- Physical therapy
- Module 6
- Section 16 of 17
Sections C and D in Depth, and a Field Guide to the Rest of the Instrument
This module already teaches you how to run these interviews. The SOC OASIS: Patient History, Screening & Cognitive Assessments walks the three memory words, the orientation questions, the recall procedure, the two mood questions and the seven that follow them, and the score bands those answers produce, and the OASIS-E2 Recording Prompts sheet carries the wording to read aloud in the home. None of that is repeated here.
What this section adds is the machinery around the interview, which is where the coding errors live: the item that decides whether the interview happens at all, the rules for stopping one that is going nowhere, the difference between a score of zero and an interview that never finished, the four rows of the delirium item that are not a yes or no answer, and the gateway that decides whether two mood questions become nine.
Then it closes with a field guide to the sections of the instrument this module has not given a section of its own, so you can open any OASIS screen in the system you are charting in and know what it wants and who normally answers it.
Section C, Cognitive Patterns, is nine items. Section D, Mood, is three.
| Item | What it asks |
|---|---|
| C0100 | Should the Brief Interview for Mental Status be conducted |
| C0200 | Repetition of three words |
| C0300 | Temporal orientation, scored as year, month, and day of the week |
| C0400 | Recall, scored once per word |
| C0500 | Summary score |
| C1310 | Signs and symptoms of delirium, scored as four separate rows |
| M1700 | Cognitive functioning |
| M1710 | When confused |
| M1720 | When anxious |
| D0150 | Patient mood interview |
| D0160 | Total severity score |
| D0700 | Social isolation |
Appendix B puts every one of them at start of care, resumption of care, and discharge, and none of them on a follow-up assessment. It also flags M1700, M1710, and M1720 for quality reporting and for value based purchasing, which is the short answer to why a cognitive item on a therapy note is worth getting right. How that scoring works for an agency is taught in HHVBP, Star Ratings, and QAPI, under Home Health Operations in Bonus Content, and is deliberately not re-taught here.
The Item That Decides Whether the Interview Happens
C0100 is an item, not a judgment call you make silently. It asks whether the Brief Interview for Mental Status should be conducted, and its answer controls everything after it.
The manual's instruction is to interact with the patient in their preferred language, make sure they can hear you or have access to their preferred method of communication, and offer alternatives such as writing, pointing, sign language, or cue cards if the patient appears unable to communicate. Then you determine whether the patient is rarely or never understood verbally, in writing, or using another method.
- Code 0, no, if the interview should not be conducted because the patient is rarely or never understood, cannot respond verbally or in writing or using another method, or an interpreter is needed but is not available. This skips C0200 through C0500.
- Code 1, yes, if the patient is at least sometimes understood and, if an interpreter is needed, one is available.
- A dash is valid here, and CMS expects dash use to be a rare occurrence.
The coding tip under this item is one line long and it is the whole philosophy of the section: attempt to conduct the interview with ALL patients. The item exists because a structured cognitive test is more accurate and reliable than observation alone, and because without an attempted structured interview a patient might be mislabeled based on their appearance or assumed diagnosis. A clinician who decides on the doorstep that this patient clearly cannot do a memory test has coded C0100 by not asking.
Stopping, Scoring Zero, and the Difference Between Them
The scoring rules that get missed turn on what counts as an answer rather than on the point values.
Dash is valid on the interview items. C0100, C0200, C0300, C0400, and C0500 all accept a dash, always with the same warning attached, that a dash indicates no information and CMS expects dash use to be a rare occurrence. That is a different rule from the one you carry out of the functional and skin sections, where dashes are mostly prohibited.
C1310 Is Four Observations, Not a Verdict
The delirium screen you learned as a four-question tool with a positive or negative result is not how C1310 is coded. C1310 is four separate rows, and three of them have a three-level answer that turns on whether the behavior fluctuates.
- C1310A, acute mental status change. Code 0, no, when there is no evidence of an acute change from baseline. Code 1, yes, when the patient has an alteration in mental status, observed or reported or identified, that represents an acute change from baseline. At discharge you compare current mental status to baseline before the discharge assessment period, not to the day of admission.
- C1310B, inattention. C1310C, disorganized thinking. C1310D, altered level of consciousness. Each carries Code 0 when the behavior is not present, Code 1 when it is continuously present and does not fluctuate, and Code 2 when it is present and fluctuates.
Two rules decide most of these codes and neither is intuitive.
Disagreement between sources is fluctuation. If information sources disagree about whether the behavior was present, the row is coded as fluctuating. If as few as one source notes fluctuation, the behavior is coded as fluctuating. And for inattention specifically, if any information source reports the symptom as present, the row cannot be coded as behavior not present.
Code 1 requires agreement. To code a behavior as continuously present, all sources must agree that it was consistently there. That is a higher bar than the word continuously suggests, and it is why a patient who was attentive for the family this morning and unfocused for you this afternoon is a Code 2 rather than a Code 1.
