SOPHID Frequently Asked Questions

• Introduction
• Data dissemination
• Data collection
• Confidentiality

Introduction

What is SOPHID?

The Survey of Prevalent HIV Infections Diagnosed (SOPHID) began in 1995 and is a cross-sectional survey of all individuals with diagnosed HIV infection who attend for HIV-related care within the NHS in England, Wales, and Northern Ireland (E, W & NI) within a calendar year. SOPHID is funded by the Department of Health and the London Specialised Commissioning Group (LSCG) and is conducted by the Heath Protection Agency's Centre for Infections (CfI). Scottish data is collected separately in Scotland by Health Protection Scotland (formerly SCIEH), and is incorporated into UK totals.

The survey is run twice a year in London and annually outside London. The London survey includes Brighton, Hastings and Eastbourne and covers attendances from January to June and from July to December, whereas the survey outside London covers attendances for the whole calendar year. Annual national results are based on data from the two London surveys, the survey run outside London and paediatric data (children under 15 years of age) from the Institute of Child Health. Duplicates are removed from the final amalgamated dataset.

What does SOPHID aim to do?

• Provide an epidemiological profile of, and determine the prevalence of, individuals living with diagnosed HIV in England, Wales and Northern Ireland.

• Provide information to health areas (Primary Care Trusts (PCT)/Strategic Health Authorities (SHAs) and Health Protection Units) for assessing and predicting demands on local health and social services and for developing targeted local health prevention strategies.

• Support the allocation of national funds for the treatment and care of HIV-infected individuals and help target national HIV interventions.

• Support the commissioning of HIV treatment, care, and prevention services.

How does SOPHID complement other national HIV surveillance systems?

Difference between newly diagnosed and prevalent diagnosed infections.
The HIV and AIDS patient reporting system (HAP) collects information on individuals with newly diagnosed HIV and AIDS. Neither annual, nor cumulative reports of new HIV diagnoses can give a measure of individuals who are currently living with diagnosed HIV in the UK since these cannot take into account movement out of the UK or unreported deaths. SOPHID collects reports of individuals living with HIV who have been in contact with the NHS and is therefore a good measure of the annual prevalent HIV diagnosed population and the demand on services.

Estimating the number of HIV-infected but undiagnosed individuals
Some individuals infected with HIV and living in the UK have not yet been diagnosed. These individuals cannot be detected by either surveillance of new diagnoses (HAP) or SOPHID. The total estimated national number of HIV-infected individuals (diagnosed and undiagnosed) can be estimated using multiple sources of information including SOPHID, the Unlinked Anonymous Prevalence Monitoring Program (UAPMP), the National Survey of Sexual Attitudes and Lifestyles II (NATSAL) and census data. Annual estimates of the total number of individuals living with (diagnosed and undiagnosed) HIV in the UK are available in the HIV and STI annual report.

How is SOPHID run?

• Each survey, providers of HIV treatment or care are requested to collate a list of all individuals who have attended for HIV-related care. Anonymised epidemiological and demographic data collected on each patient (see section 16) are forwarded to CfI.

• Data are cleaned at CfI and missing or inconsistent data are followed up. Duplicate records and those that do not meet the inclusion criteria are excluded. The cleaned data sets are amalgamated into a single database.

• Summary tables broken down by regional, SHA and PCT level are disseminated to key health professionals in regional and local areas.

• Back to top

Data Dissemination

How are SOPHID data disseminated?

SHA and PCT level data tables are sent out annually to all Consultants in Communicable Disease Control (CCDCs), Regional Epidemiologists (REs), and our SOPHID facilitators. If you have any queries on your local data then these are the people to contact first. SHA and PCT sexual health leads are informed that the data can be requested.

No individual data are given out but aggregate data summaries at SHA level are published on the HPA website.

