Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Aug 1.
Published in final edited form as: J Nurs Meas. 2021 Apr 16;29(2):E95–E109. doi: 10.1891/JNM-D-19-00101

Reliability and Validity Testing of the Quantified Quality of Interaction Scale (QuIS)

Barbara Resnick 1, Elizabeth Galik 2, Anju Paudel 3, Rachel McPherson 4, Kimberly Van Haitsma 5, Marie Boltz 6, Jeanette Ellis 7, Karen Eshraghi 8, Liza Behrens 9, Shijun Zhu 10, Rachel Blankstein Breman 11
PMCID: PMC8324536  NIHMSID: NIHMS1718052  PMID: 33863846

Abstract

Background and Purpose:

The purpose of this study was to test the reliability and validity of the Quality of Interaction Survey (QuIS) using a quantification scoring approach.

Methods:

Baseline data from the Evidence Integration Triangle for Behavioral and Psychological Symptoms of Dementia (EIT-4-BPSD) study was used.

Results:

A total of 553 residents participated. There was evidence of inter-rater reliability with Kappa scores of .86 to 1.00 and internal consistency based on the Rasch analysis (item reliability of .98). There was some support for validity based on item fit and hypothesis testing as resistiveness to care was significantly associated with total QuIS scores.

Conclusion:

This study supports the use of the quantified QuIS to evaluate the quality of interactions over time and to test interventions to improve interactions.

Keywords: Rasch analysis, geriatrics, dementia, interactions

Interpersonal interactions and the quality of those interactions are central to providing optimal person-centered care to older adults with dementia in nursing home settings (Muralidharan et al., 2019; Williams et al., 2017; Yeung & Rodgers, 2017). Social and care interactions and communication that meet the needs of older adults with dementia have a critically important impact on the individual’s quality of life and overall sense of well-being (McGilton et al., 2012; Roberts & Bowers, 2015). Conversely, interactions that do not address psychosocial and care needs can result in distress and unhappiness among residents. Care interactions and the communication that occurs during those interactions can influence clinical outcomes such as eating (Liu et al., 2017), resistiveness to care that occurs during bathing, dressing, and other care interactions (Herman & Willimas, 2009), and can decrease the use of antipsychotic medications (Tjia et al., 2017). The quality of the interaction also has an impact on how the individual appraises him or herself.

There is an increase in numbers of older adults with serious mental illness and dementia living in nursing homes or other long term care settings. Little focus has been given to the evaluation of the quality of care interactions of staff, family, friends, and others with these residents and the ways to optimize interactions to improve care and clinical outcomes (Grabowski et al., 2010; Li, 2010). To help evaluate the quality of care interactions and test interventions to optimize the interactions, reliable and valid measures of the quality of interactions are needed.

To measure the quality of interactions, the Quality of Interaction Schedule (QuIS) was originally developed based on work done by Clark and Bowling (1989). Clark and Bowling initially categorized the interactions between staff and residents as positive, negative, and neutral in an observational study of quality of life of older adults in nursing homes and long stay wards. The ratings were based on the degree to which staff acknowledge the resident as a person, not to the adequacy of any care delivered during the interaction. This qualitative classification was expanded into five categories by Dean et al. (1993) after recognizing the important subgroups of positive and negative interactions. As described in Table 1, positive care interactions were differentiated to be either primarily social and reflective of good, constructive, beneficial conversation and companionship or related to physical care and reflective of the appropriate delivery of physical care. Negative care interactions were differentiated between those that were protective of residents and those that were restrictive and carried out to make the work of the staff easier.

TABLE 1.

Quality of Interaction Survey Items

Item Description and Examples Score
Positive social Interactions which involves good, constructive beneficial conversation and companionship. Examples include giving encouragement during care tasks; recognizing achievements; smiling and laughing with resident; showing interest; offering choices. 2 if present
0 if not present
Positive care Interactions that result in providing appropriate care and are generally task focused. Examples include verbal brief explanations and encouragement to complete a care specific task; no empathy is expressed but caregiver is not rude; telling someone what is going happen without offering any choice; not showing interest in the conversation/interaction; indifference to resident. 1 if present
0 if not present
Neutral Interactions which include those that are brief and indifferent. Examples include putting a plate down without verbal or nonverbal contact; not saying hello or interacting directly when working with a resident. 0 if present
1 if not present
Negative protective Interactions that focus on keeping the resident safe or removing dangers but in a restrictive way. Examples include providing care to the individual via protection but in a negative way such as telling someone to wait for something without any explanation; treating the resident like a child and using “elderspeak”; ignoring resident’s preferences; scolding a resident for behaviors that might be deemed risky. 0 if present
1 if not present
Negative restrictive Interactions which are those that oppose or resist residents’ freedom of action without a good reason. Examples include moving a resident without warning or explanation; telling them they can’t have something without a reason. 0 if present
2 if not present
Open in a new tab

Use of the Quality of Interaction Scale

The QuIS has been used to evaluate interactions among staff and older adults in long term care facilities, acute care, day care centers, and residential care (Allen & Turner, 1991; Jenkins & Allen, 1998; Lindesay & Skea, 1997; Olusina et al., 2003; Proctor et al., 1998; Shepherd et al., 1996; Skea, 2007, 2014; Skea & Lindsay, 1996). The use of the measure has been descriptive in nature and used to identify the percentage of care interactions that were positive social, positive care, neutral, negative protective, or negative restrictive over a predetermined observation period. The observation periods ranged from 30 seconds (Jenkins & Allen, 1998) to 3 hours (Proctor et al., 1998) although most studies observed resident interactions with staff for a 20 minute period.

