Risk Score Impact of a Diabetic Neuropathy Query
A Medicare Advantage plan is reviewing a 72-year-old female member (community, non-dual) whose demographic score is 0.400. The chart supports CHF (HCC 85, coefficient 0.350) and type 2 diabetes without complications (HCC 38, coefficient 0.105). A neurology note says 'neuropathy' and lists gabapentin, but neither the neurology note nor the primary care note states that the neuropathy is diabetic. The plan's base capitation rate is $10,000 per member per year. The plan average risk score is 1.18 versus a national average of 1.00. The coding team is deciding whether to query the physician before the submission deadline.
1. Calculate the member's current risk score using the given coefficients. What would the score be if the physician confirms diabetic neuropathy (HCC 37, coefficient 0.302) and the hierarchy removes HCC 38? Show the dollar impact at the $10,000 base rate.
Read the worked answer
Current score: 0.400 + 0.350 + 0.105 = 0.855.
If diabetic neuropathy is confirmed, HCC 37 (0.302) replaces HCC 38 (0.105) because of the diabetes hierarchy. New score: 0.400 + 0.350 + 0.302 = 1.052.
Difference: 1.052 - 0.855 = 0.197 risk-score points. At a $10,000 base rate, that is 0.197 x $10,000 = $1,970 additional capitation for this member, before any plan-level normalization.
Do not add HCC 37 and HCC 38 together; the hierarchy counts only the most severe diabetes HCC.
2. Can the coder assign E11.42 (type 2 diabetes with diabetic polyneuropathy) based on the current documentation? What action is required and why?
Read the worked answer
No. E11.42 is a combination code for type 2 diabetes with diabetic polyneuropathy. The record must show that the provider linked the neuropathy to diabetes. 'Neuropathy' plus a diabetes diagnosis, or gabapentin use, is not enough.
Query the physician: 'Is the neuropathy diabetic in etiology?' If yes, code E11.42. If no, code the diabetes and the neuropathy separately as supported. Inferring the link would create an unsupported HCC 37.
3. The plan's average risk score is 1.18. Does that alone prove upcoding? What payment and RADV considerations apply?
Read the worked answer
A score of 1.18 versus 1.00 does not by itself prove upcoding. Risk scores reflect age and sex, HCCs, model version, and the plan's population.
Payment is based on the plan's average risk score, so a higher average raises capitation. However, RADV compares each submitted HCC to the medical record. Unsupported diagnoses are deleted, error rates are calculated, and the plan may face extrapolated refunds and corrective action. The plan should audit documentation and queries, not assume the score is wrong.