529131 research-article2014

HEBXXX10.1177/1090198114529131Health Education & BehaviorCrawford et al.

Article

Community Impact of Pharmacy-Randomized Intervention to Improve Access to Syringes and Services for Injection Drug Users

Health Education & Behavior 2014, Vol. 41(4) 397­–405 © 2014 Society for Public Health Education Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/1090198114529131 heb.sagepub.com

Natalie D. Crawford, PhD1, Silvia Amesty, MD, MPH, MSEd2, Alexis V. Rivera, MPH2, Katherine Harripersaud, MPH2, Alezandria Turner, PhD2, and Crystal M. Fuller, PhD, MPH2

Abstract Objectives. In an effort to reduce HIV transmission among injection drug users (IDUs), New York State deregulated pharmacy syringe sales in 2001 through the Expanded Syringe Access Program by removing the requirement of a prescription. With evidence suggesting pharmacists’ ability to expand their public health role, a structural, pharmacy-based intervention was implemented to determine whether expanding pharmacy practice to include provision of HIV risk reduction and social/ medical services information during the syringe sale would (a) improve pharmacy staff attitudes toward IDUs (b) increase IDU syringe customers, and (c) increase prescription customer base in New York City neighborhoods with high burden of HIV and illegal drug activity. Methods. Pharmacies (n = 88) were randomized into intervention (recruited IDU syringe customers into the study and delivered intervention activities), primary control (recruited IDU syringe customers only) and secondary control (did not recruit IDUs or deliver intervention activities) arms. Results. Pharmacy staff in the intervention versus secondary control pharmacies showed significant decreases in the belief that selling syringes to IDUs causes community loitering. Conclusions. Structural interventions may be optimal approaches for changing normative attitudes about highly stigmatized populations. Keywords community-based participatory research, health behavior, health disparities, HIV/AIDS, substance use Prevention efforts to reduce HIV among high-risk populations have been successful (Crawford & Vlahov, 2010), particularly those aimed at increasing access to sterile needles/ syringes for those who are unable to stop injection drug use (Vlahov, Fuller, Ompad, Galea, & Des Jarlais, 2004). However, public health messages that advocate behavior change, without complimentary attention to the social and environmental factors that may counter behavior change, have had marginal success in reducing HIV in the United States (Friedman et al., 1992). Interventions that target relevant social and environmental factors are typically what researchers refer to as structural or multilevel interventions, where either features of the structural or social environment is the target of the intervention, or both behavioral and structural factors serve as the target (Blankenship, Bray, & Merson, 2000; Blankenship, Friedman, Dworkin, & Mantell, 2006). A growing body of evidence suggests that the most optimal HIV prevention strategies are structural/multilevel intervention designs (Manhart & Holmes, 2005).

One example of a successful structural intervention is the Expanded Syringe Access Program (ESAP) established by New York State in 2001, which allowed syringe sales in pharmacies without a prescription to increase sterile syringe access for injection drug users (IDUs) thereby reducing HIV transmission among IDUs. Evidence from ESAP evaluation studies have indicated success in improving pharmacy access to sterile syringes for IDUs (Deren et al., 2006) as well as other unintended positive effects including syringe disposal (Cleland et al., 2007; Jones & Coffin, 2002) and reduced syringe sharing (Pouget et al., 2005) among IDUs. 1

Georgia State University, Atlanta, GA, USA Columbia University, New York, NY, USA

2

Corresponding Author: Natalie D. Crawford, Division of Epidemiology and Biostatistics, School of Public Health, Georgia State University, 1 Park Place Room 630B, Atlanta, GA 30303, USA. Email: [email protected]

Downloaded from heb.sagepub.com at University of Texas Libraries on January 26, 2015

398

Health Education & Behavior 41(4)

