Abstract
Purpose: This study evaluated the comparative effectiveness of different pharmacist visit types on reducing readmission rates. Method: A single-center, retrospective cohort study was conducted from January 2015 to July 2017. Patients were 18 years or older with an index heart failure (HF) exacerbation admission. Upon hospital discharge, patients were seen in clinic by a clinical pharmacy specialist (CPS) with collaborative practice agreement (CPA) (High Intensity Bundle), medication therapy management (MTM) pharmacist without CPA (Low Intensity Bundle), or no pharmacist (Standard of Care [SOC]). The primary outcome was 30-day all-cause readmission rate. Secondary outcomes included rate of 30-day HF readmissions and average number of days until readmission in those who were readmitted. Results: Totally, 98 patients were included in the final analysis (35 High Intensity Bundle, 28 Low Intensity Bundle, and 35 SOC). The primary outcome of all-cause readmissions was lower in both the pharmacist groups compared with SOC (CPS 8.6% [3/35] vs SOC 25.7% [9/35], P = 0.046 and MTM 7.1% [2/28] vs SOC 25.7% [9/35], P = 0.057). Incremental differences were seen between visit types for the secondary outcome of 30-day HF readmissions (CPS 2.9% vs MTM 7.1% vs SOC 17.1%, P = 0.039). The average number of days until readmission was longer in the CPS versus the MTM and SOC (26.7 days vs 12.5 days vs 14.1 days, respectively). Conclusion: Post-hospital discharge pharmacist visits were associated with lower 30-day all-cause readmission. In particular, clinic visits with a Higher Intensity Bundle may be more effective in reducing HF readmissions. These exploratory findings warrant further investigation.
Keywords: ambulatory services, cardiovascular, clinical services, disease management, medication therapy management (MTM)
Introduction
Heart failure (HF) accounts for over 1 million patient admissions in the United States annually, resulting in a total Medicare expenditure exceeding $17 billion.1 The prognosis for outpatients with HF has significantly improved in the last 20 years due to advances in medical therapy. However, readmission rates within 30 days of discharge following HF hospitalizations remain higher than 20%.2-6
Studies have shown that interventions performed by pharmacists improve transitional care by reducing readmission rates.7 Bellone et al conducted a retrospective study to determine whether clinical pharmacist interventions under a collaborative practice agreement (CPA) were associated with lower all-cause readmission rates 60 days post discharge. Pharmacist interventions included medication titration, discontinuation/initiation of drug therapy, medication counseling, and laboratory monitoring. Results showed readmission rates were significantly lower in the intervention group compared to the control group (18.2% vs 43.1%, respectively, P = 0.002). Services provided by a medication therapy management (MTM) pharmacist have also resulted in decreased all-cause readmission rates.8 Luder et al assessed the impact of a MTM pharmacist versus usual care, which consisted of either no interventions or a nurse home visit, on 30-day all-cause readmission rates. Interventions performed by pharmacists included medication reconciliation, counseling regarding medications, disease states, and self-care monitoring, prescriber consultation, and referrals. Results showed readmission rates were significantly lower in the intervention group versus the control group (6.9% vs 20%, P = 0.019).
In addition, programs that are designed to target patients, prospectively and in real-time, who are at high risk of readmission using a multidisciplinary approach are sustainable and affordable. A 2013 study by Amarasingham et al incorporated an e-Model that identifies hospitalized HF patients at high risk of readmission and subsequent initiation of a multidisciplinary intervention. By using this targeted approach they were able to decrease the HF readmission rate at a large, urban tertiary care center from 26.2% to 21.2% (P = 0.01). This e-Model was later trademarked as PIECES.9
Although pharmacist involvement in transitions of care has shown benefit either alone or in multidisciplinary interventions, to our knowledge, there are no studies that have evaluated the comparative effectiveness of different pharmacist visit types on reducing readmission rates. This study will evaluate post-discharge care provided by either a clinical pharmacy specialist (CPS) with CPA, MTM pharmacist without CPA, or no pharmacist and their impact on all-cause and HF readmission rates.
