Step 4: Quality Monitoring

If your staff has accomplished Step 1 (Assessment), Step 2 (Feeding Assistance Trials), and Step 3 (Staffing), then your facility has made a significant investment in improving the quality of feeding assistance for residents. All that time will go to waste, however, unless you conduct regular checks to make sure staff continue to provide quality care. Most nursing homes skip this step only to pay a price for their negligence: studies show that in the absence of quality control assessment, nurse aides do not consistently provide or accurately document the delivery of feeding assistance either during or between meals.

Evidently, old habits are hard to break and new ones are hard to maintain if you don’t get timely feedback about how you’re doing, including reinforcement for doing things right and recommendations for improvement if you’re having trouble. While frequent quality monitoring is absolutely essential at the start of a new program, the good news is that most facilities can get by with less in just a few weeks, once new care patterns are established.

The purpose of mealtime quality control monitoring is two-fold:

1. To determine whether staff are providing consistent feeding assistance; that is, on all days of the week, for all meals; and
2. To assess the quality of feeding assistance for targeted residents

The most reliable way to monitor nursing home care—feeding assistance as well as all other types of care—is to directly observe how the care is provided in daily care practice. This method contrasts with the usual assessment method of using data from medical records (e.g., nurse aide flow sheets) and the Minimum Data Set (MDS) to evaluate care. A common problem with both medical record and MDS documentation is that the information is often tainted with inaccuracies (1-2, 4-5); in other words, you can’t trust it. Moreover, it is almost always in the direction of over-estimating care quality; that is, medical record and MDS documentation both tend to reflect better care practices and better resident outcomes than the reality based on other information sources, such as observation, independent assessments, and resident interview (1-2, 4-5). Thus, it is imperative to assess feeding assistance care quality based on an information source other than the medical record and the best source is direct observation of care delivery. Specifically, a supervisory-level staff person should be assigned to conduct quality control observations during mealtimes. Before you balk at this seemingly expensive requirement, read on: 

• Supervisors need focus their attention only on the estimated 50% of residents with low food and fluid intake who also were “responsive” to the mealtime intervention; that is, they increased their oral intake by 15% or more when provided with optimal feeding assistance (see Step 2); these are the residents who should continue to receive feeding assistance during meals.
• One supervisor can reliably observe feeding assistance for five to ten residents at a time, provided all the residents are in the dining room or in their rooms but within the same hallway.
• To start, each resident who needs feeding assistance should be observed during at least three meals per week, alternating days of the week and meals; if nurse aides provide proper feeding assistance consistently for four weeks across all scheduled mealtime periods, quality control assessments can be reduced to as few as only one meal per week but continue to alternate days of the week and meal periods from week to week.

We estimate that in a typical 100-bed nursing home, one supervisor will initially spend 5 hours per week conducting mealtime quality control observations for the estimated 25 residents with low intake who are responsive to the mealtime intervention. Once the feeding assistance protocol takes hold, the supervisor’s assessment time should drop to about 2.5 hours per week or less. It is easy to see why it would be beneficial to train more than one supervisor to conduct mealtime observations. Several supervisors could share the responsibility of conducting observations to alleviate the burden on any one individual and compensate for an individual’s occasional absence due to illness or vacation. You may want to consider including supervisory-level staff from other departments who have their own unique investment in the quality of the mealtime process (e.g., dietary, registered dietitian, assistant to the Director-of-Nursing, speech therapists, Administrator).

DOUBLE-DUTY ASSESSMENTS: Mealtime monitoring not only ensures quality feeding assistance for targeted residents, but the presence of supervisory-level staff in the dining room during meals also communicates to those supervised that feeding assistance is an important care routine that is valued by management. Quality control observations should support and reinforce educational in-service training sessions related to nutritional care and weight loss prevention (3). 

The supervisor should conduct periodic checks during both week and weekend days, if possible, and across all mealtimes—breakfast, lunch, and dinner. If this is not feasible due to work hours and schedule of the designated supervisor(s), focus on the days and meals that it is possible to do (typically week days, breakfast and/or lunch meals) and consider identifying another supervisory-level person to be responsible for other days/meals. Ideally, the supervisor should observe the entire meal, from tray delivery to tray pick-up, and use our Quality Improvement Observation Form: Meals to record pertinent information. Briefly, the observational form prompts supervisors to collect the following information:

• Total number of residents eating in the dining room

• Names of the residents targeted for observation

• Type of feeding assistance provided to each observed resident (as a practical matter, supervisors can record this information only for residents who eat in the dining room, not for those who eat in their rooms)

• The total percentage eaten by each resident as estimated by the supervisor and then as estimated by the nurse aide in the medical record for the same resident-meal

• The amount of time the nurse aide spent providing assistance to each resident

• Whether a resident consumed an oral liquid nutrition supplement during the meal

TIME-SAVING TIP: If a staff member is unable to observe the entire mealtime period (from start to finish), observations may be strategically conducted at key time points (e.g., beginning, middle, and end) during the meal to capture the same information. An alternative strategy is to observe during only the first half-hour of the meal as this is the time period during which most feeding assistance care is provided, if any is provided at all. 

