GP coding behaviour for non-specific clinical presentations: a pilot study.

Area of Science:

• Primary Care Research
• Health Informatics
• Medical Coding

Background:

• Accurate clinical coding in primary care is crucial for disease incidence studies using electronic health records (EHRs).
• Existing code validation methods are insufficient for non-specific conditions and do not capture General Practitioners' (GPs) coding decision-making.
• Qualitative research can illuminate GPs' coding behaviors, particularly for ambiguous presentations.

Purpose of the Study:

• To explore GPs' decision-making processes when assigning Read codes to non-specific clinical presentations.
• To use Lyme disease as a specific case example to understand coding behaviors in primary care.

Main Methods:

• A pilot study involving masked, semi-structured interviews with eight GPs in North West England.
• Interviews were based on 11 clinical cases simulating Lyme disease presentations.
• Thematic analysis was applied to interview transcripts to identify coding behavior themes.

Main Results:

• Key themes influencing coding included GP experience, clinical evidence, diagnostic uncertainty, professional integrity, and patient-reported information.
• GPs considered Lyme disease in their differential diagnosis for half of the cases presented.
• Specific Lyme disease Read codes were selected in only two out of eleven cases.

Conclusions:

• GPs exhibit reluctance in applying specific diagnostic Read codes for vague or unfamiliar symptoms.
• The study introduces a masked questionnaire methodology as a novel approach for validating Read codes of non-specific conditions.
• This coding reluctance presents challenges for primary care EHR research, necessitating further investigation into underlying drivers of GP coding behavior.