Abstract
Purpose.:
To understand the barriers facing primary care providers (PCPs), including nurse practitioners (NPs) and physician assistants (PAs), in the current referral-to-eye-care process and to solicit suggestions from PCPs on how to improve the current referral system.
Methods.:
Four focus groups were conducted with a total of 17 PCPs: two groups with physicians (MDs): one in a rural setting and one in an academic medical center setting and one group of NPs and one of PAs, both in an academic setting. All discussions were audiotaped and transcribed, and both authors performed content analysis of the transcripts with the assistance of qualitative software.
Results.:
The most frequently cited referral barriers included: (1) poor communication from eye care providers (ECPs), (2) patients' lack of finances/insurance coverage, and (3) difficulty in scheduling an eye care appointment. Suggestions made in all groups on ways to improve the current referral system included (1) implementing electronic medical records (EMRs), (2) receiving better communication/feedback from ECPs, (3) having ophthalmologists hold clinic days in primary care facilities, and (4) performing retinal scans in primary care clinics. We found few differences between the opinions of MDs and those of NPs and PAs.
Conclusions.:
PCPs desire change(s) in the current referral-to-eye-care system. Better communication between PCPs and ECPs, further implementation of EMRs, and increasing eye screening in primary care clinics were common themes. Implementing specific suggestions, such as modernizing medical record systems, may help to increase eye care utilization among patients at high risk for advancing eye disease and vision loss.
The population of persons 65 years of age and older (≥65) in the United States is projected to increase from 40 million to 71 million in the next 20 years.
1 Approximately half of those who survive at least 7 years will develop glaucoma, age-related macular degeneration, and/or diabetic retinopathy as they age, increasing the need for eye care services.
2 Regular eye care is associated with better disease outcomes and quality of life for those ≥65,
3 but almost half of patients with one or more diagnosed eye diseases do not receive eye care according to recommended guidelines.
4,5 In many patients, particularly those with diabetes, the disease is diagnosed and followed up regularly by primary care providers (PCPs), but often the patients do not receive regular eye care.
4 In the context of this article, we define a PCP as a physician (MD, i.e., a general internist, family practitioner, general practitioner, or pediatrician), nurse practitioner (NP), or physician assistant (PA), providers who serve as a primary contact point between the patient and the health care system, and an eye care provider (ECP) as an ophthalmologist or optometrist.
Based on our review of the literature, the overall referral-to-eye-care practices of PCPs has seldom been examined, and the practices of NPs and PAs has not been examined at all. The work that has been done has focused mainly on referral of diabetic patients for eye examinations,
6–12 which is an important but only a single aspect of referral to eye care. One of these studies noted that a breakdown in the PCP-to-ECP referral chain may explain the discrepancy between PCP reported referral-to-eye-care rates and actual rates of ophthalmic examination by an ECP.
7 A study in South Africa also cited referral system deficiencies as the probable cause that many patients do not receive eye care in that country.
13 Furthermore, referral to eye care is a key aspect of health care hand-offs that are a potential quality-of-care concern identified by the Institute of Medicine.
14 As a result, we targeted this PCP-to-ECP referral step as a point of interest and sought to obtain initial qualitative data directly from PCPs regarding their referral-to-eye-care patterns, the barriers they encounter in the process, and what can be done to improve the referral system.
Before it began, the study was granted approval by the institutional review boards of both Duke University Health System and the University of North Carolina at Chapel Hill (UNC). We first conducted a MEDLINE search surveying literature published in English from 1966 to February 2009, using combinations of key words relevant to PCP referral-to-eye-care patterns and barriers.
After reviewing relevant articles, we realized that very few publications discuss these issues, and most of the ones that do are centered on referral of a patient by an MD (not an NP or a PA) for diabetic eye examinations. We used the information gathered from the literature search to develop a semistructured script for the individual interviews and focus groups. To ensure that the posed questions were clear and relevant, we pilot tested the script on three individuals who did not participate in the focus groups: one MD, one NP, and one PA. We used the revised script to facilitate each focus group discussion, to ensure comparability between the results for each group.
