Four years after COVID emptied hospital wards, Medicare inpatient volume has not recovered. Across the 20 highest-volume diagnosis-related groups in 2019, 17 had not returned to 90% of their pre-pandemic discharge counts by 2023. This isn't a story about a system still healing. It's a story about a system that may have permanently contracted.

The Numbers That Define the Gap

The most extreme case is DRG 470, Major Hip and Knee Joint Replacement. In 2019, Medicare recorded 413,865 discharges under that code. By 2023, that figure had fallen to 71,904, a recovery rate of just 17.4%. No other top-20 DRG comes close to that collapse.

The contrast with DRG 291, Heart Failure and Shock with MCC, illustrates how uneven the picture is. Heart failure, a condition that rarely offers patients a choice about hospitalization, recovered to 80.9% of 2019 levels, with 319,367 discharges in 2023 versus 394,926 in 2019. That's still below the 90% threshold, but it's a fundamentally different trajectory than elective orthopedics. The gap between 80.9% and 17.4% is the widest among the 20 highest-volume DRGs, and it maps almost perfectly onto the distinction between conditions patients can defer and conditions they cannot.

DRG 194, Simple Pneumonia with CC, recovered to only 38.4% of 2019 levels, with 37,534 discharges in 2023 against 97,746 in 2019. Septicemia without MV (DRG 872) reached 69% recovery, with 104,269 discharges in 2023 compared to 151,210 in 2019. Even conditions that would seem to track closely with population health are running well below their pre-pandemic baselines. For hospitals that built capacity and staffing around 2019 volumes, these numbers represent a structural revenue shortfall that has now persisted for four consecutive years.

States That Never Stopped Falling

The state-level picture adds another layer of concern. Virginia showed the largest decline in recovery among the states analyzed, reaching only 67.9% of its 2019 baseline by 2023. What makes Virginia's trajectory particularly striking is that it fell 8.9 percentage points from its 2020 level of 76.8%. The state was already down sharply from 2019 at the height of the pandemic, and it kept declining.

New York and New Jersey followed similar paths. New York's inpatient discharges fell to 77.4% of 2019 levels by 2023, declining further from 79.3% in 2020. New Jersey lost 0.9 percentage points from 2020 to 2023, and Massachusetts lost 1.5 percentage points over the same period. These are not states that hit a floor and stabilized. They are states where inpatient volume continued to erode after the acute phase of the pandemic ended.

California, by comparison, recovered to 82.7% of its 2019 baseline by 2023, with 438,107 discharges against 529,503 in 2019. Hawaii gained 2.6 percentage points from 2020 to 2023. The divergence between California and Virginia (82.7% versus 67.9%) is a 14.8 percentage point gap that has real consequences for hospital systems operating in those markets.

What a 17.4% Recovery Rate Actually Means

Return to DRG 470 for a moment. A recovery rate of 17.4% means that for every six hip or knee replacements Medicare was paying for in 2019, only about one is being billed as an inpatient procedure today. Ambulatory surgical centers and outpatient settings have absorbed some of that volume, driven by CMS's removal of total hip and knee replacement from the inpatient-only list. But the scale of the shift is large enough that it has reshaped the financial model for orthopedic service lines at hospitals across the country.

The broader pattern, 17 of the top 20 DRGs still below 90% of 2019 volume four years later, suggests the inpatient channel itself has been permanently repriced in the minds of patients, physicians, and payers. Hospitals that planned capital expenditures, staffing ratios, and debt service around 2019 discharge volumes are now operating in a structurally different environment, and the 2023 data shows no sign of convergence back to the old baseline.

Whether the patients who stopped coming to hospitals in 2020 shifted to outpatient settings, delayed care indefinitely, or simply never sought treatment remains the central unanswered question in understanding what this volume loss actually cost in health outcomes.