Health economic evaluation of 2-dose and 3-dose rotavirus vaccines in children below 5 years of age in Morocco

ABSTRACT

Following the introduction of rotavirus vaccination into the Moroccan National Immunization Program, the prevalence of the disease has decreased by nearly 50%. However, evidence on the economic value of rotavirus vaccinations in Morocco is limited. This health economic analysis evaluated, from both country payer and societal perspectives, the costs and the cost-effectiveness of three rotavirus vaccines using a static, deterministic, population model in children aged < 5 years in Morocco. Included vaccines were HRV (2-dose schedule), HBRV (3-dose schedule) and BRV-PV 1-dose vial (3-dose schedule). One-way and probabilistic sensitivity analyses were conducted to assess the impact of uncertainty in model inputs. The model predicted that vaccination with HRV was estimated to result in fewer rotavirus gastroenteritis events (−194 homecare events, −57 medical visits, −8 hospitalizations) versus the 3-dose vaccines, translating into 7 discounted quality-adjusted life years gained over the model time horizon. HRV was associated with lower costs versus HBRV from both the country payer (−$1.8 M) and societal (−$4.1 M) perspectives, and versus BRV-PV 1-dose vial from the societal perspective (−$187,000), dominating those options in the cost-effectiveness analysis. However, costs of BRV-PV 1-dose vial were lower than HRV from the payer perspective, resulting in an ICER of approximately $328,376 per QALY, above the assumed cost effectiveness threshold of $3,500. Vaccination with a 2-dose schedule of HRV may be a cost-saving option and could lead to better health outcomes for children in Morocco versus 3-dose schedule rotavirus vaccines.

GRAPHICAL ABSTRACT

KEYWORDS:

• Cost-effectiveness analysis
• health economic evaluation
• rotavirus
• rotavirus gastroenteritis
• vaccination

Introduction

Rotavirus infection is responsible for approximately 40% of diarrheal-related hospitalizations and approximately 30% of diarrheal-related deaths in children aged >5 years old worldwide.^1–3 Rotavirus causes acute diarrhea which can lead to severe dehydration, hospitalization, and death.^3,4 The burden of disease is especially high in low- and lower-middle-income countries where access to safe and clean water, adequate sanitation, and emergency healthcare may be more limited.¹

One of the effective measures to prevent rotavirus is through vaccination.⁴ Many countries adopted rotavirus vaccines into their National Immunization Programs (NIP) following recommendation by the World Health Organization (WHO) in 2009.⁵ These vaccines were human rotavirus, live, attenuated, oral vaccine (HRV; Rotarix; GSK) and rotavirus vaccine, live, oral, pentavalent (HBRV; RotaTeq; Merck).⁵ In 2018, two additional rotavirus vaccines were pre-qualified by the WHO: rotavirus vaccine, live attenuated oral, freeze-dried (BRV-PV 1-dose vial; Rotasiil; Serum Institute) and live, attenuated, oral rotavirus 116E (ORV 116E; Rotavac; Bharat Biotech).⁵

In Morocco, HRV was implemented into the NIP in 2010, rapidly reaching 87% of the pediatric population.⁶ During the pre-vaccine era in Morocco, rotavirus was responsible for 44% of gastroenteritis hospitalizations in children under 5 years old.⁶ This declined to 29% in 2011, 15% in 2012, and 24% in 2013, after the introduction of rotavirus vaccination.⁶ Following updates to the Moroccan NIP at the end of 2014, HRV was replaced with HBRV, and in 2021, HBRV was replaced by BRV-PV 1-dose vial following its prequalification by the WHO.⁷

As highlighted by a recent modeling study, there is a lack of up-to-date cost-effectiveness data describing both the economic and clinical impact of rotavirus vaccines in middle-income countries, such as Morocco.⁸ This study aimed to compare the difference in total vaccination costs, rotavirus gastroenteritis (RVGE)related health outcomes and value for money of three vaccines included in the Moroccan NIP at one point in time – HRV, HBRV and BRV-PV 1-dose vial presentation – by simulating their implementation into the Moroccan NIP for the 2023 birth cohort. By comparing three vaccines included in the Moroccan National Immunization Program, this study provides further information to aid decision-making for the Moroccan government regarding which rotavirus vaccine is the most cost-effective from both a country payer and societal perspective.

Materials and methods

Analysis population

This analysis considered the population of children aged under 5 years in Morocco for the 2023 calendar year,⁹ with an estimated birth cohort of 578,358 individuals in the same year.

Vaccines included in the analysis

Three vaccines were considered in this health economic analysis: HRV (Rotarix, GSK, Belgium), HBRV (RotaTeq, Merck, USA) and BRV-PV 1dose vial (Rotasiil 1-dose vial, Serum Institute of India, India). BRV-PV 1dose vial was used in this analysis over the 2-dose vial, as only the 1-dose vial is available in Morocco, at the time of conducting this analysis. HRV and HBRV are only available in 1-dose vials. HRV requires two doses in the complete schedule, while HBRV and BRV-PV 1-dose vial both require three doses. The three vaccines were selected as they are either currently, or were previously, included in the NIP of Morocco. Therefore, this analysis may provide updated evidence on the introduction of different rotavirus vaccine options and contribute to informing healthcare decision making in Morocco. Further details, including vaccine presentations, are provided in Table 1.

Table 1. Details of vaccines included in the cost analysis and cost-effectiveness analysis.

