UTILIZATION MANAGEMENT FOR ADULT MEMBERS

UTILIZATION MANAGEMENT FOR ADULT MEMBERS Quarter 2: (April through June 2014) EXECUTIVE SUMMARY & ANALYSIS BY LEVEL OF CARE Submitted: September 2, 2014 CONNECTICUT DCF CONNECTICUT

Utilization Report for Adult Members Quarter 2, 2014 General Overview On at least a quarterly basis, the reports mutually agreed upon in Exhibit E of the CT BHP contract are submitted to the state for review. This Quarterly Report focuses on the utilization management portion of these reports, evidenced in the 4A series which reviews utilization statistics such as average length of stay (ALOS) and admissions per 1,000 members (Admits/1,000). NOTE: A detailed description of the measures can be found at the end of this document. As stated in previous submissions, results were graphed only for benefit groups that had a sufficient volume of members receiving services in each level of care (LOC). To provide better clarity when viewing the graphs, we have highlighted the benefit groups that appear on the related graph. The Quarterly Report focuses only on those levels of care in which the data warranted analysis and discussion as evidenced by significant changes and trends or in cases when changes and trends are unclear and additional data is needed. If the analysis for a LOC did not reveal results or trends that warranted discussion, the results were removed from the body of this Report and placed in an Appendix at the end. This quarter, tables associated with utilizations graphs have also been placed in the Appendix. As a result, this Report outlines/highlights the areas of interest related to certain utilization trends, as well as the underlying factors which drive the trend and associated programmatic responses taken by VO to impact/mitigate or support the trend. We also present recommendations to address remaining challenges and report progress related to these planned recommendations. The areas of focus for this quarter are listed in the table to the right. Areas of Focus for this Quarter Membership Inpatient Facilities - Admits/1,000 - Days/1,000 - Average Length of Stay (ALOS) Inpatient Detoxification: Hospital-Based - Admits/1,000 - Days/1,000 - ALOS Inpatient Detoxification: Free-Standing Detox Programs - Admits/1,000 - Days/1,000 - ALOS Home Health Services - Admits/1,000 - Medication Administration Frequency Methadone Maintenance: Admits/1,000 Outpatient: Admits/1,000 This quarter, the following utilization data points have been placed in the Appendix and are not discussed: Mental Health Group Home - Admits/1,000 - Days/1000 - ALOS Partial Hospital Program: Admits/1,000 Intensive Outpatient: Admits/1,000 Ambulatory Detox: Admits/1,000 Utilization Report for Adult Members : Page 1

Methodology The data contained in this report are based on authorization admissions and are refreshed for each subsequent set of Quarterly Reports during the year. Due to retrospective authorizations and changes in eligibility, the results for each quarter may change from the previously-reported values. The reports and analyses for all levels of care are affected by this change. Therefore the data analysis applies to the previous reporting period, though the current reporting period data is provided. To ensure that only refreshed data is the focus, data from the most recent quarter that has not been analyzed has been shaded in gray. Further details regarding the methodology used for determining membership counts, length of stay, admits/1000 and days/1000, and statistical significance of changes in percentages across time are listed in the last section of the report prior to the appendices, on page 21. Reports Used CTBH06030AL_Utilization_Statistics_(4A_2)_XRV CTBH06030AS_Utilization_Statistics_(4A_2)_XRV CTBH06030FS_Utilization_Statistics_(4A_2)_XRV CTBH06030HB_Utilization_Statistics_(4A_2)_XRV CTBH06030LS_Utilization_Statistics_(4A_2)_XRV CTBH06030ML_Utilization_Statistics_(4A_2)_XRV CTBH06030AD_Utilization_Statistics_(4A_2)_XRV CTBH06030FD_Utilization_Statistics_(4A_2)_XRV CTBH06030LD_Utilization_Statistics_(4A_2)_XRV CTBH06030ALD_Utilization_Statistics_(4A_2)_XRV Membership During Q1, 2013, in collaboration with the state departments, the decision was made to remove dually eligible members from all utilization data included in the quarterly reports, with the exception of the home health and mental health group home measures. As of Q2 13, this decision was executed in all of the Exhibit E Reports. The decision was based on the finding of inconsistencies in the authorization of services for these populations among the providers. While a small minority of providers requested authorization for the entire stay of a dual eligible member s treatment, others gained authorization only when Medicare lifetime maximums were approached and still others never sought authorization at all. As a result of these inconsistencies, utilization measures for dually eligible members were inaccurate and potentially skewed aggregate All Benefit measures. For this Report, indicators were only reported without the duals. However, we will continue to present a total membership table that includes the dually eligible and a membership table that excludes the dually eligible, for comparison purposes. Utilization Report for Adult Members : Page 2

