Measuring dementia caregiving styles to improve person-centered care.

Skip to main content An official website of the United States government Here's how you know Here's how you know Official websites use .gov A .gov website belongs to an official government organization in the United States. Secure .gov websites use HTTPS A lock ( Lock Locked padlock icon ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites. Search Log in Dashboard Publications Account settings Log out Search… Search NCBI Primary site navigation Search Logged in as: Dashboard Publications Account settings Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide PERMALINK Copy As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice Innov Aging . 2026 Jan 30;10(5):igag006. doi: 10.1093/geroni/igag006 Search in PMC Search in PubMed View in NLM Catalog Add to search Measuring dementia caregiving styles to improve person-centered care Amanda N Leggett Amanda N Leggett , PhD, FGSA 1 Institute of Gerontology, Wayne State University, Detroit, Michigan, United States 2 Department of Psychology, Wayne State University, Detroit, Michigan, United States Find articles by Amanda N Leggett 1, 2, ✉ , Jin-Shei Lai Jin-Shei Lai , PhD 3 Department of Medical Social Sciences, Northwestern University, Chicago, Illinois, United States Find articles by Jin-Shei Lai 3 , Sophia Tsuker Sophia Tsuker , BS 4 Institute of Gerontology, Wayne State University, Detroit, Michigan, United States Find articles by Sophia Tsuker 4 , Jennifer A Miner Jennifer A Miner , MBA 5 Department of Physical Medicine and Rehabilitation, University of Michigan, Ann Arbor, Michigan, United States Find articles by Jennifer A Miner 5 , Jonathan Troost Jonathan Troost , PhD 6 Department of Physical Medicine and Rehabilitation, University of Michigan, Ann Arbor, Michigan, United States Find articles by Jonathan Troost 6 , Noelle E Carlozzi Noelle E Carlozzi , PhD 7 Department of Physical Medicine and Rehabilitation, University of Michigan, Ann Arbor, Michigan, United States Find articles by Noelle E Carlozzi 7 Editor: Megumi Inoue Author information Article notes Copyright and License information 1 Institute of Gerontology, Wayne State University, Detroit, Michigan, United States 2 Department of Psychology, Wayne State University, Detroit, Michigan, United States 3 Department of Medical Social Sciences, Northwestern University, Chicago, Illinois, United States 4 Institute of Gerontology, Wayne State University, Detroit, Michigan, United States 5 Department of Physical Medicine and Rehabilitation, University of Michigan, Ann Arbor, Michigan, United States 6 Department of Physical Medicine and Rehabilitation, University of Michigan, Ann Arbor, Michigan, United States 7 Department of Physical Medicine and Rehabilitation, University of Michigan, Ann Arbor, Michigan, United States ✉ Address correspondence to: Amanda N. Leggett, PhD, FGSA. E-mail: [email protected] Roles Megumi Inoue : PhD, MSW, RN, FGSA , Decision Editor Received 2025 Jun 26; Accepted 2026 Jan 8; Collection date 2026. © The Author(s) 2026. Published by Oxford University Press on behalf of the Gerontological Society of America. This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs licence ( https://creativecommons.org/licenses/by-nc-nd/4.0/ ), which permits non-commercial reproduction and distribution of the work, in any medium, provided the original work is not altered or transformed in any way, and that the work is properly cited. For commercial re-use, please contact [email protected] for reprints and translation rights for reprints. All other permissions can be obtained through our RightsLink service via the Permissions link on the article page on our site—for further information please contact [email protected]. PMC Copyright notice PMCID: PMC13069935  PMID: 41970183 Abstract Background and Objectives Approaches to caregiving interventions are often “one-size-fits-all,” yet family caregivers for individuals with dementia have unique caregiving styles with which they enact daily care. This study describes the development of CG-STYLE, the first person-centered assessment of caregiving style. Research Design and Methods Development phases of this measure included mixed-methods exploratory research, iterative item development and refinement, cognitive interviews, expert review, literacy and translatability review, and field testing of the items in 209 family/friend caregivers for an individual with dementia. Item response theory and classical test theory analyses were run. Results Confirmatory factor analysis affirms the unidimensionality of 7 item banks of the measure: orientation toward the care partner, adaptability, understanding of dementia, emotional expression (positive, negative), and behavioral management (adaptive, maladaptive). Discussion and Implications A person-centered assessment measure can help tailor services and interventions to best fit unique styles of care, ultimately improving the well-being of the caregiver and the quality of care for the individual with dementia. Keywords: Dementia, Family caregiver, Caregiving style, Person-centered measurement, Measure development Innovation and Translational Significance: Our person-centered measure is the first to develop an assessment of cognitive and behavioral domains of dementia family caregiving styles. The CG-STYLE measure may be used by clinicians, community-service agencies, or caregivers themselves to understand caregiving approaches and to tailor and target supports and services to family caregivers’ unique needs. Background and objectives More than 16 million Americans serve in a dementia care role, with the care provided estimated at a total of 18.5 billion hours annually ( Alzheimer’s Association, 2020 ). Although many positive aspects of caregiving are identified, the stress process associated with dementia care is well defined, with dementia family caregivers reporting more depressed mood, anxiety, anger, guilt, and vulnerability to disease than their same-aged non-caregiving peers ( Cuijpers, 2005 ; Fonareva & Oken, 2014 ; Mausbach et al., 2013 ). In contrast to paid caregivers, family caregivers rarely receive disease education or guidance on managing day-to-day care; for example, half of family caregivers have never performed medical/nursing tasks before ( Alzheimer’s Association, 2020 ; McClendon & Smyth, 2013 ). This lack of readiness for care can result in a diverse array of utilized care management approaches, both adaptive and maladaptive, as caregivers attempt to manage the complex behaviors and health care needs of the person living with dementia ( Alzheimer’s Association, 2020 ; McClendon & Smyth, 2013 ). This is important