Evolution of disability in adults with hip arthroplasty: A national longitudinal study

Authors

  • Agnes Dechartres,

    1. INSERM, U738, Université Paris 7, UFR de Médecine, AP_HP Hôpital Bichat, Département d'Epidémiologie, Biostatistique et Recherche Clinique, Paris, France
    Search for more papers by this author
  • Isabelle Boutron,

    Corresponding author
    1. INSERM, U738, Université Paris 7, UFR de Médecine, AP_HP Hôpital Bichat, Département d'Epidémiologie, Biostatistique et Recherche Clinique, Paris, France
    • Département Epidémiologie, Biostatistique et Recherche Clinique, Hôpital Bichat Claude Bernard, 46, rue Henri Huchard, 75877 Paris, cedex 18, France
    Search for more papers by this author
  • Remy Nizard,

    1. Hôpital Lariboisière, Assistance Publique Hôpitaux de Paris, Université Paris 7, Paris, France
    Search for more papers by this author
  • Serge Poiraudeau,

    1. Hôpital Cochin, Assistance Publique Hôpitaux de Paris, Université Paris 5, Institut Fédératif de Recherche sur le Handicap, Paris, France
    Search for more papers by this author
  • Carine Roy,

    1. INSERM, U738, Université Paris 7, UFR de Médecine, AP_HP Hôpital Bichat, Département d'Epidémiologie, Biostatistique et Recherche Clinique, Paris, France
    Search for more papers by this author
  • Jean-François Ravaud,

    1. INSERM U750, CERMES, Institut Fédératif de Recherche sur le Handicap, Paris, France
    Search for more papers by this author
  • Philippe Ravaud

    1. INSERM, U738, Université Paris 7, UFR de Médecine, AP_HP Hôpital Bichat, Département d'Epidémiologie, Biostatistique et Recherche Clinique, Paris, France
    Search for more papers by this author

Abstract

Objective

To describe disability in individuals with hip arthroplasty and its evolution over 2 years compared with that in the general population, and to compare the degree of disability between subjects with recent and older hip arthroplasty.

Methods

We selected a national representative sample of 16,945 subjects from the 1999 French population census. This sample, interviewed in 1999 and 2001 about their level of disability, included 527 subjects with hip arthroplasty (i.e., representing 424,000 individuals in the French noninstitutionalized population): 145 who underwent the procedure between 1999 and 2001 (recent hip arthroplasty) and 382 with an older hip arthroplasty.

Results

Subjects with hip arthroplasty reported more difficulty in bending forward (odds ratio [OR] 4.5, 95% confidence interval [95% CI] 3.1–6.6), climbing stairs (OR 2.2, 95% CI 1.5–3.1), walking >300 meters (OR 1.6, 95% CI 1.03–2.6), dressing (OR 2.9, 95% CI 2.1–4.2), and getting in and out of a chair (OR 2.5, 95% CI 1.7–3.6) than the general population. However, the evolution in disability was similar to that of the general population. Compared with subjects with older hip arthroplasty, those with recent hip arthroplasty reported more difficulty walking >300 meters (OR 2.7, 95% CI 1.3–5.6), washing (OR 2.9, 95% CI 1.6–5.4), dressing (OR 2.2, 95% CI 1.2–4.2), and getting in and out of a chair (OR 2.1, 95% CI 1.1–3.9).

Conclusion

This study describes the potential future disability in the more elderly population, with implications for health-related planning.

INTRODUCTION

Total hip arthroplasty is currently the most efficient procedure to reduce disability among individuals with endstage hip osteoarthritis. The high prevalence of osteoarthritis in elderly persons and the aging of the population will lead to an increased prevalence of osteoarthritis and, consequently, total hip arthroplasty. The projections for 2030, based on changes expected in the population's age profile, predict an increase of ∼80% in total hip replacements (1). Therefore, to forecast future public health policies, activity limitations of this increasing population of elderly individuals should be assessed. Prospective cohort studies (2–12) have confirmed that the postoperative levels of pain and disability are consistently lower than preoperative levels, but only a few studies (10, 13, 14) have compared disability status of individuals who have undergone arthroplasty with that in the general population. In the Handicap, Disability, Dependence Survey, a 1999 French national community-based survey involving nearly 17,000 individuals, subjects with lower-limb arthroplasty reported a higher level of disability in mobility and self-care than did the general population (15, 16). However, this cross-sectional study did not provide any information on the evolution of disability over time. Moreover, because individuals with hip and knee arthroplasty were not initially distinguished, the specific level of disability for subjects with hip arthroplasty remained uncertain.

