Pharmacists' counseling on oral contraceptives: A theory informed analysis

Research in Social and

Administrative Pharmacy j (2015) j–j

Original Research

Pharmacists’ counseling on oral contraceptives:A theory informed analysis

Mohamed E.K. Amin, Ph.D.a,*, Betty Chewning, Ph.D., F.A.Ph.A.baFaculty of Pharmacy, Beirut Arab University, Room 102 Hariri Building, Tareek El Jadida, Beirut, Lebanon

bSonderegger Research Center, School of Pharmacy, University of Wisconsin–Madison, Madison, WI 53705-2222, USA

Abstract

Background: Providing correct and complete counseling on the use of oral contraceptives (OCs) iscentral to securing the autonomy of women in child-bearing age and is a powerful, proven tool of

social change. Pharmacists in many developing countries such as Egypt are involved in dispensing andat times prescribing pharmaceuticals, including oral contraceptives that are readily available without aprescription.

Objectives: To predict Egyptian community pharmacists’ counseling on oral contraceptives while utilizinga theoretical framework guided by the Theory of Planned Behavior (TPB).Methods: A cross-sectional, self-administered survey was completed by a random sample of communitypharmacists in Alexandria, Egypt to determine their attitudes and behaviors regarding counseling on OCs.

Multiple regression was used to predict self-reported counseling on oral contraceptives as a function of theTPB-related constructs and six other factors – “perceived importance of profit from dispensing OCs onpharmacy revenue,” “number of hours worked,” “age,” “gender,” “pharmacy practice degree” and

“marital status” of the pharmacist.Results: Of the 181 pharmacists invited to complete the survey, 168 (93%) participated. Pharmacistsindicated they talked to a slightly higher proportion of women about the importance of taking OCs at the

same time daily than about topics such as which day to start taking OCs, side effects and what to dowhen a dose of OCs was missed. Pharmacists’ reported counseling on oral contraceptives was positivelyassociated with their perception that women welcomed pharmacist initiated OC counseling (b ¼ 0.315,P ! 0.001), perceived adequacy of time available to counsel women on OCs (b ¼ 0.290, P ¼ 0.003) and

the perceived number of women who asked for their help in selecting an OC without providing a pre-scription in the past week (b ¼ 0.160, P ¼ 0.018). Pharmacists reported that women’s welcoming phar-macists initiating OC counseling was associated with the pharmacists’ reported percent who asked

pharmacists for OC advice out of the last 5 women seeking OC (r ¼ 0.45; P ! 0.0001). Male pharmacistswere less likely than female pharmacists to report that women welcomed pharmacist initiated OC coun-seling (r ¼ �0.27; P ¼ 0.0005).

Conclusions: The TPB appears to help predict pharmacists’ OC counseling. There is a need to preparepharmacists who are frequently requested to assist women with the selection of an oral contraceptive.Interventions that would facilitate women’s requests for information may be valuable to increase

pharmacists’ counseling on oral contraceptives. Future qualitative and observational studies are needed

* Corresponding author.

E-mail address: [email protected] (M.E.K. Amin).

1551-7411/$ - see front matter � 2015 Elsevier Inc. All rights reserved.

http://dx.doi.org/10.1016/j.sapharm.2015.08.009

Delta:1_given name

Delta:1_surname

mailto:[email protected]




2 Amin & Chewning / Research in Social and Administrative Pharmacy j (2015) 1–13

to assess complexities in counseling on oral contraceptives in developing countries.� 2015 Elsevier Inc. All rights reserved.

Keywords: Oral contraceptives; Family planning; Contraception; Egypt; Theory of planned behavior; Pharmacist

Introduction

Background

Oral contraceptives (OCs) are highly effectivewhen used properly. Providing correct and

complete counseling on the use of oral contra-ceptives is central to securing the autonomy ofwomen in child-bearing age and has been

proven to be a powerful tool of social change.In recent decades, Egypt, the most populouscountry in the Middle East and the third most

populous country in Africa, has witnessed aconsiderable expansion of family planning usethat was associated with an increased awareness

of family planning methods.1

Pharmacists in many developing countries suchas Egypt dispense, select and at times prescribepharmaceuticals including OCs, which are readily

available without a prescription. They have asignificant potential to ensure the safe use ofOCs where family planning is a national priority.

