Mapping Irish Primary Care Research Across the Cancer Continuum: A Scoping Review and Bibliometric Analysis

ABSTRACT

Background

The scope of primary care cancer research on the island of Ireland has not been clearly mapped, despite primary care's pivotal role across the cancer care continuum.

Methods

We conducted a systematic search of studies from the Republic of Ireland (RoI) and Northern Ireland (NI) up until October 2022 and summarized the literature using bibliometric analysis and narrative synthesis. From 3911 screened abstracts, 100 eligible outputs were identified: 60 journal articles, 38 conference abstracts, and 2 doctoral theses. Most papers (92%) were published after 2000.

Results

Research output from NI and the RoI was proportional to population size; only one cross‐border study was identified. Study designs were predominantly comparative observational (51%), with fewer non‐comparative observational (23%), qualitative (23%), and experimental studies (3%). The most frequently studied cancers were cervical, breast, prostate, lung, and colorectal.

Conclusion

This review provides a foundation for strategically directing future primary care cancer research in Ireland.

1. Introduction

1.1. Background

Cancer is a significant public health issue in the developed world, with approximately 36,000 invasive cancers diagnosed annually in Ireland between 2018 and 2020 [1]. General practitioners are the initial point of contact for the majority of patients presenting with symptoms of cancer [2], with primary care providers playing a key role before and after diagnosis, in prevention, screening, and early detection as well as survivorship and end of life care [3].

The integration of research in the continuous improvement of cancer care is recognized by the Irish Department of Health's National Cancer Strategy (2017–2026) recommendation that “clinical cancer research, and the staff who deliver it, become a fully integrated component of cancer care delivery” [4]. However, internationally it has been claimed that primary care research tends to be underfunded relative to secondary care research [5, 6, 7, 8, 9, 10, 11].

1.2. Rationale, Aims, and Objectives

In addressing this dichotomy—the importance of primary care to cancer care delivery but its relative underrepresentation in research funding—it is important to understand the current literature and where gaps exist. A 2022 analysis of general practice‐based cancer research internationally reported the distribution of research according to year, country, study design, cancer continuum stage, and cancer type, finding that such research is predominantly observational, stable across time but varying significantly in terms of per capita output between countries. However, data for Ireland was not reported separately.

Thus, the main objective of our study was to systematically map all primary care focused cancer research which has previously been conducted in both the Republic of Ireland and Northern Ireland, so as to guide and support future work. A secondary objective was to narratively synthesize the identified literature so as to better understand the drivers of primary care cancer research across the island and therefore to develop a greater awareness of potential underserved areas requiring further research.

2. Methods

We conducted a systematic search, bibliometric analysis, and narrative synthesis of primary care focused research from the island of Ireland. The scope of this review was to map and synthesize research activity rather than compare the health systems of the Republic of Ireland and Northern Ireland, although system differences are noted where they provide useful context.

2.1. Geographical Scope and Healthcare Context

International readers should note that the island of Ireland is divided between two countries: “Ireland,” the southernmost five‐sixths of the island, and “Northern Ireland,” the northernmost one‐sixth. The former is a sovereign state, often unofficially referred to as the “Republic of Ireland,” so as to distinguish it from the latter, which is one of four nations in the United Kingdom. For clarity, we will use the terms Republic of Ireland (RoI) and Northern Ireland (NI). The term “Irish” is unavoidably ambiguous as it is the adjectival form of the Republic of Ireland (e.g., the “Irish government,” as opposed to “Northern Irish”), but is often used to pertain to the island of Ireland as a whole; for clarity we will use it only in the former sense, and only when demanded by existing terminology.

The important distinction, for our purposes, is that the Northern Irish health system is well‐known internationally since it is part of the UK's National Health Service (NHS), wherein care is provided free to all at the point of access. However, the health system in the RoI is more complex.

Primary care in the RoI is characterized by mixed public/private funding—at the time of writing, GP visits are free for only 40% of residents (roughly the poorest one‐third of society, under 6s, over 70s and those with eligible medical conditions) with either a “medical card” or a “GP visit card” [12, 13]. The remainder pay €53 on average to visit the GP [12]. Population‐based cancer screening programs are offered for breast, cervical, and colorectal cancer [12]. Universal free access to cancer treatment is delivered by the Health Service Executive [14, 15], however, private healthcare also plays a role in the provision of cancer diagnostics and cancer treatment for those who can afford it [12].