Using the 2-year followup data of the Handicap, Disability, Dependence Survey, we aimed to describe activity limitations and their evolution in persons with hip arthroplasty compared with that in the general population. A secondary objective was to compare the level of disability between subjects with recent hip arthroplasty (within the previous 2 years) and those with less-recent hip arthroplasty.

MATERIALS AND METHODS

Study design

The data constituting the basis of this report were collected from the Handicap, Disability, Dependence Survey, a national, longitudinal, community-based survey by the French National Institute of Statistic and Economic Studies describing disability and handicap in France. The target population included residents in all French households (n = 57.4 million) including children (age range for the 2001 survey 2–103 years). This survey methodology is described in detail elsewhere (15, 17–19). Briefly, a 2-stage method was used according to United Nations recommendations (20) to organize the cohort in 1999. For the first stage, a representative sample of census districts (∼600 inhabitants per district) was selected. During the census taking, enumerators gave these households the standard forms of the 1999 French population census and an additional questionnaire concerning daily life and health. This screening questionnaire allowed for classifying individuals into 6 groups of increasing probability of presumed disability. This first 1999 phase involved ∼417,500 persons and had an 86% response rate. For the second stage, we selected the population by randomization, with an unequal probability of drawing, with a high sampling rate for the most severely disabled group and a minimum sampling rate for persons without daily living restrictions (the largest group). Each of the resulting groups were assigned a specific sampling coefficient that increased with the probability or severity of the presumed handicap. The sample design allowed for weighting the data to estimate representative results at a national level (21). This cohort included 16,945 subjects representative of the French population living at home, 815 of whom reported having undergone lower-limb arthroplasty (i.e., representing 691,000 individuals in the French noninstitutionalized population).

For the 2-year followup survey, 12,530 individuals were re-interviewed: 608 of the 815 subjects who had reported lower-limb arthroplasty and 11,922 of the other 16,130 original participants. Reasons for not re-interviewing are shown in Figure 1. Among the 608 subjects with lower-limb arthroplasty identified in 1999, 382 (i.e., representing 303,000 individuals in the French noninstitutionalized population) had received hip arthroplasty (older hip arthroplasty), 134 had received knee arthroplasty, 30 had received both hip and knee arthroplasty, and 62 had uncertain data. Among persons in the general population who were re-interviewed, 145 (i.e., representing 121,000 individuals in the French noninstitutionalized population) reported having undergone hip arthroplasty since 1999 (recent hip arthroplasty). Consequently, a total of 527 individuals with hip arthroplasty (i.e., representing 424,000 individuals in the French noninstitutionalized population) were selected.

Figure 1.

Study population of the Handicap, Disability, Dependence Survey.

Data collection

Computer-assisted personal interviews were conducted from October 2001 to January 2002 by 408 trained interviewers. Respondents were asked to describe changes in health status since 1999 (improvement or worsening, appearance of other chronic conditions). Chronic conditions included chronic diseases, effects of accidents, problems from birth, and impairments due to aging.

Disability was assessed from subjects' reports. Respondents were asked about the degree of difficulty or the need for help in 5 defined areas: activities involving personal care (washing, dressing, cutting toenails, manual abilities, continence), mobility (going outside, getting out of bed or a chair, climbing stairs, bending forward, walking a distance), and housekeeping (shopping, carrying, housework, meal preparation), as well as cognitive and sensory abilities. Answers were given on a 5-point Likert scale, from “no disability” to “impossible without help.” Walking distance limitation was initially a continuous variable that we transformed to a variable with several categories: walking <100 meters, 100–300 meters, 300–500 meters, 500–1,000 meters, and >1,000 meters. For the description of walking distance limitations, we selected several cutoff points such as walking no more than 100 meters, 300 meters, 500 meters, and 1,000 meters. For each task, conditions were considered improved if the answer decreased by at least 1 point and worse if the answer increased by at least 1 point. If the answers did not differ between 1999 and 2001, we reasoned no change in disability for the studied task. We did the same for the description of the evolution of walking distance limitation, using the variable with several categories. For example, if subjects reported walking no more than 300–500 meters in 1999 and reported in 2001 that they could not walk more than 100–300 meters, they were considered to have a worsened limitation for walking distance. The complete questionnaire is available at the following address: http://ifr-handicap.inserm.fr/HID/ACCUEIL_HID_New/.HTM.