as the ability to navigate and adjust to the changing demands of dementia can affect the well-being of the care dyad as well as whether the individual with dementia can be cared for in the home. For example, prior studies have found that caregivers who use avoidance, confrontation, or disengagement coping strategies have increased depression and anxiety ( Neundorfer, 1991 ), and their care partners with dementia show increased frequency and severity of behavioral and psychological symptoms of dementia ( García-Alberca et al., 2013 ; McClendon & Smyth, 2015 ). In examining dementia care management strategies more specifically, styles reflecting criticism, anger, lack of acceptance, and lower encouragement were associated with caregiver burden, a desire to institutionalize the person living with dementia, and less experienced caregiving gains ( Bakker et al., 2013 ; Davis et al., 2014 ; de Vugt et al., 2004 ; Hinrichsen & Niederehe, 1994 ; Hong et al., 2013 ; Lim et al., 2011 ). Aligning with care management research and cognitive-behavioral theoretical models more broadly, our prior research has shown that caregivers develop a “caregiving style” composed of interrelated cognitive frameworks and behavioral actions that inform their care decision-making ( Ajzen & Fishbein, 1980 ; Beck, 1976 ; Fishbein, 2008 ; Leggett et al., 2021 ). Our prior work utilized thematic coding of in-depth qualitative interviews on the cognitive and behavioral approaches used by family caregivers to manage care challenges with individuals living with dementia; see Leggett et al. (2021) for a detailed description. Focused codes were organized into themes reflecting common cognitive-behavioral approaches, which are reflected as five key domains that underlie one’s caregiving style. Caregiving styles, therefore, can be organized by common patterns of response across the following domains (many corresponding with other research on caregiver management approaches), which tended to align by cognitive framework and behavioral management (e.g., lack of adaptability tended to appear with maladaptive behavioral strategies ( Davis et al., 2014 ; Hong et al., 2013 ; Leggett et al., 2019 , 2020 , 2021 ): (1) understanding of dementia (understanding a disease process taking place, recognizing symptoms as stemming from the disease and not as intentional behaviors, understanding that previously normative forms of interaction and communication may no longer be effective), (2) emotional approach (more negative emotions such as feeling turbulent, frustrated, helpless, or guilty and more positive emotions such as feeling empathetic, optimistic, and carrying a positive attitude towards caregiving), (3) adaptability (efficacious mastery involving the ability to be flexible in one’s caregiving, willingness to change one’s management, or having multiple care strategies at the ready to navigate a challenge; this is in contrast to more rigid caregiving approaches representing inability to adjust one’s perspective or approach), (4) orientation of care (focusing more on the stress of caregiving to oneself or in contrast, focusing on the stresses of care challenges on the person living with dementia), and (5) enacted behavioral management (more adaptive caregiving behaviors such as working as a team, diverting attention, reassurance, and environmental modifications and more maladaptive caregiving behaviors such as expressing anger, placing blame, arguing, and yelling). Full domain definitions are found in Table 1 . Five distinct caregiving styles emerged from our qualitative analysis based on patterns of response (focused codes) across the five domains: Externalizer, Individualist, Learner, Adapter, and Nurturer, as summarized in Supplementary Table 1 ( Leggett et al., 2021 ). Table 1. Definitions of the five domains of CG-STYLE and spread of items across the domains. Domain Domain definition Number of items Initial item pool Final short forms Understanding Recognizing symptoms as stemming from disease and not intentional; recognizing the cognitive, behavioral, psychological, and functional aspects of dementia. Understanding caregivers would acknowledge that the “reality” of the care recipient has shifted relative to the “reality” of the caregiver (e.g., understanding that normative forms of communication prior to the disease process may no longer be effective). 10 3 Adaptability Adaptability styles include: (1) a rigid, inflexible, and direct (point A to point B) approach to care (absence of change in management strategy, adjustment, or perspective); (2) a willingness or readiness to change or be flexible in one’s approach, characterized by a growth perspective; and (3) accrued or natural adaptability. An adaptable caregiver is able to “roll with the punches” or “go with the flow,” is open to changing their approach, and is able to handle care challenges with multiple behavioral management strategies. An adaptable caregiver does not tend to get “stuck” in managing care challenges but is efficacious and able to figure out a solution. Those who are adaptable reflect on care management practices and their effects, whereas those who are not adaptable do not reflect on them. 33 3 Emotional expression: positive Emotional expression that includes showing positive emotional regulation in the face of challenges, such as feelings of empathy, optimism, and positive affect towards care. 7 2 Emotional expression: negative Emotional expression that includes more maladaptive emotions (e.g., frustration, anger, guilt, and helplessness) and being emotionally removed, repressing, or tempering one’s emotions. 18 3 Orientation to Self or Other Self-focused caregivers worry more about the stress of caregiving challenges on themselves. Other-focused caregivers worry more about the stress of caregiving challenges on the care recipient. 12 2 Behavioral management strategies: Adaptive Strategies that are person-centered, respect the personhood of the care recipient, and are more encouraging. These may include using humor (making light or fun of a situation in a positive way), comfort (reassure, calm, put at ease), and environmental modifications, working as a team with the care recipient to address a challenge, diverting the attention of the care recipient when they become upset or perseverate on something, and putting the focus on the caregiver or a medical professional or other person in a challenging situation to avoid centering focus or blame on the care recipient 13 2 Behavioral management strategies: Maladaptive Strategies that are more critical, attempt to bring the care recipient back within the reality of the caregiver, and could be potentially harmful. These may include anger, arguing (bickering with the care recipient as a result of an unsuccessful care strategy), yelling (or showing outward displeasure), placing blame/confrontation, “lone wolf”/going it alone, and sarcasm (joking in a harsher or more sarcastic way). 