Cancer Medication Consumptions in Vietnam Between 2019 and 2023: An Analysis from the National Database Authors: Mai Thi Tuyet Kieu, Duc Quang Tran, Anh Thi Phuong Le, Ha Thi Nguyen, Minh Cong Nguyen Published: 2026 ABSTRACT AbstractPurposeThis study was conducted to analyse cancer medication consumption in Vietnam between 2019 and 2023.Materials and MethodsThis was a cross-sectional study using data of cancer medication consumption that was publicly announced by Drug Administration of Vietnam (DAV) as a proxy. Data on expenditures and purchasing characteristics were extracted from every bidding package and contract implemented throughout the period. They afterwards were quantified and categorised by bidding technical groups, chemical classes, procurement levels, manufacturing origins and administration routes. Descriptive and statistical analyses were performed to explore the relationship between consumption, price and bidding mechanisms.ResultsThe total consumption of cancer medications between 2019 and 2023 in Vietnam was estimated to be about $1.4 billion. Of which the number slightly increased in 2020 before dropping sharply to 2022 and finally rebounding and reaching its peak in 2023 with $359.5 million. Targeted therapies consistently accounted for the largest proportion (44–48%), followed by chemotherapies with 34–39% and immunotherapies with 8–16%. Most expenditures were attributed to originator products and high-quality generics (Group 1), contributing 73–85% annually. Overall, the mean price per milligram decreased over time, with notable variation driven by shifts in the distribution of bidding technical groups while the centralized national-bidding was generally more efficient than regional procurements.ConclusionCancer medication consumption in Vietnam fluctuated substantially between 2019 and 2023, accompanied by significant changes in procurement patterns and the pandemic. Strengthened and targeted policies are required to optimize the economic impact of this group of medication, while broader efforts are needed to alleviate the cancer burden in Vietnam.Keywords: cost-analysis, cancer medications, drug consumption, bidding technical groups Introduction According to the Global Cancer Observatory Research (GLOBOCAN), nearly 20 million new cancer cases and 9.7 million cancer-related deaths were recorded worldwide in 2022. In Vietnam, cancer ranks among the top ten leading causes of death in both men and women, with 180,480 new cases and 120,184 deaths in the same year.1 Of which, the five most frequent cancers are breast, liver, lung, colorectum and stomach cancers, respectively, making up nearly 60% of the total new cases in 2022. Noticably, over 70% of cases are diagnosed at late stages (stage III or IV), causing most treatments only help patients prolong their survival and improve quality of life instead of curative intents. On the other hand, although the proportion of cancer-related economic burden relative to national GDP tends to be lower in low- and middle-income countries (LMICs), these settings experience substantially higher cancer mortality rates, which continue to rise even as mortality stabilizes or declines in high-income countries (HICs).2 Structural constraints—including shortages of trained oncology professionals, limited access to essential cancer medicines, and significant treatment-related financial hardship—are major contributors to these disparities, and Vietnam is no exception.3,4 A previous study indicated that 57% of patient families had to spend more than 50% of their income for cancer treatments and 37% were impoverished by related costs.5 Yet, financial and system-related challenges are often under-recognized in clinical guidelines, particularly those originating from countries with high human development index.6 Despite major advances in cancer therapeutics, medications remains a cornerstone of cancer treatment as about 40% of diagnosed patients received at least one type of cancer medication throughout their treatment.7 Noticably, studies in LMICs showed that costs for medications generally accounted for the largest proportion of healthcare expenditure, ranging from 47% to 65% in common diseases like breast8,9 or lung cancers10 and being significantly higher than HICs. Furthermore, cancer medications made up about 10% of the total drug expenditure11 while only being used by around 0.7% of the total population.12 They are also expected to increase sharply because of the launch of many expensive agents, thereby raising questions about their economic efficiency, particularly when researches in Vietnam noticed about the poor cost-effectiveness of those new medications.13–15 This reality raised the need for evidence regarding cancer medication consumptions and their price variations to optimize product selection, guide procument and payment policies in the context of constraint healthcare resources. Nevertheless, to date, there has been no studies in Vietnam provided such information at the national