A cross-sectional study examining the equitability of invitation, uptake and coverage for NHS Health Check

Abstract

Background

The evidence for access to NHS Health Check (NHSHC) varies considerably across the country. This study examined the equity in invitation, uptake and coverage of NHSHC and impact of different invitation methods.

Methods

This patient-level cross-sectional study from 52 general practices in Walsall used adjusted logistic regressions to examine the association between patient characteristics (age, sex, ethnicity and deprivation) and NHSHC access.

Results

Over the 5-year study period, 61 464 people were eligible for NHSHC, 66% were invited, uptake was 74% and coverage was 55%. Males had lower odds of: invitation (AOR: 0.78, 95% CI: 0.75–0.81), uptake (0.73, 95% CI: 0.70–0.77) and coverage (0.69, 95% CI: 0.66–0.71). Compared with White, the ‘Other’ ethnicity group (mixed backgrounds, other Asians that are not South Asians and other ethnic groups) had lower odds of: invitation (0.74, 95% CI: 0.67–0.81), uptake (0.86, 95% CI: 0.75–0.98) and coverage (0.74, 95% CI: 0.68–0.81). The most deprived areas had lower odds of invitation, uptake and coverage. Opportunistic invitation had a 25-fold increase in odds of uptake.

Conclusions

The study has highlighted areas of inequities in access to NHSHC. The group most negatively affected were men, people from particular minority ethnic groups and people from deprived communities. Further actions are needed to reduce these inequities.

health inequalities, health policy, health services, health care disparities, public health

Background

NHS Health Check (NHSHC) is a national risk assessment and management programme for the prevention of cardiovascular disease (CVD) and other related diseases.¹ The NHSHC pathway consists of identifying and inviting eligible population once every five years, completing their risk assessment at attendance, referring those with high risk for clinical or lifestyle interventions.¹

Walsall is culturally diverse with a higher proportion of Black, Asian and Minority Ethnic (BAME) compared with England. A quarter of Walsall’s 167 Lower Layer Super Output Area (LSOA) is among the most deprived 10% in England with Walsall ranking 25th (out of 317 most deprived English local authorities).² NHSHC in Walsall is delivered primarily through General Practices (GP); however, additional provision is also made through the Healthy Workplace Programme³ and more ad hoc provision through various community events such as The Health Bus.⁴

The NHSHC aims to address health inequalities by making the programme available to everyone aged between 40 and 74 years who meet the eligible criteria irrespective of their age, sex or ethnicity.¹ The potential challenge with this type of approach is that if there is bias in those who attend the NHSHC, this has the potential to inadvertently widen health inequity.⁵

As part of the Public Health Outcomes Framework,⁶ there are three process measures used to evaluate the performance of NHSHC:

• Invited: percentage of the eligible population offered an NHSHC.

• Uptake: percentage of the eligible population offered an NHSHC who received an NHCHC.

• Coverage: percentage of the eligible population who received an NHSHC.

The evidence for NHSHC attendance varies considerably across different regions, although there is consistent evidence that supports attendance being higher among older patients and female patients. The evidence is however mixed for ethnicity and deprivation. A national study by Chang et al.⁷ found significantly lower coverage in Black Africans and Chinese (compared with White ethnicity). This contrasts with most of the local peer-reviewed studies that found higher level of attendance in Black,^8,9 South Asians^8,10,11 and minority ethnicity more generally.¹²

The aim of the study was to assess the equity in the invitation, uptake and coverage of NHSHC in Walsall during the study period. The dimensions of equity assessed were age, sex, ethnicity and deprivation. A second aim of the study was to examine the impact of the different invitation methods on uptake.

Methods

Cross-sectional data for NHSHC eligible patients in Walsall between 1 October 2014 and 30 September 2019 were obtained. By allowing for a complete 5-year cycle, this ensured that every eligible participant in the study should have attended at least one NHSHC during the study window.

In line with national rules of NHSHC,¹ the study population for this study (Supplementary Fig. 1) were patients currently registered with a General Practice (GP) in Walsall who: were aged 40–74 years as of 1 October 2014, had not attended NHSHC in the previous 5 years before 1 October 2014, were not prescribed statin and did not have the following pre-existing diagnoses at the start of the study—coronary heart disease (CHD), chronic kidney disease (CKD) Stages 3–5, diabetes, hypertension, atrial fibrillation (AF), transient ischaemic attack (TIA), hypercholesterolemia, heart failure, peripheral arterial disease and stroke.

