This Article Abstract Full Text (PDF) Services Download to citation manager PubMed PubMed Citation

Caregiving in the Presence of Chronic Pain

¹ Centre for Education and Research on Ageing, Concord Hospital, University of Sydney, Australia.
² Pain Management and Research Institute, Royal North Shore Hospital, University of Sydney, Australia.
³ School of Public Health, University of Sydney, Australia.
⁴ ABC Consulting, Croydon New South Wales, Australia.

Address correspondence to Fiona M. Blyth, PhD, Centre for Education and Research on Ageing, Concord Hospital, Concord, NSW 2139, Australia. E-mail: fblyth{at}med.usyd.edu.au

	Abstract

Top Abstract Methods Results Discussion References

 
Background. Chronic pain sufferers and caregivers share the risk of higher levels of psychological distress and adverse effects on well-being. This study examined the joint impact of chronic pain and primary caregiving on older people.

Methods. Data came from the New South Wales (NSW) Older People's Health Survey 1999, a state-wide general health survey of over 9000 NSW residents 65 years old or older. Using survey logistic regression modeling, we examined the relationship between chronic pain with different levels of disability, caregiving status, self-reported physical functioning, and two dependent variables—poor/fair self-rated health and psychological distress.

Results. Caregivers with chronic pain reported more psychological distress and poorer self-rated health than caregivers without pain, when both were compared to noncaregivers without pain (age-adjusted and sex-adjusted odds ratios [ORs] for caregivers with pain were 3.4 and 2.8, respectively, both p <.001). Caregivers with pain and noncaregivers with pain had similar patterns of results. Physical function significantly declined for both caregivers and noncaregivers with pain when compared with noncaregivers without pain.

Conclusions. Older people coping with caregiving and chronic pain are a potentially vulnerable group. Chronic pain status should be ascertained in older people who are caregivers, with particular attention to the issue of caregiver psychological distress and physical well-being.

Key Words: Caregivers • Chronic pain • Well-being

Older caregivers may be particularly at risk of adverse psychological and physical health states, including pain problems (5). To our knowledge, there are no reports in the pain or caregiving literature that explore the health effects of having chronic pain and being an older caregiver. With aging populations, demand for informal caregiving by older people is likely to increase. Minimizing pain-related disability should become a priority if chronic pain adversely affects the ability to provide care. Conversely, to optimize pain management in older people, it may be important to know about caregiving responsibilities.

In this study we explore the joint impact of chronic pain and primary caregiving on older people by testing the following hypotheses:

1. That primary caregivers who have chronic pain report higher levels of psychological distress and poorer self-rated health than do caregivers without chronic pain,
   
2. That primary caregivers who have chronic pain and pain-related interference with activities have higher levels of psychological distress and poorer self-rated health than do caregivers without chronic pain or caregivers with chronic pain that does not interfere with activities, and
   
3. That among primary caregivers, poorer self-rated physical functioning is associated with pain status and self-rated health status.
   

We have also included comparative data on participants with chronic pain who were not caregivers, to provide context for the findings for caregivers with chronic pain.

	METHODS

Top Abstract Methods Results Discussion References

 
Data for this analysis come from the New South Wales Older People's Health Survey 1999 (NSW OPHS 1999), a state-wide general health survey of over 9000 community-dwelling NSW residents 65 years old or older that was conducted by the NSW Health Department in 1999–2000. Comprehensive details of the methods and overall findings of this survey have been published elsewhere (6). The target sample comprised at least 500 residents 65 years old or older from each of the 17 geographically defined health administration areas in NSW in 1999. Households were sampled from electronic telephone listings that were geocoded and assigned to health areas. Introductory letters were sent before calling. One eligible respondent was randomly sampled from each household. Standardized telephone interviews were conducted by trained interviewers using computer-assisted telephone interview (CATI) technology. This article is based on data from 8881 persons who were able to respond to the interview themselves, and excludes the 537 participants interviewed by proxy.

Questions
Caregiver status.-- Caregiver status was ascertained with the question "Do you have the main responsibility in caring for someone who has a long-term illness, disability or other problem?" with an optional prompt "Such a problem would prevent them from managing their household tasks of personal care independently."

