In the outpatient setting, it is exceedingly difficult to know what medications our patients have been prescribed and are taking. Each encounter with a specialist, hospital, or pharmacy can generate a change to a patient’s list of medications, and in most systems, this information is not communicated back to the primary care practice’s electronic health record—the exception being opiate prescriptions. Prescription drug monitoring programs in 48 states list every opiate prescription, the name of the prescriber, and the date and location the prescription was picked up. We propose that policy makers act to expand these programs to all medications, thus improving the likelihood that any provider prescribing a new medication would know what medicines their patient is already taking.

	A statewide program that collects and reports what medications individuals have been prescribed and picked up from the pharmacy has the potential to save many lives through reductions in medication errors.  

		Studies demonstrate that electronic health record medication lists often contain errors or omissions. 

		Medication reconciliation—verifying the list of medications that a patient takes&mdash;is difficult, and increasingly so, due to multiple factors. 

		The success of prescription drug monitoring programs at reducing erroneous opiate prescriptions offers hope that such a program would work for all medications.

Medication reconciliation is the process of maintaining the list of medications a patient takes. Few clinical tasks are as important—or as deceptively simple. It involves knowing which medications have been prescribed; their dosages, frequencies, and durations; interactions between medications; whether a patient has actually filled the prescriptions; and whether patients have been taking them as directed. In our clinical practice, the peeved patient will say to us, “I take a white pill, a brown pill, and a little blue pill. I don’t know what the names are, doc. Don’t you have that in the computer?” An accurate medication reconciliation relies on diligence from patients and all of their providers; however, diligence alone is not sufficient.

In 2005, research indicated that 54% of patients being admitted to the hospital, and 14% of patients being discharged, had a discrepancy in their medication lists.

 Such discrepancies can result in poor medication adherence and in preventable adverse drug events (ADEs). The Joint Commission listed medication reconciliation as a National Patient Safety Goal in 2005 and began scoring it in accreditation. In 2009, CMS enacted Meaningful Use, which includes medication reconciliation as a core objective for both inpatient and outpatient settings. The growth of electronic health records (EHRs) and prescribing systems has led to over half of new and renewed prescriptions being sent to pharmacies electronically.

 As with so much that ails healthcare, the hope and expectation was that technology and incentives would improve medication reconciliation and reduce ADEs.

Yet, 10 years later, our efforts have produced little change. Although studies in this area are typically small and single-site, most are discouraging. In a 2015 study of emergency departments, only 21.9% of medication lists matched what the patient took at home,

and in a 2014 study of primary care clinics, only 15% matched.

 Another study looking at hospital discharges found that 81% of adults 65 years or older either were subject to a provider error in their medication list or were unaware of at least 1 medication change having taken place.

The problem is inherently hard, and the failure to solve it is multifactorial. On the patient side, individuals have differing abilities to learn and stay on top of what medications they take, and how this list changes; this is a particularly difficult task for the 20% of adults 65 years or older who take more than 10 medications.

 In addition, patients may never pick up their medications, may pick them up from different pharmacies, and may take them inconsistently or not at all. On the provider side, it is cumbersome to check long medication lists at each encounter, particularly if there are more pressing matters to address. Doses are important, and high-risk medications (eg, anticoagulants, insulins, diuretics, and antihypertensives) often need frequent dose adjustments.

Moreover, providers may be unclear on who is “in charge” of a medication—for example, the cardiologist or the primary care provider&mdash;leading to inconsistent dose changes. Finally, EHRs of different health systems rarely interact, meaning that any patient seen at more than 1 health system will have at least 2 separate medication lists, which, at best, will have a time lag to mutually update, and, at worst, discrepancies leading to preventable ADEs. Therein lies the key to why interventions have not worked: medication reconciliation within a single system misses the point. Any effort with a chance at accuracy must take place at a higher level, including all systems, pharmacies, and providers that a patient encounters. For this reason, we call for a statewide medication reconciliation program (SMRP).

The ideal SMRP should have features as follows. It should be a single source—a “master list”&mdash;used by all prescribers, existing beyond a single health system, such that all health systems, prescribers, and pharmacies for the state should be enrolled. It should be electronic, with real-time updating of changes to the list, as well as what the patient has picked up. The list would have both current medications and a history of what a patient had received in the past, including at what sites, and providers would be able to make annotations. The SMRP would be organized in a way that users find intuitive and easy to use, hopefully avoiding the “600 clicks” that many EHRs require of providers.

 Individual health systems’ EHRs would be interoperable with the SMRP, avoiding double work for the provider. Additionally, a patient portal would allow patients to not only to see their current medication list, but to potentially be an active collaborator in keeping it up-to-date. All of this should be managed by a multidisciplinary board, including providers and pharmacists.

The concept does have a precedent. Every state, except Missouri, has a Prescription Drug Monitoring Program (PDMP), which is a statewide database tracking narcotic prescriptions. The goal of PDMPs is to clamp down on “pill mills” and “doctor shopping” for opiates; however, they can also serve as a tool for providers in deciding whether to prescribe opiates for a patient. To work successfully, PDMPs require comprehensive registration of providers and pharmacies, as well as the maintenance of an electronic system that is separate from the EHR system and must serve different users in different settings while maintaining patient privacy. They have also taken various forms throughout the states, with different successes, problems, and failures.

 As such, PDMPs can provide experimental evidence for how best to organize an SMRP.

