Pharmacist-initiated adherence promotion activities for persons living with HIV in ambulatory care settings: Instrument development and initial psychometric testing

Research in Social and Administrative Pharmacy xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Research in Social and Administrative Pharmacy journal homepage: www.elsevier.com/locate/rsap

Pharmacist-initiated adherence promotion activities for persons living with HIV in ambulatory care settings: Instrument development and initial psychometric testing Jennifer Kibichoa,∗, Thomas Dilworthb, Jill Owczarzakc, Florine Ndakuyaa a

College of Nursing, University of Wisconsin-Milwaukee, P O Box 413, Milwaukee, WI, USA Department of Pharmacy Services, Aurora St. Luke's Medical Center, Milwaukee, WI, USA c Bloomberg School of Public Health, John Hopkins University, Baltimore, MD, USA b

A R T I C LE I N FO

A B S T R A C T

Keywords: Pharmacists Patients HIV Adherence promotion activities Instrument development Psychometrics

Objectives: Consistent adherence to antiretroviral therapy (ART) remains a challenge for half the people living with HIV (PLWH) in the U.S. Pharmacists have the expertise in pharmacology and pharmacotherapeutics to manage ART and optimize PLWH outcomes. We developed and validated the psychometric properties of a scale to measure the breadth and depth of adherence promotion activities provided by pharmacists to PLWH in ambulatory care settings. Methods: An initial 37-item instrument was developed from 31 pharmacists' interviews and a comprehensive literature review. Psychometric properties were assessed from responses to questionnaires of 10 content experts and 184 ambulatory pharmacists in 37 States. Psychometric tests included: content validity index (CVI); Exploratory Factor Analysis (EFA); and internal reliability using Cronbach's alpha (α). Results: 26 items were eliminated in the item reduction stage. The final 11-item adherence promotion activities scale (APAS) was a single factor with high loadings (0.51–0.85), good internal consistency (α ≤ 0.93) and an explained variance of 60%. For known-groups validity, HIV-certified pharmacists had comparatively higher and statistically significant APAS score compared to non-certified pharmacists (4.00, p < .001). Conclusions: Preliminary psychometric testing—factor analysis, and high internal consistency—depict that APAS can be useful in scientific research and pharmacy practice to assess the nature and range of pharmacists’ aboveand-beyond prescription services in real-world ambulatory settings. Further validation work is needed to establish conclusive reliability and validity of the newly developed scale.

Introduction Significant advancements in antiretroviral therapy (ART) have simplified drug regimens and reduced pill burden, yet adequate ART adherence is challenging for some persons living with HIV (PLWH). Contemporary estimates from the Centers from Disease Control and Prevention suggest approximately 50% of HIV infected persons in the United States have not achieved viral suppression.1 While not all due to ART non-adherence, these data suggest an opportunity to identify and mitigate barriers to ART adherence when appropriate. Inconsistent adherence is associated with poor clinical outcomes, such as inadequate viral suppression, increased HIV replication, development of resistant virus, treatment failure, and disease progression to AIDS.2–5 While clinical trials have reported that 60–90% of participants have achieved undetectable viral loads, these same rates have not been observed in

∗

clinical practice.2,6–8 Pharmacists can play an important role in increasing adherence to ART for PLWH in ambulatory practice settings. A pharmacist is wellpositioned to reinforce adherence messages PLWH receive from other healthcare providers (e.g., physicians, nurse practitioners) because they are the last healthcare professional in contact with a patient before they take their medications.9–11 Pharmacists have expertise in the pharmacology and therapeutic effects of different drug regimens and may have access to a patient's medication history of prescriptions filled from different providers. For these reasons, pharmacists can proactively identify, prevent and resolve some of the barriers that PLWH face while attempting to adhere to ART including, but not limited to, medicationrelated barriers due to side effects, drug interactions, and ART access. Pharmacists can also provide ART and disease state counseling, assess patient ART adherence, work with patients to develop ART adherence

Corresponding author. E-mail address: [email protected] (J. Kibicho).

https://doi.org/10.1016/j.sapharm.2019.09.062 Received 27 April 2019; Received in revised form 19 August 2019; Accepted 26 September 2019 1551-7411/ © 2019 Published by Elsevier Inc.

