Advertisement

Advances in Therapy

, Volume 35, Issue 8, pp 1199–1214 | Cite as

Enhanced Safety Surveillance of Seasonal Quadrivalent Influenza Vaccines in English Primary Care: Interim Analysis

  • Simon de Lusignan
  • Gaël Dos Santos
  • Rachel Byford
  • Anne Schuind
  • Silvia Damaso
  • Vishvesh Shende
  • Chris McGee
  • Ivelina Yonova
  • Filipa Ferreira
Open Access
Original Research

Abstract

Introduction

The European Medicines Agency (EMA) requires vaccine manufacturers to conduct enhanced safety surveillance (ESS) of seasonal influenza vaccines including a near real-time evaluation of collected data. The objective was to identify whether the use of passive surveillance or active surveillance provides different results of reported adverse events of interest (AEIs) by specified age strata and AEI type. We report the weekly incidence rates of AEIs within 7 days following seasonal influenza vaccination using passive and active surveillance.

Methods

AEIs were collected within 7 days of vaccination from ten general practices predominantly administering inactivated quadrivalent influenza vaccine (IIV4, Fluarix Tetra, GSK). Vaccinees completed an adverse drug reaction (ADR) card. ADR card and medically attended AEIs data were recorded in practice electronic health records. We report the outcome of the first 5 weeks of safety surveillance (September 12, 2016–October 16, 2016); in an exploratory analysis, rates of AEI for IIV4 are compared to those passively reported through a sentinel network.

Results

Practices vaccinated 13.1% (12,864/98,091) of their registered population; 5.6% (95% CI 5.20–6.00) of them reported AEIs, none serious. The most frequent were respiratory 2.60% (95% CI 2.33–2.88), musculoskeletal 1.82% (95% CI 1.59–2.05) and neurological 1.05% (95% CI 0.88–1.23). AEIs were more frequently reported for adults than for children; 5.91% (95% CI 5.49–6.34) compared to 1.49% (95% CI 0.69–2.29); 47.18% of the adults reported AEI using the ADR card, none were returned for subjects < 18 years old. The frequency of AEIs reporting was higher, 6.88% (95% CI 6.35–7.42) vs. 3.30% (95% CI 2.68–3.96, 100/3028, p < 0.000), through ESS than passive surveillance.

Conclusion

The ESS did not reveal any safety signal and we demonstrated the feasibility of conducting ESS following EMA recommendations. The use of a customised ADR card led to a doubling of AEIs reports over passive surveillance in adults.

Funding

GlaxoSmithKline Biologicals SA, Wavre, Belgium.

Keywords

Adverse events Adverse reaction reporting systems Evaluation studies as a topic General practice Immunology Infectious diseases Influenza Medical record systems, computerized Vaccination 

Introduction

Background

European Medicine Agency (EMA) Guidance on Influenza Vaccines

The EMA set out new standards for Marketing Authorisation Holders who produce influenza vaccines for use within the EU. An interim guideline was released in August 2014 and further updated in July 2016 and came into effect in February 2017 [1, 2].

These news standards set out a requirement for safety and reactogenicity monitoring of seasonal influenza vaccines with a new strain composition. These vaccines are to be evaluated for local (at the injection site) and systemic (e.g. fever, myalgia) adverse events of interest (AEI) in specified age groups, with timely relevant findings to be provided to the competent regulatory authorities.

The previous requirement to conduct small-scale safety and immunogenicity clinical trials was withdrawn in 2015 [3]. The rationale for this change was that these trials had insufficient power to adequately evaluate safety of annual formulation changes because of the relatively low frequency of some adverse events [4, 5, 6, 7]. The trials were to be replaced by enhanced safety surveillance and post-marketing monitoring program, and vaccine effectiveness assessments.

The EMA Guidance Set Out Three Options Envisioned for Enhanced Surveillance

  • Enhanced active surveillance [post authorisation safety studies (PASS)]: The PASS should follow up defined cohorts of adults and children using web-based reporting or diary cards for 7 days [or 14 for live attenuated influenza vaccine (LAIV)] after immunisation. The PASS should include specific endpoints of AEIs.

  • Enhanced passive surveillance: The principle is to rapidly monitor vaccine usage and to facilitate rapid adverse drug reaction (ADR) reporting.

  • Data mining or other use of electronic health record (EHR).

English Primary Care as a Setting for Enhanced Safety Surveillance

Primary care in England has a registration-based list system that has been highly computerised since 2004 and data extracted from this system are widely used in research [8, 9]. Key data are recorded in EHRs using the Read coding system, a detailed classification system able to record the EMA-specified AEIs [10]. General practices are largely independent professional partnerships and make their own decision about which brand of influenza vaccines to purchase prior to the start of the influenza season; they generally follow the Chief Medical Officer’s recommendation (Fig. 1) by giving inactivated influenza vaccine (IIV) to most adults and LAIV [11] to children.
Fig. 1

UK Chief Medical Officer’s recommendations for influenza vaccination [36]

In the 2015–2016 influenza season, we demonstrated the feasibility of conducting enhanced passive surveillance to address EMA guidelines on seasonal influenza vaccines using routinely collected primary care data [12]. The methods used for data extraction and processing were those developed by the University of Surrey to support the Royal College of General Practitioners (RCGP) Research and Surveillance Centre (RSC) [13].

In the 2016–2017 season, passive surveillance was enhanced by the use of a customised ADR card distributed to patients vaccinated with GlaxoSmithKline’s (GSK) Fluarix Tetra in the ten GP practices. We designed an “orange card” which included a structured list of EMA-specified AEIs, which we grouped into body systems, as well as the option for study participants to report that no AEI or any other adverse events (AEs) occurred post vaccination (see Sect. “The Orange Card”). Our method builds on the study we conducted in 2015–2016 [14], whilst improving the usability of the orange card.

