Background
Functional status plays an important role in the comprehensive characterization of older adults. Functional limitations are associated with an increased risk of adverse treatment outcomes, but there are limited data on the prevalence of functional limitations in older women with pelvic floor disorders.
Objective
The aim of the study was to describe the prevalence of functional limitations based on health status in older women with pelvic organ prolapse (POP).
Study Design
This pooled, cross-sectional study utilized data from the linked Health and Retirement Study and Medicare files from 1992 through 2008. The analysis included 890 women age ≥65 years with POP. We assessed self-reported functional status, categorized in strength, upper and lower body mobility, activities of daily living (ADL), and instrumental ADL (IADL) domains. Functional limitations were evaluated and stratified by respondents self-reported general health status. Descriptive statistics were used to compare categorical and continuous variables, and logistic regression was used to measure differences in the odds of functional limitation by increasing age.
Results
The prevalence of functional limitations was 76.2% in strength, 44.9% in upper and 65.8% in lower body mobility, 4.5% in ADL, and 13.6% in IADL. Limitations were more prevalent in women with poor or fair health status than in women with good health status, including 91.5% vs 69.9% in strength, 72.9% vs 33.5% in upper and 88.0% vs 56.8% in lower body mobility, 11.6% vs 0.9% in ADL, and 30.6% vs 6.7% in IADL; all P < .01. The odds of all functional limitations also increased significantly with advancing age.
Conclusion
Functional limitations, especially in strength and body mobility domains, are highly prevalent in older women with POP, particularly in those with poor or fair self-reported health status. Future research is necessary to evaluate if functional status affects clinical outcomes in pelvic reconstructive and gynecologic surgery and whether it should be routinely assessed in clinical decision-making when treating older women with POP.
Introduction
Function is complex and comprised of physical, cognitive, sensory, psychological, and social elements. Physical function encompasses domains in activities of daily living (ADL), instrumental ADL (IADL), body strength, and mobility. Functional limitations reflect gaps between the person’s capabilities to perform in these domains and the demands of the environment. They are associated with increased risk of adverse treatment outcomes, including postoperative complications such as delirium, slower recovery, prolonged hospital stay, and in-hospital and long-term mortality. Functional limitations occur more frequently in women than men affecting 42-50% of women age ≥65 years. Therefore, functional status may be a crucial determinant of clinical outcomes among older women, and a comprehensive characterization of function in this population is necessary.
Recent research suggests that women with pelvic organ prolapse (POP) and impaired preoperative functional status have a longer length of hospitalization and increased probability of complications even after controlling for a variety of potential confounding factors. The Institute of Medicine has identified the maintenance of functional status as a priority in older women receiving surgical care. In addition, the Centers for Medicare and Medicaid Services are currently working on developing functional status quality measures. However, pelvic surgeons may not be inclined to evaluate functional status or promote functional independence in older women if they are not cognizant to how prevalent functional limitations are in women undergoing surgical treatment.
Using the linked Health and Retirement Study (HRS) data and Medicare files, the objective of the study was to describe the prevalence of functional limitations based on health status in women age ≥65 years with POP, a prevalent pelvic floor disorder. We hypothesized that functional limitations are common in older women with POP.
Materials and Methods
Data sources
This is a pooled cross-sectional study utilizing data from the linked HRS and Medicare files from the CMS from 1992 through 2008. This study was deemed exempt from the Case Western Reserve University Institutional Review Board. HRS is a combined effort of the Institute for Social Research at University of Michigan and the National Institute on Aging and represents the largest ongoing prospective observational study of older persons’ health in the Unites States. Data collection began in 1992 and nearly 30,000 Americans age >50 years have since been enrolled. HRS was designed to assess the changes in labor force participation and health that individuals experience in later life. Comprehensive data were collected on physical health and functional status through in-depth interviews conducted every 2 years. Validity and reliability of the HRS data collection has been studied extensively. The HRS data have been used to evaluate mobility impairment and incontinence severity, urgency urinary incontinence in older women, and depression and urinary incontinence. The HRS data were linked to Medicare files that contain International Classification of Diseases, Ninth Revision ( ICD-9 ) codes from inpatient admissions and outpatient or ambulatory surgery files. Medicare claims data were used to identify participating older women in HRS with POP.
