The study identified two main uses of electronic feedback: i) Administration of a regular disease control schedule for patients with chronic disease and ii) Monitoring of patients for purposes of resource prioritisation and medication management. Those two uses of feedback were described as routinely and as primarily based on the feedback’s standard-indications. Some GPs also used the feedback as a point of departure for broader considerations about individual patient needs and specific treatment approaches and linked interpretations of feedback data to other sources of knowledge. In the following, we first elaborate on the different types of use and then on the variations in the perceptions of feedback related to these differences in use.
Administration of a regular disease control schedule for patients with chronic disease
Most informants described how they used the feedback to support the administration of a regular disease control schedule for patients with chronic diseases, as recommended by the Danish College of General Practitioners (DSAM) since 1991 . For type 2 diabetes, this control schedule typically involves three annual check-ups with a nurse and an annual check-up with the GP. When the feedback was described by the informants in connection with such purposes of administration, it was commonly referred to as a checklist or as a quality assurance tool.
Some informants had already implemented this control schedule for their diabetes and (to a lesser extent) COPD patients prior to signing up for data capture. These GPs regarded the feedback’s overview of birth dates and latest check-up visits as useful for identifying patients who had not been scheduled to a fixed control or who had not showed up for appointments. Apart from this, these GPs did not experience much change, but often described the control of their chronic patients as more regular or more systematic:
We haven’t changed a bit! We have just become more systematic. We now have the kind of overview that facilitates that we check whether somebody has let us down and not shown up for controls. I think it is a brilliant tool for that…. I just print this and then the secretary makes sure the patient receives an appointment. [GP # 2]
In practices that had not already incorporated a regular control schedule before the implementation of data capture, the feedback had also worked as a tool for introducing the recommended schedule of controls. In these practices, the informants described how the access to the feedbacks’ population overviews and the benchmark data on average prevalence of diagnosis had made them more aware of diagnostic coding and had led them to increase diagnostic coding of patients with type 2 diabetes. These GPs experienced that they now managed to get more patients through the system. Some of them had adopted the technology specifically because of a wish to implement a framework for regular controls of chronic patients:
This was what we had been missing…. We really had hesitated a bit and been in doubt as to how we should do things. Suddenly there was a box, which could be implemented directly in practice. By Jove, this was clever! [GP #1]
It was a general assessment among the informants that the access to feedback had provided them with a better overview of their chronic patient populations. Depending on practice organisation, routines and individual preferences, the use of the feedback as administrative information for managing controls of chronic patients was either taken care of by the GPs themselves or partly delegated to practice staff.
Monitoring patients with chronic disease
For most informants, the use of feedback to administrate a systematic control schedule was combined with more or less routine monitoring of treatment conditions for patients with chronic diseases based on the feedback’s overview of patient outcomes. When the feedback was described in relation to this monitoring, it was often referred to as a “checklist” or a “script”.
The main purpose of outcome monitoring was to distinguish between medically well-regulated and less well-regulated patients. That differentiation was associated with objectives of resource prioritisations and of medication management:
Monitoring for resource prioritisations
The GPs in the study received a fixed annual fee – a so-called episode-of-care payment or bundled payment – for each patient with type 2 diabetes. This fee differs from the normal remuneration for individual contacts and is supposed to cover the various amounts of work in relation to each patient. Several informants explained that they could gain financially by using the feedback to distinguish between medically well-regulated and less well-regulated diabetes patients and by using this information to adjust the number of additional check-ups for each patient:
Probably, we have become more systematic in the sense that we now take more into regard, at the annual check-up, whether the fixed number of check-ups is actually a reasonable number […]. It makes some sense financially, when you receive a fee for the whole year and you may only need to see them once or twice a year – rather than being forced to see them four times [GP #3]
Whilst the administrative use of feedback described in the first section supported the adaption to a fixed control schedule, the use of feedback to monitor outcomes could often support resource related decisions to deviate from the standard schedule. This approach to prioritisation – need-stratification based on outcomes – was described by the informants as being novel to them.
Monitoring for medication management
Most GPs found that their use of feedback for monitoring outcomes helped to make them aware of patients who were poorly medically regulated. This could result in additional check-ups or in decisions to change or increase the medication for these patients. According to one informant, the use of feedback had resulted in an overall enhanced structure of the medical treatment, which he described as separate from deliberations on treatment strategies:
Having access to the actual data has convinced us that some of our patients are too poorly regulated: now they are in for maximum medical treatment. It’s not that I have become better at choosing other drugs or treatment strategies or the like, but the system has a script-like quality, which is really okay. [GP # 1]
It was a shared feature of the above described forms of monitoring that they were predominantly based on the feedback’s indications of satisfactory or unsatisfactory outcomes of treatment. A few GPs related that they – for reasons of mere curiosity – sometimes visited the feedback’s benchmark section to compare results of treatment with the averages of colleagues. One GP related that looking up the benchmarked data aroused his competitive inclinations, but that it did not affect his clinical priorities since he was always “in front”.