Challenges for Pharmacovigilance

Seasonal influenza vaccination presents challenges for pharmacovigilance. These include yearly immunisation in large population cohorts in a relatively short time period, and multiplicity of vaccine products on the market with the need to conduct product-specific safety surveillance and report findings in a near real-time manner. The 2016/2017 influenza plan recommended the vaccination of high-risk groups and children, who are likely to be the vectors of influenza virus transmission (Fig. 1) [15]. Healthcare practitioners are also able to apply medical judgement and recommend vaccination as prevention of exacerbation of underlying diseases. General practices in England usually start administering influenza vaccines in September each year [16]; in parallel a smaller number of vaccinations are administered at school or community pharmacies [17].

In addition to the large number of people being vaccinated and therefore at risk of minor AEIs, influenza vaccinations have previously been associated with developing Guillain–Barré syndrome [18] and narcolepsy [19]. Therefore, with the large amount of people being vaccinated in such a short period of time, and the potential for developing severe AEIs, fast and reliable reporting of AEIs is important.

This study was conducted to address the EMA requirements. Here we present interim results of enhanced passive surveillance and enhanced active surveillance (passive surveillance plus the orange card), between September 12, 2016 and October 16, 2016 (International Standards Organisation (ISO) weeks 37–41).

Objective

The objective was to identify whether the use of passive surveillance or active surveillance provides different results of reported AEI by specified age strata and AEI type. We report the weekly incidence rates of AEIs within 7 days following seasonal influenza vaccination using passive and active surveillance.

Methods

Study Design

We defined active surveillance practices as the ten practices who predominantly used GSK influenza vaccinations during the interim period of September to October 2016. Patients from these practices were invited to record adverse events within 7 days post-vaccination and were asked to return the card to their registered practice no later than 14 days post-vaccination. Any AEI derived from the orange cards was entered into the practice EHR, whereas cards returned showing no report of ADR were retained by the practice and counted.

We operationalised enhanced passive surveillance in two ways. Firstly, through the ten practices who used the orange cards, but by looking at AEIs reported and extracted from the practices’ EHRs rather than the orange cards. Secondly, by comparing adverse events reported by all other practices from a sentinel network (Royal College of General Practitioners Research and Surveillance Centre; RCGP RSC) in England, who used GSK vaccinations.

This study builds on the key learnings and gaps of an initial pilot study in England in 2015 and continuous feedback received from the Pharmacovigilance Risk Assessment Committee (PRAC) of the EMA and the Vaccine Working Party (safety task force). We did not register this study as a clinical trial as the intervention was not used to treat patients, but rather to see if it increased surveillance of AEIs.

Setting

Ten volunteer general practices who stated that GSK’s Fluarix Tetra would be the main brand of influenza vaccine for the 2016/2017 season participated in the study. We enrolled practices spread across England: three practices each in North, Midlands and East, and South NHS Regions, with the tenth practice in London. The practices used different EHR systems. Practices’ reimbursement followed National Institute of Health Research (NIHR) guidelines for industry-sponsored studies.

Participants

Inclusion Criteria

All patients receiving a seasonal influenza vaccine between September 1, 2016 and November 30, 2016 in participating practices were eligible for inclusion in the study. The period for this interim report is September 12, 2016 to October 16, 2016, International Standards Organisation (ISO) weeks 37–41.

Exclusion Criteria

Patients who had explicitly opted out of data sharing were excluded. According to a National Health Service (NHS) communication, it is estimated that around 1.25% of the population has opted out (patients registered an objection to their identifiable information) [20]. Patients without a valid NHS number (which we pseudonymise to link data) or gender recorded in the EHR were excluded.

Variables

We defined exposure to seasonal influenza vaccine as a coded EHR record of either a vaccination administration code, a vaccine prescription or both.

We created lists of Read codes [21] for the AEIs specified by EMA, using an ontological approach grouping conditions by categories as follows [22, 23].
  • Respiratory: conjunctivitis, coryza, cough, epistaxis, hoarseness, nasal congestion, oropharyngeal pain, rhinorrhoea, wheezing

  • Musculoskeletal: arthropathy, muscle aches/myalgia

  • Neurological: Bell’s palsy, Guillain–Barré syndrome (GBS), headache, peripheral tremor, seizure/febrile convulsions

  • General symptoms: drowsiness, fatigue, irritability, malaise

  • Gastrointestinal: decreased appetite, diarrhoea, nausea, vomiting

  • Local symptoms: local erythema

  • Fever/pyrexia: fever, including actual temperatures recorded, when available

  • Rash: generalised rash, or other distribution

  • Sensitivity/anaphylaxis: anaphylactic reactions, facial oedema, hypersensitivity reactions

All the clinical codes used: Read codes version 2 and Read Clinical Terms version 3 for the AEI’s are presented in the Electronic Supplementary Material 1.

The Orange Card

The orange card was a strategy implemented to improve reporting of AEIs after vaccinations. One of the issues with passive surveillance is that not everybody will report to practice with an AEI. The orange card aims to overcome this issue by asking patients to report if they have had an AEI (no matter how minor) and return the card back to the practice. This is a clear advantage of using orange cards over passive surveillance. Patients are asked to provide demographic details, vaccine details, and whether or not they had any AEIs. If participants had any AEIs they were required to fill out tick boxes indicating the symptom (written in plain English), the severity, and whether the symptom was still present.

Compliance with Ethics Guidelines

Ethics committee approval was obtained in England from the Health Research Authority (IRAS project ID 211560). GlaxoSmithKline Biologicals SA, Wavre, Belgium funded the project and was involved in study design and interpretation. This study received a favourable ethics committee approval: Integrated Research Application System (IRAS) project ID 211560; Research Ethics Committee (REC) reference 16/NE/0271. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Informed consent was obtained from all individual participants included in the study.

