Julia C. Prentice, PhD

 There is little research on the relationship between call center performance and patient-centered outcomes. In this study, we quantified the relationships between 2 measures of telephone access, average speed of answer (ASA) and abandonment rate (AR), and patient satisfaction outcomes within the Veterans Health Administration (VHA).

 We analyzed 2015 and 2016 data from the Survey of Healthcare Experiences of Patients and linked them with administrative data to gather features of the patient visit and monthly measures of telephone access for each medical center.

 We used mixed effects logistic regression models to estimate the effects of ASA and AR on a variety of access and satisfaction outcomes. Models were adjusted for patient-level demographics, time-varying facility-level characteristics, features of the patient visit, and facility-level random effects to control for care quality and case mix differences.

 The VHA made substantial strides in both access measures between 2015 and 2016. We found that a center’s ASA was inversely associated with patients’ perceptions of their ability both to access urgent care appointments and to do so in a timely manner. In contrast, telephone AR was not associated with any of the patient satisfaction outcomes.

 Our results associate decreased telephone waits with improved perceptions of urgent care access even without concomitant decreases in observed appointment waits. These findings may have important implications for regulators as well as for healthcare organizations that must decide resource levels for call centers, including hospitals, federal health insurance exchanges, and insurers.

Using survey data from 2015 and 2016, we conclude that telephone access has important consequences for patient satisfaction. We show that:

Longer telephone wait times were associated with decreases in patients’ perceived ability to access urgent care appointments and to do so in a timely manner.

There was no clear association between call abandonment rates and patients’ perceptions of healthcare access or satisfaction with their care.

If hospitals and providers are to become more patient centered, attention is needed to how patients are served when calling for appointments and medical questions.

For many patients, picking up the telephone is the first step in their engagement with the healthcare system. Hospitals and payers each have independent call centers to assist patients with medical and administrative questions, appointments, billing, and more. A 2015 survey of healthcare call center leadership found that many of these call centers are long established, and an overwhelming majority (93%) are managed in house. Most respondents predicted that service levels, staffing, and the importance of telephone-based services will grow in coming years.

 Additionally, the Affordable Care Act imposed requirements for state and federal health insurance exchanges to operate telephone hotlines for citizens, under guidelines promulgated by HHS.

Two of the largest efforts to collect data on call center performance, at least for public programs, are operated by CMS and the Veterans Health Administration (VHA). CMS has the authority to monitor call centers for Medicare Advantage organizations, Prescription Drug Plan sponsors, and Medicare/Medicaid insurers under 42 CFR 432.128(d)(1). CMS employs secret shoppers to collect a variety of call center performance metrics and conduct quarterly “timeliness studies” to evaluate call center performance. A call center is assigned a passing grade if the average hold time is less than 2 minutes and if fewer than 5% of calls are disconnected. Additionally, summary reports of other performance metrics are provided back to the call center but are not used for assessment purposes.

VHA, the largest integrated healthcare system in the United States, also provides timely telephone services, including 24/7 telephone access to clinical staff trained to provide healthcare advice and information, to all veterans receiving care at its facilities. Facility-level telephone data and a variety of care quality measures are used to summarize and improve medical center performance. To accelerate improvement, VHA established a nationwide initiative to improve telephone access in 2009-2010. The initiative included components such as the installation of automatic call distribution systems, improved training and monitoring of call center teams, and the creation of multidisciplinary teams at participating medical centers to test and implement quality improvement strategies.

 Timely telephone services are currently assessed by 2 measures: average speed of answer (ASA) and abandonment rate (AR). The VHA’s goal for each facility is to have an ASA of less than or equal to 30 seconds and an AR of less than or equal to 5%.

There is little research on the relationship between specific measures of call center performance and patient-centered outcomes. Within the healthcare literature, studies have found that hold times, staff courtesy, whether staff provided requested medical information or help,

 were all related to overall patient satisfaction with care. Other measures cited as important in the nonhealthcare literature include first-call resolution, AR, ASA, total call volume, and average talk time, among others.

 These earlier efforts had several limitations. For instance, they generally involved relatively small sample sizes and a single outcome measure (overall patient satisfaction with care).

In this study, we quantified the relationship among multiple measures of telephone access and satisfaction within the VHA. Using Survey of Healthcare Experiences of Patients (SHEP) data on primary care visits to the VHA during federal fiscal years (FYs) 2015 and 2016, we examine whether improvements in clinic-level telephone access measures led to concomitant improvements in a wide range of patient-reported outcomes. To our knowledge, no previous work has taken advantage of this large, national data set to examine these relationships. It is important for hospitals, payers, and other healthcare organizations such as CMS to understand whether telephone access is meaningfully associated with patient perceptions of care quality and access. Without such evidence, insufficient resources may be directed to call centers, and interventions to improve call center performance may not include the metrics that matter most for patient satisfaction.

These analyses used VHA telephone access administrative measures to predict self-reported patient satisfaction with care. Data for this study came from the FYs 2015 and 2016 SHEP outpatient cohorts, which were the most recent years of data available during the study period. We did not use data from before 2015 due to changes in the SHEP sampling methodology that could confound the analysis. SHEP is a nationwide mail survey distributed to veterans after a visit to a VHA facility that seeks to obtain veterans’ perceptions of their care. For outpatient care, a simple random sample of patients with completed appointments at every VHA facility nationwide is selected each month. Thus, SHEP may be considered a repeated cross-sectional survey. The overall response rates in 2015 and 2016 were 43% and 41%, respectively.

 Survey responses were linked back to VHA administrative data to gather features of the patient visit. Respondents were included in the sample if the patient completed at least part of the SHEP survey; they were not missing demographic information, such as age, race/ethnicity, or gender; and their appointment was for primary care. For the 6.2% of patients with more than 1 survey response, only the first response was included in the sample.

Telephone data that were missing responses or had a value of 0 recorded for either AR or ASA were removed from the data set, as were unrealistic outlier data including errors in data entry in the top or bottom 2.5% for either AR or ASA.

 This cutoff threshold was later tested and performed well in various sensitivity analyses. The final sample included 252,145 unique patients across 285 medical facilities.

