Causal inference concepts applied to three observational studies in the context of vaccine development: from theory to practice.

BACKGROUND: Randomized controlled trials are considered the gold standard to evaluate causal associations, whereas assessing causality in observational studies is challenging. METHODS: We applied Hill's Criteria, counterfactual reasoning, and causal diagrams to evaluate a potentially causal relationship between an exposure and outcome in three published observational studies: a) one burden of disease cohort study to determine the association between type 2 diabetes and herpes zoster, b) one post-authorization safety cohort study to assess the effect of AS04-HPV-16/18 vaccine on the risk of autoimmune diseases, and c) one matched case-control study to evaluate the effectiveness of a rotavirus vaccine in preventing hospitalization for rotavirus gastroenteritis. RESULTS: Among the 9 Hill's criteria, 8 (Strength, Consistency, Specificity, Temporality, Plausibility, Coherence, Analogy, Experiment) were considered as met for study c, 3 (Temporality, Plausibility, Coherence) for study a, and 2 (Temporary, Plausibility) for study b. For counterfactual reasoning criteria, exchangeability, the most critical assumption, could not be tested. Using these tools, we concluded that causality was very unlikely in study b, unlikely in study a, and very likely in study c. Directed acyclic graphs provided complementary visual structures that identified confounding bias and helped determine the most accurate design and analysis to assess causality. CONCLUSIONS: Based on our assessment we found causal Hill's criteria and counterfactual thinking valuable in determining some level of certainty about causality in observational studies. Application of causal inference frameworks should be considered in designing and interpreting observational studies.

Immunogenicity and safety of the RTS,S/AS01 malaria vaccine co-administered with measles, rubella and yellow fever vaccines in Ghanaian children: A phase IIIb, multi-center, non-inferiority, randomized, open, controlled trial.

BACKGROUND: To optimize vaccine implementation visits for young children, it could be efficient to administer the first RTS,S/AS01 malaria vaccine dose during the Expanded Programme on Immunization (EPI) visit at 6 months of age together with Vitamin A supplementation and the third RTS,S/AS01 dose on the same day as yellow fever (YF), measles and rubella vaccines at 9 months of age. We evaluated the safety and immunogenicity of RTS,S/AS01 when co-administered with YF and combined measles-rubella (MR) vaccines. METHODS: In this phase 3b, open-label, controlled study (NCT02699099), 709 Ghanaian children were randomized (1:1:1) to receive RTS,S/AS01 at 6, 7.5 and 9 months of age, and YF and MR vaccines at 9 or 10.5 months of age (RTS,S coad and RTS,S alone groups, respectively). The third group received YF and MR vaccines at 9 months of age and will receive RTS,S/AS01 at 10.5, 11.5 and 12.5 months of age (Control group). All children received Vitamin A at 6 months of age. Non-inferiority of immune responses to the vaccine antigens was evaluated 1 month following co-administration versus RTS,S/AS01 or EPI vaccines (YF and MR vaccines) alone using pre-defined non-inferiority criteria. Safety was assessed until Study month 4.5. RESULTS: Non-inferiority of antibody responses to the anti-circumsporozoite and anti-hepatitis B virus surface antigens when RTS,S/AS01 was co-administered with YF and MR vaccines versus RTS,S/AS01 alone was demonstrated. Non-inferiority of antibody responses to the measles, rubella, and YF antigens when RTS,S/AS01 was co-administered with YF and MR vaccines versus YF and MR vaccines alone was demonstrated. The safety profile of all vaccines was clinically acceptable in all groups. CONCLUSIONS: RTS,S/AS01 can be co-administered with Vitamin A at 6 months and with YF and MR vaccines at 9 months of age during EPI visits, without immune response impairment to any vaccine antigen or negative safety effect.

Long-term incidence of severe malaria following RTS,S/AS01 vaccination in children and infants in Africa: an open-label 3-year extension study of a phase 3 randomised controlled trial.

