Sten Olsson

A Bayesian neural network method for adverse drug reaction signal generation

AbstractObjective: The database of adverse drug reactions (ADRs) held by the Uppsala Monitoring Centre on behalf of the 47 countries of the World Health Organization (WHO) Collaborating Programme for International Drug Monitoring contains nearly two million reports. It is the largest database of this sort in the world, and about 35 000 new reports are added quarterly. The task of trying to find new drug–ADR signals has been carried out by an expert panel, but with such a large volume of material the task is daunting. We have developed a flexible, automated procedure to find new signals with known probability difference from the background data. Method: Data mining, using various computational approaches, has been applied in a variety of disciplines. A Bayesian confidence propagation neural network (BCPNN) has been developed which can manage large data sets, is robust in handling incomplete data, and may be used with complex variables. Using information theory, such a tool is ideal for finding drug–ADR combinations with other variables, which are highly associated compared to the generality of the stored data, or a section of the stored data. The method is transparent for easy checking and flexible for different kinds of search. Results: Using the BCPNN, some time scan examples are given which show the power of the technique to find signals early (captopril–coughing) and to avoid false positives where a common drug and ADRs occur in the database (digoxin–acne; digoxin–rash). A routine application of the BCPNN to a quarterly update is also tested, showing that 1004 suspected drug–ADR combinations reached the 97.5% confidence level of difference from the generality. Of these, 307 were potentially serious ADRs, and of these 53 related to new drugs. Twelve of the latter were not recorded in the CD editions of The physicians Desk Reference orMartindales Extra Pharmacopoea and did not appear in Reactions Weekly online. Conclusion: The results indicate that the BCPNN can be used in the detection of significant signals from the data set of the WHO Programme on International Drug Monitoring. The BCPNN will be an extremely useful adjunct to the expert assessment of very large numbers of spontaneously reported ADRs.

The Role of the WHO Programme on International Drug Monitoring in Coordinating Worldwide Drug Safety Efforts

The rationale for setting up the WHO International Programme for Adverse Reaction Monitoring, 30 years ago was to make it possible to identify rare adverse drug reactions (ADRs) that could not be found through clinical trial programmes. It became evident that maintaining an international database of ADR case reports and a network of institutions and scientists concerned with drug safety issues provides great additional gains when compared with operating in isolation. Thus, the scope of the WHO programme has expanded over time to accommodate the expansion of the field of drug safety monitoring, now often named pharmacovigilance. The international centre, the WHO Collaborating Centre for International Drug Monitoring in Uppsala [now known as the Uppsala Monitoring Centre (UMC)], maintains the international database and serves the national centres associated with the WHO programme; however, the role of the centre is expanding allowing it to play a leading role in global drug safety monitoring.The national centres are appointed by the governments of each of the countries participating in the WHO programme. These centres are responsible for collecting spontaneous ADR reports originating from health professionals. 49 countries are currently contributing case information and are full members of the programme; an additional 11 countries have applied for membership but have still not submitted any reports. The annual influx of reports is currently fluctuating at around 150 000 reports.In its development, the data collected by the WHO programme was guarded by strong rules of confidentiality. In some member countries, however, case data, with the important exception of reporter and patient identities, has always been public information. The UMC has made it a priority to try to create an atmosphere of openness and trust between all parties involved in drug safety assessment, which will eventually enable general sharing of available data and an extended analysis and use of the data collected. The WHO network represents the wealth of competence and experience that is at the disposal of countries wishing to join the international pharmacovigilance community.

Pharmacovigilance Activities in 55 Low- and Middle-Income Countries A Questionnaire-Based Analysis