Method
Patients
A single-center, retrospective cohort study was conducted from January 1, 2015, to July 31, 2017. Criteria for inclusion were patients with a HF exacerbation index admission and were at least 18 years old. Criteria for exclusion were patients who were identified by the PIECES e-Model as being at high risk for readmission who received an intensive bundle of coordinated multidisciplinary services, had a diagnosis of end-stage renal disease (ESRD), were actively incarcerated, had admission to the inpatient psychiatric facility, were actively pregnant, were enrolled in hospice care, had completed appointments with both a CPS with CPA and MTM pharmacist without CPA post-discharge, or were seen in congestive heart failure (CHF) clinic. Patients were randomly assigned to each bundle. This study was approved by University of Texas at Southwestern Institutional Review Board. Data were extracted from the electronic medical record (EMR).
Description of Pharmacy Services Provided in Each Study Arm
Arm 1 (High Intensity Bundle)
Patients seen by a CPS were managed under a CPA with cardiology physician supervision recognized by the Texas State Board of Pharmacy and were scheduled with the goal to be seen within 10 days after discharge. Management of HF and other chronic disease states were based on the most current national clinical practice guidelines. Patients were randomly identified when admitted in the hospital and were scheduled a post-hospital discharge appointment with the CPS near time of hospital discharge. Patients received a telephone follow-up call to notify of upcoming post-hospital discharge appointment with the CPS. Interventions provided by the CPS with CPA included similar interventions to MTM pharmacists such as providing adherence tools and educating on therapeutic lifestyle changes, medication adherence, and disease states. However, interventions also included ordering referrals (primary care physician [PCP], nutrition, smoking cessation, medication access specialist, and anticoagulation management), making therapeutic medication changes including medication discontinuations and initiations, ordering medication refills and labs, and triaging acute issues by admitting patients directly to the Emergency Department (ED). The CPA also allowed the pharmacist to provide follow-up care during the 21-days post-hospital discharge with a second visit within 7 to 21 days of the initial visit, depending on the acuity of symptoms and patients’ understanding of diet, fluid intake, and medication adherence. Patients were discharged from the CPS clinic if clinically stable with appropriate follow-up.
Arm 2 (Low Intensity Bundle)
Patients were identified during hospital admission when a consult was ordered for pharmacist medication discharge counseling. Patients were then scheduled a post-hospital discharge appointment with the MTM pharmacist with the goal to be seen within 10 days after discharge. Patients were provided information describing MTM services in the hospital and were scheduled to have a telephone or face-to-face visit after the initial hospital encounter. Interventions provided by the MTM pharmacist without CPA included reconciling medications, providing adherence tools, educating on therapeutic lifestyle changes, medication adherence, and disease states, providing information on when to seek ED care, coordinating care, and providing recommendations to physicians to optimize therapy.
Arm 3 (Standard Care)
Patients in the control arm had a pharmacist consult for discharge counseling ordered during the index admission. However, they did not receive a follow-up appointment with either a MTM pharmacist or CPS after discharge due to limited resources.
All patients may have received additional appointments with their PCP and/or other specialty clinics not mentioned in the exclusion criteria.
Outcomes
The primary outcome assessed was the rate of 30-day all-cause readmissions of HF patients seen by a CPS with CPA, MTM pharmacist without CPA, and patients who did not have a pharmacist visit after hospital discharge. Secondary outcomes included rate of 30-day HF readmissions, number of ED visits, average number of days until hospital readmission, adherence to follow-up pharmacist appointments, average number of days from discharge to the first pharmacist follow-up visit, and average number of days from discharge to the first physician follow-up visit.