BACKGROUND: Results from a previous phase 3 study showed efficacy of the RTS,S/AS01 vaccine against severe and clinical malaria in children (in 11 sites in Africa) during a 3-4-year follow-up. We aimed to investigate malaria incidence up to 7 years postvaccination in three of the sites of the initial study. METHODS: In the initial phase 3 study, infants aged 6-12 weeks and children aged 5-17 months were randomly assigned (1:1:1) to receive four RTS,S/AS01 doses (four-dose group), three RTS,S/AS01 doses and a comparator dose (three-dose group), or four comparator doses (control group). In this open-label extension study in Korogwe (Tanzania), Kombewa (Kenya), and Nanoro (Burkina Faso), we assessed severe malaria incidences as the primary outcome for 3 additional years (January, 2014, to December, 2016), up to 6 years (younger children) or 7 years (older children) postprimary vaccination in the modified intention-to-treat population (ie, participants who received at least one dose of the study vaccine). As secondary outcomes, we evaluated clinical malaria incidences and serious adverse events. This trial is registered with ClinicalTrials.gov, number NCT02207816. FINDINGS: We enrolled 1739 older children (aged 5-7 years) and 1345 younger children (aged 3-5 years). During the 3-year extension, 66 severe malaria cases were reported, resulting in severe malaria incidence of 0·004 cases per person-years at risk (PPY; 95% CI 0-0·033) in the four-dose group, 0·007 PPY (0·001-0·052) in the three-dose group, and 0·009 PPY (0·001-0·066) in the control group in the older children category and a vaccine efficacy against severe malaria that did not contribute significantly to the overall efficacy (four-dose group 53·7% [95% CI -13·7 to 81·1], p=0·093; three-dose group 23·3% [-67·1 to 64·8], p=0·50). In younger children, severe malaria incidences were 0·007 PPY (0·001-0·058) in the four-dose group, 0·007 PPY (0·001-0·054) in the three-dose group, and 0·011 PPY (0·001-0·083) in the control group. Vaccine efficacy against severe malaria also did not contribute significantly to the overall efficacy (four-dose group 32·1% [-53·1 to 69·9], p=0·35; three-dose group 37·6% [-44·4 to 73·0], p=0·27). Malaria transmission was still occurring as evidenced by an incidence of clinical malaria ranging from 0·165 PPY to 3·124 PPY across all study groups and sites. In older children, clinical malaria incidence was 1·079 PPY (95% CI 0·152-7·662) in the four-dose group, 1·108 PPY (0·156-7·868) in the three-dose group, and 1·016 PPY (0·14-7·213) in the control group. In younger children, malaria incidence was 1·632 PPY (0·23-11·59), 1·563 PPY (0·22-11·104), and 1·686 PPY (0·237-11·974), respectively. In the older age category in Nanoro, clinical malaria incidence was higher in the four-dose (2·444 PPY; p=0·011) and three-dose (2·411 PPY; p=0·034) groups compared with the control group (1·998 PPY). Three cerebral malaria episodes and five meningitis cases, but no vaccine-related severe adverse events, were reported. INTERPRETATION: Overall, severe malaria incidence was low in all groups, with no evidence of rebound in RTS,S/AS01 recipients, despite an increased incidence of clinical malaria in older children who received RTS,S/AS01 compared with the comparator group in Nanoro. No safety signal was identified. FUNDING: GlaxoSmithKline Biologicals SA.

Effectiveness of the Oral Human Attenuated Rotavirus Vaccine: A Systematic Review and Meta-analysis-2006-2016.

BACKGROUND: Gastroenteritis caused by rotavirus accounts for considerable morbidity in young children. We aimed to assess the vaccine effectiveness (VE) of the oral rotavirus vaccine Rotarix, as measured by laboratory-confirmed rotavirus infection after referral to hospital and/or emergency departments in children aged <5 years with gastroenteritis. METHODS: We performed a systematic search for peer-reviewed studies conducted in real-life settings published between 2006 and 2016 and a meta-analysis to calculate the overall Rotarix VE, which was further discriminated through stratified analyses. RESULTS: The overall VE estimate was 69% (95% confidence interval [CI], 62% to 75%); stratified analyses revealed a non-negligible impact of factors such as study design and socioeconomic status. Depending on the control group, VE ranged from 63% (95% CI, 52% to 72%) to 81% (95% CI, 69% to 88%) for unmatched and matched rotavirus test-negative controls. VE varied with socioeconomic status: 81% (95% CI, 74% to 86%) in high-income countries, 54% (95% CI, 39% to 65%) in upper-middle-income countries, and 63% (95% CI, 50% to 72%) in lower-middle-income countries. Age, rotavirus strain, and disease severity were also shown to impact VE, but to a lesser extent. CONCLUSIONS: This meta-analysis of real-world studies showed that Rotarix is effective in helping to prevent hospitalizations and/or emergency department visits due to rotavirus infection.

