Factors influencing inappropriate management of upper respiratory infections in general practice

doi:10.21203/rs.3.rs-3564540/v1

Download PDF

Research Article

Factors influencing inappropriate management of upper respiratory infections in general practice

https://doi.org/10.21203/rs.3.rs-3564540/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

Upper respiratory tract infections (URIs) are common in general practice, and inappropriate antibiotic prescribing remains a major concern that contributes to antimicrobial resistance. This prospective study aims to examine the factors influencing inappropriate antibiotic prescribing for URIs in general practice.

Methods

The study was conducted during the winter of 2017 by 15 general practice trainees in Alsace, France, in collaboration with 25 general practitioners. Data from random general practice consultations were collected, and specific grids were used for URI-related consultations.

Results

Out of the 807 consultations, 173 were related to URIs. Antibiotics were prescribed in 35% of cases, and management was deemed appropriate in 75% of cases. Amoxicillin was the most frequently prescribed antibiotic. Certain infections, such as sinusitis and bronchitis, were more likely to result in inappropriate antibiotic prescriptions. Factors influencing inappropriate management included a light clinical examination compared to systematic clinical examination (OR 4.83, 95% CI: [1.55–15.08], p = 0.007), perceiving the patient as "at risk" (OR 7.05, 95% CI [2.55–19.50], p < 0.001), repeated consultations for the same reason (OR 4.13, 95% CI [1.51–11.27], p = 0.006), and diagnostic uncertainty (OR 4.99, 95% CI [1.40–17.71, p = 0.013). Diagnosis explanation decrease the risk of inappropriate management (OR 0.41, 95% CI [0.17–0.99], p = 0.047]. Physicians who received visits from pharmaceutical industry representatives were more likely to improperly prescribe antibiotics (OR 4.59, 95% CI [1.51–13.95].

Conclusions

While inappropriate antibiotic prescribing for URIs remains significant, this study highlights the complexity of the clinical reasoning underlying this practice. The study identifies modifiable factors, including conducting a thorough clinical examination, patient explanation, and independence from pharmaceutical promotion. It also highlights the critical role of managing diagnostic uncertainty. Rapid diagnostic tests and antibiotic prescription software help to reduce uncertainty, but it is equally important to learn to work with uncertainty in daily practice. These findings remain valid in the post-COVID period and call for ongoing medical education and public health policies aimed at promoting prudent antibiotic use in general practice.

Upper respiratory infections

antibiotics

general practice

clinical decision making

Numerous efforts are underway to preserve the effectiveness of antibiotics and enhance the efficiency of patient care [1]. In France, three national plans were implemented between 2001 and 2016, resulting in an 11.4% reduction in antibiotic prescriptions [2, 3]. The majority of antibiotic prescriptions are dispensed in outpatient clinics.

Controlling antibiotic resistance requires appropriate antibiotic prescribing. Upper respiratory tract infections (URIs) and bronchitis are common reasons for consulting general practitioners, especially during the winter months, accounting for 47% of antibiotic prescriptions in urban areas [4]. However, these infections are predominantly viral and do not require antibiotic prescriptions [5–7]. As a result, these infections contribute significantly to the overuse of antibiotics.

The French Infectious Diseases Society (SPILF) has issued recommendations for antibiotic prescribing in URIs [6, 8]. Situations wherein antibiotic therapy is not immediately recommended include nasopharyngitis; cases of purulent or mucopurulent nasal secretions; congestive otitis in children; seromucosal otitis in children; purulent acute otitis media (AOM) in children over 2 years of age with mild symptoms; angina with a negative Rapid Diagnostic Test for Streptococcus (RDT) in children; or a Mac-Isaac score of less than 2 in adults. French general medicine educators have developed free online tools such as Antibioclic® and Antibiolor® to support decision making regarding antibiotic prescriptions [9, 10].

The PAAIR study, which was conducted in 2001 among general practitioners in the Ile-de-France region, analysed antibiotic prescriptions for presumed viral respiratory infections, specifically URI infections and bronchitis, which are diagnoses that are more likely to lead to inappropriate antibiotic prescribing. This study led to the development of "nonprescription" strategies (PAAIR 2 [11]) by identifying eleven "at-risk" prescribing situations [12]. These critical situations that are associated with a high risk of inappropriate antibiotic prescribing included patient requests for antibiotics, past experiences of medical failure, initiation of antibiotic treatment, patient narratives of painful personal or family ordeals, prioritization of antibiotic consultation, competition among physicians, second or third consultations for the same episode, patient fatigue, perception of a "high-risk" patient, physician's doubt about viral origin, and discordance between physician's conviction and recommendations [13]. Furthermore, nine effective strategies for avoiding antibiotic prescriptions consisted of anticipating situations where nonprescription might pose a problem, deciding not to prescribe before negotiation, explicit communication of the patient's request, a precise and ritualized clinical examination, clarification of the differences between viruses and bacteria, explanation of antibiotic actions and adverse effects, and suggesting noninfectious causes.

These findings are consistent with a qualitative meta-synthesis of the literature, which identified multiple factors influencing inappropriate antibiotic prescribing. Intrinsic factors related to physicians included elements tied to their training and years in practice, as well as their attitudes, such as complacency, fear, ignorance, and diagnostic uncertainty or the desire for expedient problem resolution. Extrinsic factors included public health recommendations and policies, patient clinical presentation and history, physician‒patient and physician-family relationships, time constraints, and financial incentives, particularly from pharmaceutical companies [14].