Chronic pain status.-- Chronic pain was defined as pain experienced for 3 months or more in the 6 months prior to the interview, using standard criteria (7). It was ascertained with the question "I would like you to think back over the last 6 months, that is, since [CATI automatically identifies the month 6 months prior to interview] and remember any episodes of pain that lasted for three months or more. This means pain experienced on all of the days over a three month period since [month]. Have you had any pain like this?"

Pain-related disability was ascertained with the questions, "In the last 6 months, how much has pain interfered with your daily activities (not at all/a little bit/moderately/quite a lot/extremely) and "how many days in the last 6 months have you been kept from your usual activities because of pain?" For this analysis, the number of days were grouped into the three categories (0, 1–6, 7) used in the Chronic Pain Grade (8).

Psychological distress.-- Psychological distress was measured with the Kessler 6 (K6), a six-item questionnaire about the level of anxiety and depressive symptoms experienced in the 4 weeks prior to interview. The K6 is the short form of the K10 questionnaire, which was developed for use in the National Health Interview Survey and the World Health Organization (WHO) World Mental Health Surveys (9) and has been used in previous population studies of chronic pain (1,2). Each item has a five-level response scale indicating how often each problem was experienced in the reference period (all of the time/most of the time/some of the time/a little of the time/none of the time).

Scores were converted to a T score with a mean of 50 and a standard deviation of 10. Scores of 1 standard deviation above the mean (60) were chosen as the level indicating psychological distress. Using this criterion, the K6 identifies similar proportions of the general population as psychologically distressed as do other recognized measures that identify anxiety and depression, such as the General Health Questionnaire 12 (GHQ-12) (10).

Self-rated health.-- Respondents were asked, "In general, would you say your health is: excellent/very good/good/fair/poor?" Responses were grouped into fair/poor or excellent/very good/good. Self-rated health has been shown to have a strong biological basis with a graded relationship to common biomarkers, and to be a useful measure of physiologic states (11).

Restricted activity or recent hospitalization.-- Respondents were classified as having significantly restricted activity if they had either stayed in hospital for at least 1 night in the last 12 months or had spent more than a complete week in bed at home because of illness or accident. These are termed intervening events in the development of disability in older adults, and are strong predictors of functional decline in community-dwelling older adults (12,13). The current study used a longer recall period than is used in the standard version of this measure.

Physical functioning.-- The Physical Functioning Scale of the SF-36 (14) was used. The 20-item scale consists of questions about ability to perform various moderate and vigorous physical activities. From these, physical functioning scores were calculated. Scores range from 0 to 100, with higher scores indicating higher levels of functioning. Scores were entered as a continuous variable in all models.

Statistical Analysis
The survey sample was weighted to adjust for differences in selection probabilities among respondents, according to the number of eligible respondents and the number of household residential telephone connections. We performed unweighted analyses and analyses using poststratification weighting that adjusted for differences between the age and sex structure of the survey sample and the Australian Bureau of Statistics midyear population estimates (excluding people residing in institutions) for each health area. Both methods produced results that were not significantly different; hence, we report the results of the weighted descriptive and logistic regression analyses.

The Survey Means procedure in SAS version 8.2 (15) was used for the initial descriptive analyses. The procedure uses the Taylor expansion method to estimate sampling errors of estimators based on complex sample designs (in particular, unequal weighting). Differences in proportions of subgroups that were in the lowest quartile for physical functioning or the highest quartile for psychological distress were tested using the two proportions test.

Survey logistic regression modeling was performed using STATA release 9.0 (16). These models were weighted to adjust for differences in selection probabilities among respondents (arising from different strata and household sizes). Models were formulated for two dependent variables—poor or fair self-rated health and psychological distress. For each dependent variable, a series of models was formulated based on caregiver and chronic pain status, with three different definitions of pain status as follows:

Model A (Chronic pain overall): Chronic pain—present versus absent

Model B (Chronic pain categorized by degree of interference with activity): Chronic pain present and causing interference with daily activities: not at all or only a little bit; or causing interference with daily activities moderately, quite a bit, or extremely

Model C (Chronic pain categorized by days kept from activities in last 6 months by pain): Chronic pain present with no days, 1–6 days, or 7 days kept from activities in last 6 months by pain.

For each formulation of caregiver/chronic pain status, a hierarchical series of models was created adjusting for the following a priori–selected variables: age group and sex; age group, sex, and restricted activity; age group, sex, restricted activity, and self-reported physical functioning (SF-36 Physical Functioning Scale).