The 2014 Egypt Demographic and Health Survey(EDHS), the latest in a series of nationallyrepresentative population and health surveys con-

ducted in Egypt, shows that pharmacies were theprincipal source for OCs.1 However, it also showsthat women obtaining their OCs from pharmacieswere much less likely than women obtaining their

OCs from other health facilities to have receivedinformation, especially about side effects, neces-sary to make an informed choice.1

While most of the literature focused on phar-macists’ attitudes and practices related to emer-gency contraception,2–9 a few studies examined

community pharmacists’ counseling on the morecommonly used low dose OCs.7,10–12 Studies thatused simulated patrons to evaluate the quality of

counseling provided to patrons indicate that asmall percentage of pharmacists provided coun-seling on OCs and that even when counselingwas provided, it was missing key information.7,12

Sattari et al’s findings indicate that at times phar-macists were not providing information to pa-trons even when pharmacists possessed the

correct knowledge.10 While these few studies sug-gest pharmacists’ counseling behavior is limited,there is little information on factors associated

with the provision of counseling to women onOCs.

Theoretical framework

The Theory of Planned Behavior (TPB) implies

that individuals carefully consider the availableinformation before acting.13 According to TPB,attitudes (feelings about a behavior), subjective

norms (perception of whether important peopleperform and approve the behavior or not) andperceived behavioral control (perception of thedifficulty of performing a behavior) determine

the individual’s behavioral intention (plan toperform behavior) and consequently determinethe likelihood of the individual carrying out that

specific behavior. A measure of perceived moralobligation (personal feelings of responsibility toperform, or refuse to perform, a certain

behavior)14 could add predictive power to themodels predicting pharmacists’ intentions to carryout behaviors.15,16 As seen in Fig. 1, this study

uses constructs derived from TPB and adaptedto the research question following pre-testing.Consistent with TPB, pharmacists’ counseling isassumed to be under their volitional control.

Perceived Welcome of Counseling (PWC), derivedfrom the subjective norm construct, is defined asthe pharmacist’s perception of the proportion of

women who would welcome the pharmacist initi-ation of a conversation on topics related to oralcontraceptive use. The Perceived Knowledge of

Women (PKW), derived from the attitudeconstruct, is defined as a pharmacist’s perceptionof the proportion of women who possess adequate

knowledge regarding OCs. Perceived Responsibil-ity for Notification (PRN), derived from theperceived moral obligation construct, is definedas the pharmacist’s perception of responsibility

to notify women about four key topics related toOCs. Finally, two measures related to theperceived behavioral control construct were

used. These included the level of confidence inone’s therapeutic knowledge regarding OCs andperceived adequacy of time available to counsel

women on OCs.Consistent with TPB higher rates of counseling

on OCs should be associated with higher rates of


Perceived Knowledge of Women(Attitude)

Perceived Welcome of

Counselling (Subjective Norm)

Perceived Responsibility for

Notification (Perceived Moral Obligation)

Confidence in therapeutic

knowledge regarding OCs

Perceived adequacy of time to counsel women on OCs

Perceived importance of dispensing OCs on pharmacy

revenue

Number of working hours in the pharmacy

Counselling on Oral contraceptives

Pharmacist’s age

Pharmacist’s gender

Pharmacist’s marital status

Pharmacy Practice Degree

Number of women who asked for pharmacist’s help in selecting an OC without providing a prescription

Fig. 1. Theory of planned behavior based theoretical framework.

3Amin & Chewning / Research in Social and Administrative Pharmacy j (2015) 1–13

women welcoming pharmacists’ initiative of OC

consultation, a stronger pharmacist sense of re-sponsibility to provide OC consultation, percep-tion of lower patron knowledge about OCs,

higher pharmacist confidence in their OC knowl-edge and higher perceived adequacy of time to doOC counseling.

Aim of the study

The aim of this study was to predict Egyptian

community pharmacists’ counseling on oral con-traceptives while utilizing a theoretical frame-work, the Theory of Planned Behavior.

Methods

Ethical approval

The High Institute of Public Health Institu-tional Review Board at Alexandria University,Alexandria, Egypt approved the study.