As seen in previous quarters, there is an increase in Membership in Q2 13. The 2.8% change in membership prior to the data refresh, gives rise to the expectation that membership will increase dramatically again in Q2 14. From Q4 13 to Q1 14, Total Membership-Duals Removed increased from 331,992 to 382,207 (a difference of 50,215 members, and a 15.1% increase) and Total Membership-Including Duals increased from 407,903 to 458,234 (a difference of 50,331 members, and a 12.3% increase). These numbers indicate that only 0.2% of new members were Dually Eligible. As expected, 43,736 of these new members were included in the HUSKY D (MLIA) benefit group (87.1% of the Total Membership increase excluding duals), a 37.5% increase in HUSKY D membership over the previous quarter. Family Single membership increased by 11,408, a 6.4% increase over the previous quarter. From Q4 13 to Q1 14, Total Membership- Duals Removed increased from 331,992 to 382,207 (a difference of 50,215 members, and a 15.1% increase Utilization Report for Adult Members : Page 3

Membership Data Refresh The refresh rate for this quarter was 7.41%, far greater than the historical refresh of less than 2%. This increase in the refresh rate is likely due to logistics associated with the rollout of the Affordable Care Act. Refresh Percentage Change by Quarters All Benefit Groups - Duals Removed Q1 13 Q2 13 Q3 13 Q4 13 Q1 14 Original 316,408 320,017 321,524 325,868 355,844 Refresh One Quarter Later 320,068 325,881 327,205 331,992 382,207 Pct Change 1.16% 1.83% 1.77% 1.88% 7.41% UM Analysis by Level of Care Inpatient Psychiatric Utilization Analysis While the total admissions increased slightly from Q4 13 to Q1 14, the total Admits/1,000 decreased slightly, led by the noticeable decrease in HUSKY D (MLIA) Admits/1,000. This is a result of the dramatic increase in membership in the HUSKY D (MLIA) benefit group, which has the effect of diluting the concentration of admissions. The Days/1,000 data follows the same pattern as the Admits/1,000. The ALOS for each benefit group has remained relatively stable after accounting for the expected seasonal increase in response to occasional inclement weather. All Benefit Groups increased admissions by 4.4% from Q4 13 to Q1 14, but still decreased in Admits/1,000 by 8.6% due to the total membership increase of 50,215. HUSKY D (MLIA) admissions increased by only 5.2% despite adding 43,736 members to the group. That resulted in a 22% decrease in Admits/1,000. Utilization Report for Adult Members : Page 4

The Days/1,000 results reflected the same pattern as Admits/1,000. That result was expected, given the effect of membership on this indicator. The greatest increase in ALOS was in the HUSKY C (ABD Single) population. That measure increased by 10% from Q4 13 to Q1 14. Conclusions In summary, Admits/1,000 and Days/1,000 decreased from Q4 13 to Q1 14. We believe that the increase in membership has had a significant impact on the per 1,000 measures. The decrease in per 1,000 measures will continue as membership continues to increase, provided that the new membership does not require services in higher levels of care and the number of admissions is stable. We would like to gain a better understanding of new member utilization. Historically, when a member gains coverage, pent up need for services results in an increase in utilization at all levels. It appears that the new membership in Medicaid through the ACA may not follow past trends related to new member utilization. The ALOS for Q1 14 was 8.60 days (see table in the appendix). The ALOS has remained near 8 to 8.5 days for the past 2 years. The small variation in ALOS from one quarter to the next has no clear pattern or explanation. It appears that in 2013 and 2014 the ALOS in Q1 is longer than the previous quarters. It is likely that inclement weather was a contributing factor in ALOS for both Q1 13 and Q1 14. Providers have also reported longer waits for state beds. Based on provider feedback, we have begun tracking the number of days members wait on inpatient units for Historically, when a member gains coverage, pent up need for services results in an increase in utilization at all levels. It appears that the new membership in Medicaid through the ACA may not follow past trends related to new member utilization. Utilization Report for Adult Members : Page 5

state beds. There is currently no ability to track these wait times retrospectively, but if ALOS has been affected recently, we will have a better understanding of the impact in the future. The new Bypass parameters and program inclusion was rolled out to the providers in June 2014. Bypass inclusion criteria included ALOS and readmission rates. In 2015, criteria will include percentage of discharges completed via the web, as well as ALOS and readmission rates. Bypass hospitals have been offered the opportunity to obtain authorizations on Connect, VO s web-based system. All Bypass providers have been trained, and now submit clinical reviews (pre-certs and concurrent) on-line which allows the reviews to be completed at their convenience. Initial authorizations for the adult units are now 7 days, up from 5 days, which should also reduce administrative burden. We believe that, as Bypass hospitals become more comfortable working in Connect, they will understand better and appreciate more the benefits of being on the Bypass Program. We have had very few provider complaints regarding the new process. Clinical staff and RNMs will continue to support providers during this time of change. Care Managers will also provide ongoing support with treatment planning and discharge planning for all members, especially those with complex needs. Inpatient providers now have access to VO s adult inpatient dashboard. The dashboard contains real-time (up to the most recent month) hospital-specific and state-wide information for ALOS and readmission rates. Providers do not appear to be utilizing the dashboard as frequently as had been hoped. However, RNMs will strongly encourage providers to utilize this innovative tool and will continue to describe its benefits. In the meantime, RNMs will review providerspecific PAR data via the dashboard during regularly scheduled quarterly meetings. Utilization Report for Adult Members : Page 6

