Hypertension, or high blood pressure (BP), is a significant public health concern; it is the main risk factor for stroke and a major risk factor for cardiovascular disease (CVD) (Murray and Lopez, 1997). An American survey showed that 31% of people with the condition were unaware of it (Wang and Wang, 2004), a figure likely to be similar in the UK; moreover, many with diagnosed hypertension have poorly controlled BP (Berlowitz et al, 1998). Yet it is the risk factor most amenable to treatment—and as such presents an ideal focus for public health promotion.
The traditional role of the ambulance service has been to concentrate on providing a timely response to those requiring urgent medical care, stabilisation of the patient's condition, and efficient transfer to hospital for definitive treatment. However, this role has evolved over the last decade or so, and ambulance service personnel are now equipped with a range of skills that can be applied in a wide variety of situations. A Department of Health (DH) publication in 2005 stated that ambulance services in England should ‘provide an increasing range of other services’, including health promotion and the prevention of emergencies (DH, 2005).
The Stroke Association website states that 40% of strokes could be prevented by appropriate control of hypertension and for this reason promotes an on-going campaign called ‘know your blood pressure’ (KYBP) (Stroke Association, 2012). In 2010, it partnered the Ambulance Service Network through the College of Paramedics in organising public events in 12 trust regions. As part of this campaign, clinicians from the London Ambulance Service NHS Trust (LAS) offered free BP testing in Greater London in the April of 2010.
This paper aims to ascertain the suitability of the ambulance service for participating in such events by assessing whether attending members of the public who were advised to seek medical attention for high BP subsequently did so, and whether treatments were initiated or changed as a result. It also examines whether event attendance increased knowledge of the risks associated with high blood pressure, and resulted in any behavioural or attitudinal changes.
Methods
Study design
A telephone survey was conducted to evaluate the effectiveness of the KYBP public health event. An attempt was made to survey all participants who had a high BP measurement (≥140/90 or ≥ 130/80 if the participant had diabetes) and were referred by LAS staff for further medical attention, together with a 20% random sample of those who were not referred. The random sample was generated by selecting every fifth non-referred participant.
Setting
Greater London covers a geographic area of 1579 km ² (609 miles ²) and is the most populous municipality in the EU with a population of 7.8 million people (representing 13% of the UK total). It is highly ethnically and linguistically diverse, with 36% of London residents identifying themselves as from a Black, Asian or other ethnic group compared to just 11% of people in the UK (Office for National Statistics, 2007), with over 300 different spoken languages (Baker and Everseley, 2000). London is served by the LAS—a single, publicly funded ambulance service
The event
The KYBP event was held on the 17 April 2010 at a total of 19 venues across Greater London including markets, outside supermarkets and mainline railway stations. Event locations were selected specifically to target ethnic minorities and older populations (and thus those thought to be most at risk of high BP). Events were promoted and advertised through local newspapers, the LAS website, and the Stroke Association website.
At each venue, members of the public were offered free BP testing by LAS clinicians, a demonstration of the ‘face arm speech test’ (FAST) that is used for identifying a stroke, and a leafet outlining the risks associated with high BP and the benefits of a healthy lifestyle. BP was assessed as a single measurement using a digital sphygmomanometer and in accordance with the European Society of Hypertension Guidelines (Guidelines Committee, 2003).
Participants who had a high BP measurement were referred for further medical attention following the clinical guidelines provided by the Stroke Association/Ambulance KYBP project. The type of referral offered depended upon the BP level: a ‘routine GP referral’ was advised for those whose BP ranged from 140/90 to 179/109 or over 130/80 if the participant had diabetes; an ‘urgent GP referral’ if BP was greater than 180/110, and ‘hospital referral’ if BP was greater than 180/110 with any associated symptoms, namely headache, retinal haemorrhage, palpitations, papilloedema, pallor or diaphoresis. Those given routine or urgent GP referrals were advised to contact their GP surgery for a follow-up appointment within one week or 24 hours respectively.
Hospital referrals were sent directly to hospital for immediate medical attention. LAS clinicians in attendance at each venue recorded, for each participant, BP measurements and the type of referral advised—using a data collection sheet, slightly amended from that suggested by the KYBP campaign, to store additional items of information needed for the evaluation.
All event participants were asked to complete a consent form before having their BP measured, through which they gave permission for the use of their data and indicated their willingness to be contacted at a later date for a telephone interview. Identifiable participant data was held in password-protected files and locked fling cabinets only accessible to the research team.
