Introduction

For the past 20 years, Dr. Sandra F. Simmons, PhD has been devising and testing non-medical interventions to improve nutrition and prevent dehydration among nursing home residents, thereby helping to prevent unintentional weight loss among this vulnerable population.  The impetus for this work derives from a substantial body of research that supports two conclusions:

1. Under-nutrition and dehydration are common problems among nursing home residents; and

2. These problems are associated with unintentional weight loss and can lead to a host of other problems for older adults including delayed wound healing and increases in the rates of hospitalizations and death.

The many causes of weight loss, under-nutrition, and dehydration in the frail elderly—depression, dementia, polypharmacy and reduced senses (taste, smell, hunger, thirst), to name a few—suggest many possible solutions to these problems.  Strong evidence, however, suggests that the amount and quality of feeding assistance provided to residents during and/or between regularly-scheduled meals is possibly the most powerful determinant of their daily food and fluid intake.  Thus, it makes sense to direct weight loss prevention efforts toward improving feeding assistance care quality during and between meals.

With this in mind we set out to first assess, and then improve, the quality of feeding assistance in nursing homes.  Our researchers have spent hours in nursing homes across the country, observing the staff, the residents, the meal service, and recording what’s done, what’s said, and what’s eaten.  When you directly observe mealtime routines, you see things that would otherwise escape notice if your only information source was resident medical charts.  Consider these findings:

• Nurse aides consistently overestimate residents’ mealtime food and fluid consumption by as much as 15% to 20%, on average so many residents who are potentially at risk for weight loss, under-nutrition, and dehydration due to low food and fluid intake are not identified by staff when examining only a resident’s “percent eaten” documented in their medical record.  Moreover, one of our studies showed that there was a systematic error rate in nurse aide estimation of residents’ oral intake; that is, the less a resident ate, the more likely staff were to overestimate the resident’s consumption.
• The majority of residents are at risk for under-nutrition and dehydration due to low food and fluid intake.
• Most facilities do not have enough nurse aide staff to adequately assist all residents who need assistance during mealtimes.
• Due to understaffing, nurse aides “triage” residents at mealtimes, with the most functionally and cognitively impaired individuals, those who wouldn’t eat a bite if someone didn’t put it in their mouth (i.e., requiring “full physical assistance”), getting the most staff attention. 
• Many others are physically capable of eating on their own but remain at high risk for under-nutrition, dehydration, and weight loss because they do not eat or drink enough on their own.
• These at-risk residents don’t consume many calories between meals either, though the staff often believe they do.  Staff usually are surprised by our findings based on direct observations, which show that residents consume, on average, fewer than 100 calories from additional foods and fluids (snacks) and oral liquid nutrition supplements between meals. However, also based on our own direct observations, staff do not consistently offer residents additional foods and fluids between meals nor do they provide appropriate assistance to encourage consumption—even when the resident has a physician or dietitian order to receive snacks or supplements.

Clearly, these findings point to a serious problem with the adequacy and quality of feeding assistance in nursing homes.  If you’re now thinking, as we did, that the obvious solution is to assign more staff to help at mealtimes, then think again.  We tried that in three nursing homes: Assigned our own highly trained staff to provide one-on-one feeding assistance over six consecutive meals to each of 74 residents who were consistently under-eating.  Working within the context of a standardized protocol, we coaxed, cajoled, and conversed with each resident for about 40 minutes per meal, doing everything we could think of to get the person to eat and drink more.  About half did consume more foods and fluids, significantly more, increasing their intake by 30% or higher, on average. 

The other half did not increase their consumption.  For a sub-sample of these residents, we provided an additional two days of individualized feeding assistance—to no avail.  Despite our best efforts, they still consumed less than half of the foods and fluids served to them during meals.

Not satisfied with these results, we offered all residents who did not increase their food and fluid consumption in response to mealtime feeding assistance a variety of between-meal snacks three times a day (10am, 2pm, and 7pm) for two days.  Again, we sat and visited with each person during the snack period, providing feeding assistance as needed.  It worked, and although the residents ate and drank more at snack time, they didn’t eat or drink less at mealtimes.  On average, these residents consumed an additional 380 calories per day from snacks.