Recommendations 1. Continue the Adult Inpatient Bypass Program Determination of bypass status will be conducted annually, and mid-year monitoring will be conducted at the end of Q2 (each year). Update: The new bypass program was implemented, effective July 1, 2014. Provider Relations held numerous webinars for providers during June and early July to explain and train hospital employees on new procedures. All hospitals are expected to submit concurrent reviews on-line and Bypass hospitals are expected to complete pre-certifications on-line. Bypass hospitals now receive a 7 day/unit authorization on pre-cert, an increase of 2 days/units from prior Bypass authorization parameters. RNMs and Provider Relations continue to work with hospitals to problem-solve issues that arise. Activities related to this recommendation are ongoing and a progress update will be described in the next quarterly report. 2. Continue Adult PAR Program RNMs will share Q2, 2014 PAR data with hospitals, and meetings will be arranged as needed. Follow up will occur specifically with those hospitals that have higher lengths of stay. The next inpatient provider meeting will be held before the end of the year. Update: The Inpatient Provider meeting was held on June 16. Providers and DMHAS discussed the process for determining how members are admitted to state beds. Detailed data from the 2013 Inpatient PT was presented and discussed. RNMs are currently scheduling PAR meetings with providers to share Q2 2014 data. Activities related to this recommendation are ongoing and a progress update will be described in the next quarterly report. 3. Develop a reporting measure to track units authorized for members who are inpatient and awaiting state beds Update: In June 2014, clinical staff began indicating in the Connect system when adult members are on an inpatient psychiatric unit awaiting a state bed. Many of the members on the state bed waitlist will remain on an inpatient psychiatric unit, still meeting criteria for IPF, until a state bed is available. Care managers will continue to work closely with providers to identify alternative discharge plans for members waiting for state beds. There are occasions when members symptoms improve and lower levels of care can be authorized, thus avoiding utilization of a state bed. The four DMHAS sites identified as long term inpatient programs for the purposes of tracking days awaiting state beds are: Connecticut Valley Hospital, Capitol Region Mental Health Center, Connecticut Mental Health Center, and Bridgeport Mental Health Center. Activities related to this recommendation are ongoing and a progress update will be described in the next quarterly report. Inpatient Detox Hospital Based Utilization Analysis The Admits/1,000 for each benefit group was stable at previous levels until Q1 14, when there were changes in authorization procedures that increased the admissions reported. Previously, medical detox admissions were authorized by the Community Health Network (CHN). Starting on March 1, 2014 authorizations were issued by ValueOptions. The dramatic increase in Admits/1,000 (and Days/1,000) appears to be largely due to the new authorization process given that the observed increase of 350% greatly exceeds the percentage increase in new membership. The precise contribution of the change in authorization process and the increase in new membership will require further analysis over time. Utilization Report for Adult Members : Page 7

The dramatic increase in Admits/1,000 (and Days/1,000) appears to be largely due to the new authorization process given that the observed increase of 350% greatly exceeds the percentage increase in new membership. The precise contribution of the change in authorization process and the increase in new membership will require further analysis over time. HUSKY A (Family Single) increased by 21 admissions, HUSKY C (ABD Single) increased by 32 admissions, and HUSKY D (MLIA) increased by 174 admissions. As described above, the Days/1,000 rates increased consistently with the Admits/1,000. Utilization Report for Adult Members : Page 8

The ALOS for All Benefit Groups has varied over time, but remains in the range of 4-5 days. Conclusions VO began authorizing medical detoxes on March 1, 2014. As a result, hospital-based detox utilization has increased in Q1 14 and is expected to increase again in Q2 14. We have worked with providers to clarify any questions or concerns regarding the process. We will gain a better understanding of hospital-based detox utilization when Q2 14 data measures are analyzed. Q2 14 will be the first quarter for which we will have been authorizing this level of care for the entire duration of the quarter. The ALOS for Q1 14 was 4.89 days (see table in the appendix). As part of the inpatient 2014 performance target activities, ICM and Peer teams have been working with St. Francis and Yale to increase connections to care and decrease readmission rates for members admitted to hospital-based detox. The Peer/ICM teams will work with St Francis and Yale for the foreseeable future, aiding hospital staff in discharge planning and supporting members on their pathway to recovery. Recommendations -See Global Recommendation #2 As recommended in the Global recommendation section below, a dedicated group of clinicians with a focus on co-management with CHN has been developed. This group of care managers works closely with hospital staff, CHN, and ABH to ensure that discharge plans are appropriate and secure for members admitted for hospital-based detox. The co-management team will continue to complete authorizations and aid in discharge planning for all members receiving hospital-based detox. VO began authorizing medical detoxes on March 1, 2014. As a result, hospital-based detox utilization has increased in Q1 14 and is expected to increase again in Q2 14. Utilization Report for Adult Members : Page 9