7 3 Open in a new tab Theoretical context Of note, although the conceptualization of caregiving styles was based more broadly on cognitive-behavioral theories highlighting the interconnectedness of one’s thoughts and behaviors ( Ajzen & Fishbein, 1980 ; Beck, 1976 ; Fishbein, 2008 ), the five domains have some alignment with other common family caregiving models such as the Stress Process Model ( Pearlin et al., 1990 ). For example, an understanding of dementia and its symptoms may affect stress appraisals, the emotional approach and orientation to oneself, or the individual with dementia may align with aspects of secondary intrapsychic strains, and adaptability may correspond with one facet of Pearlin’s conceptualization of coping, that being a more active management of more stressful care contexts. In contrast, adaptive and maladaptive behavioral care management strategies are not part of the Stress Process model and are an addition to the caregiving style model. Furthermore, our caregiving style profiles group caregivers based on unique patterns of cognitive-behavioral management strategies, which differ from the more variable-centered approach of the Stress Process Model in predicting caregiver-specific outcomes ( Pearlin et al., 1990 ). To date, however, there has been no way to assess a dementia caregiver’s caregiving style. This assessment is critically needed to aid in identifying mechanisms of action that improve care outcomes and build the foundation for personalized medicine for caregivers. For example, assessing someone’s caregiving style has the potential to improve person-centered care by treating caregivers as unique individuals, not as “one-size-fits-all” within an intervention context, and allowing for the tailoring of service and support recommendations and interventions to best meet a caregiver’s distinct style ( Molony et al., 2018 ). Disappointingly modest gains from existing caregiver interventions may be due to the limited understanding of the variability in caregivers’ style ( Agency for Healthcare Research and Quality, 2020 ; Walter & Pinquart, 2019 ). Current study This paper details the person-centered development process of the first measure of dementia caregiving style (CG-STYLE) (the measure, along with its validity and reliability characteristics, is described in Leggett et al., 2025 ). The CG-STYLE measure aims to capture the five cognitive-behavioral component domains of caregiving style: understanding of dementia, adaptability, emotional approach, orientation of care, and behavioral management, such that profiles of caregiving style can be ascertained. Research design and methods Study participants Participants included 209 family caregivers of individuals living with dementia. Inclusion criteria Caregivers were English-speaking, 18 years of age or older, and providing unpaid physical, emotional, and/or financial assistance for a family member or friend with a doctor’s diagnosis of an age-related dementia (not including mild cognitive impairment). Exclusion criteria The care partners, or individuals living with dementia, could not be bedbound, living in a long-term care facility, or have a life expectancy of 6 months or less. The sample does not include one duplicate that was removed, two participants who completed only the demographic section of the interview, and two who were dropped after being identified as fake participants due to contradictory and inconsistent survey responses. Participant recruitment process Efforts were made to recruit a diverse sample of caregivers through the [Blinded] Alzheimer’s Disease Center’s Memory and Aging Project participant pool, the Alzheimer’s Association Trial Match database, the Healthier Black Elders Center’s participant research pool, the Michigan Health Research database, the CARE registry, the Leggett Lab caregiver registry, a Michigan Medicine Facebook advertisement, participant referral, and other community events. This study was initially approved by the Michigan Medicine IRB (and subsequently by the Wayne State University IRB) in accordance with the 1964 Declaration of Helsinki. Participants provided their informed consent and received a small stipend for their participation. Study measures: the caregiver STYLE item pool Item pool development Literature review and previous empirical work have identified several primary domains of caregiving styles: (1) understanding of dementia (attributing symptoms to the disease), (2) adaptability (efficacious mastery or more rigid approaches), (3) emotional expression (positive: positive or empathetic; negative: feeling turbulent, frustrated), (4) orientation to self and others (focusing on stress on oneself or the person living with dementia), and (5) enacted behavioral management strategies (e.g., adaptive: working as a team, diverting attention, reasoning; maladaptive: anger) ( Davis et al., 2014 ; Hong et al., 2013 ; Leggett et al., 2019 ; 2020 ; 2021 ). The Principal investigator, four research staff, and a co-investigator who were involved in the study ( Leggett et al., 2021 ) that originated the caregiving styles and component domains developed item pools that captured each of these important concepts. These item pools include 7 item banks, with emotional expression split into positive and negative banks, and behavioral management strategies split into adaptive and maladaptive banks. Items were developed to fit two Likert response scales (strongly disagree to strongly agree; almost never to almost always). Items were developed using an “in vivo” coding type approach that generated items using caregivers’ wording from qualitative interviews on caregiving style and also in alignment with key facets of the domain definitions developed from the qualitative interviews (see Table 1 ). Iterative process for item pool refinement After generating as many items as possible during the initial item pool development process, an iterative process was used to refine them ( Figure 1 ). Originally, the team developed 188 items (26 Understanding, 59 Adaptability, 19 Orientation to Self or Other, 47 Emotional Expression, and 37 Behavioral Management). The first step of iterative refinement was expert review, in which three dementia caregiving content experts and three measurement development experts suggested revisions and deletions, resulting in 63 items that were modified and 66 items that were deleted. Next, cognitive interviews were conducted