2.2. Eligibility Criteria

Peer‐reviewed primary research in the form of journal articles, conference abstracts, dissertation theses, and study protocols (including clinical trial protocols) were eligible for inclusion. To be eligible, the article needed to report original research or a synthesis thereof; hence, editorials, letters to the editor, and reports by healthcare organizations (as opposed to research performed by these organizations) were not included. Review articles were deemed eligible only if they were conducted by a research group from the island of Ireland; however, all reviews were examined for eligible articles. Only English language articles were eligible for inclusion. All eligible studies, irrespective of quality, were included in the descriptive analysis.

Finally, to be eligible for inclusion, research needed to correspond thematically to three themes: cancer, primary care, and Ireland. In keeping with Milley et al.'s review, “studies where cancer was an incidental result and not related to the research question” were excluded [16]. Primary care was defined as broadly as possible as “healthcare services that you find in your community, outside of the hospital setting” [17], hence research pertaining to the work of all community‐based healthcare professionals—GPs, practice nurses, public health nurses, community pharmacists, community dentists—was eligible for inclusion. Research was eligible if the study subjects from the Republic of Ireland or Northern Ireland were not aggregated with a country outside of the island of Ireland.

2.3. Search Strategy

Searches of the following electronic databases and trial registries were conducted on the 15th November 2022: MEDLINE, Embase, CINAHL (Cumulative Index to Nursing and Allied Health Literature), Web of Science, Cochrane Library, ProQuest Dissertation and Thesis Database, ClinicalTrials.gov, and the WHO International Clinical Trials Registry Platform. The search strategy used three themes, “cancer,” “primary care,” and “Ireland,” combined using the “AND” operator. A full search strategy is included in Supporting Information 1.

2.4. Study Selection

After deduplication, the title and abstract of all identified papers were screened for eligibility independently by at least two authors. All conflicts were resolved by discussion between four authors (A.A., R.A., B.M.J., and P.R.). Where the full text of an eligible paper could not be sourced, despite contacting the authors and consulting with library staff, this was reported as a non‐retrieved paper.

2.5. Data Collection

Following full‐text review to confirm inclusion, data extraction was undertaken by two authors using the labeling facility provided by the Rayyan systematic review tool using the following headings: country, study design, cancer continuum stage, and cancer type. The year of the publication was automatically collected in Rayyan.

2.6. Data Items

Country was designated as either “RoI,” “NI,” or “Both.” Cancer continuum stage was designated as one of “Prevention,” “Screening,” “Early Detection & Diagnosis,” “Treatment,” “Survivorship,” “Palliative,” “All”. We also utilized the category, “Risk,” which formed part of Milley's schema if a paper involved the determination or management of cancer risk in the absence of a clear application (e.g., screening or early detection).

Milley et al. categorized non‐review evidence under three broad headings: “observational,” “interventional,” and “other” [16]. We split this final category into “Observational (non‐comparative/NOS)” and “Qualitative,” a breakdown which could be gleaned from the Australian review.

To account for publications comprising multiple study designs, we assigned each study design within a publication an equal weighting, such that the weightings summed to 1, to prevent overcounting of mixed methods publications. For example, if a publication contained a qualitative component and a cross‐sectional study, we would represent this as a qualitative study with a weighting of 0.5 and an observational comparative study with a weighting of 0.5. The same approach was employed for publications focusing on multiple cancer continuum stages and multiple cancer types.

3. Results

4709 publications were initially identified, yielding 3911 unique publications following deduplication; 101 studies were included in the final review (see Figure 1: PRISMA flow diagram). We identified 100 publications in total: 60 journal articles, 38 conference abstracts, and 2 dissertation theses. No eligible systematic reviews, meta‐analyses, or clinical trial protocols were identified.

FIGURE 1.

PRISMA flow diagram.

