Lorens A. Helmchen, PhD

 Limited data are available regarding the impact of the type of healthcare delivery system on technology diffusion and associated clinical outcomes. We assessed the adoption of minimally invasive radical prostatectomy (MIRP), a recent clinical innovation, and whether this adoption altered surgical morbidity for prostate cancer surgery.

 Retrospective review of administrative data from TRICARE, the healthcare program of the United States Military Health System. Surgery occurred at military hospitals, supported by federal appropriations, or civilian hospitals, supported by hospital revenue.

 We evaluated TRICARE beneficiaries with prostate cancer (

International Classification of Disease, 9th Revision, Clinical Modification

] code: 185) who received a radical prostatectomy (60.5) between 2005 and 2009. MIRP was identified based on minimally invasive surgery codes (54.21, 17.42). We assessed yearly MIRP utilization, 30-day postoperative complications (Clavien classification system), length of stay, blood transfusion, and long-term urinary incontinence and erectile dysfunction.

 A total of 3366 men underwent radical prostatectomy at military hospitals compared with 1716 at civilian hospitals, with minimal clinic-demographic differences. MIRP adoption was 30% greater at civilian hospitals. There were fewer blood transfusions (odds ratio, 0.44; 

<.0001) and shorter lengths of stay (incidence risk ratio, 0.85; 

<.0001) among civilian hospitals, while 30-day postoperative complications, as well as long-term urinary incontinence and erectile dysfunction rates, were comparable.

 Compared with military hospitals, civilian hospitals had a greater MIRP adoption during this timeframe, but had comparable surgical morbidity.

This is a retrospective review of administrative data from TRICARE, the healthcare program of the United States Military Health System. 

Compared with 1716 men in 767 civilian hospitals, 3366 men underwent radical prostatectomy in 36 military hospitals.

Minimally invasive radical prostatectomy adoption was 30% greater at civilian hospitals. 

The 30-day postoperative complications, as well as long-term urinary incontinence and erectile dysfunction rates, were comparable. 

Adoption of minimally invasive technology has not significantly improved surgical morbidity. 

This study suggests that adoption of technology may not always translate to an observable improvement in patient outcomes.

The rapid pace of clinical innovation—including the development of novel drugs, devices, diagnostic techniques, and surgical interventions&mdash;makes the practice of medicine in the United States dynamic. Hospitals and physicians constantly seek to offer new technologies that are effective and help speed recovery, thereby achieving the best possible results for patients. One example of clinical innovation is minimally invasive surgery (MIS), which is associated with fewer complications across a range of procedures.

 Recently, the introduction of robotic surgery has made MIS feasible for even complex surgical procedures such as radical prostatectomy (RP) for the management of prostate cancer in hopes of reducing its well-recognized surgical morbidity.

 Minimally invasive RP (MIRP) with laparoscopy was first reported in 1997

 and robotic MIRP with the da Vinci Surgical System (Intuitive Surgical, Inc; Sunnyvale, California) was introduced in 2001.

 MIRP has become the most common treatment for men with prostate cancer,

 with the latest 2013 estimates showing 85% of all RPs were performed using the robotic approach, with the remainder still being performed via the open radical prostatectomy (ORP) approach.

Prior investigations report that the widespread dissemination of clinical innovation, such as MIRP, in the United States has been associated with patient, surgeon, and hospital characteristics.

 These studies, however, did not directly examine the question of how the funding mechanism for healthcare services may have altered this adoption. In theory, hospitals that rely on revenue through reimbursement of services (ie, for-profit and nonprofit institutions) will preferentially adopt new clinical innovations at a faster rate because they attract patients, thereby maximizing hospital revenue.

In the current study, we analyzed a contemporary cohort of male enrollees in TRICARE, the healthcare program of the United States Department of Defense (DoD) Military Health System (MHS). These men all have prostate cancer and underwent RP at either civilian hospitals or military hospitals, which have substantially different models for healthcare delivery. We compared cohorts from these disparate healthcare environments to evaluate the hypothesis that civilian hospitals preferentially adopted MIRP compared with MHS hospitals, resulting in an improved surgical morbidity profile for patients with prostate cancer.

TRICARE is the healthcare program serving active-duty and retired service members and their dependents. This program provides 3 types of health plans: Standard (similar to traditional indemnity insurance), Extra (similar to a preferred provider organization), and Prime (similar to a health maintenance organization). Prime enrollees have a primary care manager (PCM) responsible for their care. TRICARE Plus provides enrollees 65 years or older with access to MHS facilities and benefits comparable to Prime, where available. Services provided by TRICARE are via the Direct Care System at MHS facilities where healthcare professionals are salaried, either as employees of the federal government or as hired contractors. There is also a Purchased Care System (PCS), which reimburses for services at civilian facilities, comprising for-profit and nonprofit institutions in which hospitals and surgeons maximize revenue by increasing the volume of care (ie, fee-for-service).

