Abstract
Background.
The NHS Health Check programme aims to improve prevention, early diagnosis and management of cardiovascular disease (CVD) in England. High and equitable uptake is essential for the programme to effectively reduce the CVD burden.
Objectives.
Assessing the impact of a local financial incentive scheme on uptake and statin prescribing in the first 2 years of the programme.
Methods.
Cross-sectional study using data from electronic medical records of general practices in Hammersmith and Fulham, London on all patients aged 40–74 years. We assessed uptake of complete Health Check, exclusion of patients from the programme (exception reporting) and statin prescriptions in patients confirmed with high CVD risk.
Results.
The Health Check uptake was 32.7% in Year 1 and 20.0% in Year 2. Older patients had higher uptake of Health Check than younger (65- to 74-year-old patients: Year 1 adjusted odds ratio (AOR) 2.05 (1.67–2.52) & Year 2 AOR 2.79 (2.49–3.12) compared with 40- to 54-year-old patients). The percentage of confirmed high risk patients prescribed a statin was 17.7% before and 52.9% after the programme. There was a marked variation in Health Check uptake, exception reporting and statin prescribing between practices.
Conclusions.
Uptake of the Health Check was low in the first year in patients with estimated high risk despite financial incentives to general practices; although this matched the national required rate in second year. Further evaluations for cost and clinical effectiveness of the programme are needed to clarify whether this spending is appropriate, and to assess the impact of financial incentives on programme performance.
Keywords. Cardiovascular disease, primary care, primary prevention, risk assessment, screening.
Introduction
Cardiovascular diseases (CVDs) are the commonest cause of death and a contributor to health inequalities in many countries.1 Projections suggest the worldwide total burden of CVD will double from 1990 to 2020.2 Cardiovascular risk assessment programmes, such as the ‘Million Hearts’ initiative in the USA3 and the NHS Health Check in England,4 are increasingly being implemented as part of efforts to address this burden.
NHS Health Check is a national population based cardiovascular risk assessment and management programme for all adults aged from 40 to 74 years not already on a CVD register.4 The programme was established in 2009 and aims to reduce CVD incidence and mortality; and socio-economic, ethnic and gender inequalities in health. It was designed to be implemented both in general practices and other community settings, including pharmacies and work places, to make it accessible to a wider population.5–7 Patients presenting with an elevated CVD risk factor, or at greater than or equal to 20% risk of developing CVD in the next 10 years, are provided an appropriate intervention, with the latter eligible for statins.4,8
Financial incentives are increasingly employed to promote preventative interventions, such as smoking cessation, in primary care.9 General practices are commonly paid to screen and manage patients under the Health Check programme. Often simple payments, based on fee-for-service, are structured within a local enhanced service. A small number of areas include Health Checks within a local quality outcomes framework (QOF), a more complex pay-for-performance framework.10
We aimed to assess the impact of a local financial incentive scheme on uptake and statin prescribing in the first 2 years of Health Checks. Given inequalities in CVD risk factors, including smoking and obesity,11,12 and total CVD burden,13 we also examined whether uptake and prescribing within the programme differed between socio-demographic groups.
Methods
NHS Health Check in Hammersmith and Fulham, London
The Department of Health has produced national guidance on implementation of the Health Check programme. Primary care trusts have substantial autonomy to administer the programme according to the needs of their local population, provided minimum standards are met.14
NHS Hammersmith and Fulham is a primary care trust in London serving approximately 185 000 residents. CVD accounts for 35% of mortality in the area. NHS Health Check was implemented ahead of the national schedule, beginning in July 2008 and organized under a local incentive framework for general practice (QOF Plus) (Box 1).10 The primary care trust provided guidance and training15 on implementation of the NHS Health Check programme to 31 independent general practices in the area; the programme was then largely administered by nurses and health care assistants employed by the practices.
Box 1: The QOF Plus business rules to determine Health Check uptake and exception reporting.
| Risk factor recording | YEAR 1 Blood pressure, total and HDL cholesterol, glucose and BMI recorded since 1 July 2008 / Family history of CHD and diabetes, ethnicity and smoking status records / Lifestyle advice on exercise and appropriate dietary changes, Statin prescription records since 1 July 2008. |
| YEAR 2 Blood pressure, Total and HDL cholesterol, glucose and BMI recorded since 1 Dec 2009 / Family history of CHD and diabetes, Ethnicity and smoking status records / Lifestyle advice on exercise and appropriate dietary changes since 1 Dec 2009 / NHS Health Check Complete code since 1 Dec 2009. | |
| Health Check Uptake based on National guidance Completeness of recording of blood pressure + total and HDL cholesterol + BMI + family history of CHD + ethnicity + smoking status + lifestyle advice on exercise and dietary changes + statin prescription (if appropriate). | |
| QOF Plus Payments | YEAR 1 Components of the Health Check were incentivized individually. Practices were awarded with points if meet indicators in 40–90% of their registered eligible population. |
| YEAR 2 Practices were awarded for complete Health Checks. | |
| Practices were awarded 40 points for completion of Health Checks in 70–95% of population with estimated high risk. | |
| Practices were awarded 63 points for completion of Health Checks in 5–30% and extra 23 points for 30–45% of the rest of the population not with estimated high risk. | |
| Exception Reporting | Patients who declined the Health Check offer during the reference period. |
| Patients who does not respond to serial invitations for a Health Check. | |
| Patients who fail to attend a Health Check appointment and do not respond to attempts to rebook the appointment. | |
Existing risk factor data in electronic medical records (EMRs) was used to estimate CVD risk in all individuals aged from 40 to 74 years without diabetes and CVD, (hypertensive and chronic kidney disease (CKD) patients were not excluded from the programme, as per national guidance, but were excluded from analyses) using the Joint British Societies 2 risk algorithm.16 Individuals estimated to be at 20% or greater 10-year CVD risk were prioritized and invited for a Health Check in Year 1 of the programme (1 July 2008–30 November 2009), as NICE guidance recommends.8 In Year 2 (1 December 2009–31 March 2011), practices were incentivized to provide a Health Check to 30% of the remaining non-high risk population using opportunistic methods (Figure 1).15
Figure 1.