Vaccine	Form	Presentation	Number of doses per vial	Number of doses per fully immunized child	Secondary packaging	Refrigerated storage volume per dose (cm^3)*
HRV	Liquid	Plastic tube	1	2	Carton of 50 tubes	17.12
HBRV	Liquid	Plastic tube	1	3	Carton of 25 tubes	46.25
BRV-PV 1-dose vial	Lyophilized powder	Two-vial set	1	3	Active: Carton of 50 vials (50 doses) Diluent: Carton of 50 vials	35.14

*Refrigeration storage information was obtained from the WHO information on prequalified vaccines.⁵⁷ BRV-PV 1-dose vial: rotavirus vaccine, live attenuated oral, freeze-dried; HBRV: rotavirus vaccine, live, oral, pentavalent; HRV: human rotavirus, live, attenuated, oral vaccine; WHO: World Health Organization.

Model structure

A model was developed in Microsoft Excel for use in health economic evaluations of rotavirus vaccination in different settings, and was based on a model previously developed which compared the health economic impact of rotavirus vaccination implementation in various countries.^10–12 For the purpose of this analysis, the model was adapted to the Moroccan setting to simulate vaccination of the 2023 birth cohort, and to estimate and compare the total vaccination costs of HRV, HBRV, and BRV-PV 1-dose vial.

The construct was a static, deterministic, population model of the at-risk group (children aged under 5 years old in Morocco). Briefly, a decision tree on disease management was added, comparing the costs and QALYs of vaccination both in the absence of the vaccines (natural disease pathway) and in the presence of the vaccines (intervention), as described previously.^13,14 The total costs and clinical outcomes associated with each vaccine were estimated and assessed in a feedback loop structure, as depicted in Figure 1.

Figure 1. Cost-effectiveness model construct.

This study used a model that was developed in Microsoft Excel and was based on a previous model published by Bakir et al.¹⁰ The model construct was a statistic, deterministic, population model of risk group of children aged below 5 years old, described previously.¹⁴ The results of the cost-effectiveness analysis were presented as the incremental cost-effectiveness ratio (ICER). DALY: disability-adjusted life year; ICER: incremental cost-effectiveness ratio; QALY: quality-adjusted life year; RVGE: rotavirus gastroenteritis

While HTA reimbursement decisions have been traditionally based on economic evaluations assuming the perspectives of governmental payers, it has been widely recognized that societal perspectives are required to account for healthcare, non-healthcare, and productivity costs, thus capturing a more complete value of healthcare interventions. Therefore, this analysis presents results from both payer and societal perspectives.^15,16

Four health states of rotavirus infection were used as described in previous modeling studies simulating rotavirus vaccination.^8,13,14 These four health states were: 1) RVGE home care (RVGE cases that can be treated at home without the need for healthcare provider consultation); 2) RVGE clinic visits (RVGE cases that require outpatient visits); 3) RVGE hospitalization (RVGE cases that require hospital admissions); 4) RVGE death. The outcomes used to measure the benefits of rotavirus vaccines in the model were expressed in the number of RVGE cases avoided and the accumulated Quality Adjusted Life Years (QALYs) by the birth cohort.

Analysis time horizon and discount rate

Given that RVGE is an acute disease with a short duration, and therefore the benefits offered by rotavirus vaccines would be accrued during the early stages of an infant’s life,¹⁷ the model used in this analysis estimated the costs and outcomes of rotavirus vaccine options in a 1-year cross-sectional analysis of the target population (<5 years). Therefore, no discounting was applied to the costs in the cost-effectiveness analysis as all relevant costs were assumed to be incurred within the same year of vaccination.

The model included estimates of vaccine efficacy at different RVGE health states, and data on health state utility values,^18–20 for children aged 0–5 years of age during the same year as vaccinating the birth cohort. Therefore, quality of life gains due to the prevention of RVGE cases (except RVGE-related deaths) were also not discounted. The QALY loss of early death was calculated using life expectancy estimates, and a discount rate of 5% was applied to health-adjusted life expectancy (HALE). HALE was estimated to be 63.7 in Morocco for the 0–1 year age group. As HALE is only available at birth, not for subsequent age groups, HALE declined by one year with every age group until reaching the 4–5 year age group.²¹ While there is no explicit discount rate recommended by healthcare payers in Morocco, the use of 5% rate to discount benefits only (and not costs) has been reported in previous economic evaluations of vaccines.²² Similar to a previous economic evaluation,¹⁴ this analysis assumes a steady state – the vaccine uptake period was not considered and the potential herd effect of the vaccine was not accounted for.

Cost analysis

The cost analysis considered costs from the country payer and the societal perspectives with HRV as the reference vaccine and HBRV and BRV-PV 1-dose vial as the comparators. The country payer perspective included vaccination and supply chain costs, whilst the societal perspective included country payer costs, as well as costs related to caregivers in Morocco. All costs in this analysis are reported in 2022 US dollars (USD) (1 USD = 10.16 Moroccan Dirham [MAD] 2022).²³

Vaccination costs

Vaccine acquisition costs were obtained from the Pan American Health Organization (PAHO) Revolving Fund. The most likely value for the purchase price per dose was $6.50 for HRV,²⁴ $5.20 for HBRV,²⁵ and $3.00 for BRV-PV 1-dose vial (Table S1).²⁴ Morocco is not eligible for support from Gavi, The Vaccine Alliance funding, therefore vaccine acquisition costs would be fully financed by the country payer.

Waste-adjusted vaccine acquisition costs per fully immunized child were calculated using the WHO Wastage Rates Calculator.²⁶ Due to the lack of data on the cost of vaccine waste disposal in Morocco, the costs of vaccine waste disposal reported for South Africa were used in the cost analysis. These costs are summarized in Table S2 and have been described in detail previously.¹³ The waste disposal cost was estimated as $163.5 per m³ of wasted vaccine. Vaccine administration costs were calculated per minute of healthcare worker time, summarized in Table S3.

