Trial Magazine

The vast majority of injuries suffered by nursing home residents are the direct result of understaffing.¹ Pressure injuries, falls, choking, dehydration, malnutrition, and unchecked infections are all caused by understaffing. Understaffing is intentional—it’s a business model designed at the corporate level and forced on the facility. When you handle a nursing home case based on understaffing, the progression is simple: The facility was understaffed, the decision to understaff was intentional and based solely on money, and that decision killed your client.

Nursing home cases have two major components: the abuse, neglect, and injuries of your client; and corporate malfeasance. Understaffing, which falls under the corporate malfeasance component, is perilous for defendants if you can present data supporting the proposition that the facility was understaffed and that the understaffing was due to decisions made at the corporate level—decisions that placed profits over people.

By analyzing staffing information you can establish the level of staffing that the facility reported to state and national agencies.² Then compare this “reported staffing” to “expected staffing” to determine the degree of understaffing. The Centers for Medicare and Medicaid Services (CMS) regulates nursing homes in the United States. Although CMS does not have a numerical minimum staffing requirement, it requires nursing homes to have sufficient nursing staff with the appropriate skill sets “to assure resident safety and attain or maintain the highest practicable physical, mental and psychosocial well-being of each resident, as determined by resident assessments.”³

Some states, such as Texas⁴ and Virginia,⁵ follow CMS’s lead and require only “sufficient” staffing, without delineating a specific minimum staffing level. Other states go further—they require a specific minimum staffing level and also require that the staffing level meets the needs of the individual residents.

---

Satisfying the minimum staffing requirement in a given state does not mean that the facility had sufficient staff to meet the needs of each resident.

---

For example, Vermont requires 3.0 hours of nursing care per resident per day,⁶ while California requires 3.5 hours of total nursing care with a minimum of 2.4 hours of the total provided by certified nurse assistants (CNAs).⁷ Note that satisfying the minimum staffing requirement in a given state does not mean that the facility had sufficient staff to meet the needs of each resident.

Staffing Analysis Basics

The analysis is accomplished with basic math. You want to know how much staffing the facility had, how much it should have had, and what the difference is.⁸ Ultimately, this is done with three basic data points: actual staffing data for the facility, the number of residents in the building (census), and an understanding of the care each resident requires (acuity). With that information, you can then determine what that difference was worth in savings to the facility. First, you need to understand a few key terms.

Reported staffing. This is the amount of staffing the facility reports. In practice, the analysis is very straightforward. To determine total reported staffing in a 24-hour period, add together the registered nurse (RN) time, licensed practical nurse/licensed vocational nurse (LPN/LVN) time, and CNA time. The total is the total direct care, in worked hours. For example, if the facility reports 24 RN hours, 60 LPN hours, and 210 CNA hours, this is 294 hours of direct care.

To be relevant to a case, this data needs to be converted to “hours per patient per day” (HPPD). To calculate this, take the total hours and divide by the census. In the example earlier, if the census was 82 residents, the math would be:

294/82 = 3.58 HPPD

At this point, we know the facility was staffed at roughly 3.6 hours of direct care per patient per day. You also can calculate the HPPD for each nursing category by dividing the hours by census. For example:

RN = 24/82 = 0.29 HPPD

LPN = 60/82 = 0.73 HPPD

CNA = 210/82 = 2.56 HPPD

Minimum staffing. This is the staffing needed to meet the minimum recommended staffing generally (but it may be different from the facility’s expected staffing, described later). The most recent literature on staffing indicates that the minimum for total direct care staffing should be at least 4.1 HPPD.⁹ Academic articles by nursing care experts conclude the minimum staffing a nursing home should have for RNs is 0.75 HPPD, for LPNs is 0.55 HPPD, and for CNAs is 2.8 HPPD.¹⁰

Thus, the facility in the example was understaffing RNs by 0.46 HPPD, overstaffing LPNs by 0.18 HPPD, and understaffing CNAs by 0.24 HPPD, for a total understaffing of 0.52 HPPD comparing reported staffing to minimum staffing. This is standard “for-profit” staffing: cut RN hours to almost nothing and slightly overstaff LPN hours to cover some of the RN understaffing. Since RNs are the most expensive labor, facilities may try to drop RN hours as low as possible to save as much as possible.

The understaffing numbers in this example appear small for a single patient, but in the aggregate across the facility they are substantial: Each resident is being shorted 0.46 hours of RN time per day, almost half an hour per day. Across the entire 82-resident facility that is 37.72 hours per day, 264 hours per week, and just under 14,000 hours per year. That’s seven full-time RNs that this facility is shorting its patients—and assuming an RN pay rate of $37.11 per hour, this facility saved $510,923 on RN staffing in one year.¹¹ It gave some of that savings back with overstaffing LPNs but also saved more money by understaffing CNAs. In my experience, it is common to find facilities that understaff in the $500,000 to $1 million range per year, and occasionally more with larger facilities.

