Abstract
Background We conducted research on the diagnosis and treatment of urinary tract infections in general practices in the Northern Dutch-German cross-border region. The challenges that were encountered in that study are discussed in this article, including prolonged formal ethical and legal processes, data collection, and recruitment of both general practitioners (GPs) and patients.
Aim To provide recommendations for those in the early stages of planning and conducting research in general practices.
Design & setting Opinion paper based on the authors’ experience conducting a cross-border prospective study.
Method An overview of the strategies used to collect data and recruit participants is presented, along with their respective advantages and disadvantages.
Results GPs were recruited by post, email, and newsletters, with varying success rates depending on the region. Overall, personal approaches fared better, leading to increased willingness to participate in the study. Two different measures were used to collect data: students in The Netherlands and practice assistants in Germany. While using student assistants alleviated the constraints associated with study participation for a general practice, their limited availability disrupted the continuity of data collection phases. Instructing practice assistants to collect data independently enhanced flexibility regarding patient recruitment, but the authors’ experience revealed that this approach was more prone to errors, likely as a result of heavy workloads. Designating a contact person within the practice facilitated communication between the practice and the researchers, providing reminders for ongoing study recruitment.
Conclusion With research in general practice becoming increasingly important, it is necessary to illustrate and understand its barriers and facilitators. Data collection strategies have to be carefully assessed and selected according to the specific needs of practices (such as flexibility and time constraints) to mitigate disruption of regular care.
How this fits in
Research in general practice poses unique challenges, particularly when it involves data collection across multiple sites and countries. Drawing on their prospective study as an example, the authors provide valuable insights by sharing first-hand accounts of the challenges encountered and the solutions implemented. These recommendations can be used to help researchers improve the efficiency and reliability of data collection and participant recruitment.
Introduction
GPs have access to the broadest group of patients and diseases as the first point of contact for routine health issues. Additional clinical studies in general practices are necessary to target diverse patient populations and further strengthen primary care. Navigating research in general practices can be a complex process with unique challenges, especially when it involves collecting first-hand data.1–4 This is further complicated when conducting research in multiple countries.
A cross-border study was conducted in the Northern Dutch–German cross-border region, the Ems-Dollart region. The aim was to examine the differences in diagnosing and treating uncomplicated urinary tract infections, as well as the prevalence and antibiotic resistance of uropathogens, by collecting data through patient surveys and urine samples.
Method
In this article, the authors present the challenges they experienced while developing and conducting research in Northern Dutch and German GP practices. They propose potential solutions with the aim of highlighting considerations that could assist other researchers conducting cross-border research in future.
Results
Preparing to conduct the study
Legal and ethical considerations
The research was a cooperation between the University of Oldenburg (UOL) in Germany and the University Medical Center Groningen (UMCG) / University of Groningen (RUG) in The Netherlands.
At its onset in 2019, the study did not require extensive review by the Medical Ethics Committee in The Netherlands as it did not fall under the scope of the Medical Research Involving Human Subjects Act. A confirmation letter was received from the Ethics Commission. By contrast, in Germany, any research that involves personal/medical data and biomaterial requires an ethics committee review.
Several contracts needed to be drafted to ensure data security conformity:
a data processing agreement between the universities and corresponding laboratories;
a cooperation agreement/contract between each recruited practice and UOL; and
a data transfer agreement between UOL and UMCG.
Drafting and finalising all contracts created a considerable challenge, as gaining approval from all involved parties delayed the start of data collection by approximately a year. Moreover, procedures were different in each country.
Preparation for data collection
Data were collected using a combination of self-administered questionnaires and laboratory analyses. Two separate patient questionnaires were used:
Intake questionnaire: administered in the general practice setting for patients to complete during their visit. The form used for taking a medical history in The Netherlands (‘Urinebriefje’) was inserted into this intake questionnaire to reduce the number of questionnaires completed by patients in the practice.5
Follow-up questionnaire: provided to patients to take home or complete online, with instructions to complete this within a week after their visit. Patients provided contact details to receive reminders.