Supply chain costs

Supply chain costs included the costs of storage, handling, and transportation of the vaccines, both internationally and locally. International vaccine handling and transportation costs were used in this analysis as a percentage of the total vaccine purchase price, each estimated to be 0.35% of the total price. The three vaccines were assumed to be refrigerated, with costs of refrigeration estimated to be $5 per m³ per day. In addition, refrigerated transport costs were estimated at $1.3 per m³ of vaccine per km. These costs were sourced from HRV manufacturer logistical data in Morocco.

For public central, regional, district and health facility storage locations, the average storage duration was assumed to be 180 days, 90 days, 45 days, and 30 days, respectively. The average transport distance between the central location and health facility was assumed to be 400 km.

Caregiver attendance costs

Caregiver attendance costs reflected costs incurred as a result of caregivers accompanying a child to the vaccine center; the costs included those for transportation and the productivity losses for the caregivers. A conservative approach was adopted given that other vaccines may also be administered in addition to rotavirus vaccines whilst attending the vaccine center. In this analysis, it was assumed that three vaccines would be given in one session. Therefore, the transportation costs borne by the caregiver and the average number of hours of productivity loss were distributed over the assumed three vaccines given in one session. A return trip to receive vaccination was assumed to cost $1 ($1 = 10.16 MAD 2022)²³ which was divided over the three vaccines to give an input value of $0.33. The average productivity loss was one working day (8 h) divided by three vaccines; thus, an input value of 2.67 h was used. The average hourly income of the caregiver was $1.39, calculated from the adjusted net national income per capita from the World Bank (net: $2,687; hourly income: $1.39; accessed 30 March 2023).²⁷

Cost-effectiveness analysis

The cost-effectiveness analysis incorporated the country payer and societal perspectives, consistent with the cost analysis described above. The gross domestic product (GDP) per capita for Morocco in 2021 was $3,497.²⁸ Therefore, the estimated cost-effectiveness threshold for the study was set at $3,500 per QALY gained.²⁸ In addition to the total vaccination costs outlined in the cost analysis above, the cost-effectiveness analysis also considered the costs associated with managing RVGE, along with other direct non-medical costs and productivity losses related to RVGE management. Clinical outcomes were estimated in terms of QALYs gained in the four RVGE health states described above: RVGE homecare, RVGE medical visits, RVGE hospitalization, and RVGE-related death.

Cost inputs

Relevant local data were obtained from the literature and public sources. The cost-effectiveness analysis included the same vaccine cost inputs as those in the cost analysis (detailed above: vaccination costs, supply chain costs, and caregiver attendance costs), in addition to cost inputs associated with RVGE management which were not included in the cost analysis. A summary of the cost components included in the cost analysis and cost-effectiveness analysis is presented in Table 2.

Table 2. Cost components included in the cost analysis and cost-effectiveness analysis.

Analysis	Cost analysis		Cost-effectiveness analysis
Perspective	Country payer	Societal	Country payer	Societal
Vaccination costs (Waste-adjusted vaccine acquisition, vaccine administration, waste disposal)	✓	✓	✓	✓
Supply chain costs (International transportation/handling and local storage/transportation)	✓	✓	✓	✓
Caregiver attendance costs (Transportation to receive vaccine)	X	✓	X	✓
RVGE management costs (Homecare treatment, medical visits, hospitalization)	X	X	✓	✓
Direct non-medical costs (Transportation to manage RVGE)	X	X	X	✓
Productivity losses of caregivers (To manage RVGE)	X	X	X	✓

RVGE: rotavirus gastroenteritis.

RVGE management costs were considered from both the country payer and societal perspectives; these costs included homecare treatment, cost of medical visits, and hospitalizations. The costs were calculated by multiplying the unit costs and probability of each RVGE health state (RVGE homecare treatment, RVGE medical visits, and RVGE hospitalization) and applying the result on the simulated cohort. The homecare treatment cost was assumed to be $0. The costs of medical visits and hospitalizations due to RVGE were reported in a previously published economic analysis of rotavirus vaccination,⁸ and were adjusted from 2018 to 2022 USD using the Consumer Price Index (CPI) inflation calculator to be $11 and $156, respectively.

Probabilities of RVGE events per 100,000 individuals were estimated for each of the four RVGE health states using previously published modeling study data (Table S4).⁸ To calculate the probability of each RVGE health state per person in the total population, per year of age, RVGE health state probabilities were applied to 1-year age groups from 0 to 5 years. Data on age group distribution were obtained from previously published epidemiological studies conducted in Morocco.^29,30

From the societal perspective, caregiver management of RVGE cases was considered and included costs of transportation and productivity losses associated witSSh RVGE medical visits (Table 2). The number of days’ productivity loss was based on a previously published cost-effectiveness analysis of rotavirus vaccination in France,¹⁸ in which the duration of a community case was assumed to be 4 days. Other health states and associated workdays lost were assumed to be equal and were expressed as incremental days added to the duration of community cases. The duration and workdays lost due to medical visits and hospitalization were 5 days and 7 days, respectively.¹⁸

Vaccine efficacy, coverage, and compliance

Consistent with the WHO summary of rotavirus vaccine characteristics, where HBRV and BRV-PV had equal vaccine efficacy in high mortality countries,³¹ this analysis assumed that all included vaccines would have equal efficacy. In the absence of specific vaccine efficacy data per outcome, equal vaccine efficacy was assumed for all RVGE health states described above.