Statistical Analysis
Descriptive statistics were used to assess the baseline characteristics. Nominal variables of baseline characteristics and readmission rates were evaluated using a chi-square test and/or Fisher exact test when appropriate. Continuous variables of baseline characteristics were evaluated using a Mann–Whitney test except for age in which a t test was used. A Log Rank (Mantel–Cox) analysis was used to assess the primary outcome. For the secondary outcome of HF readmissions, a trended analysis was used to identify if incremental differences existed between groups. An a priori significance level of 0.05 was used. SPSS Version 19 was utilized to assess nominal and continuous variables when applicable.
Results
A total of 683 patients were assessed for eligibility with 98 patients meeting study inclusion criteria (Figure 1). Among the excluded patients, the primary reasons for exclusion were due to enrollment through the PIECES e-Model (n = 264) and scheduled appointment in CHF clinic (n = 127). Among the included patients, 35 were in the CPS with CPA arm, 28 were in the MTM pharmacist without CPA arm, and 35 were in the control arm and had no follow-up visit with a pharmacist. Baseline characteristics were similar except for gender (Table 1). Patients in the control arm had more unfunded patients compared to those in the CPS with CPA or MTM without CPA arms (Table 1).
Figure 1.
Screening and randomization of patients.
Note. CPS = clinical pharmacy specialist; MTM = medication therapy management; CPA = collaborative practice agreement.
Table 1.
Demographics in Study Population.
| Characteristics | Clinical Pharmacy Specialist with CPA (n = 35) | Medication Therapy Management without CPA (n = 28) | Control (n = 35) | Arm 1 vs 3 P value | Arm 2 vs 3 P value |
|---|---|---|---|---|---|
| Average age (years) ± SD | 60 ± 13 | 63 ± 14 | 59 ± 14 | .85 | .27 |
| Gender | |||||
| Male | 37.1% | 53.6% | 65.7% | .02 | .33 |
| Race/ethnicity | .16 | .28 | |||
| African American | 57.1% | 57.1% | 48.6% | ||
| Caucasian | 2.9% | 3.6% | 20.0% | ||
| Hispanic | 37.1% | 35.7% | 28.6% | ||
| Other | 2.9% | 3.6% | 2.9% | ||
| Payer/insurance | .22 | .33 | |||
| Parkland Financial Assistance | 42.9% | 28.6% | 25.7% | ||
| Unfunded | 11.4% | 10.7% | 25.7% | ||
| Medicare/medicaid | 45.7% | 50.0% | 45.7% | ||
| Other | 0.0% | 10.7% | 2.9% | ||
| Pharmacy | .62 | .86 | |||
| Parkland | 60.0% | 67.9% | 65.7% | ||
| Outside | 40.0% | 32.1% | 34.3% | ||
| Comorbidities | |||||
| Diabetes mellitus | 54.3% | 57.1% | 42.9% | .34 | .26 |
| Hypertension | 85.7% | 96.4% | 91.4% | .45 | .42 |
| Hyperlipidemia | 54.3% | 67.9% | 68.6% | .22 | .95 |
| Chronic obstructive pulmonary disease | 20.0% | 14.3% | 14.3% | .53 | 1.00 |
| Chronic kidney disease | 40.0% | 57.1% | 45.7% | .63 | .37 |
| Atherosclerotic cardiovascular disease | 22.9% | 42.9% | 42.9% | .08 | 1.00 |
| Baseline Labsa | |||||
| Ejection fraction (%) | 42 (33, 60) | 43 (35, 60) | 38 (21, 56) | .11 | .15 |
| Creatinine (mg/dL) | 1.1 (0.8, 1.6) | 1.4 (1.1, 2.4) | 1.4 (1.1, 1.7) | .13 | .54 |
| N-terminal prohormone of brain natriuretic peptide (pg/mL) | 2407 (1113, 5755) | 2084 (1182, 4950) | 2236 (1619, 12040) | .46 | .35 |
| New York Heart Association Class | |||||
| I | 14% | 7% | 3% | ||
| I-II | 9% | 7% | 0% | ||
| II | 46% | 7% | 9% | ||
| II-III | 20% | 14% | 17% | ||
| III | 6% | 21% | 37% | ||
| III-IV | 0% | 7% | 11% | ||
| IV | 3% | 4% | 9% | ||
| Unknown | 3% | 32% | 14% | ||
| Type of heart failure | |||||
| Reduced EF | 34% | 32% | 37% | ||
| Preserved EF | 37% | 32% | 34% | ||
| Combined | 26% | 32% | 29% | ||
| Unknown | 3% | 4% | 0% | ||
Note. CPA = collaborative practice agreement; EF = ejection fraction.