Additionally, urine samples were collected from the patients for routine testing and subsequent antibiogram analysis in a designated laboratory. An accompanying form was used to document diagnostic information obtained by practices during the patient’s visit.
The authors decided to work with one laboratory from each side of the border to allow more comparable and homogeneous data collection regarding uropathogens. The German laboratory results were not available digitally, so data were entered manually from printed results.
Recruiting practices
The objective was to recruit 20 practices per country, each of which would recruit approximately 10–15 patients, targeting around 500 patients.
In Germany, 212 GPs were emailed using the research network of the Division of General Practice of UOL. Five practices responded, of which one was included in the study, because it worked with the selected laboratory and was located within the study region. Three more practices were recruited through personal contacts, events, and colleagues. At the same time, the laboratory was given flyers to disseminate to practices. From 18 practices that consented to be contacted, four were recruited. Approaching practices by phone was unsuccessful because of the prolonged wait times for calls to be answered. A total of eight practices in Germany took part in the study.
Similarly, on the Dutch side, the AHON network was used.6 Fifty practices with at least three GPs were initially approached by post and newsletter. However, most practices were recruited through personal contact. We used snowball sampling through GPs who were already participating in the study. Eight practices were successfully included. The challenges of recruiting sufficient practices were similar on both sides of the border.
The requirement to use the pre-selected laboratory was the main GP recruitment challenge in Germany, but not in The Netherlands, where the selected laboratory covered a wider area of the study region. Some GPs in both countries declined to participate because of time constraints in their practices and uncertainty about being able to recruit sufficient patients. One practice on the Dutch side said that they lacked room to accommodate a student for the study. Efforts to recruit GPs were hindered by limited response rates.
Assigning an individual to collect data within the general practice: student vs practice assistants
Data collection was conducted by students or practice assistants. In Germany, data were collected by practice assistants. Students were not used in participating German practices because of a lack of available workspace. In The Netherlands, data were primarily collected by students, with practices also being offered the option to independently collect data themselves (used by one practice). In the following section, the advantages and disadvantages of these approaches will be discussed.
Use of students
In The Netherlands, students were given instructions on how to collect data and subsequently rotated from practice to practice, collecting data using a study checklist. Students were either employed as assistants or were offered research internships.
The main advantage of using students was minimising disruption to routine tasks, as practices only needed to offer the student a workspace and notify students of potential participants. However, students faced challenges recruiting patients, particularly on low-traffic days, and scheduling across multiple practices was constrained by their availability. Limited workspace in smaller practices and higher costs for student assistants added to the difficulties of this option. Retention was challenging, as students’ commitments and internships ended mid-study, requiring further hiring, and disrupting consistency in data collection across sites.
Two research internship students recruited a vast majority of the patients (n = 106, 87%) during November 2022–February 2023 and June–August 2024. Two paid student assistants also contributed during summer 2023, but their patient recruitment was limited (n = 8, 13%).
Instructing practice assistants
A flowchart of study logistics and training was provided for practice assistants in Germany to facilitate data collection (see Figure 1), which required frequent in-person visits by the primary researcher to establish contact, monitor progress, and collect documents.
Practices were reimbursed €15 for each successfully recruited patient. While this approach allowed flexible patient recruitment, it increased the workload for practice assistants and was error-prone, with issues like missing or mislabelled documents, and incorrectly handled questionnaires. Simplifying instructions could have reduced errors, which improved after visits from the researcher. Recruitment challenges included patients being overlooked because of time constraints, absence of trained assistants, or lack of awareness. Labour shortages further prolonged data collection. To improve recruitment, practices were encouraged to brief all staff and train additional assistants. Data were collected from August 2023─June 2024.