level. Therefore, the aim of this study to provide an analysis of nationwide cancer medication consumption in Vietnam between 2019 and 2023, using data from the national procurement database of Drug Administration of Vietnam (DAV). This study would identify agents that need attention to assess economic efficiency, provide valuable input parameters for future health economic evaluations and information on the efficacy of drug procurement regulations in the near future. Materials and Methods Study Settings A previous national review in 2019 indicated that most cancer treatments in Vietnam were delivered at public hospitals, therefore cancer medication consumptions in those healthcare centers could be considered approximately equivalent to the national figures.16 The supply of medicines to public hospitals is carried out through bidding processes under the control of Drug Administration of Vietnam (DAV) and must comply with national regulations. Drug procurement may be conducted via: Centralized national-level bidding (National Center for Drug Procurement – Ministry of Health), Centralized national-level bidding (National Center for Drug Procurement – Ministry of Health), Provincial-level bidding (Department of Health), or Provincial-level bidding (Department of Health), or Independent procurement by individual hospitals to meet their specific needs. Independent procurement by individual hospitals to meet their specific needs. The maximum contract duration for procument at national-level biddings and provincial-level bidding is 36 months and 12 months for individual-hospital level.17 Within this period, healthcare facilities and the winning bidder must fulfill at least 80% of the value of the signed contract.17 Therefore, procurement volume is widely accepted as a reliable proxy for drug consumption in Vietnam.18 Moreover, in this system, consumption and procurement volumes are closely aligned because hospitals procure medicines strictly based on treatment needs and contractual fulfillment requirements. According to Circular 15/2019/TT-BYT17 and Circular 06/2023/TT-BYT19 of Ministry of Health, branded medicines and generic drugs are categorized into five technical groups that strongly influence bid-winning prices: G1 – generics manufactured in EMA/SRA countries or in Vietnam with EU-GMP certification, and licensed by EMA/SRA authorities; G1 – generics manufactured in EMA/SRA countries or in Vietnam with EU-GMP certification, and licensed by EMA/SRA authorities; G2 – generics manufactured in other countries meeting EU-GMP standards or PIC/s-ICH countries meeting PIC/s-GMP standards; G2 – generics manufactured in other countries meeting EU-GMP standards or PIC/s-ICH countries meeting PIC/s-GMP standards; G3 – generics with bioequivalence evidence; G3 – generics with bioequivalence evidence; G4 – generics manufactured in Vietnam meeting GMP standards; G4 – generics manufactured in Vietnam meeting GMP standards; G5 – generics licensed for circulation in Vietnam. G5 – generics licensed for circulation in Vietnam. In practice, winning bids are generally the lowest-priced products within the eligible technical groups based on predefined quotas.18 Data Resources The law on Procurement stipulates that the bidding opportunities, contract information (purpose, contractor and value) must be publicly announced, and it is encouraged to publish on the local government agencies, ministries, websites or mass media. This analysis is based on the data of the tender-winning bids that is published in the website of the Drug Administration of Vietnam (https://dav.gov.vn/), which is the Ministry of Health regulatory authority for the period January 2019 and December 2023.18 This dataset covers all publicly procured products nationwide, representing nearly the entire volume of cancer medications used in Vietnam’s public health system, where most cancer treatments occur. In the database, each drug record included the name of the active ingredient, trademarks, strength, dosage form, package, route of administration, registration identification, pharmaceutical manufacturer, country of origin of manufacturer, unit of measurement, quantity, unit price, total purchase expenditures, bidding intervals and technical bidding groups.18 Data Classification, Cleaning and Exclusions Cancer medication products were identified using the WHO Anatomical Therapeutic Chemical (ATC) classification system. They afterwards were classified into five main categories: chemotherapies, targeted therapies, immunotherapies, hormonal therapies and others.20 Records with missing information on any of the above characteristics were excluded. Duplicate entries were identified and removed. Procurement records with zero or implausible quantities or prices were removed following standard pharmacoepidemiologic practice. Data cleaning was performed using Microsoft Excel 365. Statistical Analysis Because Defined Daily Doses (DDDs) are not applicable to chemotherapy agents, average drug prices were calculated