Statistical analysis

The study design allowed for weighting the data to estimate representative results at a national level. To calculate statistical parameters and their 95% confidence intervals (95% CIs), we used SAS procedures (SAS Institute, Cary, NC) specific for handling complex sample designs to obtain correct variances estimates (SAS version 9.1, PROC SURVEYFREQ, PROC SURVEYMEAN, PROC SURVEYREG, and PROC SURVEYLOGISTIC [22]).

To account for attrition bias, we built a propensity score to assess the probability of nonresponse. To do so, we performed a logistic regression with the 2001 responses to the survey as the dependent variable. The covariates were 1999 demographic, economic, and educational variables as well as the level of disability in performing common activities of daily living from the 1999 survey. The logistic regression analysis was used to determine the probability of nonresponse (from 0 to 1), the propensity score, for each subject in the data set. Then, multiple logistic regression models adjusted for age, sex, education, number of chronic conditions, and probability of 2001 nonresponse allowed for estimating odds ratios (ORs) and their 95% CIs for 1) level of disability in subjects with hip arthroplasty compared with that in persons in the general population, 2) level of disability in subjects with recent hip arthroplasty compared with those with less-recent hip arthroplasty, and 3) evolution of disability (i.e., worsening) in subjects with hip arthroplasty identified in 1999 compared with that in the general population. Concerning the choice of the type of evolution entered into the model, we found it more interesting to know whether disability in individuals with hip arthroplasty was worse as compared with the general population so that the long-term treatment of persons with hip arthroplasty could be adapted. Separate models were created for each activity of daily living. All P values were 2-sided, and a P value ≤0.05 was considered significant. Data analyses involved use of the SAS statistical software, version 9.1 (SAS Institute).

RESULTS

Demographic characteristics

The main demographic data of the 527 subjects with hip arthroplasty representing 424,000 individuals are presented in Table 1. The mean age was 74.1 years, ranging from 29 to 98 years, with 55.6% of subjects ≥75 years of age. Women represented 60.4% of the study population.

Table 1. Demographic, educational, disability, and health characteristics of all subjects with hip arthroplasty, subjects with recent hip arthroplasty, older hip arthroplasty, and the general population*
CharacteristicsAll hip arthroplasty (n = 527)Recent hip arthroplasty (n = 145)Older hip arthroplasty (n = 382)General population (n = 11,510)
No.% (95% CI)No.% (95% CI)No.% (95% CI)No.% (95% CI)
  • *

    95% CI = 95% confidence interval.

  • Sum not always equal to total number because of missing answers.

