Health protection surveillance systems – England and Wales

Consent, confidentiality and the threat to public health surveillance
 
 
 

Definitions Ascertainment The degree of success in discovering all cases of a condition under surveillance—100% ascertainment would mean that all cases had been reported Bias Any effect causing the results of an investigation to differ from the true values. For example, ascertainment bias will result if those approached by doctors to consent, or those consenting to reporting differ from those not approached or not consenting Coding or encrypting Techniques that disguise personal identifiers or other data. For example, each patients NHS number is a form of code Duplicate reporting Reporting of the same individual or event more than once leading to over-counting—especially common with chronic conditions and rare events  Notification A special form of reporting whereby there is a statutory (legal) requirement that cases of certain infectious diseases should be reported to a designated local "proper officer" by the doctor providing care. In the UK there are no case definitions and no laboratory confirmation is required; the system is not well known by hospital clinicians; and seemingly no doctor has ever been prosecuted for not notifying. Hence it is one of the least complete and inaccurate forms of reporting. Its importance to public health is diminishing. Non-intrusive research—research that does not require patient contact and uses only pre-existing data Reconciliation Resolving potentially conflicting reports of cases of disease or infection, eliminating duplicates and incompatible data Reporting When applied to diseases or infection and public health, this is the forwarding of information about ill or infected individuals to those responsible for protecting and promoting the health of the public Surveillance The collection, collation, and analysis of health data and its dissemination along with analyses and conclusions, in a timely manner, to all those who need to know, so that important trends and events may be detected, and action necessary to promote and protect the public health may be taken Validation Confirming or improving the correctness of data about an event or individual—usually achieved by checking back to the original source or comparing data on the same event or individual from two or more sources

 
 
 

Health protection surveillance systems – England and Wales Adverse drug and vaccine reaction reporting (the "yellow card" system) Cancer registrations Congenital malformation reporting Clinician reporting of selected uncommon conditions (HIV, vCJD) Confidential enquiries into sudden deaths in infancy (CESDI) Data linkage studies of vaccine reactions Enhanced surveillance of selected conditions (tuberculosis, meningococcal disease, antibiotic resistant organisms, etc) General practice sentinel surveillance General practitioner databases Hospital admission reporting (HIPE system) Immunisation courses completed (COVER system) Diagnostic laboratory reporting National clinical surveillance units British neurological surveillance unit (BNSU) British ophthalmological surveillance unit (BOSU) British paediatric surveillance unit (BPSU) National CJD surveillance unit (NCJDSU) Welsh paediatric surveillance unit (WPSU) Outbreak detection and investigation Reference laboratory referral and reporting Screening completeness reporting Serious adverse incident reporting to regions Seroepidemiology for immunity to vaccine preventable diseases Statutory infectious disease notifications—food poisoning, tuberculosis, etc Unlinked anonymous seroepidemiology for HIV, hepatitis B and C Field investigation of disease "outbreaks," case-control or cohort studies, etc Ad hoc analytic studies using one or more of the above mechanisms

 

<posted as supplied by author>

Consent, confidentiality and the threat to public health surveillance

These ten examples are chosen from the generality of health protection mechanisms as examples of surveillance mechanisms protecting children that would be prejudiced if it became necessary to either obtain explicit patient consent or totally anonymise data.

1. Detecting and Controlling Outbreaks of Infection e.g. common source gastrointestinal infections - E. Coli O157 causing haemolytic-uraemic syndrome (HUS) and food poisoning by salmonella and campylobacter.

When gastrointestinal infections potentially associated with common sources (e.g. contaminated food products) are detected, public health action has to be undertaken rapidly. This relies on prompt reporting and the central referral of specimens to reference laboratories to detect common sub-types, even though this contributes little to direct patient care. Information gathering and case-control and other studies are undertaken immediately so as to allow rapid identification of risks and implementation of control measures to eliminate dangerous foods.^[1],[2],[3]

2. Maintaining Vaccine Safety – e.g. Measles, mumps and rubella (MMR) combined vaccine

In 1992 prompt reporting of cases of mild meningitis associated with MMR vaccine identified an adverse side-effect of a particular mumps vaccine, thereafter discontinued in the UK.^[4] This detection would have been severely inhibited if the microbiologist had to ask the paediatrician to obtain retrospective parental consent, especially when the connection was unproven.