Data Sources/Measurement

We extracted the data on a weekly basis using an automated extraction protocol and produced weekly reports of the incidence of AEIs by the variable grouping shown above; an issue of the weekly report is presented in the Electronic Supplementary Material 2.

Bias

It is likely that conditions that do not require medical attendance may have been underreported, such as local reactions or minor systemic symptoms. We tried to overcome this limitation by asking vaccinated patients to return the cards even if they did not experience any AE, reporting that no AEI occurred. However, we made no provision for recording returned negative cards.

Sample Size

We recruited ten volunteer practices whose combined list size exceeded the required sample size intended in our protocol. Using conservative assumptions, we estimated that approximately 70,000 registered patients would be required to enrol sufficient number of vaccinees (n = 5000). As mean practice size in England and Wales was 7034, we recruited ten practices. We assumed that vaccination uptake in the 2016/2017 season would be similar to the previous one where 71.0% of subjects over 65 years, 45.1% of those in a clinical risk group aged 6 months to 65 years and 42.3% of pregnant women were vaccinated [24]. With comparable vaccination uptake, at the time of the protocol development, we expected to enrol at least 5000 vaccinees, allowing a study sufficiently powered to detect most of common AEs regardless of the expected probability of occurrence. The scenarios and assumptions are described in depth previously as part of the learning from our first study [16]. Participating practices had larger list sizes than the national average resulting in a total of 98,091 registered patients at the start of the study (ISO week 37).

Quantitative variables

We included one quantitative variable, temperature recording. This variable was cleaned, as per our standard practice [25], correcting likely data entry errors, such as missing decimal points or recording in Fahrenheit.

Comparator Data

In the absence of well-defined comparator we compared the data extracted in this study with a passive surveillance system, i.e. without the enhancement of ADR cards [17]. We used a data extract for the same dates from the RCGP RSC, extracting data for people only given GSK Fluarix Tetra™ brand vaccine. This extract reported rates of AEI in people who had received any brand flu vaccine. The RCGP RSC has over 50 years involvement in influenza monitoring and vaccine effectiveness studies [26]. Practices have a dashboard [27] and receive feedback to improve data quality [28].

Statistical Methods

The rates of AEIs were computed overall, for each AEI and by category of AEI. We used the number of registered patients vaccinated with a GSK vaccination in weeks 37–41 as the denominator, and AEIs reported within 7 days after vaccination reported in EHR or through the orange card as the numerator; patients may have reported more than one AEI. We report the rates and percentages of patients vaccinated in weeks 37–41 with the GSK vaccination, and with any non-GSK vaccination for comparison (Table 1). We then report AEIs from the ten orange card practices for reported AEIs using the orange cards, and AEIs using any method. We carry out a chi-square test to see if there are statistically significant age differences in orange card reporting between groups or practices (Tables 2, 3). In the weekly report, the denominators are patients vaccinated in the reported week and the preceding week as reporting an AEI can straddle 2 weeks. We report 95% confidence intervals (95% CI) on AEI rates, using the critical binomial function in Microsoft Excel [29] (Tables 4, 6). We then compared the proportion of people administered GSK’s Fluarix Tetra (only) who reported an AEI in the study with those reported for the same brand of vaccine across the RCGP RSC using 95% CI and Pearson chi-square test. We report crude AEI rates but also compare age-standardised rates using the 2015 census population [30] (Table 7).
Table 1

Vaccination by age groups for GSK and non-GSK vaccinations (weeks 33–41, n = 12,864)

Age band

GSK vaccinations

Non-GSK vaccinations

Number of vaccinated patients (weeks 33–41)

% of GSK vaccinated patients

% of all patients vaccinated with GSK

Number of vaccinated patients (weeks 33–41)

% of non-GSK vaccinated patients

% of all patients vaccinated with non-GSK

6 months–5 years

6

0.07

0.05

559

13.16

4.35

6–12 years

12

0.14

0.09

226

5.32

1.76

13–17 years

32

0.37

0.25

40

0.94

0.31

18–65 years

2690

31.22

20.91

1308

30.78

10.17

Over 65 years

5875

68.20

45.67

2116

49.80

16.44

Total

8615

100

66.97

4249

100

33.03

GSK GlaxoSmithKline plc

Table 2

Summary of AEIs reported using the ADR orange card or any method for GSK vaccines, by age band (weeks 37–41)

Age band

Patients with an AEI reported by any method within 7 days of vaccination (weeks 37–41)

Patients with an AEI reported by orange card within 7 days of vaccination (weeks 37–41)

% of patients reporting AEIs using an orange card (95% CI)

6 months–5 years

9

0

0.00

6–12 years

4

0

0.00

13–17 years

0

0

0.00

Children total

13

0

0.00

18–65 years

256

118

46.09 (39.84–52.34)

> 65 years

452

216

47.79 (43.14–52.43)

Adults total

708

334

47.18

All ages (over 6 months)

721

334

46.32

AEIs adverse events of interest

Table 3

Summary of AEIs reported using the ADR orange card or any method for GSK vaccines, by practice (weeks 37–41)

Practice

Practice denominator

Number of GSK vaccinations

Patients with an AEI reported by any method within 7 days of vaccination (weeks 37–41)

Patients with an AEI reported by orange card within 7 days of vaccination (weeks 37–41)

% of AEIs reported using an ADR card (95%CI)

Practice 1

5769

740

78

54

69.2 (58.97–79.49)

Practice 2

4736

629

57

26

45.6 (33.33–57.89)

Practice 3

10,126

0

56

4

7.1 (1.79–14.29)

Practice 4

7383

881

41

19

46.3 (31.71–60.98)

Practice 5

11,683

9

16

4

25.0 (6.25–50.00)

Practice 6

14,689

2015

108

79

73.1 (64.81–81.48)

Practice 7

16,852

437

78

31

39.7 (29.49–51.28)

Practice 8

9531

1135

66

32

48.5 (36.36–60.61)

