Work Flow Analysis: Applying Sutton’s Law in the Diabetes Care Workplace

All of these goals may be achieved by work flow analyses. Work flow can ultimately have an impact on the effectiveness of medical staff and perceived patient care. The typical office work flow for diabetic outpatients consists of four primary stages: 1) check-in; 2) weight, vital signs, and Standards of Care assessment; 3) physician consultation; and 4) patient billing and rescheduling. Improving work flow at any stage can have a significant effect on overall office productivity. At each stage, there are opportunities for increased efficiency. A work flow analysis can help determine the degree to which staff is being optimally used and the degree to which simultaneous processing of patients through the different stages can increase the case load that may be served.

The first stage, check-in, for example, is one area in which work flow can often be improved. Offices make many common errors here that can stymie the rest of the patient care process. New patients should be asked to be present at least 30 min before their appointment time to complete necessary paperwork. Offices that fail to request this may experience persistent delay in the entry of new patients into the flow. This delay may overburden the clinical staff with new-patient and scheduled follow-up appointments. The intake process for all patients should also allow time to confirm that managed care patients have their required authorizations for service, if needed. Later in this article, we examine the application of an information technology solution that improves office work flow.

Patient-physician consultation time and interaction quality depend on several issues. One important factor is the process of communicating blood glucose readings to the physician. Traditionally, there have been several methods for delivering daily meter readings. These include 1) patients delivering handwritten log sheets to the physician at the time of the patient-physician interaction; 2) office staff transcribing from patients’ meters to a log sheet at the time of visit; 3) office staff transcribing readings from patients via telephone before the visit; 4) patients faxing a handwritten log sheet before a visit;3 5) the physician reviewing readings on the meter itself without log sheet during the visit; and 6) office staff downloading meters to a computer.

These methods vary in accuracy and efficiency based on who transcribes, what is transcribed, and when the transcription occurs. Transcribing of readings from the meter to a log sheet not only creates an opportunity for errors but also places a burden on either the office staff or patients. In fact, it has been found that log book entries frequently differ from true meter readings.4 

Although readings can be reviewed on the glucose meter, a complete log sheet, which provides the physician with the opportunity to observe patterns available only through simultaneous inspection of long-term data, facilitates efficacy of physician consultation time. Likewise, having the log sheet before the patient-physician interaction allows the office staff opportunity to prepare charts and identify problematic patterns of blood glucose fluctuations early. In this manner, the staff can perform evaluations on the patient to help explain concerns about the blood glucose fluctuations before the physician exam. This ensures that the physician can give full attention to the patient and not be burdened with incomplete or inappropriate data presentation.

Hence, one way to improve work flow is to ask patients to submit log sheets to the physician’s office before their examination. Depending on the office work flow design, this can be accomplished either immediately upon the patient’s arrival for the visit or before arriving, e.g., via fax. Thus, technology could be used here to improve the work process in three ways by 1) reducing human transcription; 2) providing a structured long-term log sheet; and 3) presenting data before the physician-patient meeting.

The use of information technology to improve processes in health care has a long history primarily in back-office operations, such as billing, record-keeping, and insurance claims. Until recently, less attention has been given to exploring opportunities for enhancing the point-of-care experience for both patients and physicians through information technology applications. Introducing a personal computer–based system brings with it the inherent complexity found in the current state of personal computers. This element, coupled with decreasing costs for computational power, has prompted the computer industry to offer “information appliances.” These products provide information technology in a form that is tailored for a specific application.

One solution for addressing the work flow improvements suggested above while minimizing the complexity of a general-purpose system is to download blood glucose readings by modem (the information appliance). In these systems, the user connects the glucose meter to a specially developed modem, either at home or in the physician’s office, and the data are immediately transferred to a fax machine in the physician’s office. The data include a diary report listing patient blood glucose readings by time and date and a trend report showing blood glucose values by time blocks for pattern analysis. These reports can be printed on the fax machine in a number of formats to facilitate physician interpretation.

To evaluate the potential impact of an information appliance on office work flow, 40 patients, 11 physicians, and 6 nurses from 5 physicians’ offices (one each in Alabama, California, Indiana, Iowa, and Ohio) participated in a work flow evaluation study. The Acculink modem developed by Roche Diagnostics Corporation was the information appliance tested for its potential to improve effectiveness of office practices in transferring blood glucose information to the medical staff.