An efficacy estimate of 41.5% was used in this model, sourced from a surveillance study of circulating rotavirus strains in children in Morocco, which reported a 41.5% decline in the prevalence of rotavirus in children hospitalized with diarrhea between the pre- and post-rotavirus vaccine eras.⁶ As noted previously, the vaccines used in this analysis had different dosing schedules; HRV has two doses in complete schedule and HBRV and BRV-PV 1-dose vial have three doses. Vaccine efficacy is driven by both vaccine coverage (completion of recommended number of doses) and compliance (receipt of all doses at the recommended time interval).³² As coverage of HRV may be higher due to the lower number of required doses, in this model, vaccine efficacy decrements due to incomplete vaccination schedules were included. These decrements were calculated by determining the percentage difference between the efficacy of each dose and the subsequent dose. The efficacy values were taken from a cost-effectiveness publication of rotavirus vaccination in Ireland (Table S6).³³

Vaccination coverage was defined as the percentage of children who received at least one vaccine dose. In this analysis, the target coverage of the first dose was 99%, based on WHO and United Nations Children’s Fund (UNICEF) estimates of immunization coverage of the Diphtheria, Tetanus and Pertussis (DTP1) vaccine in Morocco.³⁴ Vaccine compliance was defined as the percentage of children who received the 1st, 2nd, and/or 3rd dose as indicated by the vaccine type used. Second dose compliance was assumed to be 100% (as 1st and 2nd dose coverage were both 99%). Completion rate (percentage of children fully covered) of the 2-dose vaccine was 99% (100% of those who received the 1st dose). Based on data from the Moroccan Ministry of Health records, the completion rate of 3-dose vaccines was 93%, resulting in a 3rd versus 2nd dose compliance rate of 94%.

Health state utility values

To calculate health outcomes, preference-based utility values were used to determine QALYs (Table S5). Health state utility values (HSUVs) were applied to each of the RVGE health states based on previously published values.^18–20 HSUVs were fixed for each age group; 0–1 and 1–2 years age groups shared a value and the 2–3, 3–4 and 4–5 years age groups shared a value (Table S5).

Sensitivity analysis

One-way deterministic sensitivity analyses (DSA) were performed for both the cost analysis and cost-effectiveness analysis to identify the model parameters with the greatest influence on the base-case results from the country payer and societal perspectives. Upper and lower limits of the model inputs varied by ± 50% of the most likely input value. Vaccine prices were varied by ± 20% of the base case value as this was unlikely to significantly deviate from the real-life value. DSA results were represented as tornado plots, with the most influential input parameters are located at the top of the diagram. Multivariate probabilistic sensitivity analyses (PSA) with 10,000 iterations were also conducted for the cost-effectiveness analysis to test the robustness of the base-case results.

Scenario analysis

Scenario analyses were performed for the cost and cost-effectiveness analyses to test the assumption of productivity loss due to vaccination. The scenarios tested were 0.25, 0.5, 1 and 1.5 working days.

Incremental analysis

The three vaccines were compared in an incremental cost-effectiveness analysis. Incremental net monetary benefit and probability of cost-effectiveness were also calculated, comparing HBRV and BRVPV 1-dose vial with HRV. Tables were organized by differences in the cost of vaccine.

Results

Required doses for the modelled population

Based on the simulated birth cohort and target vaccination coverage, the number of infants estimated to be vaccinated in this analysis was approximately 573,000. The estimated number of doses required to vaccinate this population with HRV was 1,192,863. This was lower than the number of doses required for HBRV and for BRV-PV 1-dose vial (1,789,295 doses for each vaccine).

Cost analysis

Base-case results

Overall, HRV was significantly less costly than HBRV from both the country payer and societal perspectives, and slightly less costly than BRV-PV 1-dose vial from the societal perspective (Table 3). From the country payer perspective, the estimated total vaccination cost associated with the implementation of rotavirus vaccination into the Moroccan NIP was $8,016,259 for HRV, $9,827,482 for HBRV, and $5,875,280 for BRV-PV 1dose vial. From the societal perspective, costs increased to $12,642,660 for HRV, $16,767,084 for HBRV, and $12,814,882 for BRV-PV 1-dose vial.

Table 3. Base-case cost analysis results comparing the total vaccination costs of HRV with HBRV and BRV-PV 1-dose vial.

Cost	HRV costs	HBRV costs	BRV-PV 1-dose vial costs	∆ costs HRV versus HBRV	∆ costs HRV versus BRV-PV 1dose vial
Vaccination	$7,916,137	$9,576,567	$5,696,548	-$1,660,430	$2,219,589
Vaccination costs including wastage	$7,753,612	$9,304,334	$5,367,885
Vaccine administration costs	$159,176	$258,660	$318,351
Waste disposal	$3,349	$13,572	$10,312
Supply chain	$100,122	$250,915	$178,731	-$150,793	-$78,609
International transport	$27,138	$32,565	$18,788
International handling	$27,138	$32,565	$18,788
Storage costs	$35,228	$142,752	$108,461
Transport costs	$10,619	$43,033	$32,695
Caregiver attendance cost*	$4,626,401	$6,939,602	$6,939,602	-$2,313,201	-$2,313,201
Total vaccination costs from the country payer perspective
For birth cohort	$8,016,259	$9,827,482	$5,875,280	-$1,811,223	$2,140,979
Per FIC	$14.00	$17.16	$10.26	-$3.16	$3.74
Total vaccination costs from the societal perspective
For birth cohort	$12,642,660	$16,767,084	$12,814,882	-$4,124,423	-$172,221
Per FIC	$22.08	$29.28	$22.38	-$7.20	-$0.30

*Caregiver attendance costs were assumed to be $0 from the country payer perspective. BRV-PV 1-dose vial: rotavirus vaccine, live attenuated oral, freeze-dried; FIC: fully immunized child; HBRV: rotavirus vaccine, live, oral, pentavalent; HRV: human rotavirus, live, attenuated, oral vaccine.