Sample

This study employed a cross-sectional, descrip-tive survey design. The target population con-

sisted of all practicing community pharmacistsin the Alexandria governorate. The Ministry ofHealth (MOH) provided an updated list of 3350



potentially open community pharmacies as thesampling frame. To determine sample size it wasassumed that predictors will explain at least 15%

of the variance in the dependent variable. Analpha of 0.01, power ¼ 0.90 with 13 predictorvariables in the model was specified. Resultsindicated that 138 pharmacists would be needed.

Substantial planning went into identifying andenrolling the sample. To achieve a target sampleof at least 138 pharmacists, consideration was

given to potential pharmacist refusals, closedpharmacies, pharmacies that cannot be located,and absenteeism pharmacies where a non-

pharmacist served patrons.17 Thus, of the 3309community pharmacies in the sampling frame,230 pharmacies were selected using a randomnumber generator. One pharmacist per commu-

nity pharmacy was asked to complete the surveyforms. When more than one pharmacist was inthe pharmacy at the time of the visit, the data col-

lector asked the staff pharmacist who was aboutto interact with the next patron to fill the survey.Pharmacy students, pharmacy interns and non-

pharmacist attendants were excluded.

Study measures

Dependent variableA pharmacist’s reported amount of OC coun-

seling was measured using four items on the

following topics (1) the importance of takingOCs at the same time every day, (2) the appro-priate day to start taking OCs, (3) how to act ifone or more pills are missed and (4) OC side

effects. Pharmacists were asked to recall the lastfive women to whom they dispensed OCs andindicate the number they counseled on each of

these four topics. This included women who didor did not present a prescription for OCs to thepharmacist. A 6-point scale ranging from 0 to 5

was used. To construct the dependent measure,the four numbers reported by pharmacists wereaveraged. This average generated an overall indi-

cator of the proportion of women to which thetotal array of OC consulting topics were coveredby pharmacists (see Table 1).

Independent variablesThe Perceived Responsibility for Notification

(PRN) predictor variable was operationalized asthe pharmacist perception of their responsibilityfor notifying women about OC using the four items

and a 5-point scale ranging from “not at all respon-sible to extremely responsible” The average scorefor the four items was used in calculating the

pharmacist’s mean PRN in the analysis (seeTable 1). The Perceived Knowledge of Women(PKW) predictor variable was operationalized as

a pharmacist’s perception of the proportion ofwomen who possess adequate knowledge regardingthe same four OC topics using a 5-point scaleranging from 0 to 100%. The average score for

the four items was used in calculating the pharma-cist’s mean PKW in the analysis (see Table 1). Simi-larly, pharmacists’ perception of how much women

welcomed pharmacist initiating OC consultation,the Perceived Welcome of Counseling (PWC) wasoperationalized as the pharmacist’s perception of

the proportion of women who would welcome thepharmacist initiation of a conversation about eachof the same four topics and a 5-point scale rangingfrom 0 to 100%. The average score for the four

items was used in calculating the pharmacist’smean PWC in the analysis (see Table 1). Like thedependent variable, these three TPB measures

included women who did or did not present a pre-scription for OCs to the pharmacist.

In addition to the above-mentioned variables

derived from the perception and subjective norm,the following independent variables were includedto represent perceived behavioral control. Phar-

macists were asked about their level of confidencein their therapeutic knowledge regarding OCs andperceived adequacy of time available to counselwomen on OCs. Similar to PRN, 5-point scales

ranging from “not ———— at all” to “extremely—————”

The full adapted TPB framework is shown in

Fig. 1.

Other variablesIn relation to the extent to which pharmacists

believed women welcomed pharmacists’ raising

OC topics, pharmacists were asked about thenumber of women who solicited advice frompharmacists on any topic related to OCs. Other

variables included importance of profit fromdispensing OCs for pharmacy revenue, and de-mographic variables such as pharmacists’ age,gender, marital status, pharmacy practice degree,

and the average number of hours the pharmacistworked every day. In terms of the prescriptionstatus of requests, pharmacists were asked about

the average number of women in the past weekwho requested an OC who had an OC prescrip-tion and the average number of women who did

not have a prescription. They were also asked forthe average number of women who requested helpin selecting an OC product without providing a


Table 1

Names, items and response categories for the dependent variable (COC) and key independent variables used

Variable name Item(s) used to measure variable Response categories

COC: Counseling on Oral

Contraceptivesa� Among the last five women whom you

dispensed OCs to, how many women

have you talked to on the importance of

taking OCs at the same time every day?

