Skip to main content
NCBI home page
The Gerontologist logoLink to The Gerontologist
. 2022 Dec 27;63(8):1395–1404. doi: 10.1093/geront/gnac193

Determining Evidence for Family Caregiver Communication: Associating Communication Behaviors With Breakdown and Repair

Kristine N Williams 1,2, Carissa K Coleman 3,✉, Jinxiang Hu 4
Editor: Suzanne Meeks
PMCID: PMC10474591  PMID: 36574501

Abstract

Background and Objectives

Communication is fundamental for dementia care. The trouble source repair (TSR) framework can identify strategies that facilitate or impede communication in dyadic interactions.

Research Design and Methods

A secondary analysis of videos (N = 221) from a clinical trial of a family caregiver telehealth intervention was analyzed using sequential behavioral coding of communication behaviors and breakdowns for 53 caregiver and person with dementia dyads. Coded data from 3,642 30-s observations were analyzed using penalized regression for feature selection followed by Bayesian mixed-effects modeling to identify communication strategies associated with communication breakdown and repair.

Results

Breakdown (coded as 0) was associated with caregivers changing topic (median = −11.45, 95% credibility interval [CrI; −24.34, −4.37]), ignoring (median = −11.49, 95% CrI [−24.49, −4.72]), giving commands (median = −10.74, 95% CrI [−24.22, −3.38]), and taking over the task (median = −4.06, 95% CrI [−7.28, −1.77]). Successful repair of breakdown was associated with verbalizing understanding (median = 0.46, 95% CrI [0.09, 0.86]), tag questions, (median = 2.4, 95% CrI [0.33, 5.35]), and silence (median = 0.78, 95% CrI [0.42, 1.15]) and negatively associated with ignoring and changing topic (median = −3.63, 95% CrI [−4.81, −2.57] and −2.51 [−3.78, −1.33], respectively).

Discussion and Implications

The TSR was effective in identifying specific communication strategies to avoid (changing topic, ignoring, commands, and taking over the task) and to use to repair breakdown (verbalize understanding, tag questions, and silence). Future research is needed to test these strategies and explore the potential effects of dementia stage, diagnosis, and dyad characteristics in additional samples. Behavioral coding provides evidence of communication best practices as a basis for family caregiver communication training.

Keywords: Best practices, Care partner, Dementia, In-home care, Person-centered care

Families provide care at home for most persons with dementia, saving the health care system $232 billion annually (Alzheimer’s Association, 2021). However, the stress of caregiving causes depression, insomnia, and other negative effects that are associated with caregiver morbidity and mortality (Monin & Schulz, 2009). Progressive loss of communication abilities accompanies a cognitive decline in persons with dementia, and communication breakdown contributes to behavioral and psychological symptoms (BPSD) and presents challenges to providing care, thus increasing stress and threatening the caregiver person with dementia relationship. These factors significantly affect the quality of life and contribute to nursing home placement (Alzheimer’s Association, 2021).

As dementia progresses, expressive and receptive impairments result in communication challenges due to linguistic, personal, and relational factors influencing communication. Initial changes affecting communication include difficulty finding words and comprehending abstract language to inability to process complex communication and express feelings and needs, leading to speech that may seem meaningless, off-topic, and repetitive to communication partners (Alzheimer’s Association, 2021). In severe dementia, communication may become unintelligible with inability to process and understand language (Savundranayagam & Orange, 2014). Caregivers instinctively alter their communication approaches with the person with dementia in response to indicators of impending communication breakdown but lack knowledge about effective strategies for resolving communication challenges (Savundranayagam & Orange, 2014).

Available communication training focuses primarily on professional caregivers. The few evidence-based programs were developed from analysis of institutional and staged interactions and provide only general advice. In contrast, this study will identify effective family caregiver communication strategies through analysis of natural interactions collected in the Supporting Family Caregivers (FamTechCare) study (R01NR014737). FamTechCare caregivers (n = 88) used an in-home recording system to capture and submit more than 1,000 videos of home care interactions to receive expert feedback to improve care (Kim et al., 2019; Williams et al., 2018, 2019). Findings from the current study will be used to develop evidence-based interventions to teach caregivers evidence-based, effective communication strategies to augment a person-centered and individualized approach. Improving communication has a high potential to help caregivers provide care, and maintain relationships and quality of life for the person with dementia. Improving in-home care is also crucial for reducing negative outcomes for caregivers such as increased emotional stress and depression and the loss of intimacy and relationship (Alzheimer’s Association, 2021). Communication can play a role in managing or improving the outcome for both the caregiver and person with dementia by improving cooperation and strengthening bonds during in-home care (National Institute on Aging, 2017).

Background and Significance

Dementia involves a progressive decline in several cognitive abilities that affect communication, including processing speed, memory, language, and executive functioning (Alzheimer’s Association, 2021). Although there are specific differences among the types of dementia, problems with communication arise across all dementia types. Noticeable changes in speech fluency and conversational communication occur early in the diagnosis, and as dementia progresses, word finding, writing, and understanding complex language such as analogies and sarcasm become more problematic (Haak, 2002; Savundranayagam & Orange, 2014). In dementia’s middle stages, spoken language has reduced meaningful content, words may be inappropriately substituted, written information becomes difficult to understand, and multistep commands are difficult to follow or conceptualize (Haak, 2002; Savundranayagam & Orange, 2014). People with dementia also have difficulty expressing unmet needs, staying on-topic, and handling multiple partner conversations. In later stages, the individual may produce streams of nonsensical language or become mute (Alzheimer’s Association, 2021; Savundranayagam and Orange, 2014).