Characteristics        
 Sex        
  Female29760.4 (55.4–65.5)7752.7 (42.4–63.0)22063.5 (57.8–69.2)6,09552.0 (49.5–54.5)
  Male23039.6 (34.5–44.6)6847.3 (37.0–57.6)16236.5 (30.8–42.2)5,41548.0 (45.5–50.4)
 Age, years        
  <6511513.6 (11.1–16.0)4018.5 (11.9–25.1)7511.6 (8.7–14.6)6,48184.0 (82.7–85.3)
  65–7415730.8 (25.9–35.7)4127.2 (18.7–35.8)11632.2 (26.1–38.3)2,1889.3 (8.1–10.5)
  ≥7525555.6 (51.0–60.1)6454.3 (44.2–64.3)19156.1 (50.0–62.3)2,8416.7 (6.2–7.1)
 Education level        
  No schooling132.6 (0.9–4.4)42.0 (0–4.2)92.9 (0.6–5.2)1,50124.1 (21.9–26.3)
  Primary school40577.5 (73.1–81.9)11282.4 (75.6–89.1)29375.6 (70.1–81.0)5,90225.1 (23.2–26.9)
  Comprehensive school/junior high school5410.3 (6.9–13.7)209.9 (4.7–15.1)3410.5 (6.2–14.8)2,06826.1 (23.9–28.3)
  High school graduate and above488.5 (5.8–11.2)84.6 (1.0–8.1)4010.0 (6.5–13.5)1,40019.4 (17.4–21.4)
  Other (do not know, refused to answer)71.1 (0.2–2.0)11.1 (0.0–3.3)61.0 (0.1–2.0)6395.4 (4.2–6.5)
 Employment status        
  Employed344.5 (2.8–6.1)115.4 (1.9–8.9)234.1 (2.2–6.1)2,54551.8 (49.2–54.5)
  Unemployed60.4 (0.03–0.7)20.6 (0–1.4)40.3 (0–0.6)4934.7 (3.5–5.9)
  Retired41083.4 (73.2–93.5)10779.3 (59.5–99.1)30385.0 (73.0–97.0)5,22623.2 (20.6–25.8)
  Housewife, other homemaker7711.8 (2.2–16.0)2514.7 (4.1–25.3)5210.6 (6.2–15.0)2,11010.2 (8.2–12.1)
  Student00000035910.0 (8.3–11.8)
Disabilities and self-reported health        
 Self-care difficulties        
  Washing21739.6 (34.5–44.7)7352.7 (42.5–63.0)14434.3 (28.7–39.9)2,1414.8 (4.1–5.5)
  Dressing24345.6 (40.2–50.9)8258.9 (48.6–69.2)16140.1 (34.0–46.1)2,1824.5 (3.9–5.1)
  Cutting toenails44583.6 (79.5–87.7)12988.4 (80.6–96.2)31681.6 (76.8–86.5)4,37611.1 (10.1–12.1)
  Using the toilets8815.6 (12.1–19.2)3221.7 (13.8–29.5)5613.2 (9.4–17.0)9071.8 (1.4–2.2)
 Mobility difficulties        
  Getting in and out of a bed22139.5 (34.5–44.6)6544.8 (34.6–55.1)15637.4 (31.6–43.1)2,0254.3 (3.7–4.9)
  Getting in and out of a chair20536.6 (31.7–41.6)6846.8 (36.5–57.0)13732.6 (27.1–38.1)1,6953.6 (3.1–4.2)
  Climbing up and down stairs32461.0 (55.5–66.6)9366.5 (56.1–76.8)23158.9 (52.3–65.4)3,1797.3 (6.5–8.1)
  Bending forward41075.9 (71.2–80.6)11980.8 (72.0–89.6)29173.9 (68.3–79.5)3,9049.3 (8.4–10.2)
 Walking distance limitation        
  <100 meters8316.2 (12.2–20.2)2923.6 (13.6–33.5)5413.4 (9.5–17.2)6761.4 (1.1–1.7)
  <300 meters14027.2 (22.4–32.0)4535.5 (24.8–46.2)9524.0 (18.9–29.1)1,1392.5 (2.0–3.0)
  <500 meters21541.9 (36.4–47.3)6647.6 (36.7–58.6)14939.6 (33.5–45.8)1,8064.5 (3.8–5.2)
  <1,000 meters23645.9 (40.4–51.4)7373.0 (41.3–63.2)16343.5 (37.2–49.8)1,9875.1 (4.4–5.9)
 Housekeeping difficulties        
  Shopping23846.1 (40.6–51.6)7152.4 (41.4–63.4)16743.7 (37.4–50.0)3,0019.2 (8.1–10.3)
  Carrying 5 kg for 10 meters36667.4 (62.3–72.5)10669.8 (59.8–79.8)26066.4 (60.5–72.4)4,37212.3 (11.0–13.5)
 Self-reported health        
  Fair, poor, or very poor11318.4 (13.5–23.3)9635.0 (24.4–45.6)24464.6 (58.4–70.9)4,86820.8 (18.9–22.7)
 Use of walking devices26148.3 (43.0–53.6)8058.8 (49.0–68.6)18144.0 (38.0–50.1)1,5403.1 (2.7–3.6)

Characteristics of nonrespondents at 2-year followup

For the arthroplasty group, nonrespondents were significantly older than respondents (mean age 77.2 versus 71.2 years; P = 0.01). Nonrespondents reported significantly more difficulties in 1999 as compared with respondents for walking and bending forward (data not shown). However, the number of chronic conditions was not significantly different from that of respondents.

Description of disability levels for individuals with hip arthroplasty

The mean number of chronic conditions for all subjects with hip arthroplasty was 3.4 (95% CI 3.2–3.6, range 0–13). Activity limitations concerned mainly mobility (Table 1): of all subjects with hip arthroplasty, 75.9% reported at least some difficulties bending forward and picking up an object, 61.0% climbing up and down stairs, 67.4% carrying 5 kg for 10 meters, 46.1% shopping, and 27.2% could not walk >300 meters. Self-care was also limited, with 45.6% reporting difficulties dressing, 39.6% washing, and 39.5% getting in and out of bed. However, only 18.4% reported poor or very poor health.