While early studies have noted significantly lower use of pharmacies as a syringe source among Black and Latino/a IDUs compared with their White counterparts (Deren et al., 2006; Des Jarlais, McKnight, & Friedmann, 2002; Fuller et al., 2004), it is possible that this may have been partly because of pharmacy staff’s unfamiliarity with IDU customers, resulting in unwillingness to sell syringes and beliefs that community members would not want to patronize a business that they believed encouraged drug use (Battles, Rowe, Ortega-Peluso, Klein, & Tesoriero, 2009; Linas, Coffin, Backes, & Vlahov, 2000). Several evaluations of New York State’s ESAP showed that participating pharmacists became more supportive over time (Vlahov et al., 2003). Moreover, targeting community and pharmacy staff knowledge and awareness of drug dependence and nonprescription syringe sales in pharmacies improves support of ESAP and attitudes about the program’s community impact on drug use and HIV transmission (Fuller et al., 2007). Herein, we set forth to provide stronger evidence of the impact of ESAP on pharmacy staff and pharmacy practices related to expanded syringe access. To date, this is the first large-scale pharmacy-randomized intervention design that we are aware of to evaluate pharmacy syringe access that is community-based and targeted to Black and Latino IDUs. Previously published data from this study have shown that pharmacy staff who were engaged in offering additional medical and social service referrals to IDUs during the syringe sale were significantly more supportive of ESAP over time compared with pharmacy staff who were not engaged in offering these services (Crawford et al., 2013). In this article, we are taking a closer look at the impact of this intervention (i.e., offering medical/social service referrals during syringe sale to IDUs) on pharmacy staff opinions and attitudes, beyond general support for ESAP. Specifically, we examined pharmacy staff perceptions of ESAPs’ impact on community factors, including improper syringe disposal, loitering, increased illegal drug use and HIV transmission. We hypothesize that pharmacy staff who participate in the intervention will have more positive attitudes about ESAPs’ impact on community factors compared with those who were in the control group over time.

Method This analysis uses data from the Pharmacists as Resources Making Links to Community Services (PHARM-Link) study. The overall objectives of PHARM-Link were to test whether the delivery of HIV prevention information and other medical and social service referrals to IDUs by pharmacy staff (1a) influenced pharmacy staff perceptions of ESAP and IDUs and (b) influenced IDUs’ risk behaviors. Details of the pharmacy population and recruitment procedures have been previously described elsewhere (Rivera et al., 2010) and results of the intervention on IDU outcomes are forthcoming. Below we provide more detailed account of

the study methods that have not been previously published. This study was approved by the institutional review boards at the New York Academy of Medicine and Columbia University Medical Center.

Study Population, Recruitment, and Eligibility Geographic target areas of high drug activity were ethnographically mapped in Upper and Lower Manhattan, Bronx, Brooklyn, and Queens. All pharmacies within these areas were cross-checked against a list of all ESAP-registered pharmacies provided by the New York State Department of Health and visited by research staff to confirm that they were still participating in ESAP at study start-up. If pharmacies were no longer in business or no longer ESAP-registered they were removed from the list. Once a complete list of pharmacies was compiled, we performed screener phone surveys to determine study eligibility. Eligible pharmacies reported (a) at least one new nonprescription syringe customer each month, (b) at least one new syringe customer who becomes a regular customer each month, (c) willingness to sell syringes to IDUs without additional requirements other than those specified in the law (i.e., require proof of age ≥18 years). Of 325 pharmacies that were screened for study participation, 172 were eligible, 42 did not participate (11 declined participation following screener, 31 became ineligible following screener); and staff within the remaining 130 pharmacies completed a baseline survey. Of those, all pharmacy staff (i.e., pharmacists, pharmacy technicians, and clerks) in 88 pharmacies (67%) agreed to be randomized into one of three arms by New York City borough: intervention (n = 26; recruited IDU syringe customers into the study and delivered intervention activities), primary control (n = 29; recruited IDU syringe customers only) and secondary control (n = 33; did not enroll IDU syringe customers) group. The 42 pharmacies that declined randomization were uncertain of their ability to satisfactorily adhere to the research protocol. Consent from all pharmacy staff in each pharmacy was needed for pharmacy participation and informed consent was obtained from each member of the pharmacy staff. Although most pharmacies remained in the study over follow-up, there were slightly fewer pharmacies in the primary control arm; however, pharmacy and pharmacy staff characteristics did not differ by attrition. In the intervention group there were 18 pharmacies (69.2% follow-up rate) at the 6-month follow-up and 20 pharmacies (76.9% follow-up rate) at the 12-month follow-up. In the primary control group there were 21 pharmacies (72.4% follow-up rate) at the 6-month follow-up and 19 pharmacies (65.5% follow-up rate) at the 12-month follow-up. And in the secondary control group there were 27 pharmacies (81.8% follow-up rate) at the 6- and 12-month follow-up. We compared pharmacy staff characteristics (race, gender, position, pharmacy area) across the follow-up visits and found no statistical differences in attrition (p > .05).