The information generated by this observational protocol can be summarized as feeding assistance care quality indicator (QI) scores. The advantage of using QI scores is that they highlight clinically significant quality-of-care problems in need of improvement. Additionally, they can be scored as either “passing” or “failing,” for an individual resident and mealtime period which is useful for making comparisons within a facility over time and identifying specific aspects of care that may require more staff education and training. You can use the scores, for example, to compare the quality of feeding assistance over different meal periods or across different staff shifts. These mealtime QI scores can be calculated by hand following the directions at the bottom of the form (Quality Improvement Observation Form: Meals). Alternatively, QI information can be entered into the nutrition software and reports can be generated that summarize the QI scores by date, day of the week, meal period, even staff member.

We present below the rules and rationale that guide the scoring of six QIs related to feeding assistance, all of them based on our previous work (4-6). The scoring rule for each QI reflects a liberal approach that maximizes the opportunity for staff to “pass.”

1) Proportion of residents eating in the dining room

Score: No rule for this one; however...

Rationale: All residents should be encouraged to eat all, or at least most, of their meals in the dining room for several reasons. First, most residents report a preference to eat their meals in the dining room, if given a choice. Second, presence in the dining room allows the staff to provide time-efficient feeding assistance to small groups of residents. Third, dining in a common area promotes social interaction among residents and staff, which in turn stimulates food and fluid intake. Finally, residents who eat in the dining room also receive more attention from staff, better feeding assistance care and more accurate documentation of their oral intake during meals. (See Step 3: Implementing Staffing Strategies for more information about the importance of dining location).

Service/Training Goal: Ideally, all residents, excluding those who are bed-bound, tube-fed, or on hospice or palliative care, should eat all of their meals in the dining room. This includes breakfast and dinner, which are often served in residents’ rooms. Supervisors should work with staff to identify ways to increase the number of residents who eat in the dining room, including using non-traditional staff to help transport residents and offering two seatings per meal period, if dining space is limited. 

2) Staff ability to provide assistance to at-risk residents

Scoring Rule: Score as “fail” residents who eat less than 50% of their food and receive less then five minutes of staff assistance during the meal.

Rationale: All residents with low intake who are responsive to the mealtime intervention should receive feeding assistance for 30 to 45 minutes in small groups of three from one staff member. Thus, if any observed resident receives less than five minutes of assistance, feeding assistance is not being provided according to the protocol. Inadequate feeding assistance is particularly detrimental to residents who consistently eat less than 50% of each meal and thus are at especially high risk for weight loss and under-nutrition.

Service/Training Goal: All nurse aides should provide adequate feeding assistance to all nutritionally at-risk residents (see Step 2).

3) Staff ability to accurately document clinically significant low food and fluid intake among residents 

Scoring Rule: Score as “fail” residents who eat less than 50% of their meal based on the supervisor’s observations, but who are reported by nurse aides to have consumed more than 60%.

Rationale: While residents who consistently eat less than 75% of most meals meet the MDS criterion for low intake, recent evidence suggests that those who consistently eat less than 50% are at a significantly higher risk for weight loss. Thus, if staff document that a resident consumed more than 60% of a meal when, in fact, the resident ate less than 50%, they are likely failing to identify a clinically significant intake problem for that resident.

Service/Training Goal: All nurse aides should be trained to use the same guidelines to calculate residents’ food and fluid intake (see Step 1). Note: before and after photographs of residents’ meal trays serve as a helpful training tool for teaching staff how to conduct intake estimates.

4) Staff ability to provide verbal instruction to residents who receive physical assistance at mealtimes

Scoring Rule: Score as “fail” any resident who receives physical assistance from staff during the meal without also receiving at least one verbal prompt directed toward eating (e.g., “Why don’t you try your soup?”). As a practical matter, this QI can be scored only for residents who eat meals in the dining room due to the difficulty in observing directly multiple nurse aide-resident interactions when the resident is eating in their room.