Please cite this article as: Jennifer Kibicho, et al., Research in Social and Administrative Pharmacy, https://doi.org/10.1016/j.sapharm.2019.09.062

Research in Social and Administrative Pharmacy xxx (xxxx) xxx–xxx

J. Kibicho, et al.

literature 20–28, we developed the initial 37-item APA instrument. We asked pharmacists to indicate on a 5-point Likert scale how often (never to always) in the previous 30 days they had provided APA to PLWH: 1) starting ART (7 items); 2) established on ART (20 items); and 3) in general (10 items). The list of 37 items are included in Appendix A.

reminders (e.g. pillboxes, cell phone alarms) and schedule automatic ART refills to prevent unnecessary ART interruption. Pilot studies have concluded that pharmacist-implemented adherence promotion activities can save scarce healthcare dollars, prevent unnecessary hospitalizations and physician visits, and optimize health outcomes while providing high-quality care.2–5,12–16 Despite numerous studies that have demonstrated the efficacy of pharmacists' interventions, the role pharmacists play in promoting adherence in real world pharmacy settings remains understudied. Our study filled this gap in the literature by developing and validating an instrument to measure the range of pharmacist-initiated adherence promotion activities for PLWH. We defined adherence promotion activities (APA) as pharmacist-initiated services that are above and beyond routinely-provided and mandated prescription-dispensing services. The purpose of our study is to report the development of a pharmacy adherence assessment instrument that measures the breadth and depth of APA services provided by pharmacists to PLWH in ambulatory care settings and evaluate the instrument's psychometric properties.

Phase II: item reduction Item reduction was conducted in three stages: 1) survey pretesting with 10 key informants and 10 pilot participants; 2) content validation by 10 subject experts; and 3) item analysis using quantitative data from 184 pharmacists. Key informants received $75 gift card for participating in the study while all other study pharmacists received a $50 gift card. The research was approved by the University of WisconsinMilwaukee Institutional Review Board (IRB). Pretesting of Questionnaire: We recruited 10 key informants with expertise in pharmaceutical care of the PLWH population to review the adequacy of the questionnaire vis-à-vis study objectives, and to provide feedback on its readability, organization, appropriateness, face validity, and logical flow.29 We refined the instrument based on key informant feedback and recommendations. Pilot testing was undertaken with a convenience sample of 10 ambulatory-based pharmacists providing care to PLWH in order to test individual survey questions for clarity and interpretation, and the overall questionnaire for relevance, flow, and arrangement.30 Content Validation: Ten content experts agreed to participate in the content validation of the APA. Ten content experts validated the initial 37-item APA instrument. Seven experts were drawn from academia and three were in clinical pharmacy practice. Three had PhDs, six had PharmD, and one had a BS in Pharmacy. Most of the content experts (70%) had either a HIV-certification or completed a HIV and/or infectious disease residency/fellowship or both. We asked these content experts to rate each individual item on a 4-point Likert Scale (not relevant,1 item needs some revision,2 relevant but could be improved,3 very relevant4). We used a 4-point Likert Scale to avoid having a neutral and ambivalent midpoint.31,32 We also asked the content experts to indicate whether all the items that form the APA scale work together.31,32 Then we asked the content experts to indicate, on a 4-point Likert scale (strongly disagree1 to strongly agree4), the appropriateness of the total scale, including vocabulary, concept measured, and relevance. A text box was provided to allow content experts to provide additional feedback, including identifying any important items that may have been omitted in the survey.32,33 We evaluated the content relevance of the APA in a two-step process by calculating the content validity index (CVI) for each item (I-CVI), and the whole scale (S-CVI). The I-CVI is the proportion of experts who

Methods The study was conducted in three phases (Fig. 1): Phase I: item generation In Phase I, we recruited a purposive sample of 31 pharmacists—28 community-based pharmacists and three mail-order pharmacists—in four Midwestern cities in the United States (Chicago, IL; Minneapolis, MN; Kansas City, MO; and Columbus, OH). The pharmacists worked in different pharmacy practices: specialty-only and non-specialty which represent the range of pharmacy settings where PLWH fill their ART. We conducted individual in-depth qualitative interviews with these pharmacists in order to identify pharmacist procedures for providing care to PLWH, including those newly diagnosed, those stabilized on therapy, and those struggling with adherence. Findings from this preliminary study are reported in other publications.17–19 The literature review was conducted by the first and last authors. The following databases were searched: PubMed, CINAHL, PsycINFO, Medline, Cochrane and Google Scholar using key words including pharmacy, pharmacists, interventions, HIV, ART, adherence, compliance, physician collaboration. We also conducted manual searches of reference lists to identify English language publications that examined pharmacists' interventions to promote adherence to ART. We excluded articles that did not contain full text, were not in the English language and did not specifically examine pharmacists’ interventions. Based on findings from our preliminary study and a review of the

Fig. 1. Overview of study procedures. 2

Research in Social and Administrative Pharmacy xxx (xxxx) xxx–xxx

J. Kibicho, et al.

cut-off of 0.70 is deemed acceptable for new scales.29,37 The overall score of the final APA scale was computed by summing the scores from all the items to produce a score ranging from 11 to 55. Higher values of APA score depict more adherence promotion activities. The APA score mean and standard deviation (SD) was calculated to determine whether adequate variability and symmetry in score distributions was achieved.35 We examined the skewness and kurtosis of the APA score to see if it is within acceptable range ( ± 1).