Downloaded from heb.sagepub.com at University of Texas Libraries on January 26, 2015

399

Crawford et al.

Pharmacy Intervention The pharmacy staff intervention included a harm reduction training series which included effective strategies on how to (a) engage IDU customers in conversations while creating a comfortable and nonjudgmental environment, (b) provide referrals for service organizations useful for IDUs, and (c) inform IDU customers about our study and schedule enrollment appointments. In addition, pharmacy staff were provided information on use of pharmacy staff as public health providers, the scope of HIV and drug abuse in the community, pharmacy syringe access and the need for extended services from pharmacies, understanding injection drug use and chemical dependency, and relevant research articles and information on drug treatment modalities. The training series included a group pharmacy staff training seminar and a series of individual trainings.

Group Training The group pharmacy training was an off-site seminar developed in partnership with the Pharmacists Society of the State of New York and pharmacists received continuing education credits for participating in the trainings. The training was led by members of our Pharmacy Working Group composed of researchers, pharmacists, community members, New York State Department of Health professionals, and physicians with experience working with IDUs and/or HIV positive populations. A pharmacy training manual grounded within a harm reduction framework was developed by the Pharmacy Working Group and provided to all pharmacy staff in the intervention arm. In addition, a 10-minute video developed by the research team and the Pharmacy Working Group which depicted appropriate interactions with IDU syringe customers including providing social service referrals and study enrollment activities was shown and discussed with intervention pharmacy staff. All pharmacy staff from each pharmacy attended one group training which took place on an evening agreed upon by the participating pharmacists. A total of four evening training sessions were conducted with staff from approximately six pharmacies in attendance.

Individual Training In order to supplement the group pharmacy training, individual one-on-one trainings between each pharmacy staff member and a research staff member were conducted to (a) reiterate the overall goals of the group pharmacy training, (b) role-play an engaging interaction with an IDU customer, and (c) practice intervention activities. One-on-one trainings were conducted quarterly, when new staff were hired, or when protocol violations were observed. For example, quality control syringe “test buys” were conducted quarterly by a research staff member unknown to the pharmacy staff to ensure that pharmacy staff were adhering to the ESAP

guidelines (New York State Department of Health, 2010) and the research protocol. When a test-buy revealed that study procedures were not being adhered to (i.e., failure to provide required ESAP insert or IDU referral into the study), all members of the staff were retrained. If a second violation occurred, the supervising pharmacist was notified that the pharmacy was no longer eligible to participate in the study. Only two pharmacies (one from the intervention and one from the primary control) were removed. To facilitate pharmacy staff social and medical service referrals, we provided safe injection packets to intervention pharmacies to distribute to their IDU syringe customers which included booklets on local service providers, drug treatment options, safe injection tips, disposal options, HIV and hepatitis C testing and treatment information, as well as condoms, personal sharps disposal containers, alcohol prep pads, and hand sanitizer to promote safe use and disposal of syringes. Pharmacies that were randomized to the primary control group did not participate in the complete harm reduction training series nor extend additional services to IDUs. However, they were trained on how to engage their IDU syringe customers, offer enrollment into our study, and schedule an appointment if they were interested. Given the potential for the primary control group to develop more positive attitudes as a result of their training thereby dampening the potential for an intervention effect, if one exists, we included a secondary control group of pharmacy staff. Those randomized to this condition received no research training activities or additional contact from research staff with the exception of the follow-up surveys.