The SMRP presents barriers and limitations regarding politics and privacy. In terms of politics, making a usable SMRP would involve coordinating many parties with diverse interests, which might require significant money and incentives. Health systems may view it to their disadvantage to make their EHRs interoperable with an SMRP. Providers may balk at registering for the SMRP and at using more than a single system for their daily work, and pharmacies, too, may balk at registering—particularly as this would represent a burden for pharmacies that do not already use e-prescribing systems. If it has been difficult to get universal buy-in for PDMPs, it will be more difficult for the SMRP.

In terms of privacy, the problem is 2-fold. First, any EHR is vulnerable to cyber-attack, and an SMRP connected to many individual health systems would need appropriate encryption and security to avoid data breaches. Second, our current culture of patient privacy allows for data sharing within a health system, but not beyond; the sharing of information in an SMRP would need to hew to privacy laws. It is important to note, however, that data privacy is not a concern that is exclusive to healthcare, nor is the current system without problems.

Learning what medications a patient has been prescribed and is actually taking is an inefficient, difficult, and absolutely vital task, and our patients bear the consequences of inaccuracies. Nevertheless, despite the barriers, we believe the dangers and limitations of the current system make establishing an SMRP not only desirable, but necessary. Increasingly, the advancement of healthcare quality and safety requires coordination and interoperability to address systemic problems. The EHR has been an effective tool, although not a panacea, for what is actually required is a re-envisioning of medication reconciliation. PDMPs show us the that registry and formation of a database is realizable. Now is the right time for action.

 Department of Medicine, University of California, San Francisco (EA), San Francisco, CA; Department of Medicine, The MetroHealth System, Case Western Reserve University (DM), Cleveland, OH.

 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 (EA, DM); drafting of the manuscript (EA, DM); critical revision of the manuscript for important intellectual content (EA, DM); administrative, technical, or logistic support (EA, DM).

 Elisabeth Askin, MD, University of California, San Francisco, 1545 Divisadero St, San Francisco, CA 94115. E-mail: elisabeth.askin@ucsf.edu.

1. Coleman EA, Smith JD, Raha D, Min SJ. Posthospital medication discrepancies: prevalence and contributing factors. 

2. Cornish PL, Knowles SR, Marchesano R, et al. Unintended medication discrepancies at the time of hospital admission. 

3. DeSalvo KB, Mertz K. Broadening the view of interoperability to include person-centeredness.

. 2015;30(suppl 1):S1-S2. doi: 10.1007/s11606-014-3096-2.

4. Monte AA, Anderson P, Hoppe JA, Weinshilboum RM, Vasiliou V, Heard KJ. Accuracy of electronic medical record medication reconciliation in emergency department patients. 

2015;49(1):78-84. doi: 10.1016/j.jemermed.2014.12.052.

5. Reedy AB, Yeh JY, Nowacki AS, Hickner J. Patient, physician, medical assistant, and office visit factors associated with medication list agreement. 

2016;12(1):18-24. doi: 10.1097/PTS.0000000000000100.

6. Ziaeian B, Araujo KL, Van Ness PH, Horwitz LI. Medication reconciliation accuracy and patient understanding of intended medication changes on hospital discharge. 

2012;27(11):1513-1520. doi: 10.1007/s11606-012-2168-4.

7. Slone Epidemiology Center at Boston University. Patterns of medication use in the United States—2006: a report from the Slone Survey. Boston University website. http://www.bu.edu/slone/files/2012/11/SloneSurveyReport2006.pdf. Published 2006. Accessed September 8, 2015.

The Digital Doctor: Hope, Hype, and Harm at the Dawn of Medicine’s Computer Age.

9. Perrone J, Nelson LS. Medication reconciliation for controlled substances—an “ideal” prescription-drug monitoring program. 

2012;366(25):2341-2343. doi: 10.1056/NEJMp1204493. 

Articles

WEB EXCLUSIVE

The authors propose statewide programs that would expand the Prescription Drug Monitoring Program for all medications-not just opiates-in order to reduce outpatient medication errors.

A Call for a Statewide Medication Reconciliation Program

To determine whether treating hospitalists can identify and address early postdischarge problems through a structured telephone call.

We studied patients insured through a managed care program who were discharged from a general internal medicine service of a university-affiliated public safety net hospital (Denver Health Medical Center) between March 1, 2012, and October 31, 2013. The hospitalist who treated the patient during their hospitalization contacted them 48 to 72 hours after discharge and completed a structured telephone assessment. We assessed the type and frequency of problems identified, the proportion of calls in which problems were independently addressed by the hospitalist, the proportion referred for additional managed care services, and the duration of calls and subsequent care coordination.

Treating hospitalists identified 1 or more problems in 74 of the 131 patients (56%) contacted. The most common categories of problems were: new or worsening symptoms (41%), difficulty accessing recommended follow-up care (21%), and medication issues (20%). Hospitalists independently managed the problems identified in 68% of the calls; additional services were required in 32%. Median time spent per call was 8 minutes (interquartile range, 5-12).

 Treating hospitalists identified problems in over half of patients contacted by telephone shortly after discharge, the largest proportion of which were new or worsening symptoms. Hospitalists were able to address the majority of problems identified through the single, brief telephone encounter without utilizing additional resources.

	We studied whether treating hospitalists could identify and address early postdischarge problems through a structured telephone call made 48 to 72 hours after hospital discharge.  

		Hospitalists discovered problems in 56% of patients contacted, including a large burden of new or worsening symptoms. The problems identified in 68% of calls were managed by the treating hospitalist through the single, brief telephone encounter without requiring additional managed care resources. 

		Telephone calls by treating hospitalists may be an effective and efficient method of identifying and alleviating early postdischarge problems. Managed care organizations might consider partnering with hospitalists to enhance patient safety following discharge.

