Why healthcare quality metrics miss the mark for consumers

5 ways that IT can help providers measure performance capabilities that really matter to patients and their families.


Healthcare is falling victim to a classic misunderstanding about quality. When the “experts” in any industry define quality, they often use technical gee-whiz measures that can miss what the customer wants.

The title of a statement by the Agency for Healthcare Research and Quality (AHRQ) asked, “What is Health Care Quality and Who Decides?”. The statement goes on to say, “the Agency fulfills this mission.” A number of federal and state agencies have taken admirable strides toward collecting and using quality metrics, but do these truly measure healthcare quality? How will the best healthcare providers and payers build healthcare systems to deliver quality metrics during the coming five years?

AHRQ defines quality healthcare “as doing the right thing for the right patient, at the right time, in the right way to achieve the best possible results.” To put their definition in action, National Committee for Quality Assurance (NCQA) developed the widely used Health Plan Employer Data and Information Set (HEDIS). The Centers for Medicare and Medicaid Services (CMS) has another stet of metrics it uses for measuring physician quality, the Physician Quality Reporting System (PQRS).

However, many of the metrics developed so far miss a key aspect of quality—what outcomes patients desire.

For example, PQRS defined four outcome metrics for OB/GYN:
  • Controlling high blood pressure: percentage of patients 18-85 years of age who had a diagnosis of hypertension and whose blood pressure was adequately controlled (<140/90 mmHg) during the measurement period.
  • Proportion of patients sustaining a bladder injury at the time of any pelvic organ prolapse repair: percentage of patients undergoing any surgery to repair pelvic organ prolapse who sustains an injury to the bladder recognized either during or within 1 month after surgery.
  • Proportion of patients sustaining a major viscus injury at the time of any pelvic organ prolapse repair: percentage of patients undergoing surgical repair of pelvic organ prolapse that is complicated by perforation of a major viscus at the time of index surgery that is recognized intraoperative or within 1 month after surgery.
  • Proportion of patients sustaining a ureter injury at the time of any pelvic organ prolapse repair: percentage of patients undergoing a pelvic organ prolapse repair who sustain an injury to the ureter recognized either during or within 1 month after surgery.

Three of these measure the frequency with which patients experience an injury during a medical procedure. While no one wants to be injured by the doctor, a patient would hardly call that an acceptable outcome.

Patients want the healthcare industry to “tell me what’s wrong, fix it and keep me healthy.” In contrast, HEDIS and PRQS outcome measures are technical in nature, defined solely by the experts and related to a general service area—not to a specific cycle of care. Missing are measures of how effectively the physician diagnosed the patient’s health conditions, whether the physician was able to “fix it” or whether the patient left the physician’s care healthier.

Measuring quality merely from the perspective of a physician (or hospital, or government agency) is akin to Kodak engineers measuring the quality of photographic film and paper. Kodak was the unquestioned market leader in quality from the perspective of a chemist. However, as Kodak learned too late, the customer just wanted an easy way to capture and distribute images, and digital was a better solution. The telephone industry struggled with a similar issue.

For a century, AT&T invested heavily in accurate voice transmission and uninterrupted service. Remember when Sprint first introduced an all-optical transmission network that was “so quiet you could hear a pin drop”? The mobile phone changed customer expectations, as mobility replaced voice quality. The smartphone has pushed quality into many new areas other than voice accuracy: video, Internet access, music playback and games. Land lines died—only 40% of American homes still have one. I rue the day we lost excellent voice transmission, but the customer now generally measures quality in a new way.

In both those industries, the experts thought they knew the quality that customers valued, but the market went in a different direction. Today, healthcare experts define quality using a 20th century concept of hospital-centric medicine. Instead, healthcare outcome measures should reflect the fact that most care for health does not happen either in a hospital or a doctor’s office. Metrics need to measure, from the customers’ perspective whether whole system delivers effective outcomes.

