4 Review of pre-pilot social marketing research - UEL

Document Sample
4 Review of pre-pilot social marketing research - UEL Powered By Docstoc
					Tower Hamlets NHS Health Check
Pilot Programme Evaluation


December 2009
Prepared by

Report Authors

Patrick Tobi

Shahana Lais

Emee Vida Estacio

Jananezwary Kanaparthy


Gail Barrow-Guevara

Adrian Renton

Contact details
020 8223 6225

Institute for Health and Human Development
University of East London
Suite 250, University House
Water Lane
London E15 4LZ

Commissioned by


Many people have contributed to the preparation
of this report. We thank all the service users and
staff of Chrisp Street, Jubilee Street and Mission
Practices, and St. Stephen’s Health Centre, who
shared their views with us. Davinia McFarlane of
City Gateway provided interpreters for the patient
focus group event. John Robson, CVD GP lead in
Chrisp Street Practice and co-developer of the
QRISK assessment tool shared valuable insights.
Our special thanks go to Eleonora Merkas, CVD
Nurse Specialist, and Amanda Watson,
Commissioning Manager for Self Care and Healthy
Lifestyles at NHS Tower Hamlets for the
information they gave and facilitation in accessing
information sources.

Institute for Health and Human Development 2009       ihhd
      List of tables and figures                                          5
      Acronyms                                                            5
1     Introduction                                                        6
1.1   Wider context                                                       6
1.2   Objectives of the evaluation                                        8
2     Evaluation methodology                                              9
2.1   Overview                                                            9
2.2   Data generation methods                                             9
2.3   Participants                                                        10
2.4   Process                                                             10
2.5   Data analysis                                                       11
3     Local implementation                                                11
3.1   Service model                                                       11
3.2   Risk assessment tool                                                12
3.3   Implementation protocol                                             13
4     Review of pre-pilot social marketing research                       14
5     Programme analysis                                                  16
5.1   Practice context                                                    16
5.2   Organisational readiness                                            18
5.3   Service factors                                                     19
5.4   Service enablers and barriers                                       23
6     Service users                                                       25
6.1   Insights                                                            25
6.2   Audience segmentation                                               28
7.    Environmental analysis                                              29
8.    Conclusions and recommendations                                     32
8.1   Key learning                                                        32
8.2   Options for programme development                                   33
9     Appendix                                                            35
      Appendix 1. Ethics approval letter
      Appendix 2. Information sheet
      Appendix 3. Consent form
      Appendix 4. Invitation letter to community café
      Appendix 5. Topic guide for patients
      Appendix 6. Topic guide for practice staff
      Appendix 7. Topic guide for PCT commissioners and programme leads

List of tables and figures

Table 1.    Distribution and description of participants
Table 2.    Advantages and disadvantages of QRISK2 compared to Framingham risk tool
Table 3.    Preliminary audience segments and key insights from the pre-pilot social
            marketing research
Table 4.    Summary of insights from the pre-pilot social marketing research
Table 5     Practice context
Table 6.    Key programme enablers and barriers
Table 7.    Environmental analysis matrix

Figure 1.   Overview of NHS Health Check
Figure 2:   NHS Health Check stages and processes
Figure 3.   Health Check implementation process in Tower Hamlets
Figure 4.   Organisational readiness for the Health Check
Figure 5    NHS Health Check patient segments

BME         Black and minority ethnic
BMI         Body mass index
CEG         Clinical effectiveness group
CHD         Coronary heart disease
CKD         Chronic kidney disease
CVD         Cardiovascular disease
DASL        Drug and alcohol service for London
DM          Diabetes mellitus
DH          Department of health
DNA         Did not attend
eGFR        Estimated glomerular filtration rate
EMIS        Egerton medical information system
GP          General practitioner
HCA         Health care assistant
HCP         Health care professional
HDL         High density lipoprotein
IFG         Impaired fasting glucose
IGT         Impaired glucose tolerance
IMD         Index of multiple deprivation
LFT         Liver function tests
LES         Local enhanced service
LAP         Local area partnership
NICE        National institute for health and clinical excellence
NHS         National health service
PCT         Primary care trust
PN          Practice nurse
SOP         Standard operating procedures
TH          Tower Hamlets
VRA         Vascular risk assessment

NHS Health Check (formally known as the vascular check programme and, before that, the
vascular risk assessment and management programme) is a new national initiative for people in
England aged between 40 and 74. The objective of the programme is to identify and assess an
individual‟s risk of developing coronary heart disease (heart attack and angina), stroke, diabetes
and kidney disease; communicate the risk in a way that the individual clearly understands; and
manage the risk appropriately through tailored advice, clinical management, signposting or
referral to other lifestyle interventions and services. It includes being recalled every 5 years for

There are good reasons for bringing these diseases together under an integrated management
umbrella. First, although they affect the body in different ways, they share common risk factors -
obesity, physical inactivity and a sedentary lifestyle, smoking, high blood pressure, raised
cholesterol levels and impaired glucose regulation (higher than normal blood glucose levels) –
which all raise the risk of vascular disease. Second, using single risk factors such as weight or
serum cholesterol by themselves are less reliable than multi-factor assessment in predicting the
future risk of heart disease. Thirdly, having one of these vascular conditions increases the
likelihood of developing the others or can worsen their effect if already present. In addition, there
is evidence that this kind of integrated approach works and is cost-effective.1

Ethnicity and social disadvantage are also recognised factors in the development of vascular
disease - people of South Asian and African Caribbean origin and the lowest socio-economic
groups are most at risk. In general, the risk factors for vascular disease broadly fall into two
groups: „fixed‟ factors such as age, gender and ethnicity and „modifiable‟ factors such as
smoking, obesity, physical inactivity, high blood pressure and raised blood cholesterol which are
amenable to intervention.


The national policy context of NHS Health Check is set out in the Department of Health strategy
Putting Prevention First2 and marks a decisive shift in emphasis from secondary to primary
prevention of cardiovascular disease. Implementation of the programme is phased beginning in
2009 -2010 with full roll out of a uniform and universal programme anticipated by 2012-2013.
Figure 1 gives a diagrammatic overview of the programme approach.

Across the country, many PCTs, particularly in spearhead areas, guided by the National Service
Frameworks for Coronary heart disease, Diabetes and Kidney disease have already developed
local versions of the vascular checks programme – often formalised in Locally Enhanced Service
(LES) agreements. Building on this, the aim of NHS Health Check is to harmonise local efforts
into a single, universal, integrated check for the population.3

  University of Leicester (2008). Handbook for vascular risk assessment, risk reduction and risk
management. The UK National Screening Committee.
  Department of Health (2008). Putting prevention first – vascular checks: risk assessment and
  Department of Health (2008). Vascular Checks: risk assessment and management: „Next Steps‟
Guidance for Primary Care Trusts.

Figure 1: Overview of NHS Health Check

Although it is a preventive programme targeting those at risk, it is anticipated that NHS Health
Check will also pick up some people with previously unidentified established disease who will
then be placed on the appropriate disease registers and care pathways alongside people with
established disease. Guidance from the Department of Health is not prescriptive and the checks
will be able to be delivered by different professionals and in a variety of settings for instance GP
surgeries, pharmacies, community and specialist nurse-led services, allied health professionals,
the voluntary sector and other alternative providers. The service may be provided at PCT,
locality or practice level, the key consideration being a service model that best serves the target
population(s) identified. There are four main stages of the programme – risk assessment, risk
communication, risk management and recall. Figure 2 illustrates these in detail with the
processes involved.

    Varvel M. Putting Prevention First: NHS Health Check. NHS Improvement
Figure 2: NHS Health Check stages and processes


The evaluation was primarily a process one with four main objectives:

      To describe the logistics of delivery highlighting what works well, what does not and why
       (including an appraisal of the method of inviting patients, their journey through the
       system, the communication of risk, the appropriateness of the delivery setting, the
       capacity of providers to deliver, and the impact on other services.

      To gain insights into key patient segments and identify how these can be embedded into
       the programme (building on work already carried out from a pre-pilot social marketing
       research exercise).

      To analyse the potential impact of wider environmental factors on successful

      To develop recommendations from the evaluation findings to guide full roll out of the

2 E VALUATION                    METHODOLOGY

The evaluation methodology was based on a framework integrating different data collection
methods and stakeholder viewpoints, and was characterised by three core features:

         A whole system approach building a multi-stakeholder perspective of the programme
          from patients, GP staff, related healthy lifestyle service providers, and PCT
         A mixed methods research design combining qualitative and quantitative methods to
          collect and analyse information.
         A social marketing element to gain insight into different audience segments in relation to
          programme marketing, communication of risk and intervention mix.