Practice 9

9526

1103

94

65

69.1 (59.57–78.72)

Practice 10

10,391

1666

127

20

15.7 (9.45–22.05)

Table 4

Rate of AEIs recorded in vaccinated patients (all reporting methods) for GSK vaccines, weeks 37–41

ISO week

Patients with AEIs reported 7 days post-vaccination

Vaccinated patients

Rate (95% CI) (%)

37a

58

1927

3.01 (2.28–3.79)

38

157

3166

4.96 (4.20–5.72)

39

192

2751

6.98 (6.03–7.96)

40

165

3162

5.22 (4.46–6.01)

41

149

1849

8.06 (6.91–9.30)

Total

721

12,855

5.61 (5.21–6.01)

AEIs adverse events of interest

a91 of these vaccines were administered in week 37 and given AEI cards ahead of the start of the study

The cluster effect of general practices has not been taken into account in the present interim analysis nor in the previous year’s study. Importantly, accounting for the cluster effect may increase the relative standard error, leading to wider confidence intervals around the estimated rates; hence, it should be accounted for in the final analysis [31].

Ethical Considerations

This study received a favourable ethics committee approval: Integrated Research Application System (IRAS) project ID: 211560; Research Ethics Committee (REC) reference: 16/NE/0271.

Results

Seasonal Influenza Vaccination Exposure

Most general practices influenza vaccination began in ISO week 37 (starting 12/09/16); however, 0.01% (n = 9) of patients were vaccinated prior to this, between weeks 33 and 36. All patients receiving a seasonal influenza vaccine between weeks 37 (September 12, 2016) and 41 (October 16, 2016) in participating practices were eligible for inclusion in the analysis (n = 12,855). Practices had vaccinated 13.1% (12,864/98,091) of their registered population by week 41, including the vaccinees prior to week 37 (Table 1).

The highest rate of seasonal influenza vaccination uptake was reported in subjects over 65 years of age with more than half (62.12%) vaccinated by week 41 (Table 1). Across recruited practices most adult vaccination was with GSK brand vaccine, whereas in children under 13 years most were vaccinated with non-GSK vaccine. Most are likely to have been given LAIV as per UK recommendations, with only those with chronic conditions given the GSK vaccine.

Return Rate of Orange Cards and Use in Reporting AEIs

A total of 334 cards were returned with AEIs recorded during weeks 37–41. Of adult patients (≥ 18 years old) reporting an AEI, 47.2% did so via the orange card method. The percentage of patients reporting AEIs using orange card was similar between 18–65 years and above 65 years of age; 46.1% and 47.8% respectively. There were no significant differences in the ages of those reporting AEIs via the orange card (p = 0.66) (Table 2).

There were significant differences in the number of practices recording AEIs using the orange cards (p < 0.001). Looking at the denominator and number of patients vaccinated with GSK vaccine, no clear pattern emerges (Table 3). Whilst it was intended that cards were exclusively distributed to patients receiving the GSK vaccine over the study period, five orange cards were returned reporting AEIs from vaccinees who did not receive GSK vaccine. We believe this was likely due to human error.

Total Reporting of AEIs

We report AEIs from both routine visit and ADR card, for GSK vaccines. The overall rate of AEIs reported was 5.61% (95% CI 5.21–6.01), although this varied from 3.01% to 8.06% between weeks (Table 4). There were 943 reports of AEIs in 721 individual patients.

Adults reported more AEIs than were reported by children or on their behalf; 5.9% (95% CI 5.5–6.3) compared to 1.5% (95% CI 0.7–2.3) (Table 5).
Table 5

Rate of AEIs recorded in vaccinated patients (all reporting methods) for GSK vaccines, by age band (weeks 37–41)

Age band

Patients with AEIs reported 7 days post-vaccination

Vaccinated patients

Rate (95% CI) (%)

6 months–5 years

9

565

1.59 (0.71–2.65)

6–12 years

4

238

1.68 (0.42–3.36)

13–17 years

0

72

0.0 (N/A)

Total children

13

875

1.49 (0.69–2.29)

18–65 years

256

3992

6.41 (5.66–7.19)

> 65 years

452

7988

5.66 (5.16–6.17)

Total adults

708

11,980

5.91 (5.49–6.34)

All ages (over 6 months)

721

12,855

5.61 (5.21–6.01)

AEIs adverse events of interest

A wide range of possible AEIs were reported, none were serious. Respiratory, musculoskeletal and neurological were the most common type of AEI reported using our ontological classification. Respiratory were the most frequent, 2.60% (95% CI 2.33–2.88), then musculoskeletal 1.82% (95% CI 1.59–2.05), then neurological 1.05% (95% CI 0.88–1.23). For full details, see Table 6. No serious adverse events (SAEs) such as Bell’s palsy, Guillain–Barré syndrome or anaphylaxis were reported to the participating GP.
Table 6

Rate of AEI reported from all sources for GSK vaccines by clinical category (weeks 37–41)

AEI category

EMA surveillance condition

Instances of AEIs for GSK vaccines

Denominator for GSK vaccine practices

Rate (%)

All

 

721

12,855

5.61 (5.21–6.01)

Respiratory

Conjunctivitis

23

12,855

0.18 (0.11–0.26)

Coryza

57

12,855

0.44 (0.33–0.56)

Cough

136

12,855

1.06 (0.89–1.24)

Epistaxis

9

12,855

0.07 (0.03–0.12)

Hoarseness

42

12,855

0.33 (0.23–0.43)

Nasal congestion

180

12,855

1.40 (1.20–1.61)

Oropharyngeal pain

82

12,855

0.64 (0.51–0.78)

Rhinorrhoea

149

12,855

1.16 (0.98–1.35)

Wheezing

40

12,855

0.31 (0.22–0.41)

Subtotal

334

12,855

2.60 (2.33–2.88)