Expected staffing. This is the staffing needed to meet the acuity needs of each resident. CMS has a nursing home comparison website called CMS Compare,¹² which uses a five-star rating system. Part of the star rating system is staffing. From 2009 to 2018, CMS published expected staffing, as it was used to assign staffing stars for each facility.¹³

In 2018, over objections from its expert advisory panel, CMS changed the methodology for assigning stars in the rating system. It no longer uses expected staffing; instead, it uses a different metric called “case-mix adjusted hours.”¹⁴ But CMS has never disavowed expected staffing as a valid means for calculating the nursing needs of nursing home residents; it merely changed how it assigns stars. To zero in on the exact staffing needs at one facility over a specific time, compare reported staffing to expected staffing.

Expected staffing is determined using time studies conducted by CMS. The basic premise is that the facility is assessing each person using a CMS evaluation tool called the Minimum Data Set (MDS), a 40+ page assessment document that lays out how much assistance the individual needs with eating, dressing, toileting, their medical diagnosis, and rehabilitation.¹⁵ In Section Z of the MDS, there is an acuity score for that individual.¹⁶ Facilities submit the MDS to CMS, and the acuity score sets the daily reimbursement for this resident.

Taking a step back, this is essentially the federal government asking the facility, “how much care does this person need?” The facility responds, “this resident needs X amount of care per day,” to which the federal and state governments respond with a daily reimbursement to the facility. The rates are based on the understanding that the facility is going to meet the individual needs of each resident. Each reimbursement is directly tied to the amount of time that is expected to be needed to care for an individual with that acuity. Staffing significantly below expected staffing is fraud.

Calculate expected staffing by obtaining Section Z from every MDS done at the facility, starting six months before your client entered the facility, through the day of their discharge. This data shows when each resident entered and was discharged from the facility, and what their acuity level was each day. You can look up that acuity level in a table to determine how much RN, LPN, and CNA nursing time that person needs per day.¹⁷ If you add all the nursing time associated with each resident each day, then divide by the number of residents in the building that day (census), you will have the exact expected staffing for that facility that day.

This calculation also will reveal the exact expected amount of RN, LPN, and CNA time for each day, which can be compared against the facility’s reported staffing. This can help you show why your client fell on that Saturday or why your client developed that stage IV pressure wound in three weeks. You also can use the hourly average cost of labor from the annual cost reports (CMS Form 2540-10) that each facility must file with CMS each year¹⁸ to explain to a mediator or jury exactly how much money the facility pocketed in staff savings and how that directly and negatively impacted your client.

In sum, after determining a facility’s reported, minimum, and expected staffing, you can use this data to calculate whether it was understaffed—both generally and for the facility’s acuity level. Here are two formulas:

• Reported staffing – minimum staffing = over- or understaffing (generally)
• Reported staffing – expected staffing = over- or understaffing (specifically for this facility’s acuity level).

Sources of Staffing Data

To calculate whether a facility was understaffed, you’ll need to collect staffing data. Here are a few options for obtaining this.

Payroll Based Journal. Arguably the best staffing data is held by CMS in their Payroll Based Journal (PBJ) system.¹⁹ CMS started collecting PBJ data in January 2017, and it is usually available online 90 days after each quarter ends.²⁰ PBJ data includes daily census and daily staffing, broken down by RN, LPN, and CNA. With this information you can determine the exact reported staffing for any day or length of time.

CMS annual cost report. CMS requires each nursing home to file an annual cost report known as CMS Form 2540-10.²¹ You can obtain CMS annual cost reports via a federal Freedom of Information Act (FOIA) request to CMS.²² The report is a spreadsheet with 66 tabs, and while most of these tabs will have no relevance to your case, worksheet S-3-V contains staffing data. The annual cost report also contains the facility census on worksheet S-3 at row 1, column 7. Nursing administration hours are available on worksheet S-3, Part II, line 7.

The data is aggregated for a year, so you cannot determine what the staffing was on a certain day, but you can see what the staffing was for the year. Another caveat is that the data reported is “paid hours,” not “worked hours.” Paid hours include vacation and sick leave and therefore will be an inflated number when you are analyzing for direct care hours as compared to PBJ data, which is worked hours.²³

As mentioned above, the annual cost report includes average hourly wages paid to each classification of employee. This can be used to calculate the savings due to understaffing based on the actual pay scale the facility uses instead of using the national labor statistics. The hourly wages are located on the same lines as the hours paid information on worksheets S-3-V and S-3, Part II.