The most common journals for publication were the Irish Journal of Medical Science (n = 18, 18%) and the Irish Medical Journal (n = 6, 6%). In total, eight journals published two or more journal articles included in this review, while 33 journals published a single article. The six primary‐care journals to feature in this review were “BMC Family Practice,” “BMC Primary Care,” the “European Journal of General Practice,” the “British Journal of General Practice,” “Family Practice,” and the “Scandinavian Journal of Primary Health Care.”

3.1. Bibliometric Analysis

65% of identified journal articles were published in the period 2010–2022, compared with 35% in the period 1980–2009 (see Figure 2: the number of papers published per year).

FIGURE 2.

The number of papers published per year.

21 publications (17 journal articles and four conference abstracts) described Northern Irish research, while 77 publications (42 journal articles, two doctoral theses and 33 conference abstracts) described research in the RoI. This breakdown of journal articles corresponded to the populations of both nations. Two cross‐border studies were identified.

Studies were primarily observational, with comparative designs being utilized in 62% of journal articles, after adjusting for publications comprising multiple studies. Qualitative studies and non‐comparative study designs were utilized in 19% and 15% of journal articles, respectively. Two randomized clinical trials (RCTs) were identified. Conference abstracts reported the use of non‐comparative study designs to a greater extent, many of which were published clinical audits. The distribution of journal articles by study design corresponded approximately to the distribution seen by Milley (Goodness of fit test: p = 0.066, table reproduced in Supporting Information 2), although it appears that interventional research was relatively more common internationally.

Table 1 demonstrates that the most common cancer types to feature in the identified articles were cervical, breast, prostate, lung, and colorectal. Of note, there seems to be some deviation from the worldwide mean, with an overrepresentation of cervical cancer, prostate cancer, and oropharyngeal cancer research in Ireland (odds ratios of 3.3, 2.5, and 5.9, respectively) and a relative dearth of colorectal cancer research (odds ratio 0.2).

TABLE 1.

Distribution of publications by cancer type and cancer continuum stage.

	Prev.	Risk	Screen.	Diag.	Treat.	Surv.	EoL	All	Other	Total
Cervical	4		13							17
All	1			2	5	4	1	2	1	16
Breast			2	6	2	3				13
Prostate			4	7		2				13
Lung	1			9						10
Multiple		1	1	2		3				7
Colorectal			3	1		2				6
Skin				6						6
Oropharyngeal			2	2						4
Miscellaneous				4						4
Ovarian				2						2
Pediatric				1		1				2
All publications	6	1	25	42	7	15	1	2	1	100
All publications	6%	1%	25%	42%	7%	15%	1%	2%	1%
Journal articles	8.3%	1.7%	33.3%	33.3%	5%	13.3%	1.7%	1.7%	1.7%
Milley—Global	9.4%	1.8%	33.8%	26.7%	2%	20.2%	4.3%	0.7%	0.8%
Milley—UK	9.7%	2.3%	13.4%	52%	2.6%	13%	4.9%	0.3%	1.3%

Abbreviations: All, all parts of the spectrum; Diag., early detection and diagnosis; EoL, pallative/end‐of‐life; Other, not‐otherwise‐specified; Prev., prevention; Screen., screening; Surv., survivorship; Treat., treatment.

In terms of the cancer continuum focus of the identified papers, screening, early detection and diagnosis, and survivorship comprised 82% of all papers, with the ranking of the former two reversing when only journal articles are considered. Additionally, there was good correspondence between the cancer continuum focus of the identified journal articles from the island of Ireland and the international literature, with screening and early detection and diagnosis papers accounting for 33% and 27% of journal articles in this review, respectively, compared with 34% and 27% in Milley, respectively (see Table 1 and Figure 3: International comparison of the distribution of publication by cancer continuum, adapted from Milley et al.).

FIGURE 3.

International comparison of the distribution of publication by cancer continuum, adapted from Milley et al.

3.2. Narrative Synthesis

Here we attempt a narrative synthesis of the 60 eligible journal articles through the lens of the role of primary care at different points across the cancer continuum. Our purpose is to better understand the underlying drivers of primary care cancer research across the island of Ireland, facilitating a greater awareness of potential knowledge gaps with the ultimate aim of stimulating research in underserved areas.