 Although all active-duty service members typically receive care at military treatment facilities, some TRICARE enrollees (eg, retirees or family members) may also receive care in the PCS, depending on where they reside.

We extracted data from October 1, 2005, to September 30, 2010, from the Military Health System Data Repository (MDR), an administrative database, for all TRICARE Prime and Plus beneficiaries within the United States. We focused on Prime and Plus beneficiaries for 2 reasons. First, the assignment of a PCM who oversees care suggests it is less likely that enrollees might receive care that is not performed within the purview of TRICARE and that care provided might be documented more completely. Second, although all TRICARE benefit plans are generous, Prime enrollees have either very low or no enrollment fees and co-payments, and no deductibles, minimizing the possibility that patient out-of-pocket costs might be a factor in treatment selection. We obtained approval from the institutional review boards of the Uniformed Services University of the Health Sciences and Brigham and Women’s Hospital and established a data use agreement with the Defense Health Agency Privacy Office.

We identified 28,998 men with a diagnosis of prostate cancer (

International Classification of Disease, 9th Revision, Clinical Modification 

] code: 185). Of these, 5082 underwent RP (

: 60.5). We examined outcomes describing RP morbidity, including surgical complications and blood transfusions, length of stay (LOS), anastomotic strictures, postoperative incontinence, postoperative erectile dysfunction, and the postoperative complication rates during the 30 days following surgery. Thirty-day surgical complications were determined based on the Clavien classification system, a validated approach for reporting complications

 We evaluated for any complication (Clavien grade ≥1), major complications (Clavien grades 3-5), and mortality (Clavien grade 5). Postoperative mortality was identified through disposition codes in the MDR. Postoperative blood transfusions were also assessed in the 30 days after surgery through a review of the administrative data.

LOS was calculated based on the days between the surgery and discharge from the hospital. The development of anastomotic strictures (

: 598.2, 598.8, 598.9) was captured from 31 to 365 days after surgery, thus limiting the years of analysis from fiscal year (FY) 2005 to FY 2009. Long-term functional status was estimated based on the use of surgical procedures to address incontinence and erectile dysfunction during the 18 months following surgery, thus excluding men undergoing surgery in the latter half of FY 2009 and in FY 2010. The codes used to capture incontinence and erectile dysfunction have been previously described.

We extracted demographic information pertinent to our study cohort, including patient age, race, and marital status, as well as clinical information, such as baseline incontinence and erectile dysfunction. For missing data, efforts were made to cross-reference multiple datasets to obtain accurate demographic information. We dichotomized the surgical approach as ORP (

: 60.5 + 54.21 and/or 17.42). To adjust for patient health status, we utilized the CMS-Hierarchical Condition Category (CMS-HCC) model, which uses demographics and diagnosis profiles captured for both inpatient and ambulatory encounters in order to adjust capitation payments for health expenditure risk.

 We used CMS-HCC both as a continuous and a categorical variable (tertiles).

We summarized patient and surgical characteristics for patients who underwent RP with descriptive statistics. Categorical variables were compared using Pearson’s χ

 test and continuous variables using the Mann-Whitney 

 test. We performed difference-in-differences (DID) estimation, with bootstrapping to calculate the 95% confidence interval (CI), to assess the effect of hospital revenue on adoption trends. To compare postoperative outcomes, we created unadjusted and adjusted regression models to evaluate the odds of postoperative complications, anastomotic strictures, incontinence, erectile dysfunction, and blood transfusion, as well as the incidence rate ratio for LOS. We accounted for clinically important confounders, including patient age, race, marital status, comorbidity (CMS-HCC), surgical approach (ORP vs MIRP), and baseline incontinence and erectile dysfunction. Additionally, we performed subgroup analysis on institutions where only MIRP was performed, examining differences in outcomes. Statistical analyses were performed using SAS version 9.3 (SAS Institute; Cary, North Carolina); all tests were 2-sided, and 

 <.05 was considered statistically significant.

In the study cohort, a total of 3366 men underwent radical prostatectomy in military hospitals compared with 1716 in civilian hospitals across the study period from 2005 to 2009. These surgeries occurred in 767 civilian and 36 military hospitals across the United States. Civilian hospitals were associated with an increased MIRP adoption by approximately 30% (DID estimation: 0.29; 95% CI, 0.19-0.39; 

) over the course of the study. The disparity in MIRP utilization in favor of civilian hospitals was present during the entire study, but substantially increased beginning in 2008.

There were minimal differences between the 2 groups in terms of baseline characteristics (

). Age and comorbidity status were similar between the 2 sectors, while there was a larger proportion of single men in civilian hospitals. There was a greater number of white, black, and Asian/Native American/other patients in the MHS (

<.0001), although the observed ethnic and racial differences may have been secondary to the larger number of unknown/missing data from civilian hospitals.