The proportions of subjects with recent and older hip arthroplasty reporting at least some difficulty were 35.5% and 24.0% for walking >300 meters, 66.5% and 58.9% for climbing stairs, and 80.8% and 73.9% for bending to the floor, respectively. Moreover, 58.8% of subjects reporting recent hip arthroplasty and 44.0% reporting less-recent arthroplasty reported the use of a technical device to walk (mainly walking sticks).

Comparison of disability level between individuals with hip arthroplasty and the general population

After adjusting for age, sex, education, chronic conditions, and probability of nonresponse, subjects with hip arthroplasty (Table 2) reported significantly greater difficulties than the general population for activities involving the hip: bending forward and picking up something (OR 4.5, 95% CI 3.1–6.6), climbing up and down stairs (OR 2.2, 95% CI 1.5–3.1), walking >300 meters (OR 1.6, 95% CI 1.03–2.6), dressing (OR 2.9, 95% CI 2.1–4.2), getting in and out of a chair (OR 2.5, 95% CI 1.7–3.6), washing (OR 2.0, 95% CI 1.4–2.8), carrying 5 kg for 10 meters (OR 1.7, 95% CI 1.2–2.5), and cutting toenails (OR 5.4, 95% CI 3.5–8.4). Subjects with hip arthroplasty also more often reported the use of technical devices to walk (OR 4.8, 95% CI 3.5–6.8). However, their self-reported health was not significantly different from that of the general population.

Table 2. Odds ratio (95% confidence interval) of disability according to domains
Domains of disabilityAll subjects with hip arthroplasty (n = 527) compared with other participants (n = 11,510)*Subjects with recent hip arthroplasty (n = 145) compared with those with older hip arthroplasty (n = 382)
  • *

    Adjusted for age, sex, education, chronic conditions, and probability of nonresponse; 527 subjects representing 423,991 individuals.

  • Adjusted for age, sex, education, chronic conditions, and probability of nonresponse; 145 subjects representing 121,000 individuals.

  • Significant at P ≤ 0.05.

Self-care difficulties  
 Washing2.0 (1.4–2.8)2.9 (1.6–5.4)
 Dressing2.9 (2.1–4.2)2.2 (1.2–4.2)
 Cutting toenails5.4 (3.5–8.4)1.9 (0.7–5.4)
 Using the toilet2.0 (1.2–3.4)2.4 (1.1–5.3)
Mobility difficulties  
 Getting in and out of bed2.3 (1.6–3.2)1.6 (0.9–3.0)
 Getting in and out of a chair2.5 (1.7–3.6)2.1 (1.1–3.9)
 Climbing up and down stairs2.2 (1.5–3.1)1.6 (0.8–3.1)
 Bending forward and picking up something4.5 (3.1–6.6)1.5 (0.7–3.0)
Walking distance limitation  
 ≤100 meters2.0 (1.1–3.3)3.5 (1.5–8.4)
 ≤300 meters1.6 (1.03–2.6)2.7 (1.3–5.6)
 ≤500 meters1.6 (1.1–2.4)1.6 (0.8–3.0)
 ≤1,000 meters1.8 (1.2–2.6)1.8 (0.99–3.4)
Housekeeping difficulties  
 Shopping1.2 (0.8–1.6)2.6 (1.3–5.2)
 Carrying 5 kg for 10 meters1.7 (1.2–2.5)1.8 (0.9–3.5)
Health status  
 Having fair, poor, or very poor self-reported health1.4 (0.97–2.0)1.1 (0.6–2.1)
Use of technical devices to walk4.8 (3.5–6.8)2.3 (1.3–4.1)

Comparison of disability level between individuals with recent and older hip arthroplasty

Multiple regression models (Table 2) confirmed that subjects who underwent hip arthroplasty within the previous 2 years were more disabled in several activities of daily living than those who had hip arthroplasty more than 2 years earlier. After adjustment for age, sex, education, chronic conditions, and probability of nonresponse, results showed that subjects with recent hip arthroplasty had more difficulties walking >300 meters (OR 2.7, 95% CI 1.3–5.6), washing (OR 2.9, 95% CI 1.6–5.4), shopping (OR 2.6, 95% CI 1.3– 5.2), dressing (OR 2.2, 95% CI 1.2–4.2), and getting in and out of a chair (OR 2.1, 95% CI 1.1–3.9); these patients also more often reported the use of a technical device to walk (OR 2.3, 95% CI 1.3–4.1).