VHA collects 2 types of telephone access data for all incoming calls. Response time is measured by the average length of time elapsed before a caller reaches a staff member:

VHA medical centers also collect data on call abandonment, regardless of call length. Telephone AR is defined as the percentage of calls coming into a telephone system that are terminated by the person originating the call before being answered by a staff person. AR is measured as:

Both telephone metrics are tracked locally by individual VHA medical centers, and monthly averages are then entered by staff members into a national database. A proportion of centers that have been granted waivers are not required to make monthly submissions. Additionally, some smaller medical centers do not have their own call centers; these are referred to as “covered” facilities. Data from covered facilities are monitored and collected by a larger nearby facility, and in these cases only a combined set of telephone metrics is reported. For example, a VHA medical center may handle the telephone data collection for its affiliated community-based outpatient clinics. Approximately 11% of SHEP responses are from visits to covered facilities. In our regression models, we included a dummy variable taking on a value of 1 if the facility was a covered facility and 0 otherwise. Due to the human involvement in this process, we visually inspected histograms of monthly telephone ARs and wait times for manipulation of the access measures near the VHA’s performance thresholds.

 contains a listing of the specific outcome measures used in this analysis along with their respective answer formats. Satisfaction measures were selected and calculated following previous work in this area.

 Respondents were asked questions regarding their ability to obtain appointments for urgent care and routine care as soon as they needed and their ability to get medical questions answered within the same day. Responses for these measures included always, usually, sometimes, and never; we dichotomized responses into always/usually compared with sometimes/never. Additionally, SHEP included questions regarding how long patients had to wait for urgent care appointments, with responses on a 5-point scale from “same day” to “more than 7 days.” We dichotomized responses into wait times of 1 day or less versus more than 1 day. Lastly, satisfaction with their healthcare provider was measured by asking respondents to provide a rating on a scale of 0 to 10. We dichotomized responses into ratings of 9 or 10 compared with less than 9.

Control Variables and Model Specification

For all models, the unit of analysis was the individual SHEP response. Self-reported physical and mental health were measured on separate 5-point scales ranging from poor to excellent. To control for survey context effects, 2 binary variables were included indicating whether the respondent received help completing the survey and whether the survey was administered in English. Additional facility-level characteristics, which are time-varying and may affect satisfaction, were included, such as SHEP nonresponse rate, average wait time for a primary care appointment, average number of days between patient visit and survey return date, covered facility status, and volume of primary care visits and phone calls during the preceding month. Models also included an overall time trend to control for secular changes in wait times and a VHA medical center random effect to control for facility quality and case mix differences.

Our key predictor variables, ASA and AR, were converted into quartiles using cutoffs based on telephone performance data reported by all VHA centers in October 2014, the first month of our data. The cutoff points for these quartiles (in seconds for ASA and percentages for AR) are contained in 

We estimated separate adjusted mixed effects logistic regression models for each patient satisfaction outcome. As a sensitivity analysis, we repeated this process using fixed instead of random effects for each medical center. Analyses were performed using R version 3.3.2 (R Foundation for Statistical Computing; Vienna, Austria).

Characteristics of our study sample are described in 

. The SHEP respondents in our sample were representative of the larger VHA population. Respondents were generally older, overwhelmingly male, and mostly white. Patients were generally highly satisfied with their providers; approximately 2 of 3 rated their provider a 9 or 10 of 10 possible points. Nearly 4 of 5 patients stated they were usually or always able to receive appointments for urgent care as soon as they needed. For routine care, this proportion fell to 3 of 5. Approximately 37% of respondents who sought an urgent care appointment were able to be seen within 1 day or less, whereas only 28% were able to receive answers to medical questions within the same day.

VHA showed improvements in telephone access measures between October 2014 and September 2016 (FY 2015 and FY 2016) (

). ASA declined slightly, from an average of 87 seconds in October 2014 to 69 seconds in September 2016. Average AR fell from 12.0% to 8.3% during the same time period. There was significant variation in telephone access measures by VHA center. It is possible that the manual transposition of telephone performance data from VHA centers to the national database could introduce human error into the data-generating process. Additionally, gaming of metrics is a concern when there are institutional incentives to meet or exceed specified performance targets.

 We conducted graphical inspection via histograms of the telephone access data to check for strategic behavior due to the human role in their collection and transmission. We found no evidence of “bunching” or rounding down of ASA or AR to the VHA performance thresholds of 30 seconds or 5%, respectively. The distributions of telephone access metrics are multimodal, with small peaks occurring at many whole numbers for ASA and tenths of a percentage for AR. This suggests that individual centers sometimes rounded their access metrics before transferring them to the national database, despite greater possible precision.

Our findings reveal negative and significant associations between ASA and 2 of the 5 outcome measures (

), and these relationships exhibited a generally decreasing gradient. For instance, patients who made appointments for urgent care were less likely to respond that they could usually or always get appointments as soon as they needed if they visited a VHA center with ASA in the third quartile (odds ratio [OR], 0.93; 95% CI, 0.86-1.02) or fourth quartile (OR, 0.85; 95% CI, 0.76-0.95) compared with a VHA center in the first quartile. Patients were also less likely to report being able to get an urgent care appointment within 1 day if they visited centers in the second quartile (OR, 0.93; 95% CI, 0.88-0.99) or fourth quartile (OR, 0.84; 95% CI, 0.77-0.92) for ASA. The effects of longer telephone wait times on provider ratings were in the expected direction and showed a similar dose—response relationship but were not statistically significant. There was no clear relationship between ASA and the ability to get medical questions answered or to receive routine care appointments.

The regression results did not demonstrate a clear, significant relationship between AR and any of the 5 outcomes. Additional data on CIs and 

 values for both our main predictors of interest and covariates are included in the 

). Sensitivity analysis results using medical center fixed effects were not significantly different from the random effects models, as determined by Hausman tests, supporting the decision to prefer the random effects specifications due to their increased efficiency in the absence of bias.

In this study, we validated the relationship between the average speed of answering telephone calls and patient satisfaction with urgent care access. As our results show, facilities with lower performance in answering phone calls also have lower patient perception of their ability both to access urgent care appointments and to do so in a timely manner.

In contrast, there was no clear relationship between AR and the study outcomes. One potential issue is that the VHA includes all callers in its calculation of AR, although callers who hang up shortly after dialing have most likely misdialed and are not likely to have hung up due to exasperation with their wait time. This “noise” in the data may attenuate the effect of AR on satisfaction and access outcomes. Best practices in the industry suggest that calls shorter than 5 seconds should not be included.