Prior testing of the QuIS using this descriptive approach provided evidence of inter-rater reliability based on Cohen’s kappa ranging from .53 to .96 (Dean et al., 1993; Jenkins & Allen, 1998; McLean et al., 2017). Evidence of validity was based on significant relationships between QuIS findings (e.g., positive interactions) and positive patient experiences (Dean et al., 1993) and built on improvements in interactions following an intervention (Ballard et al., 2018). None of these studies have quantified the QuIS to describe how positive the care interactions were and if there was an overall improvement in care interactions over time. The purpose of this study was to quantify the QuIS and test the reliability and validity of this instrument using a quantification scoring approach.

METHODS

This measurement study used baseline data from the Evidence Integration Triangle for Behavioral and Psychological Symptoms of Dementia (EIT-4-BPSD) implementation study. Details of the protocol for this study have been published elsewhere (Resnick et al., 2018). The purpose of the parent study was to test the use of the EIT (Glasgow et al., 2012) to support implementation of person-centered behavioral approaches for management of BPSD among nursing home residents. The study was approved by a University based Institutional Review Board.

Sample

A total of 55 nursing homes from two states were included in this study. To be eligible to participate in the study the settings had to: (a) agree to actively partner with the research team on an initiative to change practice; (b) have at least 100 beds or at least 50 beds if the facility had a dedicated dementia care unit; (c) identity a staff member to be an Internal Champion and work with the research team in the implementation process; and (d) be able to access email and websites via a phone, tablet, or computer. Approximately 10–20 residents were recruited from each of the participating facilities. Residents were eligible to participate if they: (a) were living in a participating nursing home; (b) were 55 years of age or older; (c) had cognitive impairment as determined by a score of 0–12 on the Brief Interview of Mental Status (BIMS) (Mansbach et al., 2014); (d) were not enrolled in Hospice; or (e) were not admitted for short-stay rehabilitation care. Eligible residents were identified by staff and research evaluators approached these individuals to determine if they were able to self-consent using the Evaluation to Sign Consent (ESC) (Resnick et al., 2007). If the resident was not able to independently sign the consent, assent was obtained from the resident and the legally authorized representative (LAR) was approached to complete the consent process. A total of 1,095 residents were approached and 590 were consented. Of those approached, 38 were noncommunicative, did not understand English, or died before they could be consented, 156 refused to assent or consent to participate, 221 LARs were unavailable, and 90 LARs refused to consent. Of those that consented, 37 residents were not eligible: 19 had a BIMS score greater than 12, 7 were under the age of 55, and 11 were on Hospice services. A total of 553 residents were enrolled into the study. Of these, there were 17 individuals that died, were hospitalized, or were transferred out of the facility prior to the collection of baseline data. One individual was not available at the time of testing and therefore QuIS data was obtained on a total of 535 residents.

Procedure

All assessments were done by research evaluators based on a one time direct observation for observed measures of the resident or input from the staff for measures that were based on proxy input. Observations were done for approximately 20 minutes during periods of care or activities such as meal times, bathing, dressing or toileting, or during recreational activities. During the observation period, interactions with staff, family, friends, visitors, or other residents were evaluated. As shown in Table 1, the following interactions based on the QuIS were observed: (a) ”positive social” care which involves good, constructive beneficial conversations and companionship; (b) ”positive care” which includes interactions that result in providing appropriate care and are generally task focused; (c) ”neutral” interactions which include those that are brief and indifferent; (d) ”negative protective” which are interactions that focus on keeping the resident safe or removing dangers but in a restrictive way; and (e) ”negative restrictive” interactions which are those that oppose or resist residents freedom of action without a good reason. Either the care interaction was observed by a staff member providing care to a resident or it was not. During an observation, the resident was evaluated for exposure to each of the different types of care interactions (e.g., positive social and neutral).

Measures

Descriptive information about residents included age, race, sex, comorbidities, and cognitive status. Research evaluators that were trained in completion of the measures and in working with older adults with cognitive impairment completed all observation assessments and obtained relevant data from staff. The evaluators all demonstrated the ability to correctly complete these observations measures during their training. Cognitive status was evaluated using the BIMS (Mansbach et al., 2014). BIMS scores can range from 0 to 15 with scores 13–15 indicative of no impairment, scores 8–12 indicative of moderate cognitive impairment, and scores 0–7 indicative of severe cognitive impairment. As noted, based on inclusion criteria, participants had to have scores of less than or equal to 12 which was indicative of moderate to severe cognitive impairment. Comorbidities were based on a summation of the 14 comorbid conditions listed in the Cumulative Illness Rating Scale for Geriatrics (Linn et al., 1968).

