Measuring costs: What is counted and who is accountable?

Research output: Contribution to journalReview article

6 Citations (Scopus)

Abstract

Efficiency and COSt containment have only recently become driving concerns of health care providers and payers. But growing concerns about cost have made economic evaluations a more common and vital aspect of clinical research. Whether such studies actually enhance the decision-making capability of health care providers and insurers depends on their methodological design and rigor. Unfortunately, frequently the users of these studies and, in some cases, the producers of the studies are not sensitive to how methodological choices can influence their findings and thus bias comparisons of different interventions. Variation in the quality of cost information derived from economic evaluations arises, in large measure, because researchers use different methods to estimate costs. Although some experts in the field have advanced guidelines for estimating costs, there is no universally accepted set of rules or guidelines. As a result, it is left up to each researcher to decide how costs will be identified, measured, and reported within economic evaluations. While there is disagreement about costing methods, it is generally agreed that costs are best estimated at the individual level. This entails a unique cost estimate for each study subject based on the value of the resources used by the subject over the study period. This costing approach is referred to as micro-costing. Micro-costing involves constructing utilization histories for individuals and then valuing these services by a set of appropriate prices. Studies relying on micro- costing techniques are very time-intensive since they require assessing service use data from many different sources and tracking these data over the study period, as well as constructing prices that match the services being used. These studies can take years to complete. In addition, because services and prices may be measured with random and nonrandom error, results may be imprecise enough as to render the findings unreliable. While it is recognized that micro-costing is the preferred method for estimating costs, it is sometimes infeasible. Time and financial constraints, for example, may require a less precise study, for example one that estimates an order-of- magnitude effect or that uses a costing approach treating all patients and service settings as more-or-less interchangeable. In many instances, cruder cost analyses may even be appropriate. For example, when forecasting the effects of a possible epidemic, the cost estimate is only one aspect of social consequences. Aggregate estimates may also be sufficient in situations where there is great uncertainty about treatment performance or when costs are only incurred many years into the future and can, therefore, be heavily discounted. Cost estimates constructed from aggregate utilization statistics and average measures of resource use are referred to as macro-costing measures. The promise of economic evaluations as a decision making tool hinges on the quality of the information forthcoming from the research community. The intent of this paper is to open up the black box of costing methods and explore the options and trade-offs associated with the different ways that cost outcomes can be estimated. The beginning section of the paper presents a general discussion of the costing problem and the micro- and macro-costing techniques commonly used to estimate cost outcomes. The second section focuses attention on key methodological assumptions and choices that influence the quality of the cost outcome. At the end of this section, a framework for choosing between alternative costing techniques is developed. In the third section of the paper, procedures for microcost estimation are described in the context of a mental health application. Emphasis is on evaluating data sources, identifying measurement errors, maximizing internal consistency, and estimating unit costs. The last section discusses two ways to manage variation in quality among economic evaluations. It stresses the responsibilities of authors to disclose metholodogical information that affects the validity and reliability of their cost estimates.

Original languageEnglish (US)
Pages (from-to)114-128
Number of pages15
JournalDisease Management and Clinical Outcomes
Volume1
Issue number4
StatePublished - Jan 1 1998

Fingerprint

Costs and Cost Analysis
Cost-Benefit Analysis
Information Storage and Retrieval
Health Personnel
Decision Making
Research Personnel
Guidelines
Insurance Carriers
Research
Reproducibility of Results
Uncertainty
Mental Health

All Science Journal Classification (ASJC) codes

  • Medicine (miscellaneous)