Sensitivity analysis results

The one-way DSA highlighted that HRV, HBRV and BRV-PV 1-dose vial purchase price per dose were the biggest drivers of costs in the comparison between these vaccines from both country payer and societal perspectives (Figure S1). From the country payer perspective, parameters such as average distance to the public health facility, supply chain costs and healthcare worker time/costs for vaccination administration were less influential (Figure S1A, S1B). From the societal perspective, lesser, but still significant drivers of the cost analysis outcomes included average hours of healthcare worker productivity loss per dose administration and average income per worked hour. Other vaccine costs such as supply chain were less influential (Figure S1C, S1D).

Cost-effectiveness analysis

Health outcomes

Overall, the model estimated that vaccination with HRV could lead to better health outcomes versus HBRV and BRV-PV 1-dose vial and that more RVGE events could be avoided through implementation of HRV into the Moroccan NIP versus HRV or BRV-PV 1-dose vial. HRV was also associated with an estimated 194 fewer homecare events, 57 fewer medical visits and 8 fewer hospitalization events than HBRV or BRV-PV 1-dose vial. This translated to a discounted QALY difference of 7 QALYs gained, favoring HRV (20 undiscounted QALYs) in the analysis (Table 5).

Base-case costs

HRV was associated with lower total costs versus HBRV from the country payer and societal perspectives (Table 4), largely driven by the vaccination costs ($8,359,258 for HRV and $10,172,432 for HBRV). From the societal perspective, HRV was also associated with lower costs than the 3-dose vaccines for RVGE disease management (–$1,952) and productivity losses (–$12,467). However, there was little difference between non-medical costs (–$83). The largest difference was observed between caregiver attendance costs, with HRV costs totaling $4,626,401 and HBRV costs amounting to $6,939,602.

Table 4. Base-case cost-effectiveness analysis results comparing costs of HRV with HBRV and BRV-PV 1-dose vial.

	HRV costs	HBRV costs	BRV-PV 1-dose vial costs	∆ costs HRV versus HBRV	∆ costs HRV versus BRV-PV 1dose vial
Country payer perspective
Vaccination	$7,916,137	$9,576,567	$5,696,548	-$1,660,430	$2,219,589
Supply chain	$100,122	$250,915	$178,731	-$150,793	-$78,609
Direct medical treatment costs	$342,999	$344,951	$344,951	-$1,952	-$1,952
Total costs from country payer perspective	$8,359,258	$10,172,432	$6,220,230	-$1,813,174	$2,139,028
Societal perspective
Direct non-medical costs	$14,534	$14,617	$14,617	-$83	-$83
Productivity loss	$2,190,773	$2,203,239	$2,203,239	-$12,467	-$12,467
Caregiver attendance	$4,626,401	$6,939,602	$6,939,602	-$2,313,201	-$2,313,201
Total costs from societal perspective	$15,190,966	$19,329,890	$15,377,688	-$4,138,924	-$186,722

BRV-PV 1-dose vial: rotavirus vaccine, live attenuated oral, freeze-dried; FIC: fully immunized child; HBRV: rotavirus vaccine, live, oral, pentavalent; HRV: human rotavirus, live, attenuated, oral vaccine.

In comparison, higher total costs were observed for HRV versus BRVPV 1-dose vial from the country payer perspective, driven by the higher vaccination costs associated with HRV (HRV: $8,359,258; BRV-PV 1-dose vial: $6,220,230), amounting to a difference of $2,139,028. However, these costs were completely offset when the societal costs were considered. From the societal perspective, HRV was associated with slightly lower costs than BRVPV 1-dose vial by a difference of $186,722. This was mainly due to the lower caregiver attendance costs for HRV ($4,626,401) versus BRVPV 1dose vial ($6,939,602).

Incremental cost-effectiveness results

Base-case findings

HRV demonstrated dominance, i.e., better health outcomes at reduced cost, versus HBRV from both the country payer and societal perspectives and BRV-PV 1-dose vial from the societal perspective (Tables 5 and 6). Although HRV was associated with fewer RVGE events and more QALY gains, from the country payer perspective costs of BRV-PV 1-dose vial were lower than HRV, resulting in a discounted base-case incremental cost-effectiveness ratio (ICER) of approximately $328,376 per QALY, above the assumed cost-effectiveness threshold ($3,500).

Table 5. Base-case cost-effectiveness analysis results comparing QALYS of HBRV and BRV-PV 1-dose vial with HRV.

Vaccines	QALY loss (non-discounted)	QALY loss (discounted)	Δ QALY (non-discounted)	Δ QALY (discounted)
HRV	3,479	1,145
HBRV	3,499	1,151	20	7
BRV-PV 1-dose vial	3,499	1,151	20	7

BRV-PV 1-dose vial: rotavirus vaccine, live attenuated oral, freeze-dried; HBRV: rotavirus vaccine, live, oral, pentavalent; HRV: human rotavirus, live, attenuated, oral vaccine; QALY: quality-adjusted life-year.

Table 6. Base-case ICER and probability of HRV being cost-effective reported from the country payer and societal perspectives.

Reference	Comparator	Δ Costs	Δ QALY (discounted)	Base case ICER	Incremental net monetary benefit of HRV	Probability of HRV being cost-effective
Country payer perspective
HRV	HBRV	-$1,813,174	7	HRV is dominant	$1,837,674	91%
	BRV-PV_1d	$2,139,028	7	$328,376	-$2,114,528	0%
Societal perspective
HRV	HBRV	-$4,138,924	7	HRV is dominant	$4,163,424	99.9%
	BRV-PV_1d	-$186,722	7	HRV is dominant	$211,222	60%

BRV-PV 1-dose vial: rotavirus vaccine, live attenuated oral, freeze-dried; HBRV: rotavirus vaccine, live, oral, pentavalent; HRV: human rotavirus, live, attenuated, oral vaccine; ICER: incremental cost-effectiveness ratio; QALY: quality-adjusted life-year. Dominance represents the vaccine that is the most cost-saving, with greater QALY gains at a lower cost. Incremental net monetary benefit =incremental benefit x threshold – incremental cost, where cost effectiveness threshold = $3,500/QALY.

