The influence of fears of perceived legal consequences on general practitioners' practice in relation to defensive medicine - a cross-sectional survey in Germany

Development of a scale for defensive medicine practices in primary care

BACKGROUND

Defensive medicine, characterized by the avoidance of high-risk patients or the overuse of diagnostic and therapeutic interventions due to fear of litigation or complaints, poses significant challenges in primary care. Existing scales primarily focus on hospital settings, limiting their applicability to family physicians. This study aimed to develop and validate a scale tailored to measure defensive medicine behaviors in primary care contexts.

METHODS

Scale development involved reviewing the literature, conducting focus group discussions with family physicians, and consulting with experts. An initial 37-item draft was pilot-tested for clarity, resulting in a refined instrument. Data were collected from family physicians in two phases: exploratory factor analysis (EFA) with 252 participants and confirmatory factor analysis with 266 participants. Criterion validity was assessed by correlating the new scale with a previously validated defensive medicine scale. Reliability was evaluated using Cronbach's alpha and McDonald's omega.

RESULTS

EFA supported a five-factor structure: referral, complaint, e-report, verbal/physical violence, and interventional procedure dimensions. After item removal, the final 21-item scale demonstrated acceptable goodness-of-fit indices (χ²/df = 2.43, root mean square error of approximation (RMSEA) = 0.07, comparative-fit index (CFI) = 0.93). Criterion validity was evidenced by a moderate positive correlation (r = 0.350, P < .01) with an existing scale. Internal consistency was high, with both Cronbach's alpha and McDonald's omega at 0.92 for the total scale.

CONCLUSIONS

The Primary Care Defensive Medicine Practices Scale is a valid and reliable instrument specifically tailored for primary care settings. It offers a more nuanced understanding of defensive behaviors, guiding targeted interventions to enhance care quality, patient safety, and resource utilization.

Healthcare AI, explainability, and the human-machine relationship: a (not so) novel practical challenge

This paper focuses on the lack of explainability that afflicts machine-learning-based AI systems applied in the field of healthcare. After a brief introduction to the topic, from both a technical and legal point of view, this work aims to assess the main consequences that the lack of explainability has on the human-machine relationship in clinical care, through a practical perspective. It then questions whether explainability is truly an objective worth seeking and, if so, to what extent, taking into account the current possible solutions.

[Factors affecting diagnosis coding in primary care - A qualitative study]

BACKGROUND

Coding diagnoses is part of a physician's daily routine. Good coding quality has several functions; it is, for example, the basis for the remuneration of contract doctors. In the field of general medicine, this process can, however, pose a particular challenge, as diagnoses are naturally more often excluded than made. In addition, the process of diagnosing informs treatment.

METHODS

23 semi-structured interviews were conducted with general practitioners from Schleswig-Holstein. The interviews were recorded with a dictaphone, pseudonymized, transcribed and then analyzed by content analysis according to Mayring.

RESULTS

87 % (N = 20) of the participants were male, the mean age was 57 years, and the mean duration of work experience was 23 years. The ICD-10 (International Classification of Diseases) coding system was often criticized for not representing general medical practice and placing an increased administrative and time burden on physicians. At the same time, diagnostic pressure from external economic interests was perceived. Specifications of practice management systems or their additional programming options were used, among other things, to search for a code. Particular challenges arose in this context, e.g., complex clinical pictures and lack of time. Prescriptions without indication, among other things, have been made at the request of patients, due to a lack of time and concerns about doctor switching. In these cases, a diagnosis in the ICD-10 code that justified the therapy was sometimes used.

CONCLUSIONS

The path leading from clinical findings to therapy and finally to the appropriate diagnostic code is a complex one, and it involves various influencing factors, including non-medical factors. These influencing factors must be taken into account in future secondary data analyses. The ICD-10 is not the ideal choice for mapping these processes in general medicine.

Low-value practices in primary care: a cross-sectional study comparing data between males and females in Spain

INTRODUCTION

Overuse of medical services is a challenge worldwide, posing a threat to the quality of care, patient safety and the sustainability of healthcare systems. Some data suggest that females receive more low-value practices (LVPs)-defined as medical interventions that provide little or no benefit to patients and can even cause harm-than males. This study aims to evaluate and compare the occurrence of LVPs in primary care among both males and females.

DESIGN

A retrospective study was conducted.

SETTING

Primary care in the Alicante province (Spain) during 2022.

PARTICIPANTS

Data were extracted from the digital medical records of 978 936 patients attended by 1125 family physicians across 262 primary healthcare centres in the Alicante province.

OUTCOME MEASURES

Data on age, sex, diagnosis and treatment were extracted. The primary outcome measure was the frequency of 12 selected LVPs prescribed to male and female patients. These LVPs were expected to be relatively frequent occurrences with the potential to cause harm.

RESULTS

A total of 45 955 LVPs were identified, of which 28 148 (5.27% of 534 603, CI95% 5.20-5.32) were prescribed to female patients and 17 807 (4.00% of 444 333, CI95% 3.95-4.06) to male patients (x², p value <0.0001). The most common LVPs were prescribing treatment for overactive bladder without excluding other pathologies that may cause similar symptoms (30.87%), using hypnotics without having a previous aetiological diagnosis in patients with difficulty maintaining sleep (14%) and recommending analgesics (NSAIDs, paracetamol and others) for more than 15 days per month in primary headaches that do not respond to treatment (13.33%).

CONCLUSIONS

Future clinical training, management and research must consider biological differences or those based on gender factors when analysing the frequency and causes of LVP.

TRIAL REGISTRATION NUMBER

NCT05233852.