The following methods and analytical techniques were employed to generate intelligence about
the programme:5

    i.    Documentary review – relevant reports on the vascular checks programme were
          critically appraised including: Department of Health implementation guidelines and
          national evaluation plan, NHS Tower Hamlets vascular strategy, service specifications,
          Clinical Effectiveness Group (CEG) clinical guidelines and operating procedures, the
          Opinion Leader pre-pilot social marketing research, and documents and minutes from
          Tower Hamlets high risk CVD steering group and NHS London Vascular Leadership

    ii.    Focus groups and in-depth interviews (see appendix for the interview topic guides).
           - with clinical and administrative staff including health care assistants (HCAs), GP
              leads, practice managers and other support staff. Discussion topics included
              preparation for the service, invitation and assessment process, communication of
              risk, risk management and referrals, changes in workload and ability to cope.
           - with patients – a) those who attended - to understand their experience of being
              contacted and assessed, scheduling of appointments, acceptability and satisfaction
              with the service; and b) those who did not attend – to understand barriers to uptake.
           - with PCT commissioners and service leads – to understand the commissioning
              context, referral process and generate data for the environmental analysis.

    iii. Audience segmentation – aggregation of the practice population into homogeneous
         groups on the basis of one or more common characteristics related to the Health Check
         (such as risk for the health problem, health seeking behaviour, preferred information
         channels, psychographic and/or demographic variables).

  Another key method was the statistical analysis of QRISK data from GP EMIS registers to help a)
describe uptake rates; b) profile the socio-demographic characteristics of users; and c) provide a frame of
reference for interpreting the qualitative information. However, the evaluation team were not ultimately
able to access to the data and were unable to input it into their analysis. The CEG is responsible for
collecting and analysing this data but its findings were not available at the time of reporting.
    iv. Organisational readiness assessment – mapping of the programme delivery sites to
        determine their level of preparedness to deliver the Health Check.

    v. Environmental analysis matrix – scanning of the structural context in which the pilot
       was delivered to identify opportunities and threats to successful delivery.

Thirty eight people were interviewed either individually or in focus groups and represented the
following groups: service users (15), practice and referral service providers (21) and
commissioners (2). Table 1 provides the overall distribution and description of participants.

Table 1. Distribution and description of participants

      Participants                 Number     Summary description

      Patients                        15      Age range: 54 – 75 yrs; Gender: Men - 7, Women
                                              – 8; Ethnicity: White British –10, Bangladeshi – 3,
                                              Others – 2; Non-attenders: 4

      HCAs                            6

      Other practice staff            7       Practice managers, administrators, IT staff

      CVD specialists                 2       CVD nurse specialist, GP lead

      Commissioners                   2       PCT commissioning leads

      Healthy lifestyle services      6       Health trainers manager, Smoking cessation
                                              advisers, Neighbourhood healthy lifestyle
      Total                           38

A focus group was held with the HCAs and in-depth interviews with other practice staff - practice
managers, VRA administrators and lead nurses. Face to face interviews were also held with
healthy lifestyle managers and commissioners. A communal café event for patients from all four
practices was held in August just before the start of the Moslem Ramadan period. The event
was led by a facilitator and interpreters were available for Bangladeshi participants who required
language assistance. Through each of the four practices, an invitation letter was sent to patients
who were screened in the pilot phase of the programme. An accompanying translated version
was sent to those of Bengali background. Invitees were both those that had attended screening
and those that had not and they were selected to reflect different age, sex and ethnic
backgrounds. Owing to poor attendance at the café (only three participants of the anticipated 24
turned up), the data collection strategy switched to direct interviewing at the practice and
telephone interviews for non-attenders.6

 The evaluation period overlapped with the onset of the swine flu epidemic and preparations to tackle the
problem was the priority for practice staff.

Information sheets about the purpose of the evaluation were provided and written or verbal
consent taken from each participant before they took part. All interviews were voice recorded
(where consent was given), and transcribed. Follow-up interviews were carried out with
particular informants to validate the transcriptions and clarify specific areas of interest. Basic
background information was obtained on all patients interviewed consisting of age, sex,
ethnicity, GP practice and response status (i.e. attended/did not attend screening).

The data was analysed thematically. Thematic analysis is a strategy for identifying, coding,
analysing, and reporting patterns (themes) within data. It minimally organizes and describes a
dataset in rich detail. Analysis of the interview data obtained from each participant may be done
independently or across the whole group or defined subsets (cross-case analysis). The latter
approach was considered more appropriate for purposes of the evaluation. Some of the
advantages of thematic analysis include: flexibility, relatively easy and quick to use in
summarising key features of a large body of data, the results are accessible to the lay public,
and it is a useful method for working with participants as collaborators. In addition, it can
generate unanticipated insights and is useful for producing qualitative analyses suited to
informing policy development.


In Tower Hamlets, the delivery of both the pilot and subsequent roll-out is being overseen
internally by a steering group made up of primary and secondary care clinicians, commissioners,
and public health professionals. Preparatory work included: a) social marketing research to
understand the needs of the target audience and their views about the proposed method of
delivery of the scheme; b) service model and clinical guidelines agreed and signed off by the
steering group; c) completion of a 5 day training course for 5 Health Care Assistants (HCAs)
who are responsible for undertaking delivery of the programme; and d) preliminary work on the
programme‟s social marketing strategy7

The programme was delivered from GP practices and targeted people between 40 and 74 years
who were not currently on CHD or Diabetes registers. There were two levels of screening:

       Level 1 (by all GP practices) - the identification and ranking according to risk of patients
        over 40, compilation of up to date contact details and the loading of data post-screening.
        These were preparatory steps for graduation to level 2 at full roll out of the programme.

       Level 2 (piloted in 4 GP practices) - a comprehensive systematic screening service
        (including all the activities described at level 1).

  The work was undertaken by the Tower Hamlets Partnership social marketing unit and details are
contained in the project initiation document Cardiovascular screening: using social marketing to enhance
the scheme.
Tower Hamlets opted for the QRISK2 risk assessment tool rather than the Framingham CHD
risk score recommended by NICE. QRISK2 is based on the national QRESEARCH database of
anonymised primary care patients. The database was developed from 2.3 million patients aged
35-74 with 140000 cardiovascular events from 531 practices in England and Wales. QRISK2
uses the following parameters in its algorithm: patient age (35-74); gender; current smoker
(yes/no); family history of heart disease aged <60 (yes/no); existing treatment with blood
pressure agent (yes/no); postcode (postcode related Townsend score) - an area measure of
deprivation; body mass index (height and weight); systolic blood pressure (use current not pre-
treatment value); total and HDL cholesterol; self assigned ethnicity; rheumatoid arthritis; chronic
kidney disease, type 2 diabetes and atrial fibrillation. Missing values are statistically estimated.

Validation studies indicate that incorporating ethnicity, deprivation, and other clinical conditions
into the QRISK2 algorithm improves the accuracy of identification of those at high risk. QRISK2
is particularly recommended for use in Tower Hamlets because it is (a) a more accurate method
for identifying people who will benefit from treatment, and (b) more equitable in a socially and
ethnically diverse population. A summary comparison of the two tools is presented in Table 1.
Table 1: Advantages and disadvantages of QRISK2 compared to Framingham risk tool.

     Advantages                                       Disadvantages

      Calculated risk is calibrated to the            Although the calculator uses a
       contemporary UK population, so is likely to      complicated algorithm to estimate missing
       provide more appropriate risk estimates to       variables, cholesterol risk behaves more
       help identify high risk patients.                as expected if full values are used.

      Calculated risk is adjusted for additional      Still can't be used legitimately for "what if"
       variables - social deprivation and current       scenarios.
       treatment with antihypertensives.

      It has been validated in the UK using an
       alternative research database.