Sensitivity analyses

In the one-way DSA, HRV remained dominant versus HBRV from both the country payer and societal perspectives, except when the low input value of HBRV price per dose was used in the payer perspective. Similar to the cost analysis DSA, purchase price per dose of HRV, HBRV and BRV-PV 1-dose vial were the most influential parameters on base-case results (Figure 2). Other vaccine costs including wastage and administration were less influential.

Figure 2. Cost-effectiveness deterministic sensitivity analysis tornado plots.

BRV-PV 1-dose vial: rotavirus vaccine, live attenuated oral, freeze-dried; HBRV: rotavirus vaccine, live, oral, pentavalent; HRV: human rotavirus, live, attenuated, oral vaccine; ICER: incremental cost-effectiveness ratio; RVGE: rotavirus gastroenteritis. a&b: country payer perpective, c&d: societal perpective.

Based on a sample of 10,000 iterations, the PSA demonstrated a high probability (91% and 99.9% from country payer and societal perspectives, respectively) of HRV being a cost-effective option at the $3,500 threshold versus HBRV (Figure 3); the majority of iterations were located within the lower right quadrant of the cost effectiveness plane. When compared with BRV-PV 1-dose vial, from the country payer perspective, the probability of HRV being cost-effective was very low, with less than 1% of the 10,000 iterations estimated to be below the $3,500 threshold. However, from the societal perspective, HRV had a higher probability of being cost-effective versus BRV-PV 1-dose, with 60% of iterations below the cost-effectiveness threshold.

Figure 3. Cost-effectiveness analysis probabilistic sensitivity analysis.

BRV-PV 1-dose vial: rotavirus vaccine, live attenuated oral, freeze-dried; HBRV: rotavirus vaccine, live, oral, pentavalent; HRV: human rotavirus, live, attenuated, oral vaccine; QALY: quality-adjusted life year.

a: country payer perspective, b: societal perpective.

Scenario analysis

When the assumption of caregiver productivity time lost per vaccination dose − 2.67 h (one working day [8 h] divided by three vaccines) – was tested, HRV was dominant versus HBRV and BRV-PV 1-dose vial across all scenarios (0.25 days, 0.5 days, 1 day and 1.5 days), except 0.25 days for the BRV-PV 1 dose vial. In this scenario, ICER was $60,114 (Table 7).

Table 7. Scenario analyses testing different assumptions of productivity loss.

Caregiver productivity loss with administration of 1 dose	Vaccine	Costs (societal perspective)	QALY loss (discounted)	Δ Cost	Δ QALY	ICER
0.25 working day (2 h)	HRV	$14,034,366	1,145
	HBRV	$17,549,990	1,151	-$3,560,624	7	HRV is dominant
	BRV-PV 1-dose vial	$13,642,788	1,151	$391,578	7	$ 60,114
0.5 working day (4 h)	HRV	$17,504,167	1,145
	HBRV	$22,799,691	1,151	-$5,295,525	7	HRV is dominant
	BRV-PV 1-dose vial	$18,847,489	1,151	-$1,343,323	7	HRV is dominant
1 working day (8 h)	HRV	$24,443,769	1,145
	HBRV	$33,209,094	1,151	-$8,765,326	7	HRV is dominant
	BRV-PV 1-dose vial	$29,256,892	1,151	-$4,813,124	7	HRV is dominant
1.5 working days (12 h)	HRV	$30,047,364	1,145
	HBRV	$41,614,487	1,151	-$11,567,123	7	HRV is dominant
	BRV-PV 1-dose vial	$37,662,285	1,151	-$7,614,921	7	HRV is dominant

BRV-PV 1-dose vial: rotavirus vaccine, live attenuated oral, freeze-dried; HBRV: rotavirus vaccine, live, oral, pentavalent; HRV: human rotavirus, live, attenuated, oral vaccine; ICER: incremental cost-effectiveness ratio; QALY: quality-adjusted life-year. Dominance represents the vaccine that is the most cost-saving, with greater QALY gains at a lower cost.

Incremental analysis

When the three interventions were compared in an incremental analysis, both HBRV and BRV-PV 1-dose vial were dominated by HRV. This was due to the lower costs and more favorable outcomes associated with HRV, i.e., HRV resulted in fewer QALYs lost, compared with the other two vaccines. HRV was associated with higher incremental net monetary benefit versus HBRV from the country payer and societal perspectives (Table 6). Figure 4 shows a scatter plot with the threshold value.

Figure 4. Incremental analysis scatter plot with threshold value.

BRV-PV 1-dose vial:rotavirus vaccine, live attenuated oral, freeze-dried; HBRV: rotavirus vaccine, live, oral, pentavalent; HRV: human rotavirus, live, attenuated, oral vaccine; QALY: quality-adjusted life year. Threshold: $3,500/QALY.

Discussion

RVGE has been responsible for significant morbidity and mortality in North Africa.³⁵ Demographic indicators from the United Nations Children’s Fund (UNICEF) estimated that the average mortality rate in North Africa was 39 per 100,000 individuals, at least four times higher than in Europe (below 10 per 100,000).³⁵ This highlights the importance of implementing effective rotavirus vaccination in the region, which has the potential to alleviate the burden of RVGE.