Testing Hypotheses – Autism

The suggestion that childhood autism might be caused by MMR vaccine has generated huge anxiety in parents and doctors.^[5] One of the most definitive population-based studies showing no relationship used records of children diagnosed with autism and linked these with an independent dataset of vaccine records.^[6] The study is widely cited^[7] and its results are crucial in informing parents when making choices about immunisation for their children.² This was a non-intrusive investigation linking previously recorded information held in medical notes, special school records and immunisation records.^[8] A requirement for explicit parental consent to gathering data from their child’s record would have resulted in considerable delay and probably a major loss of data, which would have threatened scientific validity. Linkage would have been impossible without patient identifying details.

Detecting True Vaccine Adverse Reactions

Data linkage techniques can also identify and confirm rare vaccine risks.^[9] A similar study to that on autism linking hospital admissions and vaccine records has confirmed a rare (1 in 22,000) risk of idiopathic thrombocytopoenia purpura with MMR.^[10]

3. Monitoring vaccine effectiveness^[11],[12]

Sub-acute sclerosing pan-encephalitis (SSPE) is a rare late complication of measles resulting in an untreatable progressive deterioration of the nervous system. Combined clinician and laboratory reporting of SSPE has provided an important final measure of the effectiveness of measles control since 1970. Sufficient patient identifiers are required to reconcile cases from these two sources and eliminate duplicates. Elimination of measles in Europe is a WHO target and complete ascertainment of SSPE has been used to monitor the UK’s progress. It had been suggested that measles vaccine can cause SSPE. Wild and vaccine-related measles virus can be distinguished by genotyping isolates from SSPE cases and complete ascertainment of SSPE has been essential to exhaustively demonstrate the absence of vaccine-related SSPE following the millions of doses of MMR and MR vaccines used in the UK in the 1990s. This has only been possible retrospectively by having patient identifiers.^[13]

Immunisation against rubella is to prevent the severe sequelae that can follow fetal infection, congenital rubella syndrome (CRS). Surveillance for CRS through laboratories and clinicians has monitored the effectiveness and safety of rubella immunisation since 1971.^[14] Duplicate reporting is common (the international record is one case reported 6 times – A Medegalia, Canadian Paediatric Surveillance Programme 1998). Elimination of CRS is a WHO European target and complete ascertainment is necessary. Surveillance has revealed that recent CRS cases are due to circulation of wild virus in the indigenous population (not rubella immunisation) and that recent migrants are at heightened risk.⁹ This has allowed targeting of health protection to this vulnerable group but the validity of such findings would be threatened by an explicit consent requirement.^[15]

Immunisation against Haemophilus influenzae type b infection was introduced in 1992 resulting in a dramatic fall in invasive disease in children.¹⁰ The UK’s immunisation strategy, a primary course in infants without a booster, is unique in Europe. The strategy is economical, but it was necessary to monitor its effectiveness through complete ascertainment, which was again provided by complementary reporting by laboratories and clinicians.^[16] Similarly vaccination against type C meningococcal disease from 1999 onwards has required complete laboratory and clinician ascertainment of cases to monitor vaccine effectiveness and safety.^[17]

4. Preventing Further Cases of Serious Infectious Disease – e.g. Meningococcal disease and Haemophilus influenzae type b infection

Public health action has to take place rapidly after diagnosis of either meningococcal or haemophilus disease to reduce risk of further cases.^[18],[19] This requires rapid reporting and reconciliation of reports from clinicians, public health doctors, peripheral and reference laboratories. It is also essential that meningococcal specimens are sent on to the reference laboratory to demonstrate whether or not the type of N. meningitidis requires immunisation and to detect outbreaks of especially virulent strains. Most recently this surveillance has demonstrated increasing numbers of an unusual type of meningococcal disease (W135) among ethnic minority groups travelling to the Middle East and their contacts, requiring a different vaccine for those at risk.^[20]

5. Detecting Adverse Effects of Medicines - Reye Syndrome and avoiding aspirin in childhood^[21]

Surveillance using clinicians’ reports first detected an association between childhood aspirin ingestion and Reye syndrome and then demonstrated that a warning against aspirin use by children resulted in a substantial fall in cases. ^[22] These reports come through many routes and need reconciliation to ensure complete ascertainment and elimination of duplicates. The Medicines Control Agency has stated that confidential reports through the ‘Yellow Card’ system do not require consent. The basis for this statement is unclear but gaining consent would further reduce ascertainment and would probably add significantly to any bias in reporting.

