Achieving Functional Goals during a Skilled Nursing Facility Stay: A National Study of Medicare Beneficiaries

Abstract

Objective:

To describe the frequency that functional goals are documented on the Minimum Data Set (MDS) and identify resident characteristics associated with meeting or exceeding discharge goals.

Methods.

We selected Medicare fee-for-service beneficiaries admitted to a skilled nursing facility (SNF) within three days of hospital discharge from 10/01/2018–12/31/2019 (n=1,228,913). The admission MDS was used to describe the discharge goal scores for seven self-care and 16 mobility items. We used the eight self-care and mobility items originally included in a publicly reported quality measure to calculate total scores for discharge goals, admission performance, and discharge performance (n=371,801).

Results.

For all self-care items, over 70% of residents had a goal score of 1–6 points documented on the admission MDS. Chair/bed-to-chair transfer had the highest percentage of residents with a score of 1–6 points (77.1%) and walking up/down 12-steps had the lowest (23.2%). Approximately 44% of residents had a discharge performance score that met or exceeded their goal score. Older age, urinary incontinence, and cognitive impairment had the lowest odds of meeting or exceeding discharge goals.

Conclusions.

Assessing a resident’s functional goals is important to providing patient-centered care. This information may help SNFs determine if a resident has made meaningful functional improvements.

Introduction

In 2020, Medicare covered 1.7 million skilled nursing facility (SNF) stays by 1.2 million Medicare fee-for-service beneficiaries.¹ SNFs provide nursing, physical and occupational therapy, speech-language pathology, intravenous therapy, and other skilled services that are medically necessary to improve a resident’s health. Fee-for-service Medicare spending on SNF services in 2020 was $28.1 billion.¹

Regaining function and independence are common goals of older adults receiving rehabilitation after a hospital stay.² However, not all older adults can realistically expect to regain their functioning or to be discharged home.³ This makes it important for SNFs to work with residents and families to establish clear, realistic goals at the start of care.⁴

In October 2016, a section (Section GG) was added to the Minimum Data Set (MDS) that included new items to document discharge function goals in their care plan. In October 2018, SNFs started publicly reporting this information with the process quality measure for the percentage of residents with a complete admission and discharge functional assessment and care plan that addresses function.⁵ As part of this quality measure, SNFs must document a valid discharge goal score for at least one self-care or mobility item in Section GG of the resident’s admission assessment.⁶ A valid score is a functional score of 1 (dependent) to 6 (independent) or a code indicating that the resident did not attempt the task (07, 09, 10, 88).⁶ SNFs are performing well on the quality measure. In 2019, SNFs, on average, submitted the required information for 99.2% of all resident stays.⁷ Furthermore, the percentage of SNFs with a perfect quality measure score (i.e., submitted the required information for all resident stays) increased from 33.1% in 2017 to 64.1% in 2019.⁷

Effective goal setting involves communicating with patients and their families to develop evaluation criteria and identifying outcomes the patient can achieve within a realistic timetable.⁸ As such, setting and communicating feasible goals is essential to delivering effective patient-centered care.⁹ Evidence from hospital, long-term care, and end-of-life care settings indicates that effectively setting and communicating goals are associated with better overall satisfaction with care.^10–12 Unrealistic, unclear, or poorly communicated goals can contribute to hospital readmissions from SNFs.^{3, 13}

SNFs have high completion rates for documenting residents’ functional performance at admission and discharge,^{7, 14} but documentation of the individual functional goal items has not been described. Such a description will provide evidence of if SNFs are using the goals of care items as CMS intended.¹⁵ Additionally, the goals of care items can be used to identify resident characteristics associated with meaningful recovery during an SNF stay. Our objective was to describe the documentation frequency of self-care and mobility goals at SNF admission and identify resident characteristics associated with meeting or exceeding functional goals.