After 15 years of national efforts to combat antibiotic resistance, data from Santé Publique France indicate a recent resurgence in antibiotic prescriptions since 2021, especially among children [15]. France continues to be one of Europe's largest consumers of antibiotics, largely due to prescriptions originating from outpatient clinics. Given this concerning trend, we aim to assess how factors contributing to inappropriate antibiotic prescribing by general practitioners have evolved 15 years after the PAAIR studies, focusing on upper respiratory infections, including angina, otitis, rhinopharyngitis, bronchiolitis, tracheitis, laryngitis, and acute community sinusitis. Additionally, we include bronchitis, categorized as a lower respiratory infection7, as it presents similar prescribing challenges.

Therefore, our study seeks to determine the proportion of recommended and nonrecommended antibiotic prescriptions by general practitioners in Alsace for URIs and to identify the clinical, contextual, and physician-related factors influencing these prescribing decisions.

A prospective observational study was conducted by fifteen volunteer interns in general medicine who were undergoing training with general practitioners. These practitioners provided their consent to participate in the study, and they were not informed of the specific objectives, thus preventing any influence on their clinical practice.

The data were collected as part of the ECOGEN study, which was conducted between 2011 and 2012 by the CNGE (National College of General Practitioner Teachers). The primary objective of the ECOGEN study was to characterize the distribution of reasons for consultation and associated health issues in general practice in France [16, 17]. Trainee investigators observed their GPs’ consultations with all patients and collected data, including patient socioeconomic information, reasons for consultation, procedures performed, consultation outcomes, and prescriptions (see Appendix 1). All data were coded using the second International Classification of Primary Care (ICPC2) [18]. Trainee investigators did not participate in the consultations, and the consultation procedures and results were solely the responsibility of the observed general practitioners.

In cases of consultations for URI, a specific grid was added to the ECOGEN grid. This supplementary grid was developed based on findings from the PAAIR study [13] and relevant literature (see Appendix 2). Physicians were asked to rate their satisfaction with the consultation, the diagnosis, and their estimate of whether the URI had a viral or bacterial origin using a visual analogue scale.

Furthermore, each of our GPs provided information about their sociodemographic characteristics and practice methods (see Appendix No. 3).

The required number of URI consultations for this study was determined based on an expected rate of inappropriate antibiotic prescribing of 50% [19]. Including 200 observations would provide an estimate of this rate with a 95% confidence interval width of +/- 7% (95% CI: 43% − 57%). This level of precision was considered sufficient. The proportion of consultations involving URI during the winter period in a prior study conducted in the same region was 12.5%, which required a total of 1,600 consultations to be observed. The trainees investigators observed 3 to 6 randomly selected half-days of consultations with each of their GPs.

Prior to data collection in November 2016, a half-day training session on data collection and ICPC2 coding was organized, followed by a test phase. Regular working meetings with the trainees investigators were held throughout the data collection period to assist them in coding the data.

All data were anonymized at the source, with an alphanumeric code randomly assigned to each intern, doctor and consultation.

For statistical analysis, we defined "appropriate" and "inappropriate" management of URI. Appropriate management aligned with the prescription or nonprescription of antibiotics following recommendations from the French Infectious Diseases Society (SPILF) for each URI. We evaluated appropriateness based on the indication for antibiotic therapy, without considering the specific molecule, dosage, or duration.

Statistical analysis was performed using Shiny Stats software, with support from statisticians at the Groupe de Méthode en Recherche Clinique (GMRC) at Strasbourg University Hospitals [20]. Bivariate analyses included Fisher’s exact test, the chi-squared test, and the Mann‒Whitney test.

Multivariate analyses were conducted using a mixed logistic regression model, including a random "doctor" effect to account for the nonindependence of observations among patients consulting the same doctor. We built three separate multivariate models due to the large number of variables, categorized as follows: patient clinical characteristics, contextual elements of the consultation, and physician profiles.

This study was approved by the Strasbourg Faculty Ethics Committee, and no conflict of interest is to be declared. Consultations were collected after the verbal consent of each patient.

A total of 807 usable consultations were gathered between December 2016 and April 2017 (see the follow chart in Figure 1). The intended goal of 1600 consultations was not achieved, and this discrepancy was attributed to the withdrawal of three interns and the refusal of participation by three doctors. On average, the interns observed a lower number of consultations per half-day than anticipated. However, it is noteworthy that the proportion of consultations involving URI of 21.4% was higher than expected. Of the 807 general consultations collected, a total of 173 were related to URIs (see Table 1). Among these URIs, 57 consultations (33%) were for paediatric patients (aged < 18 years), while 116 consultations (67%) were for adults. URIs represented 40.7% of the paediatric consultations and 17.4% of the adult consultations.

Table 1. Study population and overall characteristics of consultations.

		Total population n = 807	Adult population, n = 667	Child population, n = 140
Age - average (wks)		32.19 (1.8)	45.31 (1.6)	5.47 (0.57)
Gender M - n (%)		77 (44.5)	50 (43.1)	30 (52.6)
Seen in consultation (vs. visit) - n (%)		157 (90.8)	104 (89.7)	53 (93.0)
Number of URI consultations - n (%)		173 (21.4)	116 (17.4)	57 (40.7)
Consultation time in min - average (sem)		13.98 (0.5)	14.5 (0.62)	13.04 (0.82)
No. of reasons for consultation - n (weeks)		3.35 (0.1)	3.66 (0.13)	2.70 (0.14)
No. of procedures - n (sem)		2.87 (0.09)	3.13 (0.12)	2.33 (0.09)
No. of consultation results - n (sem)		1.68 (0.08)	1.81 (0.11)	1.42 (0.09)
No. of drugs prescribed - n (sem)		3.29 (0.14)	3.67 (0.18)	2.53 (0.19)
No. of URIs/patient - n (%)	1	132 (76.3)	87 (75.0)	45 (78.9)
	2	38 (22)	27 (23.3)	11 (19.2)
	3	3 (1.7)	2 (1.7)	1 (1.9)
Rhinopharyngitis - n (%)		94 (54.3)	57 (49.1)	37 (64.9)
Angina - n (%)		19 (11)	13 (11.2)	6 (10.5)
Acute otitis media - n (%)		7 (4)	4(3.5)	3 (5.2)
Other ear infections - n (%)		20 (11.6)	6 (5.2)	14 (24.5)
Otitis total - n (%)		27 (15.6)	10 (8.6)	17 (29.7)
Laryngitis - n (%)		7 (4)	3 (2.6)	4 (7)
Sinusitis - n (%)		13 (11.2)	13 (11.2)	0
Bronchitis - n (%)		43 (24.9)	41 (35.3)	2 (3.5)
Bronchiolitis - n (%)		3 (1.7)	1 (0.9)	2 (3.5)
Tracheitis - n (%)		10 (5.8)	9 (7.8)	1 (1.7)
Antibiotic prescription rate - n (%)		76 (44)	56 (48.3)	20 (35)
Appropriate care rate - n (%)		130 (75.1)	83 (71.6)	47 (82.5)