Depression, Agitation, and Resistiveness to Care

Depressive symptoms were measured using the Cornell Scale for Depression in Dementia (CSDD) (Alexopoulos et al., 1988a, 1988b).The CSDD is a 19 item survey that is used to describe depressive symptoms in individuals with dementia. Informants, which included the staff working with the resident on the day of testing, were asked to rate the resident’s mood and behavior over the past 2 weeks as either absent, mild/intermittent, or severe. Examples of items include such things as evidence of sadness, anxiety, irritability, and lack of reactivity to pleasant events. Prior testing provided evidence of the reliability and validity of the measure (Alexopoulos et al., 1988a, 1988b). Aggressive and/or resistive behaviors during care were measured using the Resistiveness to Care Scale (RTC) (Mahoney et al., 1999). The RTC scale includes 13 items that assess residents’ behaviors during activities of daily living. Research evaluators observe staff providing care to residents over a 15–30 minute period and record evidence of any resistive behaviors. Examples of behaviors recorded include evidence of hitting or kicking at staff during care or refusing care. Prior testing provided evidence of reliability and validity (Galik et al., 2017; Mahoney et al., 1999).

Agitated behaviors were measured using the Cohen-Mansfield Agitation Inventory (CMAI) (Cohen-Mansfield, 2014; Finkel et al., 1992). The CMAI is a survey of behavioral symptoms that frequently challenge caregivers and are commonly found in residents with dementia. Like the CSDD, the staff working with the resident on the day of testing were asked to respond to the items on the CMAI and rate if any of the activities were never noted, seen less than once a week, once or several times a week, once or several times a day, a few times an hour or continuous for half an hour or more. Examples of items on the 14 item version of the CMAI include frequency of cursing or verbal aggression, hitting, complaining, or general restlessness. Prior testing provided evidence of reliability and validity of this measure (Cohen-Mansfield, 1997).

Quality of life was evaluated using the Quality of Life in Late-Stage Dementia (QUALID) (Logsdon et al., 2002). The QUALID is also based on input from the staff providing care to the resident on the day of testing. These informants are asked to rate the resident’s quality of life over the past 2 weeks in each area using one of four words: poor, fair, good, or excellent. The QUALID scale includes 11 observable behaviors thought to be indicative of quality of life such as whether or not the individual smiles, appears sad, cries, has facial expressions of discomfort, appears emotionally calm and comfortable, or is irritable or aggressive. A lower score is indicative of better quality of life. Prior testing supported the reliability and validity of this measure (Logsdon et al., 2002).

Measurement of the Quality of Care Interactions

The quantified measurement of quality of interactions between residents and staff was based on the QuIS (Dean et al., 1993). As noted above, to complete the QuIS residents and staff interactions were observed for approximately 20 minutes. A scoring format was developed to quantify the QuIS as delineated in Table 1. If there was evidence that the resident received positive social care (e.g., interactions which involve good constructive beneficial conversation and companionship such as providing encouragement during care tasks), which was considered to be the most desirable type of care residents could receive, 2 points were allocated. If there was evidence of positive care (e.g., interactions that result in providing acceptable, appropriate care such as giving brief explanations and encouragement to complete a care specific task), which is consistent with acceptable care, a score of 1 point was given. If care was neutral (brief interactions that are indifferent such as putting a plate down without verbal or nonverbal contact), a score of 0 was given. If there was no evidence of neutral care, participants received a score of 1 as the individual was given credit for providing some interaction and not being neutral. If there was no evidence of negative protective care (interactions that focused on keeping the resident safe by reprimanding them for doing something potentially unsafe or restricting their movement) a score of 1 was given. If there was evidence of negative protective care a score of 0 was given. Negative restrictive care interactions (interactions that decrease a resident’s freedom of action without a good reason such as moving a resident from one dining room table to another without warning or explanation) are the least desirable care interactions provided to residents thus a score of 2 was given when this care was not provided. If there was evidence of negative restrictive care a score of 0 was given. Negative protective care is not as undesirable as negative restrictive care as it focused on resident safety. Thus the score of 1 for not providing negative protective care versus a 2 for not providing negative restrictive care. For each observation period scores could range from 0 to 7 with higher scores indicating a better, or more positive, quality of interaction between the staff member and the resident.

Data Analysis

Descriptive statistics were done to describe the sample using SPSS version 24.0. The RTC and CMAI were positively skewed and a Log 10 transformation was done to achieve a normal distribution. As described below, to evaluate the reliability and validity of the quantified QuIS measure a Rasch analysis was done using the Winsteps statistical program. A linear regression analysis was done to test the validity of the measure based on hypothesis testing.

Reliability Testing.

Testing of the internal consistency of the quantified QuIS was done using a Rasch measurement model and item reliability (Smith & Smith, 2004). Winsteps provides a person separation index which is the equivalent of internal consistency based on logit values (Fox & Jones, 1998; Tennant & Conaghan, 2007). Although there are no universally acceptable cutoff values, a minimum value of 0.7 is considered sufficient evidence of internal consistency (Tennant & Conaghan, 2007). Inter-rater reliability based on Cohen’s Kappa (Fleiss et al., 2003) was completed with 16 randomly selected evaluations of the QuIS completed by a second evaluator. Cohen’s Kappa results are interpreted as follows: values ≤ 0 indicate no agreement; 0.01–0.20 as none to slight agreement; 0.21–0.40 as fair agreement; 0.41–0.60 as moderate agreement; 0.61–0.80 as substantial agreement; and 0.81–1.00 as perfect agreement (Landis & Koch, 1977).