VHA made substantial strides in reducing telephone wait times and abandoned calls between 2015 and 2016, with average wait times and ARs decreasing by approximately 7% and 22%, respectively. Despite these improvements, nearly 80% of centers still did not meet 1 or both of the VHA’s stated performance thresholds by the end of the study period. Based on our findings, further improvements in telephone access could be expected to lead to improvements in patients’ self-reported access and satisfaction.

These results demonstrate that decreased telephone waits are associated with improved patient perceptions of urgent care access even without concomitant decreases in observed appointment waits. The strength of this association may occur because telephone call centers are often the first point of contact for patients, and this contact immediately precedes the patient visit. However, we did not find evidence that either AR or ASA were associated with patient ratings of routine care access, of their ability to get medical questions answered the same day, or of their providers. These limited findings may have implications for insurers, CMS, and other healthcare organizations that must make resourcing decisions for call centers for hospitals and federal health insurance exchanges. Specifically, if these organizations are to become more patient centered and improve patient satisfaction, managing the promptness of service for patients calling for appointments and medical questions is likely to be an important component of a successful strategy to improve particular satisfaction measures.

This study has several limitations. Due to the observational nature of the analysis, the significant relationships between telephone access measures and patient satisfaction may be due to the presence of omitted demographic or time-varying facility-level variables and thus not causal. However, these results comport with prior research demonstrating associations between telephone access measures and caller satisfaction.

 Further, the relationships we found with ASA were similar to those that were found in prior research validating access metrics with health outcomes and self-reported patient satisfaction.

 found similar negative relationships between satisfaction and longer average appointment wait times for new patients and consult wait times for returning patients.

The SHEP sample consists predominantly of middle-aged and elderly white men, so the results may not generalize to other demographic groups. Further, patients with multiple visits in a short period may not always remember which appointment the survey was referencing.

The VHA does not collect data on other telephone access measures, such as the number of transfers, first-call resolution, and average talk time, that have been cited as important in the literature. Due to widespread availability of technology to track these additional measures, we recommend that the VHA and other organizations collect these data so that future research may study their effects on patient satisfaction as well.

Our study is the first to look at the relationships between 2 measures of telephone access, wait times and ARs, and patients’ satisfaction with their care experience. Within the VHA system, ASA was associated with lower patient ratings of their ability to access urgent care appointments and to do so in a timely manner. We observed these effects even after controlling for a variety of demographic characteristics, institutional factors, and average medical center wait times. However, ASA was not associated with patient ratings of their ability to access routine care or their providers. AR was associated with neither patients’ perceptions of their ability to access care nor satisfaction with their care. Our results suggest that hospitals and providers could achieve modest improvements in patient satisfaction by reducing the time patients spend on hold.

 Partnered Evidence-Based Policy Resource Center, Veterans Health Administration (KNG, SDP), Boston, MA; Department of Health Law, Policy & Management, Boston University School of Public Health (KNG, SDP), Boston, MA; Center for Healthcare Organization & Implementation Research, Veterans Health Administration (DL, JCP), Boston, MA; Office of Veterans Access to Care, Veterans Health Administration (MLD), Washington, DC; Department of Psychiatry, Boston University School of Medicine (JCP), Boston, MA.

 Funding for this research was provided by the Department of Veterans Affairs, Veterans Health Administration Office of Veterans Access to Care. The views expressed in this article are those of the authors and do not necessarily represent the position or policy of the Department of Veterans Affairs, Boston University, or Northeastern University.

 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 (KNG, MLD, SDP, JCP); acquisition of data (KNG, DL, JCP); analysis and interpretation of data (KNG, DL, MLD, SDP, JCP); drafting of the manuscript (KNG); critical revision of the manuscript for important intellectual content (KNG, MLD, SDP, JCP); statistical analysis (KNG, DL, SDP); provision of patients or study materials (MLD); obtaining funding (MLD, SDP, JCP); administrative, technical, or logistic support (MLD, SDP, JCP); and supervision (SDP, JCP).

Articles

Greater telephone wait times, but not abandonment rates, were associated with lower patient perceptions of their ability to obtain urgent care in a timely manner.

Call Center Performance Affects Patient Perceptions of Access and Satisfaction

The Medicare accountable care organization (ACO) program financially rewards ACOs for providing high-quality healthcare, and also factors in the patient experience of care. This study examined whether administrative measures of wait times for specialist consults are associated with self-reported patient satisfaction. 

 Analyses used administrative and survey data from a clinically integrated healthcare system similar to an ACO. 

 Veterans Health Administration (VHA) data from 2012 was obtained. Administrative access metrics included the number of days between the creation of the consult request and: 1) first action taken on the consult, 2) scheduling of the consult, and 3) completion of the consult. The Survey of Healthcare Experiences of Patients—which is modeled after the Consumer Assessment of Healthcare Providers and Systems family of survey instruments used by ACOs to measure patient experience&mdash;provided the outcome measures. Outcomes included general VHA satisfaction measures and satisfaction with timeliness of care, including wait times for specialists and treatments. Logistic regression models predicted the likelihood of patients reporting being satisfied on each outcome. Models were risk adjusted for demographics, self-reported health, and healthcare use. 

 Longer waits for the scheduling of consults and completed consults were found to be significantly associated with decreased patient satisfaction.  

Because patients often report high levels of powerlessness and uncertainty while waiting for consultation, these wait times are an important patient-centered access metric for ACOs to consider. ACOs should have systems and tools in place to streamline the specialist consult referral process and increase care coordination.

The American Journal of Accountable Care. 2017;5(1):23-28

The Medicare accountable care organization (ACO) program financially rewards ACOs for providing high-quality healthcare. Participating providers are then financially rewarded if healthcare spending is kept below targets set by Medicare, which can be achieved by the prevention of medical errors and duplication of services.

 The success of ACOs assumes that a structure of clinical integration and appropriately targeted incentives will improve coordination of care and quality.

 A key measure of quality of care is patient experience of care, including perceived coordination. Using the Consumer Assessment of Healthcare Providers and Systems (CAHPS) family of survey instruments, ACOs are required to collect information on patient experiences, such as the ability to obtain timely care and access to specialists.

Care coordination between primary and specialty care has received little attention despite the fact that specialty care accounts for more healthcare resource use than primary care, and specialists outnumber primary care physicians in the United States.

 Primary care and specialist providers report inadequate communication between each other about referrals, which compromises their ability to provide high-quality care

 and may also have a negative impact on self-reported patient satisfaction. 