Methods

Data and Sample

This study was a secondary analysis of Medicare fee-for-service data files from the Centers for Medicare and Medicaid Services (CMS). All data was obtained from existing de-identified datasets. Thus, our University’s Institutional Review Board waived the requirement for informed consent. The datafiles were accessed after submitting a data use agreement to CMS. This study conforms to all STROBE guidelines and reports the required information (see Supplementary Checklist).

We used 100% national Medicare claims files from 10/01/2018 to 12/31/2019. These files included the Master Beneficiary Summary File, the Medicare Provider Analysis and Review (MedPAR) file, and Version 3.0 of the MDS. The Master Beneficiary Summary file was used for information on age, sex, information for if a beneficiary was enrolled in Medicare and Medicaid, the original reason for Medicare eligibility, and to determine Medicare fee-for-service enrollment status. The MedPAR file was used to identify the number of comorbidities in the past year and characteristics of the index hospital stay. The MDS was used for information on residents’ health characteristics and functional status during the SNF stay.

We created two samples (Figure 1). Sample 1 was used to describe the number and percentage of SNF residents with a valid response for each discharge goal for the seven self-care and 16 mobility items on the admission MDS. First, we identified the first SNF admission for all Medicare fee-for-service beneficiaries between 10/01/2018 to 12/31/2019. Next, we selected beneficiaries who were admitted to an SNF within 3-days of hospital discharge, were age 66 or older at SNF discharge, who had continuous Medicare fee-for-service enrollment in the year before SNF discharge and had a 5-day MDS assessment. Sample 1 included 1,228,913 residents.

Figure 1.

Selection of Medicare fee-for-service beneficiaries admitted to a skilled nursing facility between October 1, 2018, and December 31, 2019.

Sample 2 was used to describe residents with a total score for eight self-care and mobility items at discharge that met or exceeded their total goal score for those eight items. We first selected residents who survived their SNF stay. Next, we selected residents whose functioning was assessed for eight self-care and mobility items at SNF admission and discharge. SNFs assess self-care and mobility at discharge for residents who had a planned discharge, have a SNF Medicare Part A discharge assessment, have a length of stay longer than two days, and are discharged from the SNF to a location other than an acute hospital.¹⁶

The eight self-care and mobility items we selected were eating, oral hygiene, toileting hygiene, sitting to lying, lying to sitting on the side of the bed, sitting to standing, chair/bed to chair transfer, and toilet transfer. We chose these items because they were originally included in the quality measure for the percentage of residents with a complete admission and discharge functional assessment and care plan that addresses function.¹⁵ These eight items also had the highest percentage of residents with a valid discharge goal score of 1–6 points. Using these eight items allowed us to calculate a total discharge goal score comparable to the total scores for functional performance at admission and discharge. We identified 393,153 residents who had a goal score between 1–6 points for all eight self-care and mobility items. We excluded residents with a valid not attempted code (07, 09, 10, or 88) or who were missing a score for one or more goal items so the total score for all residents used the same individual items. The total admission score ranged from 8 (dependent on all items) to 48 (independent on all items). To remove residents who were independent in at least one self-care or mobility item on admission to a SNF, we excluded residents with a total admission score of more than 40 points. This exclusion criterion ensured that all residents in the final sample could improve on all eight self-care and mobility items. Sample 2 included 371,801 residents.

Supplemental table 1 shows the characteristics of the 297,654 residents we excluded because of having a score of 07, 09, 10, or 88 for one or more self-care and mobility items on the admission or discharge MDS assessment, did not have a goal score between 1–6 points for one or more self-care and mobility items, or had an admission total score greater than 40 points compared to those in the final sample. Most residents we excluded did not meet the criterion for having a goal score between 1–6 points on all eight self-care and mobility items. We did not detect meaningful differences in the demographic or health characteristics between the 371,801 residents in the final sample to the 297,654 residents we excluded from the analysis (Supplemental table 1).