The same relationships between pain status and the two dependent variables poor/fair self-rated health and psychological distress were then tested separately in caregiver and noncaregiver subgroups in a hierarchical series of models. In these models, the referent category in each subgroup was respondents with no pain.

To further test the third hypothesis, we used unweighted regression to test models in which physical functioning was the dependent variable. Adjusted mean differences (least squares mean) in physical function (Physical Functioning Scale, SF-36) for each caregiver/pain status were calculated using Proc GLM in SAS version 8.2 for Windows (14). The models adjusted for age group, sex, restricted activity, and self-rated health as well as a Pain xby Self-rated health interaction. All models were assessed for goodness-of-fit.

	RESULTS

Top Abstract Methods Results Discussion References

 
The analyses reported here are based on 8881 survey respondents. The response rate was 70.7%. Nine percent (9.2%) of older people identified themselves as primary caregivers. Women were more likely to be primary caregivers in all age groups except the oldest (80 years). Care recipients were predominantly spouses (73.4%). In 85.5% of cases, the reason for care was physical illness or disability, including frailty and blindness; in 19.4%, memory problem or intellectual disability (including dementia) was the reason for care (NSW OPHS, 1999).

Overall, 19.2% of men and 23.3% of women 65 years old or older reported having chronic pain (Table 1). In almost all age and sex categories, more than three quarters of participants with chronic pain reported some degree of interference with activities due to pain. Prevalence of pain and pain-related interference was higher in women than in men.

View larger version (8K): [in this window] [in a new window]   Figure 1. SF-36 physical functioning scale: proportion in lowest quartile by pain and caregiver status. Pain/caregiver categories are not mutually exclusive. Percentages are weighted. Missing data: pain status (n = 29); caregiver status (n = 16)

View larger version (10K): [in this window] [in a new window]   Figure 2. Psychological distress (Kessler 6 [K6] score 60): proportion above the threshold for distress by pain and caregiver status. Pain/caregiver categories are not mutually exclusive. Percentages are weighted. Missing data: pain status (n = 29); caregiver status (n = 16)

View this table: [in this window] [in a new window]   Table 2. Univariate Odds Ratios (OR) for Pain/Caregiver Status, and Poor/Fair Self-Rated Health or Psychological Distress (Kessler 6 Score 60).

In Model C, chronic pain is categorized by number of days of interference with activities. Compared with participants who were not caregivers and did not report chronic pain, caregivers with chronic pain who reported 1–6 days or 7 days of interference with activities were significantly more likely to report poor/fair self-rated health (age-adjusted and sex-adjusted ORs 3.3, 95% CI, 1.9–5.8 and 13.6, 95% CI, 4.8–38.5, respectively). Adjusting for restricted activity and physical functioning attenuated this association, most notably in caregivers with 7 days of interference (adjusted OR 3.5, 95% CI, 0.8–14.4). Note that this subgroup contained only 30 participants. For noncaregivers with chronic pain, there was also a strong association between number of days of interference with activities and self-rated health. This association was strongest in participants reporting 7 days of interference (age-adjusted and sex-adjusted OR 5.2, 95% CI, 3.8–7.2). Adjusting for restricted activity and physical functioning reduced the OR (1.6, 95% CI, 1.1–2.2).

Table 4 shows a similar hierarchical series of logistic regression models, with psychological distress as the dependent variable. In Model A, for caregivers with chronic pain, there was a strong relationship with poor/fair self-rated health in the age-adjusted and sex-adjusted model (adjusted OR 3.4, 95% CI, 2.6–3.6), which was somewhat reduced when restricted activity and physical functioning were added to the model (adjusted OR 1.7, 95% CI, 1.2–2.3); the effect of adjusting for physical functioning was smaller than in the corresponding model in Table 3. In comparison, caregivers without chronic pain were less likely to report psychological distress (age-adjusted and sex-adjusted OR 1.5, 95% CI, 1.1–2.2), and further adjustments did not change this association. There was also a strong relationship with psychological distress in the age-adjusted and sex-adjusted model for those who were not caregivers and who experienced chronic pain (OR 3.1, 95% CI, 2.6–3.6), and additional adjustment for restricted activity and physical functioning produced a similar effect.