Caregiver Communication Education and Current Evidence

Educating professional caregivers in communication strategies can significantly improve caregiver knowledge, increase the use of effective verbal and nonverbal communication strategies, and reduce BPSD as well as need for psychotropic medication (Eggenberger et al., 2013; Machiels et al., 2017; Morris et al., 2018; Nguyen et al., 2019; Shaw et al., 2018; Tjia et al., 2017; Williams et al., 2016, 2017). Reports of communication interventions specifically targeting family caregivers have increased in recent years. Perkins et al. (2022) identified nine communication interventions targeting family caregivers in their systematic review. All the studies demonstrated some benefits to the family caregiver and the person with dementia; however, three were rated the highest in quality. Barnes and Markham (2018) conducted an RCT with 55 caregivers providing three 60-min face-to-face sessions focusing on knowledge of dementia and communication difficulties, insight into communication difficulties, thoughts and feelings, environment, personhood, how to be the caregiver, reminders and encouraging conversation, communication and activities, and challenging behaviors. The treatment group had significant increases in quality of life, communication self-efficacy, and caregivers reported fewer care challenges related to the person with dementia. Klodnicka-Kouri (2011) conducted an RCT with 50 caregivers providing five 90-min face-to-face sessions focusing on effective communication models and skills, self-efficacy, and diverse approaches to reduce the caregiver’s anxiety. Significant increases in communication knowledge, skills, self-efficacy, and significant decreases in perceived communication-related behavioral symptoms occurred in the treatment group compared to control. Finally, Liddle et al. (2012) conducted an RCT with 29 caregivers providing two 45-min video training. Caregivers were taught “RECAPS” in the first video which stands for (R)eminders, (E)nvironment, (C)onsistent routines, (A)ttention, (P)ractice, (S)imple steps, whereas the second video taught “MESSAGE” or (M)aximize attention, (E)xpression and body language, keep it (S)imple, (S)upport conversations, (A)ssist with visual aids, (G)et their message, (E)ncourage and engage in conversation. Significant improvement in knowledge of strategies, near-significant increases in positive aspects of caring, and decreases in perceived communication-related behavioral symptoms were found after the training. Overall, communication interventions for both professional and family caregivers show promise for improving dementia care and BPSD management. However, training content varies widely and is not always empirically based, impeding the ability of researchers and practitioners to know which behaviors and strategies to target.

Although experimental studies evaluate the effectiveness of interventions to improve caregiver communication, descriptive studies assess and explain specific caregiver communication strategies and approaches. Studies solely focused on family caregiving with sample sizes of at least 15 caregivers currently recommend a mix of practical and interpersonal strategies including eye contact, smiling, short sentences, slow approach, verbal and physical reassurance, respectfulness, and acknowledgment of personhood (Alsawy et al., 2017). Small and Gutman’s research indicates that communication strategies such as asking one question at a time, using short sentences, speaking slowly, using yes/no questions, paraphrasing, and repeating, are frequently used if caregivers perceived them as effective (Small & Gutman, 2002; Small et al., 2003). Studies by Savundranayagam and Orange (2011, 2014) mirror these findings, reporting that communication strategies rated as effective by caregivers reduce caregiver stress, relationship burden, and communication breakdown. Specific communication strategies that caregivers rated as effective were repeating, trying to figure out the meaning, redirecting, giving more information, asking questions, simplifying, speaking slowly, and rephrasing. Communication strategies used in developing the Verbal–Nonverbal Interaction Scale for Caregivers (VNVIS-CG) added nonfacilitative and nonverbal behaviors (Williams & Parker, 2012). Although these studies rate effectiveness based on caregiver self-report, empirical knowledge based on predictive modeling and larger samples is needed to determine which strategies work best with individuals, within familial relationships, and in specific settings such as the home.

Frameworks for Dementia Communication, Breakdown, and Trouble Source Repair

Communication breakdown contributes to BPSD and presents challenges to providing care, increasing caregiver stress, and threatening caregiver person with dementia relationships. A number of theoretical frameworks conceptually targeting person centeredness provide both professional and family caregivers with guidance in overcoming communication challenges in dementia care (O’Rourke et al., 2022). Communication Accommodation Theory (Coupland et al., 1988) frames communication as an interactive interchange between two communication partners that consider social and personal contexts, integrating both linguistic and relational and personal elements (McGilton et al., 2009).

As explained in the Communication Enhancement Model (Ryan et al., 1995), effective communicators make an individualized assessment of communication abilities and then make modifications to their own communication only in response to needs of their communication partner. As part of this feedback loop, family caregivers may rely on longstanding relationships with and knowledge of the person living with dementia. However, caregivers may need assistance with communication challenges indicating a need to identify additional evidence-based communication strategies that are commonly effective. Caregivers can use strategies to prevent breakdown and repair communication to further adapt and improve communication.

The trouble source repair (TSR) framework (Orange et al., 1996, 1998) provides a theoretical framework for examining communication breakdown in dementia care and was used to guide this study. The TSR framework provides a systematic evaluation of what occurs during a communication sequence including the occurrence of and reason for a breakdown, what actions are taken to overcome the breakdown, and the outcome in terms of successful communication (see Figure 1). TSR was first used by Orange et al. to analyze transcribed conversations of caregiver person with dementia dyads and to determine how dyads corrected and repaired misunderstandings or mishearings. Orange found that over a third of the utterances by people with dementia in moderate-to-severe stages of dementia were TSR-related communication breakdowns, almost twice as many for normal controls. Within dyads, the person with dementia produced more total TSR-related utterances than their unimpaired conversational partner (Orange et al., 1996, 1998). Most repairs were successful, and both the person with dementia and their partner relied on self-repair, nonspecific terms (e.g., What? or Huh?), and understanding checks (e.g., What do you mean?). However, over time, dyads including a person with dementia were less successful in resolving communication breakdowns compared to cognitively intact dyads (Orange et al., 1996, 1998).

Figure 1.

Figure 1.

Open in a new tab

Illustration of the trouble source repair (TSR) framework. Notes: As the caregiver person with dementia dyads enter communication breakdown, the trouble source initiates the misunderstanding, the flag indicates the communication partner does not understand, and repair strategies are used to resolve the breakdown. Most successful repairs take only one repair strategy, successful repairs take more than one repair strategy, and unsuccessfully repaired breakdowns are abandoned or lead to additional breakdown.

The evaluation of effective communication strategies for dementia care within the framework of communication breakdown and repair is relatively young but provides a new perspective. An early study found fewer breakdowns were observed when specific strategies were used compared to when they were not, whereas a more recent study using TSR identified effective strategies for successful resolution of breakdown and determined that the use of these strategies resulted in reduced stress and relationship burden (Savundranayagam & Orange, 2011). Caregivers’ ratings of strategy effectiveness agreed with TSR evaluation 45%–68% of the time. However, caregivers incorrectly evaluated effectiveness 32%–55% of the time. This mismatch indicates caregivers lack knowledge of what works best. Most mismatched appraisals were for effective communication strategies caregivers rated as ineffective (Savundranayagam & Orange, 2014).