Telephone surveys
A clinically qualified member of LAS staff attempted to contact all consenting participants who had had a high BP and therefore received a referral at the event, as well as the 20% random sample of those who had a normal BP. Telephone respondents were asked to answer a number of questions incorporating a variety of multiple-choice, scaled and open-ended formats designed to determine effectiveness of the campaign. Participants who were not referred at the event were asked questions on personal medical history, basic demographics, what their major take home message was from the event, and whether participation at the event had increased their knowledge regarding BP.
Referred participants were asked additional questions about whether they had contacted their GP surgery as advised, the outcome of the referral, and whether event attendance had influenced their behaviour and attitude towards B P. Answers recorded for the open-ended questions were later categorised for analysis purposes. Attempts were made to contact participants up to a maximum of five times after which they were excluded from this phase of the study if these were unsuccessful. Participants were surveyed an average of 53 (± 23) days after the event.
Data analysis
All continuous data were analysed using the Shapiro-Wilk test to determine normality. The Mann-Whitney and Pearson's Chi-Square tests were used, where appropriate, to identify differences in the demographics of participants that were referred and not referred. A P value of < 0.05 was considered significant. Statistical analyses were performed using SPSS 18.0 (SPSS Inc, 2009; Chicago, IL, USA).
Results
Overall, 2 274 members of the public attended the KYBP event in London and had their BP measured by the LAS. Overall, 522 (23.0%) of event participants had a high BP measurement and were referred for further medical attention. Of these, 471 (90%) had a routine referral, 46 (9%) had an urgent referral, and five (1%) were referred immediately to hospital.
In total, 28% (n=625) of event participants contributed to the telephone survey, from whom more detailed knowledge is therefore available. This sample represented 57% of those who were referred for further medical attention (n=299/522) and 19% who did not require referral (n=326/1718). The remaining 43% (n=223/522) of referred participants did not contribute to the survey largely because they could not be contacted, although a small number refused.
Of the 299 survey respondents who received a referral, 91% (n=273) were given a ‘routine’ referral (i.e. requiring a GP appointment within one week), 8% (n=24) an ‘urgent’ referral (advised to contact their GP surgery within 24 hours), and 1% (n=2) were advised to attend hospital immediately.
Given the small sample sizes of some referral groups and for ease of comparison, in some instances routine, urgent and hospital referral groups have been combined and are collectively described as ‘referrals’.
68% (n=204) of ‘referred’ survey respondents were unaware of any possible blood pressure problem. The remaining 30% (n=89) had already been diagnosed with hypertension by their general practitioners (GPs).
Comparison of characteristics between referred and non-referred surveyed participants
As presented in Table 1, ‘referred’ participants were significantly older than those who were not ‘referred’ (means=56 vs 52 years; U(614)=35220, Z=-5.38, P < 0.001). Ethnicity was found to be associated with referral status (X2(4)=11.20, n=592, P< 0.005).
| All surveyed n=625 | Referrals n=299 | P value | |
|---|---|---|---|
| Age (years) | <0.001 | ||
| Median | 52 | 56 | |
| Range | 13–91 | 17–91 | |
| Gender | 0.218 | ||
| Male (%) | 307 (50) | 154 (50) | |
| Female (%) | 314 (50) | 142 (45) | |
| Missing | 4 | 3 | |
| Ethnicity | 0.024 | ||
| White (%) | 353 (60) | 184 (52) | |
| Black African/Caribbean (%) | 82 (14) | 37 (45) | |
| South Asian* (%) | 71 (12) | 30 (42) | |
| Other Asian (including Chinese) (%) | 50 (8) | 15 (30) | |
| Other (inc Mixed) (%) | 36 (6) | 14 (39) | |
| Missing | 19 | ||
| Smoking status | 0.316 | ||
| Smokers (%) | 78 (13) | 33 (42) | |
| Non-smokers (%) | 525 (87) | 254 (48) | |
| Missing | 22 | 12 | |
| Regular exercise † | |||
| Yes (%) | 336 (56) | 28 | |
| No (%) | 261 (44) | 127 (49) | |
| Missing | 28 | 17 | |
| Diagnosed hypertension | <0.001 | ||
| Yes (%) | 142 (23) | 89 (63) | |
| No (%) | 469 (77) | 204 (44) | |
| Missing | 14 | 6 | |
| Personal medical history | <0.001 | ||
| Stroke/TIA (%) | 10 (2) | 7 (70) | |
| Diabetes (%) | 31 (5) | 26 (84) | |
| Heart attack (%) | 4 (1) | 4 (100) | |
| Angina (%) | 13 (2) | 5 (39) | |
| Two or more of the above (%) | 10 (2) | 6 (60) | |
| None(%) | 530 (89) | 237 (45) | |
| Don't know/ Missing | 27 | 14 (52) |
Key: Percentages in the referred column indicate the percentage of participants in each category who were referred.
