Thursday, October 17, 2013
The Dartmouth Institute has just published its Atlas of areal differences in utilization of prescription drugs by Medicare Part D recipients. The Atlas--unsurprisingly but disturbingly--details significant differences. Pharmaceutical interventions are classified as effective, discretionary (where there is diagnostic or therapeutic uncertainty), and likely to be harmful in the patient population at issue. A caveat, however, is that the report measured prescriptions filled and thus may underestimate actual provider behavior.
An initial variation involved sheer numbers of prescriptions, with a high average of 63 per year in Miami and a low average of 39 per year in Colorado (overall, the average was 49 standardized 30 day prescriptions filled per year per Part D beneficiary). In general, the Mountain West had the lowest prescription average and the Rust Belt and Appalachian states the highest. These differences could not be explained primarily by overall burden of disease but instead appear to reflect variations in provider prescribing practices. For example, the American Heart Association recommends use of beta blockers in heart attack patients for three years post-attack. However, rates of prescriptions for these drugs in the first six months ranged from highs of 94% to lows of under 68%, and persistence in the next six months was only slightly lower, ranging from highs of 92% to lows of under 68%. Variations in statin use were even greater, ranging from just over 91% in Ogden, Utah, to below 45% in Abilene, Texas. Interestingly, there was little correlation between effective use of beta blockers and effective use of statins.
The other two therapies analyzed in the Atlas were treatment of diabetes and treatment of patients with fragility fractures. Diabetic patients fared somewhat better than heart attack patients, albeit still with significant variations. Osteoporotic patients, however, fared dismally, receiving a high of 28% and a low of 7% with filled prescriptions for drug to combat osteoporosis after fragility fractures in sites other than the hip (such treatment is recommended to decrease the risk of future hip fractures).
Most interesting of all, there was no correlation between drug expenditures and measures of effective care. In other words, patients in some regions may be spending a great deal on their drugs (paid for under Part D), but receiving far less benefit that patients in other regions who spend a great deal less.
October 17, 2013 in Access, Chronic Care, CMS, Consumers, Cost, Drug and Device, Health Care, Health Care Costs, Medicare, Prescription Drugs, Quality, Spending | Permalink | Comments (0) | TrackBack (0)
Monday, September 23, 2013
Chilmark Research produces evidence-based reports of health IT and market trends in the health IT industry.
A recently issued Chilmark report, 2013 Clinical Analytics for Population Health Market Trends Report, which I have not read because it costs $4500, details the conflicting interests of clinicians and payers with respect to insights gleaned from data analytics. The hope of EHRs in combination with data analytics is better patient health, for example through alerts about needed preventive measures or care management strategies. But different payment may reimburse categories of care differently--so a diabetic covered by one type of payment structure might get reminders when her counterpart with different coverage might not. Even worse, patients whose prognosis is seen as "hopeless" through the predictive lens of analytics might get very different treatment recommendations under cost-conscious reimbursement structures.
Cora Sharma's post on the Chilmark blog details these likely conflicts with chilling precision.
September 23, 2013 in Access, Accountable Care Organizations, Chronic Care, Consumers, Cost, Coverage, Disparities, Electronic Medical Records, Health Care Costs, Insurance, Prevention, Private Insurance | Permalink | Comments (0) | TrackBack (0)
Thursday, September 5, 2013
Don't miss a fascinating article in the August 30th issue of Science, "Poverty Impedes Cognitive Function." The article contends that there is a causal explanation for the correlation between poverty and disfunctional behavior, such as the failure to keep medical appointments or to employ healthy behaviors. Put crudely, the connection is that people in poverty have to think about so much just to keep going that they don't have the cognitive bandwidth to make carefully reasoned decisions.
The authors of the article, Anandi Mani, Sendhil Mullainanthan, Eldar Shafir, and Jiaying Zhao, present two studies in support of their claim. The first study involved four experiments in which shoppers at a New Jersey mall were paid participants. The income level of the shoppers varied, from the bottom quartile of US income to over $70,000. In the first experiment, participants were asked to think about a decision about how to pay for car repairs, and were randomized to inexpensive ($150) or expensive ($1500) costs of the repair. They were then asked to perform simple cognitive tests on a computer. Among those asked to think about the inexpensive repair, there were no significant differences by income level in performance of the cognitive task. By contrast, there were significant differences in performance by income among those confronted with the more expensive repair. Variations on this experiment involved problems where sums of money were not involved (to control for math anxiety), incentives in the form of getting paid for getting the right answers on the cognitive tests, and situations in which participants came to a decision about the financial problem, engaged in intervening activities, and then were asked to perform the cognitive tests. Each of these variations produced results similar to the initial experiment: the performance of people in poverty on the cognitive tests was significantly associated with the expensive repair, but the performance of those in higher income groups was not.
