Home Health & Hospice Week

Accuracy, specificity for comorbidity codes another looming problem.

If you aren’t beefing up your diagnosis coding practices now to prepare for the Patient-Driven Groupings Model’s arrival in January, you could already be behind.

“Coding is going to be absolutely critical under PDGM,” stresses consultant Pam Warmack with Coding Connections in Ruston, Louisiana.

“With PDGM, correct coding is essential to obtain accurate reimbursement,” says coding expert Joan Usher with JLU Health Record Systems in Pembroke, Massachusetts. “PDGM relies more heavily on the patient’s principal diagnosis and comorbidities, allowing up to 25 diagnoses on the claim,” Usher explains.

“Diagnosis coding will have a much greater impact in PDGM” than it does under the current prospective payment system, says Lisa Woolery, director of coding education for Fazzi Associates in Northampton, Massachusetts.

Why? Under PDGM, a patient’s primary diagnosis will determine which of 12 clinical groups the patient is placed in under the first step of the case mix system, explains Corinne Kuypers-Denlinger with Quality in Real Time in Floral Park, New York. And then the other 24 secondary diagnoses listed on the claim can contribute to the case mix system’s comorbidity adjustment.

Based on the secondary diagnoses, a patient’s episode can be assigned no adjustment, a low adjustment, or a high adjustment for comorbidities (see more details in Eli’s HCW, Vol. XXVIII, No. 39-40). Woolery estimates a low adjustment will average about a $100 increase, and a high adjustment will average about a $300 bump-up per 30-day billing period.

Beware The Many Boomerang Codes

Along with the new emphasis on diagnosis coding will come a host of problems, experts warn.

Number one on the list is claims that will Return to Provider (RTP) because their primary diagnosis codes are not on the PDGM grouper list. That means they can’t be placed into one of these 12 clinical case mix groups: Musculoskeletal Rehabili­tation, Neuro/Stroke Rehabilitation, Wounds — Post-Op Wound Aftercare and Skin/Non-Surgical Wound Care, Behavioral Health Care (including Substance Use Disorder), Complex Nursing Interventions, and seven Medication Management, Teaching and Assessment (MMTA) subgroups.

“With PDGM, most non-specific and all symptom codes will no longer be allowed primary,” explains Kyle Johnson, owner and director of coding operations with Home Health Coding Solutions in Brigham City, Utah.

Data has suggested that up to 40 percent of the diagnoses currently used as primary under PPS will not be allowed under PDGM, Warmack cautions. That accounts for up to 15 percent of current claims, she says.

The problem: Home health agency staff must derive the code from the referring physician’s information. Johnson expects that come Jan. 1, docs will “send over referrals to home health with invalid, unallowed primary diagnoses,” he tells Eli. “Diagnoses like Osteoarthritis non-specific or Muscle weakness or unsteady gait or abnormality of gait are no longer allowed as primary,” he emphasizes. (See 30 common unallowed codes in chart, p. 155.)

“The agency cannot submit a claim with a code that does not fit into a grouper,” stresses attorney and coding expert Lisa Selman-Holman with Selman-Holman & Associates in Denton, Texas. And “the coder cannot just change the code to fit into a grouper without verifying the diagnosis with the physician,” Selman-Holman warns.

The Centers for Medicare & Medicaid Services formerly called the diagnoses that weren’t in the grouper “Questionable Encounter” codes. Now, “we laughingly refer to them as ‘boomerang codes’ because they will fly right back to you if you try to use them as primary,” Selman-Holman tells Eli.

Sometimes, the problematic codes will not be defined with enough specificity. But sometimes they will be codes that are not allowed at all. Either way, the HHA must reach out to the physician to determine correct coding.

The problem with the unallowed codes is that sometimes, they really are the correct ones. “There are situations where by coding conventions, a symptom code is appropriate as primary (for symptoms not attributed to a specific diagnosis and adverse effects), but the grouper will not permit their use,” Woolery explains.

For example: Epistaxis as an adverse effect of Coumadin can be the correct code because “the residual effect is a symptom code which must be coded before the adverse effect code,” Woolery offers. The same goes for Bradycardia as an adverse effect of Digoxin toxicity.

Another example: “A patient with pain or a gait disturbance as a late effect of a fracture” will be returned under PDGM. “The pain and gait codes must be coded before the fracture late effect code, but these are not acceptable as primary diagnoses” under PDGM, Woolery notes.

Warmack hopes to see CMS tweak some of its allowed primary diagnosis codes in the 2020 final PPS rule that is expected in late October or early November.

One good point: At least claims with unacceptable primary diagnoses will RTP instead of deny, Usher points out. “That means once corrected, they may be resubmitted,” she tells Eli.

But watch out: Even when HHA staff do secure a new, valid primary diagnosis from the physician, they then have to make sure that the codes, the face-to-face physician encounter documentation, the plan of care, and the HHA documentation all agree, Woolery highlights.

Note: Links to CMS’s PDGM grouper tool and the 2019 PPS final rule that finalized PDGM are at www.cms.gov/center/provider-Type/home-Health-Agency-HHA-Center.html.