In Model C, compared with participants who were not caregivers and did not report chronic pain, caregivers with chronic pain who reported 1–6 days or 7 days of interference with activities were significantly more likely to report psychological distress (age-adjusted and sex-adjusted ORs 4.7, 95% CI, 2.6–8.5 and 6.1, 95% CI, 2.6–14.3, respectively). As in Table 3, adjusting for restricted activity and physical functioning had the largest effect in the small subgroup of caregivers with 7 days of interference (adjusted OR 2.1, 95% CI, 0.7–5.8). For noncaregivers with chronic pain, there was a similar association between reporting 1–6 days or 7 days of interference with activities and psychological distress (age-adjusted and sex-adjusted ORs 4.9, 95% CI, 3.9–6.2 and 4.7, 95% CI, 3.4–6.4, respectively). Adjusting for restricted activity and physical functioning reduced OR values in both these groups; however, the relationship with psychological distress remained statistically significant (for example, the adjusted OR for noncaregivers with 7 days interference was 1.9, 95% CI, 1.3–2.6).

Adding an interaction variable (Physical functioning x Pain status) confirmed that there was a significant interaction between greater severity and lower physical functioning in these models (data not shown). We also found that the same relationships between pain status and the two dependent variables (poor/fair self-rated health and psychological distress) were maintained when tested separately in caregiver and noncaregiver subsets (data not shown). Similar patterns of changes in OR values were seen with adjustment for age group, sex, restricted activity, and physical functioning.

The third hypothesis was tested in linear regression analysis adjusting for age, sex, comorbidity and self-rated health, with physical functioning as the dependent variable. Physical function significantly declined for both caregivers noncaregivers with pain when compared with noncaregivers without pain. This effect was greater for the caregiver group, regardless of pain status. However, caregivers with pain had on average lower scores. In particular, caregivers (OR 69.7, 95% CI, 67.5–72.0) and noncaregivers (OR 67.0, 95% CI, 65.8–68.2) without pain, on average, had significantly (p <.01) higher physical function scores than caregivers with pain (OR 55.4, 95% CI, 52.4–58.5) and noncaregivers with pain (OR 51.8, 95% CI, 50.4–53.3). This trend was evident for all definitions of pain. In addition, the significant (p <.005) Pain x Self-rated health interaction in all models further supported this association.

	DISCUSSION

Top Abstract Methods Results Discussion References

 
Our results showed that older primary caregivers who had chronic pain reported poorer self-rated health than did caregivers without pain. Modeling revealed that this relationship was partly mediated by self-reported physical functioning status. The explanatory power of self-reported physical functioning in the models increased as the level of pain-related disability increased. Caregivers with moderate to severe levels of pain interference were more than three times more likely to report poor/fair self-rated health than were caregivers without chronic pain, when both were compared to a common referent category (noncaregivers without pain). However, there was little difference between caregivers with lower or higher levels of severity of interference. Caregivers reporting any days of interference with activities were much more likely than caregivers without chronic pain to report poor/fair self-rated health when compared to noncaregivers without pain. Although OR values increased with the number of days reported, the small numbers and resulting wide CI values mean that an incremental relationship is not proven. The results for noncaregivers with pain and poor/fair self-rated health were generally similar to those for caregivers with pain, except that there was a relationship between severity of interference and self-reported health. There was some suggestion that caregivers reporting 7 days of interference reported worse health than noncaregivers with the same amount of interference, but this would need confirmation with a larger sample size.

In overall terms, both caregivers with chronic pain and noncaregivers with chronic pain who reported moderate to severe interference with activities were around four times as likely to report higher levels of distress compared to noncaregivers without pain. Caregivers without chronic pain were also more likely to report higher distress, but not as much as the two chronic pain groups. There was no clear relationship between level of severity of interference and psychological distress in the caregiver with chronic pain group. Caregivers reporting any days of interference with activities were more likely than caregivers without chronic pain to report psychological distress when compared to noncaregivers without pain. Again, the small numbers in the analysis mean that an incremental relationship is not proven. In noncaregivers with chronic pain, there was an association between level of severity of interference and distress; however, once again the results for days of interference were equivocal.