Care for a medical condition involves a network of various clinician services, including obtaining and using medications, and patient adherence. These are typically out of a single doctor’s or hospital’s control. As such, it may not be possible to place specific responsibility for the quality all of OB/GYN (using the example above) healthcare outcomes on a specific doctor. Measuring quality of one doctor may be the equivalent of measuring the quality of your smartphone’s speaker—ignoring that it is but one element of a complex system delivering high quality mobile phone service.

Contrast PRQS’ OB/GYN outcome metrics to the Pregnancy and Childbirth outcomes defined by the International Consortium for Health Outcomes Measurement (ICHOM). These include 14 measures in four groups—survival, morbidity, patient-reported health and wellbeing and patient satisfaction with care. These measure the care for one cycle of care, not an entire specialty. A few examples of ICHOM Pregnancy and Childbirth metrics include: confidence with role as mother, pelvic pain and discomfort, and shared decision making and confidence with care providers. ICHOM developed these with input from patients as well as a spectrum of providers and experts. Compared with those of PRQS, these get more to the point of the “job to be done” in delivering a healthy baby.

The 2001 healthcare quality exposé, “Crossing the Quality Chasm” called for “changing systems of care” in order to “improve the health…of people”. Yet, some efforts to implement those bellwether observations appeared to have wandered off message. CMS seems to invoke quality as a tool to penalize doctors a few percent out of their Medicare reimbursements. Quality should instead be a tool by which healthcare providers differentiate their services. Healthcare providers who embrace quality will set the standard low-quality providers must, in turn, strive to meet. All this suggests that many early efforts at measuring healthcare quality are missing the real need.

IT systems to measure quality over the full patient care cycle will be vastly more challenging than those needed to report compliance metrics. Healthcare IT executives have the opportunity to build systems that measure and report how effectively and efficiently their organizations tell the patient what’s wrong, fix it and keep the patient healthy. IT systems supporting quality measurement will excel in five strategic dimensions:
  1. Broad data integration. Metrics will require data from all aspects of the patient’s care network—from many provider networks and one or more payer networks. Other parts of the patient network may be involved as well—pharmacies, nursing homes, device manufacturers and perhaps even fitness centers. Thus, a challenge for IT executives will be to develop new forms of data collaboration and partnerships.
  1. Electronic health record data. Measuring quality will require semantic data analysis of health records full of data not fielded for easy analysis. Quality reporting may also require longitudinal data analysis, not generally possible with today’s islands of information. This will require new software tools and data analysts.
  1. Data structured by patient, not event. Today, quality metrics are focused on hospital, doctor or discharge, rather than the patient. To measure quality of the job done for the patient, the focus must be the patient. Healthcare quality systems could develop as the equivalent of industry shifting from transaction processing to Customer Relationship Management (CRM) and Enterprise Resource Management (ERM). Unlike manufacturing in the 1990s, healthcare has no packaged solution to turn to, as manufacturing turned to SAP. IT executives will collaborate with other organizations and best-of-breed solutions vendors to assemble quality analysis capabilities across the full patient cycle of care
  1. Network, not individual performance. IT systems to measure quality of the full job to be done will measure how well individual providers participate as part of the patient’s overall network. Deming taught industry in the 1980s and 1990s that systems, not individuals, deliver quality. Thus, effective quality management cannot focus on individual provider’s medical performance. Instead, IT systems will collect systemic data rather than data suited to affix quality performance to the work of any one individual.
  1. Competitive advantage, not compliance. Quality for the purpose of complying with a mandated program misses the point altogether. IT systems built to generate government reports will not support real quality. As they face meeting the current need for compliance metrics, IT executives will work with clinicians to develop strategic approaches that measure how well the network delivers on the patient job to be done.

Better quality metrics are imperative—as much today as before we started “crossing the quality chasm.” However, systems measuring quality of the job done for the patient will be misguided if developed merely for the purpose of compliance with a government mandate. Instead, providers and payers have the opportunity to cherish quality metrics as a way to demonstrate their superior patient care. To IT executives building the systems that will measure and report quality standards, patients holler, “can you hear me now?”

More for you

Loading data for hdm_tax_topic #better-outcomes...