Background: The WHO Programme for International Drug Monitoring aims to develop a comprehensive global pharmacovigilance strategy that responds to the healthcare needs of low- and middle-income countries. However, first there is a need to measure and understand existing conditions and pharmacovigilance initiatives intended in these settings. Very few investigations have carried out such a systematic assessment of the pharmacovigilance landscape in recent years in low- and middle-income countries.Objective: To assess current and planned pharmacovigilance activities in low-and middle-income countries, identify gaps and the most urgent pharmacovigilance priorities at national and international levels, and define the elements of a sustainable global pharmacovigilance strategy.Methods: A standardized questionnaire was sent to 114 representatives of countries participating in the WHO Programme for International Drug Monitoring (but excluding Australia, Canada, New Zealand, Switzerland and the International Conference on Harmonization countries, i.e. countries in Europe, Japan and the US) and to a few other identified contacts from non-member countries. The questionnaire was sent out between March and July 2008 and was designed to collect information on the structure, resources, functions and achievements of pharmacovigilance systems in low- and middle-income countries, with a focus on pharmacovigilance activities supported by national health authorities including public health programmes. All questionnaires that were returned by the end of July 2008 were used in the analysis.Results: Fifty-five completed questionnaires were received by July 2008, representing a response rate of 55.5%. Forty-five percent of the pharmacovigilance centres in the analysis were established during the 1990s and 49% were set up later; 69% were affiliated with their Drug Regulatory Agency, 20% with the Ministry of Health and 9% with a university or other scientific body. Few countries (23 of 55) have any budget allocated for pharmacovigilance. Public health programmes (44%), the Global Fund to fight AIDS, Tuberculosis and Malaria (36%), universities (26%), poison centres (21%), Management Sciences for Health (18%) and Rational Use of Drugs networks (15%) sponsor some pharmacovigilance activities. In addition to direct pharmacovigilance activities, many centres are also involved in other activities such as drug information (63%), promoting patient safety (52%), rational use of drugs (46%) and poison information (15%). Some countries have sentinel sites to monitor HIV/AIDS patients (in seven countries) and other special groups. Information gathered through pharmacovigilance activities is used to assist regulatory functions in most countries (n = 42), lack of training and funding were mentioned as being major challenges to pharmacovigilance in many countries.Conclusions: This study has helped identify some of the special challenges and barriers to promoting pharmacovigilance in low- and middle-income countries. A pharmacovigilance strategy in these settings needs to help build health systems that can serve the purpose of multiple health conditions. It needs to identify and implement feasible systems, governance, infrastructures, human resource, training and capacity building, sustainable methodologies and innovations in pharmacovigilance; a key component will be the dissemination of medicines safety information to policy makers and regulators and knowledge sharing with healthcare professionals through high quality informatics and learning tools, with rational use of medicines and patient safety as the ultimate goal of pharmacovigilance.

WHO Strategy for Collecting Safety Data in Public Health Programmes: Complementing Spontaneous Reporting Systems

Globally, national pharmacovigilance systems rely on spontaneous reporting in which suspected adverse drug reactions (ADRs) are reported to a national coordinating centre by health professionals, manufacturers or patients. Spontaneous reporting systems are the easiest to establish and the cheapest to run but suffer from poor-quality reports and underreporting. It is difficult to estimate rates and frequencies of ADRs through spontaneous reporting. Public health programmes need to quantify and characterize risks to individuals and communities from their medicines, to minimize harm and improve use, to sustain public confidence in the programmes, and to track problems due to medication errors and poor quality medicines. Additional methods are therefore needed to monitor the quantitative aspects of medicine safety, to better identify specific risk factors and high-risk groups, and to characterize ADRs associated with specific medicines and in specific populations. The present paper introduces two methods, cohort event monitoring and targeted spontaneous reporting, that are being implemented by the WHO, in its public health programmes, to complement spontaneous reporting. The advantages and disadvantages of these methods and how each can be applied in clinical practice are discussed.

Assessment of global reporting of adverse drug reactions for anti-malarials, including artemisinin-based combination therapy, to the WHO Programme for International Drug Monitoring

