Summary health measures

No single measure has been agreed upon as the best approach. Two key contender metrics are the Health-Adjusted Life Expectancy (HALE) and the Disability Adjusted Life Year (DALY) .

HALE is an expansion of the concept of life expectancy. Years lived with illness or disability, in the population, are weighted for severity and subtracted from the total years lived, in a life table. It is most commonly used to measure the health of a population compared with others using a single number. The drawback of this measure is that it cannot easily be decomposed into its contribution by diseases and risk factors. For that reason, a health-gap measure, DALY was constructed.

The DALY was introduced in 1993 when the Harvard School of Public Health, in collaboration with The World Bank, assessed the global burden of disease for the year 1990 in the inaugural Global Burden of Disease (GBD) Study .

This study generated the most comprehensive and consistent set of estimates of mortality and morbidity by age, sex and world region ever produced. DALYs used in burden measurement are the gap between current health status and an ideal situation where everyone lives into old age free of disease and disability. One DALY can be thought of as the loss of 1 year of ‘healthy’ life. The DALY combines in one measure the time lived with disability and the time lost due to premature mortality.

DALYs are calculated as the sum of the years of life lost (YLL) due to premature mortality in the population and the years lived with disability (YLD) for incident cases of the health condition:

YLL corresponds to the number of deaths multiplied by a standard life expectancy at the age at which death occurs. For this purpose, a standard life table with a life expectancy at birth of 80 years in males and 82.5 years in females has been chosen.

To estimate YLD for a particular cause in a particular time period, the number of incident cases in that period is multiplied by the average duration of the disease and a weight factor that reflects the severity of the disease on a scale from 0 (perfect health) to 1 (disease state equivalent to death) .

Dismod’s general disease model

The model is modified as necessary for specific diseases and conditions ( Fig. 1 ). The four parameters specific to the model include

• •
  

  incidence or prevalence rate;

  
  
• •
  

  remission rate;

  
  
• •
  

  cause-specific mortality rate; and

  
  
• •
  

  natural mortality rate.

Disease model.

The last parameter can be obtained through population-based all-cause mortality tables, while the other three must be determined for each specific condition.

DALYs are limited somewhat by the lack of population-based incidence data for many of the conditions; but, the DISMOD software has been developed to estimate a consistent set of epidemiological parameters (incidence, prevalence, mortality risk, cure and duration) for each disease. It is more realistic than assuming prevalence is the product of incidence and duration, as it accounts for competing causes of mortality in an iterative fashion. This is particularly important for chronic conditions with low rates of remission (defined for GBD study purposes as medication-free cure) and cause-specific mortality.

Case definitions

The measurement of burden needs clear definitions to be able to identify cases with consistency. A drawback of the majority of chronic musculoskeletal conditions is that there is no single diagnostic test or pathognomonic clinical or self-reported feature that defines them. For the purpose of measuring burden or disability related to a condition, the case definition also needs to incorporate some description of the impact on daily activity and loss of function; yet, this is rarely recorded routinely in national data collections or health surveys.

Case finding may be through health interview surveys (HISs), without the opportunity for examination or investigation to establish cause. Many health surveys will ask about ‘arthritis’ or ‘rheumatism’ generally, but, for the purposes of health policy and research planning, there is also a need to know who has certain musculoskeletal conditions such as rheumatoid arthritis or osteoporosis, and, for each condition, a specific case definition is required. Case definition that is based on clinical features requiring an examination, laboratory tests or imaging and X-ray studies may be more accurate, but are also more costly to acquire in a large-scale population-based sample. In addition, they may be of limited value in assessing the burden unless symptoms are also present, as there is no real impact on functioning.

Incidence and prevalence

The number of people affected by a musculoskeletal condition can be considered in different ways. Incidence refers to the number of new cases occurring over a predefined time period while prevalence refers to the number of existing cases for a population at either a particular point in time (point prevalence) or during a specified period (period prevalence). For the purpose of estimating potential for the burden of disease on health services, the point prevalence is the most useful statistic. Conditions in which the point of onset is clearly identifiable, such as fractures resulting from trauma, lend themselves well to measures of incidence. However, as most musculoskeletal conditions have a gradual progressive onset, it is problematic to determine when a condition such as osteoarthritis or osteoporosis becomes a definable case. In this regard, measures of prevalence are more easily measured. Many musculoskeletal conditions, for example, rheumatoid arthritis or osteoarthritis, are chronic diseases for which we have no cure and as such are considered to be forever prevalent after onset. Yet we know that the associated burden with these conditions will demonstrate considerable variability and fluctuations in the level of symptoms and loss of function over time. Other conditions such as back and neck pain and soft-tissue disorders may have much shorter duration sometimes without recurrence, or other times with multiple episodes of disability but long disability-free periods interspersed.