Although these studies only assessed a small number of communication strategies with a small sample, the current study evaluated a comprehensive battery of verbal and nonverbal caregiver communication strategies (56 variables) in a large sample of naturally occurring, in-home recorded interactions (221 video observations) to determine which strategies prevent communication breakdown and promote successful repair. This evidence can add to the evidence base for teaching family caregivers’ communication strategies that work best.

Study Purpose

The purpose of this study was to explore effective communication strategies for family caregivers of people with dementia using predictive modeling with a larger sample than the previous studies. The aims included characterizing communication behaviors used by caregivers in daily care in relation to breakdown and repair through sequential behavioral coding analysis of video-recorded caregiver person with dementia interactions and identifying effective strategies that prevent and repair communication breakdown. A further aim was to evaluate the effectiveness of caregiver communication strategies that prevent or repair communication breakdown in relation to different types and stages of dementia and other dyad characteristics such as relationship, age, and gender.

Research Design and Methods

A sequential behavioral analysis was completed on archived video recordings of daily care recorded during the FamTechCare clinical trial (Coleman et al., 2022; Williams et al.. 2019). A behavioral coding scheme was developed to identify communication strategies used by caregivers and persons with dementia, as well as communication breakdown and types of repairs (Most successful, successful, or unsuccessful). Penalized regression, a feature selection method called Least Absolute Shrinkage and Selection Operator (LASSO), was first used to explore which specific strategies were the most important predictors of breakdown/repair (Tibshirani, 1996). Based on the best features selected by LASSO, Bayesian mixed models were used to identify significant predictors of occurrences of breakdowns and of repair outcomes. Finally, demographic data were added to the models as covariates, and subgroup analyses were used to determine whether dyad characteristics predicted strategies facilitating or impeding successful communication.

Sample

Videos archived from the FamTechCare clinical trial parent study provided the sample for this study. The FamTechCare study collected in-home video recordings from caregiver person with dementia dyads (N = 88) of daily care interactions that were reviewed by an expert panel who gave feedback to improve care. Enrolled caregivers provided home care at least weekly to the person with dementia who was diagnosed with any type of dementia. Exclusion criteria were diagnoses of Huntington’s disease, alcohol-related dementia, schizophrenia, bipolar disorder, deafness, or developmental delay. Caregivers captured recordings of spontaneous, unscripted interactions using an application (app) on an iPad Mini set to record video when triggered by the caregiver. Additional protocol details are provided elsewhere (Williams, 2006). The parent study collected 1,127 in-home videos from 88 dyads totaling 7,514.5 min and averaging 6.7 min per video.

A subsample of 221 videos from FamTechCare was selected for this study. The subsample included videos for 53 dyads who had at least three videos (allowing analysis of within-dyad differences) of adequate quality to observe verbal and nonverbal communication lasting at least 30 s (an adequate amount for communication analysis; The Cambridge Handbook of Group Interaction Analysis, 2018; Brauner, 2018). The 221 videos averaged 5.8 min in length (range 49.61 s–13.02 min). The videos were segmented into 30-s intervals providing 3,642 segments for analyses. This duration was selected to support focused evaluation of complete, discreet communication breakdown and repair sequences including antecedent communication behaviors.

Procedure

Behavioral coding

We developed a coding scheme based on our theoretical model to include (a) verbal caregiver communication behaviors; (b) nonverbal caregiver communication behaviors; and (c) communication breakdown and repair sequences. The coding scheme was developed using an iterative process and the videos were coded using Noldus Observer XT 15.0 software during 2020 and 2021. The Observer program provides a template for a behavioral coding scheme in which a computer key corresponding to each behavior is depressed when behaviors occur in the real-time video recordings. Multiple coders can access the data set to code different variables, and interrater reliability can be calculated among multiple coders. The duration of each behavior state is tabulated by the length of time a key remains depressed until the key corresponding to the alternate, mutually exclusive behavior state is pressed. Within each state, specific events were also coded. Verbal and nonverbal behavior codes were adapted from the VNVIS-CG for frame-by-frame behavioral coding (Coleman & Williams, 2018; Coleman et al., 2022; Williams & Parker, 2012). This scale was selected because it includes a comprehensive list of verbal and nonverbal communication behaviors, had adequate test–retest scores ranging from 0.70 to 0.91 depending on the behavioral scale, and established validity (Williams & Parker, 2012). Dyadic verbal communication is coded as mutually exclusive states of silence, person with dementia verbalizations, and caregiver verbalizations acknowledging the interaction inherent in communication. Caregiver verbalizations are further categorized as (a) affirmations (supportive statements); (b) repetitions/paraphrasing; (c) acknowledging emotion (statements to the person with dementia’s emotional expression); (d) verbalizing understanding (the caregiver confirms they understand the person with dementia); (e) providing direction or information; and (f) asking questions. An additional modifier is coded for questions to specify type: (a) clarification; (b) request for assistance; (c) offer’s choices; (d) asks permission; (e) shows interest (asks the person with dementia’s opinion or perspective); (f) encourages emotional expression (how the person with dementia is feeling); (g) quizzing (memory check); or (h) tag question.

Caregiver nonverbal communication was coded for the emotional tone or affective quality of the caregiver’s communication throughout the interactions as well as other observable nonverbal behaviors used by the caregiver. Emotional tone is categorized as (a) respectful (indicating equality of the communication partner); (b) controlling (dominating); or (c) overly nurturing (patronizing and infantilizing). We used operational definitions for emotional tone that were developed and validated in our prior research to evaluate the qualitative affective qualities of communication (Williams et al., 2012). Additional nonverbal behaviors include (a) laughter/joy (delight); (b) compassion (affectionate touching); (c) assistance (environmental changes to help the person with dementia); (d) gestures/positive postures (body orientation, eye contact, face-to-face, instructive gestures); (e) aggression (disrespectful facial expressions, touching, or yelling); (f) rejecting (no reaction, walking or turning away); and (g) negative postures (back turned to the person with dementia or being too far away). The coding scheme thus included a variety of verbal and nonverbal, quantitative and qualitative, and interpersonal and linguistic variables. After coding was completed, event data with time notations were for analyses. Details about the coding scheme, development, and operational definitions for coded variables are provided elsewhere (Coleman et al., 2022).