      QRISK identifies a different high risk group
       of patients than Framingham, with 1 in 10
       patients being reclassified into high or low

  QResearch Database, University of Nottingham.
  Hippisley-Cox J, Coupland C, Vinogradova Y, Robson J, May M, Brindle P (2007). Derivation and
validation of QRISK, a new cardiovascular disease risk score for the United Kingdom: prospective open
cohort study. BMJ; 335(7611):136.
   Hippisley-Cox J, Coupland C, Vinogradova Y, et al (2008). Predicting cardiovascular risk in England and
Wales: prospective derivation and validation of QRISK2. BMJ; 336(7659):1475-82.
   Hippisley-Cox J, Coupland C, Vinogradova Y, et a (2008). Performance of the QRISK cardiovascular
risk prediction algorithm in an independent UK sample of patients from general practice: a validation
study. Heart; 94(1):34-9.
   Colins G, Altman D (2009). An independent external validation and evaluation of QRISK cardiovascular
risk prediction: a prospective open cohort study. BMJ; 339:b2584.
   Although we have compared the Framingham and QRISK2 tools, there are other CVD risk calculators
available including JBS2 and ASSIGN. A detailed comparison of their strengths and weaknesses is
beyond the scope of this exercise, but generally, the evidence shows that Framingham-based calculators
tend to overestimate risk in lower-risk populations and underestimate risk in high-risk populations

Figure 4 illustrates the Heath Check implementation process developed by the Clinical
Effectiveness Group. The starting point is an estimation of the CVD risk of people on GP
registers based on the existing information held therein. Missing data is estimated where
possible and the risk scores are then ranked from highest to lowest. People at highest risk (i.e. ≥
20%) and those with no data are prioritised for assessment. They are contacted by phone and/or
mail with an explanation of the programme and invitation to visit. The following parameters are
collected in the assessment age, gender, social group via postcode, self-reported ethnic group,
first degree family history of premature CHD, smoking status, medical history (on treatment
hypertension; atrial fibrillation, rheumatoid arthritis or CKD), height, weight and BMI, blood
pressure and alcohol consumption

Figure 3: Health Check implementation process in Tower Hamlets

                     Computer prior assessment:
                      - Use QRISK2 batch processing tool to generate a risk
                        score for patients.
                      - Missing data is imputed where possible
                      - Rank patients from highest to lowest CVD risk

                     Start with highest risk (≥ 20%) and those with no data
                     Contact patients for a full CVD assessment
                     (Include publicity/explanatory material)

                     VISIT 1:                                                      SUPPORTING
                     - Explanation                                                   HEALTHY
                     - Full assessment + blood for total/HDL cholesterol            LIFESTYLES
                     - Advise healthy lifestyles
                     - ≥ 20% CVD risk: appoint for 2nd visit

                     ADMIN: Enter total and HDL cholesterol: Calculate CVD
                     and diabetes risk (when this is available).              NHS Midlife LifeCheck

                     All levels of risk should have written information for
                     Healthy Lifestyles and local support services.

    Fast track       VISIT 2: ≥ 20% risk or high diabetes risk                Practice and locality
    co-morbidity     - Inform CVD and diabetes risks and co-morbidity         Network
                     - Arrange fasting bloods: glucose, lipids, eGFR, LFT
                     - Healthy lifestyles advice
                     - Appointment with GP/PN
  Positive family
                     ADMIN: Fasting bloods                                    Programmes for:
  Raised blood
  pressure                                                                     Smoking
                     GP/PN                                                     Diet
                     - ≥ 20% Statin                                            Physical activity
                     - Raised blood pressure                                   Obesity
                     - Increased diabetes risk, pre-diabetes or diabetes       Alcohol
                     - Chronic kidney disease                                  Self-management
                     - Family history CHD+ve
                     - Other co-morbidity

The assessment process specifies two visits but practices may vary in their arrangements for
blood tests and follow-up. With high risk people some practices may take fasting bloods at the
first visit and make an appointment for a second visit. The first visit may be omitted if the person
already had the relevant blood tests done within the last 3 months. In low risk people (< 10%),
one visit is also sufficient to complete the assessment. If a new co-morbidity is identified at the
first visit, the person is managed in the same way as someone with a high risk score (i.e. fasting
bloods and referral to the GP or practice nurse). The definitive CVD risk is calculated once the
results of the lipids are available (people with <10% CVD risk do not need to have this test as it
makes little difference to their risk score). In all cases, regardless of risk, written information on
healthy lifestyle and local services is given.


Preparation for implementation of the Health Check included the commissioning of social
marketing research with local people. This involved a series of workshops (focus groups) and
face-to-face depth interviews with 54 participants around four themes: a) current views and
experiences of healthcare; b) information received about staying healthy; c) understanding of
screening and the link between healthy lifestyle and reduced risk of diabetes and CVD; and d)
designing an ideal programme (“what we want”). While the research report did not specifically
identify key audience segments, it contained sufficient information for us to construct broad
segments that could be further built on and refined in the course of the evaluation. Important
insights from the research are summarized in Tables 3 and 4 (a more detailed rapid review is
contained in the appendix).

Table 3. Preliminary audience segments and key insights from the pre-pilot social marketing

     Audience segments               Key insights

      1. Enthused uninformed              Low awareness about CVD/diabetes particularly
                                           among BME people
      2. Risk denial
                                          Resistance to behaviour change especially among
      3. Change resistant                  older people

      4. NHS/GP averse                    Varied mix of advertising methods to inform and
                                           educate people

                                          Need for the programme to have follow through and
                                           not one-off media/ advertising campaigning

                                          Preference for mixed settings – community and GP
                                           (but primarily GP based)

                                          Where GP based, the Health Check should be a stand
                                           alone service and not seen as part of routine health
                                           services (because of the association with illness)

Table 4. Summary of insights from the pre-pilot social marketing research

Barriers                Health care practitioners      Knowledge /Awareness                                 Vascular screening

Language                                                                                                      Was viewed favourably and majority of participants
                         Participants‟ wanted HCPs     Knowledge / awareness about screening
 Somali participants                                                                                          stated that they would attend the screening
                          to be more proactive in        services, CVD/diabetes prevalence in Tower
  expressed that          advocating health related      Hamlets and health interventions available
  there was a lack of
                          messages and screening         was low. Knowledge /awareness was                    Wanted the service to be easy-to-use, quick and
  information on                                                                                               informal. Also wanted out of hours services
                          services available             particularly low in ethnic minorities
  health and
  wellbeing tailored                                    Participants wanted a variety of advertising         Preferred stand-alone service that is not associated
                         GPs are viewed as being        campaigns to provide education on health
  for their community                                                                                          with existing local health services as attending GP
                          overloaded with work,          and screening:
                          hence providing a                                                                    surgery/hospitals was associated with feeling ill
Behaviour change                                          -   local media
                          screening service within
 Older participants                                      -   minority press                                  Participants wanted letter of invitation from G.P
                          the GP practice would
  were less                                               -   referral from GP                                 (clearly stating the screening process to facilitate an
                          increase their workload
  supportive of the                                       -   door-to-door leaflet                             informed decision)
  screening as they                                       -   adverts in public places
                         Some patients had bad
  were unsure if they                                     -    leaflets, posters and brochure               Screening venue
                          experience with HCPs in
  would be ready for                                          (preferred when the message being               Some participants wanted the screening to be
                          the past; hence avoiding
  behavior change                                             delivered is concise, practically inclined,      available at the GP surgery.
                          using the NHS unless
  /modification                                               easy to read, eye catching and full of          Some other participants preferred community centres
                          absolutely necessary
                                                              imagery)                                         and mobile units
                         Participants viewed HCPs                                                            Participants did not want screening to be conducted
                          as extremely busy to          Partnership work with community and                   in pharmacies as they were doubtful of the medical
                          explain results and            voluntary sectors could be crucial to                 expertise of the staffs at the pharmacy
                          treatment options              educate / enlighten ethnic minority patients
                          adequately                     on health and wellbeing                            Screening staff
                                                                                                              Participants wanted staff conducting the screening to
                         Participants wanted           Most common sources of health information             be bright, welcoming and friendly
                          personally focused support     cited by participants were TV, radio,                Bangladeshi participants preferred screening to be
                          and practical advice to        newspaper, leaflets and posters seen at the           conducted by same sex staff
                          facilitate behaviour           GP surgery                                           Participants wanted bad news about screening to be
                          modification                                                                         delivered by GP or any other HCP who is able to
                                                        Participants were keen on local workshops             allocate time to explain the results adequately.
                                                         on health. Hence health workshops could be           Participants were willing to work with suitably trained /
                                                         organized to educate the community on                 qualified persons.
                                                         health and wellbeing.
5 P ROGRAMME                     ANALYSIS

Although the practice sites operated in fairly similar contexts – e.g. life expectancy for men and
women, and the top vascular conditions seen in GP practices (i.e. hypertension, diabetes and
ischaemic heart disease) were similar - there were aspects where they differed, for instance, the
relative proportion of non-white populations (lowest for St Stephen‟s practice) and CVD and
stroke mortality rates (highest for Mission practice). Table 5 below highlights selected indicators
of the practice environment.

