Research has shown that primary care providers do not adhere to the full range of preventive care clinical practice guidelines (CPGs). A number of reasons have been formulated by healthcare analysts to explain such nonadherence, including competing clinical demands, time constraints and provider attitudes. The VA has implemented electronic clinical reminders (CRs) in primary care to prompt providers to address CPGs during clinical encounters. A body of research indicates, however, that CRs have had only a modest impact, at best, on primary care providers' addressing the full range of preventive care CPGs. What remains poorly understood is how primary care providers prioritize preventive care tasks during clinical encounters and what impact this may have on which preventive care services are offered to patients.
We conducted a qualitative study to examine how front-line primary care providers at facilities in two VISNs prioritize preventive care tasks. As investigators in the HIV/HCV QUERI, we were also interested in exploring how providers prioritize offering patients HIV testing, given that CRs for HIV testing, triggered for at-risk patients, were implemented in both facilities. In particular, we examined whether there are differences in how providers prioritize HIV testing and other preventive care tasks. In summary, we conducted in-depth qualitative interviews with primary care providers to address the following objectives:
a.Describe how providers make decisions about prioritizing preventive care in general;
b.Examine how providers make decisions about prioritizing HIV testing in relation to the range of perceived preventive care tasks that comprise their work;
c.Describe providers' perceptions of CPGs and CRs and their perceived relationship to preventive care.
Over the course of one year, we conducted a qualitative study comprised of in-depth, semi-structured telephone interviews with primary care providers at facilities in two VISNs. We sent email invitations to 71 providers, and a total of 31 participated in interviews (44% participation rate). All interviews were conducted by the principal investigator (Dr. Solomon). Interviews ranged in length between half an hour and one hour. Interviews were audio-recorded with the consent of all participants. Audio files of interviews were then transcribed verbatim by a professional transcription company with a VA-approved contract. Transcriptions were stored on a secure VA network in a software program designed for qualitative data management (NVivo 8). Analysis-conducted by a sub-set of three study investigators with expertise in qualitative methodologies-consisted of utilizing several techniques from a grounded theory approach and drawing upon the tools of thematic analysis . Consensus was reached among the three investigators regarding themes and findings, which were then shared with the entire research team for discussion and further refinement.
The two site investigators on the study team-one on the East Coast (VISN 3), the other on the West Coast (VISN 22)-facilitated recruitment by announcing the study at a regularly-scheduled primary care team meeting. The site investigators notified providers that: 1) participation in the study was voluntary; 2) refusal to participate would not affect professional status, pay or benefits; 3) the principal investigator would be sending an email invitation to participate in an interview and 4) the site investigator would not be informed of whether or not providers chose to participate. All physicians and nurse practitioners in primary care were eligible to participate. After the principal investigator sent an initial email invitation to all primary care providers at both facilities, follow-up emails were sent every two weeks, on three occasions, to providers who did not respond. Nine providers from the East Coast site, and 22 from the West Coast site, participated in telephone interviews.
Interviews were designed to elicit the perspectives and experiences of primary care providers in relation to prioritizing preventive care tasks during clinical encounters. Questions were semi-structured and dealt with topics such as: The range of preventive care tasks faced by providers; how providers become aware of preventive care tasks; how providers prioritize among multiple preventive care tasks, including HIV testing; and, among others, the roles of CRs and CPGs in prioritizing preventive care tasks (including the HIV testing CR).
General Preventive Care
Most primary care providers said they prioritize general preventive care tasks according to their independent clinical judgment about the extent to which preventive care is aligned with patients' most relevant or pressing health needs. By "clinical judgment" we mean, "The ability to work out how general rules-scientific principles, clinical guidelines-apply to one particular patient..to determine the best action to take to cure or alleviate" illness . More specifically, clinical judgment is an iterative process that entails providers' ongoing assessment of overlapping salient contexts, including perceptions of patients' desires, preferences, concerns and broader life contexts; the purpose of patients' visits (follow-up vs. urgent); patients' health status, history, and prognosis; and patient/provider short- and longer-term goals.