The period immediately following hospital discharge has been identified as a major patient-safety gap in which misunderstandings and medical errors frequently occur.

 Approximately 20% to 30% of medical patients experience an adverse event after discharge, resulting in a variety of suboptimal outcomes.

 The majority of these adverse events may be preventable, but they frequently occur early with respect to discharge and prior to patients being seen in the outpatient setting.

 Thus, managed care organizations are seeking feasible strategies to bridge this important safety gap.

Although telephone calls to patients following hospital discharge have garnered interest as a way to assist them with a variety of issues related to their hospitalization, the results of studies examining the effectiveness of such calls have been mixed.

 Very few of these studies have described calls made by the physicians who treated the patients during their hospitalization

and, to our knowledge, none have described such calls to adult patients by treating hospitalists. We theorized that hospitalists may be well-positioned to recognize and assist their patients with problems and concerns arising shortly after discharge. Accordingly, we studied whether the same hospitalist who cared for the patient during their hospitalization could effectively identify and efficiently address early postdischarge problems through a structured telephone call.

We conducted a prospective cohort study of patients who were discharged from a general internal medicine service by one of 19 participating hospitalists at Denver Health Medical Center, a 525 bed university-affiliated, urban public safety net hospital, between March 1, 2012, and October 31, 2013. We included English- or Spanish-speaking patients aged 18 to 89 years who were insured by Denver Health’s own managed care organization. We excluded patients who lacked telephone access, those with a physical or cognitive impairment precluding their participation in a telephone encounter, those who were incarcerated, known to be pregnant, and those discharged to a skilled nursing facility or hospice. The study was reviewed and approved by the Colorado Multiple Institutional Review Board with a waiver of consent.

 Participating hospitalists assessed patients for eligibility and verified their telephone number prior to discharge on selected study days. Study days were selected such that the hospitalist knew they would be available to make the telephone calls 48 to 72 hours after discharge. Beginning at 48 hours post discharge, the hospitalists attempted to contact eligible patients by telephone. They were instructed to make 2 attempts to reach each patient, beginning at approximately 48 hours after discharge and making a final attempt at approximately 72 hours. Hospitalists either spoke with the patient directly or with a proxy selected by the patient. Telephone interpreters were utilized for Spanish-speaking patients.

A structured assessment based on key components of Project Reengineered Discharge

was utilized by the hospitalist during the postdischarge calls (eAppendix, available at 

). Questions were organized within specific domains and prompts were used to encourage providers to perform “teach-back” at key intervals.

 Hospitalists also had access to the patients’ electronic health record (EHR), including information pertaining to their hospitalization, discharge instructions, and medication list. If the hospitalist determined that additional services were needed beyond what they could provide, then an electronic referral was sent to the Department of Managed Care at Denver Health Medical Center for resolution.

Within 2 weeks of the hospitalist call, patient navigators employed by the Department of Managed Care attempted to contact each patient who had received a hospitalist postdischarge call in order to evaluate their perceptions regarding the utility of the call and to determine whether the problem addressed in the initial call had been adequately resolved.

We collected demographic and clinical data from our EHR and assessed the following variables: a) the category and frequency of problems identified in the postdischarge call; b) the severity of the most significant problem, as perceived by the hospitalist, using a 100-point visual analog scale (with 1 representing the least serious problem and 100 representing the most serious); c) the proportion of calls in which a problem was identified that were independently addressed by the hospitalist; d) the proportion of calls for which additional managed care services were requested; e) the proportion of calls resulting in a recommendation to seek urgent/emergent care; f) the time spent by the physician on the call and any subsequent care coordination; g) each hospitalist’s a priori assessment of the likelihood that each patient would have postdischarge problems identified in the call (assessed at hospital discharge using the previously mentioned visual analog scale); h) the types of managed care interventions requested; i) patients’ perceptions regarding the usefulness of the call; and j) the navigators’ determination of whether the problem(s) identified by the hospitalist had been adequately resolved.

REDCap, a secure Web-based application, was used to collect and manage all study data

; analyses were performed using SAS Enterprise Guide version 5.1 (SAS Institute, Inc, Cary, North Carolina). A Student’s 

 test or Wilcoxon Rank Sum test was used to compare continuous variables depending on the results of normality tests. Chi-square tests were used to compare categorical variables; 

 <.05 was considered to be statistically significant. We constructed a receiver operating characteristic (ROC) curve to evaluate the accuracy of the physician assessments in predicting postdischarge problems.

We assessed 200 hospitalized patients for eligibility, and 27 were excluded (Figure). Of the 173 eligible patients, 131 (76%) received a call from their treating hospitalist and agreed to participate in the survey. Assessments were completed on all 131 patients who received a call. We found no significant differences between any of the variables listed in the Table for patients who were successfully contacted versus those who were not.

A total of 107 problems were identified in 74 of the 131 patients (56%). A single problem was identified in two-thirds of patients while the remaining one-third had multiple problems identified. Forty-eight of these 74 patients (65%) initially denied having had any problems following discharge, but problems were subsequently discovered as a result of further questioning by the treating hospitalist using the structured survey.

Forty-one percent (44 of 107) of the problems identified were: new or worsening symptoms. The next most common categories were: difficulties obtaining follow-up appointments (23 of 107; 21%), medication issues (21 of 107; 20%), and problems understanding or executing discharge instructions (9 of 107; 8%). The remainder of the problems were related to home care services (5 of 107; 5%) and the capacity for self-care (5 of 107; 5%). Physician assessment regarding the severity of what they considered to be the most significant problem identified during the call was a median of 49 points (95% CI, 25-69) on a 100-point visual analog scale.