We recruited study participants for the academic-setting MD, NP, and PA groups from Duke Primary Care (DPC) and UNC Health Care in Durham and Chapel Hill, North Carolina, respectively. The rural MD group was recruited through a personal PCP contact of one of the authors (CDH) in rural southwest Arkansas. We were unable to recruit a sufficient number of NPs and PAs in rural areas of North Carolina and Arkansas to conduct focus groups with those provider types in the rural setting.
When a PCP expressed interest in joining the study, we confirmed that the PCP had primary care experience and practiced primary care either within the academic setting (for the academic groups) or the rural setting (for the rural MD group). We conducted four focus groups: an academic-setting MD group, a rural-setting MD group, an academic-setting NP group, and an academic-setting PA group. We conducted all the focus groups over dinner in an area that was easily accessible and familiar to all participants. All participants received a free multicourse dinner (approximate value, $75) during the discussion as compensation for their time. Before the start of each group, we obtained informed consent from each participant, in accordance with the Declaration of Helsinki. All the focus groups were moderated by one of the authors (CDH).
We began each group by expressing appreciation for the individuals' participation, stating the objectives and ground rules for the focus group, and asking for introductions by the participants. Throughout the discussion, the participants were able to speak freely about their perceptions of barriers in the referral-to-eye-care system and ideas regarding the changes needed in the current referral system. We recorded all the focus groups with two microcassette tape recorders for accuracy and clarity and later transcribed each recording with word-processing software. We used qualitative software (NUD*IST Vivo; QSR International, Cambridge, MA) for content analysis of the transcripts. In addition, we both reviewed and analyzed each of the scripts from individual interviews and focus groups for content and key concepts. The purpose of the present study was to identify a range of issues regarding eye care and ideas on how to improve care delivery that can later be explored quantitatively. Therefore, the results of the focus groups represent all the information obtained from the discussions, whether it was supported by the majority of the participants or by only one.
There have been few studies assessing the referral-to-eye-care practices of PCPs and what barriers PCPs face in the referral process for eye care in general (i.e., not just diabetes eye care). Therefore, through this pilot study, we sought to identify general referral-to-eye-care barriers from the perspective of PCPs (including NPs and PAs) and to solicit suggestions from PCPs on how problems in the current referral system can be remedied. We explicitly sought to include NPs and PAs, given their growing roles in the health care delivery system, to assess whether they had different or unique concerns.
We used the focus group method for several reasons. Focus groups facilitate an atmosphere of self-disclosure in which participants can freely express ideas, opinions, and feelings about a particular topic among a group of people who share similar characteristics.
15 In our study, all participants shared the experience of providing primary care and navigating the referral system to get their patients seen by ECPs. One of the reasons we separated our focus groups into different provider types (MD, PA, and NP) was to maintain a comfortable, permissive environment for all participants (e.g., PAs might be more likely to express themselves freely if they are discussing with other PAs and not with MDs who may be their supervisors). Another reason for this separation was to reveal differences of opinion between provider types. Focus groups allow a collection of data that present a range of opinions about a particular topic in a rapid and in-depth fashion.
16 Analysis of the data from three or more focus groups commonly helps identify trends and patterns regarding the issue of interest.
15
Patients access eye care through various potential referral pathways, including self-referral (i.e., a patient contacting an ECP directly), referral from another ECP, referral from another specialist (e.g., an endocrinologist referring a patient for a diabetic eye examination), referral from an emergency or urgent care provider, and referral from a PCP. Few of these pathways have been studied in detail, especially when considering referrals within the health care system of the United States. Regarding self-referral, one study of 214 patients in a California tertiary care eye center found that almost 30% of patients received their primary contact with the health care system through their ophthalmologist,
17 suggesting that self-referral plays a large role in that patient population. Otherwise, data from the U.S. health care system on self-referrals and referrals from ECPs or other specialists are limited. We also could not find specific data on the percentage of eye care patients in the United States who are referred by their PCPs, but we noted that in the United States in 2004, more than 40% of the 1.12 billion ambulatory patient visits were made to PCPs, including NPs and PAs.