All 52 General Practices (GP) in Walsall agreed for their patient records stored in the EMIS (Egton Medical Information Systems)¹³ enterprise database to be interrogated for this study.

Outcomes

The primary outcomes for this study were invitation, uptake and coverage of NHSHC. Data on participants were extracted using universal Read codes as described in the NHSHC Business Rule.¹

Exposures and co-variates

Data on equity vectors of interest (age, sex, ethnicity, deprivation) and methods of invitation were extracted using Read codes as described in the NHSHC Business Rule.¹ Given that all eligible patients should have been invited during the study period and to avoid over-adjustment, only the variables relating to the equity vectors of interest (age, sex, ethnicity and deprivation) were adjusted for during analyses.

Data manipulation and cleansing

Anonymized individual level data were extracted from EMIS. Date of birth was converted to age at the start of the study, and these were grouped into 40–44, 45–54, 55–64 and 65–74 years. Deprivation status was obtained by linking Lower Super Output Area (LSOA, derived from individual’s postcode in EMIS) to Index of Multiple Deprivation (IMD) data 2019.

Statistical analysis

Associations between participant characteristics and each outcome measure were examined using unadjusted and adjusted multilevel logistic regression models (allowing for clustering of patients at general practice level as a random effect). Odds ratio (OR) and adjusted odds ratio (AOR) with 95% confidence intervals were reported. Intra-cluster correlation coefficients (ICC) were reported, using the null linear model.

All analyses were conducted using Stata v15. Data manipulations were performed using MS Access.

Research governance and ethical considerations

Ethical approval was not required for this service evaluation. However, approval to interrogate non-identifiable patient’s data on EMIS database was granted by the assigned committees in Walsall Clinical Commissioning Group and Walsall Local Authority.

Results

Population profile

In the eligible population, there were more women compared to men (50.9 versus 49.2%), more people aged 45–54 years (43.3%) compared to other age groups, more people who identified as white ethnicity (78.0%) and more people in the most deprived IMD quintiles (47.7%). The same patterns were observed in the invited and attended populations. Around 5% of patients had a high risk of CVD (QRISK>20) in all three groups (Table 1).

Table 1

Characteristics of study population categorized by eligible, invited and attended

Variables	Eligible	Invited	Not invited	Attended	Not attended
Total, n (% of eligible)	61 464	40 591 (66.04)	20 873 (33.96)	33 761 (54.93)	27 703 (45.07)
Sex
Female, n (%)	31 257 (50.85)	21 406 (52.74)	9851 (47.19)	18 593 (55.07)	12 664 (45.71)
Male, n (%)	30 207 (49.15)	19 185 (47.26)	11 022 (52.81)	15 168 (44.93)	15 039 (54.29)
Age
Years, mean (SD)	52.78 (8.69)	52.67 (8.47)	52.99 (9.09)	53.10 (8.55)	52.39 (8.83)
Age group
40–44, n (%)	11 998 (19.52)	7815 (19.25)	4183 (20.04)	6059 (17.95)	5939 (21.44)
45–54, n (%)	26 604 (43.28)	17 760 (43.75)	8844 (42.37)	14 501 (42.95)	12 103 (43.69)
55–64, n (%)	14 952 (24.33)	10 036 (24.72)	4916 (23.55)	8727 (25.85)	6225 (22.47)
65–74, n (%)	7910 (12.87)	4980 (12.27)	2930 (14.04)	4474 (13.25)	3436 (12.40)
Ethnicity
White, n (%)	47 924 (77.97)	33 144 (81.65)	14 780 (70.81)	27 928 (82.72)	19 996 (72.18)
Black, n (%)	1494 (2.43)	959 (2.36)	535 (2.56)	807 (2.39)	687 (2.48)
South Asian, n (%)	6868 (11.17)	4115 (10.14)	2753 (13.19)	3857 (11.42)	3011 (10.87)
Others, n (%)	2246 (3.65)	1329 (3.27)	917 (4.39)	1082 (3.20)	1164 (4.20)
Unknown, n (%)	2932 (4.77)	1044 (2.57)	1888 (9.05)	87 (0.26)	2845 (10.27)
IMD deprivation
1 - Least deprived, n (%)	6899 (11.22)	4750 (11.70)	2149 (10.30)	4227 (12.52)	2672 (9.65)
2, n (%)	6975 (11.35)	4537 (11.18)	2438 (11.68)	4010 (11.88)	2965 (10.70)
3, n (%)	7850 (12.77)	5279 (13.01)	2571 (12.32)	4517 (13.38)	3333 (12.03)
4, n (%)	10 386 (16.90)	7135 (17.58)	3251 (15.58)	5944 (17.61)	4442 (16.03)
5 - Most deprived, n (%)	29 295 (47.66)	18 853 (46.45)	10 442 (50.03)	15 046 (44.57)	14 249 (51.43)
Unknown, n (%)	59 (0.10)	37 (0.09)	22 (0.11)	17 (0.05)	42 (0.15)
QRISK
Low-risk, n (%)	30 527 (49.67)	24 680 (60.80)	5847 (28.01)	23 222 (68.78)	7305 (26.37)
High-risk > 10, n (%)	9976 (16.23)	7427 (18.30)	2549 (12.21)	7126 (21.11)	2850 (10.29)
High-risk > 20, n (%)	3261 (5.31)	2091 (5.15)	1170 (5.61)	1895 (5.61)	1366 (4.93)
Unknown, n (%)	17 700 (28.80)	6393 (15.75)	11 307 (54.17)	1518 (4.50)	16 182 (58.41)