), the median LOS between the patient cohorts was similar, although both the unadjusted and adjusted analyses demonstrated a reduced LOS for civilian hospitals (incidence risk ratio, 0.85; 95% CI, 0.83-0.87; 

<.0001). Bivariate analyses showed that the shorter LOS for MIRP compared with ORP was more pronounced for civilian hospitals (MIRP: 1.74 days vs ORP: 2.52 days) than military hospitals (MIRP: 2.27 days vs ORP: 2.52 days). The proportion of patients receiving a blood transfusion within 30 days of the surgery was lower in the civilian hospitals (3.5% vs 8%; adjusted odds ratio, 0.44; 95% CI, 0.34-0.58; 

<.0001). When we stratified this further by type of surgery performed, we found that the lower blood transfusion rate for MIRP was true for civilian hospitals (MIRP: 2.0% vs ORP: 8.9%) but not for military hospitals (MIRP: 4.4% vs ORP: 4.4%).

In contrast, the observed 30-day postoperative complication rates were highly comparable on both the unadjusted and adjusted analyses. Although the 30-day mortality rate was lower for civilian hospitals (0.03% vs 0.12%), this did not achieve statistical significance. Similarly, we found that the development of an anastomotic stricture following surgery was lower among men in civilian hospitals (6.7% vs 9.1%), but our adjusted analysis failed to identify a statistically significant difference between men in the 2 cohorts. Assessment of long-term functional outcomes revealed that the utilization of surgery to address erectile dysfunction and incontinence was similar between men cared for in either healthcare environment. On subgroup analysis including only hospitals that performed MIRP, we also found no differences in terms of postoperative complications, incontinence, or erectile dysfunction (data not shown).

In this study of TRICARE beneficiaries with prostate cancer, we found a substantially greater diffusion of MIRP among civilian institutions compared with MHS institutions between FY 2005 and FY 2010, and this difference was most pronounced after 2008, coinciding with the widespread adoption of robotic MIRP in the United States.

 Despite the increasing disparity in MIRP diffusion, we observed similarity in surgical morbidity for these 2 healthcare environments based on postoperative complications and functional outcomes (ie, incontinence and erectile dysfunction). Our findings support the hypothesis that market forces may accelerate the adoption of clinical innovation, such as MIRP, among civilian hospitals, which rely heavily on generating revenue to deliver healthcare; however, the increased tendency to adopt MIRP was not associated with a meaningful improvement in surgical morbidity.

The strength of the current study is that we assessed men from a common insured population that were managed in 2 different healthcare environments distinguished, in large part, by the role of hospital revenue source in healthcare delivery. Because the delivery of health services at civilian institutions is reliant on generating sufficient revenue, it is feasible that the motivation for technology adoption stems from an aim to maximize revenue by attracting new patients or avoiding a loss of patients.

 In contrast to civilian hospitals, the MHS is supported by appropriations from the DoD. These budgetary decisions have an unusually broad scope of maintaining the health of all military personnel, their dependents, and retirees, as well as supporting military readiness, providing medical education, and improving public health. These competing priorities, coupled with the idiosyncrasies of federal appropriations laws and budgeting, may slow the adoption of technology within the MHS.

Prior population-based studies of nonfederal hospitals report that MIRP is associated with increased surgical volume. Chang et al demonstrated that surgeons, particularly high-volume surgeons, who adopted robotic MIRP experienced an increase in overall annual volume of RP.

 Similarly, Makarov et al reported that hospitals that acquired the robotic platform were associated with a significant increase in annual RP volume, while hospitals without the robotic platform experienced a loss in volume.

 Complementing the prior investigations, this study provides data that support the hypothesis that maximizing hospital revenue may have influenced the likelihood of adopting technology such as MIRP (Figure).

Despite the argument that hospital revenue drives technology dissemination, we also observed a substantial increase in the proportion of patients undergoing MIRP in the MHS during the final 2 years of the study. We speculated that this upward trend may be associated with a motivation to maintain parity with civilian hospitals

 and to keep TRICARE enrollees from seeking care at civilian hospitals to undergo MIRP. Surgeons at civilian hospitals may feel more of a need to learn new techniques to remain competitive for patient referrals in a “private” environment. In terms of reimbursement incentives at the physician level, a common misconception among nonurologists is that perhaps MIRP offers better reimbursement for the physician than ORP. However, this has not been found to be true; reimbursement to physicians for both ORP and MIRP has been found to be similar on a per-case basis.