BackgroundIn spite of enhanced control efforts, malaria remains a major public health problem causing close to a million deaths annually. With support from several donors, large amounts of artemisinin-based combination therapy (ACT) are being deployed in endemic countries raising safety concerns as little is known about the use of ACT in several of the settings where they are deployed. This project was undertaken to profile the provenance of the pharmacovigilance reporting of all anti-malarials, including ACT to the WHO adverse drug reaction (ADR) database (Vigibase™) over the past 40 years.MethodsThe WHO Programme for International Drug Monitoring, the Uppsala Monitoring Centre (UMC) provided anonymized extracts of Vigibase™ covering the period 1968-2008. All countries in the programme were clustered according to their malaria control phase and income status. The number of individual case safety reports (ICSRs) of anti-malarials was analyzed according to those clusters.ResultsFrom 1968 to 2008, 21,312 ICSRs suspecting anti-malarials were received from 64 countries. Low-income countries, that are also malaria-endemic (categorized as priority 1 countries) submitted only 1.2% of the ICSRs. Only 60 out of 21,312 ICSRs were related to ACT, 51 of which were coming from four sub-Saharan African countries. Although very few ICSRs involved artemisinin-based compounds, many of the adverse events reported were potentially serious.ConclusionsThis paper illustrates the low reporting of ADRs to anti-malarials in general and ACT in particular. Most reports were submitted by non-endemic and/or high-income countries. Given the current mix of large donor funding, the insufficient information on safety of these drugs, increasing availability of ACT and artemisinin-based monotherapies in public and private sector channels, associated potential for inappropriate use and finally a pipeline of more than 10 new novel anti-malarials in various stages of development, the presence of well functioning national pharmacovigilance systems is vital to ensure safe and responsible scale up of ACT deployment. Bringing together the competencies of national pharmacovigilance centres and various types of organizations in the NGO, academic and private sectors with global coordination to create short- and long-term solutions may help address the lag between rapidly growing ACT use and poor ADR reporting.

Effective vaccine safety systems in all countries: A challenge for more equitable access to immunization

Serious vaccine-associated adverse events are rare. To further minimize their occurrence and to provide adequate care to those affected, careful monitoring of immunization programs and case management is required. Unfounded vaccine safety concerns have the potential of seriously derailing effective immunization activities. To address these issues, vaccine pharmacovigilance systems have been developed in many industrialized countries. As new vaccine products become available to prevent new diseases in various parts of the world, the demand for effective pharmacovigilance systems in low- and middle-income countries (LMIC) is increasing. To help establish such systems in all countries, WHO developed the Global Vaccine Safety Blueprint in 2011. This strategic plan is based on an in-depth analysis of the vaccine safety landscape that involved many stakeholders. This analysis reviewed existing systems and international vaccine safety activities and assessed the financial resources required to operate them. The Blueprint sets three main strategic goals to optimize the safety of vaccines through effective use of pharmacovigilance principles and methods: to ensure minimal vaccine safety capacity in all countries; to provide enhanced capacity for specific circumstances; and to establish a global support network to assist national authorities with capacity building and crisis management. In early 2012, the Global Vaccine Safety Initiative (GVSI) was launched to bring together and explore synergies among on-going vaccine safety activities. The Global Vaccine Action Plan has identified the Blueprint as its vaccine safety strategy. There is an enormous opportunity to raise awareness for vaccine safety in LMIC and to garner support from a large number of stakeholders for the GVSI between now and 2020. Synergies and resource mobilization opportunities presented by the Decade of Vaccines can enhance monitoring and response to vaccine safety issues, thereby leading to more equitable delivery of vaccines worldwide.

Specific features of medicines safety and pharmacovigilance in Africa

The thalidomide tragedy in the late 1950s and early 1960s served as a wakeup call and raised questions about the safety of medicinal products. The developed countries rose to the challenge putting in place systems to ensure the safety of medicines. However, this was not the case for low-resource settings because of prevailing factors inherent in them. This paper reviews some of these features and the current status of pharmacovigilance in Africa. The health systems in most of the 54 countries of Africa are essentially weak, lacking in basic infrastructure, personnel, equipment and facilities. The recent mass deployment of medicines to address diseases of public health significance in Africa poses additional challenges to the health system with notable safety concerns. Other safety issues of note include substandard and counterfeit medicines, medication errors and quality of medicinal products. The first national pharmacovigilance centres established in Africa with membership of the World Health Organization (WHO) international drug monitoring programme were in Morocco and South Africa in 1992. Of the 104 full member countries in the programme, there are now 24 African countries with a further nine countries as associate members. The pharmacovigilance systems operational in African countries are based essentially on spontaneous reporting facilitated by the introduction of the new tool Vigiflow. The individual case safety reports committed to the WHO global database (Vigibase) attest to the growth of pharmacovigilance in Africa with the number of reports rising from 2695 in 2000 to over 25,000 in 2010. There is need to engage the various identified challenges of the weak pharmacovigilance systems in the African setting and to focus efforts on how to provide resources, infrastructure and expertise. Raising the level of awareness among healthcare providers, developing training curricula for healthcare professionals, provisions for paediatric and geriatric pharmacovigilance, engaging the pharmaceutical industries as well as those for herbal remedies are of primary concern.