Measures

Self-Care and Mobility

We used seven self-care items and 16 mobility items from Section GG of the MDS (Supplemental Table 2).¹⁶ Each item is rated on a 6-point scale according to a resident’s usual performance in the previous 3-days. Each item is rated as follows: 1. Dependent (helper does all the effort); 2. Substantial/maximal assistance (helper does more than half the effort); 3. Partial/moderate assistance (helper does less than half the effort); 4. Supervision (verbal or touching cues); 5. Setup/cleanup assistance (helper sets up / cleans up, but resident completes activity); and 6. Independent (resident does activity by themselves with no assistance). Residents who do not attempt an activity can be coded as refused (07), not attempted because of illness or injury (09), environmental limitations (10), or not attempted because of safety concerns (88).

Discharge Goals

Section GG of the admission MDS includes a column to document a discharge goal for each self-care and mobility item.¹⁶ A licensed clinician should document a resident’s discharge goal based on their medical expertise, conversations with the resident and family, and the resident’s medical history, functional performance on admission, motivation to improve, expected length of stay, and discharge plan.¹⁶ The discharge goal is scored between 1 (dependent) and 6 (independent). The not applicable codes of 07, 09, 10, or 88 can also be used. We used this information to describe the percentage of SNF residents (n=1,219,192) with a discharge goal score of 1–6 points, 07, 09, 10, or 88 for each self-care and mobility item. We differentiated between residents with a score of 1 to 6 and valid not attempted code (07, 09, 10, or 88).

Resident Characteristics

Demographic characteristics were age, sex, race, and ethnicity, the original reason for Medicare entitlement, and dual enrollment in Medicare and Medicaid. Race and ethnicity were categorized as White, Black, Hispanic, and Other. The original reason for Medicare entitlement was categorized as age or disability/end-stage renal disease.

We used the admission MDS assessment for information on hearing and vision impairments, urinary and bowel incontinence, the primary medical condition category, and cognitive status. Hearing and vision impairment were categorized as none, minimal, and moderate-high. Urinary and bowel incontinence were categorized as none, occasionally, frequently, always, or not rated. Incontinence is coded as not rated for residents who have a catheter, urinary ostomy, ostomy, or have no urinary output or no bowel movement for the entire 7-day reference period. The primary medical condition category is the reason for SNF admission according to information in the resident’s medical record, hospital transfer documents, discharge summary, and related sources.¹⁶ We collapsed the 14 medical condition categories in the MDS into seven primary medical condition categories according to categories that were comprised of similar conditions. We used seven categories to minimize the number of comparisons: 1. Neurological (stroke, non-traumatic brain / spinal cord dysfunction, progressive conditions); 2. Orthopedic (hip/knee replacement, fractures); 3. Trauma (traumatic brain / spinal cord dysfunction); 4. Amputation; 5. Debility and cardiorespiratory conditions; 6. Medically complex conditions; and 7. Other (other neurological conditions, other orthopedic conditions). We used the Cognitive Functioning Scale to categorize residents’ cognitive status.¹⁷ The Cognitive Function Scale combines the Brief Interview for Mental Status (BIMS) and Cognitive Performance Scale (CPS) into a single measure. The Cognitive Function Scale classifies residents as having no impairment (BIMS score 13–15), mildly impaired (BIMS score 8–12 or CPS score 0–2), moderately impaired (BIMS score 0–7 or CPS score 3–4), or severely impaired (CPS score 5–6).

Outcomes

We used eight self-care and mobility items to calculate total scores for discharge goals, functional performance at admission, and functional performance at discharge. We treated the individual items as a single measure of self-care and mobility.

Meeting or Exceeding Discharge Goals

We compared residents’ total score for performance at discharge to the total goal score. Residents with a total discharge performance score equal to or greater than their total goal score were categorized as having met or exceed their total goal score (yes/no).

Difference between Discharge Performance and Goal Score

The second outcome was the difference in the total goal score and discharge performance score. We subtracted the total discharge performance score for the eight items from the total goal score.

