorship

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Conflicts of interest

There are no conflicts of interest.