Study sample
There were 9125 fee-for-service Medicare beneficiaries in the HRS with linked Medicare claims from 1992 through 2008. The analytic sample of 918 subjects was identified through the ICD-9 POP diagnosis code (618.XX), including 618.01, 618.02, and 618.03 for anterior vaginal wall prolapse and 618.04 and 618.05 for posterior; 618.2, 618.3, 618.4, 618.1, 618.5, and 618.6 for uterovaginal or apical; and 618.00, 618.9, 618.09, and 618.84 for unspecified POP. Participants for this study included women ≥65 years of age with POP identified by the linked prolapse ICD-9 code and with self-reported functional and health status data. Women (n = 28) without data on self-reported health status or functional limitations were excluded. The HRS conducts the interviews every 2 years, which resulted in some of the 65-year-old participants being surveyed 1-2 years before their POP first diagnosis. Thus, the final sample contained a small number (n = 17) of women 63-64 years old.
Outcomes variables
While proxy respondents are sometimes used in HRS when a subject is unable to complete the interview herself, in this study there were 0 instances where a proxy respondent was used. We used the data from the interviews immediately preceding the diagnosis of POP when they were first recorded in claims data, but before pessary or surgical treatment for POP was initiated. The strength questions were based on the Guttman Scale of physical health used in the examination of older adults’ social participation. Strength evaluation consisted of questions about sitting for ≥2 hours, pulling/pushing large objects (eg, living room chair), and rising from a chair (after sitting for a long period); upper and lower body mobility included evaluation of lifting 10 lb (eg, heavy grocery bag), picking up a dime from the table, lifting arms above shoulder level, walking 1 block, walking several blocks, climbing 1 flight of stairs, and climbing several flights of stairs, respectively. Mobility and ADL were reported by participants using questionnaires originally developed by Katz et al that were utilized in all HRS interviews. IADL measurement in the HRS was based on items from the inventory developed by Lawton and Brody. ADL included bathing, getting in/out of bed, dressing, eating, and crossing the room; IADL included preparing meals, taking medications, managing money, using the telephone, and shopping. For each item, the participants were dichotomized into those with and without a limitation based on whether they reported any difficulty completing a task due to a health problem expected to last ≥3 months: “no” vs “yes” and “can’t do” or “don’t do.”
Respondent’s self-reported functional status was evaluated and stratified by their general health status that was measured as “poor,” “fair,” “good,” “very good,” or “excellent.” The primary outcomes were the prevalence of functional limitations in strength, upper and lower body mobility, ADL, and IADL domains that were stratified based on self-reported health status “good/very good/excellent” vs “poor/fair.”
We also examined the presence of comorbidities and geriatric syndromes. Participants were queried on whether “a doctor ever told” them they had a condition. The HRS measured chronic diseases and comorbidities prevalent among middle-aged and elderly persons most likely to result in functional limitations. Selected comorbidities included lifetime history of hypertension, diabetes, cancer, chronic lung disease, heart problems (eg, angina, congestive heart failure), stroke, psychiatric illness, and arthritis. Geriatric syndromes were depression (≥4 symptoms on modified 8-item Center for Epidemiologic Studies Depression Scale ), urinary incontinence, falls in the last year, low cognitive performance (eg, bottom tertile of 35-point scale measuring working memory, mental processing speed), hearing impairment (self-rated “fair” or “poor” hearing, even when “using a hearing aid as usual”), vision impairment (self-rated “fair” or “poor” eyesight, even when wearing corrective lenses, or “legally blind”), and severe pain (self-rated as whether patients are “often troubled” by moderate to severe pain).
Statistical analysis
We conducted descriptive analysis for all study variables overall and by participants self-reported general health status. Comparisons of functional limitations between good and poor self-rated health status were made using Pearson χ 2 test for categorical variables and Student t test for continuous variables. To measure differences in the odds of functional limitations by participants’ categorized age (65-74, 75-84, and ≥85 years), logistic regression was used to model the probability of disability for each category of functional limitations as the dependent variable conditional on age grouping. All analysis was performed using software (SAS, Version 9.3 for Unix; SAS Institute Inc, Cary, NC). To comply with the Medicare data users’ agreement, we masked the small cells with n = ≤10. Additional cells were masked, as necessary, to prevent complementary disclosure.