Flowchart titled “Flowchart TRACE-UTI” showing the study process for patients presenting to primary care with symptoms of a urinary tract infection.Top section: A patient comes to the practice with symptoms of a urinary tract infection. The practice informs the patient about the TRACE-UTI study and provides a study brochure. A box at upper right lists the inclusion criteria: female, adult (over 18 years old), not pregnant, able to consent, has read, understood and signed the informed consent form.A glossary at upper left states that REDCap is software for web-based data collection using questionnaires and forms. A contact person field is shown as a placeholder for name, telephone number and email.Decision point:The central decision asks, “Did the patient decide to participate?” If no, the patient will be treated as usual. If yes, the study workflow continues.Consent and enrolment: If the patient participates, the practice gives the patient the following documents:- study information- data protection information- informed consent form- contact data formThe patient reads, completes and signs the documents.The practice gives the patient a pseudonym. The pseudonym is also entered onto a list with the patient’s name.The patient fills in the contact data form.Materials split into two groups: 1. Materials taken home by the patient:- copy of the study information, data protection information and informed consent form- Week 2 questionnaire, labelled with the pseudonym- return envelope2. Materials kept in the practice for later collection by the study team:- signed informed consent form- pseudonym list with names labelled with the pseudonym- Week 1 questionnaire, labelled with the pseudonym- GP form, labelled with the pseudonymPractice data entry: The practice registers the patient with REDCap, preferably by the end of the practice day. The necessary data listed are:- pseudonym- date of visiting the practice- email address and/or phone number of the patientA crossed-out contact data form indicates that, after the data have been entered in REDCap, this form has to be destroyed.Questionnaires and urine sample: The practice gives the patient:- Week 1 questionnaire- Week 2 questionnaire- Return envelopeThe patient fills in the Week 1 questionnaire and hands it in to the practice.The patient delivers a urine sample.The practice affixes the pseudonym to the urine monovette and the request form. A note states that additional study information is entered in the practice software as “Study name + pseudonym”.Urine testing and routine care: The practice conducts routine diagnostic tests using the urine sample, with a note saying to use the tests provided by the study. A laboratory analysis box is shown to the left, connected to the sample pathway. At the lower left, a dashed box distinguishes two specimen routes:- Urine sample for routine testing, shown with a container labelled “Routine sticker” and a red “Study (pseudonym)” sticker, plus an “Order form (routine)” also marked with a red study pseudonym sticker- Urine samples only for study purposes, shown with a container labelled with a blue “Study (pseudonym)” sticker and a “Study order form”.The patient will be treated as usual.Follow-up after seven days: A horizontal bar labelled “After seven days” separates the follow-up stage.The research team contacts the patient. The patient fills in the second questionnaire.Two return options are shown:- Option 1: return in paper form with a postage-paid return envelope- Option 2: complete online via the link sent or via the QR code or link in the documents.
In The Netherlands, one practice used practice assistants, which led to the recruitment of seven patients over 2 weeks.
Duration of data collection
Two distinct strategies were used to determine the duration of data collection; restricting data collection to a predetermined duration (The Netherlands), or to a target number of recruited patients (Germany).
In The Netherlands, the recruitment period was initially set at 2 weeks to accommodate participation and time limitations, with the exception of one practice extending this to 3 weeks. The study was paused for approximately 9 months, because of a lack of students and, to a lesser extent, participating practices. After this, the authors returned to three practices to conduct second rounds of recruitment.
In Germany, no restrictions were imposed on the duration of data collection, giving the practice assistants greater flexibility. Practices recruited patients until they reached 12 patients, after which a second round of the study was initiated depending on recruitment success. The average recruitment period per practice was 22 weeks (range: 11─33 weeks), with recruitment success varying widely (range: 5─17 patients/practice); four practices conducted second rounds of recruitment.
The success of the student-based approach is evident because more patients were recruited on the Dutch side (n = 121) than the German side (n = 87) in a shorter time period, approximately five patients per week versus two per week (cumulative study duration: 24 weeks versus 43 weeks).