Regardless of the true motivation for this increased utilization by military hospitals of MIRP in the latter years of our study, we concluded that the adoption of technology is influenced by a multitude of factors, among which generating revenue is likely to represent one of the most important issues for civilian hospitals. Although both civilian and military hospitals were observed to have an increasing trend in utilization of MIRP, the proportion in military hospitals was lower (Figure) compared with civilian hospitals across all years. We hypothesized that this could be due to certain military facilities having an overall lower caseload of RP, which does not justify the establishment of an MIS program, and which has substantial upfront investment and ongoing maintenance costs. Although this may appear to be a lack of technological diffusion in the military system, it might actually point to prudent adoption of technology in select facilities that have the necessary surgical volume to allow for cost-conscious and safe dissemination.

Our study design permitted an evaluation of the possible impact of MIRP adoption on patient outcomes. We found that civilian hospital procedures were associated with a decrease in blood transfusions and a slightly shorter LOS, which are commonly noted among patients undergoing MIS procedures.

 There was, however, no difference between procedures performed in civilian and military hospitals in the 30-day major complication rate (ie, Clavien grades 3-5) or in the long-term functional outcomes of postoperative incontinence and erectile function. The incidence of anastomotic strictures was also lower in civilian hospitals, but this did not achieve statistical significance in the adjusted analysis. Even with subgroup analysis focusing only on institutions offering MIRP, we did not detect any differences in outcomes between civilian and military hospitals, suggesting that our findings are not influenced by unmeasured factors associated with the availability of MIRP (eg, the availability of the robotic platform).

Our finding of relative equivalence in morbidity between healthcare systems, despite differences in the adoption of MIRP, is consistent with contemporary population-based studies that report mixed results regarding the perioperative outcomes of ORP and MIRP.

 There are, however, unmeasured factors, such as differences in the time for convalescence, total pain medication use, and patient satisfaction, which may preferentially favor MIRP given the decrease in surgical trauma associated with MIS.

 In our adjusted model, LOS and blood transfusion were slightly reduced for civilian hospitals even after adjusting for surgical approach (Table 2), possibly reflecting advantages at the civilian hospitals that are not quantified by the Clavien classification system for surgical complications or captured by assessment of functional outcomes.

Given the absence of a morbidity improvement for MIRP, we concluded that its rapid adoption among civilian hospitals was not based on data in the medical literature. In fact, well-executed studies regarding MIRP are only recently available.

 There are challenging obstacles for critical evaluation through a randomized controlled trial, most notably: 1) the difficulty in accruing patients willing to defer their choice for management to randomization, 2) the impracticality in blinding patients and surgeons, and 3) the challenges in controlling for surgeon learning curve and variations in technique.

 We surmised that the promise of new surgical innovation to improve outcomes is intensely attractive, particularly for a procedure like RP, which is associated with detrimental functional issues, such as urinary incontinence and erectile dysfunction, thereby creating pressure for surgeons and hospitals to explore new technologies. Consequently, despite the absence of Level I evidence demonstrating superiority or equivalence of MIRP over ORP, a large percentage of surgeons in the United States has now adopted robotic MIRP,

 and the majority of RP surgeries are now performed with the robotic platform.

Our findings must be interpreted within the context of several important limitations to our study design. First, we used administrative data from TRICARE (claims for reimbursement for civilian hospitals and utilization records for military hospitals), which is not designed for documenting a patient’s clinical course and outcome. It is possible that more granular clinical information might have altered the associations we report or revealed new relationships not identified in the current study. However, the TRICARE claims data share many similarities with Medicare claims data, which have been used extensively to evaluate surgical outcomes.

Second, we could not adjust for tumor stage and grade because these characteristics are not reliably captured with claims data; however, in men with prostate cancer, tumor characteristics are unlikely to influence the postoperative complications, LOS, and long-term functional outcomes. Third, to assess for postoperative urinary incontinence and erectile dysfunction, we considered surgical intervention as a proxy for functional status. Admittedly, we cannot exclude the possibility that there may be a disparity in the threshold to undergo those surgical interventions between civilian and military institutions.

Fourth, although we had hospital volume data for military hospitals, we did not have that same reliable data for civilian hospitals, precluding us from adjusting for surgical volume as a possible confounder, especially when this has been previously shown to affect postoperative outcomes.

 Lastly, we did not have reliable estimates for costs and revenues to comprehensively assess the differential amount of revenue generated by adoption of MIRP over ORP; this may have garnered additional insights regarding the potential financial incentives of adopting MIRP.

To our knowledge, this is the first study assessing the potential impact of the type of healthcare delivery system on the adoption of MIRP in the United States. Based on TRICARE enrollee data, we observed that adoption was more rapid among civilian hospitals, which are comprised of institutions heavily reliant on generating hospital revenue to deliver health services. Further analysis, however, showed that the increased use of MIRP did not translate into a marked improvement in outcomes, thus highlighting the potential dangers of embracing new clinical innovations prior to the availability of evidence showing benefit. Our findings are particularly relevant as the delivery of healthcare in the United States is undergoing dramatic change. Future work should focus on determining if the goals of cost containment and of revenue based on quality, the principles underlying accountable care organizations, promote or repress clinical innovation in the United States.