Materials and Methods
Data sources
This is a pooled cross-sectional study utilizing data from the linked HRS and Medicare files from the CMS from 1992 through 2008. This study was deemed exempt from the Case Western Reserve University Institutional Review Board. HRS is a combined effort of the Institute for Social Research at University of Michigan and the National Institute on Aging and represents the largest ongoing prospective observational study of older persons’ health in the Unites States. Data collection began in 1992 and nearly 30,000 Americans age >50 years have since been enrolled. HRS was designed to assess the changes in labor force participation and health that individuals experience in later life. Comprehensive data were collected on physical health and functional status through in-depth interviews conducted every 2 years. Validity and reliability of the HRS data collection has been studied extensively. The HRS data have been used to evaluate mobility impairment and incontinence severity, urgency urinary incontinence in older women, and depression and urinary incontinence. The HRS data were linked to Medicare files that contain International Classification of Diseases, Ninth Revision ( ICD-9 ) codes from inpatient admissions and outpatient or ambulatory surgery files. Medicare claims data were used to identify participating older women in HRS with POP.
Study sample
There were 9125 fee-for-service Medicare beneficiaries in the HRS with linked Medicare claims from 1992 through 2008. The analytic sample of 918 subjects was identified through the ICD-9 POP diagnosis code (618.XX), including 618.01, 618.02, and 618.03 for anterior vaginal wall prolapse and 618.04 and 618.05 for posterior; 618.2, 618.3, 618.4, 618.1, 618.5, and 618.6 for uterovaginal or apical; and 618.00, 618.9, 618.09, and 618.84 for unspecified POP. Participants for this study included women ≥65 years of age with POP identified by the linked prolapse ICD-9 code and with self-reported functional and health status data. Women (n = 28) without data on self-reported health status or functional limitations were excluded. The HRS conducts the interviews every 2 years, which resulted in some of the 65-year-old participants being surveyed 1-2 years before their POP first diagnosis. Thus, the final sample contained a small number (n = 17) of women 63-64 years old.
Outcomes variables
While proxy respondents are sometimes used in HRS when a subject is unable to complete the interview herself, in this study there were 0 instances where a proxy respondent was used. We used the data from the interviews immediately preceding the diagnosis of POP when they were first recorded in claims data, but before pessary or surgical treatment for POP was initiated. The strength questions were based on the Guttman Scale of physical health used in the examination of older adults’ social participation. Strength evaluation consisted of questions about sitting for ≥2 hours, pulling/pushing large objects (eg, living room chair), and rising from a chair (after sitting for a long period); upper and lower body mobility included evaluation of lifting 10 lb (eg, heavy grocery bag), picking up a dime from the table, lifting arms above shoulder level, walking 1 block, walking several blocks, climbing 1 flight of stairs, and climbing several flights of stairs, respectively. Mobility and ADL were reported by participants using questionnaires originally developed by Katz et al that were utilized in all HRS interviews. IADL measurement in the HRS was based on items from the inventory developed by Lawton and Brody. ADL included bathing, getting in/out of bed, dressing, eating, and crossing the room; IADL included preparing meals, taking medications, managing money, using the telephone, and shopping. For each item, the participants were dichotomized into those with and without a limitation based on whether they reported any difficulty completing a task due to a health problem expected to last ≥3 months: “no” vs “yes” and “can’t do” or “don’t do.”
Respondent’s self-reported functional status was evaluated and stratified by their general health status that was measured as “poor,” “fair,” “good,” “very good,” or “excellent.” The primary outcomes were the prevalence of functional limitations in strength, upper and lower body mobility, ADL, and IADL domains that were stratified based on self-reported health status “good/very good/excellent” vs “poor/fair.”
We also examined the presence of comorbidities and geriatric syndromes. Participants were queried on whether “a doctor ever told” them they had a condition. The HRS measured chronic diseases and comorbidities prevalent among middle-aged and elderly persons most likely to result in functional limitations. Selected comorbidities included lifetime history of hypertension, diabetes, cancer, chronic lung disease, heart problems (eg, angina, congestive heart failure), stroke, psychiatric illness, and arthritis. Geriatric syndromes were depression (≥4 symptoms on modified 8-item Center for Epidemiologic Studies Depression Scale ), urinary incontinence, falls in the last year, low cognitive performance (eg, bottom tertile of 35-point scale measuring working memory, mental processing speed), hearing impairment (self-rated “fair” or “poor” hearing, even when “using a hearing aid as usual”), vision impairment (self-rated “fair” or “poor” eyesight, even when wearing corrective lenses, or “legally blind”), and severe pain (self-rated as whether patients are “often troubled” by moderate to severe pain).