� Among the last five women whom you


have you talked to on the appropriate

day to start taking OCs?



have you talked to on how to act when

forgetting to take one or more doses of

OCs?


dispensedOCs to, howmanywomenhave

you talked to on the side effects of OCs?

(a) 0

(b) 1

(c) 2

(d) 3

(e) 4

(f) 5

PRN: Perceived

Responsibility for

Notification

� How responsible is the pharmacist for

notifying the woman with the impor-

tance of taking OCs at the same time

every day?


notifying the woman with the appro-

priate day to start taking OCs?


notifying the woman with how to act

when forgetting to take one or more

doses of OCs?


notifying the woman with the side ef-

fects of OCs?

(a) Not responsible at all

(b) Slightly responsible

(c) Somewhat responsible

(d) Responsible

(e) Extremely responsible

PKW: Perceived

Knowledge of Women

� What proportion of women possess

adequate knowledge on the importance

of taking OCs at the same time every

day?


adequate knowledge on the appropriate

day to start taking OCs?


adequate knowledge on how to act

when forgetting to take one or more

doses of OCs?


adequate knowledge on the side effects

of OCs?

(a) “20% or less”

(b) “21–40%”

(c) “41–60%”

(d) “61–80”

(e) “81–100”

PWC: Perceived Welcome

of Counseling

� What proportion of women would

welcome you starting a conversation on

the importance of taking OCs at the

same time every day?



the appropriate day to start taking

OCs?

(a) “20% or less”

(b) “21–40%”

(c) “41–60%”

(d) “61–80”

(e) “81–100”

(continued on next page)



Table 1 (continued )

Variable name Item(s) used to measure variable Response categories



how to act when forgetting to take one

or more doses of OCs?



the side effects of OCs?

Confidence in therapeutic

knowledge regarding OCs

How confident are you in your scientific

and therapeutic knowledge regarding

OCs?

(a) Not confident at all

(b) Slightly confident

(c) Somewhat confident

(d) Confident

(e) Extremely confident

Perceived adequacy of time to

counsel women on OCs

How adequate is the available time to

counsel women on OCs?

(a) Not adequate at all

(b) Slightly adequate

(c) Somewhat adequate

(d) Adequate

(e) Extremely adequate

Soliciting pharmacists advice

by women

Among the last five women whom you


have solicited your advice on any of the

topics related to OCs?

(a) 0

(b) 1

(c) 2

(d) 3

(e) 4

(f) 5

Perceived importance of

dispensing OCs on

pharmacy revenue

How important is the profit coming from

dispensing OCs on the pharmacy

revenue?

(a) Not important at all

(b) Slightly important

(c) Somewhat important

(d) Important

(e) Extremely important

a To construct the dependent measure for COC, the four values reported by pharmacists for the items shown in the

table were averaged. This average served as an overall indicator of the proportion of women who were counseled by

pharmacists. The same process was used in calculating PRN, PKW and PWC.


prescription in the past week. All key question-naire items are shown in Table 1.

Measure refinement and pre-testing

The primary researcher conducted standardcognitive testing of all questionnaire items with

eight community pharmacists from Alexandria,Egypt who were representative of the target popu-lation. Findings from cognitive testing indicatedthat pharmacists reported their past counseling

behavior much more easily and accurately thantheir intentions to counsel women on OCs inthe future. Hence, past counseling behavior served

as the dependent variable in the study. Followingslight revisions, pre-testing was done in nine com-munity pharmacies in Alexandria. The respondents

were asked to complete the questionnaire and thengive comments and feedback on the relevance andclarity of items.

Data collector training

Data collectors were instructed to adhere tothe agreed upon introductory script and maintain

neutrality at all times regarding all topics coveredin the questionnaire in order not to influencepharmacists’ responses. They were also instructed

not to answer respondents’ questions regardingthe meaning of survey questions to ensure consis-tency. Role-playing was used in training the data

collectors before the start of the survey.

