Everyone who works in healthcare knows stories of clients/consumers/patients who seemed to miss out on the benefit of care or who lacked access to it. Sometimes the consequences are tragic not only for the consumer but for others, as well. One of the best known cases of the latter is Andrew Goldstein, who pushed a young woman to her death in front of a subway train in New York City. He never connected sufficiently to care. No one knew how he was doing until it was too late because the healthcare system “lost” him.
To better ensure that care is connected and meaningful, it is crucial to measure the results of care and learn from the data how to improve. Of course, we all have a general sense of what we want to achieve: to help people recover their ability to function effectively and to return them to the state of health they had prior to their disease or dysfunction. Yet healthcare has been slow to measure results in ways that help us understand whether we are achieving the basic goals consumers have of treatment.
When a consumer enters a hospital, his/her immediate goal usually is to leave in better health, but that is not the hospital's only goal. The hospital measures whether the stay was profitable and how much reimbursement it received, if the stay involved nonreimbursable services, if the consumer was discharged within the number of days anticipated by payers, if the consumer left alive, if complications created demands on the hospital that were difficult to meet, if reports were required because of the consumer's stay, and so on.
In too many cases, measures of success are related to the provider's goals, not consumers'. Behavioral healthcare has been especially slow in assessing and addressing outcomes meaningful to consumers. A provider can tell you how many visits a consumer made (and whether any appointments were missed). A provider can describe the consumer's state of dysfunction at enrollment using instruments such as the PHQ-9. But in most cases a provider can't determine whether the consumer is now better able to hold a job, stay in school, be close to friends and family, or live a comfortable life.
Some providers do use functional assessments, but these usually are complicated and time-consuming, so they are not widely used. In the end, most providers don't have a real knowledge of whether consumers are functioning in their daily lives as well as they could. (There are exceptions, such as the Center for Behavioral Health in Bloomington, Indiana, which measures its results based on consumers' functional improvement.) A provider can determine whether specific symptoms have abated, but it usually doesn't know whether that made a difference in functioning or what other problems might have developed along the way, such as diabetes or heart failure, as a result of its interventions.
“But why is it our fault if the consumer has developed some other issue, such as diabetes, that detracts from his/her quality of life?” you might wonder. “Isn't treating behavioral symptoms our job?” Yes, that's part of our job, but we have to be prepared to deal with the whole person. If we don't measure the full effects of treatments, then we will continue to have outcomes that don't truly represent consumers' level of functioning. Treatment isn't truly effective if reduction of a particular behavioral problem comes with a serious side effect that creates an equal or greater disability. Thus, Descartes was wrong: There is no separation between mind and body. And as the gateway to their consumers' support and care, behavioral healthcare providers' job is to ensure that consumers can function better after services than before.
“Oh, but we already use best practices,” you might say, but best practices by themselves don't ensure any outcome. They don't tell us how well we are using them, and they don't help us learn from others how we might do them better. Best practices should be better than other practices at achieving a result, with evidence demonstrating their efficacy, but they are no guarantee of any result.
Of course, we all want to do as good a job as possible, but it isn't possible to know how well we are doing unless we know how our efforts stack up against others'. Only Plato's “philosopher king” has perfect knowledge, benevolently directing his subjects toward the best “outcomes.” The rest of us have to learn from trial and error, ours and our peers'. Yet there is essentially no universe against which a provider can compare its results with those of other similar organizations.
In an effort to begin addressing this major gap, in 2004 and 2005 the Substance Abuse and Mental Health Services Administration began to adopt the National Outcome Measures (NOMs). These include areas such as employment/education, crime, housing, and social connectedness. Unfortunately, they are only helpful as comparative benchmarks for providers within their own system. Without specific comparisons there is no way to find out where to learn how to do a better job or even whether one is doing a good job.
There also has been a tendency to suggest that data measurement and comparison are impossible, irrelevant, and unprofessional. Professional associations have promoted mostly process measures, and often propose only general directions to members regarding measuring results meaningful to consumers or useful in comparisons between providers.