Legend: sem: standard error of mean

We observed that the rate of inappropriate management was 24.9%, with a 95% confidence interval of 18.6% to 32.0%.

The study involved a total of 25 practitioners, all of whom were university training supervisors practising across the Alsace region (as shown in Table 2). Notably, the physician population in our study had a higher representation of females compared to the general Alsatian physician population. The average age of the practitioners was 53 years, and their average period of practice was 21.7 years, ranging from 3 years to 43 years.

Table 2. Description of the practitioner population compared with Alsatian practitioners (source ERASME).

	Study GP	Alsatian GPs in 2017
Total (MG workforce)	25	1977
Age (sem)	53,2 (1,95)	56,1*
Gender (H) (%)	13 (52%)	1223 (61,9%)
Rural	5 (20%)	323 (16,3%) **
Urban	10 (40%)	-
Semirural	12 (48%)	-
Alone	6 (24%)	1167 (59,0%)
Group	19 (76%)	810 (41,0%)
Cslts/year (sem)	5014,71 (247,50)	4866

*Sem could not be calculated

**Municipalities with fewer than 2,000 inhabitants. The rural, semirural and urban breakdown is a declarative estimate by doctors, not comparable with health insurance data.

Most of these practitioners were part of group practices. Additionally, several practitioners possessed specialized skills or were engaged in specific medical practices, including geriatrics, working with disabled individuals, firefighting, osteopathy, service with doctors’ service, and medical pedagogy.

Regarding their practice statistics, the average number of annual consultations reported by these practitioners was 5014 (range: 2500 to 7384), which was similar to the regional average. These physicians also maintained an average weekly work volume of 41 hours, with hours ranging from 20 to 56 hours.

Furthermore, the majority of these practitioners were proactive in their professional development, with 87.5% attending continuing medical education courses and 73% subscribing to medical journals. Nearly two-thirds of the practitioners were receiving visits from medical representatives.

Observations of upper respiratory tract infections (URIs)

The majority of URIs observed in our study were cases of rhinopharyngitis, and 32 cases were combined with an additional diagnosis of URI (as detailed in Table 3). Notably, our findings indicated that the management of bronchitis and otitis cases was the least appropriate. Our analysis did not reveal any statistically significant influence stemming from factors such as the patient's social status, age, gender, socioprofessional background, social category, or the presence of an ALD (long-term condition) on the appropriateness of management for URIs.

Table 3. Antibiotic prescription and appropriate management by type of HRI.

	Total	Atb n (%)	Appropriate care n (%)
Total URI/patients	173	76 (44%)	130 (75.1%)
Rhinopharyngitis	94	23 (24.5)	72 (76.6)
Rhinopharyngitis alone	62	10 (16.1)	52 (83.9)
Angina	19	9 (47.3)	17 (89.5)
OMA	7	7 (100)	6 (85.7)
Other ear infections	20	10 (50)	11 (55)
Otitis tot	27	17 (63)	17 (63)
Laryngitis	7	0 (0)	7 (100)
Maxillary sinusitis	9	6 (66.7)	6 (66.7)
Frontal sinusitis	4	3 (75)	3 (75)
Total sinusitis	13	9 (69.2)	9 (69.2)
Bronchitis	43	29 (67.4)	23 (53.5)
Bronchiolitis	3	2 (66.7)	1 (33.3)
Tracheitis	10	8 (80)	6 (60)

Legend: rhinopharyngitis alone: diagnosis with no other associated URI. Adapted management: management according to SPILF recommendations in terms of whether or not to prescribe antibiotics, regardless of choice of molecule, dosage or duration. Patients could have several URI diagnoses at the same time.

Prescription patterns and decision-making factors

In our study, the most frequently prescribed antibiotics were amoxicillin, followed by amoxicillin-clavulanic acid, as summarized in Table 4. Bronchitis displayed the highest diversity in prescribed antibiotic molecules.

Table 4. Antibiotic molecules prescribed by type of URI.

Atb prescribed	n	Rhino-pharyngitis only	Angina	Otitis	Sinusitis	Laryngitis	Bronchitis	Bronchiolitis	Tracheitis
Diagnostics (n)	173	62	19	27	13	7	43	3	10
Antibiotics prescribed n (%)	76 (44)	10 (16.1)	9 (47.3)	17 (63)	9 (69.2)	0 (0)	29 (67.4)	2 (66.7)	8 (80)
amoxicillin	33	2	7	9	1	-	11	1	2
amoxicillin-clavulanic acid	14	2	1	1	7	-	3	-	-
amoxicillin/Fucidin	1	1	-	-	-	-	-	-	-
cefixime	3	2	-	1	-	-	-	-	-
cefpodoxime	6	1	-	3	-	-	2	-	-
ceftriaxone	1	-	-	-	-	-	1	-	1
cefuroxime	2	-	-	-	1	-	1	-	-
clarithromycin	6	2	-	-	-	-	4	1	3
ofloxacin	1	-	1	1	-	-	-	-	-
ofloxacin auric.	2	-	-	2	-	-	-	-	-
pristinamycin	2	-	-	-	-	-	2	-	2
roxithromycin	5	-	-	-	-	-	5	-	-

A patient could have several URIs. Atb: antibiotic

Factors influencing clinical decisions

Table 5 shows the clinical factors influencing GP management.

