The long-stay depressive symptoms measure counts residents whose mood interview, or the staff assessment that stands in for it, shows depression on their most recent assessment. It is on Care Compare but not in the Five-Star rating. Texas and New York both pay on it.
Where the measure is used
- CMS ID N030.03, a long-stay measure built from the MDS. A long-stay resident is one with 101 or more cumulative days in the building.
- Care Compare: yes. Five-Star: no. The long-stay QM rating uses seven MDS measures and two claims measures, and this isn’t one of them.
- Texas QIPP scores it as Component 3, Metric 1 for state fiscal year 2027. See Texas QIPP, explained.
- New York’s NHQI scores it among the 13 quality measures in its 2025 methodology.
What puts a resident in the numerator
“The measure reports the percentage of long-stay residents who have had symptoms of depression during the 2-week period preceding the MDS 3.0 target assessment date.”
There are two ways in, one for each version of the PHQ. Whichever one was done on the target assessment, both parts have to be true.
| Resident interview (PHQ-2 to 9) | Staff assessment (PHQ-9-OV) | |
|---|---|---|
| Part 1: a core symptom | D0150A2 or D0150B2 coded 2 or 3 (half or more of the days) | D0500A2 or D0500B2 coded 2 or 3 |
| Part 2: the total | D0160 from 10 to 27 | D0600 from 10 to 30 |
Item A is little interest or pleasure in doing things. Item B is feeling (or appearing) down, depressed or hopeless. A total of 10 or more without either of them at 2 or 3 does not count, and neither does a core symptom with a total of 9.
The PHQ-2 does some of this sorting for you. If both D0150A2 and D0150B2 are 0 or 1, the interview ends after two questions and D0160 is their sum, so that resident can’t meet the measure.
Who is left out
- Residents who are comatose, or whose comatose status (B0100) is missing.
- Residents not in the numerator whose resident interview and staff assessment are both incomplete: a core item dashed or skipped, or a total that is missing or, for the interview, coded 99.
There is no risk adjustment. The rate CMS reports is the observed rate.
When it counts, and when it clears
Only the target assessment matters. That is the resident’s latest qualifying assessment (an OBRA assessment, a PPS 5-Day or an OBRA discharge) in the episode, dated no more than 120 days before the episode ends or the quarter closes. Earlier assessments are not scanned.
So the measure follows the resident’s mood. The PHQ asks about the last 2 weeks, and the interview is done during the assessment’s look-back period. When symptoms ease, the next target assessment shows it.
A worked example. On Resident A’s quarterly, D0150A2 is 2 (7 to 11 days), D0150B2 is 1, and the PHQ-9 total is 11. Part 1 and Part 2 are both met, so Resident A is in the numerator. Three months later, after treatment, both D0150A2 and D0150B2 are 1. The interview stops at the PHQ-2 with a total of 2, and Resident A is out.
New in the October 2026 RAI Manual
Version 1.20.11 of the RAI Manual, effective October 1, 2026, adds a rule for buildings that interview more than once in the look-back period:
“If multiple resident mood interviews (PHQ-2 to 9©) are conducted during the look-back period, code the MDS with the resident mood interview (PHQ-2 to 9©) conducted closest to the ARD.”
Common coding mistakes
Using the staff assessment instead of the interview. The staff assessment is only for residents who can’t be interviewed (D0100 = 0), such as a resident who is rarely or never understood, or who needs an interpreter who isn’t available. It isn’t a backup when the interview was missed or went badly.
“Do not complete the Staff Assessment of Resident Mood items (D0500) if the resident interview should have been conducted but was not done, or if the assessment being completed is a stand-alone Part A PPS Discharge assessment.”
Explaining the answer away. A resident who says they’re tired because of their COPD is still tired. The frequency is coded as the resident gives it.
“Record the resident’s responses as they are stated, regardless of whether the resident or the assessor attributes the symptom to something other than mood.”
Losing the interview to blanks. If three or more of the nine frequency items are blank or dashed, the interview is not complete and D0160 is coded 99. The staff assessment isn’t done in that case either, so the resident drops out of the measure for the quarter, and the record has no usable mood score.
Reading the score as a diagnosis. The PHQ records symptoms. It doesn’t diagnose depression, and a coded total isn’t a reason to add one. What the score should do is reach the physician so the symptoms get a follow-up.
Two smaller rules worth knowing: if the resident can’t choose between two frequencies, code the higher one; and if an item has two parts with different frequencies, use the higher.
See your building’s depressive symptoms rate against the nation, free.
SuperQM reads each resident’s latest PHQ against the two-part rule, so the team sees who is in the measure and why before the quarter closes.
Sources
- MDS 3.0 Quality Measures User’s Manual v18.0 (effective January 1, 2026), Table 2-26 and Chapter 1
- MDS 3.0 RAI User’s Manual v1.20.11 (October 2026), Chapter 3, Section D
- Care Compare Five-Star Quality Rating System: Technical Users’ Guide, September 2026
- Texas HHSC, Quality Incentive Payment Program (QIPP)
- New York State Department of Health, 2025 Nursing Home Quality Initiative Methodology