Study participants who were eligible for and received a Health Check in Year 1 and Year 2
Data
Data were extracted from EMRs in 27 of the 31 practices on all patients aged from 40 to 74 years, registered on 30 November 2009 for Year 1. Data were available from 29 practices in Year 2, covering eligible patients registered on 31 March 2011. Data on demographic characteristics (e.g. age, sex, ethnicity), clinical characteristics (e.g. BMI, blood pressure, disease status), prescribing and exclusions from the Health Check that determine the payments to general practices (exception reporting) were extracted automatically from EMRs. Patients with diagnosed diabetes, CVD [Coronary heart disease (CHD), stroke/transient ischaemic attack, atrial fibrillation], hypertension and CKD were excluded.17 Newly registered patients i.e. those registered with general practices in the last 3 months of the reference periods, were also excluded, because there was insufficient time to complete a Health Check. The most recent record of each CVD risk factor was extracted.
Outcome measures
Uptake of a complete Health Check was the primary outcome, with prescription of statins in eligible patients8 and exception reporting secondary measures. The business rules of the local financial incentive scheme were applied to determine completeness of all Health Check components and exception reporting.18 The uptake of Health Checks was determined based on the national guidance; patients with all components of the Health Check were counted as having a complete Health Check. The national criteria for determining Health Check uptake and the three valid criteria for exception reporting of patients are outlined in Box 1.
Predictor variables
Age, sex, ethnicity, deprivation (the 2007 indices of multiple deprivation based on postcode of residency), family history of CHD, non-CVD co-morbidities (asthma, mental health, depression, hypothyroidism and chronic obstructive pulmonary disease), smoking status and practice list size were predictor variables.
Analysis
We assessed factors predicting Health Check uptake, exception reporting and statin prescription in patient and practice subgroups. We firstly compared the unadjusted differences in the prevalence of outcomes using two-tailed z-test for proportions. Multi-level logistic regression was then used to examine the predictors of uptake, with adjustment for covariates. We built mixed models, with patient variables at level 1 and practice at level 2, including predictor variables described above. We tested each variable individually for the best fitting model structure; no level 2 structure, random effects or random slope model, using AIC to compare. Random effects models provided the best fit for all variables. We included all variables, using random effects, into the final model, not employing model selection.19 The amount of unexplained variance, a proxy for the differences between practices not attributable to their characteristics, in uptake and statin prescribing at the practice level was assessed by calculating median odds ratio and variance partitioning coefficient; a large coefficient suggest marked practice level variation.
All analyses were conducted using STATA version 11.2 SE. Ethical approval was granted from National Research Ethics Service Committee, London—Queen Square.
Results
In the first year of the Health Check, 4748 patients aged from 40 to 74 years with an estimated risk greater than 20% were targeted and the remaining 35 364 patients were eligible for the Health Check in Year 2. Table 1 shows the characteristics of the eligible populations. In Year 1, the mean age of the eligible population was 60.9 years and more men were eligible than women (78.4%). The mean age of the eligible population was 50.0 years and a higher proportion of women (54.8%) were eligible in Year 2.
Table 1.