Table 5. Practice context

                                               Chrisp        Jubilee       Mission    St Stephens
                                               Street         Street
     Local Area Partnership                    LAP 7         LAP 4         LAP 1         LAP 5
     Location                                 Southeast    Southwest      Northwest     Northeast
     LAP population                            36,619       28,011         42,951        24,873

   Ethnicity (% of total popn)
     White                                      41.9           51.0          47.8          59.7
     Bangladeshi                                33.3           32.6          32.4          20.0
     Other                                      24.8           16.4          19.8          20.3

   IMD (2007)                                   52.0           39.9          45.5          43.0

   Life expectancy (yrs): 2002-06 pooled
      Male                                      73.5           73.4          74.2          73.2
      Female                                    78.7           81.2          80.6          79.8

   Mortality (per 100,000): 2002-07 pooled
    CVD                                         62.5          124.2         218.5           -
    Stroke                                      14.2           24           59.1            -

   Practice population                         11,407        10,136         9770         10,790

   CVD risk % of practice population             34            35             35            37

   Top vascular conditions seen in all GP
   practices in LAP (as % of CVD-related
         Hypertension                          51.4           51.8          51.1          54.5
         Diabetes                              26.5           27.0          26.8          24.4
         Ischaemic Heart Disease               12.8           12.6          12.8          12.3
         Stroke                                 5.8            5.3           6.1           5.4
         Heart Failure                          3.4            3.3           3.2           3.4

Sources: LAP Profiles: Update May 2009, LDP data 2009, Annual Public Health Report 2008-09, CEG
baseline data.

The QRISK2 batch processing tool was used to identify patients on practice registers who were
at risk of vascular disease. Each practice then sent invitations for assessment to the top 100
patients with CVD risk ≥ 20% on its list in accordance with the PCT‟s strategic decision to
prosecute targeted rather than general screening. Practices used either Trust or self-designed
invitation letters. At the first visit, patients had their vital signs (blood pressure, weight, height)
measured and blood test (total and HDL cholesterol) done. On obtaining the test results, their

risk score was calculated and they were invited for a second visit where they were informed of
their risk and subsequently referred to the GP or signposted to appropriate healthy lifestyle
services. Practices took slightly different approaches to the vascular screening to suit their


The QRISK2 search and list of patients was generated by the GP VRA lead. The practice initially
had only one HCA who handled the contacting and assessment aspects of the screening. Owing
to the workload, the HCA was unable to follow up invitation letters to patients with phone calls,
but the overall response rate was reportedly good. An extra HCA was recruited in the later
stages. After receiving invitation letters, patients then booked an appointment with the HCA for
the screening, which was usually on a Wednesday afternoon. During this time, the blood test
was done and the Health Check explained. When the blood test results were obtained, patients
were invited to the practice once again to come and collect their results.


After an initial poor response to letters of invitation the practice adopted a more tailored
approach. Two HCAs were involved in screening and were supported by administrators who
handled the initial risk screening and sent out invitation letters. Reminder calls were made a day
prior to appointments. Blood tests were arranged for convenient times and once the results were
available, an appointment was set up with the HCA where their exact risk was explained and
they were signposted either to the GP or other health improvement services. Patients who failed
to turn up after the first letter of invitation were sent the same letter a further two times. Although
the HCA team was not optimally supported by the allocated nurse mentor,14 they were
experienced (both HCAs had medical qualifications) and worked well with the PCT CVD nurse
specialist, and this, together with GPs who had a clear picture of the screening programme,
contributed to the smooth running of the Check. The HCAs screened approximately 20 patients
a week. Practice meetings were frequently held to discuss what worked well and what
improvements needed to be made to the screening.


The programme was well organised. Patients between 35-74 years old were targeted, a wider
age band compared to other practices. One HCA handled the vascular screening, although
another was later recruited. Uptake by patients was reported to be good at first but later
dropped. Uptake by ethnic minority patients however, was particularly low. The HCA was
supervised by a lead GP and also supported by other practice staff. Administrative and IT
support for the HCAs was very good. Patients received an invitation letter, followed by a phone
call to book them for VRA screening. During the initial assessment, bloods were taken where
indicated and if at that stage the HCA felt there was need to have direct medical intervention, the
patient was booked for an appointment with the GP. Screening took place four days a week and
approximately 6-7 patients were screened in a day.


The practice had a well organised programme with two HCAs involved in the vascular screening,
one of whom also handled the administrative work. Their workload was considered manageable.

  The nurse mentor supervises and signs off the HCAs competencies after which they are allowed to
practice on their own.

Patients were invited to take part by letter followed by a phone call to schedule an appointment
for the screening. A QRISK2 assessment allowed the HCAs to invite the highest scoring patients
first and then work their way down the list. At screening, the HCA went through the template with
the patient; height and weight measurements were taken, nutritional and exercise advice offered
and referrals made where necessary. One of the clinics ran on Wednesday late in the afternoon
when blood tests could not be conducted so some patients needed to re-book appointments to
do blood tests. When test results were available, the patients were called up by the HCA and the
results discussed. If necessary, an appointment with the GP for follow up was arranged.

To determine how prepared the practices were to deliver the Health Check we assessed the
extent to which they provided clinical and IT infrastructure/administrative support for the HCAs.
These two areas were operationally the most mission critical and represented the overarching
dimensions for a cluster of factors, namely:

                     existing clinical guidelines and standard operating procedures
                     mentoring support for HCAs
                     HCA working relationship with the GP lead and other clinical staff
                     existing administrative framework
                     availability of administrative staff to manage the invitation process (or in lieu of that,
                      administration time allocated for the HCA)
                     the integration of the QRISK2 tool with practice systems

Collectively, these determined the seamlessness of the patient journey from invitation through to
assessment, risk communication and subsequent referral. The assessment was done
qualitatively by collating and transforming interview comments and responses from informants
into an aggregated score based on a 5 point scale. Figure 4 maps each practice against these
two areas.

Figure 4. Organisational readiness for the Health Check

                       5                                 C


                                                                                                 Jubilee Street
                                                                                                 Practice (J)
   Clinical support

                                                                                                 Practice (M)

                       1                                        J
                                                                                                 St. Stephens
                                                                                                 Health Centre (S)

                                      1             2           3            4            5      Chrisp Street
                                     IT infrastructure/admin support                             Practice (C)

An organisation readiness assessment looks at different aspects of an organisation and
evaluates how they are planned or handled to deliver an operation. It gives a sense of what it will
take to implement the operation, identifies potential problem areas and helps shape the initial
direction for change. While the map may be used to make comparisons between practices, that
is not its primary purpose as the practices had differences (albeit minor) in their approaches and
served different populations. Rather the primary intention is to provide an organisational context
within which programme barriers and facilitators and process and outcome measures (such as
uptake rates, user satisfaction, etc) might be explained at the individual practice level.

Ten main themes were identified from the interviews and focus group held with practice staff
around the delivery of the Health Check: i) targeting; ii) awareness; iii) behavioural change; iv)
BME engagment; v) support; vi) workload; vii) referral monitoring; viii) HCA experience and
training; ix) coding; and x) success indicators. Each of the themes is evidenced and discussed in
detail below.

There were two subordinate themes related to inviting people for screening a) how to invite
(targeted versus general approach) and b) who to invite. Support was expressed for the
programme‟s strategic approach to target the highest risk people, even though they are harder
to reach, than screen opportunistically. It was felt that while in the short term targeting might not
generate large numbers of people being screened, it would ultimately have more impact in
reducing CVD related outcomes. The issue of who to invite was more contentious and captured
in the comment of one informant.
    There are two issues really here, one, whether somebody should have a check or should
    they be excluded?

Patients who were already on statins for instance were observed to be among those screened
although the programme was not intended to include them as they were already on clearly
established care pathways. There was no consistent position by the practices on how to handle
this or clear understanding of the logic behind national guidelines that exclude people with
diabetes, CHD or stroke but include people with atrial fibrillation or hypertension. People with
co-morbidities thought to constitute around 25% of all those at high risk. They therefore
represent an important demographic group.

