Expectation of sickness absence duration: A review of European and North American guidelines
Background and goals:
Sickness absence refers to a decision by a worker who for physical, psychological, social, administrative, or cultural reasons does not return to work. Sickness absence is a major concern of health care systems and economies. Both insurers and certifying professionals express the need for evidence-based guidelines on duration of sickness absence. We carried out a review on existing guidelines to understand their methodology through exploring statements on expected sickness absence duration and their evidence.
Methods:
We used a convenient sample of recommendations from a previous project supplemented by material from personal contacts within the EUMASS (European Union of Medicine in Assurance and Social Security). We studied the methodology as published or, if this was lacking, we interviewed a person who was responsible for the guideline. We compared statements and evidence among guidelines.
Results:
We identified 8 guidelines, 5 from private sources (4 USA, 1 Netherlands) and 3 from social insurance institutions (France, Spain, Sweden).
Statements:
All guidelines made statements about expected sickness absence duration according to diagnosis and type of work. Expected sickness absence duration was expressed as mean or optimum in guidelines without definition. US guidelines stratified duration of sickness absence by job class (sedentary/light, medium, heavy, and very heavy). European guidelines added further categories such as specification of work requirement per disease (Sweden), sector (profit, education, civil service, health care Netherlands), or 17 pre-defined job groups (Spain). All but one (France) guidelines used either ICD-9 or ICD-10 to specify disease.
Evidence:
All guidelines reported the use of registered data and scientific literature about sickness absence to produce statements. Type and source of primary data varied considerably: MD Guidelines used data from health insurances, companies, and surveys among different countries, while ODG used a series of data from different sources only in the USA. The AMA guides used the data of MD Guidelines, and ACOEM used ODG’s data. The data of MD Guidelines and/or ODG were implemented in Swedish and French guidelines, while the Dutch guideline was based on a national monitor. Spain used the data of the National Institute of Social Security. How the data from different sources were merged to present the statements was not defined. Almost all guidelines reported the use of a literature review without specifying how this was done or how it was integrated in the statements. All but the Spanish guideline unclearly referred to literature review and consensus as a tool to establish the expectation of sickness absence duration.
Discussion:
Guidelines state the duration of sickness absence in different und unclear manners. The heterogeneous sources of the data, the lack of information on how these data were analyzed, how the scientific literature entered the process and how statements came about raise doubts about the credibility of the statements of sickness absence. The available information does not allow us to determine to what extent these statements are evidencebased.
Conclusion:
The current guidelines on sickness absence duration made different and undefined statements based on indistinct evidence. Hence, whether their results are representative to the population at risk or can be used by clinicians and insurers remain unanswered.