Inpatient Detox Free-Standing Utilization Analysis The Admits/1,000 and Days/1,000 have remained steady for all benefit groups, with the exception of the HUSKY D (MLIA) benefit group, which decreased on these measures from Q4 13 to Q1 14. The decrease is believed to be due to the dramatic increase in membership in the HUSKY D (MLIA) population as a result of the Affordable Care Act. The Admits/1,000 for All Benefit Groups decreased by 11.5%. This decrease is primarily driven by the HUSKY D (MLIA) results, which showed that admissions remained relatively unchanged, despite nearly 90% of new membership entering that group. As a result, HUSKY D (MLIA) Admits/1,000 decreased by 24.6%. The Days/1,000 measure showed similar results, as is expected, given that this measure is affected by changes in membership, as well. Utilization Report for Adult Members : Page 10

The ALOS has remained flat across all time periods and benefit groups. Protocol driven detox regimens drive the length of stay at free-standing facilities. Conclusions The HUSKY D (MLIA) population consistently utilizes free-standing detox services at a higher rate than other HUSKY members. There were 1,887 (see table in the appendix) HUSKY D (MLIA) admissions in Q1 2014, representing 82% of the total admissions (2,303). HUSKY D (MLIA) per 1,000 measures for this quarter were greatly affected by the increase in membership during Q1 14. Admits/1,000 and Days/1,000 decreased from Q4 13 because the number of admissions remained stable while membership grew significantly. The ALOS has remained stable over time. Despite providers reporting that each member is assessed and medicated based on their unique needs, the ALOS remains unchanged at about 4 days. RNMs and Clinical staff will continue to meet with providers and ABH to develop interventions that support members transitioning from one level of care to the next. The HUSKY D (MLIA) population consistently utilizes freestanding detox services at a higher rate than other HUSKY members. There were 1,887 (see table in the appendix) HUSKY D (MLIA) admissions in Q1 2014, representing 82% of the total admissions (2,303). Utilization Report for Adult Members : Page 11

The number of members receiving Home Health services increased by 8.8% from Q1 11 to Q4 13, (4,617 to 5,021). During this same period, the number of members receiving twice daily (B.I.D.) medication administration services decreased by 15.6% (from 1,196 in Q1 11 to 1,009 in Q4 13). Recommendations 1. Continue to focus on discharge planning We recommend continuing to conduct Connect to Care regional meetings focused on substance abuse services to identify gaps in service at the regional level. Update: RNMs continue to meet and develop relationships with providers of substance abuse services and to identify gaps in service at the regional level. Some of the barriers to care that have been identified during meetings with detox providers are: lack of residential rehab beds and time consuming referral processes for lower levels of care. Providers also reported a desire to view the inpatient dashboard data which includes ALOS and readmission rates for Free-standing and Hospital-based detox programs. VO will work with regional providers, ABH and DMHAS to develop solutions to the problems identified during provider meetings. Activities related to this recommendation are ongoing and a progress update will be described in the next quarterly report. 2. Continue to Coordinate with ABH We will continue to meet monthly with DMHAS and ABH to review OATP outcomes and develop strategies to improve outcomes. We also will continue to hold a monthly ICM strategy meeting with ABH regional managers (and VO ICMs) to ensure that transitions within the substance abuse continuum for our shared members are smooth and timely. Update: We continue to meet with DMHAS and ABH monthly to review OATP outcomes and develop strategies to improve those outcomes. In May, ADRC became an OATP facility and Rushford is currently working towards OATP status. In July, ABH and VO teamed up to revisit OATP detox programs to identify ways to reduce recidivism and increase connections to methadone programs. Efforts by the ABH/DMHAS/VO team reach beyond OATP development. We are interested in understanding better how to improve the transition from detox to methadone maintenance, based on provider feedback. During our two ICM strategy meetings held in April and May, changes to the VO ICM program related to 2014 performance targets were described and possible shifts in roles and responsibilities were discussed. We also viewed readmission rates for OATP and non-oatp programs. The next strategy meeting is scheduled for August 20th and the agenda includes a review of changes based on VO s 2014 PT, detox dashboard, updates on new OATP providers, and barriers to methadone identified by OATP providers. The overarching purpose of this strategy meeting is to improve outcomes for our shared members through coordination, communication, and intervention. Activities related to this recommendation are ongoing and a progress update will be described in the next quarterly report. Utilization Report for Adult Members : Page 12