   QRISK includes people that have comorbidities. They have atrial fibrillation, rheumatoid
   arthritis, and if you include them, people with diabetes……..It may well be that these
   people are already being seen in continuing treatment programmes. So if you’ve got
   atrial fibrillation for example, you will already probably be on warfarin, in which case it
   might be better to look at your list ………. …so there is quite a big discussion about who
   shouldn’t be called. Should people already on statins, should they be called? Should
   people already on hypertension treatment programmes, should they be called? So
   practices, you know, needed quite a lot of discussion really about who they wanted to
   call, and who would be useful to call. And I think that was one of the, sort of, learning
   points from the pilot that needed to be considered by practices that we hadn’t fully
   addressed beforehand.

Almost all the staff observed that patient awareness of the NHS Health Check was low, exposing
the need for marketing/advertising measures to be scaled up to enhance awareness and
increase screening uptake.

   .. not a lot of people are responding to it. I don’t know why, I don’t know.. when you
   explain it to them they think oh you going to tell me that I’m gonna die very soon. We are
   trying to explain to them very clearly but still…... we’ve been sending out letters to them
   and everything and calling patients to invite them, but still no response; its quite low.

The need for sustained marketing to reinforce the initial messages communicated at invitation
and during assessment was also stressed. Earlier campaigns for smoking cessation and 5 a day
were also mentioned as examples of good marketing practice worth emulating.

   It's about education, it's about messaging, it's about media push. All of that, education,
   information, reinforcement, works very well.

   I’ve seen the Tower Hamlet's 5 a day advert, I think we need something like that.

   Do more posters, billboards, and TV adverts, I think that would be good. For example
   with smoking, because of the advert on the television, most patients come in for smoking


It was recognised that engaging individuals and motivating diet and lifestyle changes was much
more than a matter of letting them become aware of the availability and benefits of the service.
The particular challenge of shifting behaviour when a person felt well and had no physical
symptoms of ill health, and therefore no compelling reason to take action, was also highlighted.
This pattern of behaviour was widespread and not restricted to any particular ethnic group.

   Most of the health professionals do think that this is a useful programme that will benefit
   high risk people, and so I hope that that will convince people to do it …..but it remains to
   be seen how many people will take this up. Conversely, these are well people that
   actually have nothing wrong with them by and large and they find it difficult to
   understand, why if their cholesterol is not particularly high.


The view was expressed by some that ethnic minority patients were less enthusiastic about
having the Health Check than their White counterparts. While sociocultural influences were
broadly held responsible for them not understanding the importance of screening or prioritising it
over other competing demands, specific factors could not be pinpointed. However, language
barrier was not perceived as a problem as most non-English speaking patients attended with
relatives to act as interpreters. Practices also had patient advocates available to provide
translation assistance.

   Everyone I’ve seen is White British, I think I’ve seen about three Bengali people, one
   Chinese, (and) one Turkish person.

   The barrier is the Bengali patients; some have engagements somewhere else, so they
   always make excuses that they are fit and well and they don't need to come………I won't
   say its culture but they are just used to one lifestyle and you explain to them, you try and
   it's not working.

Support was a recurrent theme in many interviews. Enhanced support within the practice was
perceived to have a positive impact on the delivery of the Health Check. Two main dimensions
of support were identified - clinical and IT infrastructure/administrative (see section 5.2 above).
In practices where HCAs were well supported, they were able to fully concentrate on screening
patients and completing the VRA paperwork. Where administrative support was deficient, the
HCA struggled to juggle clinical and administrative workloads. The latter usually was sacrificed
which meant that the HCA did not have the time to properly monitor or follow up patients who did
not respond to the letters of invitation.
Strongly positive feedback on the supportive role of the CVD nurse specialist and some practice
nurses was expressed:
   The support from the practice is very good , my line manager , she is the head nurse,
   she supports me and I've got Elenora at the end of the line, so if I have any problems I
   can ring Elenora. Therefore, I think the support is good.

   I had support from Elenora who I was calling very often at the beginning. My practice
   nurse who is my line manager, she is very supportive, I know that I can go to her about
   anything. She is new; she just started as we finished the training. Therefore, she does
   not know a lot, she tends to ask me more rather than I can ask her. She is a great
   support in a personal way. However, I had no support from anyone else.

In three of the practices, a further HCA or administrator was recruited, an indication of the extra
workload generated by the screening. A practice manager observed that undertaking the pilot
had highlighted some shortage in workforce within the practice. All staff acknowledged that the
screening had implications for the workload of GPs, but overall, the load was manageable.

   It's motivated us now. At that time we only had one phlebotomist, because we were doing
   this and we were seeing how HCA and phlebotomy team tie up, I've now got another
   member of the team to go off and do the phlebotomy course, and now we have
   employed a third phlebotomist. So, it's been useful in that sense, because it's highlighted
   an area of potential weakness.

   The workload was's half an hour appointment, so it gave me less time for
   all the things I needed to do, but we have a new HCA now, so it's been managed by that.


All practices had in-house smoking cessation teams to which they could refer patients after
assessment if necessary. Referrals were also made to external healthy lifestyle services
including the Health Trainers, Young @ Heart, PCT smoking cessation team, GP exercise on
referral, and the Drug and Alcohol Service for London (DASL). The Health Trainers service
received the bulk of referrals. However, a proper system for tracking referrals from the practices
to these services was lacking. While practices had records of the patients they referred, they are
unable to tell whether the patients actually attended the referral. At the other end of the process,
the receiving services did not hold information on referred patients nor provide feedback to the
practices. Some HCAs felt that they were too busy to keep track of patients‟ transition/progress.

   I refer patients to the Health Trainers, but I do not monitor if they went there. It will be
   good to do it, but it’s difficult as its extra work and I don’t have the time to do it


The HCA posts were largely filled by people already working in the practice in some capacity
(receptionist, clerical). This was helpful (and perhaps inevitable) because it enabled the practice
fulfilled the requirement by the PCT that people nominated for the vascular training course
should already have some level of HCA training. Using existing practice staff also meant that the
HCAs were familiar with practice systems and standard operating procedures (SOPs) right from
the start, obviating the need for lengthy induction.

There was general agreement that the training provided for the HCAs was excellent. However,
some HCAs felt that the three days training was intense and would prefer it be delivered over
more days. Furthermore a desire for top-up training was expressed.

   Positive things (about the programme) are excellent training. We have shortened the
   training to 3 days now. I think Tower Hamlet was very proactive in starting with this is
   January. Training people and getting this on board, I think that was a really positive thing

   It is an intensive course, you are taught many things, and unfortunately, you don’t learn
   everything. You can only learn one thing at the time, so I think the training could be
   prolonged for a few more days.

Although they were provided with motivational interview trainings, some HCAs felt insufficiently
equipped to persuade patients to make behaviour modifications.

   A lot of the patients are very reluctant you know. People who are overweight, they know
   they are overweight and know they should do something about it but you know people
   don't for whatever reasons, so I feel that you can't really push them

Concern was expressed that it would be difficult to retain the trained HCAs because a number of
them were „over qualified‟ for the job and would quickly become dissatisfied with the level of
remuneration and responsibility and move on to other jobs. One informant felt that the person
specifications for entry into HCA training were set too high and precluded more practice staff
from applying.

   Most practices have staff who would like to take on broader roles and for a receptionist,
   for instance, moving to a HCA role will be seen as a move upwards. But the obstacle is
   that Tower Hamlets requires that anyone taking up the role should already have (some
   level of) HCA training.

In primary care as a whole, there are known to be large variations in the accuracy and
completeness of the clinical information stored in electronic patient records. Generally, the
evidence indicates that prescriptions have the highest rate of recording while the recording of
diseases (i.e. diagnoses) varies, with completeness generally highest for diseases with clear
diagnostic criteria. In comparison, lifestyle and socio-economic data have lower rates of
recording. It was not unexpected therefore to find that the NHS Health Check was not exempt
from coding problems. Even though it was understood that accurate coding of monthly activity
data was important both for proper analysis of collected data and reimbursements to the
practices, the integrity of the coding process was questioned by informants.

   People were confused, and I think still remain confused, about some of the coding used
   on our templates for the administration of the enhanced service. So, it’s not enough to do
   the check, you have to save the check with the code.