Impact of history of febrile convulsions on the risk difference of febrile convulsions with the tetravalent measles-mumps-rubella-varicella vaccine: Post-hoc exploratory analysis of results from a matched-cohort study.

This post-hoc analysis of data from a matched cohort study investigated the risk of febrile convulsions (FC) 5-12 days post-first dose of measles-mumps-rubella-varicella vaccine (MMRV) in a low-risk population, compared to measles-mumps-rubella (MMR) and varicella (V) vaccines administered separately. The low-risk population excluded children with personal history of FC (Scenario 1) and children with personal or/and family history (≥1 parent/sibling) of FC (Scenario 2). Incidence of FC post-MMRV in Scenario 2 (excluding at risk children) (36.3-49.5/100,000) and post-MMR+V in the whole cohort including children with personal/family history of FC (43.6/100,000) were similar. The risk difference of FC increased by 0.2 case/100,000 in Scenario 1 and decreased by 5.3-8.6 cases/100,000 of vaccinated children in Scenario 2, compared to the whole cohort. The overall risk of FC post-first dose MMRV vaccination could be lowered by administering MMRV only to children with no personal or family history of FC.

Common Evaluations of Disease Activity in Rheumatoid Arthritis Reach Discordant Classifications across Different Populations.

OBJECTIVES: The classification of disease activity states in rheumatoid arthritis (RA) can be achieved through disease activity indices, such as the Disease Activity Score in 28 joints erythrocyte sedimentation rate (DAS28-ESR), the Simplified Disease Activity Index (SDAI), and the Clinical Disease Activity Index (CDAI). Subjective measurements, such as patient reported outcomes have been incorporated into several of these indices alongside more objective assessments, such as increases in the ESR and C-reactive protein. Moreover, while they use similar criteria, different indices weight these criteria to different extents. Therefore, the classifications based on each evaluation may not always be the same. We aim to compare the performance of the three indices and their individual components in two different populations. METHODS: Data from Dutch and Portuguese adherent centers were extracted from the METEOR database, a multinational collaboration on RA. We included a total of 24,605 visits from Dutch centers (from 5,870 patients) and 20,120 visits from Portuguese centers (from 3,185 patients). We compared the disease activity states as evaluated by the DAS28-ESR, CDAI, and SDAI across the two populations. In addition, we analyzed the individual components of each evaluation, including their respective contributions to the outcome, in each population. RESULTS: We found significant differences in the disease activity states classified with the DAS28-ESR between the two populations. SDAI and CDAI had more congruous results. While the proportion of visits to Dutch and Portuguese centers that were classified as "in remission" was very similar between the CDAI and SDAI, the DAS28-ESR gave discordant results. Dutch patients had lower ESRs, which is more heavily weighted in the DAS28-ESR. In addition, even though the mean physicians' global assessment values did not vary significantly for Dutch vs Portuguese physicians, we found that doctors at Portuguese centers overall scored the physician's global assessment lower than Dutch physicians for patient visits classified by disease activity state. CONCLUSION: While the CDAI and SDAI assigned disease activity states that were largely similar, the DAS28-ESR was often discordant across the two populations. Moreover, we found that physicians, more than patients, evaluated disease activity differently among the Portuguese and Dutch populations.

When rheumatologists report that they agree with a guideline, does this mean that they practise the guideline in clinical practice? Results of the International Recommendation Implementation Study (IRIS).