This finding suggests yet another reason why some nursing home residents do not eat or drink enough on a daily basis: They have a small appetite, which means they will eat and drink only a small amount at any one time.  Thus, offering between-meal snacks three times a day doubles the number of opportunities that residents have to eat to six times per day, which leads them to increase their overall daily caloric consumption.  The results of a separate study showed that offering residents a choice among a variety of foods and fluids was more cost-effective in increasing residents’ between meal caloric intake than offering residents oral liquid nutrition supplements alone – the most common nutrition intervention.  Residents preferred alternative foods and fluids to supplements, and due to residents’ preference for snacks, this approach required less staff time.  Offering residents a choice of assorted fluids between meals also leads to increased fluid intake and a decrease in dehydration, an important outcome because residents who are not eating enough during meals generally are not drinking enough either.

When paired together, our mealtime and snack interventions combine to create a single very powerful and, equally important, feasible weight loss prevention intervention.  This dual intervention offers several advantages:
 

• Nearly 90% of residents with low intake will significantly increase their food and fluid consumption with either the mealtime or snack intervention protocols.
• Both the mealtime and snack interventions can be implemented with groups of three (during meals) or more residents (during snacks) and still effectively prompt residents to significantly increase their intake.  This group model is a more practical alternative for most nursing homes, though it requires staff to transport residents to the dining room or another common area for group delivery.
• Nursing home staff need not provide intensive feeding assistance to all residents at mealtimes.  Residents who are responsive to mealtime assistance can be identified in a two-day, or six-meal, assessment trial.  Staff should concentrate their efforts on helping these residents during meals; that is, residents who are not eating well on their own and who will eat significantly more when staff spends time providing the appropriate level and amount of assistance.  Residents who are not responsive to this approach become the focus of the snack intervention.
• The snack intervention fits in well with most organized social activities programs, as part of which snacks can be efficiently provided in larger groups (four or more residents).  Many residents who are responsive to snacks require only verbal encouragement and social stimulation to increase their food and fluid intake.  In our experience, social activities coordinators are willing, even eager to take on the extra responsibility of a snack program because the intervention adds a new dimension to their existing social programs, one the residents seem to appreciate (after all, who doesn’t enjoy snacks at a social event?).  This arrangement leaves nurse aides free to attend to other duties between mealtimes.  Residents not appropriate for mealtime assistance (e.g., those with a strong preference to dine in their rooms for most meals or those who refuse to alter their dining room seating arrangement to allow for group delivery) also may be good candidates for the snack intervention.
• If there is still not enough staff available, then decisions must be made about which residents will receive assistance (e.g., those at highest risk for weight loss) or if other staff (e.g., social activities personnel, volunteers, non-nursing staff trained as “dining assistants”) could help (see Step 3 for a discussion of staffing strategies).

NOTE: The forms referenced in the links within this and other sections of the module can be used to document a resident’s feeding assistance care needs.  Each of the protocols (mealtime assistance or between-meal snack delivery) should be attempted with the resident for a two-day trial (6 meals or 6 snacks) to determine if an individual resident is appropriate.  For meals, if a resident increases their average total percent eaten by 15% or more (i.e., estimated gain of 300 additional daily calories based on a 2000 calorie/day diet) in response to mealtime assistance (compare average total percent eaten during prior week or previous 2 days to average total percent eaten during the 2-day trial), then s/he should continue to receive mealtime assistance.  For snacks, if a resident accepts at least 2 of 3 snack offers per day and consumes roughly 100-150 calories per snack in response to a 2-day (6 snack) trial, then s/he should continue to be offered snacks between meals at least twice daily and preferably three times daily.

In sum, our weight loss prevention intervention enables nursing homes to individualize care so that residents get what they need without overwhelming the staff.  It’s a practical, efficient alternative to providing sub-optimal feeding assistance to all residents, which is the usual practice in many nursing homes.