Home Health Services Utilization Analysis The reported Admissions as well as Admits/1,000 for each benefit group decreased slightly from Q4 13 to Q1 14. B.I.D. & Q.D. Utilization Analysis Note: Utilization analyses of service frequency is based on claims data. Since the claims lag is approximately 90 days, the most recent data available for analysis is from Q4 13. The number of members receiving Home Health services increased by 8.8% from Q1 11 to Q4 13, (4,617 to 5,021). During this same period, the number of members receiving twice daily (B.I.D.) medication administration services decreased by 15.6% (from 1,196 in Q1 11 to 1,009 in Q4 13). The number of once daily (Q.D.) MedAdmin visits increased by 20.5% during this same period (from 1,362 in Q1 11 to 1,641 in Q4 13). Quarterly results for Q.D. and B.I.D. service frequencies are shown in the table below: Members Served Q1 11 (Baseline) Q3 12 Q4 12 Q1 13 Q2 13 Q3 13 Q4 13 Change from Q3 13 to Q4 13 4617 4883 4792 4886 4951 4897 5021 2.5% B.I.D Users 1196 1016 1016 987 985 960 1009 5.1% Q.D. Users 1362 1597 1562 1617 1683 1606 1641 2.2% % at B.I.D 25.9% 20.8% 21.2% 20.2% 19.9% 19.6% 20.1% 0.5% % at Q.D. 29.5% 32.7% 32.6% 33.1% 34.0% 32.8% 32.7% -0.1% Utilization Report for Adult Members : Page 13

From Q3 13 to Q4 13 there was a slight increase in the percentage of members receiving twice daily (B.I.D.) Med- Admin (from 19.6% to 20.1%). During this same period, the total number of active utilizers increased by 2.5% (from 4,897 to 5,021). Six of the fifteen PAR providers had an increase in the percentage of B.I.D. MedAdmin users in Q4 13 as compared to Q3 13. Of these six providers who increased, three remained below the statewide average. Five provider B.I.D. rates were unchanged (less than 0.5% difference from previous quarter). One provider decreased its B.I.D. rates more than two percentage points during this same period. By comparison, five of the fifteen PAR providers had an increase in the percentage of Q.D. MedAdmin users from Q3 13 to Q4 13, while five providers decreased their Q.D. rates during this period. Three of the PAR providers have lower than average utilization rates for both B.I.D. and Q.D., while five others are above the statewide average for both B.I.D. and Q.D. The PAR providers account for approximately 86% of the members receiving medication administration services in Q3 13. These results may indicate the extent to which providers are integrating skills transfer and recovery efforts, and also identify providers who may require additional attention and training. This information is shared with the providers during individual PAR meetings. ED/Inpatient/OBS Utilization Analysis Claims for ED, Observation and Inpatient admissions, both psychiatric and medical, were evaluated for members who had one or more medication administration visits during the period from Q1 11 to Q4 13. No patterns of increased ED or inpatient utilization were identified during this period and, in fact, both saw a slight decrease. The inpatient utilization rate for the 15 PAR providers was 10.4% in Q1 11 and 9.6% in Q4 13. This is the lowest inpatient rate seen since the start of the contract in 2011. Across the 12 measured quarters, inpatient utilization rates ranged from a low of 9.6% in Q4 13 to a high of 11.6% in Q2 12. The ED utilization rate for the 15 PAR Providers was 28.5% in Q1 11 and 27.7% in Q4 13. This is the lowest ED rate seen since Q2 11. Across the 12 measured quarters, ED Utilization rates ranged from a low of 27.3% in Q2 11 to a high of 30.1% in Q3 12. While some providers had variation in ED rates from quarter to quarter, no provider showed a clearly trending increase or decrease in either Inpatient or ED utilization between Q1 11 and Q4 13. The reduction in B.I.D. services has not resulted in an increase in inpatient or ED utilization, and rates for inpatient and ED use have decreased slightly in the most recent quarter. 23 Hour Observation Bed rates have also remained fairly stable across the 12 measured quarters with a slight increase in the most recent measured quarter. In Q1 11 the 23 hour observation rate for the 15 PAR providers was 1.7%, while in Q4 13 it was 3.0%. Prior to Q4 Utilization Report for Adult Members : Page 14