Allocating a contact person
In Germany, a personal meeting was arranged with contact persons from practices, which typically included one practice assistant and one GP. GPs and practice assistants were instructed together on study procedures in five practices and separately in one. Practice assistants were instructed alone in two practices where GPs were unavailable. Depending on the practice, either a GP or a practice assistant was the direct contact person for all matters pertaining to the study, who was contacted regularly to provide reminders and ensure accurate data collection.
In The Netherlands, practice assistants were the contact people for the students conducting the data collection.
Recruiting patients
On the German side, flyers advertising the study were distributed at the reception or occasionally placed in waiting rooms. Practice assistants informed eligible patients about the ongoing study and asked them to participate. Personal contact with practice assistants revealed that participants’ willingness to participate varied. Several practices encountered difficulties recruiting older participants, who were more hesitant to provide contact information for reminders to complete the second questionnaire because of concerns about data security and unfamiliarity with research. Some participants were unable to provide e-mail addresses. Additionally, some individuals felt overwhelmed by the study steps and did not consider themselves capable or willing to participate. Rural practices reported having bigger challenges as their patients skewed towards older age groups.
On the Dutch side, flyers were not provided for patient recruitment. Patients were informed of the ongoing study at the reception or occasionally via telephone during appointment scheduling. Reasons for non-participation were not recorded. Students on the Dutch side were only present in the practices during the morning, when most patients submit their urine samples, and thus missed opportunities to recruit potential patients in the afternoon.
Discussion
Summary
During preparation for data collection, specific needs of practices (such as flexibility and time constraints) have to be considered. Allowing practices to choose between using student assistants or collecting data themselves appears promising. This flexibility can enable practices to select the least intrusive methods based on their capabilities, potentially increasing their willingness to participate. The choice should consider the specific circumstances and resources of each practice to ensure efficient data collection while minimising disruptions to daily operations. The presence of a study team member may have helped to alleviate patients’ concerns, but this was not feasible in German practices because of limited space.
The authors noticed a difference in recruitment success when approaching GPs, as personal approaches fared better. When approaching GPs, selecting a more personal and tailored approach seemed to be more effective in recruiting practices as opposed to sending mass emails: two GPs in Germany were recruited through emails and only one Dutch GP was recruited through post or newsletter.
Furthermore, the use of student assistants can help to reduce strain on practices and ensure efficiency and consistency in data collection. Directly engaging GPs in research can also facilitate patient recruitment. GPs explained the process to the different practice assistants, decreasing the risk of error in implementing the study steps. However, practice assistants were easier to contact and exchange information with compared to practices where the only contact points were GPs.
Strengths and limitations
A main strength was that this overview offers a direct perspective of the challenges of research in general practice, drawing from the authors’ experiences of conducting a prospective study within general practices in two different countries.
This is an opinion paper based on a prospective study featuring a limited sample size of participants and practices, limiting the generalisability and representativeness of the report. Nevertheless, these recommendations may still be valuable for future researchers in the early stages of planning a similar endeavour.