where each item was reviewed by five family and friend caregivers, with each caregiver reviewing one or more complete sets of items by domain (seven total caregivers). Caregivers were asked to answer each item, explain how they came up with their answer, and were asked if they could think of a better way to word the item. Further, for each domain set, they were asked if they were comfortable with the set of items, how they would categorize the set of items, whether there were any other questions we should ask, and if they had any other suggestions to improve the questions for future use. Upon review of interview responses by the full research team, 19 items were modified, and 4 were deleted. A translatability review of Spanish was conducted by FACIT ( https://www.facit.org/ ) to ensure that wording and phrases that could impede future translations were removed (e.g., the phrase “things go wrong” was flagged as vague and changed to “when there are problems”), resulting in 28 items modified and 4 items deleted. Finally, items were evaluated for item literacy level with the Lexile Analyzer ( https://hub.lexile.com/text-analyzer/ ) to ensure ≤6th-grade reading level; edits but no deletions were made. The final item pool for field testing was comprised of 99 items; see Table 1 . The final measure and its reliability and validity data are published in Leggett et al. (2025) . Figure 1. Open in a new tab Iterative process for item pool development. Analysis plan In accordance with established patient-reported outcome measurement development standards, classical test theory and item response theory (IRT) analytical approaches were used to develop the CG-STYLE Questionnaire ( Cappelleri et al., 2014 ; Carlozzi et al., 2023 ; PROMIS, 2013 ). Confirmatory factor analysis (CFA) was used to assess the unidimensionality of the content included in each of the 7 item banks (Understanding Dementia, Adaptability, Positive Emotional Approach, Negative Emotional Approach, Orientation to Self or Other, Adaptive Behavioral Management, Maladaptive Behavioral Management), using Mplus version 7.4 ( Muthén & Muthén, 2015 ). We evaluated model fit using the following criteria: comparative fit index (CFI) >0.90, root mean square error of approximation (RMSEA) <0.1 ( Hu & Bentler, 1999 ). Items were considered excluded from the measure under the following criterion: sparse cells (i.e., n < 10 respondents for a response category), residual correlations between item-pairs >0.15 ( Cook et al., 2009 ; Reise et al., 2007 ), and low R -squared ( R 2 ) (i.e., <0.30) ( Moore, 1996 ). Analyses were iterative, and input from the principal investigators (A.N.L. and N.E.C.) was used to make final item selection/deletion decisions ( Cook et al., 2009 ; McDonald, 1999 ; Reise et al., 2007 ). Next, Samejima’s graded response model (GRM) was used to establish item parameters ( Samejima, 1969 ) using IRTPRO (version 3.1.2) ( Cai et al., 2015 ). Items were removed if they displayed significant misfit (S- X 2 / df effect size >3) ( Stark et al., 2006 ). For each domain, final items were selected using item calibration and calibration-related statistics (e.g., item slope, thresholds, average item difficulty, and item information) and clinical expertise to ensure domain content coverage. For each of these domains, IRT-scaled scores were converted into a standardized score utilizing a T score (mean = 50, SD = 10), with the caregiver population represented by the current sample as reference. An advantage of IRT-calibrated item banks is the ability to construct short forms tailored to research or clinical needs. Scores of short forms remain directly comparable to those from the full-item banks if scoring uses the original calibration parameters. If short forms utilize the same scale and item parameters, their reliability and validity reflect those of the full calibrated banks ( Cella et al., 2019 ). In cases where IRT analysis was precluded, classical test theory data were used in conjunction with clinical expertise for the selection of final items. In this case, raw sum scores were generated. Sample size considerations Our sampling size ( N > 200) supports the utilization of GRM, which estimates a common discrimination/slope parameter and individual item threshold parameters across the domains of the CG-STYLE Questionnaire. Although sample sizes of 200-1,000 have been proposed when using a GRM ( Orlando, 2004 ), convention dictates that a minimum of 100 individuals is needed to provide stable parameters when using a restricted GRM; thus, our sample of over 200 participants exceeded the minimal sample size requirements to ensure that statistical assumptions are met. Results A total of 209 care partners participated in this study; see Supplementary Table 2 for a summary of the sample descriptives. CG-STYLE measurement development First, an essentially unidimensional set of items was established for each domain except Orientation to Self or Other (see Table 2 ). Understanding was winnowed from 10 items to 6 by dropping one item for residual correlation, 2 for sparse cells, and one for its differing rating scale. Adaptability was narrowed from 32 items to 26 by dropping 4 items for residual correlations and 2 for high R 2 and residual correlations. For Positive Emotional Expression, 7 items were reduced to 4 by dropping one item for not meeting the sparse cell criterion and 2 items for being on a different rating scale. Negative Emotional Expression was winnowed from 18 items to 14 by dropping 3 items based on the residual correlation criterion and one item for both high R 2 and residual correlation. For Adaptive Behavioral Management, items were reduced from 13 to 5 (2 dropped for R 2 only, 3 for the residual correlation, 1 for high R 2 and residual correlation, and 2 for sparse cells). Finally, Maladaptive Behavioral Management was reduced from 7 items to 6 (1 item did not meet the sparse cell criterion). No items were dropped based on differential item functioning (DIF) or IRT item misfit. For Orientation to Self or Other, IRT analysis was abandoned, and classical test theory was used in the final scoring of this domain. Table 2. Unidimensional modeling and analyses for CG-STYLE domains. Domain Initial item pool CFA modeling Initial item performance IRT modeling R 2 only (>0.3) Residual correlation only (<0.40) Both high R 2 and residual correlations Sparse cells (<10) Different rating scale IRT item misfit DIF Interim/final item bank Understanding 10 0 1 0 2 1 0 0 6 Adaptability 32 0 4 2 0 0 0 0 26 Positive emotional expression 7 0 0 0 1 2 0 0 4 Negative emotional expression 18 0 3 1 0 0 0 0 14 Adaptive behavioral management strategies 13 2 3 1 2 