Aggregated measures of functional disability in a nationally representative sample of disabled people: analysis of dimensionality according to gender and severity of disability. (Cabrero-García & López-Pina, 2008)

Cabrero-García, J., & López-Pina, J. A. (2008). Aggregated measures of functional disability in a nationally representative sample of disabled people: analysis of dimensionality according to gender and severity of disability. Quality of Life Research, 17(3), 425-436.

Department of Nursing, University of Alicante, Campus de San Vicente del Raspeig, Ap. 99, 03080 Alicante, Spain. julio.cabrero@ua.es

OBJECTIVE: To determine (i) the dimensional invariance of instrumental and basic activities of daily living (IADL/ADL) by gender subgroups, and (ii) the extent to which ADL dimensionality varies with the inclusion or exclusion of nondisabled people.

METHODS: Data were taken from the 1999 Spanish Survey on Disability, Impairment and State of Health. The analysis focused on 6,522 people aged over 65 years who received help to perform or were unable to perform IADL/ADL items. Unidimensional and multidimensional item response theory (IRT) models were applied to this sample.

RESULTS: In the female sample, IADL/ADL items formed a scale with sufficient unidimensionality to fit a two-parameter logistic IRT model. In the male sample, the structure was bidimensional: self-care and mobility, and household activities. When the sample was composed of IADL/ADL disabled people, ADL items formed a unidimensional scale; when it was composed only of ADL disabled people, they formed a bidimensional structure: self-care and mobility.

CONCLUSIONS: IADL/ADL items can be combined in a single scale to measure severity of functional disability in females, but not in males. Separate aggregated scores must be considered for each subdomain, basic mobility and self-care, in order to measure the severity of ADL disability.

Notes

Introduction

• Knowledge about the prevalence and incidence of functional disability is important in estimating the service to be provided as well as the planning for accommodating programs at the population level
• At the individual level, on the other hand, outcome assessment concerning functional disability is essential in deciding one's eligibility to participate in long term care programs as well as other types of care and assistant.
• The subdomains for ADL are basic mobility and self-care (Avlund, 1997; Lindeboom, Vermeulen, Holman, & De Haan, 2003) and Kempen, Miedema, Ormel & Molenaar, 1996) while that for IADL were not clearly specified (Lindeboom et al., 2003) (Coster, Haley, Andres, Ludlow, Bond & Ni., 2004).
• The most commonly identified subdomains for IADL are household activities, outdoor mobility and cognitive activities (Lindeboom et al., 2003) (Lawton & Brody, 1969).
• Spector et al (1987) were the first ones to combine ADL and IADL into one measure.
• Yet, there is still need for further research to examine the practice of using one single composite score to summarize ADL and IADL (Spector & Fleishman, 1998, Coster, Haley, Andres, Ludlow, Bond & Ni., 2004; Breithaupt & McDowell, 2001).
• Studies of unidimentional IRT model: Spector & Fleishman, 1998; Saliba, Orlando, Wenger, Hays & Rubenstein, 2002; Kempen, Myers & Powell, 1995
• Studies supporting the multidimensionality of the scale: Breithaupt & McDowell, 2001; Thomas, Rockwood & McDowell, 1998; Johnson & Wolinsky, 1994; Ng, Niti, Chiam & Kua, 2006
• There have also been concerns regarding the compatibility of the IADL items for men and women such as DIF studies (See paper for gender related literature for I will not list them here given that it is not my interest)
• Lazaridis (1994) questioned the appropriateness of implementing the deterministic Guttman scale, which could be considered as a limited version of the probabilistic Rasch model, to summarize the ADL data.
• On page 626 and in the first paragraph on the right column, the author provides a good explanation about "extreme scores" such as those with a zero score.
  

  
  

  Method
  

• Instrument: The 1999 Spanish Survey on Disability, Impairment and State of Health (EDDES). Through information gathered through interviews concerning whether the interviewee is dependent or not on the others to perform the IADL/ADL tasks, each interviewee is assigned either a disabled or not disabled status for one of the 14 IADL/ADL items.
  

  Analysis method
  

• 1 PL and 2 PL
  
  • The fit of the two models are calculated using the difference between the -2 log likelihood values for each of the model
  • The percentage of reduction in the Chi square is calculated to evaluate model fits
  • Infit and outfit statistics were used to evaluate the item fit for Rasch model with the critical range between 0.7 to 1.3 (Linacre & Write, 1994)

  Results

• An important finding for me in this paper is that the dependency on sample size for the likelihood ratio chi square statistics is very evident. The authors found more misfit items when the sample size was 4618 then when the sample size is 500. No wonder all items are misfitted in my analyses since the sample size is something like 21574.
• Page 429 gives an example on how to write up IRT report
• Table 5 on Page 431 shows an example of presenting item parameter estimations.

  Discussion

• Different results were obtained when analyses included data from only people with ADL, ADL/IADL and the whole sample.
• Also, for females, items are sufficiently unidimensional while bidimensional for male (self-care and mobility, and household activities.
  

  
  

  Avlund, K. (1997). Methodological challenges in measurements of functional ability in gerontological research. A review. Aging, 9(3), 164-174.
  

  Cabrero-García, J., & López-Pina, J. A. (2008). Aggregated measures of functional disability in a nationally representative sample of disabled people: analysis of dimensionality according to gender and severity of disability. Quality of Life Research, 17(3), 425-436.
  

  Lindeboom, R., Vermeulen, M., Holman, R., & De Haan, R. J. (2003). Activities of daily living instruments: optimizing scales for neurologic assessments. Neurology, 60(5), 738-742.