Evolution of disability status over 2 years for subjects with hip arthroplasty

Data on evolution of activity limitations from 1999 to 2001 for the subjects with hip arthroplasty identified in 1999 are shown in Table 3: 25% of subjects with hip arthroplasty had worse ability for walking, 24.4% for bending forward and picking up something, 22.8% for climbing up and down stairs, 23.5% for carrying 5 kg for 10 meters, and 20.3% for cutting toenails. However, only ≤10% reported worse cognitive, sensory, and manual abilities (data not shown).

Table 3. Evolution of disabilities over 2 years reported by subjects with hip arthroplasty identified in 1999 (382 subjects representing 303,226 individuals)*
 No.WorseningNo changeImprovement
  • *

    Values are the percentage (95% confidence interval) unless otherwise indicated.

Self-care difficulties    
 Washing38116.9 (12.6–21.3)67.9 (62.3–73.5)15.1 (10.9–19.3)
 Dressing38018.2 (13.6–22.7)69.7 (64.3–75.1)12.1 (8.5–15.7)
 Cutting toenails37920.3 (14.8–25.9)68.7 (62.6–74.7)11.0 (7.4–14.7)
 Using the toilet3797.8 (4.9–10.8)86.4 (82.8–90.1)5.7 (2.4–8.1)
Mobility difficulties    
 Getting in and out of bed37917.4 (13.1–21.6)71.9 (66.7–77.0)10.8 (7.3–14.2)
 Getting in and out of a chair37916.2 (12.1–20.4)71.0 (65.7–76.4)12.7 (8.8–16.7)
 Bending forward and picking up something34324.4 (18.8–30.0)52.1 (45.5–58.7)23.5 (17.9–29.1)
 Climbing up and down stairs34122.8 (17.5–28.1)58.0 (51.7–64.4)19.2 (14.4–23.9)
Walking distance limitation33425.0 (19.5–30.5)57.4 (50.9–63.9)17.6 (12.9–22.4)
Housekeeping difficulties    
 Shopping34118.9 (14.0–23.8)69.3 (63.4–75.1)11.8 (7.8–15.8)
 Carrying 5 kg for 10 meters37723.5 (17.9–29.1)59.6 (53.4–65.8)16.9 (12.4–21.4)
Health status35120.9 (15.2–26.6)58.3 (51.9–64.7)20.8 (15.7–25.8)
Use of technical devices to walk37913.1 (8.9–17.2)79.1 (74.3–83.9)7.8 (4.9–10.6)

After adjusting for age, sex, education, chronic conditions, and probability of nonresponse, the evolution of disability over 2 years (Table 4) in subjects with hip arthroplasty was similar to that in the general population in all areas, including mobility. Nevertheless, use of technical devices increased more in individuals with hip arthroplasty than in the general population (OR 1.7, 95% CI 1.02–2.9).

Table 4. Odds ratio (95% confidence interval) of disability worsening over 2 years according to domains for subjects with hip arthroplasty identified in 1999 (382 subjects representing 303,226 individuals) compared with the other participants (11,510 subjects)
Domains of disabilitySubjects with older hip arthroplasty compared with other participants*
  • *

    Adjusted for age, sex, education, chronic conditions, and probability of nonresponse.

  • Significant at P ≤ 0.05.

Self-care difficulties 
 Washing0.9 (0.5–1.5)
 Dressing1.2 (0.8–2.0)
 Cutting toenails0.8 (0.5–1.4)
 Using the toilet0.7 (0.4–1.6)
Mobility difficulties 
 Getting in and out of bed1.2 (0.8–1.8)
 Getting in and out of a chair1.3 (0.8–2.0)
 Climbing up and down stairs1.1 (0.8–1.7)
 Bending forward and picking up something1.0 (0.7–1.5)
Walking distance limitation1.2 (0.8–1.8)
Housekeeping difficulties 
 Shopping0.7 (0.4–1.1)
 Carrying 5 kg for 10 meters0.9 (0.6–1.5)
Health status1.2 (0.8–1.9)
Use of technical devices to walk1.7 (1.02–2.9)

DISCUSSION

This study assessed disability in activities of daily living experienced by subjects with hip arthroplasty from a national random sample and the evolution of disability over 2 years. Such subjects reported a higher level of disability in mobility and self-care activities than the general population. However, the evolution of disability over 2 years was similar between the 2 groups for all areas assessed. Our results also highlighted that subjects with recent hip arthroplasty (during the previous 2 years) were more disabled than those with less-recent hip arthroplasty.