Validity Testing.

Validity testing was based on construct validity of the measure and evidence that each item fit the data and was associated with the concept of quality of interactions. The Winsteps statistical program was used to establish item fit based on INFIT and OUTFIT statistics. INFIT and OUTFIT statistics were considered acceptable if they were between 0.4 and 1.6 (Linacre, 2006). An INFIT or OUTFIT value of less than 0.4 indicates that the item may not provide additional information beyond the rest of the items on the scale. An INFIT or OUTFIT value of greater than 1.6 indicates that the item may not define the same construct as the rest of the items in the instrument, is poorly written and thus may have been misunderstood by participants, or is ambiguous (Smith & Smith, 2004). In addition to establishing item fit, item mapping was also considered to determine if the items covered the full scope of quality of care interactions.

Support for the validity of the measure was based on hypothesis testing. It was hypothesized that there would be a significant relationship between the quantified QuIS and evidence of depressive symptoms, agitation, resistiveness to care, and quality of life. Using linear regression and controlling for age, sex, race, cognition, and comorbidities, it was hypothesized that those experiencing more positive quality of interactions would have fewer depressive symptoms, less agitation, less resistiveness to care, and better quality of life. A stepwise analysis was used with an entry level of <.05 and a removal level of >.10. A stepwise approach was used to monitor the order by which variables were removed or added as it was hypothesized that all variables would be significantly associated with QuIS scores. A significance level of p ≤ .05 was used in all analyses. A sample size of 535 residents observed with 9 variables in the regression analysis was sufficient to assure a reliable model and power was sufficient to identity an R2 of 0.04 (van Bell et al., 2004).

RESULTS

A total of 535 of the recruited 553 residents were observed during care interactions. As shown in Table 2, the mean age of the participants was 83.88 (standard deviation [SD] = 10.44) and the mean score on the BIMS was 4.31 (SD = 3.50) indicating that overall the participants had severe cognitive impairment. The majority were White (N = 419, 76%) and female (N = 398, 72%). There was little evidence of agitation with a mean score of 21.02 (SD = 8.36), little resistiveness to care with a mean score of .60 (SD = 2.07), few depressive symptoms with a mean of 4.26 (SD = 4.55), and fairly good quality of life with a mean of 17.16 (SD = 6.72).

TABLE 2.

Descriptive Outcomes for Residents

N (%) Range Mean Standard Deviation
Race
 White 419 (76%)
 Black 134 (24%)
Sex
 Male 398 (72)
 Female 155 (28%)
Cohen-Mansfield Agitation Inventory 14–54 21.02 8.36
Resistance to Care 0–19 .60 2.07
Quality of Life With Alzheimer’s Disease 10–48 17.16 6.72
Cornell Scale for Depression in Dementia 0–28 4.26 4.54
Age 56–105 83.88 10.45
Quality of Interactions Score 0–7 5.56 1.73
Brief Interview for Mental Status 0–12 4.31 3.50
Comorbidities 2–12 7.10 2.16
Open in a new tab

Two hundred and thirteen (39%) observations for the QUIS were done in the dining room, 41 (7%) were done in living/activity rooms, 11 (2%) were done by the nurses station, 54 (10%) were done in the hallway, 202 (36%) were done in the resident’s room, 7 (1%) were done in the bathroom, 6(1%) were done in the shower, and 19(3%) were done in other locations such as the therapy room. The majority of the observations included interactions with nursing staff (N = 405, 76%), 65 (12%) were with activity staff, 43 (8%) were with other staff such as dietary or housekeeping, 13(2%) were with other residents, 9 (2%) were with family or other visitors.

Overall the quality of the interactions were positive with a mean of 5.56 (SD = 1.72, range 0–7). As shown in Table 3, positive social and positive care interactions were noted in over half of the observations. Neutral care was observed in 33% of the observations and negative protective care and negative restrictive care was observed in less than 10% of the observations.

TABLE 3.

Frequency of Quality of Interaction Survey Outcomes During the 535 Observations

Item Present N (%) Not Present N (%)
Positive Social 360 (67%) 175 (33%)
Positive Care 312 (58%) 223 (42%)
Neutral 124 (33%) 411 (77%)
Negative Protective 31 (6%) 504 (94%)
Negative Restrictive 23 (4%) 512 (96%)
Open in a new tab

Reliability Testing

Kappa results for the 16 observations done by 2 observers were .86 for positive social care interactions, 1.00 for positive care interactions, 1.00 for neutral care interactions, 1.00 for negative protective care interactions, and 1.00 for negative restrictive care interactions. Internal consistency based on the Rasch analysis resulted in an item separation of 6.33 and an item reliability of .98.