Data Collection Following the baseline survey, pharmacy staff in the intervention and control pharmacies completed 6- and 12-month follow-up surveys using computer-assisted personal interviews, all taking approximately 40 minutes to complete. The baseline survey instrument, which assessed pharmacy staff beliefs about nonprescription syringe sales, syringe (ESAP) customer sales, beliefs about other public health services being offered in the pharmacy, and the role of syringe sales on HIV transmission can be found online (Rivera et al., 2010). The 6- and 12-month surveys also ascertained pharmacy staff beliefs about syringe sales, ESAP customers, and public health services to determine the effect of the pharmacy intervention on the following outcomes: negative beliefs about IDU syringe sales (yes/no), and number of prescription and nonprescription customers (continuous).

Variable Definitions and Analysis In this analysis, we assessed whether or not the intervention influenced pharmacy staff beliefs that selling syringes to IDUs (a) influenced loss of business, (b) resulted in crime or

Downloaded from heb.sagepub.com at University of Texas Libraries on January 26, 2015

400

Health Education & Behavior 41(4)

Table 1.  Differences in Pharmacy-Level (n = 88) and Pharmacy Staff (n = 383) Characteristics by Study Arm, PHARM-Link 2009-2011.   Characteristics Pharmacy characteristics   Pharmacy type   Independent   Chain  Borough   Bronx   Brooklyn   Queens   Upper Manhattan   Lower Manhattan Individual characteristics  Position   Pharmacist   Technician  Gender   Female   Male  Race   Black   White   Hispanic   Other   Perceived neighborhood drug level   High   Moderate   Low

Study Arm, n (%) Intervention

Primary Control

Secondary Control

p Value

18 (69.23) 8 (30.77)

18 (62.07) 11 (37.93)

17 (51.52) 16 (48.48)

.3741  

6 (23.08) 5 (19.23) 3 (11.54) 3 (11.54) 9 (34.62)

7 (24.14) 7 (24.14) 4 (13.79) 5 (17.24) 6 (20.69)

9 (27.27) 6 (18.18) 3 (9.09) 7 (21.21) 8 (24.24)

.9632a          

53 (40.15) 79 (59.85)

53 (40.46) 78 (59.54)

46 (38.33) 74 (61.67)

.9341  

71 (53.79) 61 (46.21)

87 (66.41) 44 (33.59)

72 (60.00) 48 (40.00)

.1126  

10 (7.63) 22 (16.79) 57 (43.51) 42 (32.06)

34 (25.95) 18 (13.74) 44 (33.59) 35 (26.72)

22 (18.49) 9 (7.56) 36 (30.25) 52 (43.70)

.0002      

64 (57.14) 35 (31.25) 13 (11.61)

79 (68.70) 26 (22.61) 10 (8.70)

56 (54.37) 37 (35.92) 10 (9.71)

.1988    

Note. PHARM-Link = Pharmacists as Resources Making Links to Community Services. a Fisher’s exact test used.