Statistical analysis
We conducted descriptive analysis for all study variables overall and by participants self-reported general health status. Comparisons of functional limitations between good and poor self-rated health status were made using Pearson χ 2 test for categorical variables and Student t test for continuous variables. To measure differences in the odds of functional limitations by participants’ categorized age (65-74, 75-84, and ≥85 years), logistic regression was used to model the probability of disability for each category of functional limitations as the dependent variable conditional on age grouping. All analysis was performed using software (SAS, Version 9.3 for Unix; SAS Institute Inc, Cary, NC). To comply with the Medicare data users’ agreement, we masked the small cells with n = ≤10. Additional cells were masked, as necessary, to prevent complementary disclosure.
Results
Descriptive characteristics
The study sample included 890 women with POP. The majority (n = 772; 86.7%) were Caucasian, while a smaller proportion were African American (n = 53; 6%) and Hispanic (n = 58; 6.5%). The mean age of the cohort was 74.5 years (SD 6.5). The mean annual income of participants was $27,012 (SD $37,705) and the majority (n = 524; 59%) were married. Obesity (body mass index ≥30) was present in approximately one fifth (n = 171; 19.2%) of women, and comorbidities included arthritis (70%), hypertension (57%), heart disease (27%), stroke (11%), cancer (14%), chronic obstructive pulmonary disease/lung disease (10%), and psychiatric illness (15%). The mean number of comorbidities per person was 1.9 (SD 1.2).
Self-reported health status
The self-reported general health status was poor or fair in 29% of the sample. The majority of Caucasians reported good or excellent health status, whereas a larger proportion of African Americans and Hispanics reported poor or fair health status.
Participants’ demographics of women with POP by their self-reported general health status are shown in Table 1 . There was no difference in the mean age between women reporting good or excellent health and those who reported poor or fair health (74.2 ± 6.4 vs 75.2 ± 6.6 years, respectively, P = .06). Participants with worse health status were more likely to be obese compared to women with good or excellent health status (27.9% vs 15.7%, respectively, P < .01). Women with poor or fair health status were also less likely to be married and had lower income. In addition, those with poor or fair health status had a higher prevalence of all chronic comorbidities and geriatric syndromes, except falls ( Table 2 ).
| Demographic variables | Health status | P value a | |||
|---|---|---|---|---|---|
| Poor/fair | Good/very good/excellent | ||||
| Total | 258 | 632 | |||
| Race | <.001 | ||||
| Caucasian | 192 | (74.4) | 580 | (91.8) | |
| African American | 26 | (10.1) | 27 | (4.3) | |
| Hispanic | 39 | (15.1) | 19 | (3.0) | |
| Age, y | .09 | ||||
| 63–64 | 17 | (6.6) | 30 | (4.7) | |
| 65–69 | 46 | (17.8) | 147 | (23.3) | |
| 70–74 | 68 | (26.4) | 194 | (30.7) | |
| 75–79 | 64 | (24.8) | 138 | (21.8) | |
| 80–84 | 40 | (15.5) | 89 | (14.1) | |
| ≥85 | 23 | (8.9) | 34 | (5.4) | |
| Prolapse | .59 | ||||
| Anterior | 49 | (19) | 128 | (20.3) | |
| Posterior | 21 | (8.1) | 77 | (12.2) | |
| Uterine/apical | 133 | (51.6) | 345 | (54.6) | |
| Other | 96 | (37.2) | 218 | (34.5) | |
| BMI | <.001 | ||||
| Normal/overweight, 18.5–29.9 | 174 | (67.4) | 523 | (82.8) | |
| Obese, ≥30 | 72 | (27.9) | 99 | (15.7) | |
| Married | 125/255 | (49.0) | 399/630 | (63.3) | <.001 |
| Mean income, $, ±SD | 15,821 ± 14,232 | 31,581 ± 42,992 | <.001 | ||
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