The objective of this study was to compare differences in total vaccination costs, RVGE-related health outcomes, and cost-effectiveness of HRV, HBRV, and BRV-PV 1-dose vial presentation by simulating the inclusion of these vaccines within the Moroccan NIP. The cost analysis demonstrated that HRV was associated with lower total vaccine costs than HBRV from both the country payer and societal perspectives. In comparison, BRV-PV 1-dose vial was associated with lower total vaccine costs than HRV from the country payer perspective, but this was completely offset when societal costs were considered, resulting in lower costs for HRV versus BRV-PV 1-dose vial from the societal perspective. Furthermore, in the cost-effectiveness analysis, HRV demonstrated dominance (i.e., better health outcomes at reduced cost) versus HBRV from both perspectives, and BRV-PV 1-dose vial from the societal perspective.

Importantly, the cost-effectiveness analysis considered additional costs which were not evaluated in the cost analysis. These included costs of RVGE management from both country payer and societal perspectives, as well as direct nonmedical costs (transportation to manage RVGE) and productivity losses to caregivers from a societal perspective. Usually, economic analyses are based on the payer’s perspective. However, the effects of vaccination may have broader societal impacts. We therefore also included a societal perspective in this analysis.^36,37 The model used in the current analysis has the capacity to analyze these societal impacts, including broader implications of disease management such as caregiver productivity losses.

Debellut et al. conducted a modeling study which included 63 middle-income countries not eligible for GAVI, The Vaccine Alliance funding, including Morocco.⁸ The study highlighted the lack of existing literature examining the cost-effectiveness of rotavirus vaccination in these countries, especially the impact of the newly prequalified vaccines, BRV-PV 1-dose vial and ORV 116E.⁸ The model used in the current analysis can produce economic evaluations of rotavirus vaccination adapted for specific settings, offering the option to include additional costs that may otherwise be omitted. For instance, in this study tailored to Morocco, specific costs related to healthcare provider (HCP) administration costs, transportation, and caregiver attendance were incorporated. As a result, the outputs of this model could be used to generate more realistic simulations of rotavirus vaccination implementation in different settings for healthcare policymakers.

In addition to total costs, this analysis also estimated the RVGE-related health outcomes for HRV, HBRV, and BRV-PV 1-dose vial presentation. Although the efficacy of the vaccines included in this study were assumed to be equal when given in their completed dosing schedule, vaccination with HRV was estimated to result in better health outcomes than vaccination with HBRV or BRV-PV 1-dose vial. HRV was associated with fewer RVGE events, RVGE medical visits, and RVGE hospitalizations, which translated to 7 QALYs gained versus HBRV and BRV-PV 1-dose vial.

The effectiveness of a vaccine is impacted by the timely completion of all doses in the vaccination schedule, and vaccines requiring two doses may offer earlier protection compared to those requiring three doses.¹ A systematic review of the data gathered in the first decade of HRV and HBRV postlicensure across 24 countries concluded that the effectiveness of a partial vaccination series was not as high as a complete vaccination series.³⁸ Another systematic review and meta-analysis reported a ‘protection boost’ from one to two doses of HRV, one to two doses of HBRV and one to three doses of HBRV.³⁹ As compliance rates of HRV are likely to be higher than the 3-dose vaccines,^40–43 it may offer earlier efficacy and protection against RVGE. However, as there were no available data on vaccine efficacy decrements for the vaccines used in this analysis in the Moroccan setting, efficacy estimates of HRV and HBRV were developed by independent investigators.

This analysis assumed equal efficacy (41.5%) for the three vaccines. Uncertainty of this estimate was explored in the DSA and PSA using the range of 21–62% (±50%). The assumption of equal efficacy is consistent with previous evaluations of rotavirus vaccines, including economic and health impact evaluations of rotavirus vaccines in 63 middle-income non-Gavi countries⁸ and in 73 Gavi countries.⁴⁴ In the latter analysis, pooled vaccine efficacy was assumed to be 44% in settings where mortality for children <5 years is high. The assumption of similar vaccine efficacy is consistent with the WHO summary of rotavirus vaccine characteristics, where HBRV and BRV-PV 1-dose vial were reported to have equal efficacy (44%) in high mortality countries.³¹ Whilst HRV efficacy was reported to be 10% higher in the same summary, a more conservative approach was taken in this analysis.

The efficacy assumed in this analysis is lower than other published literature, such as Henschke et al.⁴⁵ However, rotavirus vaccine efficacy is lower for less severe RVGE compared with severe RVGE.^46–48 Therefore, a lower efficacy estimate was deemed appropriate for this analysis as health states other than severe RVGE were included and due to the scarcity of evidence for less severe RVGE vaccine efficacy in Morocco.

A limitation of this analysis is the use of a static, rather than a dynamic, model. Dynamic models can account for both direct and indirect effects of vaccination on vaccinated and unvaccinated individuals, including herd immunity.^49,50 However, this is a limitation with minimal impact on the current analysis as there may be instances where the use of dynamic models are not required to conduct a robust analysis. For example, when the intervention is not expected to impact the rate unvaccinated people become infected or when the vaccine protects individuals against the disease but doesn’t necessarily prevent asymptomatic infections.⁵⁰ Asymptomatic rotavirus infections are common among newborns, children and adults who are likely protected against symptomatic disease by maternal antibodies during the first 3–4 months of life or an immune response triggered by previous infections.⁵¹ Evidence on the ability of rotavirus vaccines to prevent asymptomatic transmission is inconclusive.^52,53 Thus, the use of a static model in this analysis is expected to provide robust and reliable evidence. A further limitation involved the use of South Africa to estimate the cost of vaccine waste disposal in Morocco. This choice is not expected to significantly influence the analysis results as the cost of the unit volume of disposed vaccine vials will be the same across the vaccines. Another limitation of this analysis is the assumption of equal vaccine efficacy. However, this is consistent with the WHO summary of prequalified rotavirus vaccines and was used in previous economic evaluations of rotavirus vaccines.^8,13,14,44 While a sensitivity or scenario analysis using differential vaccine estimates would be informative, the Roxette model used to conduct this analysis is built on the assumption that all vaccines would have equal efficacy in their complete dosing schedule and that a differential vaccine efficacy would arise from differential completion rates. However, this limitation can be expected to have minimal impact on the direction of the model results as the WHO reports comparable efficacy of the prequalified vaccines in high mortality settings for HRV (54%) and for HBRV and BRV-PV 1-dose vial (44%).³¹