Statistical Analyses

Chi-square tests and independent sample t-tests were used to describe SNF residents’ demographic and health characteristics according to having met their discharge goals. Logistic regression was used to identify demographic and health characteristics associated with having a total discharge score that was equal to or greater than the total goal score. All analyses were adjusted for the demographic and health characteristics in Supplemental table 2. We estimated the marginal adjusted mean to determine the average difference in the total discharge score minus the total goal score. These marginal means were also adjusted for the demographic and health characteristics in Supplemental table 2. The marginal adjusted mean is interpreted as the mean difference for each level of the categorical variable averaged over the levels of all other variables.

Results

Documentation frequency of self-care and mobility goals at SNF admission.

Table 1 presents the demographic characteristics of the 1,228,913 residents (sample 1) we used to describe the number and percentage of SNF residents with a valid response for each discharge goal for the seven self-care items and 16 mobility items on the admission MDS. Less than 10% of residents were age 65–69, 14.4% were 70–74, and 17.5% were 75–59. Nearly 20% of participants were age 80–84, 85–89, and 90 years of age or older. Most residents were female (62.6%). Eighty-four percent of residents were white, 8.3% were black, and 4.1% were Hispanic. Finally, 28.6% of residents were enrolled in Medicare and Medicaid.

Table 1:

Demographic characteristics of the 1,219,192 residents admitted to a skilled nursing facility within three days of hospital discharge.

Characteristics	Sample, N (%)
Total	1,228,913 100)
Age
65–69	107,920 (8.8)
70–74	176,930 (14.4)
75–79	215,502 (17.5)
80–84	243,274 (19.8)
85–89	241,853 (19.7)
90+	243,434 (19.8)
Sex
Male	459,033 (37.4)
Female	769,880 (62.6)
Race/ethnicity
White	1,033,329 (84.1)
Black	102,493 (8.3)
Hispanic	50,528 (4.1)
Other	42,563 (3.5)
Dual enrollment
No	877,563 (71.4)
Yes	351,350 (28.6)

Note: Dual enrollment refers to whether a resident was enrolled in Medicare and Medicaid.

Tables 2 and 3 show the number and percentage of the 1,228,913 residents with a goal score of 1–6 points, 07, 08, 10, or 88 on the seven self-care and 16 mobility items. All seven self-care items had a discharge goal score between 1 and 6 points for more than 70% of residents. Less than 5% of residents had a not applicable goal score (07, 09, 10, 88) for any self-care item. For the mobility items, roll left/right, sit to lying, lying to sitting, sit to stand, chair transfer, and toilet transfer, more than 70% of residents with a goal score between 1–6 points. Items for car transfer, walking 10, 50, and 150 feet, and picking up an object had between 50% and 70% of residents with a goal score between 1–6 points. Between 4% and 20% of residents had a valid not applicable score for toilet transferring, to wheel 150 feet, walking 50 feet with two turns, walking 150 feet, transferring to or from a car, and picking up an object. Over 20% of residents had a valid not applicable score for walking up or down a curb, walking up or down four steps, and walking up or down 12 steps.

Table 2.

Number (percentage) of the 1,219,192 residents admitted to a skilled nursing facility within three days of hospital discharge with a valid goal score for each self-care task on the admission Minimum Data Set assessment.

Task	Code for Discharge Goal (N=1,228,913)
	Any valid code n (%)	Code 1 to 6 n (%)	Code 7, 9, 10, 88 n (%)
Eating	959,388 (78.1)	924,983 (75.3)	34,405 (2.8)
Oral hygiene	940,007 (76.5)	910,573 (74.1)	29,434 (2.4)
Toileting hygiene	995,414 (81.0)	970,307 (79.0)	25,107 (2.0)
Shower/bathe self	931,784 (75.8)	893,618 (72.7)	38,166 (3.1)
Upper body dressing	944,922 (76.9)	917,275 (74.6)	27,647 (2.2)
Lower body dressing	948,375 (77.2)	920,266 (74.9)	28,109 (2.3)
Putting on/taking off footwear	921,934 (75.0)	883,744 (71.9)	38,190 (3.1)

The valid code for the discharge goal is a score of 1 to 6 or 07, 09, 10, and 88. The scores of 07, 09, 10, and 88 indicate the resident did not attempt the task because the resident refused to perform an activity (07), did not perform this activity prior to the current illness, exacerbation, or injury (09), did not attempt the activity because of environmental limitations (10), or did not attempt the activity because of a medical condition or safety concern (88).