Patients report high levels of uncertainty and powerlessness during the period of time between a requested referral and subsequent action, as they wait for clarity on disease outcomes.

 Consequently, if patients experience inadequate coordination between primary and specialty care, their experiences may suffer. Previous research has found that shorter wait times for appointments are not always the most important priority for patients. Patients are willing to wait longer to get an appointment at a convenient time or to see a preferred provider, especially for low-worry, long-standing conditions. Yet, when there is a new health concern, faster access becomes a higher priority.

The Veterans Health Administration (VHA)—the largest clinically integrated system similar to an ACO in the United States that coordinates primary and specialty care&mdash;is a source of data that can provide important insights into the effect of the consult process on patient experience. In 2014, the VHA had over 9 million enrollees and provided or coordinated over 92 million outpatient encounters.

 Since 2009, the VHA has consistently measured patient experiences and satisfaction with the Survey of Healthcare Experience of Patients (SHEP) Survey that is modeled after the Consumer Assessment of Healthcare Providers and Systems (CAHPS) Survey that ACOs use to measure patient experiences.

 This paper investigates the effect of several different measures of VHA consult wait times on self-reported patient satisfaction. This effect on patient satisfaction provides a specific point of intervention for ACOs to consider when thinking of ways to measurably improve patient experiences. 

As discussed in detail below, analyses used administrative data on wait times for consults in the VHA to predict self-reported patient satisfaction with care. Study approval for human subjects was obtained from the VA Boston Healthcare System Institutional Review Board. 

Administrative Consult Wait Time Measures

The VHA’s electronic consult system was implemented in 1999 and its use is mandated for all consultation requests (according to internal VHA data). Data from this study was extracted from fiscal year (FY) 2012. The consult system automatically records time stamps when consult-related administrative events occur. These events can be used to understand the number of elapsed days between consult creation and first action and/or scheduling. An additional measure included total consult resolution time (

); in other words, consults are considered resolved when the appointment (if performed) has been completed and the report is written and signed. Time stamps are also recorded when the consult is updated, discontinued, or returned to the sending service for clarification. When consults are returned for clarification of the request, the consult wait time clock is not reset. The time stamps include the time from consult creation through appointment scheduling, and eventually to consult completion. In contrast, discontinued consults stop the clock.

Standardization of the consult system enables distinctions between documents used for traditional clinical consultation and those used for other administrative purposes, such as for requesting transportation.

 With this in mind, we narrowed the 2012 data to focus on consults for clinical services by excluding administrative consults and non-VHA care consults. 

Following our previous work, the wait times were weighted by a national proportion based on FY2011 data. Weights were developed based on the frequency of different consult appointments. If a station did not have a consult request for every type of appointment in a month, the remaining appointment weights were adjusted so they would sum to 100.

 Wait time measures were used in 2 ways in statistical models: 1) as a continuous variable; and 2) categorized roughly into quartiles, with the lowest quartile used as the reference group.

Consult wait times were linked to self-reported satisfaction using the 2012 Survey of Healthcare Experiences of Patients (SHEP), which is modeled after the Consumer Assessment of Healthcare Providers and Systems (CAHPS) family of survey instruments. Managed by the VHA Office of Analytics and Business Intelligence, SHEP is an ongoing nationwide survey that seeks to obtain patient feedback on recent episodes of VHA inpatient or outpatient care. For outpatient care, a simple random sample of patients with completed appointments at VHA facilities is selected each month (according to internal VHA Support Service Center data). The overall response rate was 53% and respondents came from all VHA medical centers (n = 130). 

Different sample selection rules were applied to each consult wait measure. First, all individuals who had the visit date in SHEP match the date for a completed/updated consult were flagged. In addition, we required the station (medical center code) in SHEP to match the station of the completed consult. This was the sample for the completed consult wait measure (n = 28,328). The wait computed was the facility average for all resolved consults in that month. Not surprisingly, because a clinic visit triggers a patient to be eligible to be contacted for SHEP, 90% of the individuals in this sample had a completed/updated status compared with a discontinued or canceled status. 

For the next 2 measures—days to first action and days to scheduled consult&mdash;all individuals who had a visit date in SHEP match the date a consult was initiated were flagged (n = 44,387). The receiving (not the sending) station for these requests was linked because receiving stations actually do the scheduling. We computed a facility average wait time looking forward for all consults requested at that receiving station in the month. 

Satisfaction measures were selected and operationalized following previous work.

 Satisfaction with timeliness of care was measured by asking respondents how often they were able to get VHA appointments as soon as they thought they needed care, excluding times they needed urgent care. Access to VHA tests, treatments, and appointments with VHA specialists was measured by asking how easy it was to get this care in the last 12 months. Response options for the above 3 measures included “always,” “usually,” “sometimes,” and “never”; we estimated the likelihood of answering always/usually compared with sometimes/never. We also examined more general satisfaction measures that wait times for consults may influence. General satisfaction is measured by asking respondents to rate VHA healthcare in the last 12 months on a scale of 0 to 10 and their satisfaction with their most recent VHA visit using a Likert scale ranging from 1 to 7 (higher numbers indicate greater satisfaction). We estimated the likelihood of a 9 or 10 rating compared with less than a 9 on the first measure, and the likelihood of a 6 or 7 compared with less than a 6 on the second measure. 

Risk adjustors included age, gender, race/ethnicity, education level, number of visits to a doctor’s office in the last 12 months, and self-reported health status—all obtained from SHEP FY2012. Models also included month fixed effects to control for secular changes in wait times and a VHA medical center random effect to control for facility quality and case-mix differences. 

STATA version 10.0 (Statacorp, College Station, Texas) was used to estimate logistic regression models that predicted the dichotomized patient satisfaction variables. 

The SHEP respondents selected for this sample reflect the larger VHA patient population. Respondents were predominantly male, in poor health, and frequent healthcare users. There is evidence of high satisfaction with VHA care, with nearly 80% of respondents reporting they usually or always received appointments, treatment, or specialist care in a timely fashion. Additionally, 81% expressed the top 2 highest satisfaction levels for the most recent visit and 58% expressed the highest satisfaction levels with VHA care in the last 12 months (

There is significant variation between facilities regarding consult wait times. Facilities in the top quartile have waits that are more than 10 days longer than facilities in the lowest quartile (33.5 days vs 23 days) for consult completion. Facilities in the highest quartile took about a week longer to schedule appointments in response to a consult request than facilities in the lowest quartile. There was very little variation in time to first action, with less than a half-day difference in the highest quartile compared with the lowest quartile (

The measures for completed consult and time to scheduled consult have strong and consistent relationships with patient satisfaction (

). Generally, there is a linear relationship with satisfaction decreasing for patients who visit facilities with longer waits (the higher quartiles of wait times). Patients who visit facilities in the highest quartiles of wait times are significantly less satisfied than patients who visit facilities in the lowest quartile for these 2 measures. Sensitivity analyses that included wait times as a continuous measure found that longer waits were significantly associated with lower satisfaction on these measures for every outcome except the model using the completed consult wait to predict the overall VHA satisfaction measure (data not shown). There was no significant relationship between the days to first action measure and patient satisfaction.