Note: Ethnicity: White = White and Other White; Black = Black African, Black Caribbean, Other Black background; South Asian = Indian, Pakistan, Bangladesh; Others = Other ethnic group, other Asians, mixed backgrounds); SD = standard deviation.

Invitation, uptake and coverage

Table 2 shows analyses of invitation, uptake and coverage by sex, age, ethnicity and deprivation. In total, 61 464 people were eligible for NHSHC during the study period, and 40 591 (66.0% of eligible) were invited to participate with 74.4% uptake among those invited. Overall, 33 761 (54.9% of eligible) attended NHSHC (Fig. 1). There was variation between general practices for invitation (8.6–94.0%), uptake (33.5–97.9%) and coverage (10.8–93.7%). The ICC for the null linear model (before adjustment) shows that invitation, uptake and coverage were slightly correlated within the general practice clusters with ICC of 25, 21 and 18%, respectively.

Table 2

Regression analysis of invited among eligible, uptake among those invited and overall coverage among eligible population

	Invited (%)a	OR (95% CI)	AOR (95% CI)b	Uptake (%)c	OR (95% CI)	AOR (95% CI)b	Coverage (%)d	OR (95% CI)	AOR (95% CI)b
Total	40 591 (66.04)	—	—	30 190 (74.38)	—	—	33 761 (54.93)	—	—
Sex
Female	21 406 (68.48)	—	—	16 630 (77.69)	—	—	18 593 (59.48)	—	—
Male	19 185 (63.51)	0.80 (0.77, 0.83)	0.78 (0.75, 0.81)	13 560 (70.68)	0.69 (0.66, 0.72)	0.73 (0.70, 0.77)	15 168 (50.21)	0.69 (0.67, 0.71)	0.69 (0.66, 0.71)
Age group
40–44	7815 (65.14)	—	—	5401 (69.11)	—	—	6059 (50.50)	—	—
45–54	17 760 (66.76)	1.07 (1.03, 1.12)	1.04 (0.99, 1.10)	13 008 (73.24)	1.22 (1.15, 1.30)	1.26 (1.18, 1.34)	14 501 (54.51)	1.17 (1.12, 1.23)	1.18 (1.12, 1.24)
55–64	10 036 (67.12)	1.09 (1.04, 1.15)	1.06 (1.00, 1.12)	7799 (77.71)	1.56 (1.46, 1.67)	1.61 (1.49, 1.74)	8727 (58.37)	1.37 (1.31, 1.44)	1.37 (1.30, 1.45)
65–74	4980 (62.96)	0.91 (0.86, 0.97)	0.81 (0.75, 0.86)	3982 (79.96)	1.78 (1.64, 1.94)	1.72 (1.57, 1.89)	4474 (56.56)	1.28 (1.21, 1.35)	1.20 (1.12, 1.28)
Ethnicity
White	33 144 (69.16)	—	—	25 117 (75.78)	—	—	27 928 (58.28)	—	—
Black	959 (64.19)	0.80 (0.72, 0.89)	0.85 (0.75, 0.96)	729 (76.02)	1.01 (0.87, 1.18)	1.23 (1.04, 1.45)	807 (54.02)	0.84 (0.76, 0.93)	1.01 (0.90, 1.13)
South Asian	4115 (59.92)	0.67 (0.63, 0.70)	1.00 (0.94, 1.08)	3309 (80.41)	1.31 (1.21, 1.42)	1.32 (1.19, 1.45)	3857 (56.16)	0.92 (0.87, 0.97)	1.18 (1.10, 1.26)