Patient and practice characteristics of the study populations eligible for NHS Health Check in the first 2 years
| Year 1—High risk | Year 2—Others | ||||
|---|---|---|---|---|---|
| n | % | n | % | ||
| Sex | Male | 3721 | 78.4 | 12 996 | 45.2 |
| Female | 1027 | 21.6 | 19 368 | 54.8 | |
| Age group | 40–54 | 1035 | 21.8 | 26 875 | 76.0 |
| 55–64 | 2087 | 44.0 | 6407 | 18.1 | |
| 65–74 | 1626 | 34.2 | 2082 | 5.89 | |
| Ethnicity | White | 3389 | 71.4 | 20 086 | 56.8 |
| Black | 277 | 5.83 | 2998 | 8.48 | |
| South Asian | 137 | 2.89 | 818 | 2.31 | |
| Other | 482 | 10.2 | 5454 | 15.4 | |
| Missinga | 463 | 9.75 | 6008 | 17.0 | |
| Local deprivation thirdb | 1 | 1422 | 30.0 | 11 349 | 32.1 |
| 2 | 1669 | 35.1 | 12 050 | 34.1 | |
| 3 | 1657 | 34.9 | 11 965 | 33.8 | |
| Non-CVD co-morbiditiesc | No | 3658 | 77.0 | 26 635 | 75.3 |
| Yes | 1090 | 23.0 | 8729 | 24.7 | |
| Family history of CHD | No | 3395 | 71.5 | 29 152 | 82.4 |
| Yes | 1353 | 28.5 | 6212 | 17.6 | |
| Smoking Status | No | 2340 | 49.3 | 27 892 | 78.9 |
| Yes | 2408 | 50.7 | 7472 | 21.1 | |
| Practice list size | <6000 | 1768 | 37.2 | 12 021 | 34.0 |
| 6000–10 000 | 1776 | 37.4 | 11 942 | 33.8 | |
| Total | >10 000 | 1204 | 25.4 | 11 401 | 32.2 |
| 4748 | 35 364 | ||||
aPatients with either no ethnicity recording or who did not want to state their ethnicity.
bIndex of Multiple Deprivation 2007 (1 = most deprived, 3 = least deprived).
cNon-CVD co-morbidities—asthma, mental health, depression, hypothyroidism and chronic obstructive pulmonary disease.
CVD risk factor recording and risk factors are shown in Table 2. Smoking status and ethnicity recording were high in both years. Recording of blood pressure, cholesterol and family history of CHD were 67.0%, 55.9% and 69.1%, respectively in high risk patients and 57.7%, 36.2% and 42.9% in non-high risk patients in Year 2. Mean blood pressure was 133.7/80.4 mmHg in patients with a complete Health Check in Year 1 and was 125.6/78.2 mmHg in Year 2. Smoking prevalence was 44.3% in patients who had a complete Health Check in Year 1 and 19.7% in Year 2.
Table 2.
Recording of cardiovascular risk factors in eligible study population and levels of risk factors in eligible patients and those with a complete Health Check
| Year 1 | Year 2 | |||||
|---|---|---|---|---|---|---|
| All eligible patients | Complete Health Check | All eligible patients | Complete Health Check | |||
| % Recorded | Mean | Mean | % Recorded | Mean | Mean | |
| n = 4748 | n = 1551 | n = 35 364 | n = 7076 | |||
| Smoking Status | 83.1 | 45.7 (44.1–47.3)%a | 44.3 (41.8–46.8)%a | 85.5 | 16.2 (15.8–16.6)%a | 19.7 (18.8–20.6)%a |
| Ethnicity | 89.3 | – | – | 84.7 | – | – |
| Systolic blood pressure (mmHg) | 67.0 | 134.7 (134.2–135.3) | 133.7 (133.0–134.5) | 57.7 | 125.5 (125.3–125.7) | 125.6 (125.3–126.0) |
| Diastolic blood pressure (mmHg) | 67.0 | 80.6 (80.3–81.0) | 80.4 (80.0–80.9) | 57.7 | 78.0 (77.8–78.1) | 78.2 (78.0–78.4) |
| BMI (kg/m2) | 58.1 | 26.7 (26.5–26.9) | 26.9 (26.7–27.2) | 46.0 | 26.4 (26.3–26.5) | 26.5 (26.4–26.7) |
| Total cholesterol (mmol/L) | 55.9 | 5.37 (5.32–5.41) | 5.29 (5.23–5.34) | 36.2 | 5.29 (5.28–5.31) | 5.31 (5.29–5.33) |
| HDL (mmol/L) | 55.9 | 1.28 (1.63–1.29) | 1.27 (1.25–1.29) | 36.2 | 1.45 (1.44–1.46) | 1.45 (1.44–1.46) |
| Family history of CHD | 69.1 | 41.3 (39.6–42.9)%b | 40.0 (37.6–42.5)%b | 42.9 | 39.9 (39.2–40.7)%b | 27.1 (26.0–28.1)%b |
aMean percentage of patients who smoke.
bMean percentage of patients with a family history of CHD.
Uptake of complete Health Check
In Year 1, 32.7% of high risk patients had a complete Health Check, with 20.0% uptake in Year 2 (Table 3). Older patients were more likely to have had a complete Health Check in both years. In Year 2, uptake was higher in women [adjusted odds ratio (AOR) = 1.27 (95% confidence interval (CI): 1.20–1.35)], in ethnic minorities (Black, South Asian and other) compared with white patients, and in patients living in deprived areas. Patients with non-CVD co-morbidities were more likely to have a complete check [Year 1 AOR = 1.53 (95% CI: 1.31–1.80), Year 2 AOR = 1.75 (95% CI: 1.64–1.87)] and smokers less likely than non-smokers. High median odds ratio and variance partitioning coefficient demonstrated large unexplained variance in uptake between practices. 19.4% (95% CI: 15.2–24.4%) of the total variance in Year 1 and 37.3% (95% CI: 30.6–44.6%) in Year 2 were attributable to unexplained practice-level factors.
Table 3.