'south Asian’ includes Indian, Pakistani and Bangladeshi.
Defined as at least 30 min per week. TIA=transient ischemic attack. ‘Missing’ denotes no answer unless otherwise specified.
NB. Due to rounding up, percentages may not add up to 100.
A higher proportion (52%) of ‘referred’ patients who could be surveyed by telephone were white, compared with 45% to 30% of other ethnic groups. Little difference was noted in relation to smoking status or exercise habits. Of surveyed patients referred for raised blood pressure, 68% had no previous diagnosis of hypertension.
In total, 26 of the 31 individuals with known diabetes were found to have a raised blood pressure using the low threshold set in the protocol of 130 mm Hg systolic pressure for this condition. In response to the question relating to known diagnoses of cardiovascular disease, previous stroke, or diabetes, 45% of those with no such history were referred.
Did referred participants seek further medical attention as advised?
More than half of surveyed ‘referred’ participants (56%) contacted their GP's surgery as advised.
Almost all did so within four weeks (Table 2). When broken down by referral type, 42% of urgent referrals were followed-up within one week, compared to 24.5% of routine referrals. The 81 individuals who did not follow LAS advice to contact their GP reported this was because they did not consider their BP to be high, did not have the time, or were already self-monitoring their BP.
| Referrals n=299 | |
|---|---|
| Follow-up at GP surgery | |
| Yes (%) | 166 (56) |
| No (%) | 81 (27) |
| NA* | 2 |
| Missing | 50 |
| Follow-up within (time) † | |
| 1 week (%) | 77 (46) |
| 2 weeks (%) | 52 (31) |
| 3 weeks (%) | 7 (4) |
| 4 weeks (%) | 13 (8) |
| More than 4 weeks (%) | 8 (5) |
| Missing/Can't remember | 9 |
| Actions taken since campaign | |
| Lifestyle changes (%) | 66 (22) |
| Medication‡ (%) | 41 (14) |
| Regular BP checks (%) | 13 (4) |
| More than one of the above (%) | 4 (1) |
| None (%) | 79 (26) |
| Missing | 96 |
Note: *Not applicable as these patients were taken to hospital for immediate medical attention.
†n=166 (participants who followed-up).
‡Adherence to a medical antihypertensive prescription. GP=general practitioner; BP=blood pressure.
‘Missing’ denotes no answer unless otherwise specified. NB. Due to rounding, percentages may not add up to 100.
Referral outcomes—routine and urgent
An outcome was known for 154 of the 166 individuals who contacted their GPs. A high BP was confirmed in 55, and in at least 11 instances antihypertensive medication was prescribed for the first time. Others may have been treated subsequently because 59 of those not on medication at the time of the enquiry were asked to return for a follow-up or were referred to hospital for additional checks.
More than half of all survey respondents who reported they had been diagnosed with hypertension prior to the event (89/142) were referred for a routine or urgent GP follow-up by the LAS (Table 3), despite 83 of this group stating they were currently on antihypertensive medication. Some responses are not available, but at least 14 individuals had an increase in dose or an alternative antihypertensive agent prescribed as a result of the referrals.
| Referrals with previously diagnosed hypertension n=89 | |
|---|---|
| BP confirmed at GP surgery * | |
| Yes (%) | 32 (53) |
| No (%) | 25 (42) |
| Missing | 3 |
| Gender | |
| Male (%) | 45 (51) |
| Female (%) | 43 (48) |
| Missing | 1 |
| Smoking status | |
| Yes (%) | 10 (11) |
| No (%) | 78 (88) |
| Missing | 1 |
| Regular exercise † | |
| Yes (%) | 49 (55) |
| No (%) | 37 (42) |
| Missing | 3 |
| Reason for participation | |
| History of BP (%) | 55 (62) |
| Curiosity (%) | 16 (18) |
| Self-concern (%) | 7 (8) |
| Friend/family concern (%) | 3 (3) |
| Other (%) | 8 (9.0) |
| Actions taken since campaign | |
| Medication‡ (%) | 30 (34) |
| Lifestyle changes (%) | 15 (17) |
| Regular BP checks (%) | 1 (0) |
| None (%) | 22 (25) |
| Missing | 21 |
Note: *n=60 (participants who followed-up).