0 0 0 5 Maladaptive behavioral management strategies 7 0 0 0 1 0 0 0 6 Open in a new tab Note. CFA = confirmatory factor analysis; DIF = differential item functioning; EFA = exploratory factor analysis; IRT = item response theory. A final CFA supporting the unidimensionality of each item bank, as well as the internal consistency reliability for each item bank, is provided in Table 3 . As previously noted, all domains, except for Orientation to Self or Other, met the criteria of CFI (>0.90), TLI (>0.90), RMSEA (<0.15), and Internal Consistency (>0.70). Table 3. Final model fit for the CG-STYLE domains. Domain Theta scores M ( SD ) CFI (>0.90) TLI (>0.90) RMSEA (<0.15) Internal consistency (>0.70) Understanding 50.0 (9.3) 0.985 0.975 0.097 0.840 Adaptability 50.0 (9.8) 0.949 0.945 0.081 0.940 Positive emotional expression 50.0 (9.1) 0.991 0.972 0.105 0.720 Negative emotional expression 50.0 (9.7) 0.975 0.971 0.090 0.940 Adaptive behavioral management strategies 50.0 (9.8) 0.995 0.990 0.046 0.710 Maladaptive behavioral management strategies 50.0 (9.2) 0.991 0.985 0.120 0.900 Open in a new tab Note. Chi-square = chi-square test of model fit; CFI = comparative fit index; RMSEA = root mean square error of approximation ( Bentler, 1990 ; Cook et al., 2009 ; Hatcher, 1994 ; Hu & Bentler, 1999 ; Kline, 2005 ; Lai et al., 2011 ; 2014 ); TLI = Tucker–Lewis index. Scoring Psychometric and clinical considerations were balanced in the selection of the short forms that would comprise the final version of this measure. Specifically, the “best” 2 to 3 items for each domain were selected using both the psychometric performance data of the item as well as the item-specific content and representativeness of the associated domain. See Supplementary file “Caregiver Style – 18 item Short Form” for the final short-form version of the CG-STYLE Measure, including its scoring conversion tables and instructions. Full reliability and validity data for CG-STYLE are reported in Leggett et al. (2025) . Discussion and implications The 2023 NIH Dementia Care Summit identified a core research gap (G&O 1.1) relating to the need for measures that assess outcomes important to persons living with dementia and their care partners using standardized methods ( National Institute on Aging, 2020 ). This study presents the first person-centered measure of dementia caregiving style with attention to validated methods and the ability for harmonization (e.g., translatability review, literacy review). Standardized measures for each of five domains of caregiving style were developed (results on unidimensional modeling and fit for full item banks presented here) alongside short forms for use by community service agencies, clinical practitioners, and caregivers themselves interested in assessing how someone is managing caregiving challenges and how care management may relate to care outcomes or provide context for tailoring and targeting services and interventions. Measurement development was conducted with mixed-methods approaches in accordance with established standards, including qualitative interviewing, classical test theory, and IRT analyses ( PROMIS, 2013 ). Seven calibrated short forms were also developed and tested amongst caregivers managing care for individuals living with dementia across disease stages. The short forms allow each domain to be captured while minimizing participant burden. Further, the domains (excluding Orientation to Self or Other) are scored with a T-metric indicating how a dementia caregiver’s cognitive and behavioral style attributes compare to those of other dementia caregivers, where a higher score indicates a better score on a particular attribute (e.g., more understanding, less negative emotionality). With the t -score standardized to a mean of 50 and a standard deviation of 10, a score of 60 thus represents one standard deviation above the mean for family/friend caregivers of individuals living with dementia. Additionally, our study emphasized person-centered approaches in the development of the CG-STYLE measure. The CG-STYLE measure was built upon mixed-methods research, classifying five distinct caregiving styles across the cognitive domains of understanding of dementia, adaptability, emotional approach, and orientation of care toward oneself or the individual with dementia, as well as a domain of behavioral care management. Iterative item development and refinement utilized “in vivo” codes where possible to capture caregivers’ own wording of key concepts. Item refinement incorporated literacy review, expert review, and translatability review to ensure that items captured key facets of style, were readable by a diverse audience of caregivers and practitioners, and were translatable to other languages (particularly Spanish). Caregivers’ own feedback was included through in-depth cognitive interviews, in which they provided perspectives on their comfort with the items and suggested any concepts not captured within the domain. The CG-STYLE measure aligns in several fundamental ways with established person-centered measurement development principles as adapted from Zimmerman and colleagues’ (2023) “Person-Centered Measure Evaluation Tool (PC-Met)” and established person-centered principles ( Zimmerman & Fazio, 2022 ). First, regarding co-creation of the measure with individuals with dementia and their caregivers ( Miah et al., 2019 ), caregivers and relevant stakeholders had a role in decision-making across the measurement development stages (e.g., member checking interviews conducted with clinicians working with family/friend caregivers to form proof of concept, items were developed from qualitative interviews with family/friend caregivers, cognitive interviews were conducted with caregivers and their feedback was incorporated in the refinement of items). The concept of pragmatism ( Glasgow & Riley, 2013 ) centers on reducing the burden of use and resulting in actionable data. To this end, we conducted a translatability review to ensure broad applicability, designed the measure so that training is not needed for administration, and increased accessibility by allowing the assessment to be administered online, on paper, or verbally. Our research team also aimed to accommodate caregivers’ needs by fully disclosing the topics to be covered, accounting for different language speakers, obtaining and using caregiver feedback, and ensuring that interviewers were trained in dementia and dementia care ( Øksnebjerg et al., 2018 ). We worked to ensure that goals prioritized by family/friend caregivers were incorporated into the measure ( Zimmerman et al., 2023 ). For example, caregivers tended to prefer items that were more