13, the highest observation rate recorded was 2.3% which was reported in quarter 4 of 2012 and quarters 1 and 3 of 2013. We will continue to monitor OBS utilization to determine if there is an ongoing trend. Conclusions Current home health services data indicates that improvements made in the B.I.D. rates are holding steady, with most PAR providers demonstrating minimal changes from the previous quarter. The reduction in B.I.D. services has not resulted in an increase in inpatient or ED utilization, and rates for inpatient and ED use have decreased slightly in the most recent quarter. It is too early to determine if there has been a change in OBS utilization, and we will continue to monitor that level of care. The home health team reviews provider-specific challenges at weekly meetings, in order to develop more effective shaping and teaching strategies. The home health RNM and clinical staff will continue to work with providers to identify and address barriers to increasing member self-reliance in medication administration. Updates to Recommendations from Quarter 1 1. Continue planned focus on claims data analysis. We will continue analysis of the relationship between reduction in Med Admin frequency, re-hospitalization rates, and connection to other community services for members to ensure that further reductions in Med Admin frequency are not causing an increase in utilization of inpatient, OBS and ED services. We will continue cohort tracking of members receiving B.I.D. Med Admin services to refine our knowledge and understanding of utilization patterns. We will continue to engage providers in exploration of the variances in frequency reduction rates and hospitalization/obs and ED rates through semi-annual group and individual PAR meetings with the PAR 15 providers. Update: We have continued the analysis of claims data and cohort tracking, and data has been shared with all 15 PAR providers during individual meetings. The next round of individual PAR meetings will take place in September and October, and will include data from the 1st and 2nd quarters of 2014. 2. Continue and expand provider based recovery and skills transfer training for home health nurses to enhance nursing skills related to motivational interviewing and member centered, goal focused care plan. A. Expanded webinars will be offered on a monthly basis by the home health clinical staff. Topics will include member centered care plans, non-traditional resources, member rights and responsibilities, motivational interviewing techniques and substance abuse. Update: Three webinars were provided this quarter, addressing the Level of Care Guidelines and nontraditional resources. Webinars will continue to be provided monthly at least through the end of 2014. B. Live, on-site training will continue to be offered to home health supervisors/managers to enhance their effectiveness in communicating about recovery and skills transfer with the direct care nurses.update: As anticipated, the need for on-site trainings has decreased with the implementation of monthly webinars as well as individual PAR provider meetings during this period. Clinical staff will continue to make this service available on an asneeded basis. C. Non-PAR providers who have shown consistent growth in the number of utilizers and higher than average Med Admin utilization rates will be targeted for on-site training and encouraged to attend webinars and provider meet- Utilization Report for Adult Members : Page 15

ings. Update: Providers have been identified but individual trainings were not scheduled during this period. Several non-par providers have participated in the monthly webinars. Clinical staff will continue to address utilization concerns with providers on an on-going basis during the review process. D. Target providers for training based on PAR data and reviewer findings, with a focus on providers whose frequency of visits has increased or remains above the statewide average. Update: During this quarter, one PAR provider with B.I.D. and Q.D. utilization that is consistently above average received on-site training on specific recovery and skills transfer that can be used by nursing in more effectively working with their members. The provider serves approximately 150 behavioral health members, and seven direct care nurses were present for the training. Response was positive with some return demonstration of techniques provided by the nurses during this session. Q1 and Q2 2014 data will be evaluated by the RNM and clinical team to determine if there are higher volume providers that are not part of the current PAR 15 and that have utilization patterns that indicate a need for recovery and skills transfer training. 3. Ongoing collaboration with prescribers and provider agencies. A. Provide training regarding new Level of Care guidelines to home health providers through individual PAR meetings. Update: The new Level of Care Guidelines were reviewed with all 15 of the home health PAR providers. No additional training needs are anticipated. B. Provide training regarding new Level of Care guidelines to high volume prescribers and identify opportunities to increase collaborative practices and coordination of care between home health agencies and prescribers. Update: Training on new Level of Care Guidelines to five high volume prescribers at four mental health and substance abuse treatment agencies was completed during this quarter. Prescribers were actively engaged and provided useful feedback regarding home health services and care coordination. C. Present findings from prescriber meetings to the home care providers to identify opportunities to improve coordination of care and enhance member outcomes. Update: Recommendations and observations made by the prescribers were shared with the home health providers at the June 16th, 2014 provider meeting. 4. Implementation of enhanced CCM role. Initiate provider site visits by VO home health Clinical Care Managers (CCMs). Update: CCMs have been working closely with providers to identify opportunities to offer recommendations for additional resources. Site visits have been offered to providers but did not take place during this quarter due to provider resources and scheduling demands. CCMs will continue to work with providers to encourage and facilitate scheduling of onsite visits with the nurses and nursing supervisors to review authorization requests and support development of client centered care plans, and identify opportunities to offer recommendations and referrals for community based and non-traditional services. New Recommendations 5. Re-analysis of PAR program participants to identify new PAR 15 provider group for 2015. Member volume among the original PAR 15 has changed over time, with some providers consistently Utilization Report for Adult Members : Page 16