Others	1329 (59.17)	0.65 (0.59, 0.70)	0.74 (0.67, 0.81)	975 (73.36)	0.88 (0.78, 1.00)	0.86 (0.75, 0.98)	1082 (48.17)	0.67 (0.61, 0.72)	0.74 (0.68, 0.81)
Unknown	1044 (35.61)	0.25 (0.23, 0.27)	0.23 (0.21, 0.25)	60 (5.75)	0.02 (0.01, 0.03)	0.02 (0.02, 0.03)	87 (2.97)	0.02 (0.02, 0.03)	0.02 (0.02, 0.03)
IMD
1 (Least deprived)	4750 (68.85)	—	—	3714 (78.19)	—	—	4227 (61.27)	—	—
2	4537 (65.05)	0.84 (0.78, 0.90)	1.00 (0.91, 1.09)	3603 (79.41)	1.08 (0.97, 1.19)	0.94 (0.83, 1.05)	4010 (57.49)	0.85 (0.80, 0.91)	0.94 (0.87, 1.02)
3	5279 (67.25)	0.93 (0.87, 1.00)	0.94 (0.86, 1.02)	4073 (77.15)	0.94 (0.86, 1.04)	0.90 (0.80, 1.01)	4517 (57.54)	0.86 (0.80, 0.92)	0.88 (0.81, 0.95)
4	7135 (68.70)	0.99 (0.93, 1.06)	0.92 (0.84, 1.00)	5458 (76.50)	0.91 (0.83, 0.99)	0.83 (0.75, 0.93)	5944 (57.23)	0.85 (0.79, 0.90)	0.81 (0.75, 0.88)
5 (Most deprived)	18 853 (64.36)	0.82 (0.77, 0.86)	0.86 (0.80, 0.93)	13 326 (70.68)	0.67 (0.62, 0.73)	0.79 (0.71, 0.87)	15 046 (51.36)	0.67 (0.63, 0.70)	0.77 (0.72, 0.83)
Unknown	37 (62.71)	0.76 (0.45, 1.29)	1.19 (0.62, 2.28)	16 (43.24)	0.21 (0.11, 0.41)	0.30 (0.14, 0.66)	17 (28.81)	0.26 (0.15, 0.45)	0.39 (0.20, 0.74)
Measures of clustering (random effects)
Practice level variance		1.06 (0.87, 1.29)	1.07 (0.88, 1.30)		0.93 (0.76, 1.13)	0.91 (0.74, 1.11)		0.85 (0.70, 1.03)	0.85 (0.70, 1.03)
Intra-cluster correlation		0.25 (0.19, 0.34)			0.21 (0.15, 0.28)			0.18 (0.13, 0.25)

^aInvitation rate is the proportion of eligible population who were invited for an NHSHC.

^bAdjusted for age, sex, ethnicity and deprivation, clustered by practice (52 clusters).

^cUptake is the proportion of invited people who attended NHSHC.

^dCoverage is the proportion of eligible people (regardless of whether they were invited or not) who attended NHSHC.

Fig. 1

Flow showing eligible, invited and attended for NHS Health Check.

Equitability of access

Sex equity

There was a statistically significant 20% reduction in unadjusted odds of invitation for male participants, persisting after adjustment for confounders (AOR: 0.78, 95% CI: 0.75–0.81) (Table 2 and Supplementary Fig. 2). Odds of uptake and coverage were also significantly lower for males compared with females in adjusted and unadjusted analyses (AOR: 0.73, 95% CI: 0.70–0.77 and AOR: 0.69, 95% CI: 0.66–0.71, respectively) (Table 2 and Supplementary Fig. 2).