Cite this

@article{a5fe8f99f7d7408f9c9c6233a833b06a,
title = "Measuring costs: What is counted and who is accountable?",
abstract = "Efficiency and COSt containment have only recently become driving concerns of health care providers and payers. But growing concerns about cost have made economic evaluations a more common and vital aspect of clinical research. Whether such studies actually enhance the decision-making capability of health care providers and insurers depends on their methodological design and rigor. Unfortunately, frequently the users of these studies and, in some cases, the producers of the studies are not sensitive to how methodological choices can influence their findings and thus bias comparisons of different interventions. Variation in the quality of cost information derived from economic evaluations arises, in large measure, because researchers use different methods to estimate costs. Although some experts in the field have advanced guidelines for estimating costs, there is no universally accepted set of rules or guidelines. As a result, it is left up to each researcher to decide how costs will be identified, measured, and reported within economic evaluations. While there is disagreement about costing methods, it is generally agreed that costs are best estimated at the individual level. This entails a unique cost estimate for each study subject based on the value of the resources used by the subject over the study period. This costing approach is referred to as micro-costing. Micro-costing involves constructing utilization histories for individuals and then valuing these services by a set of appropriate prices. Studies relying on micro- costing techniques are very time-intensive since they require assessing service use data from many different sources and tracking these data over the study period, as well as constructing prices that match the services being used. These studies can take years to complete. In addition, because services and prices may be measured with random and nonrandom error, results may be imprecise enough as to render the findings unreliable. While it is recognized that micro-costing is the preferred method for estimating costs, it is sometimes infeasible. Time and financial constraints, for example, may require a less precise study, for example one that estimates an order-of- magnitude effect or that uses a costing approach treating all patients and service settings as more-or-less interchangeable. In many instances, cruder cost analyses may even be appropriate. For example, when forecasting the effects of a possible epidemic, the cost estimate is only one aspect of social consequences. Aggregate estimates may also be sufficient in situations where there is great uncertainty about treatment performance or when costs are only incurred many years into the future and can, therefore, be heavily discounted. Cost estimates constructed from aggregate utilization statistics and average measures of resource use are referred to as macro-costing measures. The promise of economic evaluations as a decision making tool hinges on the quality of the information forthcoming from the research community. The intent of this paper is to open up the black box of costing methods and explore the options and trade-offs associated with the different ways that cost outcomes can be estimated. The beginning section of the paper presents a general discussion of the costing problem and the micro- and macro-costing techniques commonly used to estimate cost outcomes. The second section focuses attention on key methodological assumptions and choices that influence the quality of the cost outcome. At the end of this section, a framework for choosing between alternative costing techniques is developed. In the third section of the paper, procedures for microcost estimation are described in the context of a mental health application. Emphasis is on evaluating data sources, identifying measurement errors, maximizing internal consistency, and estimating unit costs. The last section discusses two ways to manage variation in quality among economic evaluations. It stresses the responsibilities of authors to disclose metholodogical information that affects the validity and reliability of their cost estimates.",
author = "Nancy Wolff",
year = "1998",
month = "1",
day = "1",
language = "English (US)",
volume = "1",
pages = "114--128",
journal = "Disease Management and Clinical Outcomes",
issn = "1088-3371",
publisher = "Elsevier BV",
number = "4",

}

Measuring costs : What is counted and who is accountable? / Wolff, Nancy.

In: Disease Management and Clinical Outcomes, Vol. 1, No. 4, 01.01.1998, p. 114-128.

Research output: Contribution to journalReview article

TY - JOUR

T1 - Measuring costs

T2 - What is counted and who is accountable?

AU - Wolff, Nancy

PY - 1998/1/1

Y1 - 1998/1/1

N2 - Efficiency and COSt containment have only recently become driving concerns of health care providers and payers. But growing concerns about cost have made economic evaluations a more common and vital aspect of clinical research. Whether such studies actually enhance the decision-making capability of health care providers and insurers depends on their methodological design and rigor. Unfortunately, frequently the users of these studies and, in some cases, the producers of the studies are not sensitive to how methodological choices can influence their findings and thus bias comparisons of different interventions. Variation in the quality of cost information derived from economic evaluations arises, in large measure, because researchers use different methods to estimate costs. Although some experts in the field have advanced guidelines for estimating costs, there is no universally accepted set of rules or guidelines. As a result, it is left up to each researcher to decide how costs will be identified, measured, and reported within economic evaluations. While there is disagreement about costing methods, it is generally agreed that costs are best estimated at the individual level. This entails a unique cost estimate for each study subject based on the value of the resources used by the subject over the study period. This costing approach is referred to as micro-costing. Micro-costing involves constructing utilization histories for individuals and then valuing these services by a set of appropriate prices. Studies relying on micro- costing techniques are very time-intensive since they require assessing service use data from many different sources and tracking these data over the study period, as well as constructing prices that match the services being used. These studies can take years to complete. In addition, because services and prices may be measured with random and nonrandom error, results may be imprecise enough as to render the findings unreliable. While it is recognized that micro-costing is the preferred method for estimating costs, it is sometimes infeasible. Time and financial constraints, for example, may require a less precise study, for example one that estimates an order-of- magnitude effect or that uses a costing approach treating all patients and service settings as more-or-less interchangeable. In many instances, cruder cost analyses may even be appropriate. For example, when forecasting the effects of a possible epidemic, the cost estimate is only one aspect of social consequences. Aggregate estimates may also be sufficient in situations where there is great uncertainty about treatment performance or when costs are only incurred many years into the future and can, therefore, be heavily discounted. Cost estimates constructed from aggregate utilization statistics and average measures of resource use are referred to as macro-costing measures. The promise of economic evaluations as a decision making tool hinges on the quality of the information forthcoming from the research community. The intent of this paper is to open up the black box of costing methods and explore the options and trade-offs associated with the different ways that cost outcomes can be estimated. The beginning section of the paper presents a general discussion of the costing problem and the micro- and macro-costing techniques commonly used to estimate cost outcomes. The second section focuses attention on key methodological assumptions and choices that influence the quality of the cost outcome. At the end of this section, a framework for choosing between alternative costing techniques is developed. In the third section of the paper, procedures for microcost estimation are described in the context of a mental health application. Emphasis is on evaluating data sources, identifying measurement errors, maximizing internal consistency, and estimating unit costs. The last section discusses two ways to manage variation in quality among economic evaluations. It stresses the responsibilities of authors to disclose metholodogical information that affects the validity and reliability of their cost estimates.