This study finds a consistent relationship between measures of consult wait times and patient satisfaction. Longer waits between initial request and either scheduling of the consult or the completion of the consult are associated with poorer satisfaction. Generally, there was a stronger negative relationship between waits and satisfaction measures that were specifically related to accessing care, treatments, or specialists compared with more general satisfaction measures. There was no relationship between the waits for time to first action and patient satisfaction. 

These findings are consistent with previous research that validated access metrics for appointments using patient satisfaction as an outcome. Prentice et al (2014) found that different types of access metrics predicted patient satisfaction for new and returning patients.

 Specifically, the wait time between appointment creation and appointment completion for new, but not established, patients strongly predicts patient satisfaction. One reason this relationship pertains to new patients but not established ones may be because new patients typically want to be seen right away due to emerging health concerns. This previously validated measure is consistent with the time to scheduling and time to completion measures used here that predict the same relationship between consult waits and patient satisfaction. Patients being referred to specialty care are likely to have new problems and want to be seen as soon as possible after the need is identified.

The findings from this study also expand our understanding of administrative access metrics that are patient centered. In contrast to the measures, days to scheduled consult and days to completed consult, the days to first action measure had no relationship with patient satisfaction. This metric largely measures “behind the scenes” processes of transfer and scheduling of consults. Patients repeatedly reported a sense of powerlessness and uncertainty, as well as a feeling of “living their life on hold” when waiting for diagnosis or treatment that is compounded by a lack of information from the healthcare system.

 As ACOs and the VHA put a greater emphasis on the experiences of patients, these findings suggest that metrics should focus on measuring tangible processes that patients easily understand as action being taken on their behalf, such as scheduling appointments. 

Given that longer waits for consults negatively impact patient experience, ACOs should have systems and tools in place to streamline the consult referral process and increase care coordination. These tools include policy, training, appropriate technological support tools, and organizational culture. Greenberg et al (2014) argue for the importance of developing a collaborative care agreement that delineates the roles of the referring and consulting provider in the preconsult exchange, the actual consultation, and any co-management of the patient required after the consult is completed.

 Others argue that ACOs need to specifically focus on improving the structure of care coordination. This includes appropriate training to all providers and care teams on best practices in care coordination, technological support tools to help providers get a complete picture of a patient’s care, as well as to achieve care coordination goals and a culture that prioritizes such coordination, including protected time during the workday for care coordination activities.

The main limitation of this study is that we cannot definitely state that the relationship between longer consult wait times and lower patient satisfaction is causal because omitted variables may be responsible. Our models attempted to minimize this possibility by including month fixed effects to control for secular changes in wait times and a medical center random effect to control for facility quality and case-mix differences. On the other hand, research has repeatedly found a relationship between longer wait times and poorer outcomes, including decreased patient satisfaction and poorer health outcomes in a variety of veteran populations and time periods, thus strengthening the inference that the relationship may be causal.

Another limitation is that the study sample is largely elderly and male, so results may not be generalizable to other patient populations. Finally, wait times for all types of consults may not have the same impact on patient satisfaction. Due to data availability at the time of the study, our measures included all clinical consults. Long waits for a recommended preventive screening (eg, colonoscopy) may not have the same effect on patient satisfaction as waits for consults that are a result of new health concerns. Future work should consider these nuances to determine consult access that has the largest impact on patient experiences. 

The consult process occurs at an anxiety-producing time for patients. Findings from this study suggest that certain types of consult waits that can be easily obtained from the scheduling system are strong predictors of patient satisfaction. As ACOs reorganize to become more patient-centered, better management of consult waits has the potential to significantly improve patient satisfaction. 

	The authors are indebted to Aaron Legler for programming support. 

Partnered Evidence-based Policy Resource Center (SDP, JCP), Boston VA Healthcare System, Boston, MA and Office of Veterans Access to Care (MLD), Department of Veterans Affairs

Washington, DC; Department of Pharmacy and Health System Sciences, Northeastern University (SDP), Boston, MA; School of Medicine and Public Health, Boston University (JCP), Boston, MA.

Funding for this research was provided by the Access Clinic and Administration Program (now the Office of Veterans Access to Care) in the Department of Veterans Affairs.

Concept and design (SDP, MLD, JCP); acquisition of data (SDP, MLD, JCP); analysis and interpretation of data (SDP, JCP); drafting of the manuscript (MLD, JCP); critical revision of the manuscript for important intellectual content (SDP, MLD); statistical analysis (SDP); obtaining funding (MLD); administrative, technical, or logistic support (SDP, MLD); and supervision (SDP, JCP). 

Julia C. Prentice, PhD, VA Boston Health Care System, 150 South Huntington Ave, Mail Stop 152H, Boston, MA 02130. E-mail: jprentic@bu.edu.

1. Accountable care organizations (ACOs): general information. CMS website. https://innovation.cms.gov/initiatives/aco/. Updated December 15, 2016. Accessed December 9, 2015.

2. McWilliams JM, Landon BE, Chernew ME, Zaslavsky AM. Changes in patients’ experiences in Medicare accountable care organizations. 

. 2014;371(18):1715-1724. doi: 10.1056/NEJMsa1406552.

3. Press MJ, Michelow MD, MacPhail LH. Care coordination in accountable care organizations: moving beyond structure and incentives. 

4. CAHPS Survey for Accountable Care Organizations (ACOs) Participating in Medicare Initiatives—2016 ACO-9 Survey version required (English). CMS website. http://acocahps.cms.gov/globalassets/aco---epi-2-new-site/pdfs-for-aco/survey-instruments/2016-aco-survey/english/2016_aco-9_mail_survey_english.pdf. Accessed December 9, 2015.