Age equity

The lowest invitation level (63%) was observed in the older population (65–74 years); however, this group had the second highest attendance of all the age-groups (Table 2). After adjusting for confounders, the odds of invitation was 19% lower in oldest age group 65–74 compared with those aged 40–44 years, which is statistically significant at 5% level (AOR: 0.81, 95% CI: 0.75–0.86). However, when invited, participants in the older people age groups had significantly increased odds (72% increased odds) of attending (e.g. AOR: 1.72, 95% CI: 1.57–1.89 for 65–74 compared with 40–44). Similarly, participants in all the older age groups had a significantly higher odds of coverage compared with those aged 40–44 years (Table 2).

Ethnicity equity

Before adjustment, all non-white ethnicity groups had significantly lower odds of being invited compared to White ethnicity group; however, after adjustment, only Black (AOR: 0.85, 95% CI: 0.75–0.96) and Other ethnic groups (AOR: 0.74, 95% CI: 0.67–0.81) remained significant.

Compared with White, South Asian had significantly higher odds of uptake (AOR: 1.32, 95% CI: 1.19–1.45) and coverage (AOR: 1.18, 95% CI: 1.10–1.26). Black ethnicity group had significantly lower odds of invitation but significantly higher odds of uptake (AOR: 1.23, 95% CI: 1.04–1.45). The most consistent inequity was observed in the Other ethnicity category (consisting of mixed backgrounds, other Asians not from south Asians and other ethnic groups). After adjusting for confounders, the ‘Other’ ethnicity group had significantly lower odds for invitation (AOR: 0.74, 95% CI: 0.67–0.81), uptake (AOR: 0.86, 95% CI: 0.75–0.98) and coverage (AOR: 0.74, 95% CI: 0.68–0.81) (Table 2 and Supplementary Fig. 3C). Sensitivity analyses showed that removing ‘unknown’ ethnicity group had no impact on the overall result observed.

Socioeconomic (deprivation) equity

The odds of invitation, uptake and coverage were significantly lower for the most deprived group compared with least deprived, and there was a gradient observed with reduction in odds in each group as the level of deprivation increases (Fig. 2).

Fig. 2

Adjusted odds ratio by deprivation.

Invitation method

The most common method of invitation used was opportunistic (face-to-face), which was used in 63.8% of those invited (Supplementary Table 1). Both opportunistic and telephone invites were significantly associated with higher odds of uptake. Compared with invitation by letter, opportunistic had a 25-fold increase in odds of uptake (AOR: 25.31, 95% CI: 23.38–27.39) and telephone had over 3.5-fold increase (AOR: 3.60, 95% CI: 3.23–4.02).

After adjustment, odds of opportunistic invitation were significantly lower for male participants (AOR: 0.79, 95% CI: 0.75–0.83) but significantly higher for South Asian ethnicity (AOR: 1.19, 95% CI: 1.09–1.31) and oldest age group (e.g. AOR: 1.36, 95% CI: 1.25–1.49) (Supplementary Table 2). There was no significant difference in this method of invitation by deprivation.

Discussion

Main finding of this study

Over the 5-year study period, 66% of Walsall’s eligible population were offered (invited) NHSHC, uptake of those invitations was high at 74% and overall coverage was 55%. The study found that male participants and people from most deprived part of the communities had lower odds of invitation, uptake and coverage. The odds of invitation were also significantly lower for people from Black ethnic group. Compared with White, the ‘Other’ ethnicity group (mixed backgrounds, other Asians that are not South Asians and other ethnic groups) also had lower odds of invitation, uptake, and coverage.

This study found that younger age group 40–44 years had significantly lower odds of taking up an NHSHC invite compared to all the other age groups (odds were up to 72% higher for those aged 65–74 years). Opportunistic invitation had a 25-fold increase in odds of uptake compared with invite by letter.

Proportion of eligible population invited for NHSHC in Walsall was lower than Public Health’s England (PHE) 5-year national average of 71.8% for 2016/17 Q1–2020/21 Q4.⁶ The uptake and coverage in Walsall were considerably higher than PHE national average of 47% and 33%, respectively, although lower than the ambition of 75% set out at the start of the programme.¹⁴

What is already known on this topic

Previous studies have reported variations in attendance across general practice,^8,7 and this study also found variations in general practices for invitation (8.6–94.9%), uptake (33.5–97.9%) and coverage (10.8–93.7%) of NHSHC. However, the analysis showed around 25% of variation in invitation, 21% in uptake and 18% in coverage were attributable to potentially unmeasured practice-level factors. The literature on NHSHC attendance by age is unequivocal; almost all studies have found that older participants are more likely to attend,¹⁵ which is consistent with the findings in this study. Similar to other studies,^9,15–20 this study found that male participants had significantly lower odds of being invited and attending NHSHC. This is an area of concern especially as men are at greater risk of CVD.²¹

What this study adds

The picture from literature on attendance by deprivation has been mixed with some reporting higher attendance in the most deprived population,^10,17 or higher attendance in the least deprived^9,22 and others finding no difference.^7,19 The findings from this study show that the more deprived the area, the lower the odds of invitation, uptake and coverage.