INTRODUCTION: The European League Against Rheumatism (EULAR) recommendations for the management of rheumatoid arthritis (RA) and the treat-to-target (T2T) principles have been developed in order to improve the treatment outcome of patients with RA, and have received broad attention. It is not clear, though, whether these recommendations are indeed followed up in clinical practice. OBJECTIVE: To investigate if rheumatologists that report to agree with existing guidelines indeed follow them up in clinical practice. METHODS: The International Recommendation Implementation Study (IRIS) included 132 participating rheumatologists from 14 countries. Participating rheumatologists received a questionnaire measuring their awareness/commitment with the EULAR/T2T recommendations and followed a dedicated educational programme. Subsequently, they were asked to enrol 5-10 patients with new-onset RA in the online IRIS database and monitor disease activity and treatment for a period of 1-2 years. Four recommendations (3 from the EULAR recommendations and one from the T2T recommendations) were selected on the basis of testability, and analysed with regard to compliance by participating rheumatologists. RESULTS: In total, 72 of the 132 participating rheumatologists contributed 378 patients to the database. Of these participants, 70 (98%) agreed upfront with the recommendation that disease-modifying antirheumatic drug (DMARD) therapy should be started as soon as possible after diagnosis in every patient; 69 (96%) of the rheumatologists agreed with the recommendation that methotrexate (MTX) should be part of the first treatment strategy. When measuring the actual performance, it was found that the recommendation on early DMARD start was met in 253 (67%) of the recorded patients, and the recommendation on MTX in 225 (60%) of the recorded patients. Of the participants, 60 (83%) agreed that composite measures should be recorded regularly, but only in 134(54%) of the patients were composite scores actually recorded in ≥50% of patient visits. CONCLUSION: Reporting to be compliant with EULAR recommendations and T2T principles, even after dedicated education does not mean they actually comply with it in clinical practice.

DAS steered therapy in clinical practice; cross-sectional results from the METEOR database.

BACKGROUND: Little is known on how well targeted treatment, for instance targeting towards low DAS, is implemented in clinical practice. Our aim was to evaluate treatment adjustments in response to DAS in RA patients in clinical practice. METHODS: We used data from one referral centre, multiple rheumatologists, from the METEOR database. Generalized Estimating Equations (GEE) were used to assess whether in case of non-low disease activity (DAS > 2.4) treatment intensifications in DMARD therapy occurred ((change or increase in dose or number of DMARDs, including synthetic (s)DMARDs, biologic (b)DMARDs and corticosteroids compared to the visit before)). Determinants of not intensifying the treatment when DAS > 2.4 were investigated using GEE. RESULTS: Five thousand one hundred fifty-seven registered visits of 1202 patients were available for the analyses. A DAS > 2.4 was weakly (OR: 1.19; 95% CI 1.07-1.33) associated with a treatment intensification. In 69% (n = 3577) of the visits patients were in low disease activity. In 66% (n = 1028) of the visits with DAS > 2.4 treatment was not intensified. These patients had a higher tender joint count and received more often methotrexate plus a bDMARD, or csDMARD monotherapy, as compared to patients that received treatment intensification. CONCLUSION: In the majority of visits in the METEOR database patients were already in a state of low disease activity, reflecting appropriate treatment intensity. When DAS was greater than 2.4, treatment was often not intensified due to high tender joint count or specific treatment combinations. This data suggest that while aiming for low DAS, physicians per patient weigh whether all DAS elements indicate disease activity or will respond to DMARD adjustment or not, and make treatment decisions accordingly.

Assessment of Global Disease Activity in Rheumatoid Arthritis by Patients and Physicians: Differences Across Countries in the METEOR Database.