5. Greenberg JO, Barnett ML, Spinks MA, Dudley MJ, Frolkis JP. The “medical neighborhood”: integrating primary and specialty care for ambulatory patients. 

. 2014;174(3):454-457. doi: 10.1001/jamainternmed.2013.14093. 

6. O’Malley AS, Reschovsky JD. Referral and consultation communication between primary care and specialist physicians: finding common ground. 

. 2011;171(1):56-65. doi: 10.1001/archinternmed.2010.480.

7. Fogarty C, Cronin P. Waiting for healthcare: a concept analysis. 

. 2007;61(4):463-471. doi: 10.1111/j.1365-2648.2007.04507.x.

8. Hansen BS, Rørtveit K, Leiknes I, et al. Patient experiences of uncertainty—a synthesis to guide nursing practice and research. 

. 2012;20(2):266-277. doi: 10.1111/j.1365-2834.2011.01369.x.

Given that accountable care organizations (ACOs) will be rated on patient experience and wait times for specialist consults are associated with patient satisfaction, ACOs should monitor this process.

Consult Coordination Affects Patient Experience

Long-acting insulin analogues (eg, insulin glargine and insulin detemir) are an alternative to neutral protamine Hage-dorn (NPH) insulin for maintaining glycemic control in patients with diabetes. Clinical trials comparing analogue insulin and NPH have neither been adequately powered nor had sufficient follow-up to examine long-term health outcomes.

 To compare the effects of NPH and long-acting insulin analogues on long-term outcomes.

This retrospective observational study relied on administrative data from the Veterans Health Administration and Medicare from 2000 to 2010. Local variations in analogue insulin prescribing rates were used in instrumental variable models to control for confounding. Outcomes were assessed using survival models.

The study population included US veterans dually enrolled in Medicare who received at least 1 prescription for oral diabetes medication and then initiated long-acting insulin be-tween 2001 and 2009. Outcomes included ambulatory care—sensi-tive condition (ACSC) hospitalizations and mortality.

: There was no significant relationship between type of insulin and ACSC hospitalization or mortality. The hazard ratio for mortality of individuals starting a long-acting analogue insulin was 0.97 (95% CI, 0.85-1.11), and was 1.05 (95% CI, 0.95-1.16) for ACSC hospitalization. Differences in risk remained insignificant when predicting diabetes-specific ACSC hospitalizations, but starting on long-acting analogue insulin significantly increased the risk of a cardiovascular-specific ACSC hospitalization.

 We found no consistent difference in long-term health outcomes when comparing use of long-acting insulin analogues and NPH insulin. The higher cost of analogue insulin without demonstrable clinical benefit raises questions of its cost-effectiveness in the treatment of patients with diabetes.

The progressive nature of type 2 diabetes mellitus requires many patients to initi-ate insulin. This research compares the effects of the neutral protamine Hagedorn (NPH) and analogue insulin on long-term outcomes. n Analogue insulin is significantly more expensive than NPH.

Clinical trials comparing analogue insulin with NPH have not been adequately powered nor had sufficient follow-up to examine long-term health outcomes.

There were no significant differences in the risk of ambulatory care—sensitive condition hospitalization or mortality for patients who initiated analogue insulin compared with NPH.

The higher cost of analogue insulin without demonstrable clinical benefit raises questions of its cost-effectiveness in the treatment of patients with diabetes.

The progressive nature of type 2 diabetes mellitus requires many patients to use insulin to maintain glycemic control.

 Neutral protamine Hagedorn (NPH) insulin was the most commonly used intermediate- to long-acting insulin until the introduction of the long-act-ing insulin analogues: insulin glargine in 20002,

 The long-acting insulin analogues were designed with properties that prolonged absorption, and with a flattened activity peak that extended duration of ef-fect and also resulted in a reduced risk of hypoglycemia.

	The significantly higher cost of analogue insulin compared with NPH

 led researchers and policy makers to compare the efficacy and cost of these medications. Several reviews found no significant difference between analogue insulin and NPH on glycemic control or severe hypoglycemia (ie, low glucose level requiring assistance from another person) but did show reduced likelihood of nocturnal hypoglycemia for patients using analogue insulin.

	Despite these short-term differences, potential long-term benefits have been difficult to test due to the short time frame of studies comparing analogue insulin with NPH.2

Some cost-effectiveness studies were based on modeling that used clinical trial results as predictors of long-term treatment ef-fects (eg, Center for Outcomes Research Diabetes Model).1 Other cost-effectiveness studies used retrospective claims that may better reflect clinical settings outside of trials. Results from these studies are mixed, with some concluding analogue insulin was cost-effective

 and others concluding analogue insulin was not an efficient use of healthcare resources.

 However, an important limitation of these claims-based stud-ies was the assumption that no systematic, unobserved differ-ences existed between the groups being compared (eg, those that started on NPH compared with analogue insulin).

	To bridge this gap in the evidence, we used national Vet-erans Health Administration (VA) records to systematically compare long-term outcomes for patients using NPH or ana-logue insulin. We used a novel approach to an established observational comparative effectiveness method, instru-mental variables, to obtain unbiased estimates despite the possibility of unobserved differences between groups.

Patient-level national data from the VA were used and supplemented with data from Medicare to ensure com-pleteness in the measures of outcomes, as VA patients often use non-VA facilities for hospital care.

The study was reviewed and approved by the Institutional Review Board at the VA Boston Healthcare System.

All prescription claims for metformin, sulfonylurea, thiazolidinedione, and long-acting insulin (ie, NPH, insu-lin glargine and insulin detemir, and mixtures of short- and long-acting insulins) between 2000 and 2007 were extracted from VA pharmacy files (Figure). Eligible indi-viduals had a history of being prescribed at least 1 oral diabetes medication, and initiated long-acting insulin between February 1, 2001, and December 31, 2009. The initiation date for the long-acting insulin was the start of the study period (“index date”) for each patient and the prior 12 months was the baseline period. To ensure that all hospitalizations were recorded, we further limited the cohort to include only those enrolled in Medicare as well as the VA. The last index date permitted was at the end of 2009, to allow a minimum 12-month outcome period. This resulted in a cohort of 142,940 patients, including 118,878 who initiated NPH and 24,062 who initiated ana-logue insulin.