3.2.1. Prevention

Of the five publications pertaining to cancer prevention in primary care, three relate to cervical cancer and the 2010 introduction of HPV vaccination in the RoI for female students in their first year of secondary school (age 13) and boys of the same age in 2019. Research detailed the parental views on the barriers to vaccination [18] as well as education about cervical cancer prevention‐related behaviors in both patients and HCPs [19, 20].

Two other studies from the RoI were concerned with the perception of the risk of second‐hand smoke exposure to children [21], and men's preference about the delivery of cancer prevention information [22].

3.2.2. Screening

Of the 20 publications and one doctoral thesis written about cancer screening, 11 were about cervical screening in Ireland and related directly to the roll‐out of a cervical cancer screening program (“CervicalCheck”) in the RoI in 2008 [23, 24, 25, 26, 27, 28, 29, 30, 31, 32]. Another study from 1984 related to cervical cancer screening in Northern Ireland [33].

In general, there was reasonable balance in the cross‐cutting themes between the 12 cervical and 7 non‐cervical studies. Firstly, several papers examined the importance, knowledge, and practice of primary care professionals in relation to cervical screening [29, 32], as well as prostate screening [34, 35], breast screening [36], oral cancer screening by dentists [37], and cancer screening generally [38], with four of the five non‐cervical cancer studies being conducted in Northern Ireland. Secondly, a significant amount of research examined the barriers and facilitators to participation in screening for cervical cancer [23, 31], breast cancer [39], and colorectal cancer [40], with some authors focusing on specific hard‐to‐reach groups for cervical screening [25, 31] and oral cancer screening [41]. Finally, patient preferences about screening were assessed for cervical [30, 42] and colorectal [43] cancer. However, the factors required to roll out a screening program [24] and the efficacy of screening [26, 27, 28] were studied for cervical screening only, and in the RoI only.

3.2.3. Early Detection and Diagnosis

20 studies were categorized under the heading of “early detection and detection” with 18 of these concerning a specific cancer type.

All cancers (2): One study assessed the needs of GPs in the RoI in relation to early cancer detection by means of a focus groups and national survey of 929 GPs [44]. The second examined the use, detection rate and conversion rate of the “2‐week‐wait referral pathway” (2WW) in Northern Ireland, using data on 185,000 referrals and 29,000 cancer diagnoses in the Years 2011 to 2013.

Skin cancer (5): Three studies considered the adequacy of GP management of suspicious skin lesions—a case series of melanomas diagnosed by a GP with specialist training in skin cancers [45], and a comparison between the suspected GP diagnosis and Dermatologist diagnosis in 493 patients seen in a rapid access clinic [46], and comparison of survival rates in melanoma patients who had their initial diagnostic excision biopsy in primary care versus secondary care [47]. Also studied was the effect of an education seminar on GP trainee diagnostic accuracy [48], and the use of dermoscopy by GPs in Northern Ireland [49].

Prostate cancer (4): Four papers from the RoI focused on prostate cancer, with three evaluating the awareness of prostate cancer symptoms and factors influencing the tendency to undergo testing [50, 51, 52]. The final study reported the sensitivity, specificity, and positive predictive value of a positive digital rectal exam in 103 men referred to a rapid access prostate clinic [53].

Breast cancer (3): Three studies from the RoI dealt with the detection of breast cancer in primary care. Two of the studies examined diagnostic delay the factors associated with the promptness of help‐seeking [54, 55]. The third study derived and validated a clinical prediction rule for breast cancer using data from women attending an Irish national symptomatic breast unit in secondary care. While the rule was trained using secondary care clinic data, its intended purpose is to support the triaging of primary care referrals for further breast cancer assessment [56].

Lung cancer (3): Three qualitative studies were conducted to explore the early detection of lung cancer in a primary care setting in the RoI. The first study examined awareness of signs and symptoms and help‐seeking behaviors among high‐risk individuals [57]. The second and third examined the beliefs of healthcare professionals about the perceived barriers to help‐seeking by patients with symptoms and the clinical features motivating a referral [58, 59].