Center for Surgery and Public Health (JJL, JSW, WJ, NH, SRL, LLN, SLC), and Division of Urology (JJL, SLC), and Division of Vascular Surgery (LLN), Brigham and Women’s Hospital, Harvard Medical School, Boston, MA; Department of Preventive Medicine and Biostatistics, Uniformed Services University of the Health Services (LK, AH), Bethesda, MD; Department of Health Policy and Management, Milken Institute School of Public Health, The George Washington University (LHH), Washington, DC.

 Henry M. Jackson Foundation HU0001-11-1-0023 The Comparative Effectiveness and Provider Induced Demand Collaboration (EPIC): A Clinical and Economic Analysis of Variation in Healthcare.

 The authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

 Concept and design (JJL, JSW, LK, LHH, WJ, NH, SRL, LLN, SLC); acquisition of data (LK, AH, LLN, SLC); analysis and interpretation of data (JJL, LK, AH, LHH, WJ, NH, SRL, LLN, SLC); drafting of the manuscript (JJL, JSW, AH); critical revision of the manuscript for important intellectual content (JJL, LK, LHH, WJ, NH, SRL, LLN, SLC); statistical analysis (JJL, WJ, NH, SRL, LLN, SLC); provision of patients or study materials (LK); obtaining funding (LLN, SLC); administrative, technical, or logistic support (AH, LLN, SLC); and supervision (JSW, LK, LHH, WJ, NH, SRL, LLN, SLC).

 Steven L. Chang, MD, MS, Brigham and Women’s Hospital, Division of Urology, 45 Francis St, Boston, MA 02115. E-mail: slchang@bwh.harvard.edu. 

1. Xu T, Hutfless SM, Cooper MA, Zhou M, Massie AB, Makary MA. Hospital cost implications of increased use of minimally invasive surgery. 

 2015;150(5):489-490. doi: 10.1001/jamasurg.2014.4052.

3. Sanda MG, Dunn RL, Michalski J, et al. Quality of life and satisfaction with outcome among prostate-cancer survivors. 

 2008;358(12):1250-1261. doi: 10.1056/NEJMoa074311.

4. Shore N. Management of early-stage prostate cancer. 

5. Schuessler WW, Schulam PG, Clayman RV, Kavoussi LR. Laparoscopic radical prostatectomy: initial short-term experience. 

 Laparoscopic radical prostatectomy with a remote controlled robot. 

7. Binder J, Kramer W. Robotically-assisted laparoscopic radical prostatectomy. 

 Perioperative outcomes of robot-assisted radical prostatectomy compared with open radical prostatectomy: results from the nationwide inpatient sample. 

 2012;61(4):679-685. doi: 10.1016/j.eururo.2011.12.027.

9. Gandaglia G, Sammon JD, Chang SL, et al

 Comparative effectiveness of robot-assisted and open radical prostatectomy in the postdissemination era. 

 2014;32(14):1419-1426. doi: 10.1200/JCO.2013.53.5096.

Articles

Minimally invasive radical prostatectomy was more commonly performed in civilian hospitals compared with military hospitals among TRICARE beneficiaries, with comparable postoperative outcomes.

Radical Prostatectomy Innovation and Outcomes at Military and Civilian Institutions

Nearly 4 in 10 Americans with diabetes currently fail to undergo recommended annual retinal exams, resulting in tens of thousands of cases of blindness that could have been prevented. Advances in automated retinal disease detection could greatly reduce the burden of labor-intensive dilated retinal examinations by ophthalmologists and optometrists and deliver diagnostic services at lower cost. As the current availability of ophthalmologists and optometrists is inadequate to screen all patients at risk every year, automated screening systems deployed in primary care settings and even in patients’ homes could fill the current gap in supply. Expanding screens to all patients at risk by switching to automated detection systems would in turn yield significantly higher rates of detecting and treating diabetic retinopathy per dilated retinal examination. Fewer diabetic patients would develop complications such as blindness, while ophthalmologists could focus on more complex cases.

Am J Manag Care. 2014;20(11 Spec No. 17):eSP48-eSP52

Automated image-based detection of retinal disease is rapid and can obviate the labor-intensive method of image interpretation and data entry by physicians.

Low-cost retinal imagers that reliably detect diabetic retinopathy and are designed for use by patients have been prototyped and may soon aid in the early detection of other conditions.

By raising the number of retinopathy cases detected per office visit, automated screening systems could compensate for the emerging shortfall in office-based retinal exams.

