On 2017 Apr 06, Cicely Saunders Institute Journal Club commented:

This paper was reviewed in the Cicely Saunders Institute Journal Club on 1st March 2017.

The paper reports on the independent associations of income, education and multimorbidity with aggressiveness of end of life care, using rich data from the Health and Retirement Study (HRS) linked to the National Death Index (NDI) and Medicare data. We enjoyed discussing this paper and agree with the authors about the importance of understanding social determinants alongside clinical determinants of care at the end of life. We liked the measure of multimorbidity used, comprising of items related to comorbidity, functional limitations and geriatric syndromes, and thought this comprehensive approach was useful in this population. We were not sure why the sample was limited to fee-for-service patients and whether this may have disproportionately excluded some socio-economic groups. As a non-US audience we would have welcomed some further justification for restricting the sample in this way and discussion of potential limitations, perhaps using a CONSORT diagram to explain the steps. We enjoyed the presentation of the bivariate associations in the bar charts, helping us to understand the U and J shaped relationships between some of the variables. More information about what exactly the income variable was capturing (i.e. including pensions or not, and whether the household income was total or averaged across the number of people in the household) would have been useful. We also felt the race variable was broad and interpretation of the results would have benefited from more refined categories. Overall the paper sparked a good discussion about the importance of measuring social determinants and illness and function related factors in end of life populations and how best to capture these.

On 2017 Apr 06, Cicely Saunders Institute Journal Club commented:

This paper was reviewed in the Cicely Saunders Institute Journal Club on 1st March 2017.

The paper reports on the independent associations of income, education and multimorbidity with aggressiveness of end of life care, using rich data from the Health and Retirement Study (HRS) linked to the National Death Index (NDI) and Medicare data. We enjoyed discussing this paper and agree with the authors about the importance of understanding social determinants alongside clinical determinants of care at the end of life. We liked the measure of multimorbidity used, comprising of items related to comorbidity, functional limitations and geriatric syndromes, and thought this comprehensive approach was useful in this population. We were not sure why the sample was limited to fee-for-service patients and whether this may have disproportionately excluded some socio-economic groups. As a non-US audience we would have welcomed some further justification for restricting the sample in this way and discussion of potential limitations, perhaps using a CONSORT diagram to explain the steps. We enjoyed the presentation of the bivariate associations in the bar charts, helping us to understand the U and J shaped relationships between some of the variables. More information about what exactly the income variable was capturing (i.e. including pensions or not, and whether the household income was total or averaged across the number of people in the household) would have been useful. We also felt the race variable was broad and interpretation of the results would have benefited from more refined categories. Overall the paper sparked a good discussion about the importance of measuring social determinants and illness and function related factors in end of life populations and how best to capture these.