The main objective of the study was to compare the long-term effects of NPH and analogue insulin. Patients who initiated NPH or an NPH/regular in-sulin mixture were compared with patients who started glargine, detemir, or short-act-ing and long-acting mixtures that included analogue insulin. To improve comparabil-ity among patients, we attempted to isolate patients at a similar point in the progres-sion of diabetes. Consequently, patients were required to have a prescription for an oral diabetes medication during the base-line period before initiating insulin. Eighty percent of the individuals remained on the same insulin they started (NPH or analogue insulin). Among those who started on analogue insulin, 99% re-ceived glargine.

Provider Practice Pattern as Instrumental Variable

The principal threat to a simple comparison of out-comes is that the selection of treatment by patients and providers could be influenced by unmeasured differences in patient risk (selection bias or confounding by indica-tion). To address this, we developed instrumental vari-ables models, which identify a factor (the instrumental variable [IV]) that influences treatment but is effectively random with respect to patient risk and other potential confounders.

The statistical model isolates the com-ponent of treatment variation attributable to the IV and measures the relationship between this component and outcomes.

 The success of the approach depends on the effective randomness of the IV (controlling for all the other variables in the model) as well as the strength of the IV’s influence on treatment status.

	VA patients are assigned to their primary care physi-cian by variable and often arbitrary methods, so in this study, the IV was the proportion of long-acting insulin prescriptions written for analogue insulin by each pro-vider during an individual’s baseline period19—the 12 months before the index date on which analogue insulin was prescribed. The instrument assigned to each individ-ual was the proportion of analogue insulin prescriptions written by the provider who prescribed the initial insulin prescription on the index date.

For example, if an individual started on long-acting in-sulin on January 1, 2003, their baseline period was January 1, 2002, to December 31, 2002 (see 

). We identified the provider that prescribed insulin on Jan-uary 1, 2003, and calculated their proportion of analogue insulin prescriptions in that baseline period (January to December 2002). An individual’s insulin prescription does not contribute to their instrumental variable calcu-lation, eliminating concerns about selection bias. Provid-ers and patients were aligned based on the index date to minimize confounding that could occur if patients later switched providers.

	If a provider prescribed insulin to less than 10 unique patients during the baseline period (53% of the time), the rate at the community-based outpatient clinic (CBOC) or VA medical center (VAMC) at which the provider prac-ticed was used.

The IV will not be effective if correlated with other provider characteristics that might also affect outcomes, causing biased estimates of the treatment-outcome rela-tionship. Therefore, we included 3 provider-level process quality variables: percent of glycated hemoglobin (A1C) labs ≥9%,

 percent of blood pressure readings ≥140/90 mm Hg,

 and percent of low-density lipoprotein choles-terol labs >100 mg/dL.22 These variables were computed at the same provider, CBOC, or VAMC level, and time periods as the IV prescribing rate.

Additional control variables computed at baseline included patient age, sex, race, A1C, serum creatinine, urine microalbumin, body mass index (see Table 1 for cat-egorizations), 29 indicator variables for comorbidities,23 8 indicator variables for the components of the Young dia-betes severity index,24 and indicator variables for calen-dar years corresponding to index dates. Individuals can be diagnosed with multiple comorbidities (eg, obesity and congestive heart failure [CHF]).

Outcomes included mortality and hospital admission (VA or Medicare) for any of 13 ambulatory care—sensitive conditions (ACSCs) as defined by the Agency for Health-care Research and Quality.

These hospitalizations are hypothesized to be preventable with high-quality outpa-tient care and include several diabetes and cardiovascular complications such as uncontrolled diabetes, short- and long-term complications of diabetes, CHF, and chronic obstructive pulmonary disease. Sensitivity analyses es-timated models for specific ACSC hospitalization types most closely related to diabetes and cardiovascular dis-ease. The VA Vital Status File, which determines the date of death from VA, Medicare, and Social Security Admin-istration data, was used to determine mortality.

 Patients were censored either at the end of 2010, or when they ex-perienced either outcome (mortality or ACSC hospitaliza-tion), or the date they first switched between long-acting insulin types (NPH and analogue insulin). Consequently, the modeled outcome was the amount of time between the index date and each individual’s censoring date.

We used Stata version 10 (StataCorp LP, College Sta-tion, Texas) to estimate the effects of analogue insulin use on the risks of outcomes using Cox proportional hazards models. This model was chosen for 2 reasons: first, a time-to-event analysis has more statistical power than a logistic regression, because more information is used. In a logistic regression, estimates are based on dif-ferences between individuals who had events and those who did not. In a time-to-event model, estimates are based on differences in time to event as well. The second reason to use a Cox model was to mimic a clinical trial study design. In this study, individuals started analogue or NPH insulin and we then pre-dicted time until death or an ACSC hospitaliza-tion, using the provider’s prescribing pattern as an instrumental variable to control for selection bias. This is analogous to individuals starting on the treatment or placebo in a randomized clinical trial.

	The IV approach estimates a pair of simultane-ous equations: one for the likelihood of receiving analogue insulin compared with NPH, and the sec-ond for the likelihood of the outcome. The treat-ment equation modeled the likelihood of receiving analogue insulin as a function of the provider-level prescribing rate and control variables. The outcome equations related treatment and control variables to probabilities of ACSC hospitalization and death. Because the outcome equations were nonlinear, we used the 2-stage residual inclusion technique

 to es-timate IV specifications. Two-stage models typically require computationally intensive bootstrapping to calculate accurate standard errors.

 We bootstrapped standard errors for the mortality model and deter-mined that the large samples and strong instruments involved resulted in bootstrapped values that were virtually identical to the asymptotic values generated automatically by the statistical software.

	The Cox models assume that analogue insulin treatment increases or decreases outcome risk by a constant proportion over time. We tested this as-sumption using scaled Schoenfeld residuals from the mortality and hospitalization equations.

	Finally, to control for facility quality differences, we included a facility-level fixed effect in the treat-ment equation and a facility-level random effect in the outcome equations. Including a fixed effect in the treatment equation and a random effect in the outcome equation improved identification between the 2 stages and strengthened computational effi-ciency of the estimation algorithms. In sensitivity analyses, we estimated models with fixed effects in both the first and second stage; results were quali-tatively unchanged but computationally intensive so we chose to present the random effects version.