Reported as median and interquartile range.
All-cause 30-day readmission rates were lower for patients seen by a pharmacist, whether it be a CPS with CPA or MTM pharmacist without CPA. However, patients seen by a CPS with CPA led to a statistically significant reduction in all-cause 30-day readmission rates compared to standard care (8.6% vs 25.7%, P = 0.046) while patients seen by a MTM pharmacist without CPA did not (7.1% vs 25.7%, P = 0.057) (Figure 2). Secondary outcome of 30-day HF related readmission rates showed incremental differences between visit types (CPS 2.9% vs MTM 7.1% vs control 17.1%, P = 0.039) (Table 2). None of the patients in the control arm had an ED visit within 30 days of hospital discharge. However, ED visits occurred for 5 patients in the CPS arm and 6 patients in the MTM arm (Table 2). The average number of days until readmission was longer in the CPS versus control arms (26.7 days vs 14.1 days, P = < 0.01) but similar in the MTM versus control arms (12.5 days vs 14.1 days, P = 0.82). For patients scheduled to have a second visit with the CPS, adherence to that follow-up appointment was 68%. The average number of days to the first pharmacist follow-up visit was 11 days for both the CPS and MTM pharmacist arms. The average number of days to the first physician follow-up visit was similar among all 3 study arms (Table 2). The types of pharmacist interventions provided are shown in Table 3.
Figure 2.
A log rank (Mantel–Cox) analysis of all-cause 30-day readmission rates by follow-up visit type.
Note. CPS = clinical pharmacy specialist; MTM = medication therapy management.
Table 2.
Primary and Secondary Outcomes.
| Variable | Clinical Pharmacy Specialist with CPA (n = 35) | Medication Therapy Management without CPA (n = 28) | Control (n = 35) | Arm 1 vs 3 P value | Arm 2 vs 3 P value |
|---|---|---|---|---|---|
| 30-day all-cause readmission rates | 8.6% | 7.1% | 25.7% | .046a | .057a |
| Acute HF exacerbation | 2.9% | 7.1% | 17.1% | ||
| COPD exacerbation | 2.9% | 0% | 0% | ||
| Chest pain | 0% | 0% | 5.7% | ||
| Syncope | 2.9% | 0% | 0% | ||
| Supratherapeutic international normalized ratio | 0% | 0% | 2.9% | ||
| 30-day HF readmission rates | 2.9% | 7.1% | 17.1% | .11b | .28b |
| Number of emergency department visits | 5 | 6 | 0 | —c | —c |
| Acute HF exacerbation | 1 | 1 | —d | ||
| COPD exacerbation | 1 | 1 | —d | ||
| Subarachnoid hemorrhage | 0 | 0 | —d | ||
| LifeVest firing | 1 | 0 | —d | ||
| Ear bleeding | 1 | 0 | —d | ||
| Low blood pressure/dizziness | 0 | 0 | —d | ||
| Other | 1 | 4 | —d | ||
| Number of days until readmission (mean ± SD) | 26.7 ± 2.1 | 12.5 ± 7.8 | 14.1 ± 8.8 | < .01 | .82 |
| Adherence to follow-up appointment | 68% (17/25) | 50% (1/2) | —d | —d | —d |
| Number of days to first pharmacist visit (mean ± SD) | 11 ± 5 | 11 ± 6 | —d | —d | —d |
| Number of days to first physician visit (mean ± SD) | 17.4 ± 19 | 20.4 ± 23 | 18.5 ± 18 | —d | —d |
Note. CPA = collaborative practice agreement; HF = heart failure; COPD = chronic obstructive pulmonary disease.