The Monitoring Medicines Project: A Multinational Pharmacovigilance and Public Health Project

The Monitoring Medicines project (MM), funded by the FP-7 EU framework, was carried out between 2009 and 2013 by a consortium of 11 partners. The objectives were to support and strengthen consumer reporting of adverse drug reactions (ADRs); expand the role and scope of national pharmacovigilance centres concerning medication errors; promote improved use of pharmacovigilance data; and develop methods to complement spontaneous reporting. The work was organised into four themes: patient reporting; medication errors; drug dependence, counterfeit and substandard medicines and clinical risk estimation; and active and targeted spontaneous pharmacovigilance. MM differed from some other major pharmacovigilance initiatives by having participants from developing countries in Asia and Africa and in leaning towards public health and communicable diseases. MM brought together stakeholders including WHO, drug regulators, pharmacovigilance centres, consumers, public health and disease specialists and patient safety networks. Resources and methodologies developed directly by, or with support from, MM include electronic systems/tools for consumer ADR reporting and cohort event monitoring; publication by WHO of handbooks on consumer reporting, medication errors and pharmacovigilance for TB medicines; methodologies for detecting drug dependence and substandard or counterfeit medicines in ADR databases; and a database on HIV treatment risks with a risk assessment tool. MM enabled stakeholders to achieve more than if they had worked alone in pursuit of patient safety.

Pharmacovigilance in resource-limited countries

In the past 20 years, many low- and middle-income countries have created national pharmacovigilance (PV) systems and joined the WHO’s global PV network. However, very few of them have fully functional systems. Scientific evidence on the local burden of medicine-related harm and their preventability is missing. Legislation and regulatory framework as well as financial support to build sustainable PV systems are needed. Public health programs need to integrate PV to monitor new vaccines and medicines introduced through these programs. Signal analysis should focus on high-burden preventable adverse drug problems. Increased involvement of healthcare professionals from public and private sectors, pharmaceutical companies, academic institutions and the public at large is necessary to assure a safe environment for drug therapy. WHO has a major role in supporting and coordinating these developments.

Botanical Nomenclature in Pharmacovigilance and a Recommendation for Standardisation

Nomenclature of plants in pharmacology can be presented by pharmaceutical names or scientific names in the form of Linnaean binomials. In this paper, positive and negative aspects of both systems are discussed in the context of the scientific nomenclatural framework and the systems’ practical applicability. The Uppsala Monitoring Centre (UMC) runs the WHO Programme for International Drug Monitoring and is responsible for the WHO Adverse Drug Reaction (ADR) database that currently contains 3.6 million records. In order for the UMC to monitor pharmacovigilance through ADRs to herbal medicine products the following nomenclatural criteria are important: (i) the name should indicate only one species of plant; (ii) the source for this name must be authoritative; (iii) the name should indicate which part of the plant is used. Based on these criteria, the UMC investigated four options: (i) adopt main names used in recognised (inter-) national pharmacopoeias or authoritative publications; (ii) adopt option 1, but cite the publication for all names in abbreviated form; (iii) three-part pharmaceutical names consisting of Latinised part name plus Latinised genus name, plus Latinised specific epithet; (iv) scientific binomial names, optionally with author and plant part used. The UMC has chosen the latter option and will at its adoption utilise the scientific botanical nomenclature as defined by the International Code of Botanical Nomenclature. This decision satisfies all criteria set by the UMC and renders the necessity of creating a new system or upgrading an old inconsistent system obsolete. The UMC has also issued an extensive synonymy checklist of vernacular, pharmaceutical and scientific names for the herbals in the WHO ADR database. We strongly recommend the adoption of scientific names to denote plant ingredients in medicine.