In Search of a Better Way to Measure Quality Primary Care

Tom G. Bartol, NP


April 26, 2017

A Complex Process Needs Flexible Metrics

Payers and regulators insist that primary care clinicians report numerical quality metrics, yet primary care is not by nature a linear process nor isolated to a single disease. Primary care is a complex adaptive system involving patients with individual backgrounds and needs. Although designed to promote accountable performance, quality reporting has instead contributed to burnout among healthcare clinicians. This has, in turn, undermined primary patient care.

The primary care clinician addresses not only disease, but human factors and priorities, which vary widely and may not be aligned with those of caregivers. Patient goals, attitudes, and beliefs often confound guideline writers and clinicians. Complex interactions and interdependencies emerge that cannot be understood or predicted by simply measuring individual elements of the system. Traditional quality measures represent only a small fraction of the work delivered in primary care.

It would be more appropriate to measure outcomes than processes. Young and colleagues[1] suggest implementing transforming feedback loops across systems, not merely compiling lists of single disease measurements. This would help to absorb uncertainty as well as promote learning from the data. They suggest that shared decision-making be taken into account, and that instead of absolute numbers, goals should encompass ranges. Such outcomes as clinician/patient continuity and increased time with complex patients are quality measures that improve care and reduce costs. Adaptability rather than standardization should be the cornerstone of disease management in primary care.


Regulatory policies, such as meaningful use of the electronic health record (EHR), patient-centered medical homes, and pay-for-performance, are all demanding more quality measures, and therefore require more data from clinicians. Although the intentions are good, the results of these measures are less clear. Measuring and reporting quality, as currently accomplished, has effects and consequences that negatively influence quality of care, patient outcomes, and clinician job satisfaction.[2]

In my experience, quality measures may set the agenda even before a patient arrives. Clinical staff become focused on meeting quality measures and may not ascertain the patient's concerns or goals. A patient with family issues, economic struggles, or social stresses may not care at that moment about meeting quality measures, preferring to deal with the current life crisis. The focus on quality measures may distract from developing the relationship and inquiring about the patient's goals and concerns for that visit.

Adherence to quality measures may not improve outcomes. Many traditional quality measures are process measures. Measuring urine microalbumin in a patient with diabetes gets "credit," but there is no accountability for what follows if the results are abnormal. Quality measure reporting has become an assembly line or game, producing widgets or check marks to create profits, or keeping a scorecard that has little positive impact on patient outcomes. The EHR has become a tool for gathering data and meeting reporting requirements rather than for facilitating continuity of care.

Guidelines do not easily take into account individual attitudes, beliefs, values, and comorbid conditions. A patient may choose to exclude a screening test, medication, or lab test that is recommended by quality measures. This does not indicate poor quality care if it was a shared and informed decision made by the patient. We treat patients, not diseases, and must get beyond the idea that there is only one correct treatment for a patient.

Since the advent of the EHR and quality measure reporting, there has been an increase in burnout among clinicians.[3] With the emphasis on a scorecard for physicians, excessive time is required to collect data, often distracting from the therapeutic and supportive relationship the clinician has with a patient. Traditional quality measurement is time-consuming and expensive; the average primary care clinician spends an average of 3.9 hours per week on quality measures, with an annual cost to the practice of $40,000 per clinician.[4] The United Kingdom started pay-for-performance more than 10 years ago and is reconsidering this process. They have found that pay-for-performance based on quality measures has not improved the outcomes of care in their country.[5]

Accountability for giving appropriate and quality care is necessary and important. Doing it in a useful way that leads to growth and improved care requires a different approach. Creating data as incentives or penalties will not improve care but will standardize it—something that will not help individual patients.

Data collection must take into account patient preferences and values and focus not only on numbers, but also on relationships created between clinician and patients. The goal must be quality care with data collection a result of, not the motivation for, patient care. Overburdened clinicians trying to meet quality care documentation requirements have little left to help meet the needs of the patient.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: