Cost and availability of selected medicines after implementation of increased import verification fees

This study assessed the impact of the 12% verification fees on cost and availability of selected essential medicines in Uganda. This was a cross sectional study and both qualitative and quantitative data were collected. Quantitative data were collected using a checklist. An in-depth interview guide was used to collect qualitative data from Key informants. The study was carried out in the seven (7) National Drug Authority (NDA) regions of Uganda i.e. Central, Eastern, South Eastern, Northern, West Nile, Western and South Western. The study included wholesale and retail pharmacies in each of these regions. By 2019, Uganda had a total of 1,735 registered pharmaceutical outlets according NDA management information System. These included 539 wholesale pharmacies and 1196 retail pharmacies [11]. The study included licensed wholesale and retail pharmacies that existed for at least five years and the managers of the manufacturing facilities and other key informants from relevant Ministries, Departments, Agencies and pharmaceutical importers. This study was part of a bigger study that also assessed the impact of the verification fees on local production capacity of pharmaceuticals.

Using the Slovin’s formula for sample size calculations, n = N/ (1 + Ne²), the calculated minimum sample size for the pharmacies was 326. The total number of wholesale and retail pharmacies N = 1735, e = maximum allowable error = 5%. Sample size n = 1735/ (1 + 4.337) n = 326 (number of wholesale and retail pharmacies). The number of wholesale and retail pharmacies and their distribution in the different regions was proportionately determined as shown in Tables 1 and 2. In each region, a city and two other districts were selected. The two other districts were selected based on the existing number of pharmacies in each district in the region. Furthermore, the pharmacies selected were those that had been in existence for more than five (5) years. These were selected by simple random sampling. For regions with more than one city, simple random sampling was used to select the city. For the selected districts and cities, the number of the pharmacies was also proportionately determined. For each district, the pharmacies were randomly selected. Excel was used to generate the random sample of pharmacies.

The final sample size used in this study was n = 328.

The key informants were purposively selected to participate in the study. These included 7 Key informants from 2 pharmaceutical industries manufacturing any of the selected 37 essential medicines, 1 from Uganda Pharmaceutical manufacturers association,1 from Uganda Pharmacy owners Association, 1 from an importing company of the selected medicines and 2 from the central medical stores.

At the retail and wholesale pharmacies, quantitative data were collected using a pre-tested data checklist. Information about the cost and availability of the selected medicines were obtained. The In-depth semi-structured interview guide that was used for the key informant interview was developed from available literature (S11). The key informant interviews lasted 45–60 min and were conducted by the principal investigator with the help of a research assistant. Through the interviews, the views and perceptions of the respondents was sought regarding the impact of the increase in verification fees on cost and availability of the medicines and any challenges and recommendations regarding the import verification fees increment.

Data was cleaned to ensure that all relevant and correct data were collected. The principal investigator (PI) and study team oversaw accuracy and completeness of all data entered on the checklist before submission for data entry. A web based Open Data Kit (ODK) tool was used for data entry and a number of range checks, logical skips, missing data checks were incorporated at data entry level. The data sets entered were cross referenced and errors, inconsistencies were resolved by checking against the source documents after which one data set was produced. No names were used; identification codes were allocated to each facility checklist. Pre-testing of data collection tools was done in five pharmacy outlets. The in-depth interviews were audio recorded and transcribed later to ensure all information given by respondents were not lost. In addition to audio records, notes were also taken during the interviews. Data collection was done by trained research assistants supervised directly by a member of the lead research team. Computers used for data entry were password protected.

The quantitative dataset was transferred from the ODK tool to Microsoft Excel 2017 for further cleaning and validation. Data analysis was done using SPSS V25.0. The median prices before and after the introduction of the 12% increase were computed. Paired sample T-Test was used to test if there is a significant difference in these prices before and after the 12% increase. International reference unit prices were obtained from Management Sciences for Health (MSH) International Medical Products Price Guide 2015 [12]. For conversions, 1 US$ = 3670 Ugandan Shillings (Uganda Revenue Authority (URA) median exchange rate from 2017 to 2020). Price adjustment for inflation/deflation was taken at 5.63 in 2017 (Uganda Bureau of Statistics (UBOS) Consumer Price Index (CPI) Publication, 2017) and 3.79 in 2020 (UBOS CPI Publication, 2021). The international comparisons of median unit prices were computed using median price ratios (MPR) as follows;

Using the Consumer Price Indices (CPI) for 2017 and 2020, the median unit prices were adjusted for inflation/deflation using the following formula.

Therefore, a deflation rate of 33% was used to adjust the median unit prices.