The manual also separates two things clinicians blur. Assess attention separately from level of consciousness, and if you want another way to test attention, ask the patient to count backwards from twenty. Disorganized thinking is about the shape of the speech, not its accuracy: a patient who gives the wrong year but converses normally about their own life is coded as behavior not present, because the answer was related to the question even though it was wrong.
Altered level of consciousness has four named states. Vigilant means startles easily to any sound or touch, lethargic means repeatedly dozes off but responds to voice or touch, stupor means very difficult to arouse and keep aroused, and comatose means cannot be aroused despite shaking and shouting.
Fluctuation itself has a definition worth carrying, because it is broader than what you see in the visit. Behavior that tends to come and go, or to increase or decrease in severity, counts, and it may be noted by you, reported by staff or family, or documented in the medical record.
The Mood Gateway and the Dash That OASIS-E2 Added
D0150 is scored in two columns, and almost every mood coding error is a column error.
Column one is symptom presence. Code 0, no, if the patient indicates the symptom is not present, and then enter zero in column two as well. Code 1, yes, if the symptom is present, and then code the frequency in column two. Code 9, no response, if the patient was unable or chose not to complete the item or responded nonsensically, and leave column two blank. A dash in column one means the symptom presence was not assessed, and column two stays blank.
Column two is symptom frequency, and the manual gives it four codes: Code 0 for never or one day, Code 1 for several days, Code 2 for half or more of the days, and Code 3 for nearly every day. Read the item as written and do not define the words for the patient, because the meaning has to be the patient's own. If the patient gives different frequencies for different parts of one item, or cannot choose between two frequencies, take the higher one.
The gateway is decided by column two, not by yes or no. After the first two questions, there are exactly two ways to end the interview early, and they are not the same ending.
- If both D0150A1 and D0150B1 are Code 9, leave both frequency cells blank, end the interview, and skip D0160 entirely.
- If both D0150A2 and D0150B2 are Code 0 or Code 1, end the interview, but still add those two frequency values and enter the total in D0160.
- In every other case, continue through D0150C to D0150I and complete D0160.
That middle rule is the one that catches people. A patient who answers yes to feeling down but reports it on only a few days has a symptom present and still does not trigger the remaining seven questions, and the assessment is not finished until the total severity score is entered anyway.
D0160 has its own arithmetic, and it is not always a simple sum. The maximum patient score is twenty-seven. The score is computed only when the patient answered the frequency for at least seven of the nine items, which is another way of saying that no more than two of them are missing. When all nine carry a value, D0160 is the simple sum of those values.
When one or two are blank or dashed, you omit the missing values, sum the rest, and then scale the result so a partial interview stays comparable with a complete one. With one missing item, sum the eight that remain and multiply by nine eighths. With two missing, sum the seven that remain and multiply by nine sevenths. Round to the nearest whole number either way.
If symptom frequency is blank or dashed for three or more items, the interview is deemed not complete and D0160 carries Code 99. And unlike almost everything else in these two sections, a dash is not a valid response for D0160. Most systems run that multiplication for you, but a clinician who hand-totals an interview with a gap in it and enters the raw sum has just under-reported the patient's mood.
What OASIS-E2 changed here. In OASIS-E1, a dash was not a valid response in column two. OASIS-E2 permits it, and adds a coding tip for the situation it exists to cover: in the rare situation that the patient cannot provide a frequency after answering yes in column one, enter a dash in column two, and CMS expects a dash response to be rare. The D0160 rules were rewritten to match, so a dashed frequency now counts as missing in the same way a skipped one does. Some systems have not caught up, so if yours still refuses a dash in that column, it is running the old rule.
The remaining Section D item is D0700, social isolation. Like B1300 health literacy in Section B, it is a patient self-report item and the manual says no other source should be used to identify the response. You do not code it from the daughter's opinion and you do not code it from the chart.
A Field Guide to the Rest of the Instrument
Nine sections of the instrument do not get a section of their own here, either because a therapist rarely owns them or because they are short. Here is what each one asks, who normally answers it, and the one rule worth carrying.
Section A, Administrative Information. Two halves: patient tracking, which is names, numbers, dates, and payment sources, and administrative information, which is the assessment itself. Office staff often seed the tracking half and the assessing clinician owns the rest, and agencies draw that line differently, so it is worth asking where it falls before the first assessment. Tracking items only need to be revised if the information changes during the episode.
Two things belong to you. M0090, date assessment completed, is the date the last information used to complete the assessment was gathered, and OASIS-E2 added a coding tip saying it may not coincide with a home visit at all: if you have to call the family or the physician to finish an item the patient could not answer, M0090 reflects that date even though no visit happened. M0100, assessment reason, is what tells the rest of the system which assessment this is. OASIS-E2 also renamed the old gender item to A0810, sex.
The demographic, language, and transportation items are taught in The SOC OASIS: Patient History, Screening & Cognitive Assessments, earlier in this module, and are not repeated here. At transfer and discharge the section also asks whether the current reconciled medication list went to the next provider and to the patient, in A2121 and A2123, which is the discharge end of the work described in Section N in Depth: Medications, the Drug Regimen Review, and the Clock.
