Database Evaluation for Muscle and Nerve Diseases - DEMAND: An academic neuromuscular coding system

Abstract

Background: A database which documents the diagnosis of neuromuscular patients is useful for determining the types of patients referred to academic centers and for identifying participants for clinical trials and other studies. The ICD-9 or ICD-10 numeric systems are insufficiently detailed for this purpose. Objective: To develop a database for neuromuscular diagnoses Methods: We developed a detailed diagnostic coding system for neuromuscular diseases called DEMAND: Database Evaluation for Muscle and Nerve Diseases that has been adopted by neuromuscular clinics at University of Texas Health Science Center San Antonio (UTHSCSA), Ohio State University (OSU), University of Kansas Medical Center (KUMC), and University of Texas Southwestern (UTSW). At the initial visit, patients are assigned a diagnostic code which can be revised later if appropriate. Fields include patient’s name, date of birth, and diagnostic code. The neuromuscular database consisted of 457 codes. Each code has a prefix (MUS or PNS) followed by a three-digit number. Depending on whether muscle or nerve is primarily involved, there are eight broad groups: motor neuron disease (MUS codes 100-139); neuromuscular junction disorders (MUS 200-217); acquired and hereditary myopathies (MUS 300-600s); acquired and hereditary polyneuropathies (PNS 100-400); mononeuropathies (PNS 500s); plexopathies (PNS 600s); radiculopathies (PNS 700s); and mononeuritis multiplex (PNS 800s). Results: During a period of 10 years, 17,163 of patients were entered (1,752 at UTHSCSA, 1,840 at OSU, 3,699 at KUMC, 9,872 at UTSW). The number of patients in several broad categories are: 3,080 motor neuron disease; 1,575 neuromuscular junction disease; 1,851 muscular dystrophies; 633 inflammatory myopathies; 1,090 hereditary neuropathies; 1,001 immune-mediated polyneuropathies; 620 metabolic/toxic polyneuropathies; 535 mononeuropathies; 296 plexopathies; and 769 radiculopathies. Conclusion: A detailed diagnostic neuromuscular database can be utilized at multiple academic centers. The database should be simple without too many fields to complete, to ensure compliance during busy clinic operations. This database has been very useful in identifying groups of patients for retrospective, observational studies and for prospective treatment studies including trials for Amyotrophic Lateral Sclerosis (ALS), Muscular Dystrophies (MD), Myasthenia Gravis (MG), and retrospective studies of Primary Lateral Sclerosis (PLS), chronic inflammatory demyelinating neuropathy (CIDP), etc.