We found no statistically significant differences for any of the demographic or clinical variables assessed for patients in whom a problem was identified versus those in whom no problem was identified (Table). The hospitalist assessment at hospital discharge, regarding the likelihood of a patient subsequently experiencing problems, was a mean score of 67 of 100 (95% CI, 50-76) when a problem was reported versus a mean score of 49 of 100 (95% CI, 28-67) when a problem was not reported (both 

 = .002). However, the ROC curve depicting the accuracy of hospitalist assessment in predicting postdischarge problems had an area under the curve (ie, C statistic) of only 0.66.

Hospitalists were able to independently address the problems identified in the majority of calls (50 of 74; 68%), while the problems identified in 24 of 74 calls (32%) required referral to the Managed Care Department. The most common managed care services requested were: a nurse follow-up call (15%), transportation assistance (9%), and health coaching (8%). Eight patients—representing 11% of those in whom problems were identified&mdash;were referred to the emergency department (ED) or the urgent care clinic for evaluation as a result of the problem identified in the call.

Hospitalists spent a median of 8 minutes per call (interquartile range [IQR], 5-12) and a median of 10 minutes per case, including any subsequent care coordination (IQR, 6-16). Median time spent per call and per case when a problem was identified was 10.5 minutes (IQR, 7-16) and 13 minutes (IQR, 9-22) versus 6 minutes (IQR, 4-8) and 7 minutes (IQR, 5-10) when no problems were identified, respectively (both 

Eighty-four patients (64% of the patients who received a hospitalist call) were contacted by a patient navigator within 2 weeks and completed an additional survey. Of these, 77 (92%) believed the hospitalist call was “helpful” and 47 (56%) had problems discovered in the original call. Of the 47, 39 (83%) reported that they had “received the help necessary to resolve the problem(s) discussed” in the initial call. Navigators believed that the problem had been adequately resolved in 43 of 47 cases (91%).

Patients who received a hospitalist postdischarge call were less likely to be rehospitalized for inpatient or observation stays within 30 days (19 of 131; 15%) than those who were not able to be contacted (12 of 42; 29%) (

 <.05). Patients who had problems identified in the call were more likely to have an ED or urgent care visit within 30 days of discharge (15 of 74; 20%) than those who did not (4 of 57; 7%) (

Telephone contact with patients after discharge has been proposed as a way to address problems occurring in this period, but a large systematic review examining studies of postdischarge calls by a variety of healthcare providers found there was insufficient evidence to conclude that this intervention was effective.

 The results of several more recent studies have also been mixed.

 Most of these studies utilized nurses and pharmacists as callers. We know of only 1 other study describing postdischarge calls by treating hospitalists and it was performed in a pediatric population.

Our study of postdischarge telephone calls by treating hospitalists revealed several important findings. First, hospitalists discovered problems in over half of the patients they contacted between 48 and 72 hours after discharge, and new or worsening symptoms accounted for the largest proportion of these problems. Second, hospitalists could not reliably predict which patients were likely to encounter problems. Third, the majority of problems identified were addressed by the treating hospitalist through the single, brief telephone encounter.

Many of the problems we identified were not trivial, and 11% of the patients in whom problems were discovered were referred to the ED or urgent care clinic as a result. The spectrum of problems we discovered is similar to previous studies; however, we identified a larger burden of new or worsening symptoms than in some previous studies describing calls by primary care nurses and unlicensed call center representatives surveying medical patients within a similar time frame.

 Although the reason for this difference unclear, it is possible that patients may be more comfortable discussing symptoms with their treating physicians or, conversely, that physicians are better at eliciting such symptoms compared with other types of providers, either because they are physicians and/or because they treated the patients during their hospitalization.

We had theorized that treating hospitalists might be able to accurately predict which patients were likely to experience problems following discharge, but our data did not confirm this supposition. Our results indicate that hospitalists’ predictions are marginally better than chance. However, we found that treating hospitalists were able to independently address the majority of problems they identified during the call without requiring additional managed care resources. Moreover, they were able to do so relatively efficiently (eg, median = 10 minutes per case), perhaps because the treating hospitalist was familiar with the patient and had firsthand knowledge of the discharge plan.

The frequency with which postdischarge problems occur, together with the fact that they are difficult for physicians to predict and often initially unrecognized by patients, suggests that early telephone contact with a healthcare provider after hospital discharge should be considered as a routine practice. Given the relatively high burden of new or worsening symptoms signaling possible deterioration in a patient’s condition, we would also suggest that the provider making the calls should possess the clinical acumen necessary to quickly and effectively evaluate such complaints. Although some of the problems we discovered could likely have been dealt with by allied healthcare providers, many were seemingly addressed more directly and efficiently because of the treating hospitalists’ implicit knowledge of the patient gained during their hospitalization.

Our study has several limitations. First, it was performed in a single university-affiliated public safety net hospital, such that the results might not be generalizable to other types of institutions and/or other patient populations. Second, because we studied patients who were insured through a managed care program, our patients may have had more resources available to them than would be typical of many patients cared for at a safety net hospital. Third, we did not design our study to compare the effectiveness of postdischarge calls by hospitalists with calls made by other types of healthcare providers or with no intervention, such that we cannot conclude that calls by hospitalists are superior to other strategies. Fourth, although fewer patients who received a hospitalist call were rehospitalized, this result may have been influenced by factors unrelated to the call (eg, patients may have been rehospitalized at the time of the call). Lastly, because our study is not a randomized controlled trial, the results may be confounded by unmeasured differences, such as variable health literacy or self-investment.