rate an item as content valid (a rating of relevant but could be improved3 or very relevant4).31,34 Based on I-CVI criteria we deleted lowrated items.31 We calculated the S-CVI as the average I-CVI across items (S-CVI/Ave). The S-CVI is the proportion of items on a scale that achieved a relevance rating of 3 or 4 by all the experts.31 A scale that has excellent content validity should be composed of items with I-CVIs that meet the criteria that Lynn (1986) and Polit et al. (2007) recommended: I-CVIs ≥ 0.78 and S-CVI/Ave ≥ 0.90.31,34 CVI was done in Microsoft Excel 2016. Item Analysis: We recruited two hundred and twenty-five ambulatory pharmacists to participate in the survey through email invitations using member email lists from the American Academy of HIV Medicine, the Society of Infectious Diseases Pharmacists and Healthcare Data Solutions, and through a snowballing referral process. We sent out 444 emails to members of American Academy of HIV Medicine (AAHIVM) pharmacist listserv. 197 participants used the email links sent to complete the survey and 70 participants used the anonymous links to complete the survey. Eligibility criteria include: licensed/board certified, work in ambulatory pharmacy settings and providing care to at least 10 PLWH filling prescriptions in their pharmacy each month and/ or filling at least 10 ART prescriptions each month. We sent unique links for participants to complete the questionnaire. The survey questions were displayed one question at a time. Participants could go back and revise an answer to the questions, and had to answer all questions as applicable to complete the questionnaire. We followed up via email with individuals who had either not started the questionnaire or had started and not completed the questionnaire. We closed the study after 225 participants who met eligibility criteria had completed the questionnaire. The participants took approximately 1 h to complete the questionnaire. Pharmacists consented to the study before participating in an online Qualtrics survey. We used item analysis to identify the initial set of items for inclusion in the APA scale. First, the inter-item correlation matrix using Pearson Correlation coefficient was examined to identify items that were highly correlated. Because highly correlated items can lead to the problem of multicollinearity, we retained only one item for each highly correlated pair with interitem correlations of ≥ 0.75. The first author examined each pair of highly correlated items and chose which item to retain based on which item better captured the concept of interest. For example, we removed item #7 “Assess patient's non-HIV related laboratory values (e.g., serum creatinine, liver function tests)” which was highly correlated to item #30 “Assess patient's HIV-related laboratory values (e.g., CD4+ cell counts, viral load)”. To further validate the results of item analysis, we examined the corrected item-to-total correlations, being the correlation between the item score and the score for remaining items. Items with a corrected item-to-total correlation ≥ 0.5 were retained. The APA scale was revised after item analysis.

Construct validity Construct validity is the degree to which the scale operationalized the APA construct. To test the construct validity of the APA scale, we used the known-groups validity test to determine whether the APA scale could discriminate between pharmacists likely to differ in the level of APA based on HIV certification status.36,45 Known-groups validity provides a measure of construct validity when there is no standard criterion to measure a construct of interest (in our case APA).46 To establish known-groups validity, we examined the APA scores of two different groups that would logically have different levels of adherence promotion activities, to confirm whether the hypothesized difference is reflected in the scores of the two groups. We hypothesized that pharmacists who are HIV-certified would have higher levels of APA compared to their non-HIV-certified counterparts. We used an independent sample t-test to test this hypothesis. SPSS software was used to conduct known-groups validity testing. RESULTS Pharmacists characteristics Two hundred and twenty-five pharmacists completed the online survey. We dropped 41 surveys that did not meet the criteria for inclusion: seven mail-order pharmacists and 34 ambulatory pharmacists who indicated either filling less than 10 ART prescriptions per month or having less than 10 PLWH filling prescriptions in their pharmacy. Of the 267 participants who started the online questionnaire, 3 participants declined to participate, and 39 participants did not complete the questionnaire mostly because they were ineligible to participate (did not provide care to at least 10 PLWH) or chose not to answer all questions. The study was conducted between October and December 2014. Our study response rate was 84% (225 of 267) participants. The final count of 184 surveys (82% of total responses) are included in the validation analysis. Phase I: item generation The APA instrument was constructed from items generated from a literature review and a qualitative study with 31 pharmacists. Results from this formative work have been published elsewhere.17,18 The 37 items that make up the original APA instrument are summarized in Appendix A. Descriptive statistics of the 37 items that formed the initial APA depicted that both the skewness and kurtosis were within acceptable range ( ± 1). There were no missing values in the data. Item generation in Phase I resulted in 37 items.