per milligram of active ingredient. For fixed-dose combinations (eg, tegafur + gimeracil + oteracil potassium or tegafur–uracil), the milligram calculation was based on the sum of all active components (eg, 100 mg + 224 mg = 324 mg). All prices were adjusted for inflation using Vietnam’s Consumer Price Index (CPI) from 2019 to 2023 to Vietnamese Dong 2023 and then converted into US dollars based on the World Bank’s 2023 annual average exchange rate ($1 = 23,787.32 VND).21 Statistical analyses were conducted using R version 4.4.3. Descriptive statistics were reported as counts and percentages for categorical variables, and as means, standard deviations, medians, interquartile ranges (Q1–Q3), minimums, maximums, and 95% confidence intervals (using the 2.5th and 97.5th percentiles) for continuous variables. Interfential statistics were performed including weighted t-test for comparisons between drug price at national-bidding level and provincial-bidding/hospital-bidding level as well as across bidding technical levels; log-linear regression to evaluate the fluctuations of drug price, consumption, distribution of bidding technical groups within the total number of milligram of active ingredients each year throughout the given period and between drug prices and bidding technical groups; log-log regression to evaluate the correlation between drug consumption and the total number of active. In the log-linear and log–log regression, the group of brand-name drugs was encoded with 1 and generic drug groups 1 to 5 were encoded with numbers 2 to 6, respectively. Patients and Public Involvement Patients or the public were not involved in the design, conduct, reporting or dissemination plans of this research. Results Characteristics of Cancer Medication Consumptions in Vietnam Table 1 presents result about the number of active ingredients and total expenditures for cancer medications throughout the period of 2019–2023 while Supplementary Material 1 provides figures for each year. In terms of number of active ingredients, there were in total 77 different active ingredients consumed in five years and generally consistent across years and over the following period. A total of 77 active ingredients has been recorded with the largest number for chemotherapies with 41 (53.2%) active ingredients, followed by targeted therapies with 13 (16.9%), hormonal therapies 12 (15.6%), immunotherapies (10.3%) and pamidronate was the only classified into the “Others” group. Thirty-seven (48.1%) active ingredients were tendered in the brand-name drug bidding packages while all active ingredients were included in five generic drug bidding packages. Table 1Active Ingredient and Consumption Characteristics of Cancer Medication Between 2019 and 2023YearNumber of Active Ingredients (%)Total Consumption in 2019–2023 ($) (%)Overall77 (100.0)1,405,731,510 (100.0)Cancer medication group Chemotherapies41 (53.2)517,381,939 (37.0) Targeted therapies13 (16.9)645,685,678 (46.2) Hormonal therapies12 (15.6)63,841,375 (4.7) Immunotherapies10 (13.0)176,390,185 (12.0) Others1 (1.3)2,432,333 (0.2)Bidding technical group Brand-name37 (48.1)798,819,617 (55.7) Generic group 1 (G1)67 (87.0)316,088,775 (23.2) Generic group 2 (G2)50 (64.9)110,058,498 (8.0) Generic group 3 (G3)33 (42.9)34,016,034 (2.3) Generic group 4 (G4)37 (48.1)40,712,585 (3.1) Generic group 5 (G5)60 (77.9)106,036,001 (7.8)Bidding organisation unit Ministry of Health23 (29.9)615,679,283 (40.8) Provincial Department of Health74 (96.1)233,749,427 (17.4) Hospitals75 (97.4)556,302,801 (41.7)Orgin of manufacturer Vietnam42 (54.5)76,373,880 (5.5) Europe67 (87.0)1,005,172,040 (71.1) Asia51 (66.2)210,293,849 (15.4) Australia7 (9.1)51,057,719 (3.5) America14 (18.2)62,834,022 (4.5)Route of administration Oral32 (41.6)634,791,403 (44.6) Injection51 (66.2)770,940,107 (55.4)Bidding contract duration ≤1 year77 (100.0)853,597,723 (63.5) > 1 - ≤ 2 years59 (76.6)552,125,515 (36.5) >2 years3 (3.9)8273 (0.0) Active Ingredient and Consumption Characteristics of Cancer Medication Between 2019 and 2023 On the other hand, only 23 (29.9%) active ingredients with large consumptions were purchased by the National Center for Drug Procurement of Ministry of Health, being different from the provincial and hospital levels where almost active ingredients were included in the bidding process (74 (96.1%) and 75 (97.4%), respectively). Regarding the origin of manufacturer, while all active ingredients contained products from foreign manufacturers, there were 42 (54.2%) active ingredients that had products manufactured in Vietnam. Thirty-two (41.6%) were available for oral use and 51 (66.2%) for injection use. Bidding contract duration of 1 year was the most popular option for all active ingredients, followed by 1–2 years with 59 (76.6%) and only 3 (3.9%) have this duration longer than 2 years. In terms of expenditure on cancer medications throughout this period, the total figures for five years were above $1.4 billion, which increased from $286.4 million in 