Availability of medicines were captured as the average percentage of the medicines that were available at the time of data collection. These were aggregated to get the average availability of all the drugs by region, Vital, Essential and Necessary (VEN) classification and by formulation and these were compared across the pharmacy type [Retail, wholesale and dual (both retail and wholesale)]. Analysis of Variance (ANOVA) was used to test the difference in the average availability and whether there was a difference in availability by region, VEN classification and by formulation. Availability of medicines was also categorized based on WHO categorization into Very low (< 30%), Low (30–49%), fairly high (50–80%) and high (> 80%).

The qualitative data were thematically analyzed following the steps proposed by Braun and Clarke which included transcription, coding of the data, and generation, reviewing and naming of themes [13]. The voice records were transcribed verbatim as text and the transcripts were then compared with the notes taken during the interviews and any missing information due to unclear audio recordings were added. The next step was to read the transcript so as to get familiar with all responses from the interviews. Meanings were then generated from transcribed data, during a second perusal of the transcribed data, following which emergent sub themes were developed. This was then followed by categorization and alignment of each of the quotes with their respective sub themes, and finally, with their predetermined themes. Precedence of a given response was established based on the recurrence of responses. Emerging quotes from the interviews were highlighted and marked for referencing and reported verbatim.

This study involved 328 pharmacy outlets and seven key informants. The key informants included 2 members of the Uganda pharmaceutical manufacturers association, 2 people from the central medical stores, 2 members of the Uganda pharmacy owners’ association and 1 person from a pharmaceutical importing company.

The characteristics of surveyed pharmacies are presented in Table 2. Retail pharmacies accounted for a large proportion (n = 223, 68.0%) of the pharmacies and majority of the surveyed pharmacies had been in existence for 5 to 10 years (n = 256, 78.0%).

The main objective of a national medicine policy is to ensure the availability, affordability, accessibility and rational use of essential medicines that are safe, effective and quality assured. As one of the means to achieving this, governments in LMICs support local medicine production, expecting it to result in increased availability and lower prices for medicines in addition to industrial and economic benefits [14]. A few studies have highlighted the impact of local production on availability and prices of medicines unlike the general information on medicine availability and prices where extensive literature exists.

In the current study, the overall availability of imported medicines (54.8%) was higher than the local medicines (37.1%). A study comparing the price and availability of locally produced and imported medicines in Tanzania and Ethiopia found a similar pattern in Tanzania but the reverse in Ethiopia [15]. In Tanzania, it was found that availability of local and imported products was 21% and 70%, respectively. From the same study, the availability of local and imported products was 54% and 35%, respectively in Ethiopia. Unlike this study, the study conducted in Tanzania and Ethiopia assessed availability in both private and public sector.

Locally produced parenteral preparations (54.6%) were the only formulations more available than imported products. This could be due to the fact that Uganda has a local manufacturing facility specifically for production of parenteral products and the ability of the industry to meet the market demand and their wide spread distribution network.

The availability of local brands was mainly low (45%) and very low (41%) while the imported brands were mainly fairly high (74%). Generally, 65% of the brands available in the surveyed pharmacies were imported. This means that the imported products still dominate the market after the imposition of the verification fees. This is in agreement with previous literature which suggests that the Ugandan market is dominated by imports which constituted close to 90% [16] of Essential Medicines and Health Supplies in the country [17]. On the other hand, it also indicates a reduction in dominance of the Ugandan market with the imports of the selected medicines and improved availability of locally produced medicines since the dominance has reduced to 65%. However, a better indication of improvement or no improvement in availability would have been shown by changes in availability before and after the introduction of import verification fees. This was not possible in this study because of inadequacies in record keeping.

Very few local (4.7%) and imported (25.6%) medicines met WHO global action plan target of 80% availability of essential medicines by 2025 in public and private sectors. A baseline assessment of WHO’s target for availability of essential medicines revealed 18.9% and 5.2% availability of the lowest-priced generics and originator brands respectively that met WHO’s target in private sector in low income countries [18]. Although local and imported medicines may not be comparable to lowest priced and originator brands, the lowest priced generics in this study can be compared to the local medicines since most of them had the lowest prices. Therefore, the availability of local medicines was much lower compared to the findings of the baseline assessment study. A similar trend was reported by a study conducted in Pakistan that suggested poor availability of lowest priced generics (20.3%) and originator brands (55.0%) in private sector facilities [19].

The key informants reported no increase in availability of the selected products. Some of the reasons noted by key informants contributing to this picture included lack of capacity of local manufacturers to satisfy the local market, sourcing of raw materials from other countries and failure to process and deliver orders timely. It was reported by key informants that the high cost of importation discouraged importation of some of the products which affects availability of the products.