Table 3.

Number (percentage) of the 1,219,192 residents admitted to a skilled nursing facility within three days of hospital discharge with a valid goal score for each mobility task on the admission Minimum Data Set assessment.

Task	Code for Discharge Goal (N=1,228,913)
	Any valid code n (%)	Code 1 to 6 n (%)	Code 7, 9, 10, 88 n (%)
Roll left/right	941,503 (76.6)	906,809 (73.8)	34,694 (2.8)
Sit to lying	929,094 (75.6)	893,451 (72.7)	35,643 (2.9)
Lying to sitting on side of bed	929,002 (75.6)	892,361 (72.6)	36,641 (3.0)
Sit to stand	960,449 (78.2)	914,096 (74.4)	46,353 (3.8)
Chair/bed-to-chair transfer	979,056 (79.7)	947,946 (77.1)	31,110 (2.5)
Toilet transfer	938,233 (76.3)	881,409 (71.7)	56,824 (4.6)
Car transfer	901,950 (73.4)	692,489 (56.3)	209,461 (17.0)
Walk 10 feet	945,721 (77.0)	839,271 (68.3)	106,450 (8.7)
Walk 50 feet with two turns	947,956 (77.1)	800,292 (65.1)	147,664 (12.0)
Walk 150 feet	963,733 (78.4)	759,503 (61.8)	204,230 (16.6)
One step (curb)	907,809 (73.9)	580,388 (47.2)	327,421 (26.6)
Four steps	887,892 (72.3)	418,871 (34.1)	469,021 (38.2)
Twelve steps	883,067 (71.9)	285,359 (23.2)	597,708 (48.6)
Picking up object	897,558 (73.0)	656,812 (53.4)	240,746 (19.6)
Wheel 50ft w/two turns	525,790 (42.8)	410,223 (33.4)	115,567 (9.4)
Wheel 150ft	523,058 (42.6)	390,648 (31.8)	132,410 (10.8)

The valid code for the discharge goal is a score of 1 to 6 or 07, 09, 10, and 88. The scores of 07, 09, 10, and 88 indicate the resident did not attempt the task because the resident refused to perform an activity (07), did not perform this activity prior to the current illness, exacerbation, or injury (09), did not attempt the activity because of environmental limitations (10), or did not attempt the activity because of a medical condition or safety concern (88).

Resident characteristics associated with meeting or exceeding their total goal score.

Supplemental table 3 shows the demographic and health characteristics of the 371,801 residents (sample 2) we used to describe residents with a total score for the eight self-care and mobility items at discharge that met or exceeded their total goal score. Eight percent of residents were age 65–69, 14.4% were 70–74, 18.1% were 75–79, and approximately 20% of residents were 80–84, 85–89, and 90 years of age or older. Most (65.3%) residents were female. Approximately 87% of residents were white, 6.3% were black, and 3.8% were Hispanic. Nearly 30% of residents were admitted to a SNF for a medically complex condition, 19.9% for other type of neurological or orthopedic condition, and 19.6% for a trauma condition (traumatic brain / spinal cord dysfunction). Finally, 20.1% of residents had mild and 12.0% had moderate-severe cognitive impairment.

Overall, 43.9% of residents had a total discharge performance score equal to or greater than their total goal score (Supplemental table 3). Approximately one-third of residents aged 90 or older, with moderate to high hearing or vision impairment, primary medical condition was a neurological condition or amputation, or mild impairment in cognitive function met or exceeded their total goal score. Less than 25% of residents with no continent urinary voids and moderate to severe impairment in cognitive function met or exceeded their total goal score.

