Summary of findings
This study showed that recruiting older people to a randomised-controlled study of WCH from primary care was not feasible due to several factors. These factors were: established approaches of primary care teams towards managing WCH in older people, target study patient demographics, study design complexity, patient-facing study documents, limited research resources in primary care which all made identification of eligible patients.
Established approaches of primary care teams towards managing WCH
The evidence for treating hypertension in older adults is well established5. However, treatment of the WCH phenotype remains controversial due to inadequate evidence in existing literature. Limited evidence for WCH in older people includes two post-hoc sub-group analyses of large trials showing benefits6,8,25. This contention could have led to established perspectives about WCH among primary care practitioners. Although more than 80% of respondents agreed that WCH was an important clinical issue and treatment was relevant to clinical practice, nearly half (42%) of the respondents had concerns about treating WCH with medications. There is a recognition among primary care physicians of the need for intervention in WCH management compared to the current status quo. Our findings are consistent with a study among 7263 GPs across Europe, which showed that nearly half (52%) of GPs treated WCH only in patients with high CVD risk. In their study, exploring the impact of WCH in primary care in Europe and Canada, the practice of non-treatment of WCH was the same (98%) across both study areas26.The approach to management of WCH remains unclear; while recommendations have been made for lifestyle modifications similar to the management of pre-hypertension, a more intensive approach using medications has recently been suggested27–30. Healthcare workers’ perceptions are known factors influencing recruitment of ‘vulnerable’ patients31. These include physicians’ personal biases, concerns of additional investigations and medications, and their views about patients’ functional status32.
Articles exploring the reasons for poor recruitment in RCTs have not fully explored the positionality and reflexivity of recruiters33–35. These factors are relevant because they emphasize recruiters’ personal knowledge and experiences, which ultimately impact on their attitudes toward the entire research process36–38. The dichotomy between a researcher’s interest and a recruiter’s view on the phenomena under study including scepticism about a study’s usefulness and misunderstanding the trial’s purpose, has undesirable bearings on the success of an RCT especially in areas which remain contentious such as WCH. Patients’ participation in trials is greatly influenced by their healthcare providers’ views about the study especially when recruitment occurs in primary care as in HYVET-239. This arises from how information would be shared with potential participants, thus raising the need for early exploration, addressing and incorporating care providers’ and recruiters’ beliefs and perceptions in study design and planning40–43.
Target patient demographics
Shenoy et al. recommend that the age limit for elderly care trials be moved from the standard 65 years and above, to 75 years due to improving life expectancy worldwide44, a view that informed the inclusion criteria of HYVET-213, though the original HYVET focused on those 80 years and over5. However, our findings showed that more than half of the respondents (>50%) were of the view that the age limit 75 years and older, should have been lower, allowing for more participants to be recruited. This viewpoint is justified by the notion that older adults in the 65-75-year age group are more likely to be interested in health promotion and disease prevention measures (e. g. medical management of WCH), compared to those older than 75 years45. Some of the study sites recommended a lower age as it was less likely that the participants would be taking medications for WCH, but in the very elderly age group many were already on antihypertensives with some sites treating WCH patients in this age group. Bourgeois et al reports that only 12.3% of adults 75 years and older have enrolled in CVD clinical trials compared with 42.5% of those older than 65 years46,47. Clearly, those 65 years and older may consider themselves to benefit more in this RCT compared to those in the 75 years and older category. Personal benefits and altruism are known to be the most important motivators for older adults’ participation in research- a view reported by Tolmie et al.48,49. Although the compromise of reducing the age inclusion criteria to 65 years could have helped in recruiting more patients to HYVET-2, it would defeat the aim of the study, the focus of which was on the very elderly, since it is a sequel to the original HYVET study. There is a rising advocacy globally for the definition of older age to begin at 70 or 75 years instead of 6550,51, with few limited studies in the literature exploring WCH in older adults using less than 75 year as an inclusion criterion52–56.
With an older adult study population in research, there are additional challenges such as co-morbidities and frailty. Functional status and demographics of patients are determinants for recruitment and retention in research in primary care32. Our study showed that more than 50% of respondents considered the level of frailty and existing co-morbidities among eligible participants to be too high, making potential participants unwilling to participate. This finding is consistent with 46% of older adults with neurologic and cardiovascular diseases refusing participation in trials57. Although the HYVET-2 study protocol did not consider frailty status as part of its exclusion criteria, recruiters felt this was important especially with the age group under-study13. Number of medications already being taken, mobility challenges, reliance on others due to frail status of potential participants were cited as inhibitors to willingness to participate.