Other (3): Two studies looked at detecting oral cancer, one examined the confidence of GPs in doing so, while the other was a randomized trial assessing the effect of a patient information leaflet on patients' knowledge and risk perception of oral cancer in a dental practice [60]. There was also a case report of an unusual GP presentation of multiple myeloma [61].

3.2.4. Treatment

Two journal articles and one doctoral thesis from the same first author evaluated the cost‐effectiveness [62] and acceptability [63] of community‐based cancer treatment in the RoI [64].

The other paper in this category was a cross‐sectional study of 20,000 general practice consultations in Northern Ireland in 1980 that examined healthcare utilization in patients with a known cancer diagnosis [65].

3.2.5. Survivorship

Three of the eight survivorship papers focused on all or multiple cancers: a cross‐sectional study of “late effects” of cancer in Northern Irish survivors [66], a randomized clinical trial of a mobile health (mHealth) behavior change intervention for cancer survivors with overweight or obesity [67], and an examination of GP and outpatient clinic utilization in cancer survivors [68].

In the case of prostate cancer, a large Northern Irish postal survey sought to derive background rates of prostate cancer symptoms and treatment side effects in the general population so as to enable future evaluation of prostate cancer strategies [69]. A similar study assessed the rates of treatment side effects in prostate cancer survivors on both sides of the border. Of note, this was the only identified study which made an RoI‐NI comparison [70].

The concerns of colorectal [71] and breast cancer survivors [72] were assessed in NI and the RoI, respectively, while the dental complications of Northern Irish children in remission from acute lymphoblastic leukemia were compared against healthy controls [73].

3.2.6. End of Life

Only one study dealt with community‐based end‐of‐life care in cancer patients—a survey exploring 55 carers' experiences of both specialist‐led and GP‐led palliative care services in Northern Ireland [74].

3.2.7. Miscellaneous

Three journal articles out of 60, all from Northern Ireland, did not fit neatly into one of the above well‐established cancer continuum stages. One study examined the agreement between diagnostic data held by GPs and the “Northern Ireland Cancer Registry” [75]. While the argument might be made that this fits the “Early detection & Diagnosis” category, the diagnosis was not made in primary care, so this is simply a study of how well primary care records cancer diagnoses from secondary care correspondence, hence we allocated it to the “Other” category in the cancer continuum schema used by Milley [16]. Another assessed the experience of cancer diagnosis, treatment and survivorship in Northern Ireland via a survey distributed online and by telephone to 3750 recent cancer survivors [76], thus fitting into Milley's “All” category. The third examined GPs responses to hypothetical scenarios involving patients with elevated genetic risk for breast and colorectal cancer [77], fitting simultaneously into prevention, screening and early detection categories, but best fitting the “Risk” category in Milley's schema.

4. Discussion

4.1. Summary

This bibliometric analysis demonstrated the large increase in the number of primary care‐focused cancer research publications on the island of Ireland since 2010, both in NI and RoI, and in proportion to the size of each nation. Comparative observational study designs were used in more than half of the identified journal articles, while two out of sixty articles reported RCTs. The most common cancer types focused on in the publications were cervical, breast, prostate, lung, and colorectal; however, this differed significantly between NI and the RoI. In terms of the cancer continuum, most articles (42%) related to “early detection and diagnosis,” followed by “screening” (25%) and survivorship (15%), corresponding well to the role played by primary care in cancer control. The ranking of the top two reversed when only journal articles were considered.

A number of insightful themes also emerged from our narrative synthesis. First, there is a regrettable lack of cross‐border research (a single study) since this would enable exploration of the effects of the differences between these health services. Second, it appears that research which leverages large national healthcare databases is reasonably common in NI, and non‐existent in the RoI, suggesting the need for infrastructure development in this area. Third, it appears that much primary care cancer research in Ireland is published in Irish general medical journals rather than high‐impact primary care journals with international readership. Fourth, there seems to be poor representation in the literature of primary care disciplines other than general practice and community dentistry. And last, primary care research into cancer in Ireland lacks strategic direction, leaving important areas of research underserved.