The significantly higher number of retinal abnormalities that automated systems can help detect, coupled with the rising prevalence of diabetes, will largely sustain the demand for ophthalmology services.

Given the rapidly growing prevalence of a number of common chronic conditions, especially diabetes, it is difficult to conceive how the resulting sharp increase in the demand for recurring diagnostic and therapeutic services can be met without significant increases in physician productivity. Information technology (IT) in particular has been advanced as a potential source of productivity increases in healthcare.

Another approach centers on the implementation of electronic health records (EHRs). Yet, some healthcare providers remain hesitant to adopt EHR systems because these typically entail large upfront costs of installation, training, and work flow redesign; can be difficult to use

; and to date, reports of the success of EHRs in increasing productivity have been mixed.

 Another approach centers on the increased use of computeraided diagnostics, which involves sophisticated computer algorithms that analyze a patient’s medical images and mark suspicious elements of the images, thereby assisting the examining physician’s detection of disease.

Because these systems continue to require image interpretation and data entry by physicians, however, their productivity potential is limited.

The current standard for detection of diabetic retinopathy is a dilated retinal examination by an ophthalmologist or optometrist, using indirect ophthalmoscopy and a handheld lens (Figure).

A patient’s typical visit takes about 2 hours to allow time for pupil dilation, consultation with the ophthalmologist or optometrist, the patient’s recovery, and documentation of the results in the patient’s medical chart or EHR.

By contrast, automated image-based detection of disease takes less than 2 minutes

 and does not require the labor-intensive interpretation of medical images and data entry by ophthalmologists or optometrists. Low-cost, compact retinal imagers have been prototyped that allow patients to perform retinal scans on their own.

Moreover, “smart” retinal imagers are equipped with image-analysis algorithms that evaluate the camera-generated digital images for retinal abnormalities and automatically flag patients for a followup assessment by a physician, including a confirmatory dilated retinal examination and possibly treatment by an ophthalmologist.

 To date, patient-operated “smart” retinal imagers have only been used on an experimental basis in research settings. Once approved by the FDA, however, these devices could be made available for sale, for lease, or even for free in primary care offices, minute clinics, pharmacies, and other healthcare facilities, where trained personnel could instruct patients on how to use them properly. The widespread availability of a device that enables patients to screen themselves for retinal abnormalities will offer the potential of reaching more persons with diabetes more frequently than is possible by following the current practice of relying exclusively on dilated retinal examinations.

Although the precise integration of results from automated disease-detection devices into existing work flow designs remains to be determined, one option would follow the model of current diagnostic home-use tests for HIV or pregnancy: patients would periodically screen themselves for retinal abnormalities and save both the images and their automated interpretation, possibly following the Continuity of Care Document standard

Patients will be able to access the results electronically and share them with an ophthalmologist for further examination and for inclusion in their EHRs. Analogous to the follow-up care for positive HIV and pregnancy tests, if the device detected a retinal abnormality, patients would first consult with their primary care physicians who in turn might refer them to an ophthalmologist for a confirmatory dilated retinal examination and possibly treatment.

The Public Health Potential of Automated Diagnostics

Although diabetic retinopathy can be detected early and treated effectively, it remains a serious and widely feared complication that accounts for more than 24,000 largely preventable new cases of blindness and nearly 50,000 new cases of vision loss annually in the United States.

 Although current guidelines recommend that all persons with diabetes in the United States undergo a retinal exam every year, in 2010 ophthalmologists and optometrists examined only about 63% of the 25.8 million Americans with diabetes, or about 16.

 Potential causes include the limited number and availability of ophthalmologists and optometrists, many of whom operate at full capacity; the limited number of and access to high-quality digital retinal cameras

; patients’ ignorance of screening recommendations; and the reluctance of asymptomatic patients to incur the monetary and time costs associated with an examination.

Automated retinal imaging could be used to screen all persons with diabetes annually and thereby identify those who require further examination and possibly treatment. To assess the impact of such automated detection systems on patient health and the cost of care, it is useful to compare their sensitivity and specificity statistics to those achieved by ophthalmologists and optometrists. As with any screening process, minimizing false negatives (high sensitivity) is vital to ensure that patients excluded from further assessment as a result of the screen will not suffer from the lack of indicated treatment, which in this case would prevent vision loss and improve the patient’s quality of life.

More reliable early detection of the disease could also reduce the risk of a professional liability claim arising from a missed diagnosis. On the other hand, by minimizing false positives, highly specific detection systems reduce unwarranted dilated retinal examinations by ophthalmologists or optometrists, thus lowering the cost to patients without the disease (and their insurers). The sensitivity levels of detecting referable retinopathy achieved by ophthalmologists, optometrists, and retinal specialists have been reported to range from 34%

 with specificity levels reaching 98%.9 By comparison, in a recent study the sensitivity and specificity levels achieved by an automated system reached 97% and 59%.