6. Detecting and Controlling Very Rare Infections - variant Creutzfeldt-Jakob disease (vCJD) in children

This is being carried out through the British Paediatric Surveillance Unit (BPSU). The aim is to identify all children with conditions causing progressive intellectual and neurological deterioration (PIND) so as to detect any cases of vCJD. It is necessary to achieve the best ascertainment possible in order to estimate the extent to which children in the UK are at risk of vCJD.^[23] In almost 6 years paediatricians have reported over 1000 children with neurological disorders among whom there have been only six cases of vCJD (four definite and two probable), all in teenagers. Specimens from children with PIND are frequently sent by their paediatricians to specialised laboratories in the UK or elsewhere. It is essential to have the child’s name and date of birth in order to track these complex pathways of investigation. As with all BPSU studies the information about patients is held confidentially by the surveillance team and is not passed on or published in a form that would allow identification of individual patients.

7. Auditing the Impact of New Treatments - Haemorrhagic disease of the new-born (vitamin K deficiency)

Surveillance for this condition has become necessary because following concern over the safety of injected Vitamin K. New-borns now receive oral preparations, the effectiveness of which are uncertain.^[24] Surveillance is being carried out by the BPSU and complete ascertainment is needed to determine if the change in therapy is resulting in an increase in disease. Significant under-reporting would invalidate comparison with an earlier survey.

8. Researching Possible Child Abuse – e.g. Sub-dural haematoma

Surveillance for sub-dural haematoma was undertaken by the BPSU to estimate the incidence and proportion of cases due to non-accidental injury (child abuse) so as to inform key opinions given in courts that determine the future of children.^[25] It seems very likely that parental consent would not have been given for cases of non-accidental injury which would have seriously biased the results. Equally it is unclear whether the GMC expects parental consent to be given before children can be placed on child protection registers.

9. Auditing of Hospital Practices to Inform NHS Planning – e.g. Biliary atresia

A study undertaken through clinician reporting found that children treated surgically for biliary atresia in centres undertaking substantial numbers of operations annually did considerably better than those at centres treating smaller numbers.^[26] This led directly to NHS guidance limiting the centres undertaking such surgery.^[27] Complete ascertainment was essential and clinicians might have been unwilling to seek parental consent to report cases when children had done poorly.

10. Informing Screening Policy and Monitoring Practice – e.g. Antenatal HIV screening

Surveillance demonstrated that HIV positive mothers often had their infection diagnosed only when a child developed AIDS.^[28] This led to a multidisciplinary iniative to improve antenatal screening and the government’s adoption of an HIV screening target the monitoring of which will be based on clinician reporting.^[29],[30] Complete ascertainment is needed or the success of screening will be underascertained while patient details are required to eliminate duplicate reports.²⁸ Other surveillance studies have informed national screening policy by providing evidence that antenatal syphilis screening should be retained but that other infection screening should not be adopted because the burden of fetal disease was too low.^[31]

References

1 British Paediatric Surveillance Unit, 12^th Annual Report 1997-8, Royal College of Paediatrics and Child Health, London 1998.

2 Killalea D, Ward L, de Louvois J, Majiid F, Stuart JM, Wall PG, Susman M et al. International epidemiological and microbiological study of Salmonella agona infection from a ready to eat savoury snack. I. England and Wales and the United States. BMJ 1996; 313: 1105-7.

3 Communicable Disease Surveillance Centre. Salmonella anatum infection in infants linked to dried milk. CDR Weekly 1997; 7: 33,36.

4 Miller E, Goldacre M, Pugh S, Colville A, Farrington P, Flower A, Nash J, Macfarlane L, Tettmar R. Risk of aseptic meningitis after measles, mumps and rubella vaccine in UK children. Lancet 1993; 34:979-82.

5 Elliman D, Bedford H. MMR Vaccine: the continuing saga. BMJ 2001; 322:183-4.

6 Taylor B, Miller E, Farrington CP et al. Autism and measles, mumps and rubella vaccination: no epidemiological evidence for a causal association. Lancet 1999:353:2026-9.

7 Anonymous. The safety of MMR (Committee on Safety on Medicines and the Medical Controls Agency) . Current Problems In Pharmacovigilance 1999; 25: 9-10.