The coding scheme for communication breakdown was developed using the TSR framework. Communication in the video footage was coded as one of four states: silence; talking to self or someone other than the person with dementia; interactive (the caregiver and the person with dementia understanding one another); or breakdown (misunderstanding). Once a breakdown was identified, the trouble source (cause of misunderstanding), the flag (indicator of misunderstanding), and the repair types were identified. Once the breakdown was resolved (communication continues with evidence of understanding), the breakdown repair sequence was coded as Most successful, successful, or unsuccessful repair based on the number of repair attempts or the lack of resolution. A most successful repair only required one repair strategy to resolve.

Coders were trained on practice materials until inter-rater reliability was achieved. Coding reliability for all three coding schemes was excellent; communication breakdown and repair behaviors (Kappa = 0.85); verbal caregiver communication (Kappa = 0.85); and nonverbal communication behaviors (Kappa = 0.86). Continued evaluation of interrater reliability with a standard to exceed 85% agreement continued during coding of study videos on a biweekly basis and included 10% of the sample.

Analyses

The coded data included 3,642 30-s observations. Descriptive statistics were used to describe the type, frequency, and duration of communication strategies in the interactions between and within dyads (Coleman et al., 2022).

Our primary analyses evaluated different communication strategies used during each 30-s video clip as predictors of communication breakdown and repair. Considering the large number of communication behaviors, breakdowns, and repairs (56 variables), we first used the feature selection method LASSO to explore which specific caregiver behaviors were the most important predictors of breakdown/repair. LASSO is a machine learning method that allows feature selection from large number of predictors in the model by shrinking the sum of the absolute values of the coefficients and forcing some coefficients to 0; thus, the nonzero variables were selected by the model (Tibshirani, 1996). We used the glmnet package in R to select the best tuning parameter lambda via the cross-validation (cv.glmnet) function and performed the logistic LASSO regression. Breakdown (coded as 0), successful repair (coded as 1), and most successful repair (coded as 1 vs successful repair) were included in the logistic LASSO as the dependent variables. We entered all 56 variables as the predictors; all 56 predictor variables were included as binary variables (see Supplementary Material). Dyadic structure exists in our data set as patients and caregivers are interdependent within dyads.

Based on the LASSO findings and eliminating coding variables that were not behaviors caregivers could modify in dyadic interactions, we conducted a Bayesian mixed-effects model analysis with the dyad as the random effect to examine the relationship between caregiver communication behaviors, breakdown, and repair. We completed subgroup analyses to explore how behaviors differed according to relationship type, dyad gender combinations (male–male, female–male, male–female, and female–female), and dementia type and severity. Dementia type, severity, and relationship type were also entered as covariates in the three models to examine differences in model strength. Only data segments including a breakdown were analyzed in the models evaluating the repair outcomes (N = 820 for successful compared to not successful repair; and N = 650 for most successful compared to successful repair). We used weakly informative priors in the Bayesian mixed-effects models (Gelman et al., 2008).

Results

Demographic data for the 53 dyads are included in Table 1. A total of 221 videos were included. Videos averaged 7.69 min in length. Caregivers were primarily female (83%) with a mean age of 65.2 (standard deviation [SD] = 12.9) years, and most were spouses of the person with dementia (70%). The person with dementia was primarily male (55%) with a mean age of 76.2 (SD = 8.9), with moderately severe (64%) Alzheimer’s dementia (55%), and with an average of 4.4 (SD = 3.2) years since diagnosis.

Table 1.

Caregiver Person With Dementia Dyad Demographics (N = 53 Dyads)

Variable Caregivers Person with dementia
Mean (SD) Range n %a Mean (SD) Range n %a
Age (years) 65.2 (12.9) 32.0–90.0 53 76.2 (8.9) 54.0–93.0 53
Years since dementia diagnosis 4.4 (3.2) 0.0–15.0 53
Gender
 Female 44 83.0 24 45.3
 Male 9 17.0 29 54.7
Race
 White 49 92.5 50 94.3
 African American 4 7.5 3 5.7
Ethnicity
 Not Hispanic/Latino 50 94.3 51 96.2
 Unknown/not reported 3 5.7 2 3.8
Marital status
 Married 46 86.8 43 81.1
 Single/widowed/divorced 7 13.2 10 18.9
Education level
 Less than bachelor’s degree 22 41.5 30 56.6
 Bachelor’s degree 20 37.7 10 18.9
 Master’s degree or higher 11 20.8 13 24.5
Relationship to the person with dementia
 Spouse 37 69.8
 Child/spouse of child 13 24.5
 Otherb 3 5.7
Primary dementia diagnosis
 Alzheimer’s disease 29 54.7
 Vascular 5 9.4
 Lewy bodies 7 13.2
 Frontotemporal 3 5.7
 Parkinson’s related 1 1.9
 Unknown dementia type 8 15.1
Functional assessment staging (FAST) disability categoryc
 Mild dementia 12 22.6
 Moderate dementia 4 7.5
 Moderately severe dementia 33 64.2
 Severe dementia 3 5.7
Open in a new tab

Note: SD = standard deviation.

aPercentages may total more than 100% due to rounding.

bOther relationships with the person with dementia were friend or paid caregiver.

The accuracy scores for each of the LASSO models were no breakdown (0.97), successfully repaired (0.97), and most successfully repaired (0.83), indicating the model performed well in correctly classifying the outcome variables. A final set of variables for each analysis (no breakdown, repair, and most successfully repaired) was identified for inclusion in the models based on the logistic LASSO results (see Supplementary Material). Variables were chosen based on both positive and negative magnitudes and relevance as modifiable behaviors for dyadic interaction. The model for predicting the occurrence of breakdown versus no breakdown included 10 variables, the successful repair–no repair model included 11, and the model evaluating successful compared to most successful repairs included nine variables used in the final models.

