Patient satisfaction (PS) has long been the semi-ignored stepchild of outcomes management. This may be the result of its awkward fit within outcomes since PS doesn't have a proverbial "pre-treatment" baseline like outcomes do, as it would make no sense to query someone's pre-treatment level of satisfaction with the care he has yet to receive. Or it could be due to the fact that the majority of responses to such surveys are generally quite positive (85+% endorsement rates of being satisfied) and thus, any discriminative value seems weak at best.1 Or some clinicians may find that the systems available to purchase off-the-shelf, rather than internally developed, don't provide all the information desired (also known as the "something is better than nothing syndrome").
This said, not looking at PS is a mistake. From a quality-improvement and management perspective, such PS tools, in spite of their weaknesses, are often considered essential to better understanding patients' experience of care. Thus, the challenge is to efficiently collect as much data as possible for the most relevant concerns, process it in a timely way, and make it actionable for employee and management development.
The outpatient physical therapy practice that the authors work for is one of the larger organizations nationally and there is a shared belief that a great deal of the company's ability to grow is due to the provision of quality care with a marked focus on PS.
Almost a half-century ago, Donabedian argued that "quality of care is determined by... its effectiveness in achieving or producing health and satisfaction."2 While many healthcare institutions might be tempted to ignore PS due to its inevitable subjectivity, Donabedian believed that caregivers "must view [it] as a reality." Therefore, in order to provide the highest quality of care to patients, practitioners should solicit patient feedback via a defined and consistent survey process.
As the practice has grown, its surveying methodology has evolved. When the very first PS letters were sent out in 2001, each clinic contacted its former patients. The front office staff sent returned surveys to the corporate office, where the data was entered manually, one-by-one into a database.
While this method was impressive for the organization's size at the time, it had inevitable validity, efficiency, accountability and consistency issues. With company growth, this old surveying process also became unsustainable. In order to stay competitive, the procedural standards were raised by adopting an improved, in-house, centralized surveying method.
The most significant change was a shift of the workload from the clinics to the corporate office. Instead of clinics sending and receiving the surveys, a centralized survey production and data-collection process was implemented. In addition, automated data entry replaced the manual process, as the surveys are now bar-coded and scanned into an electronic data-viewing application. This accelerated process also mitigated the data-entry errors associated with manual input and the data-viewing application now allows clinic directors and other clinical operations leadership near-instantaneous access to survey results and information used to manage their staff.
The in-house nature of the method allows for routine survey content management, ensuring the relevance and "freshness" of the ensuing findings from patients. The centralization of surveys combats the aforementioned problems associated with the previous surveying methodology. Now that surveys are sent from and returned to the corporate office instead of to and from each clinic, the temptation for selection bias is eliminated, increasing the validity of the findings. Since the process is now automated, it's much more consistent. The survey viewer application yields an increase in visibility and accountability that clinic directors use as a management tool to improve delivery of care.
Strengths of the Model
This centralized, in-house survey method works particularly well for at least two reasons: the company's culture and the operations infrastructure. In terms of company culture, there's a strong commitment to "exceed customer expectations." Consequently, the clinician/staff dedication to this ethos enables such an exhaustive PS survey method to succeed.
The operations infrastructure also complements this surveying method. There are scalable layers of operations leaders - all of whom are also clinicians. This unique aspect of the organization is helpful in determining survey content management, driving employee development based on specific patient feedback and implementing actionable improvements based on returned surveys.
Build Versus Buy
Rather than internally develop their own PS methodology, many rehabilitation facilities choose to contract with vendors. While such companies have some impressive features, it's the bias of these authors that a centralized, in-house approach has more benefits, especially in regard to obtaining true data and useful information.
The protocol of most vendor-based solutions requires that someone at the clinic administers the survey to the patient. This presents a fundamental bias problem as the very person being "graded," so to speak, is the one deciding whether or not to survey a patient and is also the one entering potentially unfavorable data into the system. Consequently, if a centralized, non-clinical location sends and receives the surveys, clinicians and staff do not come into contact with surveys or survey data before the information is entered, which presumably results in more reliable data.
Most vendor-based systems require the clinic to approach patients upon discharge. However, some of the most unsatisfied patients simply stop attending their appointments without being formally discharged. In such instances, it's impossible for the clinic to survey them and thus they are never accounted for in the data. The clinic then is never able to benefit from the lost knowledge about what resulted in the premature self-termination from care. The approach used by the authors' organization automatically sends every patient with two or more visits a survey, ensuring that all patients have the chance to participate.
When patients are required to complete a survey within the clinic, they may feel pressured to answer the survey questions positively. A centralized system avoids this bias altogether because patients receive the survey from an address other than their clinic's address, fill out the survey in their homes, and send the survey to the corporate office, thus sidestepping any anxiety that could taint the veracity of patient input.
The in-house nature of a patient satisfaction methodology is beneficial because, unlike many one-size-fits-all surveying services, the survey developed herein provides qualitative data as well as quantitative. The survey questions that off-the-shelf services provide are necessarily somewhat vague, with Likert responding options since they have to be as general as possible in order to try to "fit" the highest number of customer sites. This doesn't allow for actionable intelligence about why a score is high or low.
For example, if a clinic suddenly started to score poorly on quantitative items related to the quality of the facility, it's easy to find the culprit with qualitative questioning to learn that the handicapped door opener is on the fritz and then fix it.
One can never know from a purely quantitative checklist survey (as employed universally by vendor companies) what a patient suggests for improvement.
Furthermore, it has been the experience of these authors that many patients additionally write or type accompanying letters with their surveys, which is a valuable source of feedback. Authoring such thoughtful narrative while in a clinic or on a pad or table device is impossible for a patient.
Finally, the in-house nature of a quality-improvement and PS surveying methodology allows for content management. This means that we change the wording of questions to be clearer, add new questions to gather information on new programs or practices, and remove obsolete or irrelevant questions.
What the Future Holds
The future looks bright for innovative quality-management perspectives powered by approaches that mitigate bias and optimize sampling in improving the patient experience.
Such information has always been important, but in these turbulent times of healthcare reform, increasing penetration of accountable care organizations (ACOs) on the horizon and ever-shrinking healthcare spending, it's incumbent upon today's creative clinical leaders to develop and share new concepts in ensuring quality patient experience in addition to optimal patient outcomes.
1. Timothy, L., Keiningham, B., et al. (2007). A longitudinal examination of net promoter and firm revenue growth. Journal of Marketing, 71(3), 39-51.
2. Hinton, J., & Stout, C. (1993). Patient satisfaction methods. In M.B. Squire, C.E. Stout, and D.H. Ruben (eds.). Current advances in inpatient psychiatric services. Westport, CT: Greenwood Publishing Group Inc.
Chris Stout is director of the department of research at ATI Physical Therapy and on the faculty of Northwestern University's Feinberg School of Medicine. Grace Wang is CPM/caseloads administrator at ATI Physical Therapy. Julie Roper is director of clinical performance management at ATI Physical Therapy.