†Defined as at least 30 min per week.
‡Adherence to a medical antihypertensive prescription. BP=blood pressure; ‘missing’ denotes no answer unless otherwise specified.
NB. Due to rounding, percentages may not add up to 100
An additional test to check cholesterol levels was undertaken on 68 of 166 participants who had contacted their GPs after the event. Of these, 56 individuals knew their test results and 21 reported that it was found to be high.
Referral outcomes—hospital
Both of the surveyed participants who were referred to hospital directly from the event were confirmed as having high BPs. By the time of the telephone survey, one of these patients had been prescribed antihypertensive medication, and the other had been diagnosed with suspected angina.
Influence on attitude and behaviour
Referred participants were asked if they had taken any actions to lower their BP as a result of the campaign and 42% (n=125/299) answered ‘yes’ to this question.
Of these, 66 said they had made lifestyle changes (e.g. started exercising or increasing its frequency, stress management) while 30 individuals already diagnosed with hypertension stated that they were now adhering to their antihypertensive prescription (Table 2 and 3).
Impact on knowledge and awareness
The majority (73%, n=103/141) of surveyed participants who said they had ‘poor’ knowledge of high BP and the associated health risks (including stroke) before the event, said that their knowledge and awareness of the risks of high BP had increased to either ‘medium’ or ‘high’ as a result of the campaign.
No single participant of the study reported a decrease in their level of knowledge.
Perceived benefits of attending the event
Nearly half of surveyed participants (45%, n=283/625) took part in the campaign because of ‘curiosity’ (Table 4). When asked which aspect of the campaign was most enjoyable or beneficial, the two most popular answers were ‘information and advice’ (n=199) and ‘checking BP’ (n=203). Many people (n=102) commented on the ‘friendliness of LAS staff’.
| All surveyed (n=625) | |
|---|---|
| Reason for participation | |
| Curiosity | 283 (45) |
| History of BP | 123 (20) |
| Self concern | 81 (13) |
| Friend/family concern | 33 (5) |
| Family history of BP | 26 (4) |
| Other | 68 (11) |
| Missing | 11 |
| Aspect most enjoyed/beneficial | |
| Information and advice | 199 (32) |
| Checking BP | 203 (33) |
| Friendliness of EMS staff | 102 (16) |
| Identifying medical problems | 14 (2) |
| Other | 6 (1) |
| Don't know/not sure/missing | 101 |
| Major take home message | |
| High BP and health risks including stroke | 157 (25) |
| Stroke symptoms/FAST test | 68 (11) |
| What BP is/importance of regular check-ups | 86 (14) |
| The effect of lifestyle* on BP | 26 (4) |
| Everything (all of the above) | 53 (8) |
| Other | 40 (6) |
| Don't know/not sure | 150 (24) |
| Missing | 45 |
Note: *Includes diet, exercise and stress levels. BP=blood pressure; EMS=emergency medical staff; FAST=face, arms, speech test. ‘Missing’ denotes no answer unless otherwise specified.
NB. Due to rounding, percentages may not add up to 100
Discussion
The estimate by the Stroke Association, that around 40% of stokes could be prevented if all patients with hypertension were treated effectively, provides powerful motivation for seeking out those with the condition who are unware of their risk. Unfortunately too little is done to remedy the situation. GPs and pharmacies make BP checks readily available but many who are not health-conscious do not benefit from these excellent initiatives. Taking the facility to places frequented by the general population offers another complementary approach that the ambulance service is well placed to offer.
Paramedics are healthcare professionals well used to dealing with individuals in their own surroundings; they are also widely respected and trusted. Results from the KYBP campaign set up by the Stroke Association in cooperation with the College of Paramedics are therefore of importance. This study presents a snapshot from London of a national effort that also involved 11 other ambulance trusts. The successes that are reported will have been replicated many times over and confirm a role for a truly novel approach.
The main findings from this evaluation demonstrate that the KYBP one-day event, conducted by the LAS, was effective in terms of identifying high BP among members of the public and in making people take high BP seriously as a health concern. It illustrates the role of the ambulance service in preventative health promotion and provides an example of the way in which ambulance services may have functions that are additional to those of emergency medical care. Many participants, who attended the event and responded to the survey, sought further medical attention after being advised to do so by the LAS. The majority also claimed to have taken positive steps towards controlling their BPs as a direct result of the event, for example by increasing exercise frequency, and better compliance with their medical prescriptions.
The reported increase in the knowledge and awareness of high BP following the event further indicates the campaign's success, as does the finding that the high BP measurements taken by the LAS were confirmed by GPs in over a third of referred individuals. Furthermore, these measurements prompted many GPs to undertake further tests and assessments (including cholesterol testing).
Almost all of those surveyed said they would be happy to participate again if similar events were to be held in the future. This positive feedback suggests that ambulance services can and should be involved in similar future events.
The full impact of the KYBP initiative is impossible to measure accurately, even using the system of telephone follow up adopted by LAS. If one considers those surveyed who had medical checks after the initial arranged referral, those whose BP was treated for the first time, those found to have undesirable cholesterol levels, and all who claimed to have made a change in life style, then few would deny that the initiative had been a success. But only a minority of those who attended the event could be surveyed, indicating that the impact was likely to have been appreciably greater than these numbers suggest—and from only one of the ambulance trusts that participated.
‘there remain a continuing need to increase public awareness about the importance of having regular BP checks’
There remains a continuing need to increase public awareness about the importance of having regular BP checks and the risks associated with high levels. This is especially important as hypertension is not usually associated with any obvious symptoms. Before the event, almost four-fifths of surveyed participants felt that their knowledge of the health risks of high BP was ‘low or medium’.
Undiagnosed hypertension carries a number of risks including stroke, vascular disease, and kidney disease. Diabetic patients with undiagnosed hypertension are at even greater risk of developing serious problems (Maahs et al, 2005). Our evaluation revealed that 83.9% of participants with diabetes were found to have high BP on the day of the event, 40.0% of whom were not already diagnosed with hypertension. These facts highlight the critical need to explore all reasonable ways of increasing public awareness of the health implications of high BP.
Hypertension is not the only readily treatable risk factor for stroke. The most important by far of the others is atrial fibrillation (AF) that is particularly common in the elderly, increasing the overall risk five-fold (Wolf et al, 1991). Almost 10% of men in the UK over 80 years-of-age are known to have AF (DeWilde et al, 2006), but as with raised blood pressure, many of those who are affected have no symptoms and are unaware of their condition. For this reason, a discretionary second part was added to the Stroke Association/ambulance protocol. It involved an electrocardiogram (ECG) for those with irregular pulses, a requirement considerably more demanding than simple blood pressure measurement. Of the patients who participated in the telephone survey, only ten reported having an ECG. It remains an important option, however, if the initiative is repeated and resources can be made available.
Two findings were unexpected. The first was that nearly two thirds of event attendees who had previously been diagnosed with hypertension required referral, many of whom subsequently had their treatment modifed. A number of other studies have raised the issue of poorly controlled hypertension, with factors including poor medication compliance, inadequate patient understanding of risk, and unsatisfactory treatment targets (Berlowitz et al, 1998; Knight et al, 2001).
Innovative approaches of healthcare integration might aid some of the problems currently associated with compliance and patient awareness (Carter, 2010; Chisholm-Burns et al, 2010).
Use of the ambulance service in campaigns such as KYBP is one step towards this goal. The second unexpected finding was the greater percentage of ‘referrals’ in white patients compared with other ethnic groups, at least among those surveyed. As only a minority of referred patients could be contacted by telephone, an element of bias could be at least partly responsible.
Limitations
There were a number of limitations to the study. Firstly, hypertension cannot be diagnosed from one blood pressure reading. We acknowledge that it is good practice to obtain a second reading should the first reading obtained be high. However, in this study, only one reading was obtained as we followed the protocol drawn up by the Stroke Association. Basing the decision to refer participants for further medical attention on one reading increases the risk that unnecessary anxiety may have been generated for participants for whom a second reading might have been within the normal range.
Secondly, the survey gave information of only short term results; we do not know if changes in behaviour were sustained or whether the campaign had only a limited impact on BP awareness. Future campaign evaluations would benefit from longer term follow-up of event attendees.
Finally, the cost-effectiveness of this campaign also remains unknown—while the campaign did seem to have a number of positive outcomes, a careful analysis of cost against benefit would be very useful in determining the real value of ambulance participation in KYBP campaigns. An accurate assessment would not be easy, however, because the cost of strokes is large and may continue in some cases over several years