below the threshold of 75 members. At the same time, other providers not currently in the PAR 15 are now above the 75 member threshold and would benefit from the focused attention of inclusion in the PAR program. 6. Institute home health discharge rounds, to review cases open longer than 9 months and identify potential discharge planning activities to providers. During trainings with providers, we have noted that discharge planning for members receiving med admin services is infrequently addressed by the nurses or supervisors. Training has focused on skills transfer, which does not always result in decreased dependency on med admin services by a home health provider. Opportunity exists for a more robust review of alternative plans for members who are either capable of self-administration, or for whom less costly and/ or less restrictive options may exist. 7. Initiate research pilot with one or more provider agencies to identify characteristics of home health utilizers who have successfully become independent with medication management. One provider, Gentiva, has expressed interest in a data sharing pilot as in a collaborative effort to identify success factors among members who have reduced or eliminated dependence on home health services. The QM department will evaluate the available data and determine how it might be used to improve outcomes on a systemic basis. Methadone Maintenance Utilization Analysis The Admits/1,000 for each benefit group has remained steady from Q4 13 to Q1 14. Along with Outpatient Services, methadone maintenance admissions in the HUSKY D (MLIA) benefit group have increased proportional to the total membership increase, indicating that these two services are the primary need of new members in the HUSKY D (MLIA) population. HUSKY D (MLIA) Admissions have increased by 300 from Q4 13 to Q1 14. Utilization Report for Adult Members : Page 17

Conclusions Members with HUSKY D (MLIA) continue to utilize methadone maintenance at the highest rate when compared to members with HUSKY C and HUSKY A. We were expecting, based on the observed trends in utilization of higher levels of care, to see a drop in per 1,000 measures based on the increase in membership. However, overall admissions increased from Q4 13 to Q1 14 (1,130 to 1,396), resulting in a small increase in Admits/1,000. That result is different than the trends seen for higher levels of care. The total admissions to outpatient care in the HUKSY D (MLIA) population have increased by 25.6% from Q4 13 to Q1 14. Recommendations - See Global Recommendation #5 Gain additional information about members receiving authorization for methadone maintenance during Q1 14. We would like to understand better if the increase in admissions is related to new membership. We also have been working more closely with detox providers, methadone providers, and LogistiCare (transportation ASO) to ensure that members are receiving methadone at the closest clinic. These two potential contributing factors need to be examined further to draw conclusions about methadone maintenance utilization. Outpatient Care Utilization Analysis The Admits/1,000 for each benefit group has remained steady over the past 2 years despite the increase in membership in Q1 14, with the exception of the HUSKY C (LTC Single) benefit group. Several OTP authorizations for the LTC Single population were issued retroactively in Q4 13 and Q1 14, and are expected to reach a natural equilibrium in the near future, similar to those figures of the other benefit groups. The total admissions to outpatient care in the HUSKY D (MLIA) population have increased by 25.6% from Q4 13 to Q1 14. Coupling this increase with 37.5% increase in total MLIA membership, the Admits/1,000 remained relatively unchanged from Q4 13 to Q1 14. Utilization Report for Adult Members : Page 18

Conclusions Outpatient services are utilized far above any other treatment service. In Q1 14, there were 18,676 new registrations for outpatient services (see table in the appendix). This is an increase of almost 3,000 registrations from Q4 13 (see table in appendix section). We will continue to support the use of outpatient services for those members who experience a wide range of behavioral health symptoms and/or problems with addiction. Authorizations for outpatient services are granted when providers register members via VO s Connect system. At this time, trends in utilization are based solely on registrations submitted by providers. Changes made to authorization and billing practices for providers working with members in Long-term Care programs account for the Admits/1,000 peaks in 2013. We expect registrations for HUSKY C (LTC Single) members to stabilize and a new baseline will emerge. Recommendations - See Global Recommendation #5 Gain additional information about members receiving authorization for outpatient during Q1 14. Similar to methadone maintenance registrations, the number of outpatient admissions and Admits/1,000 increased from Q4 13 to Q1 14. It appears that these increases may be due to increases in new membership but more data and further analysis will be required to confirm/disconfirm assumptions. Global Recommendations Each quarterly report includes recommendations specific to each level of care analyzed and global recommendations. The intent in writing Global recommendations is to indirectly or directly respond to system change and plan for system improvement. Below are the recommendations from the previous reports and the updates. New recommendations for this quarter are also proposed. Global Recommendations and Updates 1. Support Regions in the Development of Community Care Team (CCT) Meetings - RNMs will continue to support each region/hospital in the planning and development phase and ICMs will participate in follow-up meetings. Update: RNMs continue to support and assist hospitals in the planning and development of Community Care Teams. In those regions where CCTs have been established, the RNMs and/or ICMs have begun to attend the CCT meetings, and provide member-specific information to inform the discussion. Activities related to this recommendation are ongoing and a progress update will be described in the next quarterly report. 2. Develop a care management team specific to Co-management with CHN - Co-management clinicians will build relationships with CHN nurses and hospital staff to ensure members receive quality care and individualized treatment planning with a goal of safe transitions to community providers. This group will also be responsible for provider outreach related to discharge planning for members admitted to medical floors for detox. Update: A care management team specific to co-management with CHN has been created. This team is comprised of 5 Clinical Care Managers, a Clinical Liaison and a Clinical Supervisor who are in daily communication with CHN to assist in utilization management, discharge planning and connection to care activities. Utilization Report for Adult Members : Page 19