Musculoskeletal

Arthropathy

53

12,855

0.41 (0.30–0.53)

Muscle aches/myalgia

222

12,855

1.73 (1.51–1.95)

Subtotal

234

12,855

1.82 (1.59–2.05)

Neurological

Headache

130

12,855

1.01 (0.84–1.19)

Peripheral tremor

17

12,855

0.13 (0.07–0.20)

Bell’s palsy

0

12,855

0.00 (N/A)

Guillain–Barré syndrome

0

12,855

0.00 (N/A)

Seizure/febrile convulsions

0

12,855

0.00 (N/A)

Subtotal

135

12,855

1.05 (0.88–1.23)

General symptoms

Drowsiness

41

12,855

0.32 (0.23–0.42)

Fatigue

102

12,855

0.79 (0.65–0.95)

Irritability

13

12,855

0.10 (0.05–0.16)

Malaise

3

12,855

0.02 (0.00–0.05)

Subtotal

126

12,855

0.98 (0.82–1.15)

Gastrointestinal

Decreased appetite

34

12,855

0.26 (0.18–0.36)

Diarrhoea

36

12,855

0.28 (0.19–0.37)

Nausea

47

12,855

0.37 (0.26–0.47)

Vomiting

14

12,855

0.11 (0.05–0.17)

Subtotal

101

12,855

0.79 (0.64–0.94)

Local symptoms

Local erythema

75

12,855

0.58 (0.46–0.72)

Subtotal

75

12,855

0.58 (0.46–0.72)

Fever/pyrexia

Fever (no numeric value)

28

12,855

0.22 (0.14–0.30)

Fever (value recorded)

57

12,855

0.44 (0.33–0.56)

Subtotal

85

12,855

0.66 (0.52–0.80)

Rash

Generalised rash

27

12,855

0.21 (0.13–0.30)

Rash

8

12,855

0.06 (0.02–0.11)

Subtotal

35

12,855

0.27 (0.19–0.37)

Sensitivity/anaphylaxis

Facial oedema

1

12,855

0.01 (0.00–0.02)

Anaphylactic reactions

0

12,855

0.00 (N/A)

Hypersensitivity reactions

7

12,855

0.05 (0.02–0.10)

Subtotal

8

12,855

0.06 (0.02–0.11)

AEIs adverse events of interest, EMA European Medicines Agency

For patients coded as having a fever or who had their temperature recorded (n = 57), in 91% of cases a valid numerical value was recorded. Most temperatures were recorded as tympanic measurements and near to normal, median 37 °C (interquartile range 0.5).

Comparison with Contemporaneous RCGP RSC Surveillance Network Data

No formal comparison was performed between the data collected from the ten GPs and the RCGP RSC surveillance network. Nevertheless, this evaluation provides insight into the added value of the enhanced methodology.

The overall rates of AEIs for those vaccinated with GSK vaccine in this study were higher than that reported by the rest of the RCGP RSC for the same weeks 6.88% (95% CI 6.35–7.42, 593/8615) vs. 3.30% (95% CI 2.68–3.96, 100/3028, p < 0.0001), respectively. All but two of the AEI categories were reported more frequently in the study practices using ADR cards than in the passive surveillance RCGP RSC practices (Table 7). Age-standardised rates were 7.41% (95% CI 4.97–12.74) and 2.62% (95% CI 1.98–4.05), respectively, for study and RCGP RSC practices.
Table 7

AEI rates reported via any method for 2016 ten orange card practices and RCGP RSC network 2016 (weeks 37–41)

AEI category using ontological classification

Vaccine brand

AEI rate for orange card practices weeks 37–41 (95% CI) 7 days post-vaccination (%)

AEI rate for RCGP RSC weeks 37–41 (95% CI) 7 days post-vaccination (excluding pilot) (%)

Pearson chi-square (p)

All AEI

GSK

6.88

3.30

< 0.0001

6.35–7.42

2.68–3.96

 

Respiratory

GSK

3.31

1.12

< 0.0001

2.94–3.69

0.76–1.52

 

Musculoskeletal

GSK

2.16

0.46

< 0.0001

1.86–2.47

0.23–0.73

 

Neurological

GSK

1.34

0.23

< 0.0001

1.10–1.58

0.07–0.43

 

General symptoms

GSK

1.20

0.13

< 0.0001

0.98–1.43

0.03–0.26

 

Gastrointestinal

GSK

1.06

0.07

< 0.0001

0.85–1.28

0.00–0.17

 

Local symptoms

GSK

0.77

0.00

< 0.0001

0.59–0.95

NA

 

Fever/pyrexia

GSK

0.69

0.96

0.136

0.51–0.86

0.63–1.32

 

Rash

GSK

0.34

0.50

0.221

0.22–0.46

0.26–0.76

 

Sensitivity/anaphylaxis

GSK

0.08

0.00

NA

0.02–0.15

NA

 

AEI adverse event of interest, CI confidence interval

Discussion

Key Results

This report shows that it is possible to collect data on AEIs occurring within the 7 days post vaccination in near real time from GP EHR systems. The ADR card system was used by adults and led to a doubling of reported events compared to passive surveillance. In adults, the rate of reported AEIs was consistent across age groups, with 46% of patients reporting AEI by means of an ADR card. The other half were recorded as part of standard healthcare. There was no difference between the working age and retired age in the rate of AEI reporting via orange cards; however, a lower proportion was recorded in children. There were differences in the number of orange cards recorded by practices, but this did not seem to be impacted by practice denominator or number of patients vaccinated. Further, there was no clear pattern in the AEIs across weeks 37–41; AEIs increase from weeks 37 to 79, decrease at week 40, and then increase again. The most commonly reported AEIs were respiratory, musculoskeletal and neurological symptoms (Table 6). Sensitivity and anaphylaxis were the least frequently recorded AEIs. For data related to GSK vaccines where both surveillance systems were assessed, a higher rate of AEIs was recorded using the enhanced passive surveillance over the standard surveillance by practices in the RCGP RSC sentinel network. Whilst the pattern of reported AEIs was similar in both surveillance systems, all but two AEI categories were recorded more often in practices participating in the enhanced passive surveillance and thus using ADR cards. It is likely that ADR card use stimulated more ADR recording.