Tariffs such as import verification fees are vital determinants of prices and they can considerably increase the prices of imported goods [7]. According to WHO, taxes account for 20 to 30% of the final price patients pay for medicines [19]. A number of countries have adopted removing or reducing taxes and tariffs on medicines as a strategy of realizing a decrement in the retail cost and subsequent affordability and availability of medicines to the final consumer [20, 21]. The Ugandan government’s initiative to increase import verification taxes on 37 selected medicines from 2 to 12% while leaving locally manufactured products at fees of 2% was expected to increase the prices of imported medicines giving comparative price advantage to the locally produced products. However, the median prices of both local and imported medicines generally increased or remained the same after the introduction of import verification fees. The increase in median price of local products was significant for only one product while for the imported products, the increase was significant for four (4) products and the median prices of local products were generally lower than the imported products. Considering lowest priced medicines, the prices of 16 (36.4%) of the local and 13 (29.5%) of the imported medicines remained the same and the increase in price of local products was significant for 9 (20.5%) of the products while for the imported products, the increase was significant for 7 (15.9%) of the products. Whereas it makes logical sense to expect decrease or no increase in cost of locally produced medicines accruing from economies of scale and price advantage, the increase in verification fees has not resulted in reduction in prices of locally produced medicines. This was also noted by most of the key informants who reported that, the prices of the selected locally manufactured medicines remained the same or increased after the increment in verification fees but remained competitive in the market. Similarly, in Peru, removal of indirect taxes on a range of cancer medicines and antiretroviral medicines resulted in little reduction in prices [22].

As per the findings and observation of the key informants, the prices of medicines were within acceptable international prices. The MPR for most of the medicines were within the acceptable range of 3 or less of private sector for both local and imported medicines. WHO considers patient prices to be high when MPRs exceed four (4). The medicines with high MPR before remained high. However, the number of medicines with MPR (1 or more) were more in 2020 compared to before introduction of verification fees in 2017 for both imported and local brands. In 2020, the median unit prices of 12 (28.6%) locally produced medicines and 20 (47.6%) imported medicines were higher than the international median unit prices compared to 11 (25.0%) locally produced medicines and 20 (45.5%) imported medicines before introduction of import verification fees. This is an indication of increase in prices of medicines, high prices paid by patients to access these medicines and imported medicines having higher median prices. The study comparing the price and availability of locally produced and imported medicines in Tanzania and Ethiopia found a similar pattern of higher median price ratios for imported medicines compared to local produced medicines [15]. From the study, patient prices for local products (median MPR = 1.85) were lower than imported products (median MPR = 5.42) in Ethiopia but almost identical in Tanzania (Median MPR = 2.27 for locally produced products and Median MPR = 2.29 for imported products). However, this study could not establish whether these prices are a reflection of manufacturer’s selling prices. It is possible that local manufacturers may be selling at lower prices but add-ons and manipulation of prices by distributors increase the prices making the medicines expensive for the patients. This study didn’t assess the medicine price components such as manufacturer’s selling prices, mark-ups and other add-ons in the supply chain that make up the final patient price.

It is therefore important as governments support local production, it should as well assess price components and then regulate the markets to ensure their support results in more affordable medicines for patients [23]. For example South Africa adopted a Single Exit Price (SEP) mechanism that bans discounts and rebates and provides transparent information about the prices of medicines sold in the private sector [24].

The increase or no change in prices for local products was attributed to export subsidies from the countries where products are imported reducing the competitive power of local manufacturers in the market, presence of local manufacturers who are importing at the same time, high cost of raw materials and lack of price control in the market. It was noted that even at 12% verification fees, some importers were still importing some of the 37 capped products from some countries cheaply where they are given export subsidies. According to the key informants, presence of local manufacturers who also import medicines led to monopolistic tendencies where few players with high volumes control the market and manipulate prices without interference. The increase in prices for imported products was attributed to increase in verification fees.

High prices of medicines reduce utilization particularly by the poor and elderly, and reduce compliance with preventive and chronic disease treatment regimens leading to poor health outcomes. Therefore, in promoting local production, dual policy objectives need to be explored. In the short to medium term, governments need to develop and implement policies through which they can continue to support local production but, at the same time, prevent high prices being passed on to patients.

Because of poor records, only point availability was determined therefore comparison of availability of medicines before introduction of the verification fees was not established. The assessment was limited to the private sector; public sector pharmacies were not included. However, since all government facilities are supplied by only one supplier, the prices of medicines in public pharmacies would be uniform throughout Uganda and the perspectives of key informants from public institutions were sought. It is possible that COVID-19 pandemic might have had an impact on the findings but shouldn’t have been significant since most of the time period 2017, 2018 and 2019 were not affected by the pandemic and supply chain activities continued throughout COVID-19 pandemic.