Table 5. Clinical decision factors.

	Total 173 n (%)	Appropriate n (%) 130	Inappropriate n (%) 43	p	OR [95% CI]
Number of HRIs - average (sem)	1.25 (0.04)	1.21 (0.04)	1.36 (0.08)	0.07	0.5 [0.3-1.1]
Significant general signs	14 (8.1)	9 (64.3)	5 (35.7)	0.46	0.6 [0.2-2.0]
Fever	63 (36.4)	43 (68.3)	20 (31.7)	0.31	0.7 [0.4-1.4]
Purulent discharge	3 (1.7)	1 (33.3)	2 66.7)	0.17	0.2 [0.1-2.1]
Clear expectoration	4 (2.3)	1 (25)	3 (75)	0.07	0.1 [0.01-1.2]
Purulent nasal secretion	7 (4.O)	2 (28.6)	5 (71.4)	0.007	0.1 [0.02-0.73]
Evolution ≥ 10 days	24 (13.9)	17 (70.8)	7 (29.2)	0.81	0.9 [0.3-2.3]
Good condition	116 (67.1)	90 (77.6)	26 (22.4)	0.03	2.1 [1.1-4.3]
Patient at risk	40 (23.1)	22 (55.0)	18 (45.0)	0.005	0.3 [0.2-0.7]
Difficult social context	7 (4.1)	2 (28.6)	5 (71.4)	0.008	0.14 [0.03-0.73]
Consultation same reason	35 (20.2)	16 (45.7)	19 (54.3)	<0.001	0.2 [0.1-0.5]

Legend: Nb URI: average number of upper respiratory infections recorded per patient. The results are given as numbers and percentages for categorical variables and as means and sem for continuous variables.

Patient's General Condition: We found that the practitioner's perception of the patient being in "good condition" was associated with more appropriate prescribing practices (odds ratio [OR] = 2.1, 95% confidence interval [CI]: 1.1-4.3, p=0.03). This indicates that when physicians perceived patients as generally healthy, they were more likely to make appropriate prescribing decisions.
Patient's "At-Risk" Status: Conversely, when a patient was considered "at risk," which typically meant they had underlying comorbidities (e.g., diabetes, heart failure, kidney failure), the management was less likely to be appropriate (OR = 0.3, 95% CI: 0.2-0.7, p=0.005). Patients deemed "at risk" due to their medical conditions received less appropriate care.
Difficult Social Context: Patients facing challenging social contexts, often indicated by their status as beneficiaries of universal health coverage, were associated with less appropriate care (OR = 0.14, 95% CI: 0.03-0.73, p=0.008). This suggests that social factors influenced prescribing decisions.
Frequent consultations for the same reason were linked to less suitable care. If a patient had a history of prior consultations for the same medical concern, the appropriateness of care decreased (OR = 0.2, 95% CI: 0.1-0.5, p<0.001).

Notably, other clinical decision factors, such as the presence of fever, duration of symptom evolution, number of upper respiratory infections presented by the patient, and smoking status, did not demonstrate statistical significance in bivariate analysis.

When examining decision-making elements related to physicians' profiles and clinical reasoning, our bivariate analyses did not uncover any significant differences associated with the age of the physicians, their practice mode in solo or group settings, or in rural or urban locations, the number of years of practice, or the workload.

In our study, the majority of doctors conducted commented or ritualized clinical examinations, as presented in Table 6. Notably, when physicians performed light clinical examinations, this was associated with a 3.3-fold increase in inappropriate management (p=0.002, OR: 0.3, 95% CI: 0.1-0.6). This indicates that a more comprehensive clinical examination was linked to more appropriate care decisions.

Table 6. Practitioner reasoning, uncertainty and adherence strategies.

		Total 173	Appropriate n (%) 130	Inappropriate (%) 43	p	OR
Terms and conditions the consultation	Demand for antibiotics	13 (7.6)	7 (53.8)	6 (43.2)	0.1	0.4 [0.1-1.2]
	Clinical examination with commentary	126 (72.8)	93 (73.8)	33 (26.2)	0.34	1.5 [0.7-3.2]
	Ritualized clinical examination	106 (61.3)	73 (68.9)	33 (31.1)	0.14	0.6 [0.3-1.2]
	Brief clinical examination	29 (17.7)	14 (48.3)	15 (51.7)	0.002	0.3 [0.1-0.6]
Consultation conclusion stage	Explanation of diagnosis	124 (72.1)	96 (77.4)	28 (22.6)	0.03	2.2 [1.1-4.6]
	Explanation of evolution	90 (52.0)	66 (73.3)	24 (26.7)	0.9	1.1 [0.5-2.1]
	Helping you choose the right treatment	10 (7.0)	7 (70.0)	3 (30.0)	0.89	1.1 [0.3-4.4]
	Explanation of the decision	101 (58.4)	74 (73.3)	27 (28.7)	0.87	1.1 [0.5-2.1]
	Alternatives	28 (16.2)	23 (82.1)	5 (17.9)	0.2	1.9 [0.7-5.3]
	Follow-up proposal	22 (12.8)	14 (63.6)	8 (36.4)	0.3	0.6 [0.2-1.6]
	Deferred prescription	7 (9.7)	1 (14.3)	6 (85.7)	0.2	0.3 [0.03-2.4]
Doctor's overall impression	EVA diagnostic certainty - avg (sem)	82.2 (1.7)	83.5 (2.1)	78.6 (2.7)	0.2	1.0 [0.99-1.03]
	Viral orientation - n (%)	90 (57.3)	81 (90.0)	9 (10.0)	<0.001	10.5 [4.5-24.2]
	Bacterial orientation - n (%)	42 (26.8)	24 (57.1)	18 (42.9)	0.02	0.4 [0.2-0.9]
	Situations of diagnostic doubt - n (%)	25 (15.9)	7 (28.0)	18 (72.0)	<0.001	0.1 [0.04-0.26]
	EVA Satisfaction with ATB prescription - avg (sem)	87.1 (1.4)	91.3 (1.3)	76.5 (3.3)	<0.001	1.06 [1.03-1.08]
	EVA Satisfaction with consultation as a whole - avg (sem)	85.6 (1.3)	87.9 (1.3)	79.7 (2.8)	0.005	1.03 [1.01-1.05]