• Regional Epidemiologists receive their SHA tables and eight PCT tables for each PCT in their SHA.
• SHA and PCT Sexual Health Leads are sent an email by the DH informing them that epidemiological data for their residents in the form of tables giving breakdowns by ethnicity, clinical stage, age group and SHA of treatment, are available for downloading.
• The Department of Health receives a copy of all tabular summaries at the SHA and PCT level.
• Data requests from NGOs, researchers, the media and members of the public involving SOPHID information broken down to sub-regional level are, where appropriate, referred back to local holders of the data, as they are more aware of the local confidentiality issues. Although only aggregate data are released, there may be sensitivities, particularly in low prevalence areas.

SOPHID data are distributed to the following groups in the first instance:

Any queries about local data should be addressed to these people in the first instance

What routine outputs does SOPHID produce? What level of geography is data produced by?

Data are produced by regional, SHA and PCT levels. The following tables are produced annually and distributed to key health professionals (see above) for further dissemination at local level. Suggested citation for use is "Centre for Infections, Health Protection Agency".  (National and SHA tables are available online)

a. By PCT of residence

Table 1:   Numbers of diagnosed HIV-infected individuals seen for care by most advanced clinical stage, gender and age group.
Table 2a:  Numbers of diagnosed HIV-infected individuals seen for care by probable route of HIV-infection, sex and survey year.
Table 2b:  Numbers of diagnosed HIV-infected individuals seen for care by ethnic group, sex and survey year.
Table 2c:  Numbers of diagnosed HIV-infected individuals by index of multiple deprivation** and survey year.
Table 3a:  Numbers of diagnosed HIV-infected individuals seen for care by site of treatment and survey year.
Table 3b:  Number of infants who received HIV-related care because they were born to HIV-infected women in the survey year, but who are uninfected or whose infection status was indeterminate (at least 98% of indeterminate infants will subsequently be confirmed as uninfected), by site of treatment and survey year.
Table 4:   Numbers of diagnosed HIV-infected individuals seen for care by Local/Unitary Authority and postal district of residence.
Table 4a: Numbers of diagnosed HIV-infected patients by middle super output area (MSOA) of residence.
Table 5:   Numbers of diagnosed HIV-infected individuals seen for care by most advanced clinical stage and level of anti-retroviral therapy.
Table 6:   Numbers of diagnosed HIV-infected individuals seen for care by last CD4 count and level of anti-retroviral therapy.

b. By SHA of residence

Table 7a: Numbers of diagnosed HIV-infected individuals seen for care by site of treatment and survey year.
Table 7c: Number of infants who received HIV-related care because they were born to HIV-infected women in the survey year, but who are uninfected or whose infection status was indeterminate (at least 98% of indeterminate infants will subsequently be confirmed as uninfected), by site of treatment and survey year.
Table 8a: Numbers of diagnosed HIV-infected individuals seen for care, by Primary Care Trust of residence and survey year.
Table 8b: Numbers of diagnosed HIV-infected individuals seen for care seen, by Local Authority/Unitary Authority of residence and survey year.
Table 10: Numbers of diagnosed HIV-infected individuals seen for care by route of infection, ethnic group and gender.
Table 11a: Number of diagnosed HIV-infected individuals seen for care by site of treatment and PCT of residence.

 c. By SHA of treatment

Table 7b: Numbers of diagnosed HIV-infected individuals seen for care, by Strategic Health Authority of residence and survey year.
Table 9: Numbers of diagnosed HIV-infected individuals seen for care by site of treatment and survey year.
Table 11b: Number of diagnosed HIV-infected individuals seen for care by site of treatment and PCT of residence.

 d. National level tables, by SHA 

Table 1a: Numbers of diagnosed HIV-infected individuals seen for care by SHA of residence, England, Wales and Northern Ireland.
Table 1b: Numbers of diagnosed HIV-infected individuals seen for care by PCT of residence, England, Wales and Northern Ireland.
Table 2: Numbers of diagnosed HIV-infected individuals seen for care by area of residence and exposure category, England, Wales and Northern Ireland.
Table 3: Numbers of HIV-infected individuals seen for care by area of residence, ethnic group and sex, England, Wales and Northern Ireland.
Table 4: Numbers of diagnosed HIV-infected individuals seen for care by area of residence, age group and sex, England, Wales and Northern Ireland.
Table 5: Numbers of diagnosed HIV-infected individuals seen for care by area of residence and level of antiretroviral therapy, England, Wales and Northern Ireland.
Table 6: Numbers of diagnosed HIV-infected individuals seen for care by SHA of residence and SHA of care, England, Wales and Northern Ireland.
Table 7: Number of infants who received HIV-related care because they were born to HIV-infected women in the survey year, but who are uninfected or whose infection status has not yet been established, by area of residence, England, Wales and Northern Ireland.