4.2. Comparison With Existing Literature

Our bibliometric analysis revealed broad correspondence between the Irish literature and the worldwide literature in terms of cancer continuum stage, with “screening” and “early detection and diagnosis” papers accounting for 33% and 27% of journal articles in this review, respectively, compared with 34% and 27% in Milley, respectively.

Furthermore, there was good general agreement in terms of study design, although there was some evidence that interventional studies were less common in this review (4% of journal articles) than globally (10.5% of journal articles).

In terms of cancer type, there seemed to be some deviation from the worldwide mean, with an overrepresentation of cervical cancer, prostate cancer, and oropharyngeal cancer research in this review (odds ratios of 3.3, 2.5, and 5.9, respectively, see Table 2) and a relative dearth of colorectal cancer research (odds ratio 0.2, see Table 2).

TABLE 2.

Distribution of publications by cancer type, compared with the international literature from Milley et al.

Cancer type	All articles	Journal articles only	Milley et al.
All/Multiple	23%	23.3%	31.5%
Colorectal	6%	3.3%	17.5%
Breast	13%	10%	12.9%
Cervical	17%	25%	8.1%
Prostate	13%	13.3%	5.8%
Lung	10%	6.7%	5.4%
Skin	6%	8.3%	5.3%
Pediatric cancers	2%	2%	2.6%
Oral/Nasal/Pharyngeal	4%	6.7%	1.2%

In order to reconcile the per capita publication rate in our study with that reported by Milley, we calculated a rate of 3.6 publications per million population between 2013 and 2019 (based on 23 journal articles and a combined population of 6.7 million in Northern Ireland and the Republic of Ireland in 2019). While Milley et al. only reported data from the top 10 countries study over the same period, we know that since Ireland was not in the top 10, its rate in Milley's study must have been less than the rate of the tenth ranked country (0.6 publications per million population). This highlights that our search strategy or eligibility criteria were, in effect, more expansive than Milley.

4.3. Strengths and Limitations

The major contribution of this review is that it is the first attempt to systematically map primary care cancer research in the RoI and NI. Secondly, we conducted it, as much as possible, in a manner which allows international comparison. Furthermore, we attempted to be as inclusive as possible, expanding the scope beyond general practice to all primary care disciplines, which we hope will help progress the conversation on multidisciplinary primary care research.

Although our search strategy encompassed multiple bibliographic databases (MEDLINE, Embase, CINAHL, Web of Science, and the Cochrane Library) alongside gray literature sources (ProQuest Dissertations & Theses, ClinicalTrials.gov, and the WHO International Clinical Trial Registry Platform) to minimize publication bias, relevant studies may still have been missed because of incomplete indexing or non‐publication. Consequently, publication bias cannot be entirely excluded.

A number of limitations of this review stem from our inability to define exactly how we would determine eligibility pre hoc. Firstly, we have no way to tell that search terms such as “Ireland” or “Irish” capture all research from the RoI and NI, and our eligibility criterion (“the paper separately reports data from RoI or NI study subjects”) was defined post hoc to accommodate the unanticipated observation that some research about Irish subjects is conducted by overseas research teams.

Secondly, while the search theme of cancer is easy to operationalize, the distinction between included studies which deal significantly with cancer and excluded studies which deal with cancer only in passing is impossible to define reliably. Our working definition for this eligibility criterion, “the paper would be unrecognisable if all mentions of cancer were removed,” while easy‐to‐use, was again defined post hoc and remains reasonably subjective.

The same is true of the boundaries of primary care, where we encountered the issue of secondary care research featuring primary care prominently. We chose to exclude audits of the breakdown of diagnoses at rapid access cancer clinics (in secondary care) where less than 100% of the referrals were from GPs and include articles about community‐based oncology, while recognizing that the former is more likely to inform practice by primary care professionals. Lastly, and more fundamentally, we are suspicious that our search strategy failed to pick up certain primary care disciplines such as community pharmacy, a limitation which stemmed from our use of Milley et al.'s GP‐focused bibliometric analysis as a guide.