18 Because PCPs play such a large and important role in the U.S. health care system, referral to eye care by them is an important pathway to understand.
Previous work has focused mainly on MD (not NP or PA) referral-to-eye-care patterns of patients with diabetes,
6–12 and our results extend these prior findings in several ways. As found in other studies,
7,9,11 our results show that most PCPs say they refer patients for annual diabetic eye examinations but also acknowledge that many of their patients (30%–50%, according to the PCPs in this study) do not actually receive an annual examination. In addition, we discovered potential causes of the discrepancy between stated rates of referral and actual eye examination visit rates (e.g., reliance on patients to self-schedule eye care appointments). Our findings regarding (1) the need for better communication between PCPs and ECPs and (2) further implementation of EMRs to facilitate easier appointment scheduling and communication agree with results from two other studies of diabetic eye referral.
10,12 The PCPs in our study gave some clear examples of how ECPs and PCPs can improve communication. For example, the PCPs wanted more professional interaction with ECPs in the form of periodic teaching or lecture sessions. The PCPs stated that such sessions (e.g., during a noon conference at the primary care facility) would help them stay current on advances in eye care and allow them to get to know the ECPs in their area, which could result in better referral practices and communication between the two provider types.
Communication limitations underlie many PCP concerns about the current referral-to-eye-care system. In all four focus groups, multiple PCPs cited lack of feedback from ECPs as a barrier in the referral process. Most PCPs said they had trouble getting any kind of feedback about their patients from ECPs (i.e., no phone call, fax, e-mail, note in the EMR, or hard copy via mail). Two other PCPs said they receive feedback from ECPs, but the feedback is often difficult to understand because of the ECPs use of “ophthalmology abbreviations and jargon” (Rural MD4) or “ophthalmology mumbo jumbo” (NP1). Our findings are not the first to note poor communication on the part of ECPs
13,19 and specialists in general.
20
The PCPs in our study suggested that feedback from ECPs be concise and understandable. A suggestion from one PCP was to receive a referral follow-up note from the ECP saying, “I saw your patient, diagnosed him/her with X, and Y is what we're doing [for treatment and follow-up]” (Rural MD4). This suggestion closely parallels both our own and published
21 recommendations to ECPs of (1) returning a report to the PCP in a format understandable to non-ECPs, (2) including in the report results of the eye examination and treatment plan, and (3) adding any recommendations for follow-up. A few PCPs in our study also mentioned that they would like to receive a no-show notice from ECPs when their patients do not show up for eye care appointments, which is another published recommendation.
21
Review articles and studies
12,19,21,22 identify effective collaboration and communication between PCPs and ECPs as essential for improving delivery of eye care, enhancing patients' vision and quality of life, and providing continuity of care. Persistent failure of ECPs to give useful feedback to PCPs could reduce PCP referral-to-eye-care rates or at least alter referral patterns, as one PCP suggested: “Coming here [to an academic medical setting], I was shocked at the lack of communication from the academic ophthalmologists. I've been here for a little over a year, and I have yet to receive any communication…. And so, I'm constantly a little tainted. It's not that I tell patients not to go to the academic eye center, but I certainly do not look down on referring elsewhere because I have a higher chance of actually getting communication from an outside provider than one at the academic eye center” (Academic MD2). Patients may be less likely to attend their eye care appointments if their PCPs are forced to refer them to eye care clinics that are farther away from their places of residence and/or employment.