What if my SOPHID PCT/SHA totals are not what I expected?

a) Why does the total of all the PCTs not add up to my SHA total? Why does the total of the SHAs not add up to my regional total?

On occasion, the site of treatment is aware of the SHA of residence for an individual but not the PCT. In these cases, the individual will be added to the SHA total but not to any PCT, therefore the sum of the PCTs totals may be slightly less then the total for the SHA. The same is true for why the total of the individuals seen at all the SHAs in a region may be slightly less than the total for that region.

b) Why does the number of records I reported to SOPHID differ from the number of records in the tables I received back?

The following procedures may often cause final SOPHID figures to differ from those originally reported:

• The SOPHID survey can be used to allocate an individual to a PCT if the residence information provided by the site of treatment maps to one PCT only. Records where PCT cannot be ascertained are marked as ‘not known’ and not allocated to an area of residence. This may result in a discrepancy between the number of individuals reported and final number included into PCT/SHA tables.
• A single site of treatment sometimes reports duplicate records of the same individual - these are identified and removed.
• If insufficient information was provided for an individual to enable de-duplication (no soundex, date of birth or sex) then the record is removed following consultation with the reporting site.
• If an individual has been reported but later found to have actually been seen outside of the survey period then the individual is removed following consultation with the reporting site.
• More than one record of the same individual are often reported from more than one site of treatment - these are identified and removed.
  

Can I get additional data information?

The first port of call should always be your local data holder (see above). However if the data that they hold are not appropriate for your needs (e.g. national breakdowns), bona fide data requests will be considered by the SOPHID team. The SOPHID advisory group has agreed a level of breakdown of the data that can be made available without concerns over confidentiality. Requests for non-standard data breakdowns should be made by completing our data request form.

Download our data request form (Word Document, 406 KB)

What are the limitations?

Although every effort is made to include every individual seen for HIV related care in every site in England, Wales, and Northern Ireland, there are likely to be a small number of individuals who are not reported. Also, some individuals who are well may decide not to receive any treatment for a year or more, in which case, they may not be reported to the survey.

What is the smallest breakdown I can receive?

For confidentiality reasons, no information can be broken down smaller than PCT level and all data will be in aggregate form. No identifiers can be provided.

• Back to top

Data Collection

Who is included in SOPHID?

Each centre/NHS trust providing HIV treatment or care services in England, Wales, and Northern Ireland.

Individuals (15 years of age and over at date last seen for care) with diagnosed HIV infection, who received HIV-related treatment or care within the survey period. This includes individuals who were newly diagnosed and those who have died during the survey period.

Note: Children reported to the Institute of Child Health (ICH) are added to the final dataset and not collected as part of SOPHID returns. These will include all children regarded as prevalent cases for the survey year, as well as all infants reported to have been born to HIV infected mothers during the survey year regardless of their infection status. These infants will have received HIV-related care while their infection status was being established (many of whom will subsequently be confirmed as uninfected). Children who are uninfected or whose infection status has not yet been confirmed will not be included in the final tables we produce, apart from in tables 3b and 7c, which solely represent these infants.

Note: Individuals receiving HIV-related care in Scotland are added to the final dataset and not collected as part of SOPHID returns.

Who is excluded

• Individuals seen only for specialist referral (e.g. dentistry)
• Individuals not seen in the survey period
• Individuals seen only in the voluntary sector or by social services

What data are collected for the SOPHID survey?

Fields we require on each individual are as follows:

For information on the codes we use for these feilds please request our SOPHID protocol.