The potential for coding errors was identified at several levels – completeness of entries,
accuracy of entries and use of the correct templates. Mistakes were thought to be more likely to
occur where HCAs had to combine clinical and administrative roles. Continuous training for
practice staff involved in the process was seen as the solution to the problem.


While informants saw the programme as successful, there was uncertainty about the most
appropriate way of measuring this. Achieving the target of screening 100 patients per practice
over the 3 month pilot, while important, did not appear to be seen as the major measure of
success and the view was expressed that the programme needed to have clearly defined
indicators that both the PCT and practices were agreed on.

   The question I think we still need to answer is what are the measures of success? What
   is success? Is it the number of people who’ve been counted irrespective of their risk? In
   which case a PCT having ten thousand people at low risk is going to be seen as being
   more successful than a PCT having a thousand people at high risk………I think the
   whole point of this is to identify people at more than 20% risk and that needs to be a kind
   of measure of success

The uncertainty also seemed to be linked to lingering questions around the impact of targeting.
For implementation, the government‟s cost effectiveness modelling is based on the assumption
of a 75% uptake and this may perhaps represent one of the metrics for practices.

Table 6 summarises the key factors that enabled or constrained successful delivery of the
programme emerging from the information provided by practice staff and other providers. It is
not practice specific but synthesises common issues across all practices. Brief remarks are
provided (where available) to explain the effect of, (and/or mechanism underlying) the factor. A
factor could work well in certain aspects but not in others in which case it would act as both an
enabler and a barrier. But for the purpose of clarity, the factors are identified only under their
predominant effect.

Table 6. Key programme enablers and barriers

(a) What worked well?          Why?
1. Clinical and                 Provide structure and guidance; particularly useful for a new
    administrative (LES)         service where roles and processes are still evolving
    frameworks                  Flexibility within the guidelines for individual practices to modify
                                 the CEG guidance to fit their circumstances

2. Functional IT                QRISK2 tool is appropriate for local population, practical to use
   infrastructure                and integrated with the GP EMIS systems

3. HCAs                         HCAs well trained and confident; able to call on specialist support;
                                 experienced and already familiar with practice systems/protocols

4. Spreadsheet of action        Easier/quicker to distinguish b/w different categories of patients
                                 such as attenders, DNAs and people already on statins

5. Pre-appointment for          Time saving, minimized the number of visits

6. Communication of risk        Availability of advocate services for patients
                                Patients keen to know about their health

7. Call up (letter + phone)     Patients appreciate being contacted; improves uptake

8. Teamwork                     All members of the team working together to provide effective

9. Drug management              Referral and uptake of medication(e.g. statins) to reduce risk

10. Targeting                   Prioritised people at highest risk

11. Screening process itself    Opportunity to reinforce message about risk and what can be
                                 done to reduce it
                                Patients keen to know about their cholesterol /glucose

(b) What didn’t work well?     Why?
1. Search and audit             Included people already on statins

2. Uptake                       Inappropriate or non-sustained publicity/marketing resulting in low
                                 public awareness, message not reaching intended targets.
                                Poor follow-up of invitation letters by telephone (due to lack of
                                 admin time)
                                Resistance to behavioural change

3. Workload                     The Health Check had generated a new admin role and the
                                 workload of HCAs was high

4. Coding                         More room for coding errors where admin support was lacking

5. „Internal‟ support for       Inadequate admin support
   HCAs                         Inadequate mentoring of HCAs
                                Low awareness by other practice staff about the screening
6. Referral monitoring          Practice unable to ascertain whether patients actual attended the
                                 healthy lifestyle referral
                                Non-integration of primary and community data systems

7. Recruitment and              High risk of HCA churn as some are over qualified and will be
   retention of HCAs             attracted to better paying positions and roles elsewhere.

6 S ERVICE             USERS

Insights into the characteristics of service users were generated from the communal café event,
direct interviewing at the practice and telephone interviews. The main thrust of the interviews
was to understand the beliefs, attitudes and behaviours that might explain why people attended
or (more importantly) failed to attend the Health Check and what could be done to make it easier
for them to use the service. Four main themes were identified within the data analysed: (i)
miscommunication; (ii) information avoidance; (iii) resistance to change; iv) patient-GP
relationship. Each of the themes is evidenced and discussed below.

This theme consisted of three sub-themes – brand ambiguity, inadequate communication and
staff insensitivity. An early observation during the process of contacting service users for
interviews was their confusion about the name of the programme. Many associated the term
„Health Check‟ with a visit to their GP surgery for a variety of reasons (e.g. cholesterol, blood
pressure or weight measurement) and not with a specific service. This was further evidenced by
the fact that despite prior written information being given to them and further explanations on the
day, some patients who participated in the café event or telephone interviews were not quite
sure what aspect of their healthcare service they had been invited to discuss. As one participant

   I really don’t have a clue what I am supposed to be here for.

As a result some responses related to their experiences of accessing care in general and not the
Health Check specifically; for instance:

   I never get the results from my test, I went for my ECG and I heard nothing. I phoned two
   weeks later to obtained my results and they would not give it to me over the phone , they
   booked me an appointment , the next week , because that was the only time the doctor
   was free .

For those who knew about the Health Check, the view was expressed that people were very
much interested in their health and wellbeing but there was not much awareness within the
community on the importance of screening. One reason may be that the message was not
reaching the intended audience. As there were instances mentioned of non-English speaking
patients being sent letters written in English and people not receiving invitation letters at all.

Some patients felt that the information contained in the letters they received from the practice
was not sufficient. They further expressed that clearer communication could motivate patients to
attend screening and other services. There was agreement that any test results needed to be
communicated to patients even if no red flags were raised by the test.

   I did not get much from the letter I received; I would have definitely attended if more
   information was given.

   I suppose if it was explained properly, yeah, some people might (take it up).

Another dimension of communication had to do with practice staff. There was a desire for
communication between staff and patients to be more precise and informative and handled in a
less insensitive manner.

   I was with my nurse and telling her about my health problems and she says, ‘maybe you
   got cancer’. That’s not a very nice thing to say...

   Before you start talking, they start writing and book me in for another appointment, next
   appointment for what…… you haven’t even listened to my problem.

In response to how health messages could be best communicated to people, a respondent
suggested that the presentation should not be ‘gory’ but ‘nice and firm’.


The literature recognises that people may avoid, ignore or deny information that might have
adverse implications for their health. This characteristic was evident as some patients expressed
that their decision not to attend the Health Check was because they would rather not know if
anything was wrong with them.

   I would not go to any of these health check ups because of the fear factor; I’d rather not
   know what is wrong with me. I want to buy time and die indoors

   I did not want to attend the test because I was scared that they will tell me something
   was wrong, but my daughter forced me to go and dragged me to the practice for the test

People avoid information for different reasons – it may provoke unpleasant emotions, force a
change of beliefs, or compel them to take undesired action. The responses from patients
indicated that the fear of being told bad news seemed to be the main reason in this case. They
were afraid of opening up a can of worms and for them rather than knowledge being power,
ignorance was bliss.


The NHS Health Check aims to get people to adopt healthier lifestyles and the HCAs who
conduct the screening are trained in motivational techniques to encourage behaviour change.
This is vital because an important theme that emerged was that of resistance to behaviour
change. But even among patients who were aware of the programme and were well informed
about the benefits of having their health assessed there was still reluctance to make the
necessary lifestyle adjustments. An individual who was supportive of the programme and had
attended screening was found to have a high cholesterol level illustrates this disinclination.
When asked what she planned to do about the result, she responded:

   Nothing. I would not worry about it; if I worry it’s not good for my health.

Part of the issues around resistance to change were linked to the introduction of new
technology in the GP surgeries that some patients had found difficult to adjust to.

   How do we receive information? I think me and you, we were both born in the wrong era.
   We are not into technology. No computers and things like that. We rely on letters and
   what we read in the paper.

   I mean I don’t even own a mobile phone, and I would never want a mobile phone. And
   I’ve got no like internet thing; nothing like that…I know how to work the telly, and the
   video, and all like that, and I like to play me music; I’m a music person.


Patients‟ relationship with their GP influenced their willingness to take up the NHS Health Check.
The relationship was shaped by their perception of the competence of the GP and other practice
staff, length of the consultation period, influence of new technology and past experience with the
NHS. Participants viewed GPs as qualified and competent professionals but felt that they did not
have as much time for them as they did in the past. Being extremely busy meant that the GP,
although competent, could not properly address the patient‟s concerns.