The burden and cost implications are discussed next.

As ophthalmologists and optometrists currently examine only 63% of all persons with diabetes, or about 16.25 million, at a sensitivity level as high as 91% they are currently identifying 63% × 91% = 57.3% of all patients who require treatment, for a Likelihood Ratio Positive (LR+) of 45.50 and a Likelihood Ratio Negative (LR—) of 0.43 (

). By contrast, if all persons with diabetes under went an initial retinal exam using an automated detection system, the number referred to an ophthalmologist for a confirmatory dilated retinal examination and possibly treatment would drop to 12.02 million; these systems would flag 97% of the estimated 2.58 million persons with diabetes who every year develop retinopathy that requires clinical management,

or about 2.50 million, and 100% — 59% = 41% of the remaining 23.22 million patients who do not require treatment, or 9.52 million. In combination with the follow-up examinations by ophthalmologists, this hypothetical screening practice would, in assessing the overall process, yield both a higher Likelihood Ratio Positive (108.00) and a lower Likelihood Ratio Negative (0.12) than the current screening practice. Screening all persons with diabetes for retinal abnormalities using automated detection systems would cut in half the typical patient’s number of office-based dilated retinal examinations per annum. Perhaps most important, it would raise the proportion of patients requiring treatment who receive it from 57% to 88%, thereby preventing 54% more cases of partial vision loss or blindness.

If the adoption of automated detection systems left the proportion of persons with diabetes undergoing screening unchanged at its current level of 63%, the number of referrals to ophthalmologists would drop from 16.25 to 7.57 million, while the proportion of patients requiring treatment who receive it would decline minimally from 57% to 56%. Thus, automated detection systems would yield a significantly higher rate of detecting and treating diabetic retinopathy per dilated retinal examination. A full cost-effectiveness analysis of universal automated screening is beyond the scope of this report.

Even at the comparatively lower levels of specificity, automated screening allows patients to avoid on average every other office-based dilated retinal exam they would be scheduled to undergo under the current testing recommendations. At the same time, the relatively high rate of false positives under automated screening might lead some patients to discount a positive test result and discourage them from promptly seeking a confirmatory dilated retinal examination, even if they knew that a delay in treatment might result in impairment or total loss of vision. Educational campaigns, reminders, and possibly modifications to the patient’s health insurance benefit could be considered to minimize the risk that a patient will ignore a positive test result.

A Complement to Current Screening Practice

Automated image-based disease detection could complement the important role that ophthalmologists and optometrists currently fill in diagnosing diabetic retinopathy: it could reduce the time required of these specialists to interpret medical images, automatically add the results to the patient’s record, and overcome possible biases that may skew the clinical assessments. Once approved by the FDA, such an initial examination could be performed by anyone with the device, including patients themselves, which would reduce the cost of universal automated screening even more. More frequent detection of retinal abnormalities would not just prevent substantially more cases of blindness; it would also help identify patients with badly managed diabetic disease more generally and could eventually aid in the early detection of cardiovascular disease, macular degeneration, glaucoma, malaria, and Alzheimer’s disease.

Given that the prevalence of diabetes in the United States is projected to increase by 50% over the coming decade

 and given the expected increase in insurance coverage as provisions of the Affordable Care Act are implemented, automated initial screening may represent an attractive low-cost option to meet the growing demand for routine diagnostic services.

Paradoxically, greater use of automated detection systems, coupled with the rising prevalence of diabetes, could largely sustain the demand for ophthalmologist services. Even if automated detection systems replace all initial retinal examinations currently performed by ophthalmologists and optometrists, expanding these screens to all persons with diabetes implies that ophthalmologists will continue to perform about three-fourths of the current volume of examinations. Moreover, the significantly higher number of retinal abnormalities that automated systems help detect implies that patients seeking a follow-up exam will now be much more likely to require treatment, thus allowing ophthalmologists to concentrate on more complex, and possibly more rewarding, tasks.

 From Department of Health Administration and Policy, George Mason University (LAH), Fairfax, VA; Division of Health Sciences Informatics, Johns Hopkins Bloomberg School of Public Health (HPL), Baltimore, MD; and Department of Ophthalmology and Visual Sciences, and Stephen A. Wynn Institute for Vision Research, University of Iowa Hospital and Clinics (MDA), Iowa City, IA.

 Dr Abràmoff gratefully acknowledges support through NIH EY-017989, EY019112, EY018853 and VA I01 CX000119.

 Dr Abràmoff is named inventor on patents, owned by the University of Iowa, related to some of the technology described in this manuscript, and he is the founder, shareholder, and president of IDx LLC, of Iowa City, Iowa, which has licensed these patents. He has pending grants from the National Institutes of Health (NIH), the Department of Veterans Affairs (VA), Beckman Foundation, and Research to Prevent Blindness, and has received grants from NIH, the VA, and Beckman Foundation. Drs Helmchen and Lehmann report no conflicts of interest.