Data collection

Five data collectors visited sample pharmacies,enrolled pharmacists, elicited IRB consentfollowing approved procedures, delivered and

retrieved the survey forms without intervening inhow the self-administered survey was filled. Datacollectors presented pharmacists with a support



letter from the Dean of the High Institute ofPublic Health in Alexandria University thatvalidated the identity of the primary researcher,explained the purpose of the study and encour-

aged cooperation from pharmacists. The phar-macy was visited up to two additional times tomake sure the pharmacist was located. If the

pharmacist was not found after a total of threevisits, the pharmacist was considered absent. Datawere collected in the period from the December

2014 to January 2015.

Data analysis

Data were analyzed using STATA (Version 12).

Descriptive analysesDescriptive statistics were used to describe the

characteristics of the sample as well the dependent

and independent variables. Descriptive analysesgenerated frequencies, as well as means, ranges orstandard deviations as relevant.

Bivariate and psychometric analysesScale scores were calculated by summing the

items and dividing by the number of items in thescale. Average scores were created for PWC,PKW, RRN and COC. Higher scores signified

pharmacists’ higher perceived (1) patron welcomeof pharmacist initiated OC counseling, (2) patronknowledge of OCs, (3) pharmacist responsibility

to educate women about OCs and (4) a largerpercent of the last five women they dispensed OCsto who they also counseled on OC topics.

Bivariate analyses of the independent and

dependent variables were conducted. A correlationmatrix was created to examine the relationshipsbetween different variables. Cronbach’s alpha was

used to analyze internal consistency of PWC,PKW, RRN and COC behavior scales.

RegressionIn addition to TPB predictors, the regression

model included sociocognitive or pharmacistcharacteristic variables that were hypothesized tobe theoretically associated with pharmacist coun-seling behavior. The goal was to achieve the most

parsimonious model that explains the most vari-ance in the dependent variable.

Before insertion into the regression model, the

four items of the PWC and PKW variables wereeach coded as 0–4 for the following responsecategories: “20% or less,” “21–40%,” “41–60%,”

“61–80” “81–100.” The four items of the PRNvariable were coded 0–4 for the followingresponse categories: “Not responsible at all,”

“slightly responsible,” “somewhat responsible,”“responsible” and “extremely responsible.” Foreach of these variables, the average of the scoresfrom their respective four items was inserted in the

regression model. Statistical tests were two-sidedwith alpha set at P ! 0.05.

Results

Sample identification and response

Of the 230 pharmacies in the sample, 23 had nopharmacist at any time, 16 were closed, eight were

not found, two were inaccessible to data collec-tors, and 181 were open with a pharmacist onduty at some point in the day. Of these 181pharmacists, 168 (93%) agreed to participate and

13 (7%) refused. Of the 168 returned surveyforms, one was not usable (Fig 2).

Sample demographics

Ananalysis of sample demographics anddescrip-

tive statistics is presented in Table 2. Most respon-dents were males (72%) and married (65%).Pharmacists’mean agewas 37 (SD¼ 13) and ranged

from 22 to 74 years. On average, pharmacistsworked 9 (SD¼ 2) hours per day.Most pharmacistshad a bachelor of pharmaceutical science degree

(88%) while 12% had a Pharm.D. degree.

Prescription status of dispensed OCs

The average number of women who requestedan OCs while providing a prescription in the past

week (mean ¼ 7.4; SD ¼ 6.9; range 0–50) wasclose to the average number of women whorequested an OC without providing a prescription

in the past week (mean ¼ 6.7; SD ¼ 5.3; range 0–30). The average number of women who asked forpharmacist’s help in selecting an OC withoutproviding a prescription in the past week was

smallest (mean ¼ 3.7; SD ¼ 4.4; range 0–30).

Reported rates of counseling on oral contraceptives

On average, pharmacists reported talking tonearly two out of five women on different OC

topics. Pharmacists indicated they talked to aslightly higher proportion of women about theimportance of taking OCs at the same time every

day (mean ¼ 2.31, 95% CI ¼ 2.05–2.56)compared to the appropriate day to start takingOCs (mean ¼ 2.19, 95% CI ¼ 1.94–2.45), side

effects (mean ¼ 2.03, 95% CI ¼ 1.82–2.24), andhow to act when forgetting to take one or moredoses of OCs (mean ¼ 1.78, 95% CI ¼ 1.56–2.00).