Validity Testing

There was some support for the fit of the quantified items of the QuIS to the Rasch measurement model based on INFIT and OUTFIT statistics. As shown in Table 4, the INFIT and OUTFIT statistics for positive social care interactions were just slightly high with an INFIT of 1.72 (ZSTD 9.9) and an OUTFIT of 1.71 (ZSTD 9.9) respectively. Positive care interactions had a good INFIT statistic at .83 (ZSTD-2.6) but a high OUTFIT statistic of 4.60 (ZSTD 9.9). Neutral care interactions had a good fit with an INFIT statistic of .40 (−9.9) and an OUTFIT statistic of 1.57 (ZSTD 3.9). Negative protective care interactions had a good fit as well with an INFIT statistic of .65 (ZSTD −6.7) and an OUTFIT statistic of 1.60 (ZSTD 6.1). Lastly, negative restrictive interactions had a slightly high INFIT statistic of 1.80 (ZSTD 4.4) and a good OUTFIT statistic of .75 (−.30).

TABLE 4.

INFIT and OUTFIT Statistics for the Quantified QuIS

Item INFIT MNSQ (ZSTD) OUTFIT MNSQ (ZSTD) Item Difficultya
Positive social care which involves good, constructive beneficial conversation and companionship 1.72 (9.9) 1.71 (9.9) 4
Positive care which includes interactions that result in providing appropriate care and are generally task focused .83 (−2.6) 4.60 (9.9) 1
Neutral interactions which include those that are brief and indifferent .65 (−6.7) 1.60 (6.1) 2
Negative protective are interactions that focus on keeping the resident safe or removing dangers but in a restrictive way 1.80 (4.4) .75 (−.30) 3
Negative restrictive interactions which are those that oppose or resist residents freedom of action without a good reason .40 (−9.9) 1.57 (3.9) 5
Open in a new tab

Note. MNSQ = Mean Square; ZSTD = standardized fit statistics.

a

1 is easiest item to endorse and 5 is most difficult item to endorse.

There was limited support for hypothesis testing. Controlling for age, sex, race, cognition, and comorbidities, resistiveness to care was the only variable significantly associated with total QuIS score such that those who had better overall care interactions had less resistiveness to care (beta = −.15, t = −3.73, p = .001). Taken together age, sex, race, cognition, comorbidities, and resistiveness to care explained 15% of the variance in quality of interactions. Agitation (t = −.31, p = .76), depression (t = .07, p = .95) and quality of life (t = −1.74, p = .08) were not significantly associated with quality of care interactions.

Item mapping (Table 4) showed that the least likely interaction for staff to perform was negative restrictive, then positive social, then negative protective, then neutral care interactions, and the easiest item to perform was positive care. There were no resident-staff care interactions that were so high or so low in providing overall positive care interactions that they could not be differentiated.

DISCUSSION

This study provided some evidence of the reliability and validity of the quantified QuIS. There was sufficient evidence of reliability with internal consistency based on an item reliability of .98, which is consistent with an alpha coefficient of equal size (Smith & Smith, 2004). There was also strong evidence of inter-rater reliability with each item showing a kappa score of .86 or greater. This is consistent with the early development work of the QuIS with alpha coefficients and kappa scores all being greater than .70 (Dean et al., 1993). Kappa scores were higher in the nursing home setting than when testing was done with patients in acute care settings and residents in residential or assisted living settings where the kappa scores ranged from .53 to .62 (Jenkins & Allen, 1998; McLean et al., 2017). As delineated in Table 1 there are clear descriptions of the types of interactions that are considered to be positive, negative, or neutral. Evaluators were trained in the completion of the measure using these descriptions and thus were able to reliably describe the interactions.

There was less consistent support for the validity of the quantified QuIS. Positive social and negative protective care interactions had slightly high INFIT statistics which suggest that the item may not have been as predictable as would be preferred. For both items, however, the INFIT statistic was less than two which suggests that there is reasonable predictability (Linacre, 2002). Thus continued use of the item as described is appropriate. Positive care had a high OUTFIT statistic although the INFIT statistic was within an acceptable range. OUTFIT statistics are less important than INFIT statistic as they address difficulty levels that are far from the person. OUTFIT reports overfit for imputed responses or underfit for lucky guesses or careless mistakes. These issues are easier to identify when reviewing responses. There is therefore no reason to revise this item at this time.

The majority of the observations were with staff and residents with only a small percentage (5%) including family, visitors, or other residents. Mapping of the items indicated that the items covered the concept of quality of care interactions as there were no observed interactions that were so high or so low in providing overall positive care interactions that they could not be differentiated. As has been consistently noted in prior research with staff and residents in long term care settings or patients in hospitals (Ballard et al., 2018; Clark & Bowling, 1989; Jenkins & Allen, 1998; McGilton et al., 2012; McLean et al., 2017; Mesa-Egualagaray et al., 2016), the majority of observations resulted in positive social or positive care interactions. In this study, approximately one third of the observations were neutral and 10% were negative, either negative protective or negative restrictive. It may be helpful, therefore, to teach staff, as well as families and other residents to change how care interactions occur and to decrease or eliminate the neutral and negative care interactions. Examples of useful interventions to improve care interactions include the WHELD intervention (Ballard et al., 2018) and the Culture Change Studio Engagement Programme (Guzman et al., 2017) both of which involved staff nurse training in person-centered care and social interaction. These interventions had a positive impact on the quality of care interactions. One simple intervention used in Function Focused Care interventions (Resnick, 2011) is to alter the commonly expressed negative restrictive statement, “Don’t get up you might fall” by staff and families. In contrast Function Focused Care interventions encourage staff and families and visitors to have the resident stand, march in place or ambulate if possible when he or she is attempting to get up alone.