8 Department of Health – Ethical Review of multi-centre research in the NHS where these is no local researcher (Report of a Working Group – chair Dr Hugh Davies). www.doh.gov.uk/research

9 Miller E, Waight P, Farrington. Safety assessment post-licensure in Plotkin S, Brown F, Horaud F (eds). Preclinical and clinical development of new vaccines. Dev Biol Stand Basel Karger 1998; 95: 235-243.

10 Miller E, Waight P, Farrington CP, Andrews N, Stowe J, Taylor B. Idiopathic thrombocytopenic purpura and MMR vaccine. Arch Dis Child 2001; 84:227-9.

11 British Paediatric Surveillance Unit, 12^th Annual Report 1997-8, Royal College of Paediatric and Child Health, London 1998 (pages 20-21).

12 Communicable Disease Surveillance Centre. Meningococcal disease falls in vaccine recipients. CDR 2000; 10:133,6.

13 British Paediatric Surveillance Unit, 12^th Annual Report 1997-8, Royal College of Paediatrics and Child Health, London 1998..

14 Tookey PA, Peck ham CS. Surveillance of congenital rubella in Great Britain, 1971-96. BMJ 1999;318:769-70.

15 Jacobsen SJ, Xia Z, Campion ME, Darby CH, Plevak MF, Seltman KD, Melton LJ. Mayo Clin Proc 1999; 74: 330-8.

16 Booy R, Heath PT, Slack MPF, Begg N, Moxon ER. Vaccine failures after Haemophilus influenzae type b conjugate vaccine without booster. Lancet 1997; 349:197-202

17 CDSC The impact of conjugate group C meningococcal vaccination CDR January 11^th 2001 www.phls.co.uk/publications/CDR%20Weekly/archive/news0201.html#impact

18 Communicable Disease Surveillance Centre. Guidance for consultants in communicable disease control. CDR Review 1995; 5: R189-195

19 Communicable Disease Surveillance Centre. Guidance for consultants in communicable disease control. CDR Review 1995; 5: R189-195

20 Communicable Disease Surveillance Centre. Meningococcal infection and the Hajj. CDR Weekly 2001; 11: 1-2 www.phls.co.uk/publications/CDR%20Weekly/PDF%20files/cdr0201.pdf

21 British Paediatric Surveillance Unit, 12^th Annual Report 1997-8, Royal College of Paediatrics and Child Health, London 1998 (pages 25-27).

22 Hardie RM, Newton LH, Bruce JC, Glasgow JFT, Mowat AP, Stephenson JBP, Hall SM. The changing clinical pattern of Reye’s syndrome 1982-1990. Arch Dis Child 1996; 74: 400-5.

23 Verity CM, Nicoll A, Will RG, Devereux G, Stellitano L. Variant Creutzfeldt-Jakob disease in UK children: a national surveillance study. Lancet 2000; 356: 1224-7.

24 Dwyer T. Ponsonby A-L. Vitamin K at birth. BMJ 1996; 313: 179-180.

25 British Paediatric Surveillance Unit, 12^th Annual Report 1997-8, Royal College of Paediatrics and Child Health, London 1998 (pages 31-32).

26 McKiernan PJ, Baker AJ, Kelly DA. The frequency and outcome of biliary atresia in the UK and Ireland. Lancet 2000; 355: 25-8

27 NHS National Specialist Commissioning Advisory Group. Guidance on liver services for children, Department of Health February 2000 www.doh.gov.uk/nscag/service1.htm

28 Nicoll A, McGarrigle C, Brady AR, Ades AE, Tookey, Duong T, Mortimer J, Cliffe S, Goldberg D, Tappin D, Hudson C, Peckham C for the principal collaborators. Epidemiology and detection of HIV-1 among pregnant women in the United Kingdom: results from national surveillance 1988-96. BMJ 1998; 316: 253-8.

29 Intercollegiate Working Party for Enhancing Voluntary Confidential HIV Testing in Pregnancy Reducing mother to child transmission of HIV infection in the UK. Royal College of Paediatrics and Child Health, London, 1998.

30 Department of Health. Health Services Circular. Reducing mother to baby transmission of HIV. (HSC 1999/183). NHS Executive, London August 13^th 1999

31 Nicoll A, Lynn R, Rahi J, Verity J, Haines L. Public health outputs from the British Paediatric Surveillance Unit and similar clinician-based systems. J R Soc Med 2000; 93: 580-5.