 Concept and design (LAH, MDA); acquisition of data (MDA); analysis and interpretation of data (LAH, HPL, MDA); drafting of the manuscript (LAH, HPL, MDA); critical revision of the manuscript for important intellectual content (LAH, HPL, MDA); statistical analysis (MDA); provision of study materials or patients (MDA); obtaining funding (MDA); administrative, technical, or logistic support (MDA); and supervision (MDA).

 Lorens A. Helmchen, PhD, Associate Professor, Department of Health Administration and Policy, George Mason University, 4400 University Dr — MS: 1J3, Fairfax, VA 22030. E-mail: 

1. Buntin MB, Burke MF, Hoaglin MC, Blumenthal D. The benefits of health information technology: a review of the recent literature shows predominantly positive results. 

2. Chiang MF, Boland MV, Margolis JW, Lum F, Abràmoff MD, Hildebrand PL. Adoption and Perceptions of Electronic Health Record Systems by Ophthalmologists: An American Academy of Ophthalmology Survey. 

3. Wang SJ, Middleton B, Prosser LA, et al. A cost-benefit analysis of electronic medical records in primary care. 

4. Bhargava HK, Mishra A. Electronic Medical Records and Physician Productivity: Evidence from Panel Data Analysis. Social Science Research Network; 2011.

5. O’Reilly D, Tarride JE, Goeree R, Lokker C, McKibbon KA. The economics of health information technology in medication management: a systematic review of economic evaluations. 

6. Sonka M, Hlavac V, Boyle R. Image processing, analysis, and machine vision. Thompson Learning; 2008.

7. Fenton JJ, Taplin SH, Carney PA, et al. Influence of computer-aided detection on performance of screening mammography. 

8. American Academy of Ophthalmology Retina Panel. Preferred practice patterns: diabetic retinopathy. San Francisco, CA: American Academy of Ophthalmology; 2008.

9. Abràmoff MD, Folk JC, Han DP, et al. Automated analysis of retinal images for detection of referable diabetic retinopathy. 

10. Williams GA, Scott IU, Haller JA, Maguire AM, Marcus D, McDonald HR. Single-field fundus photography for diabetic retinopathy screening: a report by the American Academy of Ophthalmology. 

11. Goatman K, Charnley A, Webster L, Nussey S. Assessment of automated disease detection in diabetic retinopathy screening using two-field photography.

12. D’Amore JD, Sittig DF, Ness RB. How the Continuity of Care Document can advance medical research and public health. 

13. Blumenthal D, Tavenner M. The “meaningful use” regulation for electronic health records. 

14. Huang ES, Brown SE, Ewigman BG, Foley EC, Meltzer DO. Patient perceptions of quality of life with diabetes-related complications and treatments. 

15. Diabetes Report Card 2012. Atlanta, GA: HHS;2012.

16. Abràmoff MD; Garvin MK; Sonka M. Retinal imaging and image analysis. 

. 2010;3:169-208. doi: 10.1109/RBME.2010.2084567.

17. Abràmoff MD, Reinhardt JM, Russell SR, et al. Automated early detection of diabetic retinopathy. 

18. Lin DY, Blumenkranz MS, Brothers RJ, Grosvenor DM. The sensitivity and specificity of single-field nonmydriatic monochromatic digital fundus photography with remote image interpretation for diabetic retinopathy screening: a comparison with ophthalmoscopy and standardized mydriatic color photography. 

19. Jones CD, Greenwood RH, Misra A, Bachmann MO. Incidence and progression of diabetic retinopathy during 17 years of a population-based screening program in England. 

20. Xu X, Niemeijer M, Song Q, et al. Vessel boundary delineation on fundus images using graph-based approach.

21. Abràmoff MD, Alward WLM, Greenlee EC, et al. Automated segmentation of the optic disc from stereo color photographs using physiologically plausible features. 

22. Joshi VS, Maude RJ, Reinhardt JM, et al. Automated detection of malarial retinopathy-associated retinal hemorrhages.

23. Xu X, Abràmoff MD, Bertelsen G, Reinhardt JM. Retinal vessel width measurement at branching points using an improved electric field theory-based graph approach. 

24. Huang ES, Basu A, O’Grady M, Capretta JC. Projecting the future diabetes population size and related costs for the U.S. 

With diabetes rates projected to rise sharply, automated retinal screening may represent an attractive low-cost option to meet the growing demand for routine screening services.

Lorens A. Helmchen, PhD

Articles

Radical Prostatectomy Innovation and Outcomes at Military and Civilian Institutions

Automated Detection of Retinal Disease