Before describing the various findings pertaining to providers' uses of independent clinical judgment, it is important to take into account relevant contextual factors that influence providers' attitudes and actions. First, providers have distinct perceptions of VA expectations for how preventive care should be addressed. The combination of regularly occurring preventive care CRs, and pay-for-performance programs that reward addressing CRs, has instilled in providers the notion that the VA expects the quality of care provided to be dependent upon satisfaction of CRs. According to providers, patient care delivered in such a manner runs the risk of becoming inappropriately homogenized (by uncritically offering all patients identical preventive care) or, even worse, resulting in harmful outcomes for patients.
A second factor that influences how providers prioritize preventive care is their widely-shared view that CRs are cumbersome, needlessly complex and too time-consuming. Providers repeatedly expressed their frustration with the process of attempting to satisfy reminders, noting the loss of valuable clinic time and mounting frustration. Some providers went so far as to express concerns about having their professional autonomy undermined by CRs.
Turning specifically to findings pertaining to exercising independent clinical judgment, the following general statement is representative among providers:
"[W]e have a desire, if we're giving the best care, to individualize...what should be done, which is, I think the best way of doing this. [It] is to give thought about what's appropriate in this particular setting, the sense that the package [CRs] does not allow you in an easy fashion to do that."
This provider invokes the notion of "individualizing" care for patients, a common refrain among our interviewees. Individualized care emphasizes organizing clinical encounters based on the perceived needs of patients, rather than using CRs to structure patient care. Note how this provider, like other interviewees, contrasts the use of independent clinical judgment with his perceptions of the limitations of CRs: "the package [CRs] does not allow you in an easy fashion to do that [individualize care]."
Providers also explained that they assign low priority to certain CRs because of their perceived misalignment with the purpose of primary care. Many providers singled out seatbelt use as an example of such a CR. As a provider explained:
"some of the reminders are very silly, really not [a] place to be there in a medical chart, such as, is patient using a seat belt? But, that's a law to use a seat belt. It's not a medical issue."
For most providers, however, structuring clinical encounters according to patient needs does not exclude attention to preventive care writ large. Indeed, many providers commented that they often address the very preventive care topics that are included in CRs. However, rather using CRs to structure patient visits, these providers raise preventive care topics with patients if and when the context of the visit-in their estimation-warrants doing so. Furthermore, providers emphasized that their attention to preventive care stemmed from their years of professional training and experience, not prompts from CRs per se. As a physician noted:
"I personally include preventive health on my problem list, and it's something that I developed during residency training.as a primary care physician.. Healthcare maintenance is one of my problems and in that I document, you know, all of the things that are pertinent to that patient, so cancer screening, vaccination, lifestyle modification..[I] make a note at every visit about which items I've discussed and.make sure all of those are up to date."
This physician's training, combined with ongoing reference to, and revisions of, each patient's problem list, guides her to determine of what is "pertinent" to raise during clinical encounters. Preventive care figures prominently in this physician's clinical practice, but it is addressed in a manner she deems most congruent with the patient's visit.
Many providers noted that once they have focused on patient needs in the first phase of the clinical encounter they do turn to CRs to address preventive care topics that have not been discussed. As one provider explained:
"So to me they're [CRs] more of a backup. I know some people use them as a trigger in terms of just what needs to be done in terms of preventative health overall, but for me they're just kind of a backup."
Another provider noted:
"I basically try to do things according to what I think is the best practice for that patient, and then I use the clinical reminders as a backup."
For these providers, CRs are not to be disregarded, but they do have secondary importance within the context of patient visits. The first priority is addressing patient needs and concerns, independent of CRs.
Finally, a related finding is that in cases when patient needs are urgent, providers described taking one of the following approaches: 1) deferring addressing preventive care until a subsequent visit or 2) resolving the urgent matter first and then addressing preventive care.