Broader deliberations about individual patient needs and treatment approaches
Some of the GPs described how looking up feedback could lead to broader and more thorough deliberations about the management of individual patient cases. In that context, the feedback was not described as a checklist or a script, but for instance as a “dialogue partner” or as an occasion for “professional reflection”.As an example, such deliberation might concern the appropriateness of medical regulation as a treatment strategy in specific patient cases:
You hold your horses, you know, and take a look at the patient. What is appropriate for you etc.… I am simply not willing to make the attempt to beat some patient’s HbA1c down to a very low level if I’m convinced it is not the proper approach in that specific case. [GP # 4]
I might easily just print this chart [feedback] and then take a look at the patients’ data to check why a patient is not getting cholesterol-lowering medication. Is it because we haven’t been thinking straight or is there a good reason for it? [GP # 2]
According to the informants, such reflections always included more knowledge about the patients than was provided by the feedback. This could be knowledge about a patient’s (current) life situation, his or her individual preferences and concerns, or the presence of co-morbidity. Such assessment processes were not described as different in content from those, which the same GPs would have carried out without access to feedback. However, several informants reported that the feedback’s overview of chronic patient populations had become a new occasion for this kind of reflections. Several GPs told that the feedback had helped to make them more aware of treatment-related problems in patients who had special needs and might not have visited their doctor on their own account.
Variations in the use and perceptions of feedback
The occasional use of feedback as a basis for broader reflections on individual patient cases was particularly described by GPs who took a special interest in a specific chronic disease area (or in chronic disease management in general) and/or by GPs from practices, where a mutual decision had been made to give priority to that kind of use. In one partnership practice, the decision to implement the DCP included a decision to give priority to “professional reflection” by allocating a monthly number of hours to each doctor. Another partnership practice had scheduled a couple of annual meetings dedicated to discussions of clinical issues on the basis of the feedback.
Approximately half of the GPs reported that their use of feedback did not extend beyond administration and routinely monitoring of outcomes. Those doctors often described their use of feedback as being separate from “strictly professional” purposes:
Well, I never use it for anything professionally. We use it for some basic, common things. Such as when they [the patients] need to come, and how it’s all going, and who takes care of what […] It’s great to have a structure to it, and it does make some financial sense, but we haven’t really used the clinical information. [GP # 5]
It’s difficult for us, the thing about using it for strictly professional purposes. When you ask others how much they use it, people tend to look the other way. You hardly ever get round to it… [GP # 6]
What was commonly referred to as a “strictly professional” use of the feedback turned out to cover overall clinical deliberations on prioritisations and specific treatment approaches as well as assessments of individual patients’ needs. The reported reasons for not giving priority to such uses included lack of time and cooperation problems in the clinics. However, the reasons also included a shared perception of the feedback as insufficient for assessing individual cases and of a disparity between what appeared in the feedback as satisfactory treatment and what – based on the GPs own professional assessment – could be considered as satisfactory treatment. Such disparities were always explained with reference to the GPs’ contextual knowledge about patients:
Here [in the feedback] you can only see that her diabetes is well-treated, but she is not well treated for her lifestyle or her competing factors or anything else […] She just hasn’t got a good life, but you simply can’t see that here! [GP # 5]
Independently of how they used the feedback, there was a shared perception among the GPs that the information presented in the feedback did not concern the most important and challenging aspects of the clinical work understood as the care of and personal contact with individual patients:
There are so many other important aspects of my work, which cannot be measured in the same way as long-term blood glucose levels. The other sort of quality grows out of the ordinary and demanding daily work. The really, really difficult bit is the lifestyle thing, but that is not really evident from this [the feedback]. That is the challenging part… [GP # 2]
The informants displayed different reactions to the perceived limited relevance and incomplete presentation of patients’ health condition. Some GPs simply operated with a dual definition of quality: “Quality as defined by the feedback data” on the one hand and “the reality with the patients” on the other. Those GPs described their efforts to give priority to both kinds of quality and to maintain a balance:
We have experienced becoming prone to focus too strongly on data. That way it easily becomes an impediment […]. We have to remember the other things…. Quality is just as much about talking properly with the patients. Data capture comes with the risk that it may make you concentrate on measurable aspects and forget about everything else. [GP # 7]
For others, the experience of a disparity between good clinical quality according to the feedback and good clinical quality according to their own professional judgment had led to disappointment and frustration with the technology, and had decreased their overall motivation to look at feedback. As a consequence, one practice had abandoned the use of feedback for anything else than administration delegated to practice staff.