Controlling for age, sex, comorbidities, race, and cognition only resistiveness to care was associated with the quality of care interactions. The more resistiveness to care there was the lower, or less positive, the quality of care interactions. This study was correlational in nature and thus causality cannot be inferred. It is possible that there was more resistiveness to care because of the quality of the care interaction, or that the quality of care interaction was more negative because the resident was resistive to care and the staff responded negatively to this behavior. Describing the details of the interaction in terms of whether the resistiveness to care preceded the negative interactions on the part of the staff versus if the negative interaction resulted in resistiveness to care in future work will help to inform causality.

Together the variables in the model only explained 15% of the variance in quality of care interactions. There are, therefore, other factors to consider that might influence quality of care interactions. These may include staff, family, and visitor characteristics such as their knowledge and experience caring for individuals with severe cognitive impairment and environment factors such as a noisy or unpleasant work environment, or staff burnout.

Quality of life, agitation, or depressive symptoms were not associated with quality of care interactions in this study. These findings are consistent with the general lack of evidence between QuIS findings and positive patient experiences (McLean et al., 2017). Prior research has, however, noted that there is a relationship between quality of care, although not measured by the QuIS, and quality of life (McGilton et al., 2012; Roberts & Bowers, 2015). The lack of significance between these variables and QuIS outcomes in this study may be reflective of the sample which included those with severe cognitive impairment. The QUALID, the CMAI, and the CSDD were all developed specifically for those with moderate to severe cognitive impairment as these individuals are unable to respond to questions about quality of life, depressive symptoms, or agitation and thus assessments are done based on informants. The informant assessments, however, maybe biased by staff recall and opinions about the resident. Despite this measurement challenge, research should continue to explore the impact of the findings from the QuIS and patient related factors, particularly quality of life.

STUDY STRENGTHS AND LIMITATIONS

This study was limited in that it was a secondary data analysis from sites in just two states and included residents with severe dementia. Observations were obtained at a single point in time and specific details such as the time of day of the observation were not recorded. Resident outcome data was based on input from the staff in terms of behavior of the resident and direct observation. These results may be biased and not be a true reflection of the resident’s mood or quality of life. The observations to obtain the quality of the interactions were based on a brief 20 minute observation period and consequently did not capture all of the interactions between residents and staff. Despite these limitations this study provides some support for the use of the quantified QuIS when evaluating the quality of care interactions between residents, staff, families, friends, and visitors. A quantified measure can help with the description of the overall quality of interactions between staff and residents and the evaluation of interventions to improve the quality of interactions provided to residents.

Funding.

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Institute of Nursing Research R01NR015982-01A1, awarded to Barbara Resnick.

Footnotes

Disclosure. The authors have no relevant financial interest or affiliations with any commercial interests related to the subjects discussed within this article.

Contributor Information

Barbara Resnick, University of Maryland School of Nursing, Baltimore, MD.

Elizabeth Galik, University of Maryland School of Nursing, Baltimore, MD.

Anju Paudel, University of Maryland School of Nursing, Baltimore, MD.

Rachel McPherson, University of Maryland School of Nursing, Baltimore, MD.

Kimberly Van Haitsma, Pennsylvania State University College of Nursing, University Park, PA.

Marie Boltz, Pennsylvania State University College of Nursing, University Park, PA.

Jeanette Ellis, University of Maryland School of Nursing, Baltimore, MD.

Karen Eshraghi, Pennsylvania State University College of Nursing, University Park, PA.

Liza Behrens, Pennsylvania State University College of Nursing, University Park, PA.

Shijun Zhu, University of Maryland School of Nursing, Baltimore, MD.

Rachel Blankstein Breman, University of Maryland School of Nursing, Baltimore, MD.