AB - Efficiency and COSt containment have only recently become driving concerns of health care providers and payers. But growing concerns about cost have made economic evaluations a more common and vital aspect of clinical research. Whether such studies actually enhance the decision-making capability of health care providers and insurers depends on their methodological design and rigor. Unfortunately, frequently the users of these studies and, in some cases, the producers of the studies are not sensitive to how methodological choices can influence their findings and thus bias comparisons of different interventions. Variation in the quality of cost information derived from economic evaluations arises, in large measure, because researchers use different methods to estimate costs. Although some experts in the field have advanced guidelines for estimating costs, there is no universally accepted set of rules or guidelines. As a result, it is left up to each researcher to decide how costs will be identified, measured, and reported within economic evaluations. While there is disagreement about costing methods, it is generally agreed that costs are best estimated at the individual level. This entails a unique cost estimate for each study subject based on the value of the resources used by the subject over the study period. This costing approach is referred to as micro-costing. Micro-costing involves constructing utilization histories for individuals and then valuing these services by a set of appropriate prices. Studies relying on micro- costing techniques are very time-intensive since they require assessing service use data from many different sources and tracking these data over the study period, as well as constructing prices that match the services being used. These studies can take years to complete. In addition, because services and prices may be measured with random and nonrandom error, results may be imprecise enough as to render the findings unreliable. While it is recognized that micro-costing is the preferred method for estimating costs, it is sometimes infeasible. Time and financial constraints, for example, may require a less precise study, for example one that estimates an order-of- magnitude effect or that uses a costing approach treating all patients and service settings as more-or-less interchangeable. In many instances, cruder cost analyses may even be appropriate. For example, when forecasting the effects of a possible epidemic, the cost estimate is only one aspect of social consequences. Aggregate estimates may also be sufficient in situations where there is great uncertainty about treatment performance or when costs are only incurred many years into the future and can, therefore, be heavily discounted. Cost estimates constructed from aggregate utilization statistics and average measures of resource use are referred to as macro-costing measures. The promise of economic evaluations as a decision making tool hinges on the quality of the information forthcoming from the research community. The intent of this paper is to open up the black box of costing methods and explore the options and trade-offs associated with the different ways that cost outcomes can be estimated. The beginning section of the paper presents a general discussion of the costing problem and the micro- and macro-costing techniques commonly used to estimate cost outcomes. The second section focuses attention on key methodological assumptions and choices that influence the quality of the cost outcome. At the end of this section, a framework for choosing between alternative costing techniques is developed. In the third section of the paper, procedures for microcost estimation are described in the context of a mental health application. Emphasis is on evaluating data sources, identifying measurement errors, maximizing internal consistency, and estimating unit costs. The last section discusses two ways to manage variation in quality among economic evaluations. It stresses the responsibilities of authors to disclose metholodogical information that affects the validity and reliability of their cost estimates.

UR - http://www.scopus.com/inward/record.url?scp=0031737344&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0031737344&partnerID=8YFLogxK

M3 - Review article

AN - SCOPUS:0031737344

VL - 1

SP - 114

EP - 128

JO - Disease Management and Clinical Outcomes

JF - Disease Management and Clinical Outcomes

SN - 1088-3371

IS - 4

ER -