Health Check uptake variation by patient and practice characteristics; multivariable analysis
| Year 1—High risk | Year 2—Others | ||||
|---|---|---|---|---|---|
| % | AOR | % | AOR | ||
| Sex | Malea | 32.6 | 1.00 | 17.0 | 1.00 |
| Female | 33.0 | 0.80 (0.67–0.94) | 22.5* | 1.27 (1.20–1.35) | |
| Age Group | 40–54a | 26.9 | 1.00 | 17.7 | 1.00 |
| 55–64 | 30.5** | 1.34 (1.11–1.61) | 25.6* | 1.79 (1.67–1.93) | |
| 65–74 | 39.2* | 2.05 (1.67–2.52) | 33.1* | 2.79 (2.49–3.12) | |
| Ethnicity | Whitea | 35.7 | 1.00 | 22.5 | 1.00 |
| Black | 31.8 | 1.05 (0.78–1.41) | 28.9* | 1.58 (1.43–1.75) | |
| South Asian | 47.4* | 1.27 (0.88–1.87) | 29.0* | 1.50 (1.25–1.78) | |
| Other | 34.2 | 0.94 (0.76–1.17) | 24.3* | 1.16(1.07–1.25) | |
| Missingb | 4.75* | 0.11 (0.07–0.17) | 2.13* | 0.08 (0.07–0.10) | |
| Local deprivation thirdc | 1a | 32.5 | 1.00 | 22.9 | 1.00 |
| 2 | 32.8 | 0.94 (0.79–1.13) | 19.7* | 0.84 (0.78–0.90) | |
| 3 | 32.7 | 0.84 (0.69–1.01) | 17.5* | 0.80 (0.73–0.87) | |
| Non-CVD co-morbiditiesd | Noa | 30.2 | 1.00 | 17.3 | 1.00 |
| Yes | 40.8* | 1.53 (1.31–1.80) | 28.3* | 1.75 (1.64–1.87) | |
| Family history of CHD | Noa | 27.4 | 1.00 | 17.7 | 1.00 |
| Yes | 45.9* | 2.49 (2.15–2.90) | 30.8* | 2.01 (1.87–2.16) | |
| Smoking Status | Noa | 36.9 | 1.00 | 20.3 | 1.00 |
| Yes | 28.5* | 0.71 (0.61–0.83) | 18.6* | 0.83 (0.77–0.90) | |
| Practice list size | <6000a | 32.6 | 1.00 | 15.7 | 1.00 |
| 6000–10 000 | 30.3 | 0.74 (0.37–1.50) | 14.9 | 1.32 (0.23–7.57) | |
| >10 000 | 36.1 | 1.16 (0.51–2.65) | 29.9* | 6.05 (0.84–43.3) | |
| Total | 32.7 | 20.0 | |||
| Random effects | |||||
| Median odds ratio | 2.38 (2.11–2.73) | 3.91 (3.23–4.89) | |||
| Variance partitioning coefficient | 19.4 (15.2–24.4) | 37.3 (30.6–44.6) | |||
Note: Odds ratios are adjusted for all variables in the table.
aReference group in z-test, *P < 0.01 and **P < 0.05 for z-test.
bPatients with either no ethnicity recording or patients who did not want to state their ethnicity.
cIndex of Multiple Deprivation 2007 (1 = most deprived, 3 = least deprived).
dNon-CVD co-morbidities – asthma, mental health, depression, hypothyroidism and chronic obstructive pulmonary disease.
Exception reporting
In Year 1, 46.4% of patients were exception reported and 5.7% in Year 2. Exception reporting varied between patient groups (Table 4). Older patients were less likely to be exception reported in both years. Patients with missing or not stated ethnicity record were more likely to be exception reported [Year 1 AOR = 5.23 (95% CI: 3.92–6.99), Year 2 AOR = 3.00 (95% CI: 2.50–3.60) compared with white]. Patients living in the least deprived areas were more likely to be exception reported in Year 2 [least deprived third AOR = 1.21 (95% CI: 1.03–1.43) compared with most]. Smokers, patients with non-CVD co-morbidities and a family history of CHD were all less likely to be exception reported. There was very high variation in exception reporting between practices in both years, demonstrated by median odds ratio and variance partitioning coefficient values.
Table 4.