The implementation and/or better utilization of EMRs was the most frequently cited suggestion for referral system improvement among the PCPs in our study. Academic PCPs in our study noted that the system in which they work has EMRs, but the ECPs do not use it properly (e.g., academic ECPs do not place their progress notes in the EMRs for all other providers to access). The use of EMRs could facilitate easier and more consistent communication between PCPs and ECPs and also reduce the difficulty of scheduling eye care appointments, the reliance on patients to schedule their own eye care appointments, and the overwhelming number of referrals to process in a paper-based system—respectively, the third, fourth, and fifth most frequently cited barriers in this study. Academic PCPs suggested that if their EMR system had a function that allowed primary care clinics to view the academic eye center's appointment schedule online, scheduling could be completed by the clinic's office staff before the patients leave the primary care clinic. This type of scheduling could reduce both paperwork by primary care clinics and reliance on patients to self-schedule appointments after they leave the primary care appointment. In addition, properly implemented EMRs can provide automated referral reminders to PCPs, which can be important considering the time constraints and the complex patient medical conditions that PCPs must manage.
10,11,23
EMR systems can be expensive and time-consuming to implement and maintain, but almost all PCPs in our study expressed strong support for moving the health care system toward the implementation of EMRs. Even PCPs in the rural area of our study, where practices are not supported by an academic medical system and thus there is no outside financial and technical support to pay for installing an EMR system, were supportive of EMRs. In fact, at least two of the four PCPs in the rural MD group had already purchased and implemented EMRs in their own practices, despite the cost and time required to do so. With the recent $19 billion allocation to the health information technology program of the American Recovery and Reinvestment Act of 2009, EMR implementation could become a more realistic goal in the near future, even for those PCPs practicing in rural and underserved areas.
24
Patients' lack of finances and/or insurance coverage was the second most frequently identified barrier to referral. This finding is in line with those in other qualitative studies about overall barriers to eye care (i.e., not solely barriers to referral) for PCPs,
12 ECPs,
12,25 and patients.
12,25 Although PCPs from all focus groups noted financial concerns as a barrier to referral, few offered suggestions on how to remedy the situation. One PCP, however, suggested the following: “Given that a lot of my patients cancel their specialty appointments because they can't afford the $35 co-pay, I would say: Bring the patients in once a month and we have eye clinic at the primary care facility. You, the PCP, do your diabetic visit at the same time; it's a $10 co-pay or whatever lower co-pay. Bring eye care to the patients. In family practice, I've always viewed us a more of a ‘medical home’…. So, in my mind, [specialty services] are something that we should provide, and I've worked in practices that have done that. We've brought in specialists to our facility to provide better services for our patients, and it works really well. Our patients are more familiar with us, and so it's much more likely that they're going to come and get their care there [at the primary care clinic]” (Academic MD2). Other PCPs in that focus group thought that this was a good idea, although some wondered whether their clinic space was sufficient to support an ECP practicing even part time in the same facility.
Having ECPs hold clinic hours in the same facility as the PCPs was suggested by at least one provider in all four focus groups. The PCPs said that it would be much easier to refer to someone who could see patients the same day in the same facility, as was suggested by Academic MD2. Alternatively, many but not all PCPs were open to the idea of performing more eye care in their own clinics, with a few stipulations. Most PCPs stated that proper financial and/or temporal reimbursement would have to be in place before they would consider taking on more eye care duties. Academic MDs tended to want technicians or other clinic staff to perform the extra eye care duties. Rural MDs were more willing to take on the duties themselves, whereas NPs and PAs tended to want more training in eye care before performing extra eye care duties. In fact, to sharpen their eye care skills, NPs, PAs, and some MDs were interested in attending continuing medical education (CME) conferences taught by ophthalmologists. For example, PA1 said, “If there would be a hands-on CME [with an ophthalmologist], I would absolutely attend—like a half a day Saturday kind of thing,” to which PA4 added, “I think a lot of midlevels would attend that.”
The NPs and PAs cited most of the same eye referral barriers and recommendations as did MDs. One notable exception was in the amount of additional eye care and/or screening they would be willing to perform. Most of the NPs and PAs tended to favor receiving additional eye care training and taking on more eye screening and care in their clinics than did the MDs. Most of the MDs did not want to take on additional eye care duties, even if they received additional training and appropriate reimbursement, unless they were somehow given more time in the clinic day to take on such duties. Aside from this issue, we did not note significant differences in the comments and suggestions of the NPs and PAs compared with those of the MDs. We note, however, that we were only able to analyze data from one focus group of NPs and one focus group of PAs because these provider types were difficult to recruit in sufficient numbers in rural areas. We could discover unique issues and opinions among NPs and PAs if more of them were surveyed, and we will attempt to further assess potential differences between provider types in the ongoing quantitative portion of this study.