Log rank (Mantel–Cox) analysis.
Fisher exact test.
Not enough information for comparison.
Not applicable.
Table 3.
Pharmacist Interventions for Post-discharge Follow-Up Visits.
| Interventionsa | Clinical Pharmacy Specialist with CPA (n = 35) | Medication Therapy Management without CPA (n = 28) |
|---|---|---|
| Primary care physician referral | 5 | 0 |
| Nutrition referral | 2 | 0 |
| Smoking cessation referral | 5 | 0 |
| Medication access specialist referral | 7 | 0 |
| Anticoagulation clinic management referral | 3 | 0 |
| Medication dose change | 69 | 0 |
| Discontinuation of medication | 7 | 0 |
| Addition of medication | 22 | 0 |
| Labs ordered | 15 | 0 |
| Medication ordered | 58 | 0 |
| Therapeutic lifestyle change education | 69 | 0 |
| Disease state education | 69 | 4 |
| Medication adherence education | 69 | 20 |
| Pill box/water bottle provided for adherence | 12 | 4 |
| Admitted to emergency department | 2 | 0 |
| Provided recommendation to physician to optimize therapy | 0 | 21 |
| Coordinated care | 0 | 9 |
Note. CPA = collaborative practice agreement.
Total from all follow-up visits, patients may have had more than one visit post discharge.
Discussion
This comparative analysis of post-hospital follow-up visits by High or Low Intensity Bundle identified that both visits were associated with lower all-cause readmission when compared to control. The insignificant difference seen for those in the MTM pharmacist without CPA arm may have been due to the smaller sample size. The difference in 30-day readmission rates for those seen by a CPS with CPA could be explained by the scope of the CPA disease states which was HF in this clinic setting. In addition, the CPS with CPA was able to modify therapy whereas the MTM pharmacist without CPA could only make recommendations to physicians to optimize therapy. Post-hospital discharge pharmacist visits, regardless of pharmacist type, decreased readmission rates likely due to decreasing the number of HF exacerbation readmissions.
This positive impact is consistent with previous literature demonstrating pharmacist involvement post-hospital discharge results in reduced HF readmission rates.10,11 Kalista et al found one in-home pharmacist visit followed by two follow-up telephone calls at 1 week and 4 weeks after the initial visit reduced 30-day HF readmissions when compared with the overall visiting nurse association patient population (10% vs 38%). In addition, Jackevicius et al found pharmacist involvement in a multidisciplinary HF post-hospitalization clinic decreased 90-day HF readmissions compared to a control group that did not receive care in the clinic (7.6% vs 23.3%; HR = 0.30, 95% confidence interval [CI] = 0.15-0.59). Results of our prespecified subgroup analysis suggest the type of pharmacist may affect the magnitude of the reductions seen in disease specific HF readmissions. More specifically, patients may have a higher chance of avoiding readmission if there is follow-up with a CPS with CPA compared to follow-up with a MTM pharmacist without CPA or no pharmacist follow-up.
Patients seen by a pharmacist had more ED visits compared to the control group. This is likely due to closer follow-up that provided opportunities for early identification of management needs which may have also contributed to decreased readmissions. In addition, patients in the control arm may have presented to the ED with signs and/or symptoms severe enough to warrant hospital admission rather than discharges from the ED. Higher rates of readmission may have been seen in the control group due to the higher proportion of unfunded patients. Unfunded patients may have had more difficulty in obtaining medications and regular outpatient follow-up compared to funded patients, leading to higher readmission rates. However, in terms of comorbidities, all groups were fairly similar. Patients in the MTM group had more patients with chronic kidney disease (CKD) which may have presented a challenge when titrating and optimizing guideline-based medication therapy for HF.