Several resident characteristics were associated with the adjusted odds for meeting or exceeding discharged goals (supplemental table 3). Residents aged 90 and older had 0.61 lower odds of meeting or exceeding their total goal score than those aged 65–69. Compared to residents admitted to a SNF because of an orthopedic condition, residents who had an amputation had the lowest adjusted odds for meeting or exceeding discharge goals (OR=0.60, 95% CI=0.54–0.67). The remaining primary medical condition categories all had approximately 25% lower odds for meeting or exceeding discharge goals. Moderate-high hearing and vision impairment were associated with 0.90 and 0.86 lower odds of meeting or exceeding discharge goals, respectively. Other resident characteristics associated with lower odds for meeting or exceeding discharge goals were frequent urinary incontinence, no continent urinary voids, and moderate to severe impairment in cognitive function.

Supplement table 4 shows the average total scores for self-care and mobility performance at admission and discharge and the total goal score. All three total scores decreased with older age, greater sensory impairment, more frequent incontinence, and greater cognitive impairment. Among the primary medical conditions categories, residents in the orthopedic category had the highest average total score at admission, discharge, and total goal score whereas residents in the trauma category had the lowest total score at admission and residents in the neurological category had the lowest total score at discharge. Residents aged 65–69 had the highest average total score at admission, discharge, and total goal score, whereas residents with moderate to severe cognitive impairment had the lowest average total scores.

Supplement table 4 also shows the mean difference between the discharge performance total score and the total goal score. The widest range in the adjusted mean difference was for age, urinary and bowel incontinence, primary medical condition category, and cognitive status. The adjusted mean difference for residents aged 65–69 was −3.3 points and −4.7 points for residents aged 90 and older. The adjusted mean difference was −3.4 and −3.7 for residents with no urinary or bowel incontinence, respectively, and −5.5 and −5.6 for residents who were always incontinent. Residents in the orthopedic category had an adjusted mean difference of −3.6 points whereas residents in the amputation category had an adjusted mean difference of −4.6 points. Lastly, the adjusted mean difference for residents with no cognitive impairment was −3.6 points and −5.5 points for residents with moderate-severe cognitive impairment.

Discussion

We used a national sample of Medicare fee-for-service beneficiaries to describe the documentation of self-care and mobility goals at SNF admission and to identify resident characteristics associated with functional performance at discharge that meets or exceeds a resident’s goals. We found that SNFs routinely document functional goals in self-care and mobility on the admission MDS. However, a lack of training on how to have goals of care conversations, ¹⁸ and intangible characteristics, such as a resident’s belief they can improve in function¹⁹ can make it difficult for rehabilitation professionals to determine realistic rehabilitation goals at the start of a resident’s stay. We found that less than half of residents had a functional performance score at discharge that met or exceeded their total goal score on their admission assessment. Prior studies have reported that over 60% of SNF residents have any functional improvement between admission and discharge.^{14, 20, 21} Our finding suggests it is less common for residents to improve to the degree expected when admitted to an SNF.

The self-care tasks, four bed mobility items, and transferring to or from a chair had the highest percentage of residents with a discharge goal score between 1–6 points. Except for some higher-level mobility items (e.g., stepping up a curb, going up four or twelve steps), it was uncommon for residents to have a code of 07, 09, 10, or 88 documented as a discharge goal. This finding is encouraging because it suggests facilities are not using these codes to meet the quality measure reporting requirements for the percentage of residents with a complete admission and discharge functional assessment and care plan that addresses function.

It is not always realistic to expect a SNF resident to show functional improvement, and some residents may strive to maintain their level of independence or slow their functional decline. For SNF residents who have goals to regain independence in self-care and mobility, doing so can decrease the likelihood of being readmitted to the hospital after SNF discharge,²² and is associated with lower mortality after a critical illness.²⁰ For SNFs, publicly reported quality measures for functional improvement in the Care Compare 5-star rating system²³ and SNF Quality Reporting Program²⁴ are incentives to provide care that maximizes residents’ functional recovery during a post-acute care stay.