Impact of musculoskeletal conditions

The full impact on those who have been identified as affected must be measured in terms of the individual consequences as well as those on the population and the consequences for society ( Table 1 ). However, depending on the purpose for measuring, only some of the components may be chosen. For example, in the GBD 2005 Study ( http://www.globalburden.org ), only the individual physical and psychological disability and mortality burden are being measured, as these are aspects that are more objective and directly measured and are also more likely to be sensitive or responsive to measure change.

Physical and psychological impact on the individual

The health consequences for the individual vary from short-term pain and impaired activity to premature death. This impact can be considered within the construct of the WHO International Classification of Functioning, Disability and Health (ICF) .

The ICF considers the impairment of body functions and structures, limitation of activities and the restriction of participation in all of its aspects of social, family and occupational roles that may relate to the health condition. All these will be influenced by both environmental and personal contextual factors. The impact on body structures can be assessed, for instance, by loss of bone mass or fracture in osteoporosis or loss of cartilage in osteoarthritis. Generic and disease-specific instruments are used to measure limitation of activities and restriction of participation.

Mortality

Mortality associated with musculoskeletal conditions must also be measured. Mortality for individual conditions will be addressed in each article. For the majority of musculoskeletal conditions, the case fatality rate and condition-specific mortality is low, with some notable exceptions. Mortality is increased from all osteoporotic fractures , yet the death will be attributed to the injury rather than to the underlying osteoporosis. In the GBD Study 2005, age- and gender-specific population-based measures of bone mineral density will be evaluated as a risk factor for hip fracture. While this association is well established, the link between bone mineral density and subsequent death from fracture is not so straightforward, but will be explored in the GBD Study comparative risk analysis. Conditions such as active, untreated rheumatoid arthritis and systemic vasculitis will also lead to premature loss of life, yet these conditions are significantly underreported on death certification, with the death usually being attributed to end-organ failure and associated co-morbidities rather than to the underlying musculoskeletal condition . For others, such as osteoarthritis where there is a low level of increased mortality in a very prevalent condition, there may be a much greater mortality burden than has previously been recognised.

Impact on the individual as defined by the ‘Health-state’

While it is important to consider musculoskeletal conditions as a whole to describe their overall burden in the population, it remains methodologically challenging to account for the variability in welfare and social consequences of the different conditions in different communities. The nature of the impact on the individual will vary at each stage of a condition and this is described by the ‘health-state’. A summary measure of the burden of musculoskeletal conditions needs to incorporate multiple variables including the range of different health states of each condition and the numbers of individuals within and moving between the different stages over time. This then needs to be coupled with the value that society places on the loss of function (referred to by the GBD Study as the ‘disability weight’), associated with each stage. Many musculoskeletal conditions are persistent and progressive, and the individual will move from an early and/or mild stage to a late and/or severe stage. Other conditions such as back pain and gout fluctuate significantly with quite abrupt transitions between health states with short periods of extreme burden interspersed between much longer periods of no or relatively low level burden. Transitions between health states are not well addressed in current summary health measures.

The health state is measured by the impact on a relevant set of health domains. The most important health domains for musculoskeletal conditions have been considered to be overall well-being, general health, physical health, social health, mental health and barriers to participation . For the purposes of the GBD Study, health loss associated with individual loss of physical and emotional function is measured rather than general welfare loss and loss of participation.

The Musculoskeletal Expert Group for the GBD Study 2005 ( http://www.globalburden.org/ ) has described a range of the most frequently occurring health states or sequelae for the main musculoskeletal conditions including osteoarthritis, rheumatoid arthritis, back pain, neck pain, gout and other musculoskeletal conditions to encompass the remaining broad spectrum of musculoskeletal disorders.

Economic impact to individual and society

The cost to society as well as to the individual is another factor for consideration. This should incorporate the impact on carers and families and both the direct costs related to living with the disease conditions and indirect costs related to productivity losses. This economic cost (or cost-of-illness) of a condition often has the most influence on priority setting for strategies for prevention and treatment. Ideally, they should take a societal perspective encompassing all costs no matter who incurs them, and all benefits regardless of who receives them; however, this is not often the case and they will usually take a more restricted perspective, for example, from the government or health sector who are the payers. Yet, many ‘cost-of-illness’ studies and economic impact assessments are done in isolation, largely for advocacy purposes, and may over-inflate the burden by the inclusion of overlapping conditions and indirect cost estimates. In a world where there will always be constraints on the resources for health and social care, there will also be ‘competition’ to be the disease with the ‘biggest’ burden and the ‘most’ economic loss; hence, it is important that these studies are conducted with standardised rigorous methodology. The marked variability in quality and quantity of economic cost information in different societies and this tendency for attribution to be generously applied for individual illnesses are some of the reasons put forward by the GBD 2005 Study core team to measure all diseases simultaneously and to focus disease burden estimates on individual health-related loss of function rather than incorporating societal economic losses.