Our results on level of disability are consistent with those from other studies demonstrating that, despite large improvements in mobility, subjects with hip arthroplasty did not reach the level of mobility reported by the general population. For instance, Jones et al (10) showed that the 6-month Short Form 36 and Western Ontario and McMaster Universities Osteoarthritis Index scores for physical function did not reach the values reported in the general population for age and sex. Franzen et al (14) also demonstrated that, compared with control subjects, hip arthroplasty patients described worse pain (P < 0.001) and physical mobility (P < 0.001) 10–20 years after the procedure. Moreover, in a prospective study, Rissanen et al (23) found that 20.1% of hip arthroplasty patients considered their physical ability poor 2 years after surgery; only 36.2% and 41% had no problem walking and negotiating stairs, respectively. Concerning the evolution of disability, Fortin et al (7) found in a prospective study that patients with low baseline function status had a lower function score 6 months postoperatively as compared with patients with a high baseline function status; however, at 24 months, the evolution in disability was comparable. Moreover, in a case–control study, Goldsmith et al (13) found that walking activities were reduced with advancing age for both subjects with hip arthroplasty and the general population. However, ours is the first study providing data on disability and its evolution over 2 years for individuals with hip arthroplasty from a national representative sample. Our results may be generalizable to the entire noninstitutionalized hip arthroplasty population, which is contrary to results of most studies performed in a few specialized centers by highly qualified surgeons (24–26). Moreover, this study allows for comparing levels and evolution of disability with that in the general population after adjustment for potential confounding factors. Finally, results of the comparison between recent and less-recent hip arthroplasty are also consistent with other studies reporting a high level of functional status 3–6 months after surgery (6, 27–30), even up to 1 year after surgery (12).

The impact of our findings for clinicians is 3 fold. First, a precise description of functional limitations and disabilities after total hip arthroplasty could help rehabilitation physicians better define objectives of physical therapy and rehabilitation, and therefore better adapt therapeutic programs to these objectives. Second, the finding that individuals who underwent total hip arthroplasty <2 years earlier have more disabilities than those with less-recent total hip arthroplasty suggests that more intensive rehabilitation programs should be tested soon after surgery to shorten the period of disability. Finally, a precise description of disability after total hip arthroplasty should help surgeons better inform patients about the outcome of hip arthroplasty so that patients have realistic expectations after the procedure.

Our study has some limitations. First, data on nonrespondents could have biased results by misestimating the level of disability in the hip arthroplasty population, because nonrespondents were significantly older and more disabled in walking than respondents. To limit this attrition bias, we performed a propensity score to adjust all multivariate analyses for the probability of nonresponse. Second, our study concerned only the French noninstitutionalized population, which could lead to a possible underestimation of the level of disability reported for the hip arthroplasty population, because disabled persons often live in an institution. Our study also has the usual limitations of descriptive studies based on self-reported outcomes. However, the patient's perspective is recognized to be central to assessing the effects of procedures (31, 32). Another limitation is the lack of information about the indication for surgery, the surgical procedures, and the rehabilitation programs. Our hip arthroplasty population may include very different types of patients, with different diseases leading to arthroplasty and different types of treatments. Nevertheless, all of these limitations were counterbalanced by our use of a national survey, which provided a detailed description of disabilities experienced by the entire hip arthroplasty population, consequently reflecting their functional status and information on 2-year evolution of their disability. In conclusion, this national representative survey may help clinicians, researchers, and service planners understand and address what the increasing number of individuals who will have hip arthroplasty need from health interventions to reduce their level of disability.

AUTHOR CONTRIBUTIONS

Dr. Boutron had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study design. Jean-François Ravaud, Philippe Ravaud.

Analysis and interpretation of data. Dechartres, Boutron, Nizard, Poiraudeau, Roy, Jean-François Ravaud, Philippe Ravaud.

Manuscript preparation. Dechartres, Boutron, Philippe Ravaud.

Statistical analysis. Dechartres, Roy.

Acknowledgements

The authors thank the National Institute of Statistic and Economic Studies division Enquêtes et études démographiques and especially Pierre Mormiche, who is responsible for the Handicap, Incapacité, Dépendance project task force.

Ancillary