The highest percentage of SNF residents who had a total discharge score that met or exceeded their total goal score was for those admitted for a hip replacement, knee replacement, or fracture. Joint replacements are among the five most common conditions for hospital discharges to SNF.¹ Most Medicare beneficiaries discharged to a SNF after a joint replacement procedure or fracture are independent in daily activities prior to being admitted to a SNF,²¹ and may expect to regain function during a SNF stay. An expectation or belief that a resident will improve in function can be an important predictor of positive SNF outcomes.¹⁹

Conversely, residents older than 90 years, frequent urinary incontinence, amputation, and moderate-severe cognitive impairment, had the lowest adjusted odds of meeting or exceeding their discharge goal score. These characteristics are known risk factors for less functional recovery during an SNF stay.^{21, 25, 26} We also observed strong gradients in the total goal score according to age, frequency of urinary incontinence, primary medical condition category, and cognitive status. This finding is evidence that SNFs consider these characteristics when scoring discharge goals. However, these characteristics also had the largest adjusted mean difference between the discharge performance and goal scores. This difference may reflect the difficulty of establishing goals that are ambitious enough to motivate a resident but are also realistically achievable. Clinicians must use their best judgement to assign one or more discharge goals according to the resident’s medical history, the treatments they will receive, and expected length of stay. Clinicians can also discuss with the resident and their family about realistic rehabilitation goals and outcomes,¹⁶ but some residents are not able to engage in such discussions and family members are not always present. Decision aids can help develop realistic care goals for nursing home residents, but these aids are typically designed for residents diagnosed with dementia or at the end of life.^{27, 28}

Study limitations

One limitation is that our analysis comparing residents’ functional status at discharge to their discharge goal score could only include residents who had a complete SNF stay. Residents who have an unplanned discharge and residents discharged to an acute hospital do not have their functional status assessed by the discharge MDS.²⁹ Another potential limitation is that we required participants to have a functional goal score of 1–6 points on all eight items used to calculate the total score. Consequently, we do not consider residents with fewer than eight discharge goals documented on the admission MDS. Residents could also have met their goal for some individual self-care or mobility items but still have a performance discharge total score that did not meet or exceed their total goal score. Residents may also have different functional trajectories for self-care versus mobility items. Future research should investigate separate total scores for the three self-care and five mobility items as opposed to summing these eight items into one total score. Such investigations would give additional insights into SNF residents’ functional recovery and the likelihood a resident meets or exceeds their rehabilitation goals.

Also, our sample only included Medicare fee-for-service beneficiaries. Medicare Advantage beneficiaries are less likely to be admitted to an SNF with a 4- or 5-star rating than fee-for-service beneficiaries,³⁰ but have higher community discharge rates than fee-for-service beneficiaries.³¹ These differences may bias our findings for documenting discharge goals on the admission MDS and for residents’ functional improvement during an SNF stay. Finally, we could not consider that a resident’s discharge goals may change based on their progress during the SNF stay.

Conclusions and implications

Our study shows that SNFs document discharge goals for self-care and mobility on the admission MDS. While it is common for older adults to improve their functioning during an SNF stay, less than half are meeting or surpassing their discharge goals. Meeting or exceeding discharge goals may be a meaningful outcome for investigations of functional recovery during an SNF stay.

What is known.

Skilled nursing facilities are required to document a discharge goal for at least one self-care and mobility task at the start of care for each resident. A licensed professional uses their clinical judgement and discussions with the resident to assign a discharge goal of dependent to independent.

What is new.

Less than half of the skilled nursing facility residents in our sample had a discharge performance score for daily functioning that met or exceeded their goal score. Older age, urinary and bowel incontinence, and cognitive impairment were associated with lower odds of meeting or exceeding discharge goals.

Data availability:

The Centers for Medicare and Medicaid Services data use policies do not allow for the datasets used for this study to be made available.