Comparison with existing literature
Cross-border research is becoming increasingly important, yet it faces unique challenges alongside those of general practice research. Infectious diseases transcend national borders, and differences in health systems and regulations further complicate healthcare provision and research efforts.7 The study started approximately 1 year earlier in The Netherlands owing to complications with the legal and ethical documents on the German side. Data security and ethical concerns can differ between countries and should therefore be addressed as early as possible.8
Contact through a person known to the practice, such as a GP, proved to be beneficial. Other studies also found personalised approaches to be most successful.4,9 Similar to this study, a qualitative study found that a barrier to recruitment for some GPs was the lack of personnel resources or time in their practices but personal contact to the university was regarded positively.10 Moreover, GPs interested in the study topic are more likely to participate.11 Providing GPs support through research staff dedicated to patient recruitment can be beneficial if they do not interrupt daily routines.2
If a practice collects data themselves, it is essential to inform all practice staff that have access to potential participants about the ongoing study to decrease missed recruitment opportunities. An Irish study found that involving more practice employees helped increase the likelihood of successful patient recruitment.12 Overwork can be a challenge to successful recruitment, as evidenced by a qualitative study from the UK.13 In Germany, there is a labour shortage of practice assistants.14 Compared to German practice assistants, Dutch practice assistants are delegated more tasks related to diagnosing and treating patients, potentially allowing them to take on additional responsibilities for study recruitment and participation.15 Another potential barrier was the number of documents used for data collection. An increased amount of work in addition to time limitations can deter GPs and patients from participation.2 The challenges faced may be further exacerbated by the more rural location of this cross-border study region, as urban practices are more likely to successfully recruit patients.12
Having one dedicated contact person in the practice facilitated communication between the practice and the researcher conducting the study. Providing practices with periodic reminders reduces the risk of the study being overlooked amid competing routine tasks.1 A qualitative study found that frequent discussions with practice staff and assigning a person responsible for the study improved patient recruitment.13
Potential role of research infrastructures
In 2020, a general practice research practice network called DESAM-ForNet was established in Germany with the aim of building a national research infrastructure in general practice, funded until 2025 by the Federal Ministry of Education and Research.16,17 Encompassing GPs, practice staff and university departments for general practice, the aim of the network is to create standard operating procedures, and support research and scientific exchange. This nationwide network did not include north-western Germany when this study started, but has since expanded to include three university general practice departments in this region. Following the network's establishment, funding has continued primarily through university and professional society channels, though its long-term sustainability remains uncertain. The Netherlands is currently building a clinical trial network to overcome the issues associated with clinical research in general practice. Ensuring continued governmental funding of these networks can contribute to their longevity, building sustainable and standardised research that protects the public’s interest.
Implications for research
The authors explored various methods of participant recruitment and data collection through the cross-border setting of this study, enabling them to evaluate the merits of different strategies. They recommend accounting for the time required to finalise contracts, and considering legal and ethical concerns when developing the research proposal, study protocol and overall study timeline. Future cross-border studies should strengthen early collaborations between researchers, practices and institutions, by addressing differences in language, healthcare systems, and regulations through flexible and well-communicated study designs.
The data collection process should be designed to integrate smoothly into the daily operations of the practice without causing significant disruptions to clinical activities. Using research networks and allowing practices flexibility on data collection can improve their willingness to participate. Providing targeted training and support for both research teams and practices will help maintain engagement and ensure data quality. Busier practices with increased patient caseloads may benefit from using student assistants to support patient recruitment. The study should be visible to all staff and potential participants to promote patient recruitment. By addressing these challenges and establishing clear communication and responsibilities, the recruitment and data collection process can be made more efficient and reliable.
Notes
Funding
This project was funded by the Ministry of Science and Culture of Lower Saxony (MWK) as part of the ‘Niedersächsisches Vorab’ Program. (Grant Agreement No. ZN3730).
Ethical approval
No ethical approval was required for this opinion paper. The underlying study was approved by the Medical Ethics Committee of UOL (number: 2022-070) and the Medical Ethical Committee of the UMCG (2019/678) prior to the beginning of the study.
Provenance
Freely submitted; externally peer reviewed.
Data
N/A
Acknowledgements
This study was conducted in partnership with the Cross-Border Institute of Healthcare System and Prevention (CBI), a joint initiative of the Carl von Ossietzky University of Oldenburg, University of Groningen and University Medical Center Groningen (UMCG). The authors want to acknowledge the cooperation and support of the CHARE-GD subgroup.
Competing interests
The authors declare that no competing interests exist.
Disclosure
ChatGPT-4o was used to improve language and sentence structure of this manuscript.
- Received December 10, 2025.
- Accepted December 29, 2025.
- Copyright © 2026, The Authors
This article is Open Access: CC BY license (https://creativecommons.org/licenses/by/4.0/)
References
LinkedIn
Jump to section
Keywords
More in this TOC Section
Related Articles
Cited By...