The extent to which the findings of this study would be generalizable to other countries will depend on the extent to which the relationships demonstrated between caregiving, chronic pain, and other aspects of health are biologically determined. It is likely that societal and cultural factors that vary between countries, such as the nature and availability of health care, will also be important in this regard. In addition, the nature and severity of the care recipient's health problems is likely to be an important contributor to caregiver burden, and could not be explored in the current study. Because of the way in which caregiver status was ascertained in this study, it is possible that the care recipients in this study were a more heterogeneous group than those in studies based on samples seeking assistance with caregiving.

Older age combined with the stress and physical demands of caregiving has been identified as potentially compromising to physiological functioning in at-risk groups of older caregivers (5). In a population-based study, caregiver mental or emotional strain was identified as an independent risk factor for mortality in older adults (17). The literature on the relationship between caregiving and physical health is smaller and more equivocal than the literature on caregiving and psychological health (18,19). This is reflected in the paucity of caregiver intervention studies that address physical functioning (3,4,19).

Similarly, chronic pain is characterized by a complex relationship with psychological factors, and also a decline in physical functioning in older people (20,21). Based on these findings in the literature, older people coping with the difficulties of both caregiving and chronic pain could be seen as a vulnerable group. Our results suggest a different view: Caregivers who have chronic pain experience worse self-rated health and more distress than caregivers without chronic pain. However, they do not experience worse self-rated health or more distress than do older people with chronic pain who are not caregivers.

To our knowledge, this is the first study to explore the relationship between chronic pain status and caregiving in a general population setting. This avoids two of the limitations of using other sampling frames such as caregivers who are seeking help with their caring role. First, it reduces the problem of disentangling factors associated with help-seeking behavior from factors associated with caregiving. Second, it should not be assumed that caregivers who are not seeking help do not need help or are not affected by their caregiving role. Population-based sampling methods capture both groups, as well as provide the opportunity to compare caregivers with noncaregivers—a limitation acknowledged in studies with no comparison group (22). Few population-based studies of caregivers do more than describe the prevalence of caregiving.

Study limitations must be acknowledged. First, as the study is cross-sectional, the causal direction of relationships between caregiving, chronic pain status, psychological distress, and self-rated health cannot be determined. However, we believe that it is likely that the relationships are to some extent bidirectional. Future studies would benefit from objective measures of physical functioning rather than reliance on self-reported measures. Sources of psychological distress were not ascertained, and the inclusion of specific measures of caregiver stress and distress associated with chronic pain would be important in future studies. The recall period differed significantly for some of the variables of interest in this study, leaving additional uncertainty about the temporal relationships between them. We used a measure of the consequences of comorbidities (restricted activities and hospitalizations) that captures the aspect of comorbidities of most interest in the current study—compromised health and future risk of disability—more specifically than a checklist of common chronic conditions. However, the recall period used in this study was 12 months, and the accuracy of recall for these events over such an extended period is not known. The impact of comorbidity, either pain-related or as a measure of general disease burden, was not able to be explored in this study.

The results of this study have implications for healthy aging and societal reliance on informal caregiving. As there are effective interventions for chronic pain, systematic assessment and intervention are important in this group, with potential benefits for both the caregiver and the care recipient. However, the demands of caregiving can be considerable barriers to interventions aimed at improving physical fitness and activity levels. Recent management guidelines for health professionals fail to address the physical health of caregivers (22), despite the possibility for improving mental health by lessening physical disability. The results of this study add to this call for more focus on this issue.

	Acknowledgments

Top Abstract Methods Results Discussion References

 
This work was supported by the Australian National Health and Medical Research Council (Capacity Building Grant in Public Health No. 262120; and Grants No. 974159 and 991407) and by research grants from the Joint Coal Board, Workcover Australia, Insurance Australia Group, and the Pain Management Research Institute Ltd.

We thank the Chief Health Officer, NSW Department of Health, for giving permission to access and use the 1999 Older Persons Health Survey data. The Older Persons Health Survey was conducted by the NSW Health Survey Program in the Centre for Epidemiology and Research of the NSW Department of Health and was funded by the Department.