positively worded, which were prioritized in our measure. Finally, a systemic focus targeted the broad relevance of the measure by using consistent language and working with clinicians and caregivers to develop items with clinical and practical relevance ( Zimmerman et al., 2023 ). Our study is not without limitations. One domain, “Orientation to Self or Other,” did not align in a unidimensional way through IRT analysis; thus, classical test theory analyses were used to establish this item bank. Although our sample represents a diverse group of family/friend caregivers for individuals living with dementia, as previously mentioned, we did exclude participants ( n = 2) from our analysis whose responses did not align in a coherent story and whose responses were believed to be fabricated. Additionally, though diverse, the sample size is considered small and may not be representative of the entire population of dementia caregivers, and this study was based only on data from family caregivers of individuals with dementia. Replicability of the study results is warranted by using a different sample. Additional analyses were conducted to establish the validity and reliability of this measure and are reported elsewhere ( Leggett et al., 2025 ). Finally, Fazio and colleagues’ (2018) person-centered care practice recommendations include regular assessments of care, which align with caregiving styles that may change with accumulated experience in caregiving and the progression of symptoms of dementia. However, future longitudinal work should explore the mutability of the caregiving style domains across the caregiving journey and stages of disease, which could not be ascertained in this cross-sectional data collection. The Alzheimer’s Association’s Dementia Care Practice Recommendations ( Fazio et al., 2018 ) suggests caregiving assessments be used to promote education and support of the care dyad in addition to gathering information. The aim of the CG-STYLE measure is to enable clinicians, practitioners, community service providers, and care partners to understand an individual’s unique caregiving style and to tailor and target services and supports given this understanding. For example, a caregiver who is low in understanding and adaptability, while high in negative emotionality (i.e., an Externalizer), may benefit from an informational intervention provided by a collaborative care team at a primary care office that grounds dementia symptomatology in the disease process. A social worker or other support group facilitator may wish to pair an Adapter or Nurturer with a Learner or Individualist via peer support to facilitate adoption of more adaptive behavioral management strategies. An app-based intervention to help caregivers manage behavioral and psychological symptoms of dementia or develop skills ( Shin et al., 2022 ) might utilize CG-STYLE to tailor push notifications and recommendations sent to caregivers. A person-centered measure of dementia caregiving style is therefore vital to the development of more effective and individualized treatments that enhance caregivers’ ability to support persons living with dementia, improve outcomes, and reduce the downstream burden of illness and healthcare costs. Additionally, prior research often reports discrepancies in care values and preferences between family caregivers and individuals living with dementia ( Reamy et al., 2012 ; Shelton et al., 2017 ; Van Haitsma et al., 2020 ; Wehrmann et al., 2021 ). Hence, for interventions to be truly person-centered, a focus on the dyadic perspective of a caregiver’s style profile, as well as an individual with dementia’s preferences for their care, is critical to consider. Future work should explore the utility of the CG-STYLE measure in various groups serving individuals with dementia so that practice recommendations may be established. Supplementary Material igag006_Supplementary_Data igag006_supplementary_data.zip (208.4KB, zip) Acknowledgments We would like to acknowledge Drs Helen C. Kales, Laura N. Gitlin, and Michael Kallen for their expert review of the items and contribution to the development of the caregiving style concept. Contributor Information Amanda N Leggett, Institute of Gerontology, Wayne State University, Detroit, Michigan, United States; Department of Psychology, Wayne State University, Detroit, Michigan, United States. Jin-Shei Lai, Department of Medical Social Sciences, Northwestern University, Chicago, Illinois, United States. Sophia Tsuker, Institute of Gerontology, Wayne State University, Detroit, Michigan, United States. Jennifer A Miner, Department of Physical Medicine and Rehabilitation, University of Michigan, Ann Arbor, Michigan, United States. Jonathan Troost, Department of Physical Medicine and Rehabilitation, University of Michigan, Ann Arbor, Michigan, United States. Noelle E Carlozzi, Department of Physical Medicine and Rehabilitation, University of Michigan, Ann Arbor, Michigan, United States. Supplementary material Supplementary data are available at Innovation in Aging online. Funding This work is funded by an Alzheimer’s Association and LINC-AD “Advancing Research on Care and Outcome Measurement (ARCOM) Program” award (ARCOM-22-876659; PI: A.N.L.), which was partially supported by the Dementia Care Provider Roundtable. A.N.L. is also funded by the National Institute on Aging at the National Institutes of Health (P30AB072931). The study funder played no role in the design, methods, subject recruitment, data collections, analysis, or preparation of the paper. Conflict of interest None declared. Data Availability The data are not currently available for the privacy of the individuals who participated in the study. For questions about the data, contact the corresponding and lead author (A.N.L.). References Agency for Healthcare Research and Quality. (2020). Care interventions for people with dementia (PWD) and their caregivers . https://effectivehealthcare.ahrq.gov/products/care-interventions-pwd/protocol Ajzen I., Fishbein M. (1980). Understanding attitudes and predicting social behavior. Prentice Hall. [ Google Scholar ] Alzheimer’s Association. (2020). Alzheimer’s disease facts and figures. Alzheimer’s and Dementia, 16, 391–460. 10.1002/alz.12068 [ DOI ] [ Google Scholar ] Bakker C., de Vugt M. E., van Vliet D., Verhey F. R. J., Pijnenburg Y. A., Vernooij-Dassen M. J. F. J., Koopmans R. T. C. M. (2013). Predictors of the time to institutionalization in young- versus late-onset dementia: Results from the Needs in Young Onset Dementia (NeedYD) Study. Journal of the American Medical Directors Association, 14, 248–253. https://doi.org/ 10.1016/j.jamda.2012.09.011 [ DOI ] [ PubMed ] [ Google Scholar ] Beck A. T. (1976). Cognitive therapy and the emotional disorders. Penguin. [ Google Scholar ] Bentler P. M. (1990). Comparative fit indexes in structural models. Psychological Bulletin, 107, 238–246. 