The literature is also mixed regarding impact of ethnicity on access to NHSHC. Several studies^10,11,18,23 concur with the findings of this study that coverage was significantly higher among participants who identified as South Asians; however, other studies reported that there was no significant difference by ethnicity²² or that coverage was lower in South Asians.²⁴ This study also showed that participants from particular minority ethnic groups had significantly lower odds of invitation, uptake and coverage (Supplementary Fig. 3). The findings from this study add to the existing evidence that screening-type programmes may not be reaching aspects of ethnic minority communities and those from more deprived areas.^25,26

This study found opportunistic invitation was associated with higher odds of uptake, which is consistent with findings from an earlier study.¹⁵ Younger people aged 40–44 had the lowest odds of being invited using this approach, and this may partly explain the lower uptake in younger age group especially as this age-group are known to be less likely to visit GP.²⁷ Conversely, people aged 65–74 who are more likely to visit the GP²⁷ had the lowest odds of being invited. Further research is needed to understand these decreased odds of invitation despite the increase odds of opportunistic method being used for this age group. Contrary to finding from previous studies where it was found that the latter was the most common method of invitation used,¹⁵ the findings from this study indicate opportunistic invitation was the most used approach in Walsall. This may partly explain the higher-than-average uptake in Walsall as findings from this study, and other studies have shown non-written method of invitation such as face-to-face opportunistic invitations^9,28 and telephone invitations²⁹ were more effective for increasing uptake. However, given that certain groups such as men are less likely to attend GP,²⁷ it is possible that opportunistic invitation may also exacerbate inequality. In addition, this method of invitation can leave participants feeling forced and limits the opportunity for informed consent.³⁰ Furthermore, Cook et al. showed that effectiveness of different invitation approach differs by patient demographics,⁹ which concurs with an RCT study that found telephone invitation to only be particularly effective for younger patients.²⁹ The key message for the programme team in Walsall given the dominance of opportunistic invitation is to ensure there is adequate information and time to enable people to make an informed decision and also adopt broader recruitment approach that would tackle inequity of access to NHSHC.

One of the key objectives of NHSHC is the narrowing of health inequalities, and there needs to be equitable uptake and coverage across all parts of the population for this ambition to be realized. This study has highlighted some areas of inequities in provision of NHSHC. In addition to ineffective invitation methods, previous studies into why uptake may be low in some groups have also identified lack of interest and awareness about the programme,^20,31 lack of appreciation of the benefit³¹ and health-seeking barriers such as time constraint and wishing to avoid GP²⁸ as reasons for low attendance.

A further qualitative study in Walsall population would help identify local reasons for low attendance in males, younger people and those with mixed ethnicity. There is also potential to further utilize non-GP settings to increase access such as Walsall’s existing workplace programme³² and community outreach programme similar to the one in Greenwich,²⁰ combined with drop-in sessions outside working hours.²⁸ However, it is worth noting that a local study found that 7% of those invited through community outreach declined because they would prefer appointment through GP,²⁰ and another study found that concerns about competency of staff in non-GP settings may have hindered NHSHC attendance in those settings,³¹ which indicates a need for a clearer and more targeted messaging. Alternatively, two of the top tips from PHE are the use of financial incentives and behavior insights messages to targets priority groups; however, a double-blinded randomized study comparing behavior insights leaflets to standard leaflets found no significant difference in uptake.³³ In terms of financial incentives to patients and GPs, previous RCT study has found that offering financial incentives to patients did not increase uptake of NHSHC,²⁸ and a study assessing views of GPs reported that some GPs have expressed uneasiness from too much emphasis on finance for GPs for what is supposed to be fundamental to public health.³⁴

Limitations of this study

The level of data available meant this analytical study has been able to quantify invite, uptake and coverage of NHSHC and examined equity in the provision of NHSHC with logistic regression models that adjusted for key confounders. However, as this was an observational study, there will be residual confounders which have not been accounted for in the models.