Phase III: psychometric testing We conducted two types of psychometric tests: 1) exploratory factor analysis; and 2) construct validity. Exploratory factor analysis Exploratory factor analysis (EFA) of the remaining items was done to identify the factor structure of the scale using SPSS software. We used data from the 184 ambulatory pharmacists to conduct the EFA. EFA was conducted using principal axis factoring to identify interpretable and meaningful factors with eigen-values > 1 for direct promax rotation. For a rotated factor to be interpretable several criteria must be met: each item should have only one factor loading above 0.40; each factor should have two or more items loading above 0.40, and all items on one factor should fit together logically.35 We calculated the Cronbach alpha for the APA scale in order to establish reliability—the ability of the instrument to measure the attribute of the APA construct.36 An alpha

Phase II: item reduction Content Validation Thirty-three of the 37 items (89%) in the APA scale had an I-CVI ≥0.80. The S-CVI/Ave for the initial APA scale was 0.88. We dropped 4 items (Item 1–4 listed in Appendix A) that were not content relevant to APA based on the Lynn and Polit criterion of I-CVI > 0.8. The remaining 33 items had an I-CVI of between 0.80 and 1.00, and an S-CVI/ Ave of 0.92. Content validation reduced the APA scale from 37 items to 33 items. 3

Research in Social and Administrative Pharmacy xxx (xxxx) xxx–xxx

J. Kibicho, et al.

and most held non-managerial positions (n = 141, 77%). Participants were distributed throughout the US: North East (21%), Mid-West (23%), West (27%) and South (28%). A total of 15 items were removed from the APA scale after item analysis either because they were highly correlated (n = 8) or had lowcorrected item-total correlation (n = 7). An inspection of the 33-item correlation matrix identified 8 pairs of items that were highly correlated (r > 0.75). Next, we examined the corrected-total correlation matrix for items that did not meet the recommended ≥0.5 threshold. In an iterative process, we removed 7 items that had low-corrected total correlations (≤0.5) from the analysis. The final scale had inter-item correlations from 0.242 to 0.737 and corrected item-total correlations from 0.503 to 0.807. Item analysis further reduced the APA scale from 33-items to 18items.

Table 1 Pharmacist demographics.

Gender Age Group

Ethnicity

Post-licensure experience (years)

Post graduate education HIV Certification Membership in HIV-affiliated organization Part-time or Full-time Job Position

Female Male 17–34 years 35–50 years 51 + years Caucasian Asian African American Other ≤5 years 6–10 years 11–20 years 21 + years Yes No Yes No Yes No Full-time Part-time Manager Non-manager

N = 184

%

115 69 105 59 20 125 31 11 17 80 45 30 29 16 168 102 82 95 89 141 43 50 134

63% 38% 57% 32% 11% 68% 17% 6% 9% 43% 24% 16% 16% 9% 91% 55% 45% 52% 48% 77% 23% 27% 73%

Phase III: psychometric testing Exploratory factor analysis Initial EFA analysis in SPSS software using principal axis factoring resulted in a two-factor solution which met the eigen-value greaterthan-one criterion. An examination of the Scree plot indicated two possible breaks in the eigen-value plot: the one-factor and two-factor solutions that were selected for direct promax rotation. Both a KaiserMeyer-Olkin (KMO) measure of sampling adequacy value of 0.927 and a statistically significant Bartlett's Test of Sphericity suggested that the variables shared enough common variance to render factor analysis appropriate. A large first factor with 9 items (#27-#30; #33-#37) accounted for 59% of the variance and a smaller second factor with only 2 items (#31-#32) accounted for 10% of the variance. However, only the one-factor solution was interpretable and therefore we settled on a onefactor solution. We deleted 7 items: six with factor loading < 0.40 and one item that cross loaded on two factors. The 11-item APA scale had factor loadings ranging from 0.514 to 0.854, and an explained variance of 60%. The Cronbach alpha (α) was 0.93 suggesting high internal consistency. EFA further reduced the APA scale from 18 items to 11 items in a single factor (see items #27 to #37 in Appendix A). We summed the scores for all 11 items in order to generate the APA score. The final APA score ranged from 11 to 55 (mean 44.84, SD 8.84). Seven of the 11 items (highest to lowest—Item 27, 36, 28, 34, 35, 30 and 32) had a mean score above 4, and five items (lowest to highest—Item 30, 33, 37, 29, and 32) had SD below 1 (Table 3).

Item Analysis Data used for item analysis and psychometric testing was drawn from a quantitative study of 184 pharmacists. Tables 1 and 2 summarize the characteristics of the pharmacists and pharmacies represented in the study, respectively. The sample of participants consisted of mostly women (n = 115, 63%) which reflects the percentage of women pharmacists in the US in 2016 which was 63.4%.47 The mean age of participants was 36.4 years (SD = 9.66), which was comparatively lower than the mean average age for pharmacists in the US of 41.9 years in 2016.48 More than half (n = 102, 55%) were HIV-certified. The mean post-licensure experience was 10.2 years (SD = 9.73),

Table 2 Pharmacy characteristics.

