Large Data Sets in Primary Care Research

Abstract

The other obvious danger of jumping onto the PCMH bandwagon without reformed payment models is that the intense amount of work that is done in care management or non-offi ce-based care (ie, e-visits, phone visits) will go unrewarded. It is clearly in the interest of the insurance industry to encourage family physicians to fully embrace the PCMH model without having to pay for it. If we allow this to happen, we will doom ourselves to a practice model that is high demand but we will not be able to shrink our panel sizes or visit volumes to manageable levels and still keep our offi ce open unless we are paid in a different way.If we step back and look at what kind of payment model would best motivate physicians and their health care teams to perform at the highest level in the care of their patients, it would not be a fee-for-service model. The closer we tie the responsibility for the outcomes of care to both physician and patient, the greater the accountability. Developing primary care capitation payments to family medicine clinics based on population management with specifi c incentives for patient experience markers (a strong correlate to quality) and for key disease management and prevention measures would be our best blend of incentives for payment reform. Our European counterparts have experimented with multiple models and have found that having the bulk of a payment to physicians being a primary care capitation with careful incentives creates an optimal balance. The only way to resource clinics to carry out the work of an effective medical home is to shift more resources into the clinic via payment enhancements but how those payments are structured is critical to getting what we all wantaccessible, rational, quality primary care delivered by care teams led by family physicians.How does this impact residency training? The simple answer is that if the PCMH is the model of care for now and the future, then we need to train residents in an environment that fulfi lls that model. However, given the high stress and high burnout risk, we need to couple our PCMH implementation with education on change management, burnout prevention, and leadership skills. In doing this we will position the next generation of family medicine graduates to be the PCMH leaders of the future.

Big Questions for Social Media Big Data: Representativeness, Validity and Other Methodological Pitfalls

“…It is likely that the dynamics of such events differ significantly from those of less successful ones. In sum, hashtag datasets should be seen as self-selected samples with data "missing not at random" and interpreted accordingly (Allison, 2001;Meiman and Freund, 2012;Outhwaite et al, 2007) All this is not to argue that hashtag datasets are not useful. In contrast, they can provide illuminating glimpses into specific cultural and socio-political conversations.…”

“…Any topic of global interest about Egypt could very easily generate more absolute number of clicks outside the country even if the activity within the country remained much more concentrated in relative proportions. Finally, the size of these datasets makes traditional measures like statistical significance less valuable (Meiman and Freund, 2012), a problem exacerbated by lack of information about the denominator.…”