Our analysis resulted in higher ICERs than those in other studies.^54–56 This is mostly due to these studies comparing the cost-effectiveness of the vaccine with a no vaccine scenario, resulting in a significant difference in cost-effectiveness. In this analysis, there is a significant difference in the costs between vaccines (large numerator in the ICER equation) and a small difference in incremental QALYs (small denominator in the ICER equation), leading to high ratios. The small difference in QALYs can be attributed to the initial assumption of equal vaccine efficacy at complete dosing schedule. Therefore, the difference between vaccine outcomes was driven by the difference in compliance between 2- and 3-dose vaccines and the subsequent efficacy decrements implemented for missed vaccine doses.

The results of this analysis may have been influenced by the uncertainty surrounding the estimated costs of caregiver attendance, due to the possible concomitant administration of other vaccines with the rotavirus vaccines. Therefore, a conservative approach was taken, and caregiver attendance costs were divided over three concomitantly administered vaccine doses. Testing this assumption showed that the only scenario that influenced the output of the model was when productivity loss was 0.25 days for BRV-PV 1-dose vial. The DSA confirmed that uncertainty surrounding other model inputs, including waste disposal costs, were not major drivers of the base-case results.

Costs of introducing new vaccines included in this analysis were those with validated inputs in a Moroccan setting. Therefore, some costs without validated Moroccan inputs, such as training and communications, were not included. These costs are expected to have minimal impact on the direction of the model results as they would be consistent across the three vaccines, regardless of purchase price or number of doses.

While it can be reasonably expected that a 2-dose vaccine will provide a clear economic benefit over 3-dose vaccines, this assumption is only true under certain conditions, for example if vaccine price per dose and the effectiveness of both vaccination schedules are comparable. The current analysis provides a deeper insight on real-world implementation of rotavirus vaccination programs and the importance of full early protection. Multiple studies have consistently demonstrated higher rates of vaccine completion with 2-dose compared to 3-dose rotavirus vaccines,^40–43 and a considerable reduction in vaccine effectiveness due to incomplete dosing schedules.^38,39 Furthermore, the per-dose cost of BRV-PV 1-dose vial was around 50% lower than the per-dose cost of HRV and HBRV.^24,25 With this significant difference in vaccine price per dose, this analysis sheds light on other important cost variables that influence the results of the comparison. Such variables include caregiver attendance costs and productivity losses, with the impact of the latter investigated in the scenario analysis.

The findings presented in this study are consistent with previously published economic evaluations of rotavirus vaccine implementation into different settings. A recently published study evaluating the impact of HRV, HBRV, and BRV-PV 1-dose vial in the Mexican NIP also found that HRV could result in better health outcomes, with higher QALY gains at a lower cost than the 3-dose vaccines.¹⁴ While middle-income countries can be diverse in terms of their size, population and income level, some middle-income countries share the fact that they are not eligible to receive financial support from GAVI, The Vaccine Alliance. Given that commonality, it is likely that conducting this analysis in other healthcare jurisdictions from other regions with similar profiles, for example Thailand (Southeast Asia Region), Malaysia, and the Philippines (Western Pacific Region), will yield similar results.

Conclusions

In this analysis, the simulated implementation of HRV into the Moroccan NIP was associated with fewer RVGE-related events versus HBRV or BRVPV 1dose vial. HRV was a cost saving option in comparison with HBRV in both the cost analysis and the cost-effectiveness analysis, from the country payer and societal perspectives. Furthermore, HRV was estimated to be a cost saving option versus BRV-PV 1-dose vial from the societal perspective. Overall, the information presented in this analysis may provide updated insights on the implementation of rotavirus vaccines and may help to inform healthcare decision makers in Morocco.

Author contributions

Substantial contributions to study conception and design: AM, YL, JG, BA, MB; substantial contributions to analysis and interpretation of the data: AM, YL, JG, BA, MB; drafting the article or revising it critically for important intellectual content AM, YL, JG, BA, MB; final approval of the version of the article to be published: AM, YL, JG, BA, MB.

Acknowledgments

The authors acknowledge Laure-Anne Van Bellinghen and Ilse Van Vlaaenderen at CHESS in Health for creating the model used in this study and Bella Dragova for publication management. The authors also thank Costello Medical for editorial assistance and publication coordination, on behalf of GSK, and acknowledge Megan Thomas, Costello Medical, UK for medical writing and editorial assistance based on authors’ input and direction.

Disclosure statement

AM, YL and JG and BA are employed by and hold shares in the GSK group of companies. MB has received consulting fees from GSK.

Data availability statement

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Supplementary material

Supplemental data for this article can be accessed on the publisher’s website at https://doi.org/10.1080/21645515.2024.2353480.

Additional information

Funding

This study was sponsored by GlaxoSmithKline Biologicals SA (Study identifier eTrack VEO-000351). Support for third-party writing assistance for this article, provided by Costello Medical, UK was funded by GSK in accordance with Good Publication Practice (GPP 2022) guidelines (https://www.ismpp.org/gpp-2022).