Given our data, communication breakdown (coded as 0) was associated with caregivers changing topic (median = −11.45, 95% credibility interval [CrI; −24.34, −4.37]), ignoring (median = −11.49, 95% CrI [−24.49, −4.72]), giving commands (median = −10.74, 95% CrI [−24.22, −3.38]), and taking over the task (median = −4.06, 95% CrI [−7.28, −1.77]). Successful repair of breakdown (coded as 1) was associated with caregivers verbalizing understanding (median = 0.46, 95% CrI [0.09, 0.86]), tag questions, (median = 2.4, 95% CrI [0.33, 5.35]), and silence (median = 0.78, 95% CrI [0.42, 1.15]), and was negatively associated with ignoring and changing topic (median = −3.63, 95% CrI [−4.81, −2.57] and −2.51 [−3.78, −1.33]). Most successful repair (only one repair attempt, coded as 1) was associated with asking permission (median = 2.62, 95% CrI [0.04, 5.23]). Most successful repair was negatively associated with bossy tone (median = −1.27, CrI [−2.13, −0.48]) and asking questions (median = −2.49, CrI [−3.01, −2.02]) compared to successful repair (requiring multiple repair attempts). Results can be seen in Tables 2–4.

Table 2.

Bayesian Mixed-Effects Model Results for Caregiver Behaviors Predicting No Communication Breakdown

Model coefficients Posterior median (log) (standard error)a 95% Credibility interval (log)
Lower Upper
Intercept 1.43 (0.12) 1.21 1.66
Repair_Changing_topic −11.45 (5.39) −24.34 −4.37*
Repair_Ignore −11.49 (5.25) −24.49 −4.72*
Overly_nurturing_tone −0.23 (0.19) −0.60 0.15
Repair_Command −10.74 (5.58) −24.22 −3.38*
Use_partner_s_name −0.06 (0.22) −0.48 0.39
Asks_questions_Asks_permission 0.44 (0.35) −0.21 1.19
Repair_Taking_over_task −4.06 (1.41) −7.28 −1.77*
Laughter_Joy −0.06 (0.15) −0.36 0.25
Asks_questions_Encourages_emotional_expression −0.17 (0.34) −0.82 0.50
Acknowledges_emotion −0.16 (0.34) −0.81 0.53
Open in a new tab

Note: Zero does not occur between lower and upper intervals.

*Significant predictor.

aNo breakdown = 1, breakdown = 0.

Table 4.

Bayesian Mixed-Effects Model Results for Caregiver Behaviors Predicting Most Successfully Repaired Communication Breakdowns

Model coefficients Posterior median (log) (standard error) 95% Credibility interval (log)
Lower Upper
Intercept 1.28 (0.47) 0.35 2.17
Repair_Increased_volume 0.36 (0.29) −0.20 0.96
Asks_questions_Asks_permission 2.62 (1.35) 0.04 5.23*
Asks_questions_Tag_Question 0.77 (0.45) −0.05 1.67
Respectful_tone −0.15 (0.43) −0.97 0.67
Repair_Repeating 0.10 (0.19) −0.26 0.46
Bossy_tone −1.27 (0.42) −2.13 −0.48*
Overly_nurturing_tone −0.97 (0.54) −2.03 0.06
Repair_Questions −2.49 (0.25) −3.01 −2.02*
Verbalizes_understanding_Humor_Joking −8.12 (5.92) −22.30 0.41
Open in a new tab

Note: Zero does not occur between lower and upper interval.

*Significant predictor.

Table 3.

Bayesian Mixed-Effects Model Results for Caregiver Behaviors Predicting Successfully Repaired Communication Breakdowns

Model coefficients Posterior median (log) (standard error) 95% Credibility interval (log)
Lower Upper
Intercept 1.30 (0.59) 0.17 2.45
Repair_Ignore −3.63 (0.58) −4.81 −2.57*
Repair_Changing_topic −2.51 (0.63) −3.78 −1.33*
Aggression −0.54 (0.70) −1.94 0.80
Negative_posture −1.03 (0.70) −2.40 0.32
Repair_Command −0.47 (0.46) −1.46 0.34
Overly_nurturing_tone 0.41 (0.68) −0.84 1.77
Repair_Increased_volume 0.67 (0.49) −0.21 1.67
Verbalizes_understanding_Neutral 0.46 (0.20) 0.09 0.86*
Asks_questions_Tag_Question 2.40 (1.30) 0.33 5.35*
Silence_Vebal_Code 0.78 (0.19) 0.42 1.15*
Respectful_tone 0.69 (0.47) −0.22 1.65
Open in a new tab

Note: Zero does not occur between lower and upper interval.

*Significant predictor.

Adding dementia type, severity, and relationship type as covariates did not change the significant predictors in the models, except for dementia type in the successful repair versus no repair model. When Alzheimer’s disease was the dementia diagnosis, successful (vs unsuccessful) repair was negatively associated with caregivers ignoring (median = −3.16, 95% CrI [−4.47, −2.00]), changing topic (median = −2.33, 95% CrI [−4.02, −0.86]), giving commands (median = −2.35, 95% CrI [−5.13, −0.29]), and positively associated with verbalizing understanding (median = 0.51, 95% CrI [0.11, 0.93]), and silence (median = 1.02, 95% CrI [0.55, 1.49]). When “Other” was the dementia diagnosis (Lewy body, frontotemporal, Parkinson’s related, or unknown dementia type), successful repair was negatively associated with caregivers ignoring behavior (median = −8.16, 95% CrI [−15.15, −4.12]) and changing topic (median = −3.61, 95% CrI [−6.14, −1.39]). No significant predictors were identified when the diagnosis was vascular dementia (see Supplementary Material).

Discussion

This research used the TSR conceptual framework of communication breakdown and a sequential behavioral approach to identify the best practices in communication for dementia caregivers. In addition, LASSO analyses provided a new approach for selecting the strongest potential predictors from a complex set of communication behavior variables. Bayesian mixed-effects modeling considers the dyadic structure of the data, provides interpretations of the coefficients, and helps convergence for mixed-effects models. Both advanced statistical approaches are relatively recent, and future research should continue to use new approaches and develop additional methods to analyze increasingly complex data sets.