AIM: The aim of the study was to compare the differences between patient global disease activity (PtGDA) and physician global disease activity (PhGDA) score within and across 13 countries in the METEOR (Measurement of Efficacy of Treatment in the "Era of Outcome" in Rheumatology) database. METHODS: Data from METEOR were used to compare PtGDA and PhGDA, scored independently on a 100-mm visual analog scale from 0 (best possible) until 100 (worst possible), in 23,117 visits in 5709 anonymized patients during the period between 2008 and 2012. Linear mixed models were used to model mean differences between PtGDA and PhGDA in 13 countries (Brazil, Czech Republic, France, Ireland, Italy, Latvia, Mexico, the Netherlands, Pakistan, Portugal, Spain, United Kingdom, and the United States), adjusted for differences in Disease Activity Score in 28 joints (DAS28). Generalized estimating equations were used to model differences (>20 mm) between PtGDA and PhGDA score as the outcome and countries as determinants, adjusted for DAS28. RESULTS: Mean difference between PtGDA and PhGDA scores varied by country, from -2 mm (physician scores higher) in Mexico to +14 mm (patient scores higher) in Brazil. "Country" was a significant determinant of the difference between PtGDA and PhGDA scores, independent of differences in DAS28. With the Netherlands as reference, PtGDA and PhGDA scores for individual patients differ significantly in almost all (n = 10) countries, with the exception of France and Spain. CONCLUSIONS: Differences between patients' and physicians' assessment of GDA vary across the countries. Influence of country must be taken into account when interpreting discordances between the patient's and the physician's assessment of GDA in rheumatoid arthritis.

Intra articular injection with corticosteroids in patients with recent onset rheumatoid arthritis: subanalyses from the BeSt study.

To investigate the association between intra-articular (IA) large joint corticosteroid injections and clinical outcomes in patients with recent onset rheumatoid arthritis (RA). We compared pain (visual analog scale (VAS)), the Disease Activity Score (44 joints) (DAS), and swollen and tender joint counts before and after IA injection. Using linear mixed models (LMM), the DAS and the Health Assessment Questionnaire (HAQ) score over time were compared in IA injected versus noninjected patients. In year 1, 93 joints were injected in 44 patients treated with initial methotrexate monotherapy and 16 in patients treated with initial combination therapy (p < 0.01). Three months later, swelling and tenderness were resolved in 58-50 % of the injected joints; but within 12 months after the injection, swelling recurred in 14 % and tenderness in 41 % of the injected joints. Mean (SD) DAS decreased from 4.0 (1.4) before to 3.2 (1.2) 3 months after injection (p < 0.01) and VAS for pain from 49 (26) to 40 (27) (p < 0.01). LMM showed a higher DAS and HAQ in patients injected in year 0-1 compared to those not injected, but no difference in subsequent years, and similar treatment adjustments. Eight-year radiographs showed similar damage in injected joints (17 %) and noninjected joints (14 %). IA corticosteroid injections are associated with symptom relief, sometimes only temporarily, in 50 % of cases. Initially DAS significantly improved, but over time DAS and HAQ were similar in injected versus non-injected patients. After 8 years there was no difference in joint damage.

Assessment of global disease activity in RA patients monitored in the METEOR database: the patient's versus the rheumatologist's opinion.

The objectives of this study were to compare the patient's (PtGDA) and physician's (PhGDA) assessment of global disease activity and to identify factors that might influence these differences as well as factors that may influence the patient's and the physician's scores separately. Anonymous data were used from 2,117 Dutch patients included in the Measurement of efficacy of Treatment in the Era of Rheumatology database. PtGDA and PhGDA were scored independently on a 100-mm visual analog scale (VAS) with 0 and 100 as extremes. The agreement, intraclass correlation coefficients (ICC), was calculated and a Bland-Altman plot was created to visualize the differences between PtGDA and PhGDA. Linear mixed model analysis was used to model PtGDA and PhGDA. Logistic repeated measurements were used to model the difference in PtGDA and PhGDA (PtGDA > PhGDA versus PtGDA ≤ PhGDA). Gender patient, gender physician, age, swollen joint count (SJC), tender joint count, VAS pain, disease duration, and erythrocyte sedimentation rate (ESR) were considered as possible determinants in both models. Mean (standard deviation) age was 57 (15) years and 67 % of the patients were female. Agreement between PtGDA and PhGDA was moderate (ICC, 0.57). Patients scored on average 11 units higher (worse) than rheumatologists (95 % limits of agreement, -25.2 to 47.6). Patient's perception of pain (VAS) was positively associated with a PtGDA being higher than PhGDA. Similarly, ESR and swollen joint counts were positively associated with a PtGDA being lower or equal to the PhGDA. Patients rate global disease activity consistently higher than their rheumatologists. Patients base their judgment primarily on the level of pain, physicians on the level of SJC and ESR.