Bathing and community participation for people with mobility impairments

This study explored how people with mobility impairments spend time within their homes using data from a large national dataset, the American Time Use Survey (ATUS).

The ATUS is sponsored by the Bureau of Labor Statistics and is conducted by the U.S. Census Bureau. The survey collects data on how Americans spend their time from employment and education to social and household activities. Like the American Housing Survey (AHS), the ATUS also asks questions about impairment, using the same American Community Survey (ACS) indicators (link to page on indicators). The survey attempts to answer three main questions:

  • What were you doing?
  • Where were you?
  • How long did you do this for?

For some years the survey also asks the question (for 3 randomly selected activities): How did you feel? Respondents then indicate how they felt doing these selected activities across 6 categories (all on a scale of 0 to 6): happy, tired, stressed, sad, pain, meaningful. Finally, respondents were asked if they were interacting with anyone during these activities, including over the phone.

Study sample

This study uses data collected between 2008 and 2014 giving us a total of 87,000 observations with 2,800 of those observations being individuals with mobility impairment.

Analysis Methods

The survey allows us to compare how people with mobility impairments spend their time relative to individuals without mobility impairment, and on some occasions compare how meaningful these daily activities were. We focus on time spent engaging in personal care activities, time spend outside the home and time spent in social and recreational activities.

However, people with mobility impairments differ from people without mobility impairments. For example, 90% of mobility impaired are out of the labor force, 74% do not live with a spouse or partner, 67% are female and their average age is 63. Therefore, we include controls in our analysis for personal characteristics (sex, age, education, employment and school enrollments); family characteristics (number of children, marital status, income, home ownership); location (region, size of metro area); and finally, interview time (day or week, month, year, holiday).

Our dependent measure for this analysis is the number of minutes spent in any given activity. However, this measure is highly skewed with lots of zeros. We mitigate this skewed distribution by using a generalized linear model (GLM) with a log link and poison distribution.

Results

To start our analysis, we asked: What do individuals with mobility impairments do more/less of relative to similar individuals without mobility impairment?

In general people with mobility impairments are less likely than those without impairment to engage in shopping, educational, civic and religious, and household activities, (Table 1). However, we see more time spent in personal care and leisure activities (which are dominated by sleeping and watching television, respectively).

Taking a closer look at the category “household activities” we see that individuals without mobility impairment spend about 50% more time on household activities such as household maintenance and repairs, lawn and garden care, housework and food prep (Table 2). In addition, we see that for people with mobility impairments, household activities are associated with higher pain, fatigue and stress. However, these activities are also associated with more meaning. In other words, household activities are harder for people with mobility impairments and yet they are associated with higher meaning. This is interesting because individuals without impairment report less meaning and happiness when participating in household activities. Results from these analyses can be seen in the figures below.

Table 1. Time spent in activities relative to non mobility impaired: General linear model (GLM) regression.

This table shows the incidence rate ratios (IRRs) for a GLM regression of time spent in activities for people with mobility impairments relative to individuals who are non mobility impaired.

 

Table 2. Time spent in household activities relative to non mobility impaired: GLM regression.

 

This table shows the incidence rate ratios (IRRs) from a GLM regression of time spent in household activities for people with impairments relative to individuals who are non mobility impaired.

As noted above, mobility impaired individuals spend more time on personal care activities (health related self-care, sleep, and grooming) and more time resting (sleeping, resting, low exertion relaxing, watching television). Taking a closer look at personal care activities, focusing specifically on the subcategories of washing, dressing and grooming (WDG), we see that individuals with mobility impairments are less likely to report any time spent on these activities. However, when individuals do spend time in WDG they spend considerably more time doing so. The 6 charts below highlight this and bring in an additional comparison with those not in the labor force (LF). For individuals with mobility impairments we see a similar trend for both groups: less likely to engage in WDG, but more time spent when they do.

Chart 1. Percent who report any time washing, dressing, grooming (WDG).

Chart showing the percent of people who report spending any time washing, dressing, grooming (WDG).

 

Chart 2. For those who do report time WDG: the mean minutes reported.

Chart showing for those who do report time washing, dressing, grooming (WDG), the mean number of minutes reported.

 

 

Beyond WDG we see that individuals with mobility impairments are also less likely to leave the house and less likely to spend some time in social or recreational activities. Interestingly, these findings are similar for those not in the labor force.

Chart 3. Percent who report leaving the house.

Chart showing the percent of people who report leaving the house.

 

Chart 4. Percent who spent some time in social or recreational activities.

Chart showing the percent of people who spent some time in social or recreational activities

 

Next we can begin to examine the associations between WDG and social and recreational activities. We can see that the vast majority of people who leave home and engage in social and recreational activities spend some time in WDG. However, people with mobility impairments are more likely to go out without spending any time in WDG activities.

Chart 5. Percent who left home and didn’t report any time WDG.

Chart shows the percent of people who left home and did not report any time washing, dressing, grooming (WDG).

 

Chart 6. Percent who engaged in social and recreational activities and didn’t report any time WDG.

Chart showing the percent of people who engaged in social and recreational activities and did not report any time washing, dressing, grooming (WDG).

 

Conclusions and Implications

Our findings can be summarized in four main points. People who report mobility impairments:

  • Engage in fewer activities for some categories such as employment, education, housework, social and recreational activities and shopping and spend more time resting and watching TV.
  • Find housework both more difficult but also more meaningful
  • Are less likely to engage in washing, dressing and grooming but when they do, they spend more time doing so.
  • Are more likely to engage in social or recreational activities without washing, dressing or grooming than those without mobility impairment.

These results indicate a relationship between bathing and grooming, and participation in social and recreational activities. In general, people with mobility impairments both participate less outside the home and wash, dress and groom less than individuals without mobility impairments.  Though this does not necessarily indicate a causal relationship between the two, causation is one viable reason for the relationship. If the relationship is in fact causal, these results suggest that interventions to improve bathing capacity may have an effect on social and recreational participation.  Further research is needed to explore these relationships in controlled experiments that can evaluate causal hypotheses.