Point of Care Ultrasound Use by General Practitioners in France (Echo-MG)

There are few studies describing the use of POCUS in general medicine. Most of them are descriptive studies made more than ten years ago while recent technological development has resulted in better quality images. There is no study with high level of evidence, and the majority of the literature is made of descriptive reports based on the activity of a small number of physicians.

Unlike other countries, France has currently no recommendations on the modalities of use for POCUS. There are no guidelines on the GP curriculum nor on the appropriate equipment for general practitioners, nor on the certification process.

In the current state of knowledge, it seems relevant to establish an overview of the practice of POCUS in general medicine in France. This work is part of an international European reflection on the use of POCUS in general practice and its role in the diagnostic process, therapeutic process and the patients' care pathway in Denmark, in Norway and in France.

The current protocol was written in collaboration with the GP Research Unit of Aalborg. The Danish protocol was completed between January 2018 and July 2019 in Denmark, registered at clinicaltrials.gov NCT03375333. This part of the protocol constitutes the French study.

The aim of this study is to describe the use of POCUS and its role in the diagnostic and therapeutic process in general medicine.

Description of the use of POCUS through indication, organs scanned, findings, frequency, time consumption.

Analysis of the role of POCUS in the diagnostic process: change of diagnostic hypotheses and change of certainty in the main diagnostic hypothesis.

Analysis of the role of POCUS in the therapeutic process: change in the care pathway and the therapeutic initial plan.

The study is a prospective observational cohort study. It will take place in France among 30 GPs that use POCUS during consults in their office or during home visits.

Any patient having a POCUS scan during a consultation by a participant GP may be included in the study. Patients will be informed and asked to sign a consent form before inclusion in the study. Patients will be informed of the study terms by their GP and through an information note in the waiting room. Patients will be excluded if they refuse to participate in the study or are unable to give informed consent. Refusal to participate in the study will not change patient management.

There will be no intervention in this study. The GPs already use ultrasound during their physical examination. The data recorded for this study will reflect the daily practice of general practitioners using POCUS. There will be no additional exams or other ways to change the patient management. If a second POCUS is performed later for the same patient, this data will not be collected. Only data from the patient's first examination will be collected.

Each investigator will collect data about every POCUS examination performed during the inclusion period of 22 consecutive working days.

The data collection tool is based on the questionnaire originally developed by the Center for General Practice at Aalborg University in Denmark from literature data, a dialogue with the "Ultrasound group of the Danish College of General Medicine" and an interview study. The registration tool was developed as a questionnaire to be used before and after the GP uses POCUS in the consultation. A time log is included to ensure that the questionnaire has been completed before and after the exam.

For this study, the questionnaire will first be translated into French and then back-translated (reverse French-English translation) by two bilingual translators separately. The result of the back-translation will be analyzed and a second French version will be submitted to French general practitioners practicing POCUS and members of the "Ultrasounds group of the College of General Medicine" in France. For each theme of the questionnaire, there will be a common part of questions, to ensure the comparability of the results between the different European countries. An additional set of questions relevant to general practice in France will be discussed and validated by the French College of General Medicine. The acceptability and validity of the questionnaire will be tested during focus groups and a pilot study.

The GPs will be asked to fill a questionnaire in the online database SurveyXact each time they use POCUS during their daily work. The data will be registered electronically in SurveyXact with a time log to ensure a before and after registration.

The GPs can access the questionnaire on their mobile phone, iPad or computer, using a unique link (respondent link) allocated to each participant. A key file, connecting each GP participant with the respondent link in SurveyXact (link), will be safely stored at Center for General Practice at Aalborg University, Denmark.

The GPs will give each patient a unique ID-number. A key file connecting this ID-number and the patient's identity will be safely stored at the GP's clinic.

There will be a registration of the date and time, when the GP starts filling the questionnaire and finishes the final question. Furthermore, a variable with the exact time between "before" and "after" questions will be created. If GPs use less than one minute on this page and the GP registers a duration of the POC-US > 1 minute, there will be no "before and after" registration, but only an "after" registration. In that case, the "before" answers will be excluded.

Participants retention In order to calculate the participant retention rate, defined by the proportion of eligible patients included in the study, GPs will be asked to record age and gender for each patient eligible but not included in the study (time constraints, patient refusal or other reasons).

Data management Data will be saved electronically on the SurveyXact server and on a server at Aalborg University and will only be accessed by the research group using personal passwords. The patient key file and consent forms will be safely stored at the GPs' office and the research group will not have access to this information during the study.