2019 to $322.8 million in 2020 before dropping in the following two years to $251.6 million in 2021 and $185.4 in 2022 then rebounded to $359.5 million and reached its peak in 2023 (Figure 1). The largest proportion of cost across five years belonged to the targeted therapy group, making up about 44–48% of the total figure every year and the second group was the chemotherapy group, ranging from 34% to 39%. Immunotherapies were consistently in the third place with around 8–16%. Consumption numbers for hormonal therapies and others were minimal with only 4–6% (Figure 2). Total cancer medication consumption between 2019 and 2023 categorised by medication groups (WHO). Total cancer medication consumption between 2019 and 2023 characterisied by bidding technical groups and bidding organisation units. In terms of consumption categorised by bidding technical groups, costs for the brand name and generic group 1 took about 79% the total figures with $798.8 million (55.7%) and $316.1 million (23.2%), respectively, while costs for the other four groups only made up about 20–25% of the total expenditure. Nevertheless, proportions for brand-name medications decreased significantly from 68.3% in 2019 to 44.9% in 2022 before rising to 53.4% and $195.3 mlliion in 2023. Costs for generic group 1 medications increased from $42.9 million (15.0%) in 2019 to $77.2 million (30.7%) in 2021 then fell to $51.3 million in 2022 before recovering in 2023 with $75.1 million (20.9%). Noticably, a significant shift in procurement structure was observed. MOH-level bidding accounted for 55.4–65.9% of chemotherapy expenditure in 2019–2020 but dropped to 28.8% in 2021 and 12.8% in 2022 before rising to 41.2% in 2023. Consequently, provincial and hospital-level bidding expanded markedly, reaching their highest proportion in 2022. Products manufactured in Europe contributed to about $1.0 billion (71.1%) of the total expenditure, while domestic products only made up about 5–6% across years. The value of oral products was larger than injections in 2019 and 2021 but being reversed in the last three years when the value of injection products were about 59–61%. Finally, the value of products with bidding contract duration >1 to ≤2 years dominated in the first two years before being replaced by those with ≤1 year in the following two years, particularly in 2022 when it was only 6.7%. Despite recovering in the final year of 2023, they only were 27.4% of the total value and lower than before. Consumptions and Prices by Active Ingredients and Route of Administrations Supplementary material 2 provides results about total consumption and descriptive statistics for price per milligram of active ingredient; weighted t-test for mean price per milligram of active ingredient differences between products tendered by MOH and provincial/hospital levels across bidding technical groups, brand-name and generic group 1 products; log-linear regressions illustrating correlations between consumptions and years, average prices per milligram of active ingredients and years, the distribution of bidding technical groups and years, average prices and bidding technical groups as well as log–log regression between consumptions and total mg of active ingredients for each agent in the database. Supplementary material 3 showed graphs about total consumption, total mg of active ingredient categorised by bidding technical groups and bidding organisation units and average price per milligram of active ingredients for each agent throughout the period. Table 2 summarises those results for top ten agents with largest expenditure during that time. Table 2Summary of Key Statistical Analyses of Top ten Agents with Largest Expenditures Between 2019 and 2023Active IngredientRouteTotal Consumption (%)Log (Price per miligram of Active Ingredient) ~ YearLog (Bidding Technical Group Distribution) ~ YearLog (Price per miligram of Active Ingredient) ~ Bidding Technical Groupβ (SE)Adjusted R2β (SE)Adjusted R2β (SE)Adjusted R2ImatinibOral142,175,945 (10.0)−0.5 (0.067)**0.9324−0.04 (0.004)**0.9604−0.01 (0.006)0.3802TrastuzumabInjection121,644,100 (8.6)−0.14 (0.011)**0.97700.11 (0.04)0.6497−0.07 (0.056)0.1795BevacizumabInjection120,255,002 (8.6)−0.1 (0.06)0.3309−0.03 (0.022)0.1909−0.32 (0.051)0.9499RituximabInjection94,535,904 (6.9)−0.18 (0.024)**0.93190.05 (0.014)*0.7106−0.12 (0.092)0.1618TacrolimusOral94,457,329 (6.1)−0.03 (0.006)*0.88590.02 (0.012)0.15120.04 (0.212)−0.2385PaclitaxelInjection60,100,719 (4.2)−0.01 (0.014)−0.1190−0.01 (0.011)0.1419−0.19 (0.141)0.1528VinorelbinOral58,083,460 (3.9)−0.04 (0.01)*0.80000 (0.035)−0.33270 (0.138)−0.4999OxaliplatinInjection55,208,760 (3.9)−0.03 (0.061)−0.25590.15 (0.019)**0.9375−0.42 (0.233)0.3063MycophenolatOral51,724,428 (3.6)−0.06 (0.007)**0.94280.08 (0.013)**0.89880 (0)0.6469DocetaxelInjection49,641,404 (3.4)−0.01 (0.053)−0.32370 (0.016)−0.3188−0.59 (0.338)0.295Notes: *0.01 < p < 0.05, **0.001 < p < 0.01, ***p < 0.001.Abbreviations: β (SE), coefficient (standard error); %, percentage to