Medicaid cost reports. Medicaid cost reports are required by all states; each state uses a different form and collects different information. Most states do not collect staffing data.

Notable exceptions are Arizona,²⁴ California,²⁵ Illinois,²⁶ Oklahoma, and Pennsylvania,²⁷ which all collect good staffing data. All of these states have cost reports online except Oklahoma’s, which must be obtained by contacting the state and filing a request. This is not an exhaustive list, however, so check with your state Medicaid agency to see whether your state collects staffing data. The reporting requirements change from time to time, so data might be available in your state.

Like the CMS cost report, the staffing data contained within Medicaid cost reports is annualized data, not daily staffing data. You’ll also need to determine whether it is paid hours or worked hours—some states conveniently collect both types of data.

Staffing in Discovery

You can obtain additional staffing data in discovery. Request the facility’s “punch detail” (the industry’s term for timecards). This includes the exact times each employee clocked in and out of the facility for their individual shifts. Although normally provided in PDF format, the data needs to be in a spreadsheet to be usable. Instead of hand converting PDFs, request and insist that the production be in native format: either .csv (standard spreadsheet) or .xlsx (Excel). Native format allows for fast and efficient staffing analysis calculations.

Also request staffing schedules, which are useful to get a 30,000-foot view of the staffing as it applies to different floors or units within the facility. The schedules are aspirational, so they are not to be used for an exact staffing analysis. The fact that someone was scheduled does not mean they showed up for their shift, arrived on time, or stayed the entire shift.

---

Budgets, especially staffing budgets, are critical to link the understaffing to corporate malfeasance.

---

Budgets, especially staffing budgets, are critical to link the understaffing to corporate malfeasance, as well as policies for budgeting, setting staffing, and scheduling staffing. The budgets set the staffing. The facility employees may say they staff to acuity, but that is rarely the case. Because they are staffing to a budgetary number dictated by their corporate leadership, they staff only to census.

This means that for each individual admitted to the facility, the facility’s employees will staff an additional number of hours, no matter what that resident actually needs. This is a per se violation of CMS and state staffing regulations. The budgets also can be used to establish corporate involvement in the day-to-day operational control of the facility.

Expected staffing reveals a lot about a facility, but it can be challenging to calculate. At the facility level, you must request production of all the data in Section Z for all MDS that were performed in the six months leading up to your resident’s admission and continuing through their entire stay. Section Z is where the facility reports a final coding for the resident that tells Medicare how sick they are, and in turn, how much the facility gets paid for that resident. In short, it shows the acuity of each resident. You need the Section Z data for all the residents in the facility.²⁸ With those codes, you can look up the acuity level in a table to determine how much RN, LPN, and CNA time that person needs per day.²⁹ Doing this for every resident in the facility for each day your client was there is a sizable undertaking, but the payoff is well worth the effort.

To streamline authentication and admissibility, I suggest asking the defendants for the data and annual reports used in your analysis. If they balk, send what you’ve used from publicly available data with a request for admission asking them to authenticate it.

A staffing analysis can be tedious and time-consuming, but it is ammunition you can use when you question the director of nursing, the administrator, and corporate players about why they were not staffing to acuity. This analysis can demonstrate that your client’s injuries were not due to an isolated incident. Rather, the understaffing was intentional, done to increase profits, and done with conscious indifference to the lives of residents.

---

AAJ RESOURCES

• Nursing Home Litigation Group
• “Nursing Homes: Regulations, Discovery, and Damages” Litigation Packet
• “Understaffing Issues: Fundamentals of a Nursing Home Case (Part 3)” (On-demand webinar)

---

Ernest Tosh is a partner at Bedsore.Law and can be reached at Ernest@Bedsore.Law.