Limitations

As this was an interim study, when the results were collected we did not know if all the cards had been collected or counted, so there may be some missing data. As compared to routine pharmacovigilance systems or purely passive surveillance, our approach may have stimulated reporting of AEIs, further research is needed to test this hypothesis. Similarly, less AEIs may have been reported on the orange cards because we did not take into account health literacy of our patients; however, given the fact that the orange cards were written in plain English and the large number of patients vaccinated, any effects of health literacy are likely to be minimal. We were unable to formally compare AEI rates with information reported in the GSK’s Fluarix Tetra clinical trials and summarised in the summary of product characteristics (SmPc); the categorisation of these data does not perfectly match those used in this study. However, for the common items, the reported rates from this enhanced study were systematically below the rates reported in the Fluarix Tetra SmPc. The fact that we did not record returned ADR indicating that no AEI had occurred is a limitation of this study and should be addressed in future surveillance of this type. There is also a need to understand better about the low rate of AEI recording in children and young adults under 18 years, even though there were low numbers of individuals in this age group.

Interpretation

We did not detect any safety signal that would have required further investigation between weeks 37 and 41.

Our interpretation is that customised ADR cards with predefined fields are an efficient way to stimulate the AEI reporting, including for local reactions, compared to standard surveillance of records.

In the future, some adjustment to enhanced safety surveillance might be necessary. Investigations are needed to further improve the return rate of orange cards and extend the use of this methodology to a wider population, inclusive of children receiving nasal influenza vaccine, which is part of the UK’s vaccination policy (this study was undertaken to satisfy EMA requirement for the marketing authorisation holder that sponsored this study) [32].

Generalisability

The level of detection of AEIs in our pilot study is comparable to that found in other observation systems [33, 34]. This seems to suggest that use of ADR cards together with EHR system weekly data extractions is an appropriate tool for adverse event surveillance. Other approaches have been proposed to meet EMA requirements [35].

Conclusions

This study demonstrated feasibility of conducting enhanced active surveillance per EMA recommendations. The use of a customised ADR card led to a doubling of AEIs reports over passive surveillance in adults. This study delivers meaningful results and could be used as a mechanism for meeting the requirements set out for safety surveillance.

Notes

Acknowledgements

Fluarix Tetra is a trade mark of the GSK group of companies.

We would like to thank the patients of the participating practices for allowing their data to be used for surveillance and research and for taking the time to complete the ADR cards. In addition, we thank Apollo Medical Systems for managing secure data extraction. We thank Florence Lair and Karen Stuttard from GSK for their contribution to this study; RCGP RSC for provision of aggregated comparator data. Finally, we would like to thank Ana Correa for her major contribution to this work while Primary Care Scientist for RCGP RSC at University of Surrey.

Funding

GlaxoSmithKline Biologicals SA (Grant no. EPI FLU-046 VS UK), Wavre, Belgium was the funding source and was involved in study design and interpretation. GlaxoSmithKline Biologicals SA, Wavre, Belgium also funded the article processing charges and open access fee. All authors had full access to all of the data in this study and take complete responsibility for the integrity of the data and accuracy of the data analysis.

Authorship

All named authors meet the International Committee of Medical Journal Editors (ICMJE) criteria for authorship for this article, take responsibility for the integrity of the work as a whole, and have given their approval for this version to be published.

Disclosures

Gaël Dos Santos reports he was employed by Business & Decision Life Sciences on behalf of GSK at the time of the study and is now employed by the GSK group of companies. Gaël Dos Santos is an employee of the GSK group of companies. Gaël Dos Santos holds shares in the GSK group of companies as part of their employee remuneration. Anne Schuind is an employee of the GSK group of companies. Anne Schuind holds shares in the GSK group of companies as part of their employee remuneration. Silvia Damaso is an employee of the GSK group of companies. Vishvesh Shende reports he is a consultant for VPN Consultancy Ltd, London UK (on Behalf of GSK). Simon de Lusignan reports that he will receive an honorarium from Seqirus for attending a global influenza advisory board and that he and his team have collaborated with the funder (GSK) before within the IMI-funded projects ADVANCE and DRIVE. Simon de Lusignan also reports personal fees from Seqirus, during the conduct of the study; and other funding from GlaxoSmithKline Biologicals SA, Wavre, Belgium, outside the submitted work. Rachel Byford, Filipa Ferreira, Ivelina Yonova and Christopher McGee have nothing to disclose.

Compliance with Ethics Guidelines

Ethics committee approval was obtained in England from the Health Research Authority (IRAS project ID 211560). GlaxoSmithKline Biologicals SA, Wavre, Belgium funded the project and was involved in study design and interpretation. This study received a favourable ethics committee approval: Integrated Research Application System (IRAS) project ID 211560; Research Ethics Committee (REC) reference 16/NE/0271. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. Informed consent was obtained from all individual participants included in the study.

Data Availability

The datasets generated during and/or analysed during the current study are not publicly available due data governance constraints under the Data Protection Act 1998, United Kingdom Parliament but are available from the corresponding author on reasonable request or through the standard process of application of the RCGP Research and Surveillance Centre (http://www.rcgp.org.uk/clinical-and-research/our-programmes/research-and-surveillance-centre.aspx) or by contacting Ivelina Yonova, i.yonova@surrey.ac.uk.

