Low back pain (LBP) in occupational settings has been studied extensively. There are fewer studies on LBP in domestic settings, especially in an informal caregiving setting.
To compare the prevalence of LBP in adult female primary caregivers of children with physical disabilities who need assistance with transfers (eg, moving from a bed to a wheelchair) with the prevalence of LBP in adult female primary caregivers of children with nondisabling medical illnesses and to evaluate the factors associated with LBP.
A 15-minute, self-administered, cross-sectional survey.
Ninety consecutive adult female caregivers of children presenting to a pediatric physical medicine and rehabilitation clinic and 23 consecutive adult female caregivers of children presenting to a pediatric endocrine clinic were studied. General exclusionary criteria included the following: male sex, a history of back surgery or fracture, the caregiver was younger than 18 years old at the time of the completion of the questionnaire, or the caregiver was caring for any child younger than 2 years old. Caregivers visiting the pediatric endocrine clinic were excluded if they were caring for 1 or more children needing assistance with transfers.
Main Outcome Measures
The dependent variable was the presence of LBP. The independent variables were mood, work status, amount of lifting at work, physical functioning of the child, demographic variables of the caregiver, and demographic variables of the child.
The prevalence of having LBP (71.1%) in the physical medicine and rehabilitation group is higher than the prevalence (43.5%) in the endocrine clinic group (odds ratio, 3.2; 95% confidence interval, 1.25-8.21). The prevalence of having LBP (80.3%) when the child required physical assistance with transfers was significantly higher than the prevalence (40.5%) when the child did not require physical assistance with transfers (odds ratio, 2.56; 95% confidence interval, 2.56-14.0). Forward multiple logistic regression showed that the factors related to LBP in the caregiver were the transferability of the child, mood of the caregiver, and a history of LBP in the caregiver.
The prevalence of LBP is higher in caregivers of children needing assistance with transfers. This increased prevalence is associated with the transferability of the child and mood of the caregiver. Results of this study suggest that physical and psychological factors both contribute to the presence of nonoccupational LBP.
LOW BACK pain (LBP) in adults is a common cause of disability and financial loss. It has been estimated that the lifetime prevalence of LBP exceeds 70% and the 1-year prevalence rate is 25% to 45%.1,2 Approximately 1% (2.4 million) of the US population is chronically disabled and another 1% is temporarily disabled by LBP.1 The annual costs of LBP disability have been estimated to be about $50 billion, with the average medical and work-related cost of a single case of work-related LBP exceeding $8000.3
The risk factors for LBP have been extensively studied in the occupational setting. In nursing, risk factors include heavy lifting,4,5 poor fitness,5 and improper lifting techniques.5,6 In industry, risk factors identified include heavy physical work,7,8 frequent bending and lifting,1 repetitive work,9 vibration exposure,10 and psychosocial factors including depression.11- 14 Two studies have evaluated the prevalence of LBP with parenthood.15,16 Finkelstein15evaluated police officers and fire fighters with LBP; the results noted a higher prevalence of LBP when the civil servant had a greater number of children. The second study by Rossignol et al16 primarily looked for occupational-related LBP risk factors in aircraft assemblers and noted that being a parent was associated with an increased prevalence of self-reported work impairment. Neither study evaluated whether having a child with a physical disability was a factor for LBP.
A search of the literature reveals no studies specifically evaluating the prevalence of LBP in caregivers of children with physical disabilities. This is a unique population to examine because the caregivers studied are similar in age to that of the working population, but the heavy lifting is required at home.
The primary purpose of this study was to compare the prevalence of LBP in adult female primary caregivers of children with physical disabilities who need assistance with transfers (eg, moving from a bed to a wheelchair) with the prevalence of LBP in adult female primary caregivers of children with nondisabling medical illnesses. The secondary purpose was to identify risk factors in adult female primary caregivers of children associated with the prevalence of LBP.