Type Metropolitan Statistical Area (MSA) Region

Pharmacy Patient Volume

Pharmacy HIV Patient Volume

Pharmacy Prescription Volume

Pharmacy HIV Prescription Volume

Disease Specialty

Specialty Non-specialty Top 10 HIV prevalence Other MSAs North East Mid-West West South Low: < 200 patients/day Medium: 200≤499 patients/day High: > 500 patients/day Low: 10 < 20 patients/ month Medium: 21≤100 patients/ month High: > 200 patients/ month Low: < 200 scripts/day Medium: 200≤499 scripts/ day High: > 500 scripts/day Low: < 20 scripts/day Medium: 21≤99 scripts/ day High: > 100 scripts/day HIV/Infectious Disease Other specialties Non-specialties

N = 184

%

63 121 72 112 39 43 50 52 95 70

34% 66% 39% 61% 21% 23% 27% 28% 52% 38%

19 19

10% 10%

46

25%

119

65%

60 81

33% 44%

43 20 52

23% 11% 28%

112 125 18 41

61% 68% 10% 22%

Construct validity As hypothesized, the APA score was positively and statistically significantly correlated with HIV certification (r = 0.304, p < .001), suggesting that HIV-certified pharmacists were more likely to provide Table 3 Exploratory factor analysis (EFA): Factor loadings, mean, standard deviation and corrected item total correlations.

Item Item Item Item Item Item Item Item Item Item Item a

27 28 29 30 31 32 33 34 35 36 37

EFA Factor loading

Mean

Std. Deviation

Corrected Item-Total Correlation

0.734 0.832 0.751 0.798 0.514 0.648 0.831 0.771 0.814 0.653 0.834

4.364 4.266 3.761 4.065 3.886 4.043 3.902 4.255 4.158 4.304 3.837

0.895 0.917 1.085 1.248 0.999 1.050 1.233 0.972 0.901 0.920 1.217

0.700 0.798 0.726 0.764 0.503 0.640 0.797 0.735 0.783 0.625 0.807

See Appendix A for statements.

4

Research in Social and Administrative Pharmacy xxx (xxxx) xxx–xxx

J. Kibicho, et al.

pharmacists.52 While metrics have been developed to measure patient satisfaction with pharmacy services 54,55, fewer metrics exist for quantifying pharmacy value-added pharmaceutical care. Outside of Medication Therapy Management (MTM) services for Medicare Part D participants, and specific disease management programs (e.g., diabetes and heart disease), there are no performance indicators that measure the breadth and depth of pharmacist contributions to improvements in health outcomes and reduction in healthcare costs.56 The Ambulatory Care Summit, sponsored by the American Society of Health System Pharmacists, recommended that pharmacists in an inter-professional setting develop and validate metrics to capture pharmacists’ services across the continuum of care.51

APA to PLWH compared to non-HIV-certified pharmacists. To test construct validity, we conducted known-groups validity to assess if the APA scale could discriminate between groups that would be expected to have high versus low levels of APA. We used an independent sample ttest in SPSS to test the hypothesis that the degree of APA varied by HIVcertification status. The Levene test for equality of variances was statistically significant (p < .001) suggesting that equal variances between the HIV-certified and non-HIV-certified pharmacists could not be assumed. Under assumptions of variances not equal, the t-test for equality of means was statistically significant (p < .001) with a mean difference in APA score of 5.39 (standard error = 1.25). The mean APA score for HIV-certified pharmacists (47.24) was higher than the mean APA score for non-HIV-certified pharmacists (41.85). On the other hand, the variability of APA among HIV-certified pharmacists (SD = 0.632) was lower than among non-HIV-certified pharmacists (SD = 1.156).

Limitations Our study findings should be interpreted in light of the following limitations: First, the participant-to-variable ratio for the study of 5:1 (184 participants and 37 items) was lower than the recommended threshold of 10:1.35 Second, because of the cross-sectional design of the study, test-retest reliability and criterion-related validity were not evaluated.36 Third, the survey was self-report, which could lead to social desirability response bias. However, one study found no evidence of social desirability phenomenon in anonymous and confidential pharmacist survey responses.50 Lastly, not all items in the APAS are applicable to all pharmacists and patient populations. For example, item #32 (“pharmacists discuss adherence consideration with patients for new HIV medications within 3 months of starting therapy”) only applies when pharmacists have PLWH starting new therapy. Similarly, item #36 (“conduct individualized counseling when medication overuse or underuse is detected”) may not applicable when patients do not overuse or underuse medications.