total consumption. Summary of Key Statistical Analyses of Top ten Agents with Largest Expenditures Between 2019 and 2023 Notes: *0.01 < p < 0.05, **0.001 < p < 0.01, ***p < 0.001. Abbreviations: β (SE), coefficient (standard error); %, percentage to total consumption. Among ten agents with the highest expenditure, four with the highest figures were targeted therapies, namely imatinib (oral), trastuzumab (injection), bevacizumab (injection) and rituximab (injection), five chemotherapy agents: tacrolimus (oral), paclitaxel (injection), vinorelbin (oral), oxaliplatin (injection), docetaxel (injection) and one immunotherapy agent of mycophenolat (oral). Imatinib (oral) contributed the largest share over five years ($142.2 million, 10.0%), followed by trastuzumab (injection) with $121.6 million (8.6%) and bevacizumab (injection) with $120.3 million (8.6%). However, proportion of expenditures for imatinib dropped continuously from 16.3% in 2019 to 3.8% in 2023 and their first place was replaced by bevacizumab in 2021, 2022 and 2023 (9.7%, 11.0% and 13.1%, respectively). There was no clear trend on the consumption for most products throughout the years since p-value for the log-linear regression was >0.05 in most cases. For those with large consumption, only imatinib (oral) witnessed a significant decrease (coefficient (standard error (β (SE)): −0.390 (0.0960), p = 0.028) and an increase for bevacizumab (injection) (β (SE): 0.300 (0.0760), p = 0.030). Overall, consumption was highly correlated with total milligram of active ingredients across agents. Nevertheless, in the top ten, there were no stastically correlations for imanitib (oral) (β (SE): −1.900 (1.3840), p = 0.262) and rituximab (injection) (β (SE): 0.060 (0.4370), p = 0.896). In terms of imanitib, although the amount of consumption was relatively stable and even peaked in 2023 with over 600 million milligrams of active ingredients, a sharp decrease in price for brand-name and generic group 1 products made the total expenditures dropped significantly between 2021 and 2023 compared to the first two years. Similar patterns were observed with rituximab since the price for brand-name products fell continuously. Gemcitabine was the only case showing a negative correlation (β (SE): −0.860 (0.2660), p = 0.048) between two variables since prices for generic products were far lower than brand-names alongside with changed toward more affordable products over time. The weighted t-test for price across different bidding organisation units generally indicated that within a bidding technical group, medications procured by the national levels were generally cheaper or no expensive than provincial/hospital levels with eight out of ten products with largest expenditure following that pattern. For example, the average price per milligram of active ingredients for trastuzumab (injection) procured by MOH was 45.5% lower than those procured by provincial/hospital units (mean difference: −1.636, p = 0.001), 22.3% lower for bevacizumab (injection) (mean difference: −0.777, p = 0.001); this pattern was also observed across all bidding technical groups for tacrolimus (oral), paclitaxel (injection) and oxaliplatin (injection). The price for imanitib and mycophenolat from the national levels were higher than two others, though it should be noted that their prices lowered over time. Meanwhile, although prices of brand-name drugs in most agents were significantly higher than generic group 1, trastuzumab (injection), tacrolimus (oral) showed that the generic group prices were slightly higher and even statistically different for cetuximab (injection) and everolimus (oral). This is because for those agents, lots of brand-name medications were procured under generic group 1 bidding packages. Downward trends in average prices over time were observed by most agents and none of them showed a significant upward trend. In the top ten agents, imatinib (oral), trastuzumab (injection), rituximab (injection), tacrolimus (oral), vinorelbin (oral) and mycophenolat (oral) indicated stastically negative correlations. They came along with moves towards lower bidding technical groups for the majority of agents, except imatinib (oral) since its price had already fallen overtime. However, there were only a few agents showing strong correlations between prices and bidding technical groups and none of them included in the top ten above. Discussion This study analyses cancer medication consumption in Vietnam between 2019 and 2023 based on the national database of DAV. Results showed that the total expenditure over the five-year period was about $1.4 billion, of which targeted therapies made up the largest proportion and followed by chemotherapies while figures for the other groups were relatively minimal. Consumption generally declined during the COVID-19 period before rebounding to its peak in the final year of 2023, alongside changes in bidding values from MOH compared to provincial/hospital units. Costs for brand-name and generic group 