Open Access

This article is distributed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

Supplementary material

12325_2018_747_MOESM1_ESM.pdf (30 kb)
Supplementary material 1 (PDF 29 kb)
12325_2018_747_MOESM2_ESM.pdf (276 kb)
Supplementary material 2 (PDF 276 kb)

References

  1. 1.
    European Medicines Agency (EMA) Pharmacovigilance Risk Assessment Committee (PRAC). Interim guidance on enhanced safety surveillance for seasonal influenza vaccines in the EU. London, EMA: Ref: EMA/PRAC/222346/2014; 10 April 2014. http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/2014/04/WC500165492.pdf. Accessed 4 Sept 2017.
  2. 2.
    European Medicines Agency (EMA). Guideline on influenza vaccines non-clinical and clinical module. 2016. EMA/CHMP/VWP/457259/2014 Committee for Medicinal Products for Human Use. http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/2016/07/WC500211324.pdf. Accessed 4 Sept 2017.
  3. 3.
    European Medicines Agency (EMA). Explanatory note on the withdrawal of the note for guidance on harmonization of requirements for influenza vaccines and of the core SmPC/PL for inactivated seasonal influenza vaccines. London: EMA; 2014. Contract no.: EMA/CHMP/VWP/40560/2014. http://www.ema.europa.eu/ema/index.jsp?curl=pages/includes/document/document_detail.jsp?webContentId=WC500161022&mid=WC0b01ac058009a3dc.
  4. 4.
    Lee GM, Greene SK, Weintraub ES, et al. H1N1 and seasonal influenza vaccine safety in the vaccine safety datalink project. Am J Prev Med. 2011;41(2):121–8.  https://doi.org/10.1016/j.amepre.2011.04.004.CrossRefPubMedGoogle Scholar
  5. 5.
    Gao D, Yang H, Deng B, et al. Safety and immunogenicity of three seasonal trivalent influenza vaccines in the Chinese military. Hum Vaccin Immunother. 2016;12(10):2634–9.  https://doi.org/10.1080/21645515.2016.1194147.CrossRefPubMedPubMedCentralGoogle Scholar
  6. 6.
    Bettinger JA, Vanderkooi OG, MacDonald J, Kellner JD. Rapid online identification of adverse events after influenza immunization in children by PCIRN’s National Ambulatory Network. Pediatr Infect Dis J. 2014;33(10):1060–4.  https://doi.org/10.1097/INF.0000000000000373.CrossRefPubMedGoogle Scholar
  7. 7.
    Folkenberg M, Callréus T, Svanström H, Valentiner-Branth P, Hviid A. Spontaneous reporting of adverse events following immunisation against pandemic influenza in Denmark November 2009–March 2010. Vaccine. 2011;29(6):1180–4.CrossRefPubMedGoogle Scholar
  8. 8.
    de Lusignan S, Hague N, van Vlymen J, Kumarapeli P. Routinely-collected general practice data are complex, but with systematic processing can be used for quality improvement and research. Inform Prim Care. 2006;14(1):59–66.PubMedGoogle Scholar
  9. 9.
    de Lusignan S, van Weel C. The use of routinely collected computer data for research in primary care: opportunities and challenges. Fam Pract. 2006;23(2):253–63.CrossRefPubMedGoogle Scholar
  10. 10.
    de Lusignan S. Codes, classifications, terminologies and nomenclatures: definition, development and application in practice. Inform Prim Care. 2005;13(1):65–70.PubMedGoogle Scholar
  11. 11.
  12. 12.
    de Lusignan S, Dos-Santos G, Correa A, et al. Post-authorisation passive enhanced safety surveillance of seasonal influenza vaccines: protocol of a pilot study in England. BMJ Open. 2017;7(5):e015469.  https://doi.org/10.1136/bmjopen-2016-015469.CrossRefPubMedPubMedCentralGoogle Scholar
  13. 13.
    Correa A, Hinton W, McGovern A, et al. Cohort profile: royal College of General Practitioners Research and Surveillance Centre (RCGP RSC) sentinel network. BMJ Open. 2016;6:e011092.  https://doi.org/10.1136/bmjopen-2016-011092.CrossRefPubMedPubMedCentralGoogle Scholar
  14. 14.
    Avery AJ, Anderson C, Bond CM, et al. Evaluation of patient reporting of adverse drug reactions to the UK ‘Yellow Card Scheme’: literature review, descriptive and qualitative analyses, and questionnaire surveys. Health Technol Assess. 2011;15(20):1–234, iii–iv.  https://doi.org/10.3310/hta15200.
  15. 15.
    Public Health England, Department of Health, NHS England. Flu plan winter 2016/17. 2016. PHE publications gateway number: 2016027. https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/525967/Annual_flu_plan_2016_to_2017.pdf. Accessed 4 Sept 2017.
  16. 16.
    NHS England (2015). Enhanced Service Specification Seasonal influenza and pneumococcal polysaccharide vaccination programme 2015/16. https://www.england.nhs.uk/commissioning/wpcontent/uploads/sites/12/2015/09/seasnl-flu-pneumococcal-upd.pdf. Accessed 4 Sept 2017.
  17. 17.
    de Lusignan S, Correa A, Ellis J, Pebody R. Influenza vaccination: in the UK and across Europe. Br J Gen Pract. 2016;66(650):452–3.  https://doi.org/10.3399/bjgp16X686677.CrossRefPubMedPubMedCentralGoogle Scholar
  18. 18.
    Martin Arias LH, Sanz R, Sainz M, Treceno C, Carvajal A. Guillain–Barre syndrome and influenza vaccines: a meta-analysis. Vaccine. 2015;33(31):3773–8.  https://doi.org/10.1016/j.vaccine.2015.05.013.CrossRefPubMedGoogle Scholar
  19. 19.