A time study was performed to determine the time that patients spent in the traditional four stages of the office work flow. For the timing data, a total of 43 cycles of patients using the information appliance and 107 cycles of patients using traditional methods (e.g., log book, log sheets) were collected. A technology assessment questionnaire5 was administered to patients and medical staff, with minor wording variations, to assess their satisfaction, because patient satisfaction is an important management tool in the practice.1 The study assessed six dimensions that have been identified as the key variables in user preference for a new technology. A respondent’s score for each dimension was computed as the arithmetic average of the responses on the component questionnaire items. The dimensions are detailed in Table 1.

To evaluate the information appliance, three dimensions—system performance (work flow and quality of care/productivity), medical staff satisfaction, and patient satisfaction—were assessed.

The system performance dimension assesses resource requirements and the degree to which the new technology facilitates accomplishment of the physician’s goals efficiently. Toward that end, the timing of patient flow through the office was assessed to determine whether there was a significant time difference between using the information appliance and the traditional process at each of the four stages of an office visit. Average times for the four stages in the patient flow are presented in Table 2. Given the large variances within conditions, no conclusions regarding productivity could be made from these data. However, the questionnaire data provided additional insights into this dimension.

The definition of system performance used in this study included the questionnaire dimensions of “Quality of Patient Care” and “Productivity.” As can be observed in Table 3, there was a strong, statistically significant preference for the information appliance solution by medical staff and patients for all dimensions. Looking at physician results separately, physicians tended to give greatest support to the information appliance solution on the dimensions of “Quality of Patient Care” and “Productivity,” the two dimensions designed as indices of system performance. In fact, the mean “Quality” rating was higher than “Effort Required” and “Ease of Learning” mean ratings, at the P < 0.05 level, and higher than the “Minimizes Error” dimension scores, at P < 0.07. Physicians preferred the information appliance solution with regard to work flow efficiency and quality as compared to traditional methods. Although the work flow timing data did not identify a difference in patient work flow, physicians indicated that their productivity during the consultation time had improved. It should be noted that observation of the physician consultation was not part of this study.

The dimensions of “Errors,” “Effort,” “Learning,” and “Overall Satisfaction” were used to measure medical staff preference. For all dimensions, both physicians and nurses showed strong preference for the information appliance solution as compared to traditional methods (P < 0.01). In contrast to the physicians’ focus on productivity, nurses, when analyzed separately, were particularly sensitive to the advantages in reducing errors brought about by the information appliance. Their preference for the information appliance was strongest for “Minimizing Errors” and “Quality of Patient Care.” The data indicate that nurses, who are traditionally responsible for the transcription process, identified error reduction as an outcome of the information technology appliance.

The patients also rated the information appliance as the preferred method for all dimensions at the P < 0.01 level (Table 3). The dimension of “Error Reduction” received the highest preference for information appliance. As with the nurses, elimination of the transcription process was a key attribute. Patients also focused on “Quality of Patient Care,” which received the second-highest preference for the information appliance.

Work flow analyses and targeted uses of technology to improve work flow require further attention if we are to improve the business of diabetes care. For information technology to be effective, its complexity must be kept to a minimum while meeting a specific need for work flow productivity. Information appliances allow technology to be implemented in such a way that a specific problem is solved or efficiency gained. Our business is medical care, and thus the technology must afford us the opportunity to meet the needs of our patients while maximizing our office productivity—in short, to “go where the money is.”

In this case review, productivity, medical staff preference, and patient preference were explored with the hypothesis that the information appliance approach represented by the Acculink modem would improve productivity and lead to increased medical staff and patient preference. Physicians identified the productivity and quality of care advantages brought forth through implementation of the information appliance, and physicians preferred the information appliance overall as compared to their traditional method. However, further study is needed to evaluate elements specifically affected during the physician-patient consultation. Nurses had a preference for the information appliance on all dimensions. They focused their attention on the system’s ability to reduce errors and improve quality of care. Patients overwhelmingly preferred the information appliance and tended to identify the potential for reducing errors as their main reason. They also emphasized the beneficial effects of the information appliance on patient care quality.

The information appliance reduces human transcription, provides a structured long-term log sheet, and presents data before the physician-patient meeting. As a result, patients perceive that their physician’s office, which uses the information appliance, provides higher quality patient care than those offices using traditional systems.

The authors wish to thank Roche Diagnostics Corporation for their support of this research.