Results of our secondary analysis identified caregiver communication behaviors that are predictors of communication breakdown, and successful and most successful breakdown repair. These predictors provide a set of communication behaviors family caregivers can be taught to use or avoid in order to overcome dementia care communication challenges. Based on these findings, caregivers can be encouraged to avoid specific communication strategies including changing topic, ignoring, giving commands, and taking over a task; all these were associated with a breakdown in communication. In the case of an impending or actual communication breakdown, caregivers can try strategies likely to promote positive resolution including verbalizing understanding, using tag questions, and silence. Verbalizing understanding (confirming the person with dementia’s statement or answering the person with dementia’s question) allows communication to continue or resolves the breakdown. Tag questions provide a suggested response to the question within the question. For example, “You are ready for dinner, aren’t you?” suggests the care recipient respond affirmatively. Silence may be beneficial by allowing the care recipient time to process communication and formulate their response. Also, asking for permission (asking the person with dementia if you can assist or touch them; allowing mutual control in the interaction), avoiding bossy or controlling tone (voice quality is loud, harsh, antagonistic, argumentative, or dominating; the tone indicates the caregiver is an authority figure), and refraining from continued repeated questioning after a breakdown provides the fastest resolution of breakdown.

We hypothesized differences in the type and severity of dementia would be important covariate factors. Relationship and cohort differences (age and gender combinations) may also factor into the personal–social and cultural dimensions of communication. This study identified only a few of these relationships. Further analyses of these factors are warranted to provide evidence for tailoring training for dyads of different types and stages of dementia and for additional characteristics (Resnick et al., 2022).

Limitations

Our goal was to evaluate interactions on the micro level within the TSR framework. We set the interval of analysis at 30 s after the review of multiple video recordings examining different interval lengths with a goal of determining the interval of adequate duration to capture antecedent and consequent behavioral events related to breakdown and repair. We selected the 30-s interval as containing antecedents as well as consequent information limited to individual breakdown/repair sequences. For example, we wanted intervals that included limited breakdowns and repairs to best examine specific behaviors to identify relationships. Using a different interval for analysis may have resulted in different findings, and future research is needed to further evaluate and validate the use of a 30-s window.

Other cautions in interpreting the results of this study relate to (a) the use of secondary data of dyadic interactions that were not entirely natural representations of communication behavior (due to awareness of recording); (b) the potential influence of the design of the parent study in which videos were submitted to obtain feedback from experts on how to improve care; and (c) lack of racial and ethnic diversity in our sample (mostly White and non-Hispanic), inclusion of primarily spousal caregivers, and limited variation in type and severity of dementia and other dyad characteristics. These factors may limit generalizability of our findings and should be addressed in future research. Finally, individual and dyadic differences, validity of adapted coding schemes, and the sequential nature of behaviors within interactions will be addressed in future analyses.

Implications

Caregivers should be advised to use communication strategies that have been successful to meet individualized needs of the person with dementia. Our research identified additional strategies that caregivers may find beneficial during challenging communication. Based on our analyses, future family caregiver education will focus on assisting caregivers to prevent communication breakdowns during their interactions with their partner with dementia by avoiding commands, ignoring some verbalizations, changing topic, or completing a task themselves when necessary. Caregivers can also be advised to answer the person with dementia’s questions or convey their statement was heard when breakdowns occur (verbalize understanding), to step back and ask permission before continuing, and to provide enough time for the person with dementia to process the communication and provide a response. Caregivers should also avoid ignoring and changing the topic when a breakdown occurs because it could lead to further misunderstandings. These strategies, rather than previous strategies seen in the family caregiving literature, such as simplifying, instructing, repeating, or redirecting, lend themselves to more person-centered communication approach (Kitwood & Kitwood, 1997). These strategies focus more on the caregivers’ efforts to understand and respect the person with dementia as having autonomy and an equal voice in the activities of daily living rather than being a “task” to be accomplished. The instruction given to family members should focus less on communication to accomplish “care giving,” but on communication to accomplish relationship continuity and to be a “care partner” in the mutual, daily struggle to adapt to changes related to dementia (Coleman et al., 2013; Lann-Wolcott et al., 2011; Medvene & Coleman, 2012; O’Rourke et al., 2022).

Based on our secondary analysis of over 200 videos from the FamTechCare clinical trial, we have identified a key set of evidence-based communication strategies to improve communication for families caring for a person with dementia. Assisting families to overcome the communication challenges of dementia is important for achieving activities of daily living goals as well as for maintaining social connections and relationships in care partner person with dementia dyads. Future family education will focus on these communication strategies by providing limited content with illustrative examples supporting ease of application and reducing the recall requirements during interactions. Future research will expand this research to more diverse populations to determine if communication strategies differ based on preferences, culture, or context. Testing family communication training based on these results should evaluate the implementation of these communication best practices and confirm whether the use of these strategies relates to communication breakdown and repair comes.

Supplementary Material

gnac193_suppl_Supplementary_Material
Click here for additional data file. (32.1KB, docx)

Acknowledgments

We wish to thank Iman Aly, Ashley Dunham, Kacie Inderhees, Michaela Richardson, Paige Wilson, Amy Berkley, and Marie Savundranayagam for their valuable input into the project.

Contributor Information

Kristine N Williams, School of Nursing, University of Kansas Medical Center, Kansas City, Kansas, USA; Alzheimer’s Disease Research Center, University of Kansas, Fairway, Kansas, USA.

Carissa K Coleman, School of Nursing, University of Kansas Medical Center, Kansas City, Kansas, USA.

Jinxiang Hu, School of Medicine, University of Kansas Medical Center, Kansas City, Kansas, USA.

Funding

Research reported in this publication was supported by the National Institute on Aging of the National Institutes of Health (5R21AG066491-02). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Conflict of Interest

We have no conflicts of interest to declare.

Data Availability

The data, analytic methods or materials are available to other researchers for replication purposes by contacting the authors. This study was not preregistered.