Exception reporting by patient and practice characteristics in the first two years of NHS Health Check; multivariable analysis
| Year 1—High risk | Year 2—Others | ||||
|---|---|---|---|---|---|
| % | AOR | % | AOR | ||
| Sex | Malea | 47.5* | 1.00 | 6.77 | 1.00 |
| Female | 42.3* | 1.13 (0.96–1.34) | 4.55* | 0.78 (0.69–0.88) | |
| Age group | 40–54a | 54.8 | 1.00 | 5.82 | 1.00 |
| 55–64 | 49.8* | 0.69 (0.58–0.83) | 5.04** | 0.85 (0.72–1.00) | |
| 65–74 | 36.6* | 0.37 (0.30–0.45) | 3.70* | 0.87 (0.65–1.17) | |
| Ethnicity | Whitea | 41.9 | 1.00 | 5.14 | 1.00 |
| Black | 44.8 | 1.15 (0.86–1.54) | 4.20** | 0.63 (0.50–0.79) | |
| South Asian | 33.6 | 0.92 (0.61–1.39) | 3.67 | 1.10 (0.68–1.76) | |
| Other | 47.7** | 1.08 (0.87–1.34) | 5.72 | 0.80 (0.68–0.94) | |
| Missingb | 82.7* | 5.23 (3.92–6.99) | 7.71* | 3.00 (2.50–3.60) | |
| Local deprivation thirdc | 1a | 46.8 | 1.00 | 4.08 | 1.00 |
| 2 | 43.7 | 1.03 (0.86–1.23) | 6.59* | 1.31 (1.12–1.53) | |
| 3 | 48.7 | 1.23 (1.02–1.48) | 5.91* | 1.21 (1.03–1.43) | |
| Non-CVD co-morbiditiesd | Noa | 49.2 | 1.00 | 5.75 | 1.00 |
| Yes | 36.9* | 0.66 (0.56–0.77) | 4.96** | 0.78 (0.68–0.90) | |
| Family history of CHD | Noa | 51.6 | 1.00 | 5.50 | 1.00 |
| Yes | 33.2* | 0.41 (0.35–0.48) | 5.80 | 0.74 (0.64–0.86) | |
| Smoking Status | Noa | 42.2 | 1.00 | 5.06 | 1.00 |
| Yes | 50.4* | 1.37 (1.19–1.59) | 7.39* | 1.36 (1.18–1.56) | |
| Practice list size | <6000a | 45.1 | 1.00 | 1.02 | 1.00 |
| 6000–10 000 | 45.9 | 1.30 (0.39–4.34) | 0.52* | 1.12 (0.08–15.2) | |
| >10 000 | 48.8** | 1.80 (0.44–7.34) | 15.6* | 20.3 (1.11–372.7) | |
| Total | 46.4 | 5.72 | |||
| Random effects | |||||
| Median odds ratio | 3.10 (2.65–3.72) | 5.06 (3.85–7.06) | |||
| Variance partitioning coefficient | 29.1 (23.3–35.6) | 45.7 (36.7–55.0) | |||
Note: Odds ratios are adjusted for all variables in the table.
aReference group in z-test, * P < 0.01 and *P < 0.05 for z-test.
bPatients with either no ethnicity recording or patients who did not want to state their ethnicity.
cIndex of Multiple Deprivation 2007 (1 = most deprived, 3 = least deprived).
dNon-CVD co-morbidities – asthma, mental health, depression, hypothyroidism and chronic obstructive pulmonary disease.
Statin prescribing
Table 5 shows statin prescribing in patients who had Health Check and were confirmed as at high risk (≥20%) in both years combined. 17.7% of 1630 high risk population were prescribed statins before the Health Check, while 52.9% of high risk patients were prescribed statins after (the percentage in Year 1, from before and after the Health Check, was 16.8% and 57.7%, respectively; 19.4% and 43.1% in Year 2). Older patients were more likely to be prescribed statins before the Health Check and there was no significant difference in prescribing by age after the Health Check. Post-Health Check statin prescribing was higher amongst Black patients [AOR = 2.11 (95% CI: 1.12–3.95)] compared with white. Unexplained variation in statin prescribing between practices decreased over the Health Check, from 18.8% (95% CI: 13.0–26.7%) of total variation to 11.6% (95% CI: 7.06–18.2%).
Table 5.
Statin prescribing in patients eligible for treatment after the Health Check; multivariable analysis
| High risk patients | |||||
|---|---|---|---|---|---|
| Pre-Health Check | Post-Health Check | ||||
| % | AOR | % | AOR | ||
| Sex | Malea | 16.8 | 1.00 | 54.3 | 1.00 |
| Female | 21.6* | 1.11 (0.71–1.72) | 47.2* | 0.75 (0.53–1.06) | |
| Age group | 40–54a | 9.83 | 1.00 | 49.2 | 1.00 |
| 55–64 | 17.4** | 1.87 (1.02–3.40) | 51.9 | 1.08 (0.73–1.59) | |
| 65–74 | 21.4** | 2.49 (1.33–4.67) | 55.6 | 1.30 (0.86–1.98) | |
| Ethnicity | Whitea | 17.8 | 1.00 | 51.9 | 1.00 |
| Black | 14.2 | 0.75 (0.33–1.70) | 60.4 | 2.11 (1.12–3.95) | |
| South Asian | 25.0 | 0.92 (0.40–2.13) | 55.4 | 0.89 (0.49–1.60) | |
| Other | 16.8 | 0.83 (0.45–1.54) | 54.2 | 0.98 (0.64–1.50) | |
| Missingb | 7.14 | 0.50 (0.06–4.26) | 53.6 | 1.16 (0.40–3.40) | |
| Local deprivation thirdc | 1a | 18.6 | 1.00 | 55.4 | 1.00 |
| 2 | 16.1 | 1.15 (0.74–1.80) | 52.6 | 1.03 (0.75–1.42) | |
| 3 | 18.4 | 1.41 (0.89–2.21) | 50.8 | 0.96 (0.69–1.34) | |
| Non-CVD co-morbiditiesd | Noa | 16.2 | 1.00 | 53.2 | 1.00 |
| Yes | 21.1* | 2.06 (1.43–2.96) | 52.3 | 1.34 (1.01–1.78) | |
| Family history of CHD | Noa | 16.8 | 1.00 | 52.8 | 1.00 |
| Yes | 19.0 | 1.20 (0.84–1.72) | 53.1 | 0.89 (0.68–1.16) | |
| Smoking Status | Noa | 19.0 | 1.00 | 50.1 | 1.00 |
| Yes | 16.3 | 0.99 (0.68–1.43) | 55.8* | 1.41 (1.07–1.87) | |
| Practice list size | <6000a | 21.0 | 1.00 | 58.3 | 1.00 |
| 6000–10 000 | 17.3 | 0.67 (0.30–1.49) | 48.0** | 0.72 (0.44–1.19) | |
| >10 000 | 14.5** | 0.58 (0.23–1.46) | 52.1* | 0.92 (0.53–1.61) | |
| Total | 17.7 | 52.9 | |||
| Random effects | |||||
| Median odds ratio | 2.34 (1.98–2.91) | 1.89 (1.63–2.30) | |||
| Variance partitioning coefficient | 18.8 (13.0–26.7) | 11.6 (7.06–18.2) | |||
Note: Odds ratios are adjusted for all variables in the table.