10.1037/0033-2909.107.2.238 [ DOI ] [ PubMed ] [ Google Scholar ] Cai L., Thissen D., du Toit S. H. C. (2015). IRTPRO for Windows [Computer software]. Scientific Software International. [ Google Scholar ] Cappelleri J. C., Jason Lundy J., Hays R. D. (2014). Overview of classical test theory and item response theory for the quantitative assessment of items in developing patient-reported outcomes measures. Clinical Therapeutics, 36, 648–662. 10.1016/j.clinthera.2014.04.006 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Carlozzi N. E., Kallen M. A., Troost J. P., Bragg A., Martin-Howard J., De La Cruz B., Miner J. A., Moldovan I., Jack B. W., Mitchell S. (2023). Development and calibration data for the Healthcare Access Item Bank: A new computer adaptive test for persons with type 2 diabetes mellitus. Quality of Life Research, 32, 781–796. 10.1007/s11136-022-03278-w [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Cella D., Choi S. W., Condon D. M., Schalet B., Hays R. D., Rothrock N. E., Yount S., Cook K. F., Gershon R. C., Amtmann D., DeWalt D. A., Pilkonis P. A., Stone A. A., Weinfurt K., Reeve B. B. (2019). PROMIS( ® ) Adult Health Profiles: Efficient short-form measures of seven health domains. Value Health, 22, 537–544. 10.1016/j.jval.2019.02.004 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Cook K. F., Kallen M. A., Amtmann D. (2009). Having a fit: Impact of number of items and distribution of data on traditional criteria for assessing IRT’s unidimensionality assumption. Quality of Life Research, 18, 447–460. 10.1007/s11136-009-9464-4 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Cuijpers P. (2005). Depressive disorders in caregivers of dementia patients: A systematic review. Aging & Mental Health, 9, 325–330. 10.1080/13607860500090078 [ DOI ] [ PubMed ] [ Google Scholar ] Davis L. L., Chestnutt D., Molloy M., Deshefy-Longhi T., Shim B., Gilliss C. L. (2014). Adapters, strugglers, and case managers: A typology of spouse caregivers. Qualitative Health Research, 24, 1492–1500. 10.1177/1049732314548879 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] de Vugt M. E., Stevens F., Aalten P., Lousberg R., Jaspers N., Winkens I., Jolles J., Verhey F. R. J. (2004). Do caregiver management strategies influence patient behaviour in dementia?  International Journal of Geriatric Psychiatry, 19, 85–92. 10.1002/gps.1044 [ DOI ] [ PubMed ] [ Google Scholar ] Fazio, S., Pace, D., Maslow, K., Zimmerman, S., Kallmyer, B. (2018). Alzheimer's Association Dementia Care Practice Recommendations.  The Gerontologist. 58 (suppl_1), S1–S9. 10.1093/geront/gnx18229361074 [ DOI ] [ PubMed ] [ Google Scholar ] Fishbein M. (2008). A reasoned action approach to health promotion. Medical Decision Making, 28, 834–844. 10.1177/0272989x08326092 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Fonareva I., Oken B. S. (2014). Physiological and functional consequences of caregiving for relatives with dementia. International Psychogeriatrics, 26, 725–747. 10.1017/S1041610214000039 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] García-Alberca J. M., Cruz B., Lara J. P., Garrido V., Lara A., Gris E., Gonzalez-Herero V. (2013). The experience of caregiving: The influence of coping strategies on behavioral and psychological symptoms in patients with Alzheimer’s disease. Aging & Mental Health, 17, 615–622. 10.1080/13607863.2013.765833 [ DOI ] [ PubMed ] [ Google Scholar ] Glasgow R. E., Riley W. T. (2013). Pragmatic measures: What they are and why we need them. American Journal of Preventive Medicine, 45, 237–243. 10.1016/j.amepre.2013.03.010 [ DOI ] [ PubMed ] [ Google Scholar ] Hatcher L. (1994). A step-by-step approach to using SAS for factor analysis and structural equation modeling. SAS Institute, Inc. [ Google Scholar ] Hinrichsen G. A., Niederehe G. (1994). Dementia management strategies and adjustment of family members of older patients. The Gerontologist, 34, 95–102. 10.1093/geront/34.1.95 [ DOI ] [ PubMed ] [ Google Scholar ] Hong S.-I., Luo N., Yap P. (2013). Maximizing a nurturing care style for persons with dementia: A person-centered analysis. The American Journal of Geriatric Psychiatry, 21, 987–998. 10.1016/j.jagp.2013.01.034 [ DOI ] [ PubMed ] [ Google Scholar ] Hu L. T., Bentler P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6, 1–55. 10.1080/10705519909540118 [ DOI ] [ Google Scholar ] Kline R. B. (2005). Principles and practice of structural equation modeling (2nd ed.). Guilford Press. [ Google Scholar ] Lai J. S., Cella D., Choi S., Junghaenel D. U., Christoudolou C., Gershon R., Stone A. (2011). How item banks and its applications can influence measurement practice in rehabilitation medicine: A PROMIS fatigue item bank example. Archives of Physical Medicine & Rehabilitation, 92(Supp 1), S20–S27. 10.1016/j.apmr.2010.08.033 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Lai J.-S., Zelko F., Krull K., Cella D., Nowinski C., Manley P., Goldman S. (2014). Parent-reported cognition of children with cancer and its potential clinical usefulness. Quality of Life Research, 23, 1049–1058. 10.1007/s11136-013-0548-9 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Leggett A. N., Bugajski B. C., Gitlin L. N., Kales H. C. (2021). Characterizing dementia caregiver style in managing care challenges: Cognitive and behavioral components. Dementia, 20, 2188–2204. 10.1177/1471301220988233 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Leggett A. N., Hodgson N., Connell C. M., Gitlin L. N., Kales H. C. (2020). “I understand a fury in your words, but not the words”: The association of dementia caregiving styles and salivary cortisol. Alzheimer’s & Dementia, 16(S7), e040514. 10.1002/alz.040514 [ DOI ] [ Google Scholar ] Leggett A. N., Kales H. C., Gitlin L. N. (2019). Finding fault: Criticism as a care management strategy and its impact on outcomes for dementia caregivers. International Journal of Geriatric Psychiatry, 34, 571–577. 10.1002/gps.5052 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Leggett A. N., Troost J., Miner J. A., Tsuker S., Lai J.-S., Carlozzi N. E. (2025). Validating the CG-STYLE measure in family caregivers of individuals living with dementia. Alzheimer’s & Dementia, 21, e70492. 10.1002/alz.70492 [ DOI ] [ Google Scholar ] Lim J., Griva K., Goh J., Chionh H. L., Yap P. (2011). Coping strategies influence caregiver outcomes among Asian family caregivers of persons with dementia in Singapore. Alzheimer Disease & Associated Disorders, 25, 34–41. 