Study subjects were recruited from both a university-affiliated pediatric physical medicine and rehabilitation (PMR) clinic and from a university-affiliated pediatric endocrine clinic in November 1, 2000, through September 2, 2001, at the same hospital. Consecutive adult female primary caregivers were approached to see if they were eligible and willing to participate in the study. Informed consent was obtained face-to-face from the subjects and documented at the clinic visits prior to administration of the survey. After the consent was obtained, eligible and willing participants filled out a 15-minute, self-administered survey before leaving the respective clinic to prevent dropouts. Institutional review board approval was obtained prior to starting the study.
Caregivers were excluded from participation in the study if they had a history of back surgery or fracture, were male, were younger than 18 years old at the time of questionnaire completion, or had any child in the family younger than 2 years old. Having a child younger than 2 years old was an exclusionary criterion because most children younger than 2 years old need to be carried. Caregivers who were visiting the pediatric endocrine clinic were also excluded if they were caring for 1 or more children needing assistance with transfers. Caregivers who were attending the pediatric PMR clinic were also excluded if they were caring for a child with diabetes mellitus or caring for more than 1 child with a physical disability. It was decided to exclude males who were caring for more than 1 child with a physical disability because these groups constitute less than 10% of the parents who attend these clinics, and the size of these groups would be statistically too small to evaluate.
Ninety consecutive adult female primary caregivers in the pediatric PMR clinic participated; 6 eligible subjects who were approached refused to participate because of time commitments. Twenty-three consecutive adult female primary caregivers in the pediatric endocrine clinic participated (recruitment started later and was slower in this clinic); 2 eligible subjects who were approached refused to participate because of time commitments. The demographics of all 113 subjects are listed in Table 1.
A 15-minute, self-administered survey was provided to each subject during a clinic visit. This survey included child demographic variables (eg, age, sex, height, weight, and number of children in the family), caregiver demographic variables (eg, race, height, weight, and marital status), and the following standardized questionnaires:
The Standardised Nordic questionnaire17 is a series of 8 questions concerning whether the subject has LBP, the severity of LBP, and the frequency of LBP. The first question regarding whether the subject has LBP was modified into 2 questions asking if the subject had LBP prior to the birth of the clinic-enrolled child and/or after the birth of the clinic-enrolled child. This change was evaluated for test-retest reliability. Two other specific questions used in data analysis included the following: "Have you ever been hospitalized because of low back trouble?" and "What is the total length of time that low back trouble has prevented you from doing your normal work (at home or away from home) during the last 12 months?"
Mood was measured using the Center for Epidemiologic Studies–Depression Scale. This scale is composed of 20 questions using a 4-category Likert scale and the total score (0-60) is the sum of each individual question score.18,19 It is valid among persons with physical disabilities20 and has concurrent validity with the Diagnostic and Statistical Manual of Mental Disorders, Revised Third Edition diagnoses of depression with a cut-off score of 19.21 The subjects rated the amount of physical lifting done per day according to the US Department of Labor Classification (1 indicates sedentary; 4, heavy work).22
Physical functioning of the child was measured by the ability of the child to transfer using the "Mobility" section of the WeeFIM instrument (WeeFIM System; Uniform Data Systems for Medical Rehabilitation, Amherst, NY). The WeeFIM instrument scale is a 7-point ordinal scale ranging from 1 (needing total assistance with self-transfer) to 7 (complete independence with self-transfer). The transfers evaluated are chair-wheelchair, toilet, and tub-shower. Locomotion (walk-wheelchair) and the ability to ascend or descend stairs are also included. This instrument has been validated and has also been shown to have good interrater reliability.23,24
Body mass index, calculated as weight in kilograms divided by the square of height in meters, used the subjects' reported heights and weights. The children's heights and weights were measured in their respective clinic.
These data were analyzed using SPSS Version 10.1 (SPSS Inc, Chicago, Ill) and SAS Version 8.0 (SAS Institute Inc, Cary, NC). To determine if the demographic variables are associated with the prevalence of LBP, an odds ratio was calculated with univariate logistic regression using the presence of LBP as the dependent variable.