Any paper editions of the registration tool or questionnaire (in case of server breakdown) will be safely stored at the GPs office until the end of the data collection. Afterwards, they will be safely stored at the General Practice Department of Sorbonne University, 27 rue de Chaligny, PARIS 75012, FRANCE.

The use of ultrasound will be described in terms of ultrasound indications, organs explored, findings, frequency of use and duration of the examination.

The indications of POCUS will be described through the GP's intention regarding ultrasound:

• to answer a specific clinical question (yes / no / maybe) and document the answer in the medical file
• to conduct therapeutic education
• to monitor or control an ultrasound finding
• to screen an asymptomatic patient
• to scan to improve technical ultrasound skills without any clinical profit
• to perform a diagnostic scan with a detailed description of anatomical regions and provide a complete diagnostic report.

The frequency of the different POCUS examinations will be summarized in relation to the organs scanned. The performed POCUS examinations are defined as organs scanned and not as standardized procedures such as FAST, FATE or LUS since there might be differences in the definition and interpretation of these examinations. The organs scanned are registered on a list of organs in the questionnaire. The possibilities in this list originate from interviews with GPs using POCUS and from Danish pilot testing. This list will be discussed during a focus group of French GPs using POCUS. The GPs will be allowed to write in free text if organs are missing from the list.

The frequency of POCUS use will be described as the number of GP consultations where POCUS is used over the total number of GP consultations per day during 22 consecutive working days.

Each participant will provide information on the total number of face-to-face consultations she or he has had during the study period. If the GP uses a portable or ultraportable ultrasound device during home visits, they will be included in the total number of consultations.

The GPs will measure the time used for the POCUS examination. The time registration starts when the scanning begins and ends when the scanning has been finished. Hence, it will only include the duration of the POC-US examination, not information about the scanning or other elements of the consultation. This time registration will be described in minutes for each type of POC-US examination.

The study will describe the main diagnostic hypothesis recorded before and after POCUS, the total number of diagnostic hypotheses recorded before and after POCUS, the change in the total number of diagnoses hypotheses, the change in the GP's confidence in the diagnostic hypotheses (5-point Likert scale). To test the reliability of the GP confidence in the diagnoses, we will describe the correlation between the GP's confidence and the type of organ scanned, the change from a symptom diagnosis to a disease diagnosis and the decrease in the number of diagnostic hypotheses. A scatter plot will be used to describe the relationship between "confidence in the main diagnostic hypothesis" and "the number of diagnostic hypotheses". Frequency of incidental findings will be described separately related to the organs explored.

After the use of POCUS, the GPs will be shown their "Before-POCUS" diagnostic hypothesis and asked if these diagnoses have changed. If the diagnoses have changed, they will be asked to specify them.

The diagnostic hypotheses are registered as detailed free text in the questionnaire. The corresponding ICPC-2 (International Classification of Primary Care-2) codes will be coded from the free text by the research team.

The observed change in the diagnostic hypotheses will be described by the frequency of declared diagnostic hypotheses changes and the overall registered change in the ICPC-2 codes.

The patient's plan before and after POCUS and the planned treatment before and after POCUS will be described. The analysis will describe the relationship between the ultrasound findings and the change in the patient management (diagnostic process, treatment, care pathway). The diagnostic and therapeutic modifications will be described by the frequency of responses or by the change in the total number of responses.

The statistical analysis will be descriptive. Data collection will mainly be done using nominal variables. However, the variable "GP's confidence in the diagnostic hypotheses" will be measured by an ordinal variable, "the duration of the examination" and "the patient's age" by a continuous variable.

If the continuous variables follow the normal distribution, they will be described with means and either a standard deviation or a 95% confidence interval of the mean. If they are not normally distributed, they will be described with a median and interquartile range.

To assess the relationship between paired variables the Mac Nemar test will be used. To compare the characteristics of included and excluded patients, a Chi-2 test will be used. Results will be considered statistically significant if the p-value is less than 0.05. For highly significant results with p-values much lower than 0.05, we will simply report p < 0.01.

After using POC-US the GP is shown the "before POC-US plan for the patient" and asked if this plan has changed. If the plan has changed, the GP is asked to specify.

Change in the plan for the patient is defined as the frequency of declared change after registration from one possible answer to another or change in the total number of possible answers.

After using POCUS, the GP is shown the "before POC-US planned treatment for the patient" and asked if this planned treatment has changed. If the planned treatment has changed, the GP is asked to specify.

Change in the planned treatment of patients is defined as the frequency of declared change described before and after registration from one possible answer to another, or in the total number of possible answers.