References

  1. Alsawy, S., Mansell, W., McEvoy, P., & Tai, S. (2017). What is good communication for people living with dementia? A mixed-methods systematic review. International Psychogeriatrics, 29(11), 1785–1800. doi: 10.1017/s1041610217001429 [DOI] [PubMed] [Google Scholar]
  2. Alzheimer’s Association. (2021, October 25). Alzheimer’s disease facts and figures report. https://www.alz.org/alzheimers-dementia/facts-figures
  3. Barnes, C. J., & Markham, C. (2018). A pilot study to evaluate the effectiveness of an individualized and cognitive behavioural communication intervention for informal carers of people with dementia: The talking sense programme. International Journal of Language and Communication Disorders, 53(3), 615–627. doi: 10.1111/1460-6984.12375 [DOI] [PubMed] [Google Scholar]
  4. Brauner, E. (2018). Coding interaction: A technical introduction. In Brauner E., Boos M., & Kolbe M. (Eds.), The Cambridge handbook of group interaction analysis (pp. 165–190). Cambridge University Press. doi: 10.1017/9781316286302.010 [DOI] [Google Scholar]
  5. Brauner, E., Boos, M., & Kolbe, M. (Eds.). (2018). The Cambridge handbook of group interaction analysis. Cambridge University Press. doi: 10.1017/9781316286302 [DOI] [Google Scholar]
  6. Coleman, C. K., Aly, I. M., Dunham, A., Inderhees, K., Richardson, M., Wilson, P., Berkley, A., Savundranayagam, M., & Williams, K. (2022). Developing behavioral coding to understand family communication breakdown in dementia care. Western Journal of Nursing Research. 44(3), 250–259. doi: 10.1177/01939459211062957 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Coleman, C. K., Medvene L. J., & Van Haitsma, K. (2013). A person-centered care intervention for geriatric certified nursing assistants. Gerontologist, 53(4), 687–698. doi: 10.1093/geront/gns135 [DOI] [PubMed] [Google Scholar]
  8. Coleman, C. K., & Williams, K. (2018). Visualizing communication behaviors and breakdowns in caregiver and person with dementia dyads. Innovation in Aging, 2(Suppl. 1), 1010. doi: 10.1093/geroni/igy031.3730 [DOI] [Google Scholar]
  9. Coupland, N., Coupland, J., Giles, H., & Henwood, K. (1988). Accommodating the elderly: Invoking and extending a theory. Language in Society, 17(1), 1–41. doi: 10.1017/s0047404500012574 [DOI] [Google Scholar]
  10. Eggenberger, E., Heimerl, K., & Bennett, M. I. (2013). Communication skills training in dementia care: A systematic review of effectiveness, training content, and didactic methods in different care settings. International Psychogeriatrics, 25(3), 345–358. doi: 10.1017/S1041610212001664 [DOI] [PubMed] [Google Scholar]
  11. Gelman, A., Jakulin, A., Pittau, M. G., & Su, Y. S. (2008). A weakly informative default prior distribution for logistic and other regression models. Annals of Applied Statistics, 2(4), 1360–1383. doi: 10.48550/arXiv.0901.4011 [DOI] [Google Scholar]
  12. Haak, N. J. (2002). Maintaining connections: Understanding communication from the perspective of persons with dementia. Alzheimer’s Care Quarterly, 3, 116–131. [Google Scholar]
  13. Kim, S., Shaw, C., Williams, K. N., & Hein, M. (2019). Typology of technology-supported dementia care interventions from an in-home telehealth trial. Western Journal of Nursing Research, 41(12), 1724–1746. doi: 10.1177/0193945919825861 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Kitwood, T. M., & Kitwood, T. M. (1997). Dementia reconsidered: The person comes first (Vol. 20, pp. 7–8). Open university press. [Google Scholar]
  15. Klodnicka Kouri, K., Ducharme, F. C., & Giroux, F. (2011). A psychoeducational intervention focused on communication for caregivers of a family member in the early stage of Alzheimer’s disease: Results of an experimental study. Dementia, 10(3), 435–453. doi: 10.1177/1471301211408124 [DOI] [Google Scholar]
  16. Lann-Wolcott, H., Medvene, L. J., & Williams, K. (2011). Measuring the person-centeredness of caregivers working with nursing home residents with dementia. Behavior Therapy, 42, 89–99. doi: 10.1016/j.beth.2010.02.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Liddle, J., Smith-Conway, E. R., Baker, R., Angwin, A. J., Gallois, C., Copland, D. A., Pachana, N. A., Humphreys, M. S., Byrne, G. J., & Chenery, H. J. (2012). Memory and communication support strategies in dementia: Effect of a training program for informal caregivers. International Psychogeriatrics, 24(9), 1927–1942. doi: 10.1017/s1041610212001366 [DOI] [PubMed] [Google Scholar]
  18. Machiels, M., Metzelthin, S. F., Hamers, J. P., & Zwakhalen, S. M. (2017). Interventions to improve communication between people with dementia and nursing staff during daily nursing care: A systematic review. International Journal of Nursing Studies, 66, 37–46. doi: 10.1016/j.ijnurstu.2016.11.017 [DOI] [PubMed] [Google Scholar]
  19. McGilton, K. S., Boscart, V., Fox, M., Sidani, S., Rochon, E., & Sorin-Peters, R. (2009). A systematic review of the effectiveness of communication interventions for health care providers caring for patients in residential care settings. Worldviews on Evidence-Based Nursing, 6(3), 149–159. doi: 10.1111/j.1741-6787.2009.00155.x [DOI] [PubMed] [Google Scholar]
  20. Medvene, L. J., & Coleman, C. K. (2012). Exploring the applicability of interdependence theory to CNA-resident relationships. Research in Gerontololgical Nursing, 5(1), 43–54. doi: 10.3928/19404921-20110329-01 [DOI] [PubMed] [Google Scholar]
  21. Monin, J. K., & Schulz, R. (2009). Interpersonal effects of suffering in older adult caregiving relationships. Psychology and Aging, 24(3), 681–695. doi: 10.1037/a0016355 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Morris, L., Horne, M., McEvoy, P., & Williamson, T. (2018). Communication training interventions for family and professional carers of people living with dementia: A systematic review of effectiveness, acceptability, and conceptual basis. Aging Mental Health, 22(7), 863–880. doi: 10.1080/13607863.2017.1399343 [DOI] [PubMed] [Google Scholar]
  23. National Institute on Aging. (2017, May 17). Communication and behavior problems: Resources for Alzheimer’s caregivers. https://www.nia.nih.gov/health/communication-and-behavior-problems-resources-alzheimers-caregivers
  24. Nguyen, H., Terry, D., Phan, H., Vickers, J., & McInerney, F. (2019). Communication training and its effects on carer and care-receiver outcomes in dementia settings: A systematic review. Journal of Clinical Nursing, 28(7–8), 1050–1069. doi: 10.1111/jocn.14697 [DOI] [PubMed] [Google Scholar]