Discussion Of the 37 items that were developed through qualitative interviews with pharmacists and an extensive literature review only 11 items remained in the final APA scale and had high internal reliability. Our analysis of known-groups to establish construct validity showed that the APA scores behaved consistently with the hypothesized level of APA between groups based on HIV certification status. This finding is consistent with findings from our previous qualitative study.17,18 Preliminary psychometric testing of APA Scale—factor analysis, internal consistency, composite reliability, content, and construct, validity—suggest that the APA Scale may be a useful measure for assessing the nature and range of pharmacists' services in real-world ambulatory settings. The APA scale has relevance for scientific research and pharmacy practice.50 In scientific research, the APA scale can provide an objective measure of pharmacists provision of adherence promotion activities. In pharmacy practice, the APA scale can be used to assess provision of pharmaceutical care, to provide documentation of pharmacist involvement in improving patient health outcomes, and to evaluate the effectiveness of pharmacists' adherence promotion activities.50 When fully validated, the APA instrument could be used to document pharmacists' value-added patient-centered care, to measure the therapeutic effectiveness of pharmacists’ interventions, and to enhance provider accountability in health care cost containment as conditional for third-party payer reimbursement.35,51 This is important given the evidence suggesting pharmacists can improve ART adherence. In the current era of evidence-based medicine, value-based health care and pay-for-performance, the use of pharmacy quality metrics—that go beyond patient satisfaction and prescription dispensing error rates—is becoming increasingly important.52 Both public and private healthcare plans are incentivizing pharmacies by incorporating value-based quality metrics in order to influence pharmacy-level reimbursement and as a criterion for participation in their networks.53 Quality metrics can help pharmacists improve the quality of their work, as well as provide healthcare stakeholders—patients and third-party payers—with a better appreciation of the value-added role of

Conclusion To the best of our knowledge, this is the first study that develops and validates an instrument to measure the breadth and depth of pharmacists’ adherence promotion activities for PLWH in ambulatory care settings. The results of our study suggest that the final 11-item APA Scale we developed is a unidimensional measure of the range of adherence promotion activities to PLWH that are over and above mandated prescription dispensing services provided by pharmacists in ambulatory care settings. Our study provides preliminary evidence for the reliability and validity of the newly developed APA Scale. Given this is the initial study, more research is needed to establish external validity across broader pharmacist populations over time, and to further test construct validity of the APA scale, before conclusive reliability and validity can be established.57 Funding/support This work was supported by the National Institutes of Mental Health [Grant number R21MH098755, 2013].

Appendix A. Adherence Promotion Activities Questionnaire Likert Scale: Never = 1; rarely = 2; sometimes = 3; often = 4; and always = 5

#

37-ITEM APA INSTRUMENT

1 2 3 4

Removed during CVI (low I-CVI < 0.8) Home visit after initiation of therapy or change in drug therapy Pharmacists advocate for 90 day supply of antiretrovirals with physicians and/or insurance Patients are assigned to pharmacy teams for management of medications The pharmacy makes automatic phone calls to patients to remind them to refill their prescriptions

5

Research in Social and Administrative Pharmacy xxx (xxxx) xxx–xxx

J. Kibicho, et al. Removed during Item Analysis (correlated with another item > 0.75) 5 6 7 8 9 10 11 12 13

Make patient referral(s) for other medical services including mental health and substance abuse as needed Obtain patient's symptoms (e.g., diarrhea, nausea) Assess patient's non-HIV related laboratory values (e.g., serum creatinine, liver function tests) Identify and/or recommend management for adverse drug reactions Identify therapeutic alternatives for non-HIV medications to optimize clinical outcomes Set up individual appointments to discuss medication therapy and identify drug-related problems Consult with the patient's other healthcare providers (e.g., doctors, nurses) on patient's ability to adhere to medications Recommend alternative antiretroviral drug regimen to fit patient's lifestyle and special needs Identify therapeutic alternative according to patient's insurance to optimize medication payment coverage Removed during Item Analysis (low corrected item-total correlation < 0.50)

14 15 16 17 18 19 20

Arrange medication delivery services Assist patients in obtaining affordable HIV medications through drug assistance programs, manufacturer coupon cards, etc. Adjust timing of HIV medications (or other medications) refills to be in sync with one another Telephone call after initiation of therapy or change in drug therapy Follow-up calls during the course of therapy The pharmacist initiates phone calls to patients to remind them to refill their prescriptions Medication therapy modification per protocol (e.g., dosage adjustment or therapeutic interchange) Removed during EFA (factor loading < 0.40)

21 22 23 24 25 26

Monitor patients' HIV-related adherence outcomes to ensure that therapeutic goals are met Provide or suggest the use of medication/pill box Make patient referrals for social services including housing and food assistance Identify HIV-positive peer advocates to assist patients with medication adherence Educate patients about ways to remember to take medications Provide suggestions for medication dosing reminders (e.g., pagers, alarms) Final APA Scale