1 products accounted for the vast majority of the total expenditure. Drug prices generally decreased over time with trends toward lower bidding technical groups as well as the MOH bidding organisation unit was observed as a more efficient option compared to the others. Unlike other studies in previous periods showing a continuous increase in the total DDD and expenditure on cancer medications,22,23 our study indicates a significant downward trend throughout the COVID-19 pandemic and only recovered afterwards. A study in China24 showed that the total number of DDDs for anticancer medicines in 2019 nearly doubled the number in 2014 and the total expenditure was even about 2.2 times larger. Similar trends were also witnessed in Taiwan25 and other Southeast Asia countries,24 though their increases were much smaller. In our study, total cancer consumption dropped to only above half between 2020 and 2022, despite the fact that the number of diagnosed cases rose continually during that period. The COVID-19 pandemic has severely disrupted global pharmaceutical supply chains due to widespread lockdowns and international travel restrictions. These disruptions have contributed to delays or interruptions in access to essential cancer treatments for many patients. This underscores the critical importance of strengthening supply chain resilience for essential medicines to ensure continuity of care during emergencies such as pandemics, natural disasters, or other large-scale disruptions. From a policy perspective, these findings highlight the need for national strategies that incorporate emergency stockpiles, diversified sourcing of critical oncology medicines, improved forecasting systems, and more flexible procurement mechanisms that can rapidly respond to shortages. Although Vietnam has recently introduced several policies to promote domestic production of pharmaceuticals as a way to mitigate supply chain risks, the impact to date has been limited due to challenges in technological capacity, investment, and quality assurance.26 Strengthening coordination between central authorities, suppliers, and healthcare facilities therefore remains essential to maintain uninterrupted access to life-saving treatments. The four agents with largest expenditures were all targeted therapies: imatinib, trastuzumab, bevacizumab and rituximab, though many cost-effectiveness analyses on them showed that their efficacy was not commensurate with their price.27–29 The study by Nguyen Quynh Anh et al13 indicated that the cost of Herceptin (trastuzumab) 150 mg and 450 mg vials needed reducing by 56% and 54%, corresponding to become cost-effectiveness in Vietnam. Moreover, their prices in Vietnam were much higher than nearby countries. For example, the price of generic drug imatinib 100 mg in China was only $1.6130 while this number in Vietnam in 2023 was about $2.00 despite having much lower GDP per capital. A similar comparison was also witnessed for trastuzumab which prices for biosimilar products in Thailand was around $16031 compared to $500–2.000 in Vietnam. As both medications have already been listed in the reimbursement list of the national health insurance fund, this reality posed an urgent need to conduct and apply the result of health technology assessments to adjust drug price meeting with the Vietnamese willingness-to-pay threshold, therefore ensuring equal access to healthcare services among different patient groups. Our results also indicated important advantages of organising centrailised national drug procurement by the MOH and limitations on previous and current drug bidding regulations in Vietnam. Within a bidding technical group, the price of medications tendered by MOH was frequently lower than provincial/hospital bidding units since the amount of medicine was far larger. Furthermore, similar to several studies in LMICs,32,33 our findings also showed that there was a significant variation in drug prices purchased by different provinces or hospitals, even for the same product. This reality suggests that centralized drug procurement would help reduce delays in bidding, reduce drug shortages and bureaucratic procedures. Another important trend should be noted that the drug prices mostly reduced over time between 2019 and 2023. This is because bidding regulations before 2023 required that the bidding price across bidding technical groups were not higher than the previous year even without adjustments for inflations. This problematic regulation has been removed with the Circular 06/2023/TT-BYT, allowing healthcare facilities to become more flexible when choosing medications matching with treatment needs. In addition, our findings suggest that the current classification system is rather ineffective in achieving the intended goals. For instance, certain products such as mecaptopurin (oral), daunorubicin (injection) or methotrexat (oral) seeing the average price per mg of active ingredient of generic group 5 drugs even exceeds those of higher groups, although