Previous studies have reported poor data quality in relation to ethnicity, particularly among those who did not attend NHSHC. In this study, a key strength has been data completeness for ethnicity which was complete for 99.7% of attendees and 89.6% of non-attendees and as such has enabled analyses by ethnicity. However, small numbers in each ethnic category have restricted analyses to top-level categories.

Initial plans for this study included analyzing the data by broader characteristics such as carer status, serious mental illness and learning disability; however, low data completeness level for these variables meant this was not possible.

Finally, the primary data source for this study is Walsall population registered with GP; however, the study was not able to access or estimate the number of eligible patients who may not have been registered with a GP. This may have introduced bias as there might have been a systematic difference between those who are registered and those not registered with GP in terms of equitability of access to NHS Health Check.

Despite the above limitations, this is the first study evaluating NHSHC in Walsall. A key strength of this study is direct access to EMIS database enabling the study to include all eligible participants during the evaluation window, resulting in study population of around 61 500 participants. This study is also one of few studies that have examined impact of different invitation methods on uptake of NHSHC. Findings from this study may be relevant to other areas with similar demographics.

Conclusion

Over the 5-year study period, Walsall has achieved a high uptake of 74% among those invited compared with national average of 47% (2016/17 Q1–2020/21 Q4). The overall coverage of NHSHC in this study was 55% of Walsall’s eligible population compared with 33% nationally. Despite the high level of uptake and coverage, the study has highlighted some areas of inequities in NHSHC provision. The study found that men had lower odds of invitation, uptake and coverage. Similarly, those in the most deprived part of the population and those from particular minority ethnic groups had lower odds of invitation, uptake and coverage. People from Black ethnic group also had lower odds of invitation. Opportunistic and telephone invitations were associated with higher odds of uptake of NHSHC invites. Further actions are needed to reduce the inequities in invitation, uptake and coverage of NHSHC.

Summary Box

What is already known on this subject?

• The evidence for attendance at NHS Health Check varies considerably across different regions in England; however, there is consistent evidence that support attendance being higher among older patients and female patients. The evidence is however mixed for ethnicity and deprivation. There are also limited studies examining impact of different invitation methods on uptake.

What does this study add?

• This study found high level of uptake and coverage for NHS Health Check in Walsall; however, the study identified inequities in access to the service. Men had lower odds of invitation, uptake and coverage.

• Similarly, those in the most deprived part of the population and those from particular minority ethnic groups had lower odds of invitation, uptake and coverage.

• Opportunistic and telephone invitations were associated with higher odds of uptake.

Funding

The authors received no financial support for the research, authorship and/or publication of this article.

Patient involvement statement

Patients or the public were not involved in the design or conduct of the research.

Statement on ethical approval

The study was a service evaluation and therefore ethical approval was not required. However, approval to interrogate non-identifiable patient’s data on EMIS database was granted by the assigned committees in Walsall Clinical Commissioning Group and Walsall Local Authority.

Author’ contributions

FO was responsible for the research idea, study design, data extraction, data analysis and drafting of the manuscript. AS verified the statistical methods and provided critical feedback on the manuscript. NCL, DH and PM provided critical input into the development of research idea, methods and manuscript.

Transparency statement

The lead author affirms that this manuscript is an honest, accurate and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned have been explained.

Conflict of interest

None declared.

Data availability

No additional data available. Approval was granted to interrogate non-identifiable patient-level data and the data remain the property of the Walsall Clinical Commissioning Group and Walsall Local Authority.

Acknowledgements

The authors are grateful to Haniya Chaudhary and Sarah Smith for their advice on the EMIS dataset. This article presents independent research supported by the NIHR Birmingham Biomedical Research Centre at the University Hospitals Birmingham NHS Foundation Trust and the University of Birmingham. The views expressed are those of the author(s) and not necessarily those of the NHS, the NIHR or the Department of Health and Social Care.

Fatai Ogunlayi, Public Health Specialty Registrar

Nina Chauhan-Lall, Programme Development & Commissioning Manager

David Hughes, Public Health Technical Officer

Paulette Myers, Consultant in Public Health/Associate Director of Public Health

Alice Sitch, Senior Lecturer in Biostatistics