Limits...
Hospital expenditure at the end-of-life: what are the impacts of health status and health risks?

Geue C, Lorgelly P, Lewsey J, Hart C, Briggs A - PLoS ONE (2015)

Bottom Line: This effect is highly significant (p<0.01) from the last until the 8th quarter before death and influenced by age.Statistically significant effects on costs were found for: smoking status, systolic blood pressure and lung function (FEV1).Health risk measures obtained at baseline provide a good indication of individuals' probability of needing medical attention later in life and incurring costs, despite the small size of the effect.

View Article: PubMed Central - PubMed

Affiliation: Health Economics & Health Technology Assessment, Institute of Health & Wellbeing, University of Glasgow, 1 Lilybank Gardens, Glasgow, G12 8RZ, United Kingdom.

ABSTRACT

Background: It is important for health policy and expenditure projections to understand the relationship between age, death and expenditure on health care (HC). Research has shown that older age groups incur lower hospital costs than previously anticipated and that remaining time to death (TTD) was a much stronger indicator for expenditure than age. How health behaviour or risk factors impact on HC utilisation and costs at the end of life is relatively unknown. Smoking and Body Mass Index (BMI) have featured most prominently and mixed findings exist as to the exact nature of this association.

Methods: This paper considers the relationship between TTD, age and expenditure for inpatient care in the last 12 quarters of life; and introduces measures of health status and risks. A longitudinal dataset covering 35 years is utilised, including baseline survey data linked to hospital and death records. The effect of age, TTD and health indicators on expenditure for inpatient care is estimated using a two-part model.

Results: As individuals approach death costs increase. This effect is highly significant (p<0.01) from the last until the 8th quarter before death and influenced by age. Statistically significant effects on costs were found for: smoking status, systolic blood pressure and lung function (FEV1). On average, smokers incurred lower quarterly costs in their last 12 quarters of life than non-smokers (~7%). Participants' BMI at baseline did show a negative association with probability of HC utilisation however this effect disappeared when costs were estimated.

Conclusions: Health risk measures obtained at baseline provide a good indication of individuals' probability of needing medical attention later in life and incurring costs, despite the small size of the effect. Utilising a linked dataset, where such measures are available can add substantially to our ability to explain the relationship between TTD and costs.

No MeSH data available.


Coefficients for probability of hospitalisation by admission quarter: TTD and age interaction terms.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4352086&req=5

pone.0119035.g001: Coefficients for probability of hospitalisation by admission quarter: TTD and age interaction terms.

Mentions: In Table 3 (Columns (1) and (2)) results for the probit model are presented. An exponential increase of the probability of accessing hospital care was observed from the penultimate to the last quarter of life.


Hospital expenditure at the end-of-life: what are the impacts of health status and health risks?

Geue C, Lorgelly P, Lewsey J, Hart C, Briggs A - PLoS ONE (2015)

Coefficients for probability of hospitalisation by admission quarter: TTD and age interaction terms.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4352086&req=5

pone.0119035.g001: Coefficients for probability of hospitalisation by admission quarter: TTD and age interaction terms.
Mentions: In Table 3 (Columns (1) and (2)) results for the probit model are presented. An exponential increase of the probability of accessing hospital care was observed from the penultimate to the last quarter of life.

Bottom Line: This effect is highly significant (p<0.01) from the last until the 8th quarter before death and influenced by age.Statistically significant effects on costs were found for: smoking status, systolic blood pressure and lung function (FEV1).Health risk measures obtained at baseline provide a good indication of individuals' probability of needing medical attention later in life and incurring costs, despite the small size of the effect.

View Article: PubMed Central - PubMed

Affiliation: Health Economics & Health Technology Assessment, Institute of Health & Wellbeing, University of Glasgow, 1 Lilybank Gardens, Glasgow, G12 8RZ, United Kingdom.

ABSTRACT

Background: It is important for health policy and expenditure projections to understand the relationship between age, death and expenditure on health care (HC). Research has shown that older age groups incur lower hospital costs than previously anticipated and that remaining time to death (TTD) was a much stronger indicator for expenditure than age. How health behaviour or risk factors impact on HC utilisation and costs at the end of life is relatively unknown. Smoking and Body Mass Index (BMI) have featured most prominently and mixed findings exist as to the exact nature of this association.

Methods: This paper considers the relationship between TTD, age and expenditure for inpatient care in the last 12 quarters of life; and introduces measures of health status and risks. A longitudinal dataset covering 35 years is utilised, including baseline survey data linked to hospital and death records. The effect of age, TTD and health indicators on expenditure for inpatient care is estimated using a two-part model.

Results: As individuals approach death costs increase. This effect is highly significant (p<0.01) from the last until the 8th quarter before death and influenced by age. Statistically significant effects on costs were found for: smoking status, systolic blood pressure and lung function (FEV1). On average, smokers incurred lower quarterly costs in their last 12 quarters of life than non-smokers (~7%). Participants' BMI at baseline did show a negative association with probability of HC utilisation however this effect disappeared when costs were estimated.

Conclusions: Health risk measures obtained at baseline provide a good indication of individuals' probability of needing medical attention later in life and incurring costs, despite the small size of the effect. Utilising a linked dataset, where such measures are available can add substantially to our ability to explain the relationship between TTD and costs.

No MeSH data available.