Participants in the rural MD focus group cited two barriers that were unique to them: limited access to ECPs and patients' lack of access to transportation. They noted that the “local” ophthalmologist was only in town 1 to 2 days per week, which caused them to rely more on referral to a nearby optometrist. That rural PCPs rely more on optometrists than do their nonrural counterparts is a finding that is not unique to our study.
8 The rural MDs noted that on referrals requiring the immediate attention of an ophthalmologist (e.g., a patient with glaucoma risk factors who presents with a painful, red eye), they must call to several clinics within a 50-mile radius to determine where the local ophthalmologist is practicing that day. They also mentioned that several of their patients, particularly elderly patients, tell them not to refer them to ECPs because “I don't have a ride to get there,” a concern that was found in another study involving focus groups of patients and ECPs.
25 Patients often must have family members drive them to medical appointments, and family members do not have the time to take them to multiple appointments on different days. Thus, the PCP appointment usually takes priority and specialty appointments are not attended (Rural MD1).
We note several limitations in our study. Focus group analysis carries with it the risk of investigator subjectivity during data analysis.
15 To avoid this bias, we used systematic and reproducible methods of coding and categorizing with the use of qualitative software and two-investigator content analysis. Validity and generalizability also are common concerns in focus group studies.
15 To ensure validity, we pilot tested questions to confirm that they were understood. During focus group discussions, the moderator sought to clarify any areas of ambiguity. In addition, before the end of each focus group, the moderator gave a brief summary of what had been discussed and asked participants to verify the summary comments. As for generalizability, results of focus group studies are not intended to be generalized because such studies are, by nature, in-depth examinations of particular topics by few participants. The focus group methodology is in contrast with quantitative study methods, which usually seek to offer breadth instead of depth and therefore rely on a large sample to allow for generalizations. Instead of generalizability, authors of other focus group studies have suggested the concept of transferability, which means it is up to the receiver (i.e., you, the reader) to decide whether the results can be applied to a different situation.
15
We acknowledge that you, the reader, may not view these results as transferable to your geographical practice area. Providers in other regions of the United States may have different practice patterns that present a whole new set of challenges regarding referral to eye care. For example, PCPs in some areas may not be aware of recommended eye care guidelines and thus may not refer patients for eye care at regular intervals, whereas PCPs practicing in other rural areas may cite provider time constraints as more important than patients' access to transportation. Because the U.S. health care system differs from the health care systems of other countries in many ways, providers in other countries are likely to face different referral barriers. For example, in South Africa, patients' access to ECPs may be a larger barrier than lack of communication between PCPs and ECPs.
13 Clearly, many factors outside of this study's findings could play a role in eye care referrals. For results to be generalizable, we need a study with more breadth; therefore, we are currently developing a quantitative survey-based study to further evaluate PCP referral to eye care.
This pilot study was designed to learn more about the barriers that PCPs face in the referral-to-eye-care process and what suggestions PCPs have to remedy current referral problems. The key findings suggest that PCPs recognize major problems within the current referral system and desire better communication and collaboration with ECPs and better implementation and utilization of EMRs. ECPs can do a better job of providing prompt and understandable feedback to PCPs regarding their referred patients. The recent allocation of funds via the health information technology program of the American Recovery and Reinvestment Act of 2009 has the potential to expand EMR systems across the United States, even in rural and underserved areas. PCPs and ECPs should seek out these and other ways to improve continuity of care for the millions of patients at high risk for advancing eye disease and vision loss.
Supported by a Research to Prevent Blindness Lew R. Wasserman Merit Award and an Alcon Research Institute Award.
Disclosure:
C.D. Holley, None;
P.P. Lee, None