    Szakacs A, Darin N, Hallbook T. Increased childhood incidence of narcolepsy in western Sweden after H1N1 influenza vaccination. Neurology. 2013;80(14):1315–21.  https://doi.org/10.1212/wnl.0b013e31828ab26f.CrossRefPubMedGoogle Scholar
  20. 20.
  21. 21.
    de Lusignan S. Codes, classifications, terminologies and nomenclatures: definition, development and application in practice. Inform Prim Care. 2005;13(1):65–70.PubMedGoogle Scholar
  22. 22.
    de Lusignan S. In this issue: ontologies a key concept in informatics and key for open definitions of cases, exposures, and outcome measures. J Innov Health Inform. 2015;22(2):170.  https://doi.org/10.14236/jhi.v22i2.170.PubMedGoogle Scholar
  23. 23.
    Liyanage H, Krause P, de Lusignan S. Using ontologies to improve semantic interoperability in health data. J Innov Health Inform. 2015;22(2):309–15.  https://doi.org/10.14236/jhi.v22i2.159.CrossRefPubMedGoogle Scholar
  24. 24.
    Public Health England. Vaccine uptake guidance and the latest coverage data. https://www.gov.uk/government/collections/vaccine-uptake#seasonal-flu-vaccine-uptake. Accessed 4 Sept 2017.
  25. 25.
    Harcourt S, Morbey RA, Bates C, et al. Estimating primary care attendance rates for fever in infants after meningococcal B vaccination in England using national syndromic surveillance data. Vaccine. 2018;36(4):565–71.  https://doi.org/10.1016/j.vaccine.2017.11.076.CrossRefPubMedGoogle Scholar
  26. 26.
    de Lusignan S, Correa A, Smith GE, et al. RCGP Research and Surveillance Centre: 50 years’ surveillance of influenza, infections, and respiratory conditions. Br J Gen Pract. 2017;67(663):440–1.  https://doi.org/10.3399/bjgp17X692645.CrossRefPubMedGoogle Scholar
  27. 27.
    Pathirannehelage S, Kumarapeli P, Byford R, Yonova I, Ferreira F, de Lusignan S. Uptake of a dashboard designed to give realtime feedback to a sentinel network about key data required for influenza vaccine effectiveness studies. Stud Health Technol Inform. 2018;247:161–5.Google Scholar
  28. 28.
    de Lusignan S, Stephens PN, Adal N, Majeed A. Does feedback improve the quality of computerized medical records in primary care? J Am Med Inform Assoc. 2002;9(4):395–401.CrossRefPubMedPubMedCentralGoogle Scholar
  29. 29.
    Scholz F. Confidence bounds and intervals for parameters relating to the binomial negative binomial Poisson and hypergeometric distributions with applications to rare events. 2008. http://www.stat.washington.edu/fritz/DATAFILES498B2008/ConfidenceBounds.pdf. Accessed 4 Sept 2017.
  30. 30.
    Office for National Statistics. Population estimates for UK, England and Wales, Scotland and Northern Ireland: mid-2016. https://www.ons.gov.uk/peoplepopulationandcommunity/populationandmigration/populationestimates/bulletins/annualmidyearpopulationestimates/latest. Accessed 4 Sept 2017.
  31. 31.
    Lauer SA, Kleinman KP, Reich NG. The effect of cluster size variability on statistical power in cluster-randomized trials. PLoS One. 2015;10(4):e0119074.  https://doi.org/10.1371/journal.pone.0119074.CrossRefPubMedPubMedCentralGoogle Scholar
  32. 32.
    Pebody RG, Green HK, Andrews N, et al. Uptake and impact of vaccinating school age children against influenza during a season with circulation of drifted influenza A and B strains, England, 2014/15. Euro Surveill. 2015;20(39):1–11.  https://doi.org/10.2807/1560-7917.es.2015.20.39.30029.CrossRefGoogle Scholar
  33. 33.
    Bardenheier BH, Duderstadt SK, Engler RJ, McNeil MM. Adverse events following pandemic influenza A (H1N1) 2009 monovalent and seasonal influenza vaccinations during the 2009–2010 season in the active component U.S. military and civilians aged 17–44 years reported to the Vaccine Adverse Event Reporting System. Vaccine. 2016;34(37):4406–14.  https://doi.org/10.1016/j.vaccine.2016.07.019.CrossRefPubMedGoogle Scholar
  34. 34.
    Haber P, Moro PL, Lewis P, Woo EJ, Jankosky C, Cano M. Post-licensure surveillance of quadrivalent inactivated influenza (IIV4) vaccine in the United States, Vaccine Adverse Event Reporting System (VAERS), July 1, 2013–May 31, 2015. Vaccine. 2016;34(22):2507–12.  https://doi.org/10.1016/j.vaccine.2016.03.048.CrossRefPubMedPubMedCentralGoogle Scholar
  35. 35.
    Demeulemeester M, Lavis N, Balthazar Y, Lechien P, Heijmans S. Rapid safety assessment of a seasonal intradermal trivalent influenza vaccine. Hum Vaccin Immunother. 2017;13(4):889–94.  https://doi.org/10.1080/21645515.2016.1253644.CrossRefPubMedGoogle Scholar
  36. 36.
    Chief Medical Officer. Annual flu plan 2016 to 2017. London; Department of Health, 2016. Page 41. https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/525967/Annual_flu_plan_2016_to_2017.pdf. Accessed 4 Sept 2017.

Copyright information

© The Author(s) 2018

Authors and Affiliations

  1. 1.Department of Clinical and Experimental MedicineUniversity of SurreySurreyUK
  2. 2.GlaxoSmithKlineWavreBelgium
  3. 3.Vaccine Discovery and Clinical Research and DevelopmentGlaxoSmithKlineRockvilleUSA
  4. 4.Vaccines Clinical Safety and Pharmacovigilance DepartmentVPN Consultancy Ltd (on behalf of GSK)LondonUK
  5. 5.Department of Clinical and Experimental MedicineRoyal College of General PractitionersLondonUK

Personalised recommendations