Section B, Hearing, Speech, and Vision. Hearing and vision are covered in the patient history and screening section earlier in this module. The item worth naming here is B1300, health literacy, which asks how often the patient needs someone to help them read instructions, pamphlets, or other written material from their doctor or pharmacy. It is a patient self-report item and no other source may be used, and its time points are start of care, resumption of care, and discharge.
Section E, Behavior. Two items. M1740 lists cognitive, behavioral, and psychiatric symptoms and carries Code 7, none of the above behaviors demonstrated, for the common case. M1745 asks how often disruptive behavior occurs. Anyone doing the comprehensive assessment answers both, from interview, observation, and the record, and behaviors may be identified by a formal diagnosis or by your clinical judgment. The rule worth carrying: M1745 is not scoped to the list in M1740. Consider any and all disruptive or dangerous behaviors, including sleeplessness, sundowning, agitation, wandering, aggression, combativeness, and getting lost in familiar places. Neither item accepts a dash.
Section F, Preferences for Customary Routine Activities. M1100 identifies the patient's living arrangement and how often a caregiver other than agency staff is available. M2102 identifies the ability and willingness of those caregivers to provide each category of assistance. This is a therapist's section more often than it looks, because the discharge plan lives or dies on it.
Two rules: when the patient needs help with some parts of a category and not others, code the part that represents the most need, and when more than one response describes the caregiver, code the one that represents the caregiver's greatest barrier. A caregiver who helps but needs training is coded as needing training.
Section H, Bladder and Bowel. M1600 asks whether the patient was treated for a urinary tract infection in the past fourteen days, M1610 covers urinary incontinence or catheter presence of any type, M1620 covers bowel incontinence frequency, and M1630 covers an ostomy for bowel elimination. On a multidiscipline case these usually fall to nursing, and on a therapy only start of care they fall to you, so it is worth asking early how they are divided. The rule worth carrying is the window: the fourteen days are the two weeks immediately before the start of care date, with the start of care date counted as day zero. And M1620 asks how often the symptom happens, not what causes it.
Section I, Active Diagnoses. M1021 is the primary diagnosis, M1023 the other diagnoses, and M1028 asks whether two specific conditions are active, peripheral vascular or peripheral arterial disease and diabetes mellitus, with a none of the above answer. The assessing clinician determines the primary and other diagnoses from the assessment findings and the record, not from whatever the referral typed in.
M1028 needs a diagnosis documented by a physician or other authorized practitioner: documentation of a complication such as neuropathy does not let you assume diabetes, and a diabetic patient who also has peripheral arterial disease gets both answers. If you learn about an active diagnosis after the assessment timeframe closes, the assessment is not revised to reflect it. Those two conditions are on the form because they change functional outcomes and pressure injury risk, which is the connection to Section M in Depth: Skin, Staging, and Wounds You Do Not Own.
Section K, Swallowing and Nutritional Status. M1060 is height and weight, K0520 covers nutritional approaches such as tube feeding or parenteral nutrition, and M1870, feeding or eating, is the section's functional item and is taught in Section G in Depth: The Other Functional Ruler. The rule worth carrying belongs to M1060: only enter a height that has been directly measured by agency staff. A self-reported height, or one carried over from another provider's paperwork, is not acceptable, and both measurements round mathematically, height to the nearest whole inch and weight to the nearest whole pound.
Section O, Special Treatments, Procedures, and Programs. One item, O0110, and the screening list a therapist actually uses is already in The SOC OASIS: Clinical Systems & Functional Scoring. What that section does not say is that the item is collected at discharge as well as at start of care and resumption of care, so the special treatments question comes back at the end of the episode when the answer may well have changed.
Section Q, Participation in Assessment and Goal Setting. One item, M2401, the intervention synopsis, and it is the only place on the instrument that appears at transfer and discharge and nowhere else. It asks whether specific interventions were both included in the physician ordered plan of care AND implemented, and both halves must be true for a yes.
The part therapists should know: interventions provided by any agency staff count, including you. If the nurse identifies fall risk and the physical therapist implements fall prevention interventions that are on the plan of care, the nurse codes yes. Your work shows up in somebody else's discharge assessment, but only if it was on the plan of care and only if the record shows you did it, which is the practical argument made in The Defensible Chart: Writing for the Reader Who Was Never in the Home.
Two sections have their own deep dives here and are named here only so the map is complete: falls, pain, dyspnea, and hospitalization risk are in Section J in Depth: Risk, Falls, Pain, and Breath, and medications are in the Section N in Depth section above. The item code map for the whole instrument is in OASIS Item Codes and PDGM.
The Errors That Cost the Most in Sections C and D
Almost every rule in Sections C and D is about what counts as an answer. The interview has a gate item, a stop rule, and its own code for an interview that never happened, and none of those is the same thing as a low score. The delirium item is four observations on a fluctuation scale, where two sources disagreeing is itself the finding. The mood interview turns on the frequency column. And the rest of the instrument is mostly other people's work that you are the only clinician in the home to do on a therapy only case.