Care planning can go awry in a variety of ways. Elders and their families can refuse to be engaged; critical information about likely future course can be ignored or simply unavailable; the care team can be too hard to convene; the discussion can end up with no decisions; the care plan developed can become unavailable when needed; and so on. The communities implementing MediCaring Communities need to understand that developing efficient and effective methods to generate care plans at scale for frail elders will be a work-in-progress, with much to learn in the next few years.
One pervasive problem is that the understanding of the elderly person’s situation and choices could always be more detailed, no matter how much is already known; but overload of information makes the information as useless as a deficit of information. Sending a frail elderly person to the emergency room with one hundred pages of copied records will induce the emergency room staff to set the whole stack aside, given the press of time and the need to prioritize urgent matters. The “presentation layer” for care planning needs much more attention from creative minds, aiming to show the most salient things, as well as a general overview, depending upon the focus of the user. The elderly person viewing his or her own care plan generally does not care to read the history and may be overwhelmed with medical or administrative details, but he or she very well may want an explanation of the goals of medications and a characterization of the likely course. The emergency room physician usually does not need to know the details of nutrition support at home. A set of streamlined presentations would make the care plan and its background information so much more useful, though the user also needs to be able to pursue important issues more deeply when warranted.
Good care planning also needs much more useful information about prognosis, both for survival and for function. Predictions of outcome often depend mostly on the physician’s experience, and physicians rarely follow elderly patients over time, so they really don’t know what the likely course will be and what supportive services the person will need. An intriguing website at http://eprognosis.ucsf.edu gives some approximations of likely survival time. The website does not give a sense of the “shape of the curve,” just the point estimate for when half of the people “like this person” will have died, and it does not address function and therefore supportive service needs. With all the data that Medicare now has access to, and the enhanced data arriving with the IMPACT Act, Medicare could generate a set of interactive data analyses that would allow estimates of likely outcomes, given various treatment strategies. Having accurate data on the likelihood of future disabilities and death will be disheartening to some, but it would be very helpful to all involved, since honest and reliable information is essential to preventing emergencies and being prepared for the likely future course.
The measure of success in the care plan cannot be simply whether the elderly person achieved his or her priority goals. Some elderly persons and families will insist upon pursuing very unlikely goals—to live for another five years or to be able to live alone, perhaps. We need to establish a process that does not force elders and families to accept easy goals or encourage them to state outlandish ones. What may become the standard for good care is whether the care team helps them pursue the goals as promised and whether the elderly person and family were confident that what mattered most to them was the highest priority of the care team as well. Developing efficient ways to measure elder-driven dimensions of quality will be an important component of MediCaring implementation.
Finally, some assessment instruments automatically trigger care plan elements. The most well known of these is probably the MDS (for nursing home residents) for which certain responses about the elderly person’s situation trigger elements of the expected care plan. This approach may be useful for certain widely acceptable strategies, such as providing pressure-reducing mattresses for persons at high risk of pressure ulcers. However, most treatment strategies require input from the elderly person or their representative, so the automatic nature of the triggering leads to work-arounds and inaccuracies in the record. For example, documenting that the resident has a depressive mood disorder triggers a deficiency in MDS if the person is not on a medication to address depression. But many nursing home residents have tried anti-depressants and don’t want to do so again, so the nurse entering the data is induced to circumvent the appearance of a deficiency by noting that the resident is sad, but not acknowledging depression. This sort of adverse incentive reduces the value of the record and the care plan and must be avoided if possible in developing work processes and tools for care planning.
 (Lee, et al. n.d.)
 (Improving Medicare Post-Acute Transformation Act of 2014, Public Law 113-185 2014)
 (Centers for Medicare and Medicaid Services, MDS 3.0 RAI Manual 2016)