Patients in the sample were elderly (mean age = 69 years), about 84% white, and overwhelmingly male (Table 1). Over a quarter of the patients had average A1C ≥9% and diabetes complications such as retinopathy or nephropathy during the baseline period. Patients also had high rates of cardiovascular comorbidities, with 42% diag-nosed with obesity and 26% with CHF. Sixteen percent of the sample died and 30% had an ACSC hospitalization in the outcome period.

compare means of selected demographics and risk adjustment variables focused on diabetes severity and cardiovascular comor-bidities for patients who started on NPH compared with analogue insulin. Individuals who started on analogue in-sulin were slightly older (aged 70.6 years compared with 69 years) and had more diabetes complications and greater comorbidity burden in the baseline period. Patients on analogue insulin had lower rates of A1C ≥9% at baseline (21% compared with 27%) but were more likely to be missing A1C and BMI measurements during the baseline period. Thirty-two percent of the in-dividuals who started on NPH had an ACSC hospitalization compared with 21% of indi-viduals who started on ana-logue insulin; the figures were 17% and 11%, respectively, for mortality. providers had higher ACSC hospitalization and mortali-ty rates compared with patients assigned to high analogue prescribing providers.

There were no significant differences in the risk of ambulatory care—sensitive condition hospitalization or mortality between patients who initiated analogue insulin compared with the neutral protamine Hagedorn. 

Long-Term Outcomes of Analogue Insulin  Compared With NPH for Patients With Type 2 Diabetes Mellitus

 Medicare payment reforms will reimburse accountable care organizations (ACOs) for providing high-quality healthcare. Quality measures that reliably predict health outcomes are required.

To compare the ability of alternative wait time measures to predict glycated hemoglobin (A1C) levels among diabetes patients.

 This retrospective observational study relied on Veterans Health Administration utilization data and Medicare claims data from 2005 to 2010.

 Outcomes included an average A1C level and uncontrolled A1C. Heckman selection models simultaneously predicted the presence of an A1C value and its level. Models were riskadjustedfor prior individual health status. The main explanatory variables of interest were facility-level primary care wait times measured in days. Several measures were tested, including capacity measures and retrospective and prospective time stamp measures. The time stamp measures used either the date the appointment was created in the scheduling system (create date) or the date the patient or provider desired the appointment (desired date) as the start date for wait time computation. All measures were calculated separately for new and returning patients.

New patient capacity and create date measures significantly predicted outcomes, but desired date measures did not. The returning patient retrospective create date and desired date wait  time measures significantly predicted higher and uncontrolled A1C, but capacity and prospective create date measures did not.

Different administrative wait times predict A1C for new and returning patients. To properly measure quality, ACOs should use wait time measures that demonstrate relationships with outcomes for subpopulations of patients. 

Medicare payment reforms require measures that can reliably predict outcomes. This study compares the ability of alternative wait time measures to predict glycated hemoglobin (A1C) levels among diabetes patients.

		New patient capacity and date when appointment was created in the scheduling system (create date) measures significantly predicted outcomes, but desired date measures did not.

		Returning patient retrospective create date and desired date wait time measures significantly predicted A1C, but capacity and prospective create date measures did not.

		Accountable care organizations should use access measures that demonstrate relationships with outcomes for patient subpopulations.

A crucial piece of the Affordable Care Act legislation is the implementation of accountable care organizations (ACOs) that financially reward a group of providers for providing highquality healthcare at low cost.

ACOs may provide the greatest benefits to patients diagnosed with chronic disease due to their focus on providing high-quality, well-coordinated, and patient-centered care.

 Patients diagnosed with diabetes are likely to be targeted for disease  management by ACO managers because of the high prevalence and  significant monetary costs of managing diabetes complications. Approximately 17.5 million individuals in the United States have diabetes, and the total cost related to diabetes in 2007 was $174 billion.

The success of ACOs relies on robust quality measurement.

 Policy makers have emphasized the need to complement process quality measures (eg, the percentage of patients with diabetes who receive glycated hemoglobin [A1C] tests) with patient-centered measures that demonstrate improved patient outcomes.

Timely access to healthcare has a strong potential to meet this criterion.

Previous research suggests that access measures that do not rely on patient self-report reliably predict health outcomes. Studies using capacity wait time measures (eg, time to first next available appointment [FNA]) and comparing populations exposed to open access scheduling with controls have found patients who face shorter waits have higher primary care utilization, lower A1C levels, and a decreased risk of experiencing poor health outcomes such as mortality or preventable hospitalizations.

 Notably, in the context of ACOs, Prentice and colleagues6,7 found the effects of longer FNA waits were largest with the most vulnerable veterans—those who were older, had more comorbidities, or had higher A1C levels during baseline.

Adherence may be one factor that contributes to the observed relationship between wait times and outcomes. Several studies indicate that patients diagnosed with diabetes who are more adherent to their treatment plan experience better health outcomes and have lower healthcare cost.

 found a “white coat” effect, with adherence to medication increasing right before and right after physician visits. Consequently, longer waits between visits may decrease a patient’s ability to get the most effective treatment and lead to lower overall medication adherence. Patients with diabetes often underestimate the importance of consistent treatment adherence,

 and longer waits between appointments will decrease opportunities for patient education.

Although the importance of timely access to healthcare is widely recognized, the best method of measuring timely access has yet to be determined. The Centers for Medicare & Medicaid Services are currently using patient surveys to evaluate ACOs’ ability to deliver healthcare as soon as it is desired.

An alternative approach relies on administrative data from patient scheduling systems. For example, capacity measures (eg, the number of days until the first or third next available appointments) are commonly used when implementing Advanced Clinic Access.

 The Veterans Health Administration (VHA) found that capacity measures may not reflect the access limitations actually experienced, especially for returning patients who are typically trying to schedule follow-up appointments.

 Consequently, a variety of wait time measures have been developed. This study is the first to compare the abilities of these alternative measures of wait times to predict glycemic control among patients diagnosed with diabetes. Results could be used in Medicare and the private sector to improve upon the current quality metrics used by ACOs.

This study used administrative data from a wide variety of VHA and Medicare data sets. Please refer to 

Medicare payment reforms require valid measures of high-quality healthcare. Different types of administrative wait time measures predicted glycated hemoglobin levels for new and returning patients.

Julia C. Prentice, PhD

Articles

Call Center Performance Affects Patient Perceptions of Access and Satisfaction

Consult Coordination Affects Patient Experience

Long-Term Outcomes of Analogue Insulin Compared With NPH for Patients With Type 2 Diabetes Mellitus

Using Health Outcomes to Validate Access Quality Measures

Outpatient Wait Time and Diabetes Care Quality Improvement