Our study also has a number of strengths. First, to our knowledge, it is the first study describing postdischarge telephone follow-up of adult patients by treating hospitalists. Second, we utilized a structured survey that included a “teach-back” style of interaction in order to facilitate patient learning and retention of information, and this may have aided our ability to detect and address patients’ problems. Third, we utilized a second call to evaluate patients’ perspectives regarding the usefulness of the call made by the hospitalists.

Our strategy represents a feasible method of detecting and alleviating early postdischarge problems, which, if unaddressed, may result in poor outcomes. Managed care organizations might consider partnering with hospitalists to enhance the postdischarge safety of their members. To determine whether our approach results in better outcomes than what can be achieved by calls from providers not involved in the care of the patients during their hospitalization would require a prospective randomized trial. Given the potential for improved outcomes, the results of such a trial could substantially alter the customary approach to patient care after discharge.

 Division of Hospital Medicine (SAS, MGF, DH, MB) and Department of Medicine (SAS, AK, MGF, DH, MB), Denver Health, Denver, CO; Denver Health Managed Care (DB), Denver, CO; University of Colorado School of Medicine (SAS, MGF, DH, KM, DB, RKA, MB), Aurora, CO; Metro Hospitalists, Presbyterian St. Luke’s Medical Center (KM), Denver, CO; Rocky Mountain Poison and Drug Center (JL), Denver, CO.

 This study was unfunded; however we did provide some of the study data to the Colorado Hospital Association as a part of a secondary grant-funded project.

 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. At the time of paper’s preparation, the authors’ affiliations were accurate, as shown above; however, Dr Heppe is now affiliated with the Department of Veterans Affairs Medical Center in the Eastern Colorado Health Care System.

 Concept and design (RKA, DB, MB, AK, JL, SAS); acquisition of data (MB, MGF, DH, JL, KM, SAS); analysis and interpretation of data (RKA, MB, AK, JL, SAS); drafting of the manuscript (RKA, MB, AK, JL, SAS); critical revision of the manuscript for important intellectual content (RKA, DB, MB, MGF, DH, AK, KM); statistical analysis (RKA, AK, SAS); provision of patients or study materials (DB, DH); administrative, technical, or logistic support (MB, DH, JL, KM, SAS); and supervision (RKA, MB).

 Sarah A. Stella, MD, Denver Health and Hospital Authority, 777 Bannock, MC 4000, Denver, CO 80204. E-mail: Sarah.Stella@dhha.org.

1. Moore C, Wisnivesky J, Williams S, McGinn T. Medical errors related to the discontinuity of care from an inpatient to outpatient setting. 

2. Tsilimingras D, Bates DW. Addressing postdischarge adverse events: a neglected area. 

3. Coleman EA, Chugh A, Williams MV, et al. Understanding and execution of discharge instructions. 

. 2013;28(5):383-391. doi: 10.1177/1062860612472931.

4. Forster AJ, Murff HJ, Peterson JF, Gandhi TK, Bates DW. The incidence and severity of adverse events affecting patients after discharge from the hospital. 

5. Tsilimingras D, Schnipper J, Duke A, et al. Post-discharge adverse events among urban and rural patients of an urban community hospital: a prospective cohort study.

. 2015;30(8):1164-1171. doi: 10.1007/s11606-015-3260-3.

Treating hospitalists effectively identify and efficiently address early postdischarge problems through a single, brief telephone encounter.

Postdischarge Telephone Calls by Hospitalists as a Transitional Care Strategy

Asians are one of the fastest growing racial subgroups within the United States and constitute a relatively heterogeneous population. 

Asian American women have lower mortality following hip fracture compared with white women, but differences by ethnic origin are unknown. 

This study demonstrates similarly low 1-year mortality rates following hip fracture among Chinese, Japanese, and Filipina women, who make up nearly three-fourths of the Asian hip fracture population examined. 

Mortality risk did not differ significantly by ethnic subgroup, even after adjusting for differences in age and comorbidity burden. 

These findings support further efforts to examine similarities and differences in osteoporosis-related outcomes in larger Asian American populations.

We previously published data from Kaiser Permanente Northern California demonstrating that mortality following hip fracture varied by race/ethnicity, with 1-year mortality rates following hip fracture found to be one-third lower in Asian women (15.6%) compared with white women (23.6%).

 These findings are of interest, given that Asians make up the fastest growing racial subgroup within the United States over the past 10 years.

 Furthermore, according to the 2010 US Census, California contributes 32.1% of the US Asian population.

 Asian Americans, however, form a heterogeneous population of East, South, and Southeast Asians where selected health outcomes may vary substantially,

 and differentiating these minority populations may be important for understanding potential health disparities.

Although previous studies have examined bone mineral density and fracture risk among selected Asian subgroups in California,

 the extent to which postfracture health outcomes vary between these ethnic populations has not been examined. In this report, we conducted additional analyses within our prior study

 to describe the distribution of the 3 largest Asian ethnic subgroups among northern Californian women experiencing hip fracture and their corresponding rates of early rehospitalization and 1-year mortality post hip fracture.

Among 615 Asian women who experienced hip fracture between 2000 and 2010,

 we used available ethnic subgroup data to ascertain those of Chinese, Japanese, and Filipina ethnicity and conducted additional chart record validation of ethnicity for all women initially identified as Filipina and for 9% of Chinese and Japanese women with mixed ethnic origin data or inconsistent surnames. These efforts classified 172 (28%) Chinese, 153 (24.9%) Japanese, and 119 (19.4%) Filipina women, accounting for 72.2% (444 of 615) of women with hip fracture identified as Asian.