aReference group in z-test, **P < 0.01 and *P < 0.05 for z-test.
bPatients with either no ethnicity recording or patients who did not want to state their ethnicity.
cIndex of Multiple Deprivation 2007 (1 = most deprived, 3 = least deprived).
dNon-CVD co-morbidities – asthma, mental health, depression, hypothyroidism and chronic obstructive pulmonary disease.
Discussion
Main findings
Uptake of a Health Check was low at 32.7% among high risk patients in the first year and 20% in rest of the eligible population in the second year. In both years, uptake was lower in younger patients, smokers and patients with no ethnicity record. In the second year, uptake was higher amongst patients of South Asian and Black ethnic backgrounds. Statin prescribing increased from 17.7% to 52.9% in patients confirmed to be high risk after a Health Check. There was high practice variation in Health Check uptake, exception reporting and statin prescribing.
Comparison with existing literature
Evidence on the uptake of cardiovascular screening in routine settings is limited. In a deprived and culturally diverse area, 44.8% of high risk patients attended Health Check.20 Our study showed lower uptake among patients assessed as high risk, but unlike this previous study excluded patients with hypertension and CKD, whose reviews in general practice may improve uptake. Dalton et al.20 reported poor uptake of Health Checks in younger patients and smokers. In the OXCHECK study, men were less likely to attend for cardiovascular screening,21 consistent with findings from Year 2. Our findings support previous work, which suggest that patients from South Asian and Black groups may be more likely to attend CVD screening.20,22 Previous studies examining uptake of Health Check by socio-economic status have produced mixed results.20,21,23 We identified higher uptake in patient groups living in deprived areas, but this was only apparent in Year 2. Improved prevention in deprived groups may be due to general practitioners’ perception of greater CVD risk.24
Half the patients (52.9%) confirmed to be at high CVD risk were prescribed statins after their Health Check. This is a similar level to that found in a previous study20 and considerably lower than that anticipated in Department of Health impact modelling, which assumed the uptake of statins would be 85%. Limited statin prescribing for primary prevention may be because of patients’ and practitioners’ attitude to cardiovascular risk and beliefs about the benefits of prescribing statins in patients without an established disease.25,26
Strengths and limitations of the study
Our study covers almost all the population eligible for the NHS Health Check in one Primary Care Trust. Patients from four practices in Year 1 and two practices in the Year 2 could not be included due to technical errors in data extraction, but patients from these practices were similar in their characteristics to our study population. Although the study population is not representative of the whole UK, findings may be representative of other urban areas with similar patterns of deprivation, ethnic diversity and burden of CVD.
National evaluation of the Health Check has been initiated in 2012; however, given variation in approaches between areas, local programme assessment remains vital. Our study defined eligibility and uptake based on national guidance, however many eligible patients had some risk factors recorded, which may be a partial Health Check. Findings are therefore comparable with those from other areas. Nationally, Health Check can be carried out outside of general practice, for example in pharmacies.6,7 Our findings do not provide evidence on the effectiveness of the programme in such settings.
We could only assess statin prescribing in eligible patients, with no data concerning adherence. For primary prevention, adherence may be low.26 We could not examine the uptake and adherence to other interventions, such as weight and exercise management, and smoking cessation. Patient and practice deprivation were determined using postcode based scores. Individual measures of socioeconomic status might better predict health outcomes, but these are not included in routine medical data.
Implications for future research or clinical practice
We have shown that men without previous CVD are less likely to attend and more likely to be excluded from Health Check. This is a concern given that they are at greater CVD risk than women, but probably reflects their lower attendance at general practice27 and lower use of preventative services more generally.28 One key implication is the need for primary care teams to promote Health Check attendance in men.