  25. O’Rourke, D. J., Lobchuk M. M., Thompson, G. N., & Lengyel, C. (2022). Expanding the conversation: A person-centered Communication Enhancement Model. Dementia, 21(5), 1596–1617. doi: 10.1177/14713012221080252 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Orange, J. B., Lubinski, R. B., & Higginbotham, D. J. (1996). Conversational repair by individuals with dementia of the Alzheimer’s type. Journal Speech and Hearing Research, 39(4), 881–895. doi: 10.1044/jshr.3904.881 [DOI] [PubMed] [Google Scholar]
  27. Orange, J. B., Van Gennep, K. M., Miller, L., & Johnson, A. M. (1998). Resolution of communication breakdown in dementia of the Alzheimer’s type: A longitudinal study. Journal of Applied Communication Research, 26(1), 120–138. doi: 10.1080/00909889809365495 [DOI] [Google Scholar]
  28. Perkins, L., Felstead, C., Stott, J., & Spector, A. (2022). Communication training programmes for informal caregivers of people living with dementia: A systematic review. Journal of Clinical Nursing, 31(19–20), 2737–2753. doi: 10.1111/jocn.16175 [DOI] [PubMed] [Google Scholar]
  29. Reisberg, B. (1988). Functional assessment staging (FAST). Psychopharmacology Bulletin, 24(4), 653–659. https://www.ncbi.nlm.nih.gov/pubmed/3249767 [PubMed] [Google Scholar]
  30. Resnick, B., Galik, E., McPherson, R., Boltz, M., Van Haitsma, K., & Kolanowski, A. (2022). Gender differences in disease, function, and behavioral symptoms in residents with dementia. Western Journal of Nursing Research, 44(9), 812–821. doi: 10.1177/01939459211018822 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Ryan, E. B., Meredith, S. D., Maclean, M. J., & Orange, J. B. (1995). Changing the way we talk with elders: Promoting health using the communication enhancement model. International Journal of Aging and Human Development, 41(2), 89–107. doi: 10.2190/fp05-fm8v-0y9f-53fx [DOI] [PubMed] [Google Scholar]
  32. Savundranayagam, M. Y., & Orange, J. B. (2011). Relationships between appraisals of caregiver communication strategies and burden among spouses and adult children. International Psychogeriatrics, 23(9), 1470–1478. doi: 10.1017/s1041610211000408 [DOI] [PubMed] [Google Scholar]
  33. Savundranayagam, M. Y., & Orange, J. B. (2014). Matched and mismatched appraisals of the effectiveness of communication strategies by family caregivers of persons with Alzheimer’s disease. International Journal of Language and Communication Disorders, 49(1), 49–59. doi: 10.1111/1460-6984.12043 [DOI] [PubMed] [Google Scholar]
  34. Shaw, C., Williams, K. N., & Perkhounkova, Y. (2018). Educating nursing home staff in dementia sensitive communication: Impact on antipsychotic medication use. Journal of the American Medical Directors Association, 19(12), 1129–1132. doi: 10.1016/j.jamda.2018.09.030 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Small, J. A., & Gutman, G. (2002). Recommended and reported use of communication strategies in Alzheimer caregiving. Alzheimer Disease and Associated Disorders, 16(4), 270–278. doi: 10.1097/00002093-200210000-00009 [DOI] [PubMed] [Google Scholar]
  36. Small, J. A., Gutman, G., Makela, S., & Hillhouse, B. (2003). Effectiveness of communication strategies used by caregivers of persons with Alzheimer’s disease during activities of daily living. Journal of Speech, Language, and Hearing Research, 46(2), 353–367. doi: 10.1044/1092-4388(2003/028)) [DOI] [PubMed] [Google Scholar]
  37. Tibshirani, R. (1996). Regression shrinkage and selection via the Lasso. Journal of the Royal Statistical Society Series B (Methodological), 58(1), 267–288. doi: 10.1111/j.2517-6161.1996.tb02080.x [DOI] [Google Scholar]
  38. Tjia, J., Hunnicutt, J. N., Herndon, L., Blanks, C. R., Lapane, K. L., & Wehry, S. (2017). Association of a communication training program with use of antipsychotics in nursing homes. JAMA Internal Medicine, 177(6), 846–853. doi: 10.1001/jamainternmed.2017.0746 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Williams, C. L., & Parker, C. (2012). Development of an observer rating scale for caregiver communication in persons with Alzheimer’s disease. Issues in Mental Health Nursing, 33(4), 244–250. doi: 10.3109/01612840.2011.653040 [DOI] [PubMed] [Google Scholar]
  40. Williams, K., Blyler, D., Vidoni, E. D., Shaw, C., Wurth, J., Seabold, D., Perkhounkova, Y., & van Sciver, A. (2018). A randomized trial using telehealth technology to link caregivers with dementia care experts for in-home caregiving support: FamTechCare protocol. Research in Nursing and Health, 41(3), 219–227. doi: 10.1002/nur.21869 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Williams, K. N. (2006). Improving outcomes of nursing home interactions. Research in Nursing and Health, 29(2), 121–133. doi: 10.1002/nur.20117 [DOI] [PubMed] [Google Scholar]
  42. Williams, K. N., Boyle, D. K., Herman, R. E., Coleman, C. K., & Hummert, M. L. (2012). Psychometric analysis of the emotional tone rating scale: A measure of person-centered communication. Clinical Gerontologist, 35(5), 376–389. doi: 10.1080/07317115.2012.702648 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Williams, K. N., Perkhounkova, Y., Bossen, A., & Hein, M. (2016). Nursing home staff intentions for learned communication skills: Knowledge to practice. Journal of Gerontological Nursing, 42(3), 26–34. doi: 10.3928/00989134-20160212-06 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Williams, K. N., Perkhounkova, Y., Herman, R., & Bossen, A. (2017). A communication intervention to reduce resistiveness in dementia care: A cluster randomized controlled trial. Gerontologist, 57(4), 707–718. doi: 10.1093/geront/gnw047 [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Williams, K. N., Perkhounkova, Y., Shaw, C. A., Hein, M., Vidoni, E., & Coleman, C. K. (2019). Supporting family caregivers with technology for dementia home care: A randomized controlled trial. Innovation in Aging, 3(3), igz037. doi: 10.1093/geroni/igz037 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

gnac193_suppl_Supplementary_Material
Click here for additional data file. (32.1KB, docx)

Data Availability Statement

The data, analytic methods or materials are available to other researchers for replication purposes by contacting the authors. This study was not preregistered.

Articles from The Gerontologist are provided here courtesy of Oxford University Press

RESOURCES