27 28 29 30 31 32 33 34 35 36 37

Identify patient-specific drug-related problem including drug dosing, drug interactions, etc. Discuss patient's medical history (e.g., other comorbidities like diabetes) and medication history (e.g., current and prior medications) Assess patient's behavioral/social history (e.g., smoking, alcohol, drug use) Assess patient's HIV-related laboratory values (e.g., CD4+ cell counts, viral load) Pharmacists assess adherence considerations with patients for all prescriptions Pharmacists discuss adherence considerations with patients for new HIV medications within 3 months of starting therapy Tailor antiretroviral drug regimen to fit patient lifestyle or special needs Discuss patient care with another healthcare provider (e.g., doctor, nurse) to optimize medication therapy including recommend drug or dose adjustment and manage a side effect. Identify and/or recommend management for medication-related side effects Conduct individualized counseling when medication overuse or underuse is detected Identify patient's desired therapeutic goal for medication therapy

Appendix B. Supplementary data Supplementary data to this article can be found online at https://doi.org/10.1016/j.sapharm.2019.09.062.

12. Chisholm-Burns M, Kim Lee J, Spivey C, et al. US pharmacists' effect as team members on patient care: systematic review and meta-analyses. Med Care. 2010;48:923–933. 13. Simoni JM, Amico KR, Smith L, et al. Antiretroviral adherence interventions: translating research findings to the real world clinic. Curr HIV AIDS Rep. 2010;7:44–51. 14. Simoni J, Frick PA, Pantalone DW, et al. Antiretroviral adherence interventions: a review of current literature and ongoing studies. Top HIV Med. 2003;11:185–198. 15. Cranor CW, Bunting BA, Christensen DB. The Asheville Project: long-term clinical and economic outcomes of a community pharmacy diabetes care program. J Am Pharm Assoc. 2003;43:173–184 Washington,DC : 1996. 16. Horberg MA, Hurley LB, Silverberg MJ, et al. Effect of clinical pharmacists on utilization of and clinical response to antiretroviral therapy. J Acquir Immune Defic Syndr. 2007;44:531–539. https://doi.org/10.1097/QAI.0b013e318031d7cd. 17. Kibicho J, Owczarzak J. Pharmacists' strategies for promoting medication adherence among patients with HIV. J Am Pharm Assoc. 2011;51:746–755. 18. Kibicho J, Owczarzak J. A patient-centered pharmacy services model of HIV patient care in community pharmacy settings: a theoretical and empirical framework. AIDS Patient Care STDS. 2012;26:20–28. 19. Kibicho J, Pinkerton SD, Owczarzak J. Community-based pharmacists' needs for HIVrelated training and experience. J Pharm Pract. 2014;27:369–378. 20. Farris KB, Schopflocher DP. Between intention and behavior: an application of community pharmacists' assessment of pharmaceutical care. Soc Sci Med. 1999;49:55–66. 21. Snyder ME, Zillich AJ, Primack BA, et al. Exploring successful community pharmacist-physician collaborative working relationships using mixed methods. Res Soc Adm Pharm. 2010;6:307–323. 22. Ahmed SI, Hassali MA, Aziz NA. An assessment of the knowledge, attitudes, and risk perceptions of pharmacy students regarding HIV/AIDS. Am J Pharmaceut Educ.

References 1. Center for Disease Control and Prevention. Understanding the HIV Care Continuum. June 2018; June 2018 Washington DC. 2. Simoni J, Amico K, Pearson C, et al. Strategies for promoting adherence to antiretroviral therapy: a review of the literature. Curr Infect Dis Rep. 2008;10:515–521. 3. Osterberg L, Blaschke T. Adherence to medication. N Engl J Med. 2005;353:487–497. 4. Ma A, Chen DM, Chau FM, et al. Improving adherence and clinical outcomes through an HIV pharmacist's interventions. AIDS Care. 2010;22:1189–1194. 5. Rathbun CR, Farmer KC, Stephens JR, et al. Impact of an adherence clinic on behavioral outcomes and virologic response in treatment of HIV infection: a prospective, randomized, controlled pilot study. Clin Ther. 2005;27:199–209. 6. Becker SL, Dezii CM, Burtcel B, Kawabata H, Hodder S. Young HIV-infected adults are at greater risk for medication nonadherence. MedGenMed. 2002;4. 7. Mills EJ. Adherence to antiretroviral therapy in Sub-Saharan Africa and North America: a meta-analysis. J Am Med Assoc: JAMA, J Am Med Assoc. 2006;296:679–690. 8. Nieuwkerk PT, Sprangers MAG, Burger DM, et al. Limited patient Adherence to highly active antiretroviral therapy for HIV-1 infection in an observational cohort study. Arch Intern Med. 2001;161:1962–1968. 9. Bluml BM. Definition of Medication Therapy Management : Development of Professionwide Consensus. Washington, DC: ETATS-UNIS: American Pharmacists Association; 2005. 10. Murray MD, Young J, Hoke S, et al. Pharmacist intervention to improve medication adherence in heart failure. Ann Intern Med. 2007;146:714–725. 11. American Society of Health-System Pharmacists. ASHP statement on the pharmacist's role in the care of patients with HIV infection. Am J Health Syst Pharm. 2003;60:1998–2003.