An odds ratio (OR) was calculated with univariate logistic regression to determine if the prevalence of LBP in adult female primary caregivers in the pediatric PMR clinic group was different from that of the adult female primary caregivers in the pediatric endocrine clinic group.25 This was also used to determine if the prevalence of LBP in the subjects whose child needed help with transfers (WeeFIM instrument score ≤4) was different than that of the subjects whose child did not need help with transfers (WeeFIM instrument score ≥5). This analysis comparing the group whose children needed help with transfers with the group whose children did not need help with transfers was repeated in the pediatric PMR subjects to examine this relationship in the subgroup of primary caregivers of children with physical disabilities.
Multivariate logistic regression using forward selection was used to determine which variables are significantly associated with LBP in a multivariate model. For each model, the ability to discriminate between caregivers with and without LBP was quantified using the area under the receiver operating characteristic (ROC) curve. The ROC curve area is a suitable parameter to summarize the discriminative or predictive value and can range from 0.5 (no discrimination) to 1.0 (perfect discrimination).26
The results of the primary hypothesis are given in Table 2. The percentage of 89 subjects (71.1%) in the pediatric PMR clinic with LBP was significantly higher than that of the 23 subjects (43.5%) in the pediatric endocrine clinic with an OR of 3.2 (95% confidence interval, 1.25-8.21). The prevalence of LBP in subjects whose children needed physical assistance with transfers (80.3%) was also higher than the prevalence of LBP in subjects whose children did not need assistance with transfers (40.5%) with an OR of 6.0 (95% confidence interval, 2.56-14.0). A similar relationship was found with a subgroup analysis of the pediatric PMR clinic comparing adult female primary caregivers whose children needed assistance with transfers to adult female primary caregivers in the pediatric PMR clinic whose children did not need assistance with transfers (Table 2).
Table 3 lists the results of the univariate analyses of the pooled data comparing the prevalence of LBP with each potential predictor variable. A history of LBP prior to starting to care for the physically disabled child was associated with having LBP after the physically disabled child was born. The other factors associated with increased prevalence of LBP include decreased height of the child (OR, 0.16), decreased weight of the child (OR, 0.97), and increased depression in the caregiver (OR, 1.09). Since these predictor variables are continuous, the association should be interpreted as the change in odds per unit change in the predictor variable. The prevalence of LBP was not associated with race (P = .99), marital status of the caregiver (P = .42), relationship of the caregiver to the child (P = .97), diagnosis of the child's condition (P = .46), work status of the caregiver (P = .32), or the number of children in the family (P = .16) (None of these data are given in Table 3.)
Table 4 summarizes the results of the model selection. Model 1 used only a history of LBP. This factor was chosen for the first model since it was present before caregiving had begun for the child seen in the clinic, and it must be controlled for to determine more accurately how much the other factors contribute to the presence of LBP. The ROC curve area of model 1 was 0.59. The other demographic variables, including the child's height and weight, did not significantly add to the model. In model 2, the addition of the physical functioning of the child increased the ROC curve area to 0.75. Model 3 added the Center for Epidemiologic Studies–Depression Scale score of the caregiver and the ROC curve area increased to 0.82. In model 3, the presence of LBP in the caregiver was directly associated with a history of LBP (OR, 6.53) and mood of the caregiver (OR, 1.07 per unit change in Center for Epidemiologic Studies–Depression Scale score) and was inversely associated with the child being more independent with transfers (OR, 0.17).
The primary purpose of this study was to compare the prevalence of LBP in adult female primary caregivers of children with physical disabilities with the prevalence of LBP in adult female primary caregivers of children with nonphysically disabling medical illnesses. This study showed that the prevalence of LBP is higher in adult female primary caregivers of children with physical disabilities. This relationship was true when comparing the pediatric PMR clinic group with the pediatric endocrine clinic group as well as the subgroup analysis within the pediatric PMR clinic group. The results of regression analysis suggested that this difference was primarily related to the child needing help with transfers and was unrelated to the clinic in which the child was seen. Thus, when addressing caregiver burden with caregivers of children with physical disabilities, clinicians should also ask if the caregivers have LBP in addition to addressing the psychological aspects of caregiver burden.