), Chinese and Japanese women were similar in age, whereas Filipina women were slightly younger and more likely to be younger than 85 years (

 <.05) at the time of hip fracture. Filipina women also had greater comorbidity (comorbidity index ≥3; 

 compared with the other 2 subgroups. Nonsignificant differences in rehospitalization rates within 30 days of hospital discharge were noted, with a possible trend toward lower rates of early rehospitalization among Japanese women (

 = .06) versus Filipina. One-year overall mortality rates following hip fracture were similar (Chinese: 14.0%; Japanese: 15.0%; and Filipina: 14.3%), and mortality risk did not differ significantly by group when adjusted for age, comorbidity, fracture type, and history of fracture. 

In summary, the low 1-year mortality rate among Asian women following hip fracture was reflected similarly across the 3 Asian subgroups in the same healthcare setting, accounting for nearly three-fourths of the Asian hip fracture cohort. Although the number of women in each subgroup was relatively small, thus limiting the generalizability of our conclusions, this is one of the first studies to classify ethnic origin among US Asians experiencing hip fracture and their subsequent morbidity and mortality outcome. Future studies should be conducted in larger, diverse US populations of Asian race to determine the extent to which Asian ethnic subgroups vary with regard to key osteoporosis-related health outcomes.

 Department of Medicine (MCP, JCL), Kaiser Permanente Oakland Medical Center; Division of Research (MC, JCL), Kaiser Permanente Northern California, Oakland CA.

 Drs Lo and Chandra have received past research funding from Amgen and Dr Lo has received current research funding from Sanofi, unrelated to the current study. Dr Patel reports 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 (MCP, JCL), acquisition of data (MCP, MC, JCL), analysis and interpretation of data (MCP, MC, JCL); drafting of the manuscript (MCP, JCL), critical revision of the manuscript for important intellectual content (MCP, MC, JCL); and statistical analysis (MC).

 Joan C. Lo, MD, Kaiser Permanente Northern California Division of Research, 2000 Broadway, Oakland, CA 94612. E-mail: Joan.C.Lo@kp.org.

1. Lo JC, Srinivasan S, Chandra M, et al. Trends in mortality following hip fracture in older women. 

2. Hoeffel EM, Rastogi S, Kim MO, Shahid H. The Asian population: 2010. US Census Bureau website. https://www.census.gov/prod/cen2010/briefs/c2010br-11.pdf. Published March 2012. Accessed September 28, 2016.

3. Dumbauld S, McCullough JA, Sutocky JW. 

Analysis of Health Indicators for California’s Minority Populations

. Collingdale, PA: Diane Publishing Company; 1994.

4. Khandelwal S, Chandra M, Lo JC. Clinical characteristics, bone mineral density and non-vertebral osteoporotic fracture outcomes among post-menopausal U.S. South Asian women. 

2012;51(6):1025-1028. doi: 10.1016/j.bone.2012.08.118.

5. Morton DJ, Barrett-Connor E, Kritz-Silverstein D, Wingard DL, Schneider DL. Bone mineral density in postmenopausal Caucasian, Filipina, and Hispanic women. 

6. Kin K, Lee JH, Kushida K, et al. Bone density and body composition on the Pacific rim: a comparison between Japan-born and U.S.-born Japanese-American women. 

One-year mortality following hip fracture in Asian American women is comparable among Chinese, Japanese, and Filipina women, comprising nearly three-fourths of the Asian population examined.

Mortality Following Hip Fracture in Chinese, Japanese, and Filipina Women

To provide a comprehensive estimate of the total social value (TSV) delivered by granulocyte-colony stimulating factor (G-CSF) therapies in the United States in 2014.

Estimation of the TSV of G-CSF, based on a targeted literature review of pivotal studies.

A literature review was conducted to obtain estimates of the adverse outcomes associated with myelosuppressive chemotherapy-induced febrile neutropenia (FN) and the positive impacts of G-CSFs. We monetized each outcome into a set of mutually exclusive value components that were aggregated to estimate the TSV. To estimate the share of TSV captured by manufacturers, we estimated 2014 profits from G-CSF using measures of industry revenues and operating costs.

In 2014, approximately 314,440 patients received G-CSFs. Compared with what they would have experienced without G-CSFs, these patients were less likely to be hospitalized or die from FN, incur reductions in chemotherapy relative dose intensity, receive antibiotics, miss work, or experience reduced health-related quality of life. We estimated the social value from fewer FN hospitalizations to be $770 million; from fewer FN-related deaths, $2.65 billion; from fewer deaths due to higher effective chemotherapy doses, $4.83 billion; from reductions in antibiotics, $2.3 million; from reductions in indirect costs, $230 million; and from improvements in health-related quality of life, $1.9 million. The estimated 2014 US TSV of G-CSFs was $8.5 billion. Industry profits associated with G-CSFs were estimated at $1.3 billion, accounting for approximately 15% of the TSV.

Based on our calculations, the TSV generated by G-CSFs in the United States in 2014 was substantial, with the majority of this value accruing to patients.

	Granulocyte-colony stimulating factors (G-CSFs) can significantly reduce the risk of febrile neutropenia (FN) among certain patients receiving chemotherapy. FN is associated with significant clinical and nonclinical complications. This study provides a comprehensive analysis of the total social value of G-CSFs, accounting for the nonclinical, as well as clinical benefits captured by these therapies. 

		In 2014, approximately 314,440 patients received G-CSFs, and compared with individuals who did not, they were less likely to be hospitalized or die from FN, incur chemotherapy dose reductions, receive antibiotics, miss work, or experience reduced health-related quality of life.

		The estimated 2014 US total social value of G-CSFs was $8.5 billion.