10.1097/WAD.0b013e3181ec18ae [ DOI ] [ PubMed ] [ Google Scholar ] Mausbach B. T., Chattillion E. A., Roepke S. K., Patterson T. L., Grant I. (2013). A comparison of psychosocial outcomes in elderly Alzheimer caregivers and noncaregivers. The American Journal of Geriatric Psychiatry, 21, 5–13. 10.1016/j.jagp.2012.10.001 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] McClendon M. J., Smyth K. A. (2013). Quality of informal care for persons with dementia: Dimensions and correlates. Aging & Mental Health, 17, 1003–1015. 10.1080/13607863.2013.805400 [ DOI ] [ PubMed ] [ Google Scholar ] McClendon M. J., Smyth K. A. (2015). Quality of in-home care, long-term care placement, and the survival of persons with dementia. Aging & Mental Health, 19, 1093–1102. 10.1080/13607863.2014.1003284 [ DOI ] [ PubMed ] [ Google Scholar ] McDonald R. P. (1999). Test theory: A unified treatment. Lawrence Erlbaum Associates, Inc. [ Google Scholar ] Miah J., Dawes P., Edwards S., Leroi I., Starling B., Parsons S. (2019). Patient and public involvement in dementia research in the European Union: A scoping review. BMC Geriatrics, 19, 220. 10.1186/s12877-019-1217-9 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Molony S. L., Kolanowski A., Van Haitsma K., Rooney K. E. (2018). Person-centered assessment and care planning. The Gerontologist, 58(suppl_1), S32–S47. 10.1093/geront/gnx173 [ DOI ] [ PubMed ] [ Google Scholar ] Moore L. M. (1996). The basic practice of statistics. Taylor & Francis. [ Google Scholar ] Muthén L. K., Muthén B. O. (2015). Mplus User’s Guide (7th ed.).  Muthén & Muthén. [ Google Scholar ] National Institute on Aging. (2020, December 19). National research summit on care, services, and supports for persons living with dementia and their caregivers . https://www.nia.nih.gov/sites/default/files/2021-01/DementiaCareSummitReport.pdf Neundorfer M. M. (1991). Coping and health outcomes in spouse caregivers of persons with dementia. Nursing Research, 40, 260–265. http://journals.lww.com/nursingresearchonline/Fulltext/1991/09000/Coping_and_Health_Outcomes_In_Spouse_Caregivers_of.2.aspx [ PubMed ] [ Google Scholar ] Øksnebjerg L., Diaz-Ponce A., Gove D., Moniz-Cook E., Mountain G., Chattat R., Woods B. (2018). Towards capturing meaningful outcomes for people with dementia in psychosocial intervention research: A pan-European consultation. Health Expectations, 21, 1056–1065. 10.1111/hex.12799 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Orlando B. (2004). Critical issues to address when applying item response theory (IRT) models. RAND Corp. [ Google Scholar ] Pearlin L. I., Mullan J. T., Semple S. J., Skaff M. M. (1990). Caregiving and the stress process: An overview of concepts and their measures. Gerontologist, 30, 583–594. 10.1093/geront/30.5.583 [ DOI ] [ PubMed ] [ Google Scholar ] PROMIS. (2013). PROMIS instrument development and validation scientific standards . https://www.healthmeasures.net/explore-measurement-systems/promis/measure-development-research Reamy A. M., Kim K., Zarit S. H., Whitlatch C. J. (2012). Values and preferences of individuals with dementia: Perceptions of family caregivers over time. The Gerontologist, 53, 293–302. 10.1093/geront/gns078 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Reise S. P., Morizot J., Hays R. D. (2007). The role of the bifactor model in resolving dimensionality issues in health outcomes measures. Quality of Life Research, 16(Suppl 1), 19–31. 10.1007/s11136-007-9183-7 [ DOI ] [ PubMed ] [ Google Scholar ] Samejima F. (1969). Estimation of latent ability using a response pattern of graded scores. Psychometric Society. [ Google Scholar ] Shelton E. G., Orsulic-Jeras S., Whitlatch C. J., Szabo S. M. (2017). Does it matter if we disagree? The impact of incongruent care preferences on persons with dementia and their care partners. The Gerontologist, 58, 556–566. 10.1093/geront/gnw202 [ DOI ] [ Google Scholar ] Shin Y., Kim S. K., Kim Y., Go Y. (2022). Effects of app-based mobile interventions for dementia family caregivers: A systematic review and meta-analysis. Dementia and Geriatric Cognitive Disorders, 51, 203–213. 10.1159/000524780 [ DOI ] [ PubMed ] [ Google Scholar ] Stark S., Chernyshenko O. S., Drasgow F., Williams B. A. (2006). Examining assumptions about item responding in personality assessment: Should ideal point methods be considered for scale development and scoring?  Journal of Applied Psychology, 91, 25–39. 10.1037/0021-9010.91.1.25 [ DOI ] [ PubMed ] [ Google Scholar ] Van Haitsma K., Abbott K. M., Arbogast A., Bangerter L. R., Heid A. R., Behrens L. L., Madrigal C. (2020). A preference-based model of care: An integrative theoretical model of the role of preferences in person-centered care. The Gerontologist, 60, 376–384. 10.1093/geront/gnz075 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Walter E., Pinquart M. (2019). How effective are dementia caregiver interventions? An updated comprehensive meta-analysis. The Gerontologist, 60, 609–619. 10.1093/geront/gnz118 [ DOI ] [ Google Scholar ] Wehrmann H., Michalowsky B., Lepper S., Mohr W., Raedke A., Hoffmann W. (2021). Priorities and preferences of people living with dementia or cognitive impairment–A systematic review. Patient Preference and Adherence, 15, 2793–2807. 10.2147/PPA.S333923 [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Zimmerman S., Fazio S. (2022). Measurement to improve care and outcomes for persons with Alzheimer’s disease and dementia. Alzheimer’s & Dementia, 8, e12359. 10.1002/trc2.12359 [ DOI ] [ Google Scholar ] Zimmerman S., Mast B., Port C., VanHaitsma K., Fazio S. (2023). Development and testing of the PC-MET (Person-Centered Measure Evaluation Tool). Innovation in Aging, 7(Supplement_1), 296–297. 10.1093/geroni/igad104.0989 [ DOI ] [ Google Scholar ] Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Supplementary Materials igag006_Supplementary_Data igag006_supplementary_data.zip (208.4KB, zip) Data Availability Statement The data are not currently available for the privacy of the individuals who participated in the study. For questions about the data, contact the corresponding and lead author (A.N.L.). Articles from Innovation in Aging are provided here courtesy of Oxford University Press ACTIONS View on publisher site PDF (924.0 KB) Cite Collections Permalink PERMALINK Copy RESOURCES Similar articles Cited by other articles Links to NCBI Databases Cite Copy Download .nbib .nbib Format: AMA APA MLA NLM Add to Collections Create a new collection Add to an existing collection Name your collection * Choose a collection Unable to load your collection due to an error Please try again Add Cancel Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top