Medical Billing and Coding: A Patient-Facing Overview

Medical billing and coding sit at the intersection of clinical care and financial transaction — the largely invisible machinery that converts a doctor's observations into a claim an insurance company can pay. This page explains how that system works, what patients actually experience when it misfires, and where the logical boundaries of the process lie. Understanding the basics makes it easier to recognize errors, ask sharper questions, and avoid surprises on the way to navigating the healthcare system.

Definition and scope

Every clinical encounter generates two parallel records: a medical record describing what happened, and a billing record describing what to charge for it. Medical coding is the act of translating the first into the second — assigning standardized numeric or alphanumeric codes to diagnoses, procedures, and supplies. Medical billing is the subsequent process of submitting those coded claims to payers (insurers, Medicare, Medicaid, or patients directly) and collecting payment.

The two dominant coding systems in the US are:

  1. ICD-10-CM (International Classification of Diseases, 10th Revision, Clinical Modification) — used to code diagnoses. The ICD-10-CM system contains over 70,000 codes (CDC, ICD-10-CM Official Guidelines), ranging from broadly defined conditions to remarkably granular ones (there is, famously, a distinct code for injuries sustained from a burning water ski).
  2. CPT codes (Current Procedural Terminology) — maintained by the American Medical Association and used to describe the services a provider performed, from a standard office visit to complex surgery.

A third system, HCPCS Level II, covers supplies, equipment, and non-physician services not captured in CPT.

The scope of this system is vast. The Centers for Medicare & Medicaid Services (CMS) processed over 1.5 billion Medicare fee-for-service claims in fiscal year 2022 (CMS Medicare FFS Claims Data), and that figure represents only one payer in a multi-payer landscape that includes hundreds of private insurers, state Medicaid programs, CHIP, and the Veterans Health Administration.

How it works

After a patient encounter, a clinical coder — or, in smaller practices, a billing staff member wearing both hats — reviews the provider's documentation and assigns the appropriate ICD-10 and CPT codes. The accuracy of those codes depends entirely on the quality of the underlying documentation; a physician who writes "patient has trouble breathing" leaves the coder in a far worse position than one who specifies "acute exacerbation of moderate persistent asthma."

The coded claim is formatted into an electronic transaction (typically an 837P for professional claims, 837I for institutional claims) and transmitted to the payer. The payer's adjudication system checks:

The payer then issues an Explanation of Benefits (EOB) — or, for Medicare, a Medicare Summary Notice — detailing what was billed, what was allowed, what the plan paid, and what remains the patient's responsibility. That patient-responsibility figure flows to healthcare costs and billing territory: deductibles, copayments, and coinsurance determined by the specific healthcare coverage options the patient holds.

Common scenarios

The gap between a clean claim and a confusing bill is often traced to a handful of recurring failure modes:

Upcoding and downcoding. Upcoding means assigning a more intensive — and expensive — code than the documentation supports. Downcoding, less discussed, means assigning a lesser code, often out of caution, which underpays the provider and can distort a patient's medical history. The Office of Inspector General (OIG) at HHS identifies upcoding as one of the most common forms of healthcare fraud, with improper payment estimates for Medicare alone exceeding $31 billion in fiscal year 2023 (HHS OIG Work Plan).

Bundling errors. Some CPT codes are "bundled" — the work they describe is considered included in a more comprehensive code already billed. Billing both separately triggers an "unbundling" flag and claim denial.

Out-of-network surprises. A patient may be in an in-network hospital but treated by an out-of-network anesthesiologist, radiologist, or surgeon. The codes are identical; the reimbursement rates are not. Federal protections under the No Surprises Act (effective January 1, 2022) now limit this exposure for most emergency and scheduled procedures (CMS No Surprises Act).

Coordination of benefits errors. When a patient carries two insurance policies — common among spouses on different employer plans — determining which insurer pays first follows a defined priority sequence. Errors here routinely produce duplicate billing or phantom balances.

Decision boundaries

Medical billing and coding do not determine what care a patient receives — that remains a clinical judgment. Coding does, however, directly shape whether a payer considers that care medically necessary and therefore reimbursable.

The distinction matters because a denied claim based on a coding error is not the same as a denied claim based on a coverage exclusion. The first is correctable through an appeal with corrected documentation; the second may require a different pathway through patient rights and protections or, depending on insurer, a formal grievance process.

Patients reviewing an EOB or a bill should specifically verify that the CPT codes on the bill match the services described in their medical records and health data rights documentation. A mismatch between a provider's notes and a submitted code is the single most actionable discrepancy a patient can catch — and dispute — without any specialized credentials.

Coding also feeds downstream into quality reporting, public health surveillance, and hospital reimbursement under value-based care programs. The code assigned to a discharge from a hospital stay affects the Diagnosis-Related Group (DRG) payment the facility receives — often by thousands of dollars — which is why coding accuracy is audited continuously by payers, providers, and CMS alike.

 ·   · 

References