Despite major advances in cancer treatment, neutropenia is a life-threatening complication of chemotherapy.

 Severe neutropenia is a below-normal count of neutrophils in the blood (less than 500 per mm), which impairs the body’s ability to combat opportunistic infections.

 Neutropenia can progress to febrile neutropenia (FN) when accompanied by a temperature of 38ËšC (100.4°F) or higher, and can lead to hospitalizations, mortality, and chemotherapy dose reductions or delays, all of which adversely affect patient outcomes

 The risk of FN is determined by the chemotherapy regimen (CR) and patient factors, such as age and comorbidities.

For patients with solid tumors or nonmyeloid malignancies, granulocyte-colony stimulating factors (G-CSFs) can enable them to undergo and remain on myelosuppressive chemotherapy with lower risk of FN and infection. Despite well-established clinical benefits of G-CSFs,

 these therapies have been the subject of recent concern due to large variation in their utilization. Research has indicated that G-CSFs may be overprescribed for patients at a low risk of FN (≤20%) and underprescribed for high-risk (≥20%) patients.

Several studies have evaluated the cost-effectiveness of G-CSFs in particular tumor types, from a payer’s perspective

; however, these studies may understate the value of G-CSFs from a societal perspective. For example, reductions in lost workdays or disability provide tremendous value to society, but may be excluded from existing studies.

 Therefore, a better understanding of the total social value (TSV) of G-CSFs is needed to appropriately determine the cost-effectiveness and value afforded to patients and others affected by chemotherapy-induced neutropenia. In this study, we estimated the TSV of G-CSFs delivered to patients with cancer in the United States in 2014 by combining various components of value derived from the literature.

To identify evidence on the value of G-CSFs, we conducted a targeted literature review of pivotal studies. Specifically, we limited the search to articles published between 1991 (the year G-CSFs were approved in the United States) and 2014 that examined both adverse outcomes associated with myelosuppressive chemotherapy-induced FN, as well as the impacts of G-CSFs on these outcomes.

The searches yielded 77 potentially relevant articles that were narrowed to 58 for data extraction (Table 1 summarizes the articles by type of analysis). Articles were selected if they investigated patients with nonmyeloid cancers at risk of the clinical definition of chemotherapy-induced FN who were treated with G-CSFs. We focused on studies that compared outcomes for patients who received G-CSFs with patients who did not.

Our literature review provided information on 6 clinical outcomes: FN hospitalization, neutropenia events, length of hospital stay, reductions in relative dose intensity of chemotherapy, overall mortality, and antibiotic use. To associate these outcomes with savings or social value (SV), we mapped them to 4 clinical value components: savings from reductions in FN hospitalizations, reductions in FN mortality, reductions in mortality due to the ability to deliver higher-dose chemotherapy, and reductions in antibiotic use. Then, using cost estimates from the literature, along with clinical improvements, we estimated the SV for all 4 components and, in aggregate, the TSV of G-CSFs.

Using the limited literature on nonclinical outcomes associated with G-CSFs, we collated information on productivity loss and quality of life.

 These outcomes were mapped into similar SV components: savings from reduced indirect costs (productivity loss) and improved quality of life. Again, we calculated the value for each component and in total for the nonclinical SV of G-CSFs.

The TSV of G-CSFs was calculated as the sum of all clinical and nonclinical value components. All reported estimates were adjusted to 2014 US dollars using the Medical Care Consumer Price Index, which is a measure of changes in the price level of medical care commodities and services.

 To maintain the integrity of the outputs derived from the literature and to facilitate replication of our study methods, we kept all decimal places of the estimates, recognizing that the level of precision implied by our estimates is less than should be expected of our modeling exercise.

In both the United States and the European Union, G-CSFs are indicated for patients with nonmyeloid malignancies receiving myelosuppressive chemotherapy associated with a clinically significant (≥20%) risk of FN. They also may be considered for patients with a 10% to 20% risk and certain individual factors, such as age or stage of cancer.

 We derived an estimate of this population using the Surveillance, Epidemiology, and End Results (SEER) program, which estimated the overall cancer incidence in 2014 at 1,665,540 cases.

 To account for patients with myeloid malignancies who fall outside the approved indication, we used an incidence rate of 0.015% (15 individuals with myeloid malignancies per 100,000 population), as reported by the 2014 World Trade Center Health Program.

 Subtracting the myeloid malignancies from the SEER all-cancer incidence suggests that, in 2014, there were approximately 1,617,665 individuals diagnosed with nonmyeloid malignancies in the United States. Given that not all of these individuals would receive a CR associated with a significant risk of FN, we applied a G-CSF usage rate of 19.4%, as reported in Naeim et al (2013), to estimate a 2014 population of patients eligible for GSCFs of 314,442.

 The occurrence of FN often results in immediate hospitalization to treat the associated infection.

 Crawford et al (2008) used medical claims data to estimate that 10.7% of patients with breast, lung, colorectal, and ovarian cancers, as well as lymphoma, develop FN in the first 3 cycles of chemotherapy.

 We found that patients with nonmyeloid malignancies receive 6 cycles of chemotherapy on average. Thus, we used the growth rate in incidence of FN from cycles 1 to 2 and cycles 2 to 3, reported in Crawford et al, to predict the growth rate from cycles 3 to 4, 4 to 5, and 5 to 6 in order to estimate a 6-cycle incidence rate (I

Further, many trials have demonstrated that, when used prophylactically, G-CSFs significantly reduce the incidence of FN. Cooper et al (2011) and Kuderer et al (2007) both reviewed randomized controlled trials and reported relative risk ratios (RRs) for developing FN (RR

) for all solid tumor patients receiving primary G-CSFs of 0.51 and 0.44, respectively.