---

Notes

1. Ctrs. for Medicare & Medicaid Servs., Report to Congress: Appropriateness of Minimum Nurse Staffing Ratios in Nursing Homes Phase II Final Report (2001); Jane Bostick et al., Systematic Review of Studies of Staffing and Quality in Nursing Homes, 7 J. Am. Med. Dir. Ass’n 366–76 (2006), DOI: 10.1016/j.jamda.2006.01.024.
2. Another avenue for using data to bolster a nursing home case is a facility’s financial data. The financial data can be, and is, manipulated at the facility level, and it is often possible to show that the facility had plenty of money to properly staff the facility but chose not to. Discussion of the financial analysis component is outside the scope of this article, but I am happy to discuss it with anyone who wants more information on that subject.
3. 42 C.F.R. §483.35.
4. 26 Tex. Admin. Code §554.1001.
5. 12 Va. Admin. Code §5-371-210(B).
6. 13-007 Code Vt. Rules §7.13(d)(1)(i).
7. Cal. Health & Safety Code §1276.5.
8. Charlene Harrington et al., Appropriate Nurse Staffing Levels for U.S. Nursing Homes, 13 Health Servs. Insights, Dec. 2020, at 1–14, DOI: 10.1177/1178632920934785 [hereinafter Harrington I].
9. See id. at 5; Report to Congress, supra note 1.
10. Report to Congress, supra note 1; Charlene Harrington et al., Time to Ensure Sufficient Nursing Home Staffing and Eliminate Inequities in Care, J. of Gerontology & Geriatric Med. (June 2021), DOI: 10.24966/GGM-8662/100099 [hereinafter Harrington II].
11. U.S. Bureau of Labor Statistics, Occupational Employment and Wages, May 2022:29-1141 Registered Nurses, https://www.bls.gov/oes/current/oes291141.htm.
12. Ctrs. for Medicare & Medicaid Servs., Find & Compare Providers Near You, https://www.medicare.gov/care-compare/?redirect=true&providerType=NursingHome.
13. See Harrington I, supra note 8.
14. See id.
15. Medicare & Medicaid Programs; Reform of Requirements for Long-Term Care Facilities, 81 Fed. Reg. 68,688 (Oct. 16, 2016); Ctrs. for Medicare & Medicaid Servs., State Operations Manual: Appendix PP−Guidance to Surveyors for Long Term Care Facilities, Feb. 3, 2023, https://www.cms.gov/Medicare/Provider-Enrollment-and-Certification/GuidanceforLawsAndRegulations/Nursing-Homes.html; Am. Nurses Ass’n, Nursing Staffing Requirements to Meet the Demands of Today’s Long Term Care Consumer: Recommendations From the Coalition of Geriatric Nursing Organizations (CGNO), Position Statement, Nov. 12, 2014, www.nursingworld.org.
16. Prior to Oct. 1, 2019, this was called a RUG score; it is now called a Nursing Services Score.
17. Harrington I, supra note 8, at 6.
18. Ctrs. for Medicare & Medicaid Servs., Skilled Nursing Facility 2540-2010 Form, https://www.cms.gov/research-statistics-data-and-systems/downloadable-public-use-files/cost-reports/skillednursingfacility-2010-form.
19. Ctrs. for Medicare & Medicaid Servs., Electronic Staffing Data Submission Payroll-Based Journal: Long-Term Care Facility Policy Manual, June 2022, https://www.cms.gov/Medicare/Quality-Initiatives-Patient-Assessment-Instruments/NursingHomeQualityInits/Downloads/PBJ-Policy-Manual-Final-V25-11-19-2018.pdf; Ctrs. for Medicare & Medicaid Servs., Payroll Based Journal Daily Nurse Staffing, https://data.cms.gov/quality-of-care/payroll-based-journal-daily-nurse-staffing.
20. Payroll Based Journal Daily Nurse Staffing, supra note 19. PBJ data was voluntarily reported from July 1, 2016, to Dec. 31, 2016. It was a period for the facilities to work out any bugs they had with transmission and to make sure they were submitting their staffing data correctly—but I would not rely on it as it may not have been submitted accurately. Starting Jan. 1, 2017, the PBJ is available online without a FOIA request.
21. Skilled Nursing Facility 2540-2010 Form, supra note 18.
22. Ctrs. for Medicare & Medicaid Servs., Freedom of Information Act, https://www.cms.gov/Regulations-and-Guidance/Legislation/FOIA.
23. To estimate worked hours from paid hours, deduct 8% from paid hours (paid hours x 0.92 = worked hours).
24. Ariz. Dep’t of Health Servs., Health Facility Cost Reporting Application, https://apps.azdhs.gov/HFCR/.
25. Cal. Dep’t of Health Care Access & Info., Financial & Utilization Reports, https://sieraarchiveexternal-oshpd-web-prd.azurewebsites.net/.
26. Ill. Dep’t of Health Care & Family Servs., Enrolled Long Term Care Provider Cost Reports, https://hfs.illinois.gov/medicalproviders/costreports/icfddprovidercostreports.html.
27. Pa. Dep’t of Human Servs., Nursing Facility Report Portal, https://nfrp.panfsubmit.com/.
28. Section Z data can be produced without identifiers to be HIPAA compliant.
29. See Harrington II, supra note 10.