theft by syringe customers, (c) caused improper disposal of syringes near stores, (d) caused the community to be littered with dirty syringes, (e) caused IDUs to increase drug use, (f) encouraged drug use, and (g) reduced transmission of HIV among IDUs. All responses to the measures about pharmacy staff beliefs on selling syringes to IDUs were yes/no. We also assessed whether the intervention influenced pharmacy staff reports of the number of syringe sales in the past week (continuous) and the average number of syringe customers who patronized the pharmacy in the past month. We present comparisons of trends over time by study arm and differences at 12-month follow-up between study arms using Cochran–Armitage test for trend, and chi-square test and/or Fisher’s exact test, respectively. When significant differences in pharmacy staff attitudes, syringe sales, and customer base were present postintervention (12-month visit), we estimated the effect of the intervention at the 12-month visit accounting for the hierarchical structure of pharmacy staff clustered within pharmacies and baseline differences

using a multilevel model. We present a stepwise model that compares the support levels between the intervention and the primary control group, and between the intervention and the secondary control group accounting for clustering of pharmacy staff within pharmacies (Model 1), accounting for baseline differences (Model 2) and controlling for significant (p < .05) pharmacy characteristics (Model 3). Log-binomial regression was used to estimate the prevalence ratio and a robust error variance was specified to account for the small number of pharmacy staff within pharmacies and directly estimate the standard error. All data management and analyses were performed using SAS version 9.2 (SAS Institute Inc., 2008).

Results Characteristics of the pharmacy and pharmacy staff by study arm are shown in Table 1. At the pharmacy level, there were no differences between the intervention and control groups

Downloaded from heb.sagepub.com at University of Texas Libraries on January 26, 2015

401

Crawford et al.

Table 2.  Pre- and Postintervention Pharmacy Staff Beliefs About Selling Syringes to Drug Users and Number of Syringe Sales and ESAP Customers by Study Arm, PHARM-Link 2009-2011. Study Arm  

Baseline

Beliefs Business loss  Intervention   Primary control   Secondary control Crime  Intervention   Primary control   Secondary control Improper syringe disposal  Intervention   Primary control   Secondary control Community loitering  Intervention   Primary control   Secondary control Encourages drug use  Intervention   Primary control   Secondary control Reduces HIV transmission  Intervention   Primary control   Secondary control

6-Month

12-Month

Percentage

p Value Trend

I versus Pa .6392

0.80 0.77 0.87

2.47 1.16 2.30

5.41 3.13 2.60

.0234 .1102 .1770

12.40 15.08 19.47

0 4.65 8.33

6.76 7.94 6.49

.0319 .0288 .0026

1.000

1.000    

5.60 2.31 6.90

3.70 4.65 7.06

4.00 4.69 3.90

.2807 .1708 .2127

1.000

1.000    

36.67 37.93 36.27

30.67 28.40 26.51

15.49 32.81 29.33

.0012 .1873 .1405

.0624

.0491    

38.84 31.97 29.20

27.63 31.17 26.74

22.97 22.22 26.67

.0081 .0997 .3422

.6519

.5911    

82.40 77.52 81.03

88.89 83.13 88.24

89.19 92.19 86.84

.0741 .0060 .1126

.7201

.3651    

Median (Range) Prescription syringes sold/week  Intervention   Primary control   Secondary control ESAP syringes sold/week  Intervention   Primary control   Secondary control ESAP customers/month  Intervention   Primary control   Secondary control

I versus Sb

500 (200-1,000) 500 (300-1,000) 500 (300-1,000) 20 (5-50) 20 (10-40) 20 (10-30) 10 (5-20) 10 (6-20) 10 (5-20)



300 (100-700) 500 (100-800) 300 (100-600) 10 (3-30) 13.5 (3-25) 15 (4-30) 15 (6-30) 20 (8-50) 15 (6-50)

.6799    

200 (0.5-500) 300 (10-500) 150 (3-500)

Community Impact of Pharmacy-Randomized Intervention to Improve Access to Syringes and Services for Injection Drug Users.

In an effort to reduce HIV transmission among injection drug users (IDUs), New York State deregulated pharmacy syringe sales in 2001 through the Expan...
312KB Sizes 0 Downloads 3 Views