   Long time ago, my doctors knew me and my condition; now they only know me through
   my records and sometimes I have to keep reminding them about my conditions.

   (There is) a lack of patience on most GPs side. Unfortunately, all the good GPs round
   Bethnal Green where I’ve lived now for 38 years, they died; but the younger ones now,
   they have very little patience. I’m not saying all of them.

Some patients were willing to be attended to by other health care professionals (including the
HCAs) who could devote more time for them, and there were instances of positive experiences.

   They (HCAs) always keep an eye on me, always concerned, so that’s really nice

However, their perception of the competence of these staff was influenced by their perception of
the GP.

   If it doesn’t start with the GP, what can you expect of the people under?

There were those who expressed doubts about the competence of non-GP staff to deliver their
test results and address their concerns. In their view, the GP was the only, or best, qualified
person to do so.

   If you know someone is not a qualified person, you will not take what they say seriously,
   as you will when you hear it from the GP who is definitely more qualified and knows more
   as he is the one who referred me to the test in the first place

The introduction of IT systems and other technology had impacted on both patients and health
care professionals.

   As I said because of the new systems and what not, maybe that’s why they are losing
   interest. Mind you, you know, I’m not saying it is an individual fault of the doctors. They
   are there for hours to do what? For one quarter good result? Just because they have to
   see so many people………You might go to your GP. Tomorrow, I go back there…em,
   what was it again (name of patient)…em, they start tapping into stupid computer. What’s
   wrong with just talking?

   Well, I don’t know whose idea it was, but they want to scrap it (new technology) all.
   Because before, if you phone, ring a GP, you mention your name, he doesn’t have all
   this gadgetry and yet he knows exactly who it is, home visits, and all that. You don’t get
   none of that now.
   You see different doctors every time, so is it a wonder they don’t know you know your

   The experience I’ve had when I’ve spoken with a doctor is that while I’ve been speaking
   to him, his mobile phone’s gone off and he’s gone out of the surgery while talking to me
   to answer the phone. I shouldn’t think they would let the phone be switched on at that
   point when they are addressing people; but what can I say?.... I think it’s a bit rude but
   …..then he comes back and asks what’s wrong again.

Finally, some patients cited negative experiences with using health services in the past which
were not satisfactorily resolved. It prejudiced their views and made them reluctant to take up
current services including the Health Check.

The behavioural characteristics of the patients were analysed to construct distinct segments on
the basis of a) where patients were in relation to adoption and maintenance of the desired
behaviour and behavioural goals of the NHS Health Check; b) expressed barriers to uptake.
Figure 5 illustrates the main audience segments that were identified together with their key
descriptive features and the specific influences that shaped them.

Figure 5. NHS Health Check patient segments

        Patient          Key descriptors                               Drivers

                                                    Active interest in staying healthy
                                                    Positive experience with health services
                            informed,               „Peer‟ champions of the programme
                            satisfied               Cultural connection – access to health care
                                                     professionals with similar cultural identity (e.g.
                                                     gender, language, ethnicity or religion)

                                                 Unaware of but interested in learning more
                                                  about health and opportunities for healthy living
     Enthused                Motivated,
                            uninformed           Recipients of miscommunication - health
     confused                                     messages inappropriate or not reaching them

                                                 Passive health interest and so dependent on
                                                  persuasion (from relations, practice staff, etc)
     Passive                Unhappy,
                            uninformed           Unaware of opportunities for healthy living
                                                 Indifferent/neutral experience with services

                                                 Avoid information for fear of bad news
                                                 „Ignorance is bliss‟/„Not in my backyard‟
     Risk denial           information
                                                  (NIMBY) mentality

                                                 Poor experience in past with health services
                                                 Communication issues with practice staff
     GP averse             Unengaged,
                           technophobic          „Excluded‟ by IT innovation in health/generally

7 E NVIRONMENTAL                     ANALYSIS

The primary information from the service users and providers was supplemented with local
documentary and statistical evidence to build a picture of the wider context in which the NHS
Health Check sits. Five key environmental domains (and their sub-domains) were identified as
influencing or having the potential to influence the programme. These were:

   i. Socio-demographic - comprising population size/structure, ethnicity and deprivation.
   ii. Epidemiologic – comprising CVD and Diabetes prevalence and risk factors for CVD (i.e.
        smoking, exercise and obesity).
   iii. Regulatory – comprising clinical and administrative frameworks, and the method of risk
   iv. Capacity – comprising uptake rates and the capacity of GP and healthy lifestyle services
        to meet demand.
   v. Technological – comprising marketing and communication, data capture and collection,
        and production of manpower (HCAs).

The domains were assessed from the perspective of whether they constituted facilitators and
threats to successful delivery of the programme. The assessment derived from participants‟
responses given at interviews. The findings are presented in Table 7 below in the form of an
environmental analysis matrix indicating the nature of each domain‟s effect and likelihood of
impact. In a few instances, some domains had no discernable impact or their effect could not be
ascertained and these were categorised as „uncertain/neutral‟. With further intelligence, they
may be more definitively classified.

High impact contextual factors likely to facilitate successful implementation were judged to be
the development of clear clinical and administrative frameworks and the choice of a contextually
relevant CVD risk scoring method – the QRISK2 tool.

High impact contextual factors likely to constraint programme delivery were judged to be the
high level of deprivation in the borough, low uptake rates particularly among BME patients, and
an marketing and communications strategy not quite fit for purpose.

Table 7. Environmental analysis matrix

Environmental domain           Summary description                                                                                 Overall         Likelihood
                                                                                                                                   effect          of impact
  Population size/structure   Some 30% of the population is over 40, and ~20,000 residents are aged 65+.                      Uncertain/neutral        -
  Ethnicity                   Half of the population are from BME communities and the borough is home to Britain‟s               Impeder          Medium/High
                               largest Bangladeshi community.
    Deprivation               Tower Hamlets is the 3 most deprived local authority in England. The level of deprivation           Impeder            High
                               is reflected in higher than average mortality. Social and economic deprivation is strongly
                               associated with higher smoking rates, poorer diets, lower levels of physical activity, and
                               higher rates of alcohol/substance misuse. Ethnicity (people of South Asian and African
                               Caribbean origin) and social disadvantage are known factors in the development of
                               vascular disease.

  CVD                         CVD is the number one cause of death in the country. While mortality rates of CVD are               Impeder           Medium
                               declining nationally, TH still has one of the highest rates of CVD, stroke and smoking-
                               related deaths, and a low life expectancy particularly among its Bangladeshi population.
    Diabetes                  Higher rates than elsewhere and rising. Currently ~11,000 known cases and another                   Impeder           Medium
                               estimated 1-2,000 undiagnosed. Prevalence is higher in South Asians

    Risky behaviour           TH has among the highest prevalence of risk factors for CVD in London
     - Smoking                 In 2004 an estimated 37% of the adult population (16+) were smokers compared to an                  Impeder           Medium
                               average of 27% in England. Bangladeshi men have the highest smoking rates of all ethnic
                               groups (40%). But despite successes with smoking cessation, shifting behaviours will take
     -   Physical activity     Self-reported sporting activity for a minimum of 30 minutes 3 times per week is lower in TH         Impeder           Medium
                               compared to London and England and particularly low in women and BME groups
     -   Obesity               Around 20% of adults in TH are estimated to be obese (equivalent to 40,000 people)                  Impeder           Medium

  Clinical framework          Programme implementation is informed by locally developed (CEG) clinical guidelines. The            Enabler            High
                               programme also has designated CVD clinical leads whop have both accountable and HCA
                               mentoring roles.
    Administrative            Programme implementation is organised within LES agreement between the PCT and                      Enabler            High
     framework                 participating practices
    Risk assessment           The use of a different risk assessment system (QRISK vs Framingham) will have                   Uncertain/neutral      Low
     method                    implications for national comparability down the line. Possibility of national enforcement of
                               a common scoring tool.