REFERENCES

  1. Alexopoulos G, Abrams R, Young R, & Shamoian C (1988a). Cornell scale for depression in dementia. Biology and Psychiatry, 23, 271–284. 10.1016/0006-3223(88)90038-8 [DOI] [PubMed] [Google Scholar]
  2. Alexopoulos G, Abrams R, Young R, & Shamoian C (1988b). Use of the Cornell Scale in non-demented patients. Journal of the American Geriatrics Society, 36, 230–236. 10.1111/j.1532-5415.1988.tb01806.x [DOI] [PubMed] [Google Scholar]
  3. Allen C, & Turner P (1991). The effect of an intervention programme on interactions on a continuing care ward for older people. Journal of Advanced Nursing, 16(10), 172–177. 10.1111/j.1365-2648.1991.tb01525.x [DOI] [PubMed] [Google Scholar]
  4. Ballard C, Corbett A, Orrell M, Williams G, Moniz-Cook E, Romeo R, Orrell M, Woods B, Garrod L, Testad I, Woodward-Carlton B, Wenborn J, Knapp M, & Fossey J (2018). Impact of person centred care training and person centred activities on quality of life, agitation, and antipsychotic use in people with dementia living in nursing homes: A cluster randoised controlled trial. PLOS Medicine, 15, e1002500. 10.1371/joumal.pmed.1002500 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Clark P, & Bowling A (1989). Observational study of quality of life in NHS nursing homes and a long stay ward for the elderly. Ageing Society, 9, 123–148. 10.1017/S0144686X00013520 [DOI] [Google Scholar]
  6. Cohen-Mansfield J (1997). Conceptualization of agitation: Results based on the cohen-mansfield agitation inventory and the agitation behavior mapping instrument. International Psychogeriatrics, 8, 309–315. 10.1017/S1041610297003530 [DOI] [PubMed] [Google Scholar]
  7. Cohen-Mansfield J (2014). Instructional manual for the Cohen-mansfield agitation inventory. http://www.dementia-assessment.com.au/symptoms/CMAI_Scale.pdf
  8. Dean R, Proundfoot R, & Lindesay J (1993). The Quality of Interaction Schedule (QUIS) development, reliability and use in the evaluation of two domus units. International Journal of Geriatric Psychiatry, 8(10), 819–826. 10.1002/gps.930081004 [DOI] [Google Scholar]
  9. Finkel S, Lyons J, & Anderson R (1992). Reliability and validity of the CMAI in institutionalized elderly. International Journal of Geriatric Psychiatry, 7, 487–190. 10.1002/gps.930070706 [DOI] [Google Scholar]
  10. Fleiss J, Levin B, & Paik M (2003). Statistical methods fro rates and proportions. Wiley. [Google Scholar]
  11. Fox C, & Jones J (1998). Uses of Rasch modeling in counseling psychology research. Journal of Counseling and Psychology, 45, 30–45. 10.1037/0022-0167.45.l.30 [DOI] [Google Scholar]
  12. Galik E, Resnick B, Holmes S, Vigne E, & Nalls V (2017). Reliability and validity of the resistiveness to care scale among cognitively impaired older adults. The Journal of the American Medical Directors Association, 18(1), 59–64. 10.1016/j.jamda.2016.08.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Glasgow R, Green L, Taylor M, & Stange K (2012). An evidence integration traingle for aligning science with poliy and practice. American Journal of Preventive Medicine, 42, 646–654. 10.1016/j.amepre.2012.02.016 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Grabowski D, Aschbrenner K, Rome V, & Bartels S (2010). Review: Quality of mental health care for nusing home residents: A literature review. Medical Care Research Review, 67, 627–656. 10.1177/1077558710362538 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Guzman A, Wenborn J, Swinson T, & Orrell M (2017). Evaluation of the “Ladder to the moon, culture change studio engagement programme” staff training: Two quasi-experimental case studies. International Journal of Older People Nursing, 12(3), 1–9. 10.1111/opn.12147 [DOI] [PubMed] [Google Scholar]
  16. Herman R, & Willimas K (2009). Elderspeak’s influence on resistiveness to care: Focus on behavioral events. American Journal of Alzheimer’s Disease & Other Dementias, 24, 417–423. 10.1177/1533317509341949 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Jenkins H, & Allen C (1998). The relationship between staff burnout/distress and interactions with residents in two residential homes for older people. International Journal of Geriatric Psychiatry, 13, 466–472. [DOI] [PubMed] [Google Scholar]
  18. Landis J, & Koch G (1977). The measurement of observer agreement for categorial data. Biometrics, 33(1), 159–174. 10.2307/2529310 [DOI] [PubMed] [Google Scholar]
  19. Linacre J (2002). What do Infit and Outfit, mean-square and standardized mean? Rasch Measurement Transactions, 16, 878. [Google Scholar]
  20. Linacre J (2006). WINSTEPS Rasch measurement computer program. WINSTEPS.Com.
  21. Lindesay J, & Skea D (1997). Gender and interactions between care staff and elderly nursing home residents with dementia. International Journal of Geriatric Psychiatry, 12, 344–348. [DOI] [PubMed] [Google Scholar]
  22. Linn B, Linn M, & Gurel L (1968). Cumulative illness rating scale. Journal of the American Geriatric Society, 16(5), 622–626. 10.1111/j.1532-5415.1968.tb02103.x [DOI] [PubMed] [Google Scholar]
  23. Liu W, Jao Y, & Williams K (2017). The association of eating performance and environmental stimulation among older adults with dementia in nursing homes: A secondary analysis. International Journal of Nursing Studies, 71, 70–79. 10.1016/j.ijnurstu.2017.03.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Li Y (2010). Provision of mental health services in U.S. nursing homes, 1995–2004. Psychiatric Services, 61, 349–355. 10.1176/ps.2010.6T4.349 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Logsdon R, Gibbons L, McCurry S, & Teri L (2002). Assessing quality of life in older adults with cognitive impairment. Psychosomatic Medicine, 64, 510–519. 10.1097/00006842-200205000-00016 [DOI] [PubMed] [Google Scholar]