Legend: Results are given as numbers and percentages for categorical variables and as means and sem for continuous variables. *Decisional comfort of the practitioner measured by analogical self-assessment (VAS) from 0 to 100. VAS: visual analogical scale. VAS scores are measured from 0 to 100. VAS Diagnostic certainty 0: uncertain, 100: certain. Satisfaction VAS 0: not satisfied, 100: satisfied. The VAS odds ratio is estimated for one point of difference out of 100.

Explaining the diagnosis was relatively common and associated with more appropriate management (p=0.03, OR: 2.2, 95% CI: 1.1-4.6).

No treatment bargaining was observed in our study. Additionally, the explicit request for antibiotics by patients did not emerge as a significant decision factor influencing the appropriateness of care.

These findings suggest that thorough clinical examinations and effective communication of diagnoses played pivotal roles in determining the appropriateness of care decisions in our study.

Multivariate analysis

In the first multivariate analysis model (Table 7), we examined clinical factors associated with inappropriate patient management. In cases where "fever" was present, the care was significantly 3.65 times less appropriate (95% CI: [1.54; 8.65], p<0.01). When a "light examination" was conducted, the care was significantly 4.83 times less appropriate (95% CI: [1.55; 15.08], p<0.01). Patients perceived as "at risk" received URI care that was 7.05 times less appropriate (95% CI: [2.55; 19.50], p<0.001).

Table 7. Clinical characteristics associated with inappropriate HRI treatment: multivariate analysis.

	OR	95% CI	p
Evolution ≥ 10j	2.25	[0.62-8.09]	0.22
Fever	3.65	[1.54-8.65]	0.004**
Significant general signs	2.04	[0.43-9.77]	0.37
Brief clinical example	4.83	[1.55-15.08]	0.007**
Good condition	0.54	[0.22-1.31]	0.17
Patient at risk	7.05	[2.55-19.50]	<0.001**
Child	1.22	[0.50-2.97]	0.66

In the second multivariate analysis model (Table 8), which focused on the specific context of the consultation, we found that explaining the diagnosis was associated with a 2.44-fold higher likelihood of appropriate care (OR: 0.41, 95% CI: [0.17; 0.99], p=0.047). Offering follow-up increased the risk of inappropriate care by a factor of 5.45 (95% CI: [1.53; 19.40], p=0.009). A recent consultation for the same reason increased the risk of inappropriate care by a factor of 4.13 (95% CI: [1.51; 11.27], p=0.006). Diagnostic doubt increased the risk of inappropriate care by a factor of 5 (95% CI: [1.40; 17.71], p=0.013).

Table 8. Inappropriate management associated with contextual elements of the consultation: multivariate analysis.

	OR	95% CI	p
Social context	2.09	[0.16-27.52]	0.57
Explanation of diagnosis	0.41	[0.17-0.99]	0.047
Therapeutic alternatives	0.61	[0.20-1.89]	0.39
Nondrug alternatives	0.46	[0.20-1.1]	0.08
Follow-up proposal	5.45	[1.53-19.40]	0.009
Recent consultation (same reason)	4.13	[1.51-11.27]	0.006
Demand for antibiotics	0.74	[0.14-3.89]	0.72
Diagnostic doubt	4.99	[1.40-17.71]	0.013

Notably, variables related to social context, therapeutic alternatives, and nonmedicinal advice were not significantly associated with the quality of care.

In the third multivariate analysis (Table 9), the only variables significantly associated with inappropriate patient management were the number of drugs prescribed for conditions other than URI treatments (OR: 1.36, 95% CI: [1.07-1.73], p=0.01) and receiving visits from medical representatives (OR: 4.59, 95% CI: [1.51-13.95], p=0.007). Variables such as consultation time, the number of drugs prescribed specifically for URI, physician practice characteristics, and workload were not significantly associated with inappropriate management.

Table 9. Inappropriate management compared with practitioners' descriptive data: multivariate analysis.

	OR	95% CI	p
Time (/additional min)	0.998	[0.94-1.06]	0.96
No. of non-URI drugs	1.36	[1.07-1.73]	0.01
No. of drugs for URI	0.998	[0.71-1.41]	0.99
Age of doctor (/additional year)	0.96	[0.91-1.02]	0.91
Type of MG	1.79	[0.72-4.43]	0.21
Medical visitor	4.59	[1.51-13.95]	0.007
No. of annual consultations	1.20	[0.66-2.17]	0.55
Exercise alone	0.54	[0.09-3.26]	0.50
Weekly working hours (/additional hour)	0.96	[0.88-1.05]	0.41
Participates in shifts	0.83	[0.29-2.42]	0.74
Rural practice (vs. urban/suburban)	2.21	[0.62-7.95]	0.22

General results

This study of 807 outpatient consultations in general practice analysed practices in real-life conditions for the management of upper respiratory tract infections (URIs), which are typically not in need of antibiotic therapy. Among the 176 consultations subjected to analysis, antibiotics were prescribed in 35% of cases, and management was deemed appropriate in 75% of cases, with no notable differences observed between paediatric and adult patients.