Previous work suggests NHS Health Check have considerable workload implications for primary care29,30 and there have been concerns over the cost-effectiveness of the programme.31 Department of Health modelling assumes 75% of patients attend, with 85% uptake of statins.5 First year attendance among high risk patients was low, but 20% uptake (of the entire eligible population) in Year 2 is in keeping with the local target and programme’s goal to screen all persons aged from 40 to 74 years over a 5-year period. However, attendance may decrease in subsequent years, since highly motivated individuals may attend for screening in the first 2 years. It is therefore important that uptake of the Health Check programme be monitored over time; qualitative assessment of patients’ experiences of the Health Check may be necessary to derive lesson for improving programme performance. Achieving high uptake in hard-to-reach groups may require dedicated strategies and greater resources.32
The Health Check programme will not achieve its aims of reducing the CVD burden and health inequalities with a low uptake observed in this study. The high exception reporting of patients during the first year of the programme is concerning and needs to be monitored in future programmes with a pay for performance element. While our study design did not permit us to isolate the impact of the financial incentives on the performance of the Health Check, uptake in the study area was higher than that in a similar urban setting, where no such financial incentive scheme was in place.20 Examining the effect of different implementation and payment approaches may help identify financial, other incentives and organizational methods linked to higher uptake of the Health Check programme.
Conclusions
Uptake of the Health Check programme was low in first year in patients with estimated high risk despite financial incentives to general practices; although matched the national required rate in second year and only half of patients confirmed to be at high risk were prescribed a statin. Further evaluation of cost and clinical effectiveness is needed as programme resources may be better deployed in whole population strategies, to increase physical activity, reduce smoking and encourage healthy eating,33 for reducing cardiovascular risk. This is particularly important in the context of a financial environment in which NHS spending will be increasingly scrutinized to reduce spending on lower benefit and more costly programmes.34
Declaration
Funding: This work was supported by the National Institutes of Health Research North West London Collaboration for Leadership in Applied Health Research & Care in partnership with National Health Service Hammersmith and Fulham. CM is funded by the Higher Education Funding Council for England and the National Institutes of Health Research North West London Collaboration for Leadership in Applied Health Research & Care. The Department of Primary Care & Public Health has received funding from the Department of Health Policy Research Programme to carry out a national evaluation of the National Health Service Health Check programme.
Ethical approval: National Research Ethics Service Committee, London—Queen Square.
Conflict of interest: none.
Acknowledgements
We thank the Hammersmith and Fulham Primary Care Trust for providing us the data and all practices that participated in the study. The Department of Primary Care & Public Health at Imperial College is grateful for support from the National Institutes of Health Research (NIHR) North West London Collaboration for Leadership in Applied Health Research & Care Scheme, the NIHR Biomedical Research Centre scheme, and the Imperial Centre for Patient Safety and Service Quality.
References
- 1. Mendis S, Puska P, Norrving B.(ed.). Global Atlas on cardiovascular disease prevention and control. Geneva: World Health Organisation, 2011. [Google Scholar]
- 2. Mackay J, Mensah GA. The Atlas of Heart Disease and Stroke. World Health Organization and Centres for Disease Control and Prevention, 2004. [Google Scholar]
- 3. Frieden TR, Berwick DM. The “Million Hearts” initiative–preventing heart attacks and strokes. N Engl J Med 2011; 365: e27 [DOI] [PubMed] [Google Scholar]
- 4. Department of Health Putting Prevention First - Vascular Checks: Risk Assessment and Management. London: Department of Health, 2008. [Google Scholar]
- 5. Department of Health Putting prevention first: Vascular Checks: risk assessment and management - Impact Assessment. London: Department of Health, 2008. [Google Scholar]
- 6. McNaughton RJ, Oswald NT, Shucksmith JS, Heywood PJ, Watson PS. Making a success of providing NHS Health Checks in community pharmacies across the Tees Valley: a qualitative study. BMC Health Serv Res 2011; 11: 222 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Kaczorowski J, Chambers LW, Dolovich L, et al. Improving cardiovascular health at population level: 39 community cluster randomised trial of Cardiovascular Health Awareness Program (CHAP). BMJ 2011; 342: d442 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Cooper A, Nherera L, Calvert N, O’Flynn N, Turnbull N, Robson J. Clinical Guidelines and Evidence Review for Lipid Modification: cardiovascular risk assessment and the primary and secondary prevention of cardiovascular disease. London: National Collaborating Centre for Primary Care and Royal College of General Practitioners, 2008. [Google Scholar]