This study has several limitations, including the retrospective nature of this analysis. The sample size was small, due to a limited number of patients qualifying for study inclusion with no additional confounding factors. The primary exclusion was being enrolled in a multidisciplinary approach for HF patients through the existing transitions of care program at our institution. In addition, hospital ED visits and readmissions to outside facilities were unknown which may have underestimated readmission rates. Furthermore, the reduction in readmission rates cannot be solely attributed to pharmacist interventions as patients may have received additional appointments with their PCP or other specialty clinics.
Conclusions
This retrospective cohort study demonstrated that follow-up appointments with a CPS with CPA or MTM pharmacist without CPA after hospital discharge were associated with lower 30-day all-cause readmission rates. post-discharge follow-up visits with a CPS with CPA may be more effective in reducing HF readmissions compared to visits performed by a MTM pharmacist without CPA. Of those who were readmitted, patients not seen by a CPS appear to be readmitted sooner than those with a follow-up appointment scheduled after hospital discharge. Larger studies are needed to confirm the comparative effectiveness of different visit types by pharmacists on reducing readmission rates.
Footnotes
Declaration of Conflicting Interests: The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding: The author(s) received no financial support for the research, authorship, and/or publication of this article.
References
- 1. Desai AS, Stevenson LW. Rehospitalization for heart failure: predict or prevent? Circulation. 2012;126:501-506. [DOI] [PubMed] [Google Scholar]
- 2. Gheorghiade M, Vaduganathan M, Fonarow GC, Bonow RO. Rehospitalization for heart failure. J Am Coll Cardiol. 2013;4:391-403. [DOI] [PubMed] [Google Scholar]
- 3. Stevenson LW, Pande R. Witness to progress. Circulation. 2011;4:390-392. [DOI] [PubMed] [Google Scholar]
- 4. Krumholz HM, Merrill AR, Schone EM, et al. Patterns of hospital performance in acute myocardial infarction and heart failure 30-day mortality and readmission. Circ Cardiovasc Qual Outcomes. 2009;2:407-413. [DOI] [PubMed] [Google Scholar]
- 5. Joynt KE, Jha AK. Who has higher readmission rates for heart failure, and why? Implications for efforts to improve care using financial incentives. Circ Cardiovasc Qual Outcomes. 2011;4:53-59. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Centers for Medicare & Medicaid Services. Readmissions and Deaths—National. https://data.medicare.gov/Hospital-Compare/Readmissions-and-Deaths-National/qqw3-t4ie. Published April 2017. Accessed Jul 27, 2017.
- 7. Bellone JM, Barner JC, Lopez DA. Postdischarge interventions by pharmacists and impact on hospital readmission rates. J Am Pharm Assoc (2003). 2012;52:358-362. [DOI] [PubMed] [Google Scholar]
- 8. Luder HR, Frede SM, Kirby JA, et al. TransitionRx: impact of community pharmacy postdischarge medication therapy management on hospital readmission rate. J Am Pharm Assoc (2003). 2015;55:246-254. [DOI] [PubMed] [Google Scholar]
- 9. Amarasingham R, Patel PC, Toto K, et al. Allocating scarce resources in real-time to reduce heart failure readmissions: a prospective, controlled study. BMJ Qual Saf. 2013;22:998-1005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Jackevicius CA, de Leon NK, Lu L, et al. Impact of a multidisciplinary heart failure post-hospitalization program on heart failure readmission rates. Ann Pharmacother. 2015;49(11):1189-1196. [DOI] [PubMed] [Google Scholar]
- 11. Kalista T, Lemay V, Cohen L. Postdischarge community pharmacist-provided home services for patients after hospitalization for heart failure. J Am Pharm Assoc (2003). 2015;55(4):438-442. [DOI] [PubMed] [Google Scholar]