However, it is noteworthy that certain infections, such as sinusitis and bronchitis, were more likely to result in inappropriate antibiotic prescriptions. Several factors were identified as favouring inappropriate management, including a light clinical examination, recurrent consultations for the same reason, the patient's underlying health conditions, a noisy clinical presentation, and diagnostic uncertainty.

Interestingly, we observed few instances of patient bargaining, and there were no significant disparities in the length of consultations when care was appropriate. Importantly, appropriate care was associated with enhanced practitioner satisfaction. Finally, it was evident that interactions with laboratory sales representatives substantially increased the risk of inappropriate patient management.

In essence, our study sheds light on the complex landscape of managing upper respiratory infections in general practice, highlighting both areas of appropriate care and opportunities for improvement.

Strengths and limitations

Strengths

This study was a collaborative effort involving numerous interns and practising doctors. The strength of this approach lies in its reliance on robust data collection methods inspired by the ECOGEN study [17]. It relied on detailed data from routine clinical practice, encompassing various elements, including clinical and contextual information, detailed prescriptions, and physician characteristics. Another strength of this study is its reliance on objective data rather than declarative elements. This approach contributes to the reliability and credibility of the study's findings. Physicians who participated in the study were unaware of its specific objectives, and they were accustomed to being observed by trainees. This helped mitigate potential bias in their practice, enhancing the study's validity. The study employed a meticulous coding system using ICPC-2 codes [21], allowing for the comprehensive analysis of patient clinical and contextual data. This detail-rich approach adds depth to the study's findings.

Limitations

The doctors observed in this study were all university training supervisors, which means they may not be entirely representative of all practising GPs [22]. Key differences include their practice settings (group practices) and the gender and age distribution compared to Alsatian GPs. Despite training and refresher courses, there may still be some variability in how consultations were collected and assessed, as the coding was carried out by interns who might have had a different perspective from the practitioners. However, multiple checks were in place to validate the coding. This study was conducted prior to the COVID-19 pandemic, and medical practices may have evolved during and after the pandemic. While there were notable changes in prescribing practices at the onset of the pandemic, prescription levels returned to prepandemic levels by the end of 2022 [23]. Hence, the study's results and discussions remain relevant to current practice.

To go further

The rates of inappropriate management of antibiotics for upper respiratory infections (URIs) in this study were notably lower than the 50% rate reported in a reference study from the United States [19] but similar to European data. For instance, in England, 38.7% of antibiotic prescriptions were deemed inappropriate for infectious respiratory illnesses between 2010 and 2015, with amoxicillin being the most frequently prescribed antibiotic [24]. Additionally, a recent study in Spain estimated that 28% of antibiotic prescriptions by dentists were inappropriate [25].

France has been dedicated to a robust public health policy against antibiotic resistance, aligning with new targets set for 2025 by the World Health Organization (WHO) [26]. This policy appears to be yielding positive results, especially when compared to the United States, where 50% of antibiotic prescriptions for URIs were inappropriate in 2016. However, the significant post-COVID increase in antibiotic prescriptions, particularly in children [15], underscores the ongoing need to sustain efforts in medical training and public health policy to combat inappropriate antibiotic use.

Notably, as observed in the PAAIR 1 study, the quality of the doctor‒patient relationship did not emerge as a predictive factor for the final outcome. Furthermore, consultation time did not increase when antibiotics were not prescribed [13]. In fact, doctors prescribed fewer antibiotics when they employed nonprescribing strategies [27]. This suggests that a detailed and/or commented clinical examination can enhance patient understanding and confidence in the doctor's approach.

Antibiotics were more frequently prescribed in scenarios characterized by fever, extended symptom duration (especially upon patient return), and in cases where patients were considered medically fragile due to comorbidities such as cardiac or renal insufficiency and diabetes. Last, diagnostic uncertainty emerged as a substantial influence on the appropriateness of management. Several studies have highlighted the role of uncertainty in inappropriate antibiotic prescriptions [28–30]. Strategies aimed at reducing diagnostic uncertainty, such as rapid diagnostic tests for conditions such as sore throat [31] or in-office capillary CRP tests [32], can serve as effective tools in limiting antibiotic prescribing. Prescribing aids that provide practitioners with current data also facilitate appropriate antibiotic prescriptions and educate patients about antibiotic necessity [9, 33]. GPs can rely on these tools to open dialogue with their patients by sharing information with them about the nonindication of antibiotic therapy [34, 35]. Finally, general medicine is a medical field in which uncertainty is commonplace. Learning how to cope with this uncertainty by offering to follow patients' conditions if necessary is a way of postponing prescribing while reassuring patients [36, 37].

The impact of pharmaceutical representatives has not yet been studied specifically in this situation. It was identified as a notable influence, multiplying the risk of inappropriate prescribing by more than four. This aligns with recent research on pharmaceutical company gifts and prescriptions, indicating that doctors who did not accept gifts from pharmaceutical companies prescribed fewer drugs, including antibiotics [38].

