[Rens de Visser]

Jorrit:

It's about the use of long-term care. The research question is what are the determinants of long-term care use and to what extent do these differ over time. This can be as result of for instance policy changes. Later on I divide LTC into formal care use and informal care use, but for now LTC is sufficient.

[Jorrit Gosens]

Aha. Well, again as an example, consider the model:
LTC = a*Income + b*Yearis2013 + c*(Income*Yearis2013)
where Yearis2013 is a dummy distinguishing years 2006 and 2013. If coefficient b is positive and significant, it suggests that policy changes have made more people use LTC. If c is positive and significant it suggests that policy changes have increased the relevance of income; if c is negative it has decreased the relevance of income (i.e., made LTC more accessible to individuals with lower income).

[Rens de Visser]

Jorrit:

It's about the use of long-term care (LTC). The research question is what are the determinants of LTC and how, if so, do these determinants change over time. This might be as a result of policy changes for instance. Later on, I will further divide LTC into formal care and informal care, but that is not relevant for now. To answer this question I have panel data, in which respondents were interviewed in 2004, 2006, 2010 and 2013 (= t). Not all respondents are present in each year. In each consecutive year new respondents were added.

I have done logistic regression analyses for each wave separately (I have also used the fixed effects method (after Hausman testing). Now as I have four different regression models (1 for each year), I would like to compare the coefficients as I would like to know if for instance the effect of gender on LTC use has become stronger, weaker or remained the same in 2013 as it was in 2004. To get to know this I am going to use interaction effects, finally.

I have just typed my first guess into Stata (v13.1):
gen baseline = 1
xi: logit ltcuse i.t*age_y i.t*male i.t*registeredormarried i.t*i.hospitalization2 i.t*i.education i.t*jobsituation i.t*borninothercountry i.t*income i.t*assets i.t*livingincity i.t*havingkids i.t*i.perceivedhealth i.t*adl i.t*iadl i.t*mobility i.t*chronic2ormore i.t*eurodcat baseline, or nocons nolog

This give me a huge regression table! I figuring out how I can post that output here...

[Rens de Visser]

(two parts of the output http://postimg.org/image/v5o34pu59/ http://postimg.org/image/uhoiotrlv/)

If I for example interpret the results for age, gender (1 = male) and education, is this interpretation correct?
- The effect of age for respondents in 2013 is 1.015 times that for respondents in 2004. That means that over time, comparing 2004 with 2013, age has become a stronger predictor for LTC use.
- The odds of using LTC is 0.836 times higher for men in 2013 compared to men in 2004, meaning that men are less likely to use LTC in 2013 then they were in 2004. The effect has become weaker.
- The odds ratio for the medium educated in 2013 is 1.137, which means that the odds of using LTC is 1.137 times higher for those in 2013 compared to the medium educated in 2004. The odds ratio for the high educated in 2013 is 1.477, which means that the odds of using LTC is 1.477 times higher for those in 2013 compared to the high educated in 2004.