Evaluating eHealth How to Make Evaluation More Methodologically Robust 英文参考文献.docVIP

Essay EvaluatingeHealth:HowtoMakeEvaluationMore MethodologicallyRobust RichardJamesLilford1*,JoFoster1,MikePringle2 1DivisionofPrimaryCare,SchoolofHealthandPopulationSciences,UniversityofBirmingham,Birmingham,UnitedKingdom,2SchoolofCommunityHealthSciences, UniversityofNottingham,Nottingham,UnitedKingdom Thisisthethirdinaseriesofthreearticles onevaluationofeHealth. tion technology systems comes from 4 benchmarkresearchinstitutions’’andthat ‘‘littleevidenceisavailableontheeffectof multifunctional commercially developed systems’’ [3] such as those that are increasinglyimplementedintheNHS. Because these evaluations would inevi- tably be commissioned under intense politicalandmediaattention,NPFITtook the principled decision to contract the University of Birmingham to commission theresearchindependentlyunderDepart- mentofHealthprocurementrules.NPFIT could thus influence what was commis- sioned (i.e., research topics) but not the resultsobtained. SummaryPoints N Evaluation of information tech- nology(IT)systemsoftenrequires amixedmethodsapproach. eHealth—the organisation and delivery of health services and information using information technology (IT) systems—is playing an increasingly important role in shaping health care systems. However, as Catwell and Sheikh described in the first article in this series [1], IT systems can introduceharmsaswellasbenefits.Catwell andSheikharguedforageneralschemeof evaluation starting with careful specifica- N External evaluations have many advantages,especiallyintermsof standardisation, independence, and the possibility of using controlled before and after designs. N Difficulties arise when commis- sioners ask external evaluations to also provide formative assess- ments designed to assist in the implementation itself. Under these circumstances the summa- tive results, which encapsulate theoverallbenefitsandharmsof a system, may be rendered less generalisable. tion of need and pre-implementation testing in their article. Thi