  26. Mahoney E, Hurley A, Volicer L, Bell M, Gianotis P, Hartshorn M, Lane P, Lesperance R, MacDonald S, Novakoff L, Rheaume V, Timms R, & Warden V (1999). Development and testing of the resistiveness to care scale. Research in Nursing and Health, 22(1), 27–38. [DOI] [PubMed] [Google Scholar]
  27. Mansbach W, Macea R, & Clarka M (2014). Differentiating levels of cognitive functioning: A comparison of the Brief Interview for Mental Status (BIMS) and the Brief Cognitive Assessment Tool (BCAT) in a nursing home sample. Aging & Mental Health, 18(7), 921–928. 10.1080/13607863.2014.899971 [DOI] [PubMed] [Google Scholar]
  28. McGilton K, Sorin-Peters R, Sidani S, Boscart V, Fox M, & Rochon E (2012). Patient centred communication intervention study to evaluate nurse patient interactions in complex continuing care. BMC Geriatrics, 12, 61. 10.1186/1471-2318-12-61 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. McLean C, Griffiths P, Mesa-Equlagaray I, Pickering R, & Bridges J (2017). Reliability, feasibility, and validity of the quality of interactions schedule (QuIS) in actue hospital care: An observational study. BioMed Central Health Services Research, 17, 380–390. 10.1186/sl2913-017-2312-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Mesa-Egualagaray I, Bohning D, McLean C, Griffiths P, Bridges J, & Pickering R (2016). Onter rater reliability of the QuIS as an assessment of the quality of staff-inpatient interactions. BioMed Central Research Methodology, 16, 171–181. 10.1186/sl2874-016-0266-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Muralidharan A, Mills W, Evans D, Fujii D, & Molinari V (2019). Preparing long-term care staff to meet the needs of aging persons with serious mental illness. Journal of the American Medical Directors Association, 20(6), 683–688. 10.1016/j.jamda.2019.03.018 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Olusina A, Ohaeri J, & LOlatawura M (2003). The quality of interactions between staff and psychiatric inpatients ina Nigerian general hospital. Psychopathology, 36(5), 269–275. 10.1159/000073453 [DOI] [PubMed] [Google Scholar]
  33. Proctor R, Powell H, Bums A, Tarrier N, Reeves D, Emerson E, & Elatton C (1998). An observational study to evaluate the ipact of a specialist outreach team on the quality of care in nursing and residential home. Aging & Mental Health, 2(3), 232–238. 10.1080/13607869856713 [DOI] [Google Scholar]
  34. Resnick B (2011). Restorative care for older adults: A guide for all care settings. Springer Publishing. [Google Scholar]
  35. Resnick B, Gruber-Baldini A, Abolf-Petzer I, Galik E, Russ K, & Zimmerman S (2007). Reliability and validity of the evaluation to sign consent meausre. The Gerontologist, 47(1), 69–77. 10.1093/geront/47.E69 [DOI] [PubMed] [Google Scholar]
  36. Resnick B, Kolanowski A, Van Elaitsma K, Galik E, Boltz M, Ellis J, Behrens L, Flanagan NM, Eshraghi KJ, & Zhu S (2018). Testing the evidence integration triangle for implementation of interventions to manage behavioral and psychological symptoms associated with dementia: Protocol for a pragmatic trial. Research in Nursing and Health, 41(3), 228–242. 10.1002/nur.21866 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Roberts T, & Bowers B (2015). How nursing home residents develop relationships with peers & staff: A grounded theory study. International Journal of Nursing Studies, 52(1), 57–67. 10.1016/j.ijnurstu.2014.07.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Shepherd G, Muijen M, Dean R, & Cooney M (1996). Residential care in hospital and in the community-quality of care and quality of life. British Journal of Psychiatry, 168(4), 448–156. 10.1192/bjp.168.4.448 [DOI] [PubMed] [Google Scholar]
  39. Skea D (2007). Quality of staff service user interaction in two day centres for adults with learning disabilities. Mental Health and Learning Disabilities Research and Practice. https://www.researchgate.net/publication/272772092_Quality_of_Staff_Service_User_Interaction_in_Two_Day_Centres_for_Adults_with_Leaming_Disabilities [Google Scholar]
  40. Skea D (2014). A proposed care training system: Quality of interaction training with staff and carers. International Journal of Caring Sciences, 7(3), 750–756. [Google Scholar]
  41. Skea D, & Lindsay J (1996). An evaluation of two models of long term residential care for elderly people with dementia. International Journal of Geriatric Psychiatry, 11(3), 233–241 [DOI] [Google Scholar]
  42. Smith E, & Smith R (2004). Introduction to Rasch measurement. JAM Press. [Google Scholar]
  43. Tjia J, Hunnicutt J, Herndon L, Blanks C, Lapane K, & Wehry S (2017). Association of a communication training program with use of antipsychotics in nursing homes. JAMA Internal Medicine, 177(6), 846–853. 10.1001/jamaintemmed.2017.0746 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. van Bell G, Fisher L, Heagerty P, & Lumley T (2004). Biostatistics: A methodology for the health sciences. Wiley-Interscience. [Google Scholar]
  45. Williams K, Shaw C, Lee A, Sohyun K, Dinneen E, Turk M, Ying-Ling J, & Lui W (2017). Voicing ageism in nursing home dementia care. Journal of Gerontological Nursing, 43(9), 16–20. 10.3928/00989134-20170523-02 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Yeung P, & Rodgers V (2017). Quality of long-term care for older people in residential settings – Perceptions of quality of life and care satisfaction from residents and their family members. Nursing Praxis in New Zealand, 33(1), 28–43. 10.36951/NgPxNZ.2019.004 [DOI] [Google Scholar]

RESOURCES