Rationale: Studies show that verbal prompting encourages residents to eat independently and to eat more. There is growing consensus that verbal prompting alone or, if physical assistance is needed, verbal prompting that precedes and is coupled with physical assistance defines optimal feeding assistance. Moreover, recent research indicates that nursing home staff often provides excessive physical assistance to residents who could otherwise eat independently with just verbal prompting or encouragement. Even if a resident requires full physical assistance to eat, staff should minimally provide verbal notification (“let’s try a bite of soup next, okay?”; “I’m going to give you a bite of soup next.”).

Service/Training Goal: Ideally, all residents who receive physical assistance should also receive verbal instruction or notification from staff. Failure to provide verbal instruction or notification may reflect a language barrier or a need for staff education. Nurse aides, for example, may inappropriately assume that it is a waste of time to provide verbal instruction to residents with cognitive impairment or residents who are unable to verbally communicate.

5) Staff ability to provide social stimulation to all residents during meals

Score: Score as “fail” any resident who does not receive at least one episode of social stimulation from staff during the meal.

Rationale: Studies show that social stimulation improves food and fluid intake; thus, staff should socially interact with all residents throughout the meal. Social interaction differs from verbal instruction in that it consists of simple statements that are not specifically directed toward eating, for example, greeting a resident by name: “Hello, Mrs. Smith, it’s good to see you today.” As a practical matter, this QI can be scored only for residents who eat meals in the dining room.

Service/Training Goal: Ideally, all residents should receive at least one episode of social stimulation from staff during meals. Social interaction not only enhances residents’ oral food and fluid intake, but it also enhances their quality of life.

6) Staff ability to accurately document feeding assistance

Score: Compare how nurse aides describe the provision of feeding assistance in residents’ charts with the supervisor’s recorded observations.

Rationale: This QI enables supervisors to evaluate the accuracy of medical record documentation of feeding assistance and identify strategies to prevent documentation errors.

Service/Training Goal: A discrepancy between how nurse aides and supervisors document both the type and duration of feeding assistance may point to the need for a standardized form for charting care delivery that is more specific than a simple checklist or documentation that feeding assistance was provided “as needed,” neither of which are informative from a quality improvement perspective. Staff may also want to document reasons for not providing assistance (e.g., resident refused the meal or assistance). 

It is just as important for supervisors to monitor the quality of feeding assistance during snack periods as it is during mealtimes, especially when you consider that an estimated half of nutritionally at-risk residents need between-meal snacks to increase their daily caloric intake. Unfortunately, in many nursing homes, staff do not consistently provide snacks and beverages to residents between meals (7), and documentation of residents’ food and fluid intake between meals is typically absent or inaccurate. Quality control monitoring can identify such problems and point the way to feasible solutions. 

Here are tips for conducting snack-time quality control observations:

• Focus your observations first on residents with low intake who showed a significant gain in daily calories in response to our snack intervention (see Step 2). Staff should offer these residents snacks and beverages between meals at least twice a day (morning and afternoon) and, preferably, three times a day (morning, afternoon, and evening).

• To start, monitor each resident during two or more snack periods per week, being sure to vary the days of the week and the snack period (i.e., morning, afternoon, and evening). Reduce your observations to every other week or one snack period per week once proper care routines are firmly established.

• Use our Quality Improvement Observation Form: Between Meal Snacks to record important information about snack-time feeding assistance. Like the mealtime observation protocol, this protocol generates information that can be summarized as quality indicators (QIs), which in turn can be used to target improvement efforts. This information also can be entered into our nutrition software program (link to) to generate summary QI scores for snack delivery by date or snack period.

• Arrange for snacks to be delivered to residents during organized, social group activities so that you can conduct quality control observations in a time-efficient manner during scheduled time periods. Note: This approach will require some organization with the dietary staff to ensure that snack items are delivered to the floor at the scheduled times.

• If you can, check to make sure that all residents are offered fluids between meals. Studies show that the majority of nursing home residents are at high risk for dehydration and the overwhelming majority of residents will increase their fluid intake if prompted to drink fluids multiple times per day between meals (8).

• Be sure that a variety of food and fluid items are offered during each snack period. The availability of choices has been shown to be a particularly important component of the intervention for residents with less cognitive impairment (7, 9).

• Also if possible, monitor consumption of oral liquid nutrition supplements among all residents at snack time. Most residents have physician orders for supplements, but staff tend to offer these only during meals and often as a substitute for the served meal and quality assistance (7,9,10). If taken between meals, supplements not only increase calorie intake, but also act as an appetite stimulant so residents eat more during meals.

• Consider increasing the frequency of observations for any resident who starts to lose weight so that you can quickly correct the problem.