6

Research in Social and Administrative Pharmacy xxx (xxxx) xxx–xxx

J. Kibicho, et al.

2009;73:15. 23. Katz MD, Draugalis JR, Lai RP. HIV infection and AIDS: attitudes and knowledge of Arizona pharmacists. Ann Pharmacother. 1995;29:1218–1223. 24. Pharmacist US. HIV: A Primer for Pharmacists. Continuing Education Test Questions; 2012. 25. Herbert KE, Urmie JM, Newland BA, et al. Prediction of pharmacist intention to provide Medicare medication therapy management services using the theory of planned behavior. Res Soc Adm Pharm. 2006;2:299–314. 26. Zillich AJ, McDonough RP, Carter BL, et al. Influential characteristics of physician/ pharmacist collaborative relationships. Ann Pharmacother. 2004;38:764–770. 27. WEISS SJ, DAVIS HP. Validity and reliability of the collaborative practice scales. Nurs Res. 1985;34:299–305. 28. Hudmon KS, Prokhorov AV, Corelli RL. Tobacco cessation counseling: pharmacists' opinions and practices. Patient Educ Couns. 2006;61:152–160. 29. DeVon HA, Block ME, Moyle‐Wright P, et al. A psychometric toolbox for testing validity and reliability. J Nurs Scholarsh. 2007;39:155–164. 30. Burns KEA, Duffett M, Kho ME, et al. A guide for the design and conduct of selfadministered surveys of clinicians. CMAJ (Can Med Assoc J). 2008;179:245–252. 31. Polit DF, Beck CT. The content validity index: are you sure you know what's being reported? Critique and recommendations. Res Nurs Health. 2006;29:489–497. 32. Peirce D, Brown J, Corkish V, et al. Instrument validation process: a case study using the paediatric pain knowledge and attitudes questionnaire. J Clin Nurs. 2016;25(11–12):1566–1575. 33. Hallgren KA. Computing inter-rater reliability for observational data: an overview and tutorial. Tutorials Quant Methods Psychol. 2012;8:23. 34. Lynn MR. Determination and quantification of content validity. Nurs Res. 1986;35(6):382–385. 35. MacKeigan LD, Larson LN. Development and validation of an instrument to measure patient satisfaction with pharmacy services. Med Care. 1989:522–536. 36. Baró E, Carulla J, Cassinello J, et al. Development of a new questionnaire to assess

37. 45.

46. 47.

48. 50. 51. 52. 53. 54.

55. 56. 57.

7

patient perceptions of cancer‐related fatigue: item generation and item reduction. Value Health. 2009;12:130–138. DeVellis RF. Scale development: theory and applications (applied social research methods). Soc Res. 2003;900:14929. Baró E, Carulla J, Cassinello J, et al. Psychometric properties of the Perform Questionnaire: a brief scale for assessing patient perceptions of fatigue in cancer. Support Care Cancer. 2011;19:657–666. Davenport SJ, de Morton NA. Clinimetric properties of the de Morton mobility index in healthy, community-dwelling older adults. Arch Phys Med Rehabil. 2011;92:51–58. Bureau of Labor Statistics. Labor Force Statistics from the Current Population Survey. Labor Force Statistics from the Current Population Survey. Bureau of Labor Statistics; 2018. DataUSA. Pharmacists.. Odedina FT, Segal R. Behavioral pharmaceutical care scale for measuring pharmacists' activities. Am J Health Syst Pharm. 1996;53:855–864. Avalere Health LLC. Developing Trends in Delivery and Reimbursement of Pharmacist Services. November 2015; November 2015. Brown LM. Measurement of pharmacy quality metrics at the pharmacy level should be our goal. J Am Pharm Assoc. 2009;49:153. Fong DJ. Quality Measures Metrics: Impact on Community Pharmacy Practice and Business. . Wolters Kluwer Clinical Drug Information; September 18, 2017 2017. Sakharkar P, Bounthavong M, Hirsch JD, et al. Development and validation of PSPSQ 2.0 measuring patient satisfaction with pharmacist services. Res Soc Adm Pharm. 2015;11:487–498. Schommer JC, Kucukarslan SN. Measuring patient satisfaction with pharmaceutical services. Am J Health Syst Pharm. 1997;54:2721–2732. Dodds L. Measuring the quality of pharmacy services: we need to get it right! Lung cancer. 2018;15:05. Benson J, Clark F. A guide for instrument development and validation. Am J Occup Ther. 1982;36:789–800.