- 9. Hamilton FL, Greaves F, Majeed A, Millett C. Effectiveness of providing financial incentives to healthcare professionals for smoking cessation activities: systematic review. Tob Control 2013; 22: 3–8 [DOI] [PubMed] [Google Scholar]
- 10. Millett C, Majeed A, Huckvale C, Car J. Going local: devolving national pay for performance programmes. BMJ 2011; 342: c7085 [DOI] [PubMed] [Google Scholar]
- 11. Hiscock R, Bauld L, Amos A, Platt S. Smoking and socioeconomic status in England: the rise of the never smoker and the disadvantaged smoker. J Public Health (Oxf) 2012; 34: 390–6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. El-Sayed AM, Scarborough P, Galea S. Unevenly distributed: a systematic review of the health literature about socioeconomic inequalities in adult obesity in the United Kingdom. BMC Public Health 2012; 12: 18 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. McFadden E, Luben R, Wareham N, Bingham S, Khaw KT. Occupational social class, risk factors and cardiovascular disease incidence in men and women: a prospective study in the European Prospective Investigation of Cancer and Nutrition in Norfolk (EPIC-Norfolk) cohort. Eur J Epidemiol 2008; 23: 449–58 [DOI] [PubMed] [Google Scholar]
- 14. Department of Health Putting Prevention First - Vascular Checks: Risk Assessment and Management - Next steps Guidance for Primary Care Trusts. London: Department of Health, 2008. [Google Scholar]
- 15. NHS Hammersmith and Fulham The Handbook Year 2. Quality and Outcomes Framework Plus Year 2. London: NHS Hammersmith and Fulham; 2010. [Google Scholar]
- 16. Joint British Societies JBS 2: Joint British Societies’ guidelines on prevention of cardiovascular disease in clinical practice. Heart 2005; 91(Suppl 5):v1–52.10.1136/hrt.2005.079988 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. NHS Health Check Programme Putting Prevention First - NHS Health Check: Vascular Risk Assessment and Management Best Practice Guidance. London: Department of Health, 2009. [Google Scholar]
- 18. NHS Hammersmith and Fulham Financial and Business Rules Year 2. Quality Outcomes Framework Plus. London: NHS Hammersmith and Fulham, 2009. [Google Scholar]
- 19. Ratner B. Variable selection methods in regression: Ignorable problem, outing notable solution. Journal of Targeting, Measurement and Analysis for Marketing 2010; 18: 65–75 [Google Scholar]
- 20. Dalton AR, Bottle A, Okoro C, Majeed A, Millett C. Uptake of the NHS Health Checks programme in a deprived, culturally diverse setting: cross-sectional study. J Public Health (Oxf) 2011; 33: 422–9 [DOI] [PubMed] [Google Scholar]
- 21. Thorogood M, Coulter A, Jones L, Yudkin P, Muir J, Mant D. Factors affecting response to an invitation to attend for a health check. J Epidemiol Community Health 1993; 47: 224–8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Lambert AM, Burden AC, Chambers J, Marshall T. Heart of Birmingham Teaching Primary Care Trust Cardiovascular screening for men at high risk in Heart of Birmingham Teaching Primary Care Trust: the ‘Deadly Trio’ programme. J Public Health (Oxf) 2012; 34: 73–82 [DOI] [PubMed] [Google Scholar]
- 23. Waller D, Agass M, Mant D, Coulter A, Fuller A, Jones L. Health checks in general practice: another example of inverse care? BMJ 1990; 300: 1115–8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Willems SJ, Swinnen W, De Maeseneer JM. The GP’s perception of poverty: a qualitative study. Fam Pract 2005; 22: 177–83 [DOI] [PubMed] [Google Scholar]
- 25. Eaton CB, Galliher JM, McBride PE, Bonham AJ, Kappus JA, Hickner J. Family physician’s knowledge, beliefs, and self-reported practice patterns regarding hyperlipidemia: a National Research Network (NRN) survey. J Am Board Fam Med 2006; 19: 46–53 [DOI] [PubMed] [Google Scholar]
- 26. Mitka M. Improving Medication Adherence Promises Great Payback, but Poses Tough Challenge. JAMA 2010; 303: 825.10.1001/jama.2010.212 [DOI] [PubMed] [Google Scholar]
- 27. Hippisley-Cox J, Vinogradova G. Final Report to the NHS Information Centre and Department of Health. Trends in Consultation Rates in General Practice 1995/1996 to 2008/2009: Analysis of the QRESEARCH database. The NHS Information Centre, 2009. [Google Scholar]
- 28. Department of Health Men’s Health and Wellbeing Strategy. Melbourne: State of Victoria, Department of Health, 2010 [Google Scholar]
- 29. Dalton AR, Bottle A, Okoro C, Majeed A, Millett C. Implementation of the NHS Health Checks programme: baseline assessment of risk factor recording in an urban culturally diverse setting. Fam Pract 2011; 28: 34–40 [DOI] [PubMed] [Google Scholar]
- 30. Artac M, Dalton AR, Majeed A, et al. Assessment of cardiovascular risk factors prior to NHS Health Checks in an urban setting: cross-sectional study. JRSM Short Rep 2012; 3: 17 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Capewell S. Will screening individuals at high risk of cardiovascular events deliver large benefits? No. BMJ 2008; 337: a1395 [DOI] [PubMed] [Google Scholar]
- 32. Zelenyanszki C. Maximising Screening Attendance A Reference Guide. NHS North West London Cancer Network, 2009. [Google Scholar]
- 33. Pearson TA. Public policy approaches to the prevention of heart disease and stroke. Circulation 2011; 124: 2560–71 [DOI] [PubMed] [Google Scholar]
- 34. Majeed A. Primary care in Europe: entering the age of austerity. J Ambul Care Manage 2012; 35: 162–6 [DOI] [PubMed] [Google Scholar]