Study design complexity
The study complexity, with 9 study visits and extra medications were factors considered onerous for participants and their carers by nearly 57% of respondents, although this is the view solely from the research team members and not potential participants. As an interventional trial, the introduction of another medication leading to probable polypharmacy among potential participants is germane. Several patients might not want to take additional medications to those they are taking for their chronic conditions. Study visit challenges could be addressed by providing means of transport, payment of travel tickets and other incentivisation strategies to boost study recruitment16,48,49,58–61. Alternatively, home visits could have been used to reduce the burden of numerous study visits.
Patient-facing documents
Nearly half of the respondents (47%) observed that the information for potential participants in the PIS was lengthy and the wording could have been clearer and easier to understand. Communication is critical to successful recruitment in research. This includes, but is not limited to, how consent is sought, the PIS, use of open/closed ended questions and medical jargon in patient-facing documents15,62. For a feasible trial for managing WCH in older people, an early consultation and collaboration withholder patients and their families/ carers would be essential. This would bring to the fore the perceptions, preferences, and treatment options of potential participants for incorporation into the study design.
Limited resources in primary care for research
The challenges of recruitment in primary care settings may be more than those encountered in secondary/tertiary health centres especially during COVID times. The success of the original HYVET study5 and other blood pressure studies such as SPRINT63 could be attributable to study settings which were multi-national, multi-centre and secondary/tertiary-facility based as compared to the three-county, multi-centre, primary care settings of HYVET-2. There is more support and resources available at secondary/tertiary levels of healthcare compared with the resources and demands on staff in primary care.
Identification of eligible participants
Recruitment in RCTs across multiple centres is generally slower and more difficult than expected, due to the differences in research infrastructure and human resource across different sites33. The challenge becomes pronounced in complex trials which require additional resources. Although respondents disagreed that the study was overly complex and ambitious for their primary care settings, identifying potential participants with WCH from their clinical databases, was found to be difficult at nearly 70% of study sites. Central to recruitment is the ability to identify potential participants. Different clinical database systems and application software are used across primary care settings in the UK. This lack of uniformity makes identification of potential participants difficult especially if there are no clinical codes for WCH in software applications. This situation could lead to a decline in physician commitment since recording a diagnosis and ultimately identification of potential participants becomes difficult 57,64. Data quality within electronic records of a practice was also highlighted as an issue. Different GPs in the same practice might record data differently. Again, it was assumed that ABPM data (searching electronic ABPM records) would be the primary source of participant identification. However, most sites did not store ABPM data in a searchable way. Although differences in practice settings were contributory factors for recruitment success, the crux remains whether primary care practitioners are diagnosing WCH in their practice irrespective of the application software systems used. Humbert et al. reports that across Europe, 49.3% of GPs diagnosed WCH and 52.1% were unaware of any guidelines for WCH26.
Study Limitations
An interview between study participants and members of the research team would have provided a more in-depth data than what was captured in online questionnaire. However, the emergence of the COVID-19 pandemic disrupted the data-collection process and the four staff who initially expressed interest in being interviewed later declined due to redeployment and additional work commitments.
Recommendations for future research and policy
Due to challenges peculiar to older adults such as multi-morbidity, frailty, reduced mobility, polypharmacy among others20,21, RCTs in older age groups need careful planning, interaction with PPI, incorporating potential participants’ views, with significant attention given to research infrastructure at various multi-centre settings. It is recommended that future WCH research should adopt fewer hospital study visits.
Future research into WCH must first identify, through a scoping survey, geographical areas in the UK and worldwide with known prevalence of WCH for delivering feasible RCTs. An existing database of WCH patients should be available to make identification of potential participants easier.
With regards to policy, applications used across various primary care facilities should have specific coding for WCH. This approach should be coupled with clarity in guidelines concerning WCH diagnosis. There is a need to identify whether clinicians are diagnosing WCH in their practice and how they manage it. This knowledge would inform standardisation of practice, new guidelines, and future research. Furthermore, older patients with more co-morbidities and frailty should be supported to participate in research.