Direct and indirect costs of musculoskeletal conditions

The economic impact of any health condition can be considered in terms of the gross loss to the economy occasioned by the illness . The cost-of-illness is the monetary burden on society of the morbidity and premature mortality associated with a particular illness . It can be measured as the societal cost of providing services related to the delivery of health care and social support, but it can also be considered as the personal cost to the patient, family or immediate community. These costs can be considered in terms of the direct, indirect or intangible costs associated with the condition.

Direct costs are those health and social care costs directly associated with prevention, detection, treatment and rehabilitation, including care in the community. The value of the direct costs represents resources that would be freed for other uses in the health and social care system as well as in the individuals’ spending power, if the disease did not happen. The degree to which social care costs are included in different studies depends on who pays, and are often underestimated if the cost largely falls on individuals rather than on society. Direct costs are relatively easy to measure either by using a top-down approach, dividing the total health expenditure between different diseases, or a bottom-up approach in which the number of health-care services for individuals with a specific condition can be counted and costed. The certainty of case definition is better in the bottom-up” approach. The total costs of disease can then be estimated.

Direct costs include those related to community, outpatient and inpatient care including diagnostic tests, procedures, aids and devices, prescription drugs and over-the-counter medications. These expenditures may be borne solely or in combination by the patient, their health insurer, an employer or a local or national government agency. Direct costs also include other expenditures, generally paid for by the patient, such as transportation to and from the doctor or other allied health worker, non-physician service use such as complementary therapy or expenditures to adapt the home environment to make functioning easier. In addition, there are the direct costs associated with changes in living status. The direct costs borne by the person are called out-of-pocket expenditure and the amount will vary between countries, depending on what is covered by the health insurance system. Co-morbidities can make it difficult to know what costs to allocate to which condition, especially costs related to rehabilitation and changes in living status.

Indirect costs result from the consequences of a condition such as limitation of usual activities; and the chronic physical disability associated with musculoskeletal conditions has potential to have a major impact on these costs. Lost work productivity is important in these conditions but not often included in economic evaluations . A number of approaches including the human capital approach and the friction cost approach attempt to estimate the cost of foregone production to society. The human capital approach uses the full replacement costs irrespective of whether the worker is replaced, while the friction cost approach only considers the productivity costs during the period needed to restore initial production level and the replacement worker is from the labour market. This results in lower estimates of lost productivity costs but is probably more realistic. Indirect costs are strongly influenced by patients who are in the workforce as employment-related costs are easier to quantify. Economic impact on children, women, elderly and those unemployed is not adequately captured as these losses, such as loss of school hours or household work, are more difficult to value in monetary terms. There are also the losses attributable to the condition preventing the person from having better paying jobs or reducing their employment opportunities through the need to work part-time or to retire from the workforce early. The additional costs related to self-care, maintenance of a home, schooling or parenting are also included, which may be incurred by both patients and their carers.

There are also the intangible costs associated with loss in function, increased pain and reduced quality of life of patients, families and carers. These include the costs in terms of lost opportunities. Intangible costs are very difficult to quantify and extremely difficult to give a monetary value to. However, they are particularly important for musculoskeletal conditions, as disability is a significant outcome with limitations in activities of daily living, reduction in leisure and community activities, pain, depression and anxiety and reduced general health. The converse of the intangible costs of arthritis is the benefit a patient receives from effective treatment. If measurement of the costs of care is to have relevance for clinical decision making, it must also consider these effects of disease and therapy on the patient’s health and well-being. Quantifying ‘intangibles’ with generic summary health statistics such as a Quality Adjusted Life Year (QALY) or a DALY allows comparative economic evaluation without the need to monetarise.

There are a large number of cost items that can therefore be used in the estimation of the cost-of-illness ( Table 2 ) but not all these items are considered in health economic evaluations. Such omissions need to be recognised as they can affect the interpretation and comparability of the data. Complete data on all variables required for full economic analyses are rarely available. Modelling techniques are used to allow sensitivity analyses of a range of assumptions to be evaluated. They can incorporate real-life data and extend beyond the time of the trials to allow for the at-risk population. However, these assumptions can be open to manipulation and overestimates, making standardisation of these methods a priority. Attempts are being made to standardise the cost domains and modelling techniques for musculoskeletal conditions .

Table 2

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

The global burden of musculoskeletal conditions – Why is it important? Rheumatoid arthritis Seronegative spondyloarthritis Osteoporosis and fragility fractures Generalised musculoskeletal pain syndromes Epidemiology of gout: An update

Stay updated, free articles. Join our Telegram channel

Join