In conclusion, inappropriate antibiotic prescribing for upper respiratory infections is not a convenient prescription but is rooted in intricate clinical reasoning. GPs cope with clinical data in a context of uncertainty while considering the patient's overall health. However, modifiable factors have been identified, including conducting comprehensive clinical examinations, patient education through explanation, and maintaining independence from pharmaceutical promotion. Noisy pathologies such as bronchitis, otitis, and sinusitis, which may trigger concerns of clinical deterioration, are more susceptible to inappropriate management. To address this, reinforcing the benign nature of these conditions in initial training and promoting rapid diagnostic tests and prescribing aids are effective ways of limiting antibiotic prescriptions. Ultimately, managing uncertainty, proposing patient monitoring if necessary, and resisting pharmaceutical influence are critical facets of responsible antibiotic stewardship in general practice.

AOM acute otitis media

GP general practitioner

RTD rapid diagnostic test

URI upper respiratory infection

VAS visual analogic scale

Ethics approval and consent to participate

This study was approved by the Ethical Comitee of Strasbourg university hospital

Consent for publication

Not applicable

Competing interests

The authors declare that they have no competing interests

Funding

Not applicable

Authors' contributions

LC coordinated the study, managed the data wrote the first version of the manuscript and corrected the final version of the manuscript. FS did the statistical analysis. YH contributed to the methodology and expertise of the subject, an corrected the manuscript. JC supervised the work and wrote the final version of the manuscript. All authors read and approved the final manuscript.

Acknowledgements

We aknowledge the 15 students who collected the data and the GPs who accepted to be observed for the study.

We aknowledge Springer Nature author service for English and formatting corrections