In the authors' second study, participants were a random sample of sugar cane farmers in Tamil Nadu in southern India. They were interviewed before and after the cane harvest. Pre-harvest the farmers faced more significant financial pressures (as measured by criteria such as numbers of pawned items, numbers of loans, and the like) than post-harvest. Performance on cognitive function tests was significantly higher post-harvest than pre-harvest. Because the cane harvest extends over a considerable time period, the authors were able to control for calendar effects; the difference was similar early or later in the 5 month period of the harvest. The authors conclude that poverty has about the same cognitive consequences as the loss of a night's sleep.
To be sure, other variables might explain the authors' findings. They are careful to discuss many of these such as physical exertion, stress, nutrition, or training effects. If the authors are right, however, their findings have some impressive implications for health policy. One, which they note, is that it may just be more difficult for people who are poor to perform complex tasks needed to apply for eligibility for programs such as Medicaid (why are we surprised that so many who are eligible don't sign up?). Another is that programs designed to incentivize healthy behaviors may just not work very well if they ignore cognitive loads.
September 5, 2013 in Access, Affordable Care Act, Consumers, Health Care Costs, Health Care Reform, Health Economics, Health Reform, Medicaid, Obesity, Prevention, Public Health, Uninsured | Permalink | Comments (0) | TrackBack (0)
Saturday, August 17, 2013
This past week, the New York Times published a story about yet another delay in the implementation of the Affordable Care Act. Earlier this summer, NPR also reported the delay, which concerns total limits on out of pocket costs that consumers can be required to pay. Under ACA, beginning in 2014 consumers were supposed to have to meet only one out of pocket limit--$6,350 for an individual and $12,700 for a family--including all deductibles and co-payments. But the Times story reports that insurers have been granted a year's grace in implementing this requirement and quotes an administration official as attributing this decision to the inability of insurance plans to communicate with each other in determining out of pocket costs.
Both stories emphasize the plight of patients who are covered under separate medical and pharmacy benefit plans. Pharmacy plans in particular may have very high copayments, without annual limits. Patients with expensive drug needs for diseases such as multiple sclerosis are especially hard hit by these benefit structures.
As I ruminated on this delay, it occurred to me that the problem of the plans' inability to communicate with one another is the plan's problem, not the patient's. To say the least, it does seem rather unfair to have patients bear all of the costs of the delay.
Moreover, there is a model that could have been used to implement the single limit: submission of claims for out-of-network care. Patients do this all the time and receive reimbursement to the extent covered by their plans. The payer has a record of the claim and can credit it against the patient's deductible. Why couldn't this model have been applied to the problem of multiple plans for patients? It would be simple. These are primarily patients with employer-provided plans. All that would be needed would be to stipulate which plan is primary for the purpose of maintaining the single out of pocket total. Medical plans are used to maintaining such totals. If the medical plan were stipulated to be the primary plan, all the patient would need to do would be to submit records of out of pocket payments under their pharmacy plans. When patients meet the out of pocket total for the year, they would no longer be responsible for copays or deductibles from the primary plan. How would other plans know about this? Patients will receive records from their primary plans that they have met their deductible for the year. They would then be responsible for submitting these records to their other plans--after which the other plans would no longer be able to charge copays or deductibles.
This approach, to be sure, puts the burden on patients to solve the communication problem. But I'm surprised notbody seems to have entertained this suggestion, in a health care climate that heralds patient responsibility. Perhaps the difficulty instead is that the multiple-plan structure emerged as a way to limit health care costs for payers by shifting costs to consumers.
August 17, 2013 in Accountable Care Organizations, Affordable Care Act, Consumers, Cost, Employer-Sponsored Insurance, Health Care Costs, Insurance, Payment, Reform, Spending | Permalink | Comments (0) | TrackBack (0)