	Footnotes

Top Abstract Methods Results Discussion References

 
Decision Editor: Darryl Wieland, PhD, MPH

Received September 21, 2006

Accepted June 28, 2007

	References

Top Abstract Methods Results Discussion References

 

1. Blyth FM, March LM, Brnabic AJM, Jorm LR, Williamson M, Cousins MJ. Chronic pain in Australia: a prevalence study. Pain. 2001;89:127-134.[Medline]
2. Blyth FM, March LM, Cousins MJ. Chronic pain-related disability and use of analgesia and health services in a Sydney community. Med J Aust. 2003;179:84-87.[Medline]
3. Schulz R, Patterson TL. Caregiving in geriatric psychiatry. Am J Geriatr Psychiatry. 2004;12:234-237.[Medline]
4. King AC, Baumann K, O'Sullivan P, Wilcox S, Castro C. Effects of moderate-intensity exercise on physiological, behavioural, and emotional responses to family caregiving: a randomized controlled trial. J Gerontol A Biol Sci Med Sci. 2002;57A:M26-M36.[Abstract/Free Full Text]
5. Kiecolt-Glaser JK, Glaser R. Chronic stress and mortality among older adults. JAMA. 1999;282:2259-2260.[Free Full Text]
6. Public Health Division. New South Wales Older People's Health Survey 1999., Sydney: NSW Health Department; 2000.
7. International Association for the Study of Pain Subcommittee on Taxonomy. Classification of chronic pain. Descriptions of chronic pain syndromes and definitions of pain terms. Pain. 1986;Suppl 3:S1-S226.
8. Von Korff M, Ormel J, Keefe F, Dworkin SF. Grading the severity of chronic pain. Pain. 1992;50:133-149.[Medline]
9. Kessler RC, Andrews G, Colpe LJ, et al. Short screening scales to monitor population prevalences and trends in non-specific psychological distress. Psychol Med. 2002;32:959-976.[Medline]
10. Furukawa TA, Kessler RC, Slade T, Andrews G. The performance of the K6 and K10 screening scales for psychological distress in the Australian National Survey of Mental Health and Well-Being. Psychol Med. 2003;33:357-362.[Medline]
11. Jyllhä M, Volpato S, Guralnik JM. Self-rated health shows a graded association with frequently used biomarkers in a large population sample. J Clin Epidemiol. 2006;59:465-471.[Medline]
12. Gill TM, Allore H, Guo Z. The deleterious effect of bed rest among community-living older persons. J Gerontol A Biol Sci Med Sci. 2004;59A:755-761.[Medline]
13. Gill TM, Allore H, Holford TR, Guo Z. Hospitalization, restricted activity, and the development of disability among older persons. JAMA. 2004;292:2115-2124.[Abstract/Free Full Text]
14. Ware JE, Snow KK, Kosinski M, Gandek B. SF 36 Health Survey. Manual and Interpretation Guide. Boston, MA: The Health Institute; 1993.
15. SAS Version 8.2 for Windows. Cary, NC: SAS Institute; 1999–2001.
16. StataCorp. 2005. Stata Statistical Software: Release 9.2. College Station, TX: StataCorp LP.
17. Schulz R, O'Brien AT, Bookwala J, et al. Psychiatric and physical morbidity effects of dementia caregiving: prevalence, correlates and causes. Gerontologist. 1995;35:771-791.[Abstract]
18. Schulz R, O'Brien A, Czaja S, et al. Dementia caregiver intervention research: in search of clinical significance. Gerontologist. 2002;42:589-602.[Abstract/Free Full Text]
19. Schulz R, Beach SR. Caregiving as a risk factor for mortality: the Caregiver Health Effects Study. JAMA. 1999;282:2215-2219.[Abstract/Free Full Text]
20. Reid MC, Williams CS, Gill TM. Back pain and decline in lower extremity physical function among community-dwelling older persons. J Gerontol A Biol Sci Med Sci. 2005;60A:793-797.[Abstract/Free Full Text]
21. Onder G, Cesari M, Russo A, Zamboni V, Bernabei R, Landi F. Association between daily pain and physical function among old-old adults living in the community: results from the ilSIRENTE study. Pain. 2006;121:53-59.[Medline]
22. Bruce DG, Paley GA, Nichols P, Roberts D, Underwood PJ, Schaper F. Physical disability contributes to caregiver stress in dementia caregivers. J Gerontol A Biol Sci Med Sci. 2005;60A:345-349.[Abstract/Free Full Text]

This Article Abstract Full Text (PDF) Services Download to citation manager PubMed PubMed Citation