Le Corvoisier P, Renard V, Roudot-Thoraval F, Cazalens T, Veerabudun K, Canoui-Poitrine F, et al. Long-term effects of an educational seminar on antibiotic prescribing by GPs: a randomised controlled trial. Br J Gen Pract. 2013:63;e455-64.
Agence Nationale de Sécurité du Médicament et des Produits de Santé. La consommation des antibiotiques en France entre 2000 et 2020. In :Rapport de juillet 2023. ANSM. 2023. https://ansm.sante.fr/actualites/lansm-publie-un-rapport-sur-la-consommation-des-antibiotiques-entre-2000-et-2020.Accessed 27 oct 2023.
Menard C, Fégueux S, Heritage Z, Nion-Huang M, Berger-Carbonne A, Bonmarin I. Perceptions and attitudes about antibiotic resistance in the general public and general practitioners in France. Antimicrobial Resistance & Infection Control. 2022;11:124.
Laxminarayan R, Duse A, Wattal C, Zaidi AK, Wertheim HF, Sumpradit N, et al. Antibiotic resistance-the need for global solutions. Lancet Infect Dis. 2013:13;1057-98.
Cavalié P, Hider-Mlynarz K, ANSM. La consommation d’antibiotiques en France en 2016. France, UK: ANSM; 2017.
Société de Pathologie Infectieuse de Langue Française. Antibiothérapie par voie générale en pratique courante dans les infections respiratoires hautes de l’adulte et l’enfant. In: Recommandations de bonne pratique. SPILF. 2011. http://www.infectiologie.com/UserFiles/File/medias/Recos/2011-infections-respir-hautes-recommandations.pdf. Accessed 03 Nov 2023.
Agence Française de Sécurité Sanitaire des Produits de Santé. Antibiothérapie par voie générale en pratique courante au cours des infections respiratoires basses de l’adulte et l’enfant. In: Recommandations de bonne pratique. AFSSAPS. 2005. https://apimed-pl.org/contenu/uploads/2019/12/AIR_irb_argu.AFSSAPS.2005.pdf. Accessed 3 Nov 2023.
Société de Pathologie Infectieuse de Langue Française. 15e conférence de consensus en thérapeutique anti-infectieuse. Prise en charge des infections des voies respiratoires basses de l’adulte immunocompétent. Méd Mal Infect. 2006:36;235-44.
Antibioclic : Antibiothéraphie rationnelle en soins primaires [Internet]. Paris. 2011. https://antibioclic.com. Accessed 03 Nov 2023.
Antibiolor. Antibioguide – Référentiel lorrain d’antibiologie en établissements de soins. 9^e édition. Mediacom; Nancy: 2019. https://www.antibioest.org/wp-content/files/2019/03/Antibioguide_V9-déf.-1-1.pdf. Accessed 03 Nov 2023
Cittee D, Adeline D. Comment ne pas prescrire d’antibiotiques dans les infections respiratoires présumées virales? (les enseignements de PAAIR). France, UK: UPEC; 2013.
Attali C, Renard V. Etude PAAIR 2 premiers résultats quantitatifs. France, UK: CNGE; 2006.
Attali C, Amade-Escot C, Ghadi V, Cohen JM, Pouchain D, Huas D, et al. Infections respiratoires présumées virales: comment prescrire moins d'antibiotiques? Résultats de l'étude PAAIR. Rev Du Prat Méd Gén. 2003155-61.
Teixeira Rodrigues A, Roque F, Falcão A, Figueiras A, Herdeiro MT. Understanding physician antibiotic prescribing behaviour: a systematic review of qualitative studies. Int J Antimicrob Agents. 2013:41;203-12.
Santé publique France . Prescriptions d’antibiotiques en médecine de ville : reprise en 2021 [infographie] [Internet]. SPF. 2022. https://www.santepubliquefrance.fr/maladies-et-traumatismes/infections-associees-aux-soins-et-resistance-aux-antibiotiques/resistance-aux-antibiotiques/documents/infographie/prescriptions-d-antibiotiques-en-medecine-de-ville-reprise-en-2021-infographie. Accessed 03 Nov 2023.
Carron M. ECOGEN RESPI : étude des résultats de consultation associés à un motif respiratoire en médecine générale. exercer, la revue française de médecine générale. France, UK: Université de Nice-Sophia Antipolis; 2015.
Letrilliart L, Supper I, Schuers M, Darmon D, Boulet P, Favre M, et al. ECOGEN: étude des Eléments de la consultation en médecine genérale. Exercer. 2014:25;148-57.
Gay B. La CISP, une classification pour l'étude ECOGEN et la recherche en soins de santé primaires. Exercer. 2014:25;147.
Fleming-Dutra KE, Hersh AL, Shapiro DJ, Bartoces M, Enns EA, File TM, Jr, et al. Prevalence of inappropriate antibiotic prescriptions among us ambulatory care visits, 2010-2011. JAMA. 2016:315;1864-73.
Fabacher T, Schaeffer M, Tuzin N, Séverac F, Lefebvre F, Mielcarek M, et al. Medical biostatistics with GMRC shiny stats - learning by doing. Ann Pharm Fr. 2020:78;499-506.
Jamoulle M, Roland M, Humbert J, Brûlet JF. Traitement de l’information médicale par la classification internationale des soins primaires CISP-2 deuxième version. Belgium, UK: Care Editions ASBL; 2000.
Bouton C, Leroy O, Huez JF, Bellanger W, Ramond-Roquin A. Représentativité des médecins généralistes maîtres de stage universitaires. Santé Publique. 2015:27;59-67.
Nandi A, Pecetta S, Bloom D. Global antibiotic use during the COVID-19 pandemic: analysis of pharmaceutical sales data from 71 countries, 2020–2022. EClinicalMedicine. 2023:57;101848.
Nowakowska M, van Staa T, Mölter A, Ashcroft DM, Tsang JY, White A, et al. Antibiotic choice in UK general practice: rates and drivers of potentially inappropriate antibiotic prescribing. J Antimicrob Chemother. 2019:74;3371-8.
Rodríguez-Fernández A, Vázquez-Cancela O, Piñeiro-Lamas M, Herdeiro MT, Figueiras A, Zapata-Cachafeiro M. Magnitude and determinants of inappropriate prescribing of antibiotics in dentistry: a nation-wide study. Antimicrob Resist Infect Control. 2023:12;20.
World Health Organization. France: a broad vision of infection prevention and control and the importance of local action.https://www.who.int/news-room/feature-stories/detail/france--a-broad-vision-of-infection-prevention-and-control-and-the-importance-of-local-action. Accessed 3 May 2023.
Attali C, Rola S, Renard V, Roudot-Thoraval F, Montagne O, Le Corvoisier P, et al. Situations cliniques à risque de prescription non conforme aux recommandations et stratégies pour y faire face dans les infections respiratoires présumées virales. Exercer. 2008:82;66-72.
Vasconcelos M. Antibiotics and uncertainty of diagnosis in viral respiratory infections: point-prevalence survey across 15 European countries. Acta Paediatr. 2023:112;1970-2.
Arnold SR, To T, McIsaac WJ, Wang EE. Antibiotic prescribing for upper respiratory tract infection: the importance of diagnostic uncertainty. J Pediatr. 2005:146;222-6.
Horwood J, Cabral C, Hay AD, Ingram J. Primary care clinician antibiotic prescribing decisions in consultations for children with RTIs: a qualitative interview study. Br J Gen Pract. 2016:66;e207-13.
Humair JP, Revaz SA, Bovier P, Stalder H. Management of acute pharyngitis in adults: reliability of rapid streptococcal tests and clinical findings. Arch Intern Med. 2006:166;640-4.
Cooke J, Llor C, Hopstaken R, Dryden M, Butler C. Respiratory tract infections (RTIs) in primary care: narrative review of C reactive protein (CRP) point-of-care testing (POCT) and antibacterial use in patients who present with symptoms of RTI. BMJ Open Respir Res. 2020:7;e000624.
Burty C, Rabaud C, Empereur F, Doco-Lecompte T, Lozniewski A, Demoré B, et al. Réseau lorrain d’antibiologie: bilan après 2 ans de fonctionnement et perspectives. Metz: Antibiolor; 2006. https://www.infectiologie.com/UserFiles/File/medias/JNI/JNI06/COL/col3-Burty.pdf. Accessed 03 Nov 2023.
Vazquez-Cancela O, Souto-Lopez L, Vazquez-Lago JM, Lopez A, Figueiras A. Factors determining antibiotic use in the general population: a qualitative study in Spain. PLoS One. 2021:16;e0246506.
Coxeter P, Del Mar CB, McGregor L, Beller EM, Hoffmann TC. Interventions to facilitate shared decision making to address antibiotic use for acute respiratory infections in primary care. In: The Cochrane database of systematic reviews, 2015(11), CD010907; 2015. https://doi.org/10.1002/14651858.CD010907.pub2. Accessed 03 Nov 2023.
Liu C, Wang D, Duan L, Zhang X, Liu C. Coping with diagnostic uncertainty in antibiotic prescribing: a latent class study of primary care physicians in Hubei, China. Front Public Health. 2021:9;741345.
Coronado-Vázquez V, Canet-Fajas C, Delgado-Marroquín MT, Magallón-Botaya R, Romero-Martín M, Gómez-Salgado J. Interventions to facilitate shared decision-making using decision aids with patients in primary health care: a systematic review. Medicine (Baltimore). 2020:99;e21389.
Goupil B, Balusson F, Naudet F, Esvan M, Bastian B, Chapron A, et al. Association between gifts from pharmaceutical companies to French general practitioners and their drug prescribing patterns in 2016: retrospective study using the French transparency in healthcare and national health data system databases. BMJ. 2019:367;l6015.

No competing interests reported.

Appendix.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

Factors influencing inappropriate management of upper respiratory infections in general practice

Status:

Version 1

Abstract

Background

Methods

Results

Conclusions

Figures

Background

Methods

Results

Discussion

General results

Strengths and limitations

Strengths

Limitations

To go further

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1