different dosage strengths (eg, 10 mg, 20 mg, 50 mg) do not always scale in price according to active-ingredient content, due to variable production and distribution costs. This came from the fact that current groups are only classified by technical specifications without considering the dosage form (eg: extented-released tablets or solutions for oral use). Another issue is that many products were bidding in the generic group 1 despite being brand name to avoid the usage quotas on branded drugs at some facilities. Therefore, a promising solution is to establish a national drug-price database in addition to reforming the technical classification in the drug bidding process. This would enable procurement units to set more sensible benchmark prices and quotas when establishing bidding packages. To our knowledge, this is the first study in Vietnam—and among the few globally—to examine nationwide consumption patterns of cancer medication. The findings offer important evidence on the consumptions of high-cost oncology medicines, address the importance of revising reimbursement, price and purchasing policies on those under question about their economic efficiency. Our study also demonstrates how bidding mechanisms influence the utilization of cancer medications, thereby providing a useful basis for refining procurement regulations and improving system efficiency. Nevertheless, several limitations should be acknowledged. First, the analysis may be affected by selection bias due to incomplete publicly reported procurement data,18 which could underestimate or misrepresent actual consumption. Second, for multi-year bidding packages, we assumed stable consumption across time. Real-world utilization may fluctuate considerably with changes in disease burden, treatment protocols, or supply constraints although most bidding packages were not longer than 2 years. Third, the price metric in our study is based on milligrams of active ingredients without considering differences in dosage forms within oral or injection medicines, which heavily affected prices. However, to our knowledge, this might be the most sensible solution because the DDD is not appropriate or precise to quantify cancer medication consumptions. Fourth, due to the limitation of the database in Vietnam, we are unable to explore the consumption across different types of cancers. Finally, the study focused solely on public hospitals, excluding private healthcare providers whose role in cancer care is rapidly expanding in Vietnam. These limitations highlight the need for more comprehensive and standardised national reporting systems to better inform oncology policy and resource planning. Conclusion Cancer medication consumption in Vietnam fluctuated substantially between 2019 and 2023, accompanied by significant changes in procurement patterns and the pandemic. Changes in purchasing organisations and strategies generally affected the consumption and price of those products. Strengthened and targeted policies are required to optimize the economic impact of this group of medication, while broader efforts are needed to alleviate the cancer burden in Vietnam. Funding Statement No funding was received for this project. Abbreviations β (SE), coefficient (standard error); ATC, Anatomical Therapeutic Chemical; CPI, Consumer Price Index; DAV, Drug Administration of Vietnam; DDDs, Defined Daily Doses; EMA, European Medicines Agency; EU, European Union; GLOBOCAN, Global Cancer Observatory Research; GMP, Good Manufacturing Practice; HICs, high-income countries; LMICs, low- and middle-income countries; MOH, Minister of Health; NCCN, National Comprehensive Cancer Network; PIC/s, Pharmaceutical Inspection Co-operation Scheme; Q1 – Q3, quartile 1st – quartile 3rd; SRA, Stringent Regulatory Authorities; US$ ($), United States Dollars; VND, Vietnamese Dong; WHO, World Health Organization. Data Sharing Statement All data used in the study are publicly available at https://dav.gov.vn/thong-tin-dau-thau-thuoc-cn8.html. Ethics Approval and Informed Consent This study used publicly available, aggregated procurement data without any individual-level information. Therefore, ethics approval was not required under Vietnamese regulations or international guidelines. Informed consent is not applicable since this study used public databases. Author Contributions All authors made a significant contribution to the work reported, whether that is in the conception, study design, execution, acquisition of data, analysis and interpretation, or in all these areas; took part in drafting, revising or critically reviewing the article; gave final approval of the version to be published; have agreed on the journal to which the article has been submitted; and agreed to be accountable for all aspects of the work. Disclosure The authors report no conflicts of interest in this work. References Associated Data Data Availability Statement All data used in the study are publicly available at https://dav.gov.vn/thong-tin-dau-thau-thuoc-cn8.html.