We turn now to findings specific to how providers prioritize offering HIV testing to patients. First, most providers described a prioritization strategy similar to the broad strategy used for addressing preventive care writ large: using independent clinical judgment to gauge, on a patient-by-patient basis, when there is sufficient indication to raise the topic of HIV testing. In these cases providers talked about using their clinical judgment to offer HIV testing to patients they deemed to be at risk, independent of CRs for HIV testing. For example, a provider noted the following in response to a question about what prompts him to offer HIV testing to patients:
"I guess, you know it's generally patients who I think maybe have high risk practices or who I think have risk of exposure, so typically my younger patients who have multiple partners [and] my homosexual patients who do not have a specific one partner but who have multiple partners. And then even if they're generally older those are the ones that trigger my mind to ask about HIV testing. Or my drug-using patients and my homeless patients as well too. Those are the ones that I typically think of to just offer it and once it becomes a part of my thought process I usually include it in my..little section that I devote to preventive medicine..I'll just document you know when they need another one or, you know, how often we're gonna follow or if we're gonna continue to follow HIV testing or that type of thing."
Note that despite the presence of an HIV testing CR, this provider, like many others interviewed, offers HIV testing based on his own assessment of patient risk. In other words, his independent clinical judgment, based on what he knows about his patients, leads him to determine whether or not to offer HIV testing.
The second finding pertaining to prioritizing HIV testing differs from the findings discussed so far for both general preventive care and for HIV testing. Several providers noted that they offer HIV testing to patients as a direct consequence of receiving a CR to do so, rather than using clinical judgment alone to determine when testing is offered. As a physician explains regarding the CR for HIV testing:
"I think that it's absolutely appropriate and we definitely are missing many, many people with a treatable disease that has very significant impact on their prognosis that's treatable and can change their life expectancy..I personally think that that reminder is very appropriate, and I don't think that it's onerous. I think I understand that the recommendations in terms of the VA requirement for written consent are changing to kind of meet up with the CDC guidelines that no longer recommend that. So I think that will be a plus, because I think that's always, you know, one extra thing.to have the patient sign the form and make sure that they take it down to the lab with them. And I think that that's a barrier, so I think that's a good thing. But I think the reminder itself is not onerous at all and I think anything that only has to be done once for a patient is not that bad."
This provider begins by stating that being prompted to offer HIV testing is justified on an epidemiological basis. But she then shifts her emphasis to describe how, when various features of the CR are taken into account, fulfilling the CR requires minimal effort and time. This provider notes that the HIV testing CR is not "onerous", it is no longer necessary to obtain written informed consent from patients, and the CR itself appears just once for patients. This view of the HIV testing CR contrasts with providers' assessments of most other CRs, as described above (CRs are cumbersome, needlessly complex and too time-consuming).
Our findings reveal that primary care providers from two VA facilities prioritize general preventive care tasks according to their independent clinical judgment about the extent to which preventive care is aligned with patients' most relevant or pressing health needs. In other words, decisions about which preventive care tasks to address are based primarily on providers' ongoing assessments of overlapping salient contexts, including perceptions of patients' desires, preferences and concerns; the purpose of patients' visits (follow-up vs. urgent); patients' health status, history, and prognosis; and patient/provider short- and longer-term goals. As a result, clinical reminders for preventive care did not play a strong role in structuring clinical encounters. Indeed, most providers indicated that they refer to CRs only after having prioritized and addressed preventive care according to their independent clinical judgment.
In the specific case of offering patients HIV testing, many providers described choosing to do so on the basis of their independent clinical judgment, rather than in response to HIV testing CRs. Providers most commonly described informally assessing patient risks for contracting HIV and offering testing based on such assessments. Several providers did not fit this pattern, noting instead that they offer HIV testing in direct response to HIV testing CRs. This latter group of providers explained that they find the HIV testing CR to be practicable (not "onerous") because it occurs once per patient and because it can be fulfilled with minimal effort.
Our findings represent a tension that has been mounting in medicine writ large for a number of years. On the one hand, the widespread application of evidence-based practices offers the potential for a greater number of patients to receive recommended care. In the case of preventive care, research has shown that recommended services are unevenly distributed across the adult population in the U.S. Therefore, there is considerable need for the wider application of preventive care.
