An excerpt from How Physicians Can Fix Health Care: One Innovation at a Time.

In 2011, Shreya Kangovi hired her first two community health workers, or CHWs, at the University of Pennsylvania Health System. Their role was to develop partnerships with low-income hospitalized patients who were at high risk of readmission. The CHWs ensured that the patients understood their discharge instructions, bridged communication gaps between the patients and providers, and informed providers about socioeconomic challenges that could affect care. After discharge, the CHWs assisted patients in taking care of their health until they had gained access and transitioned to primary care.

Three years later, the results had proven powerful. Readmission rates were down, patient satisfaction was up and access to primary care was up. UPHS had created the Center for Community Health Workers, under Kangovi’s direction. The center was hiring to expand its staff to 30 people and broadening the range of situations in which CHWs worked.

I have said that the idea is the easy part, but it is certainly worthy of a brief discussion. I focus on developing a kernel of an idea into a more complete concept so that you can share it with others, build momentum and move forward to execute.

You don’t need a perfect idea to get started, and you certainly do not need an original one. In fact, it appears to me that just four fundamental ideas account for nearly all of the work that needs to be done in innovation in health care delivery. They are:

Standardize and delegate. Some aspects of care are highly amenable to standardization. If you can specify a care process, step by step, and make it routine, productivity will rise and errors will fall. System costs will fall. To further reduce costs, it may be possible to shift tasks from providers with high salaries and many years of education to providers with lower salaries and less education. Standardization and delegation are generally paired, because it is easier to delegate tasks that are scripted and consistent. These steps are the basics of industrialization, practiced by Henry Ford in automobiles a more than a century ago, as well as many others before him.

Coordinate. From the patient’s perspective, fighting illness is a continuous process that stretches out over weeks, months or even years — and the patient is intimately familiar with every minute. From the provider’s perspective, health care is a collection of discrete interactions with the patient, spread out across time, space and multiple providers. In a perfect world, these interactions collectively would constitute a coherent, planned, well-coordinated and well-executed health care intervention, one that adapts as circumstances change.

Much easier said than done, as you are already too aware. Fragmented, poorly coordinated care is commonplace, and it leads to confusion, contradictory care plans, missed care steps, wasteful duplication and errors. It can frustrate patients to the point that they disengage or give up, which of course leads to deteriorating health, more care and more waste. Improve the coordination of care, and outcomes go up while costs go down.

Prevent. Is your patient population at significant risk of needing expensive medical care in the not-too-distant future? Can you intervene to prevent it? If your prediction is accurate and the cost of the intervention is low, outcomes go up and costs go down.

Improve treatment decisions. Few treatment decisions are black and white. There are typically multiple treatment options and uncertainty about which is best. An ideal decision is a deliberate one. Both patient and doctor are clear about the latest evidence regarding the benefits, risks and side effects of each treatment option. And both are clear about what matters most to the patient. Practically speaking, such idealized decisions are difficult to achieve. Is it possible, however, to at least move closer to the ideal? Outcomes improve when the outcomes that matter most to patients are identified and considered during the decision-making process. Costs fall in cases in which well-informed patients choose less care.

One of these four pathways to improved care is typically dominant for an innovation in health care delivery. Kangovi’s central thrust was preventing patients from falling ill again and potentially needing to be rehospitalized.

Her initiative, as is commonly the case, touched on other pathways as well. She and her colleagues have standardized work routines for CHWs and delegated as much work as possible to them. CHWs endeavor to coordinate hospital care and primary care. And, they help patients and physicians improve treatment decisions in the form of more realistic care plans and goals.

Your job, as a physician innovator, is rarely to come up with a fantastic new-to-the-world breakthrough concept that nobody has ever contemplated. It is instead to apply and refine one or more of these four fundamental ideas for a specific context.

A strong first step in doing so is to develop clear and crisp one-phrase or one-sentence answers to five questions. Here are the first three:

Who are the patients you will serve?

Which outcomes will you improve for these patients?

How will you do it?

These questions are straight from the standard MBA playbook. They are tried and true. In MBA jargon, the questions are: Who is your customer? What is the value proposition? How will you deliver? Starting with these questions will help ensure that your initiative is patient-centered, as of course it should be.

Every entrepreneur is trained to develop an “elevator pitch” that answers these questions quickly — in as little time as it takes to ride an elevator. A good pitch demonstrates that the entrepreneur has clarity and simplicity of purpose. It also opens the door for conversations with potential customers, partners, suppliers and funders. The feedback from these conversations helps the entrepreneur sharpen the idea, or, in some cases, abandon it before investing too much time and energy.

You’ll benefit from a similar process. By developing a quick pitch and sharing it — with patients you want to serve, other providers you may collaborate with and institutional leaders who may support you — you’ll learn a great deal, and, potentially, sidestep a great deal of future heartache.

Inevitably these conversations will turn to money, and that brings us to the final two questions:

How will you reduce system costs?

What financial impact will you have on your practice or institution?

Chris Trimble is an adjunct professor, Tuck School of Business, Dartmouth College, Hanover, NH.  He is the author of How Physicians Can Fix Health Care: One Innovation at a Time. Reprinted with permission from the American Association for Physician Leadership.