In each of these cases, our defense is that this is the first such scoping exercise of the literature in two nations without an established primary care cancer research infrastructure, so it was difficult to anticipate the above questions, which arose only in light of unexpected search findings, where such an eventuality had not been accounted for by Milley et al.

One technical limitation was the finding that citation metrics for the Irish Medical Journal after the Year 2000 were not available on Clarivate Incites. Due to the large number of identified publications in this journal, we abandoned this element of our bibliometric analysis.

4.4. Suggestions for Future Mapping of the Primary Care Research Literature

We suggest that future attempts to map the primary care cancer literature would benefit from clear definitions of primary care research, cancer research, and research from a particular country, as these concepts are more subjective than they seem at first glance. Secondly, clearer agreement on the best way to handle publications comprising multiple study designs, a focus on multiple cancer types or cancer continuum stages, and study subjects from multiple countries would permit more reliable international comparison. Finally, a “core outcome set” for bibliometric data would allow not just reliable comparison but the development of easily digestible summaries and graphs.

In the future we would like to map identified journal articles to their funding source, the research team or discipline of the authorial team, related publications (e.g., trial protocols, conference abstracts and policy documents), a quality assessment of the studies. In an ideal world, we could create a living systematic review of primary care research into cancer displayed online as an interactive dashboard.

4.5. Implications for Future Research and Policy

This review provides researchers and policymakers with an understanding of the current state of primary care cancer research on the island of Ireland as well as a benchmark against which to monitor future activity. Its implications are primarily for the strategic prioritization of research rather than for direct changes to clinical practice or policy: no individual GP will practise differently as a result, and we make no specific policy recommendation [78]. Rather, by identifying underserved areas and mismatches between research activity and population healthcare needs, the review highlights the need for strategic direction, ensuring that the research undertaken reflects the priorities and values of key stakeholders and knowledge users rather than historical or opportunistic factors [79].

That influence is not merely hypothetical. Recognizing from this synthesis that primary care cancer research effort was comparatively light and unevenly distributed directly motivated our subsequent James Lind Alliance–style priority‐setting exercise on early detection of symptomatic cancer in Irish primary care [80]. In the same spirit, we are establishing a dedicated stakeholder group for primary care cancer research in Ireland, to coordinate researchers and knowledge users and to support capacity building and networking [79].

More broadly, the review suggests there is room for improvement in the Irish cancer research effort—in the volume, quality, and impact of the work being done—and prompts consideration of a more coordinated cancer research ecosystem for the country. It complements the broader mapping undertaken by the All‐Island Cancer Research Institute (AICRI) by supplying a primary‐care‐specific perspective [81]. One limitation bears noting here: while the distribution of publications across the cancer continuum mirrors that of the global primary care research community, we did not assess the quality of the research or its capacity to shape practice—an evaluation we intend to undertake in future.

Lastly, we recommend that all primary care researchers on the island of Ireland consider cross‐border research designs which leverage the natural experiment created by a homogeneous ethnic group being separated by two different primary care systems. Large ecological studies can help unpick the consequences of variations in primary care policy, especially the difference between free GP care for all residents in NI but only the poorest 40% in the RoI.

Author Contributions

Benjamin M. Jacob: conceptualization, investigation, writing – original draft, methodology, validation, visualization, writing – review and editing, formal analysis, project administration, data curation, supervision, software. Sarah Jacob: validation, formal analysis, investigation, data curation, writing – original draft, writing – review and editing, visualization. Brittney Adams: writing – review and editing, project administration, supervision, data curation. Ahmad Aburezq: investigation, writing – review and editing. Rabwa Ahmed: investigation, writing – review and editing. Aine Harris: writing – review and editing, project administration, supervision. Patrick Redmond: conceptualization, writing – review and editing, resources, methodology, project administration, supervision, funding acquisition.

Funding

B.M.J.'s involvement was part‐funded by a 2022 HSE/ICGP Post CSCST Fellowship in General Practice (Funder: HSE National Doctors Training & Planning).

Ethics Statement

The authors have nothing to report.

Consent

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting Information 1: Search strategy.

Supporting Information 2: Distribution of publications by study design (adjusted for publications comprising multiple studies).

Data Availability Statement

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.