On the other hand, and as we have described in this report, most primary care providers perceive VA efforts to standardize preventive care through the use of CRs as an overly prescriptive approach to patient care. Is the apparent discrepancy between VA goals and provider views reconcilable? We think so.
We argue that, when viewed through a social scientific analytic lens, providers' resistance to using CRs to structure clinical encounters speaks to a shared tacit perspective regarding the encroachment of their professional autonomy. A shared tacit perspective-sometimes called a "schema"-is a shared mental representation whose purpose is "to guide the application of concepts to particular experiences" and therefore "determine the meanings we impart to ongoing experience, and give us expectations for the future" . In our view, then, most of the providers we interviewed indexed notions of-indeed, one might say expectations for-a predictable modicum of professional autonomy that has roots in bygone eras of the profession, namely the unfettered ability to exercise independent clinical judgment. The current period of the medical profession is distinguished by the growth of influential third parties that have an increasing say in how medicine is practiced. The implementation of preventive care reminders in VA primary care is just one such example.
As we have shown, however, providers' tacit perspective regarding the encroachment of their professional autonomy to use independent clinical judgment does not exclude attention to many of the preventive care guidelines encapsulated in CRs. Providers, we have shown, often place top priority on the content of CRs, but in a manner that stems from use of their independent clinical judgment, rather than the CRs per se. More specifically, many providers consulted the CRs once they finished applying their independent clinical judgment to assess patients' needs and concerns. Indeed, some providers said they found consulting the CRs in this manner to be a helpful "backup" practice to ensure no important topics were overlooked.
All this suggests the potential for establishing "common ground" between the VA's goals of applying preventive care tasks in a widespread manner and providers' concerns about maintaining a sense of professional autonomy. Based on our findings we recommend the following next steps to work toward establishing common ground:
-VA administrators' and policymakers' developing CRs and CR policies in ongoing consultation with primary care providers to help ensure CRs take into account the exigencies of clinical practice;
-VA administrators' and policymakers' characterizing CRs as a "backup" list of tasks, some of which might not be appropriate for a particular patient during a particular clinical encounter;
-VA administrators' and policymakers' developing CRs more specifically tailored to patients' health status and medical history.
The primary limitation of our study is that our findings may not be generalizable across all primary care departments in the VA. This results from the small sample size (31 providers), the number of sites in the study (2), and the geographic qualities of the sites (urban areas, one each on the East and West coast).
Furthermore, our findings are restricted to HIV testing CRs that are triggered only for at-risk patients. Therefore, we cannot assume that our findings would apply to routine CRs for HIV testing.
Our study has shown that most primary care providers at two VA sites prioritize preventive care, including offering HIV testing, by utilizing independent clinical judgment. Providers contrasted this manner of prioritizing care with their perception that the VA emphasizes using CRs to structure clinical encounters.
Several providers stated, however, that they offer HIV testing to patients in direct response to receiving a CR to do so. These providers indicated that, unlike most preventive care CRs, the HIV CR was acceptable because it is relatively easy to fulfill.
As we have discussed, there is an apparent tension between the need for implementing preventive care tasks on a more widespread basis (hence CRs), and providers' preferences for maintaining professional autonomy by drawing upon their independent clinical judgment to individualize preventive care tasks to address a given patient's needs. Achieving "common ground" between these needs and preferences might result from collaborations between administrators/policymakers regarding the development and utilization of preventive care CRs. The impact of such collaborations would be the development and use of a CR system that is more aligned with providers'-and ultimately, patients'-ongoing needs.
- Solomon JL, Gifford AL, Asch SM, Meuller N, Thomas CM, Stevens JM, Bokhour BG. How do providers prioritize prevention? A qualitative study. The American journal of managed care. 2013 Oct 1; 19(10):e342-7.
- Solomon J, Bokhour BG, Mueller N, Gifford A, Asch SM, Thomas CM, Stevens J. How primary care providers prioritize general preventive care and HIV testing. Poster session presented at: VA HSR&D / QUERI National Meeting; 2012 Jul 16; National Harbor, MD.