 We applied the average of these RRs (0.48) to our calculated incidence measure to estimate that G-CSFs reduce the incidence of FN to 10%. Using cost estimates from Dulisse et al (2013), we used the following equation to estimate that G-CSFs prevented 35,988 FN hospitalizations in 2014, thus creating $768 million in SV derived from avoided FN hospitalizations

FN Hospitalizations Value = [{0.22 — (0.22 × 0.48)} × 314,442] × $21,341 = $768,000,000

 In addition to hospitalizations, FN is also associated with high mortality. Using claims data, Dulisse et al (2013) found that the inpatient mortality rate (MR) for patients with breast cancer, lung cancer, colorectal cancer, ovarian cancer, non-Hodgkin lymphoma, and Hodgkin lymphoma with FN was 10.6%.

 Similarly, Kuderer et al (2006) and Caggiano et al (2005) reviewed discharge data for all patients with cancer with FN and reported mortality (MR

 We used the average of all 3 rates (8.6%) to estimate that in the absence of G-CSFs, there would have been 5872 FN-related deaths in the United States in 2014.

G-CSFs reduce the chances of developing FN and the probability of death among patients who develop FN. In their systematic review of the G-CSF literature, Kuderer et al (2007) estimated a 0.55 RR of FN-related mortality (RR

 Applying this estimate to the reduced FN population derived above, we calculated that G-CSFs prevented approximately 4171 deaths in 2014.

To calculate the SV generated from the reduction in FNM, we estimated the average value of each life saved. Specifically, we used estimates of average life expectancy from diagnosis for several types of cancers.

 Weighting these values using patient populations provided in Dulisse et al (2013) and Caggiano et al (2005), we estimated that, on average, a patient who dies because of FN has lost 8.81 years of life (LYL).

 Valuing each year of life (VLY) in perfect health at $100,000

 and adjusting it to account for the average long-term quality-adjusted life-years (QALYs) weight of 0.72 for patients with cancer,

 we used the following equation to estimate a total annual value of reduced FN deaths of $2.65 billion:

 M)]} × LYL FNM] × VLY = $2,645,923,449

FNM Value = [{([0.22 — (0.22 × 0.55)] × 314,442) × [0.086 × (1 – 0.55)]} × 8.81] × 100,000 = $2,645,923,449

Mortality Due to Reduced Chemotherapy Relative Dose Intensity

Many patients receiving chemotherapy do not achieve their planned relative dose intensity (RDI) because of treatment toxicity complications. Lower RDI has been associated with adverse clinical outcomes, including reduced life expectancy.

 We identified several studies that estimated an incidence of chemotherapy dose reductions (I

) greater than 15% among some cancer types, including an estimate of 40% based on a systematic literature review by Kuderer et al (2007) in a robust cancer population.

Several studies have also found that G-CSFs significantly reduce a patient’s risk of a chemotherapy dose reduction. Although our literature review identified 3 separate estimates of the odds ratio of a dose reduction (OR

) from 2 different systematic literature reviews, these were all based on very select populations.

 Assuming that patients in our population would experience similar reductions in risk, we calculated the average of all 3 reported  estimates (0.63). This estimate suggests that G-CSFs reduced the number of patients requiring chemotherapy dose reductions (I

) by approximately 45,979 patients in 2014, or nearly 25% of our total G-CSF population.

To value survival gains associated with avoided chemotherapy dose reductions, we used estimates from Havrilesky et al (2014) who found that an RDI greater than 85% was associated with increased survival for patients with breast and lung cancers.

 Using population weights from Naeim et al (2013), we calculated an average survival gain (S

) of 17.5 months per G-CSF patient, attributable to chemotherapy dose reductions avoided.

 We then used an estimate of the average cost of CRs ($24,392) calculated from data provided in Naeim et al to account for higher treatment costs associated with this increased dosing (assuming a 15% increase).

 Clinical experts suggested these increased costs should only apply to the roughly 50% of patients receiving chemotherapy from multi-use vials. Applying the value of a cancer-adjusted life-year at $72,000, we used the following equation to estimate the value of lowering the incidence of chemotherapy dose reductions (dose-related mortality [DRM value]) to be $4.8 billion, annually:

DRM Value = {([0.4 — (0.4 × 0.63)] × 314,442) × 1.46} × 100,000 – [($24,392 × 0.15) × [0.4 – (0.4 ×  0.63)] × 314,442] = $4,827,749,599

Granulocyte-colony stimulating factors (G-CSFs) reduce the risk of febrile neutropenia in patients with cancer. This study evaluates the clinical and nonclinical value associated with G-CSFs.

October 2016

A Call for a Statewide Medication Reconciliation Program

Postdischarge Telephone Calls by Hospitalists as a Transitional Care Strategy

Mortality Following Hip Fracture in Chinese, Japanese, and Filipina Women

Estimating the Social Value of G-CSF Therapies in the United States

Periodic Health Examinations and Missed Opportunities Among Patients Likely Needing Mental Health Care

Does Medicare Managed Care Reduce Racial/Ethnic Disparities in Diabetes Preventive Care and Healthcare Expenditures?

Patients' Success in Negotiating Out-of-Network Bills

Connected Care: Improving Outcomes for Adults With Serious Mental Illness

Economic Value of Pharmacist-Led Medication Reconciliation for Reducing Medication Errors After Hospital Discharge

Benchmarking Health-Related Quality-of-Life Data From a Clinical Setting

Cost-Effectiveness of a Statewide Falls Prevention Program in Pennsylvania: Healthy Steps for Older Adults