Is reporting to SOPHID voluntary?

Participation in SOPHID is not compulsory. However Department of Health allocations for each PCT include an adjustment for HIV-infected residents, which are calculated using SOPHID data: therefore it is in each PCTs interest to respond. In addition, many London trusts have Service Level Agreements that contractually require timely SOPHID reporting.

What about individuals who are seen at more than one centre? What is de-duplication?

At the end of the year when all the data are received multiple reports of an individual are removed through the process of "de-duplication". These are identified based on matched clinic-attributed patient ID number, soundex, date of birth and sex. Individuals who are seen at more than one centre are assigned to the centre at which they were seen most recently. All routine survey summaries are then prepared using this de-duplicated dataset.

How is residence obtained?

Residence is obtained for an individual using residence-based information provided to us by the site of treatment. PCTs now have responsibility for commissioning health care for their local populations and consequently it is necessary for all individuals reported to SOPHID to have their PCT of residence allocated. All residence-based information is held on a secure database and such data is not shared. Strict attention to confidentiality is maintained at every stage of data collection, analysis and storage. For further information on safeguarding the confidentiality of patient information whilst protecting public health please see: http://www.hpa.org.uk/confidentiality/

Are children included in the final data?

Although not collected from sites of treatment by SOPHID, data on children (under 15 years of age) are reported via the National Study of HIV in Pregnancy and Childhood (NSHPC) and the Collaborative HIV Paediatric Study (CHIPS) and are included in the SOPHID tables. Infants may receive HIV-related treatment or care even if they are later found to be uninfected. For this reason, 3b, 7c and National table 8 are provided detailing the numbers of these infants.

• Back to top

Confidentiality

What confidentiality procedures are in place?

We ask that all data be sent to us electronically via the Document Gateway. If that is not possible then we will also accept it by password protected excel file via email (for instructions on password protecting data see below). When we receive data we notify the site or SOPHID facilitator. All staff are briefed on Caldicott guidelines and are aware of the sensitive nature of the data. When ad hoc queries come in we do not provide any breakdowns below PCT level, and any tables we send out at PCT level are password protected. For full confidentiality guidelines see our protocol.

To maintain patient confidentiality, soundex codes are used instead of names. All data are stored on restricted and secure databases, with strict adherence to the Data Protection Act and Caldicott Guidelines. SOPHID has approval under the section 60 regulations of the Health and Social Care Act 2001.
For confidentiality reasons, data is not broken down within a region smaller than a PCT and all outputs are presented in aggregate form. No identifiers are provided.
For further information on safeguarding the confidentiality of patient information whilst protecting public health please see: http://www.hpa.org.uk/confidentiality/

What is the Document Gateway and how do I use it?

The Document Gateway is a secure Internet site recommended by the CfI, which allows transfer of data discreetly and confidentially to a secure server located within the HPA. SOPHID facilitators can access the protocol, electronic forms, and their own area folder. The SOPHID data can be placed into this folder which is accessible only by SOPHID staff and themselves. For instructions on using the gateway please contact the SOPHID team or see the Document Gateway user guide (PDF, 134 KB).

How do I password-protect my data?

If you are not using the secure document gateway when sending HIV surveillance data to the HPA we ask that it is password protected:

To password protect the Excel file in which you send the data, go to: 'File' in the toolbar, select 'Save as', click on 'Tools', select 'General options', Enter your password in 'Password to open' and select 'OK'
Re-enter the password and click on 'OK', then 'Save'.

What is a soundex code?

A soundex code is a 4-character code of an individual's surname that we use to recognise duplicates in the data set. A Soundex code is derived from, but not unique to, the surname. Software for soundexing is available free on request or can be accessed via the document gateway. We recommend that you use our software for consistency of soundexing from all reporters.

Can you provide data showing an individual's soundex?

No, no information with identifiers will be disseminated.

Can you provide data at the individual level on those resident in our area but treated elsewhere?

We can provide numbers and site of treatment only, no further information.

• Back to top

Last reviewed: 23 February 2011