The prevalence of LBP in the pediatric endocrine clinic group (43.5%) may at first seem a little high. However, this value is reasonable when compared with the results of other studies that show the point prevalence in the general population to range from 4% to 29% and the 1-year period prevalence to range from 10% to 63%.27
The secondary purpose was to identify potential risk factors associated with the prevalence of LBP. The factors associated with LBP are having a history of LBP, increased depression, and the child needing assistance with transfers, for example, from the chair to a wheelchair, from a wheelchair to the toilet, from a wheelchair to a tub or shower, up or down stairs, or from walking to a wheelchair. This model was good at predicting the presence of LBP in the caregivers (ROC curve area, 0.82). The results suggest that both psychological and physical factors are associated with the presence of LBP.
This is an important addition to the current literature because most articles on the physical risk factors for LBP are done in an occupational setting.1 This result contrasts with the result of a major prospective study by Bigos et al28 on occupational LBP that found LBP-related disability was mostly related to psychological factors. The nonoccupational literature has focused on the relationship among demographic, anthropometric, fitness, or psychological factors and LBP.1 To our knowledge, there is no published research that looked at the relationship among daily informal caregiving, physical lifting demands, and LBP. The findings reported in this article suggest that in a nonoccupational setting both physical and psychological factors contribute to the presence of LBP.
The results of this study need to be interpreted with the following limitations: the cross-sectional design of this study can prove an association and suggests causality, but prospective studies are better at proving causality. Also, the study subjects are mostly mothers of children being followed up in medical clinics. The results may not be as applicable to fathers or to parents of children not being followed up in the medical system. Another limitation is that many other potential predictive factors (eg, support from partners, birth order of the index child, general health of the mother, availability of respite care, financial status of the caregiver, and others) were not evaluated.
Our study has several strengths. To our knowledge, this is the first study we are aware of that focuses on evaluating the prevalence of LBP in adult female primary caregivers of children with physical disabilities. The study findings of Finkelstein15 and Rossignol et al16 suggested that the prevalence of LBP in male workers was associated with parenthood or the number of children in the family. Our study suggests that, in mothers of children with physical disabilities, the prevalence is not associated with the number of children in the family. Our study findings may not be comparable to the study findings of Finkelstein or Rossignol et al because our study population is female. Another strength of our study is the high participation rate. More than 90% of the eligible subjects agreed to participate.
This is the first study we are aware of that examines the relationship between LBP and being a caregiver to children with physically disabling medical illnesses. The results of our study suggest that both physical factors (transferability of the child) and psychological factors (the mood of the caregiver) are associated with LBP in the caregiver. Further studies are needed to better define this relationship and to explore treatment or management options.
The prevalence of LBP is higher in adult female primary caregivers of children with physical disabilities. Both physical and psychological factors contribute to the presence of LBP.
Corresponding author: Henry C. Tong, MD, Spine Program, 325 E Eisenhower, Second Floor, University of Michigan, Ann Arbor, MI 48108 (e-mail: firstname.lastname@example.org).
Accepted for publication June 9, 2003.
This study was supported by research training grant H133P990014 from the National Institute on Disability and Rehabilitation Research, Washington, DC, given to the University of Michigan Health System, Michigan State University, Ann Arbor Center for Independent Living Rehabilitation. Dr Tong's salary was paid by Medical Rehabilitation Research Training Program grant T32 HD07422-10 from the National Institutes of Health to the University of Michigan.
The risk factors for LBP have been studied extensively in occupational settings. There have been few studies evaluating nonoccupational risk factors, especially in an informal caregiving setting. This study shows that, in adult female primary caregivers of children with physical disabilies, the presence of LBP is associated with depression of the caregiver and the transferability of the child. Thus, clinicians who treat patients caring for children with physical disabilities should screen the caregivers for depression and LBP.
Tong HC, Haig AJ, Nelson VS, Yamakawa KS, Kandala G, Shin KY. Low Back Pain in Adult Female Caregivers of Children With Physical Disabilities. Arch Pediatr Adolesc Med. 2003;157(11):1128-1133. doi:10.1001/archpedi.157.11.1128