Link between prescriptions and the electronic health record.

Area of Science:

• Health Informatics
• Clinical Documentation
• Medication Management

Background:

• Electronic Health Records (EHRs) are crucial for patient care but often lack complete medication indication data.
• Problem lists within EHRs are intended to track patient conditions but may not always correlate with prescribed medications.
• Accurate documentation of medication indications is essential for safe and effective pharmacotherapy.

Purpose of the Study:

• To evaluate the extent to which medication prescriptions are linked to a documented indication in EHR problem lists.
• To assess the performance of the MEDication Indication (MEDI) resource and its High Precision Subset (MEDI-HPS) in matching prescriptions to indications.

Main Methods:

• Analysis of prescriptions and problem lists for 62,191 patients between January 1 and June 30, 2015.
• Utilized RxNorm codes within MEDI to map prescriptions to International Classification of Diseases, 9th edition (ICD-9) codes for problem lists.
• Established a reference standard by manual review of 30 randomly selected patients' data by two independent pharmacists to evaluate MEDI precision and recall.

Main Results:

• MEDI matched 61% of prescriptions to a potential indication, while MEDI-HPS matched 38%.
• MEDI demonstrated a precision of 47% and recall of 57%, whereas MEDI-HPS achieved a precision of 79% and recall of 96%.
• Excluding short-term prescriptions (≤14 days) yielded slightly improved, though not statistically significant, results.

Conclusions:

• EHR problem lists appear incomplete, as most patients lacked a documented indication for their prescribed medications.
• MEDI identified indications for 61% of prescriptions, compared to 38% with MEDI-HPS.
• MEDI mapped 37% of problems to prescriptions, while MEDI-HPS mapped only 21%.