We conducted cross-sectional analyses using data from a nationwide survey of dementia family caregivers with chronic pain. Given limited available data on dementia family caregivers’ pain, we fielded a survey that focused specifically on pain's impact on caregivers’ health and caregiving outcomes. We partnered with Qualtrics to recruit the sample; Qualtrics maintains a proprietary service to garner online samples of research participants (i.e., market-research panels), which has been successfully used in behavioral science research (11).

In partnership with the Qualtrics team, we determined survey structure (e.g., length of completion) and sampling strategy (e.g., oversampling certain populations). We determined the target sample size of ∼250 caregivers via both statistical calculations of planned analyses and consideration of budgetary constraints. The sample was geographically representative based on U.S. census data, and we established a quota of at least 80 Black or Hispanic caregivers to support racial diversity in the sample given our interest in analyzing racial and ethnic differences. Additional details about the study design have been published previously (12).

Qualtrics sourced and recruited participants from pre-existing survey panels that they managed separately from our team. They determined the appropriate sampling structure based on our recruitment goals and completely managed the overall data collection process. Potentially eligible participants received an electronic invitation directly from Qualtrics to complete the survey, which included three screening questions to confirm eligibility. Eligibility criteria were as follows: caregiver experienced pain, aching, burning, or throbbing sensations on most days of every month for the past 3 months; caregiver's pain was not due to cancer; caregiver provided 20 or more hours of care per week to a relative/family member or friend with Alzheimer's disease or a related dementia.

Qualtrics fielded the survey on March 6th, 2024, and the survey closed on April 11th, 2024. The survey was designed to last no more than 15 min. Qualtrics delivered compensation to participants directly, which varied based on the participants’ pre-existing agreement with Qualtrics but averaged $7.50 per survey. Weill Cornell Medicine's Institutional Review Board deemed this study exempt from human subjects review given data were anonymous.

We first determined the number of caregivers who helped their care recipient with one or more activities and, among those caregivers, the number who experienced difficulty performing one or more activities. Among the caregivers who reported difficulty, we computed the mean rating of how much pain contributed to that difficulty across all individual BADLs and IADLs. We also created a composite score representing each caregiver's mean pain-attributed difficulty across all five BADLs and all five IADLs. We then visually reviewed histograms and Q-Q plots of the composite pain-attributed BADL difficulty variable and the composite pain-attributed IADL difficulty variable and determined both were reasonably normally distributed and thus that parametric tests were appropriate to answer our research questions (visuals of histograms and Q-Q plots are available in Supplementary Materials S1). Thus, to assess for associations between pain-related difficulty with care tasks and caregiver demographics and caregiving characteristics, we computed an Analysis of Variance (ANOVA) model (via SAS PROC GLM with estimate and contrast commands) to determine between-group differences in the extent to which caregivers attributed difficulty with BADLs and IADLs to their pain. We computed one model, such that each estimate is adjusted for all other variables, and we used Scheffe's method to account for the testing of multiple comparisons in our model.

A total of 4,400 people responded to Qualtrics’ invitation to complete the survey, the vast majority of which (4,099) were ineligible because they did not meet eligibility criteria. We removed additional participants’ responses due to data quality issues (e.g., straight lining) or missing data, resulting in a final sample of 269 dementia family caregivers living with chronic pain (Figure 1).

Caregivers ranged in age from 19 to 83, with an average age of 45 years (SD = 12 years). They provided care for an average of 66 h per week (SD = 49 h; Range = 20–168 h). Additional sample characteristics are listed in Table 1.

Of the 269 caregivers, 205 helped their care recipient with at least one BADL. Of those 205 caregivers, 127 (62%) reported difficulty helping with at least one BADL (Figure 1). Among caregivers reporting difficulty helping with at least one BADL, the average number of BADLs with which caregivers had difficulty was 2.65 (SD 1.51). When caregivers who reported difficulty helping with at least one BADL were asked about the extent to which pain contributes to the difficulty participants experienced, caregivers’ average response was 6.81 (SD 1.97) on the 0–10 scale. All caregivers (100%) attributed at least some of their difficulty to pain [i.e., no caregivers reported 0 (not at all)]. The average ratings for the extent to which pain contributed to their difficulty, for each specific BADL and for the summed BADLs composite score, are shown in Table 2.

Similarly, of the 269 caregivers, 263 helped their care recipient with at least one IADL. Of those 263 caregivers, 164 (62%) reported difficulty with at least one IADL (Figure 1). Among caregivers reporting difficulty helping their care recipient with at least one IADL, the average number of IADLs with which caregivers had difficulty was 2.90 (SD 1.45). When caregivers who reported difficulty helping with at least one IADL were asked about the extent to which pain contributes to that difficulty, caregivers’ average response was 6.49 (SD 2.02) on the 0–10 scale. Nearly all caregivers (99%) attributed at least some of their difficulty to pain [only two caregivers reported 0 (not at all)]. The data showing the extent to which pain contributed to participants’ difficulty, on average, for each specific IADL and for the summed IADLs composite score, appear in Table 2.

Compared to White caregivers, Black caregivers attributed less of their difficulty with BADLs to pain (estimate = –1.17, p = 0.04). Compared to caregivers who had been providing care for less than a year, caregivers who had been providing care for 1–2 years attributed less of their difficulty with BADLs to pain (estimate = –1.22, p = 0.05). There were no other significant differences in pain-attributed difficulty (Table 3). The statistical model for these estimates controlled for all other caregiver demographics and caregiving characteristics.

To the best of our knowledge, this is the first study to determine the extent to which dementia family caregivers with chronic pain attribute caregiving difficulty to their pain. Its nationwide reach and oversampling of Black and Hispanic caregivers are unique strengths, as is its probing of specific care tasks rather than measuring caregiving difficulty in general. Still, there are key limitations. Namely, our partnership with Qualtrics to garner a sample may have led to selection bias where certain caregivers were more likely to be recruited and to agree to complete the survey. For example, participants who willingly compete online surveys might be younger and, indeed, our sample skews younger. Though middle-aged adult-child caregivers make up a substantial proportion of dementia family caregivers across the United States, they are overrepresented in this sample. Given older caregivers are more likely to report activity-limiting pain (6), it is possible that a sample composed entirely of older caregivers would reveal even stronger pain-attributed difficulty with care tasks.

Additionally, we did not inquire about other physical (e.g., heart disease, obesity), emotional (e.g., depression, burnout) and care-recipient (e.g., dementia-related behaviors) factors that might contribute to caregiving difficulty. However, a foundational study on women's disability prompted women to report on physical symptoms that participants perceived to be the cause of their difficulty performing various activities of daily living. The study found that musculoskeletal pain was by far the most common cause of disability, even when participants were prompted with other alternative symptoms such as fatigue and balance difficulties (15). In the current study, prompting caregivers to attribute their difficulty to multiple factors, including but not limited to pain, would have offered a more complete picture of the relative role pain plays with respect to caregiving difficulty. However, the strong attribution of pain as a cause of disability in other studies as well as in the current study leads us to speculate that pain contributes significantly to the difficulty many caregivers experience helping care recipients complete various care tasks. Nonetheless, future research should ascertain factors besides pain, including other symptoms from pain (e.g., fatigue, emotional distress) that contribute to caregiver difficulty.