Clinical managers/administrators from CHN and VO meet weekly to review protocols and procedures related to authorizations and shared cases. As the system moves towards an integrated health model, efforts will focus on further developing communication plans and member specific interventions that reflect shared efforts to provide quality care and support for Medicaid members. VO CTBHP created an email box to receive and send co-management referrals in an effort to coordinate in real time with CHN regarding medically and behaviorally complex members. Monthly group supervision meetings are held to maintain consistency during the triage process and to identify ways in which co-management operations can be improved. Activities related to this recommendation have been successfully completed therefore this recommendation will no longer be included in this section. 3. Improve Care Coordination for DDS Clients - RNMs and clinical management understand that planning for members with intellectual disabilities can be challenging from a systems perspective and a clinical perspective. We need to develop a more collaborative relationship with DDS to address this challenge. Update: Building upon our success with the DDS regional office in Region 2, where we have worked with the DDS regional director and Yale to facilitate a collaborative relationship, we will begin to develop collaborative relationships with DDS staff in other regions. The Region 2 RNM has asked the DDS Region 2 director to co-sponsor a Meet & Greet with other DDS regional managers at VO. He has agreed and we will schedule the Meet and Greet in the fall. Activities related to this recommendation are ongoing and a progress update will be described in the next quarterly report. New Global recommendations 4. Increase coordination with CHN Clinical managers/administrators from CHN and VO meet weekly to review protocols and procedures related to authorizations and shared cases. As the system moves towards an integrated health model, efforts will focus on further developing communication plans and member specific interventions that reflect shared efforts to provide quality care and support for Medicaid members. 5. Analyze how increases and decreases in Admit/1,000 measures are related to the newly eligible members - We intend to examine closely the new member experience and thereby gain a better understanding of their effect on utilization. Utilization Report for Adult Members : Page 20

Further Details on Methodology The methodology for membership totals remains unchanged. For the Total Membership counts, each member is only counted once per quarter, even if he/she changes eligibility groups or experiences gaps in eligibility. For instance, if a member changes benefit groups within the quarter, that member is included in the totals for each benefit group, but only once for the total membership. This methodology is referred to in the graphs as Unique Membership. As of Q2 13, dual eligible members, including HUSKY A dual and HUSKY C dual, were removed from all membership and utilization indicators with the exception of home health and mental health group home measures. For all other levels of care, indicators were reported without the duals. However, we will continue to present a total membership table that includes the dually eligible and a membership table that excludes the dually eligible, for comparison purposes. The length of stay calculation is based upon those members who were discharged during the reporting period. This measure includes all days from the beginning of the authorization for that level of care, including those from previous reporting periods, if applicable. The numerator for Admits/1,000 and Days/1,000 are based on the total number of members and associated days in the identified benefit coverage group. Days/1,000 includes service days utilized during the reporting period. The denominator for each per 1,000 calculation is the sum of membership totals in each benefit group, in each month of the reporting period. Chi-squared tests were performed for each eligibility group to look for statistically significant differences between Q1 13 and Q1 14, as well as Q4 13 and Q1 14. Statistical significance was measured at the least restrictive 5% level. Utilization Report for Adult Members : Page 21

Appendix Mental Health Group Home Utilization Report for Adult Members : Page 22

Partial Hospital Programs Intensive Outpatient Ambulatory Detox Utilization Report for Adult Members : Page 23

Tables: Inpatient Psychiatric Utilization Report for Adult Members : Page 24

Tables: Inpatient Detox Hospital Utilization Report for Adult Members : Page 25

Tables: Inpatient Detox Free-Standing Utilization Report for Adult Members : Page 26

Tables: Home Health Services Tables: Methadone Maintenance Tables: Outpatient Care Utilization Report for Adult Members : Page 27

Tables: Mental Health Group Home Utilization Report for Adult Members : Page 28

Tables: Partial Hospital Programs Tables: Intensive Outpatient Tables: Ambulatory Detox Utilization Report for Adult Members : Page 29