  Uptake                  Although initial reports were of high uptake with over 50% of those invited attending the         Impeder          High
                           assessment, uptake subsequently reportedly fell particularly among BME patients. This
                           echoes evidence from lifestyle change programmes such as smoking cessation which
                           have shown a social gradient in uptake. For implementation, the DH‟s own cost-effective
                           modelling evaluations made a number of assumptions including a 75% uptake of the
                           programme. However, many urban areas have had uptake rates of less than 30% for
                           similar programmes.
    GP service capacity   In a practice of 10,000 patients there will be 2500 people aged 40-74 years. In an average        Enabler         Medium
                                                                     % CVD risk with additional people identified with
                           co-morbidity obesity, hypertension, renal disease, impaired glucose tolerance or a positive
                           family history. However CEG assessment indicates that in East London because of
                           ethnicity and social deprivation,
                           CVD risk with additional co-morbidity
    Demand on healthy     At the national level, DH economic modelling estimates that 20% of those assessed will        Uncertain/neutral    Low
     lifestyle services    require referral to a weight management programme and 48% will require a brief
                           intervention on physical activity. No accurate intelligence available locally on the
                           programme‟s referral rates but no evidence of capacity issues.

   Marketing and          Marketing and communication strategy informed nationally by DH guidelines and locally by          Impeder          High
     communication         pre-pilot social marketing research. However, low awareness and confusion about the
                           Health Check brand remain.
    Data capture and      TH adopted the new QRISK2 CVD risk scoring method in preference to the nationally                 Enabler          High
     collection            recommended Framingham tool. QRISK2 is considered to provide more reliable estimates
                           in higher risk areas such as TH. It is also being used in other areas.
    Manpower production   6 HCAs trained in pilot stage. Good quality training but recruitment of „over qualified‟      Uncertain/neutral   Medium
                           people puts programme at risk of high turnover (HCA churn) as they quickly become
                           attracted by more lucrative roles elsewhere. Further training has taken place to increase
                           the number as part of scaling up the programme, but no clear strategy yet developed for
                           ensuring sustainable manpower.


The NHS Health Check is a new national initiative to integrate the management of coronary
heart disease, stroke, diabetes and kidney disease, and shift the emphasis of management from
secondary to primary prevention. The programme aims to assess and enlighten people about
their CVD risk and reduce their risk of developing disease by appropriate management. Local
implementation of the programme in Tower Hamlets began as a pilot from four GP practices.
There was strong evidence of strategic planning in a number of respects including:

        Formation of a multi-stakeholder steering group
        Development of a strong business case for adoption of a GP-based service model
        Clinical leadership in designing and delivering the programme
        Selection of a contextually appropriate risk assessment tool
        Identification of the priority population groups
        Development of structured clinical and administrative frameworks
        Early recruitment and training of HCAs
        Designation of accountable clinical leads and mentors
        Implementation of a learning pilot phase prior to full rollout
        Pre-pilot social marketing research to inform marketing and delivery of the programme
        Parallel evaluation of the pilot to feed learning into the scaled up programme.

There is growing acknowledgement that conventional broad demographic classifications provide
insufficient explanations for people‟s behaviour and are even less satisfactory in informing the
design of interventions. The prominent role of social marketing in planning and evaluating the
pilot therefore aligns with current national thinking as articulated in the DH document Ambitions
for Health.15

The programme had five distinct but related operational phases – risk identification, assessment,
communication, management (including referral) and recall. The last was not a part of the
evaluation. The evaluation findings revealed on the whole a programme being delivered fairly
effectively in the face of a challenging environmental context. Assessment of organisational
readiness showed that all four pilot practice sites were well prepared in terms of the organisation
of their clinical and administrative/IT structures to implement the programme, albeit to varying

As is the case with newly introduced programmes, some aspects of delivery worked well and
others less well. They are summarised below.

        The main factors that facilitated successful implementation were i) the adoption of well
         developed local clinical guidelines; ii) a targeted approach that prioritised those at highest
         risk; iii) an integrated and functional IT infrastructure; iv) keeping the number of patient
         visits to the minimum possible, v) experienced and well trained HCAs, and vi) strong
         mentoring support (from the PCT and a clinical lead),

  Department of Health (2008). Ambitions for Health: A strategic framework for maximising the potential of
social marketing and health-related behaviour. London: NHS.

      The main constraints to implementation were i) inadequate and inappropriate marketing
       resulting in low public awareness and confusion about the programme; ii) inadequate
       administrative support for the HCAs which contributed to a high workload; iii) coding
       errors; iv) inability to monitor referrals because of non-integration of primary care and
       community data systems; and v) low uptake related to BME patients and specific
       behavioural characteristics (information avoidance for fear of bad news, NHS aversion
       following negative past experiences and difficulty in adjusting to health technology).

Drawing from the key learning points, the following options for taking forward the programme are
proposed. Much of the recommendations for improvement address the marketing strategy for
the Health Check.


Information avoidance. The assumption that people want to know underlies most models of the
information-seeking process (including psychological theory and communication practice) and
marketing strategy. As a result, there is little provision made for people with a tendency to avoid,
ignore, or deny information. This is a distinct behavioural group that emerged from the
evaluation who require altogether different marketing and communication solutions. The main
factor driving their avoidance of information was fear of being given bad news and this insight
should be used in developing appropriate marketing campaigns.

‘Rebranding’. The NHS Health Check requires a brand identity that people can readily identify
with. It suffers from a name that is too vague and creates confusion in the minds of people. The
change of the name of from vascular risk assessment was meant to make it more
understandable by people but does not seem to have achieved the purpose. It is easily confused
with general health checks and also with similar sounding DH initiatives such as NHS
MidLifeCheck. Being a national programme, there are likely to be restrictions on changing the
name but the NHS Health Check identity guidelines (Can we change the NHS Health Check
branding identity for our own local publications?) at
offer some flexibility on local branding and would be worth revisiting.

Peer champions. BME patients are a distinct group for whom targeted approaches are called
for. Evidence from other health programmes conducted in the borough has shown that one
persuasive way to get people to take action about their health is through „peer‟ champions.
These could be early adopters who have had their Health Check or locally recognised
community faces. Acceptance and uptake are further enhanced when people can make a
cultural connection with the message.

Borrowing from good practice. Lessons could be learned from effective marketing approaches
used in other programmes such as the 5 a day advert and smoking cessation which were cited
by participants in the evaluation as good examples.

Sustained engagement. At the overarching level, the big challenge for programmes like the
NHS Health Check is that they are attempting a major shift in mindset that prioritises primary
prevention over the established ways of managing CVD through secondary intervention. This will
take time. This means that one-off or periodic campaigns are unlikely to make a difference An
effective advertising strategy should emphasise continued marketing along the course of the
programme and continue to press home the marketing message (e.g. prolonging life, enhancing
patients‟ life, ensuring a healthy life, etc).

Health care professional education. Patients‟ problems with practice staff has been repeatedly
highlighted as a barrier preventing patients from engaging efficiently with health services. The

issue is well documented as are the corrective measures needed and are only highlighted here
to reemphasise the need for continued staff training on communication skills.

Patient education. The evaluation supports the findings of other studies that highlight the
authority and respect that GPs command. This is probably a factor in the success of the GP
model. But patients at the same time lack an appreciation of the competency and qualifications
of health care professionals other than GPs that work within the practice and more education is
needed in this area.


HCAs are the delivery spearhead of the Health Check and much of the programme‟s success
owes to them. More attention should be given to providing adequate administrative support (or
time) for them as the lack of this was shown to impact on following (calling) up patients to attend
assessment. Also strengthening the mentoring role of the clinical leads in particular is crucial
and will have the added advantage of improving understanding of the programme by other
practice staff.

Coding errors are common in primary care and are not unique to the NHS Health Check.
Regular retraining of all staff involved is the immediate solution to this. At a wider level, the
development of methods and incentives to improve the coding of clinical data and data quality in
electronic primary care records are a priority issue for health care information technology


The status of patients referred to healthy lifestyle services could not be ascertained because of a
lack of integration between community and primary care data systems. Currently the mechanism
for a new primary care system, EMIS Web, is being put in place to resolve this. The system
promises to deliver two key benefits to general practices: access to shared patient records
between GPs and community or secondary care, and advanced functionality for practice staff.


Wider delivery of the Health Check from community based models settings will need to carefully
consider the appropriateness of any data capture technology that will be used. The experience
of this evaluation suggests that people in the older age groups still prefer direct human contact.
Therefore the use of digital vehicles should be focused on people in the lower end of the
targeted age spectrum.

  Majeed A, Car J, Sheikh A (2008). Accuracy and completeness of electronic patient records in primary
care. Family Practice; 24(4): 213-214.