List of 3350 community pharmacies in Alexandria

230 community pharmacies

Random number generator

23 pharmacies had no

pharmacist at any visit

16 pharmacies were closed, were under

construc on or renova on

181 open pharmacies

with apharmacist

Eight pharmacies

were not found

Twopharmacies

wereinaccessible to data collectors

168 pharmacists agreed to

par cipate

13 pharmacists refused

Fig. 2. Sample identification and response.


Scale reliability

PWC, PKW, PRN and COC were all inter-

nally consistent with Cronbach’s alpha scoresgreater than 0.75 (Table 3). PWC and PRN hadthe highest reliability (alpha ¼ 0.90).

Descriptive statistics of theory of planned behaviorvariables

Perceived Responsibility for Notification (PRN)

Pharmacists’ perceptions of pharmacist re-sponsibility for notifying women about OCswere low and did not vary much by topic. On a

5-point scale, pharmacists indicated an equal

responsibility for talking about the importanceof taking OCs at the same time every day(mean ¼ 1.73, 95% CI ¼ 1.53–1.93) and side

effects (mean ¼ 1.73, 95% CI ¼ 1.55–1.91).These were very closely followed with the re-sponsibility for notifying women with the appro-

priate day to start taking OCs (mean ¼ 1.72,95% CI ¼ 1.53–1.91), The least perception ofresponsibility was for notifying the woman on

how to act when forgetting to take one or moredoses of OCs (mean ¼ 1.68, 95% CI ¼ 1.49–1.88).


Table 2

Pharmacist characteristics and descriptive statistics

(n ¼ 167)

Pharmacist characteristics Results

Gender % (n)

Male 72.46 (121)

Female 27.54 (46)

Marital status % (n)

Single 35.33 (59)

Married 64.67 (108)

Highest educational

qualification

% (n)

Bachelor of

pharmaceutical science

88.02 (147)

Doctor of pharmacy 11.98 (20)

Average number of hours

worked every day

Hours

Range 3–16

Mean (SD) 8.66 (�2.39)

Age Years

Range 22–74

Mean (SD) 37.29 (�12.5)


Perceived Knowledge of Women (PKW)Pharmacists’ perceptions of the proportion of

women who possess adequate knowledgeregarding OCs were exceedingly low and varied

by the topic examined. Pharmacists importance oftaking OCs at the same time every day(mean ¼ 1.34, 95% CI ¼ 1.18–1.51) was highest

closely followed by appropriate day to start takingOCs (mean ¼ 1.32, 95% CI ¼ 1.16–1.48) witheven lower knowledge ratings for side effects

(mean ¼ 0.87, 95% CI ¼ 0.72–1.02) and how toact when forgetting to take one or more doses(mean ¼ 0.81, 95% CI ¼ 0.66–0.96).

Perceived Welcome of Counseling (PWC)Pharmacists’ perceptions of the proportion

of women who would welcome the pharmacist

initiation of a conversation on OC topics were

Table 3

Reliability and descriptive statistics for the used scales

(N ¼ 167)

Scale Number

of items

Mean SD Min Max Cronbach’s

alpha

PRN 4 1.72 1.10 0 4 0.90

PKW 4 1.09 0.79 0 4 0.77

PWC 4 1.46 1.13 0 4 0.90

COC 4 2.08 1.34 0 5 0.89

PRN: Perceived Responsibility for Notification;

PKW: Perceived Knowledge of Women; PWC:

Perceived Welcome of Counseling; COC: Counseling

on Oral Contraceptives.

low. The highest was for addressing the appro-priate day to start taking OCs (mean ¼ 1.54,95% CI ¼ 1.34–1.75) followed by addressinghow to act when forgetting to take one or more

doses (mean ¼ 1.47, 95% CI ¼ 1.28–1.67).Pharmacists indicated an equal proportion ofwomen would welcome their initiation of a

conversation on importance of taking OCs atthe same time every day (mean ¼ 1.41, 95%CI ¼ 1.21–1.60) and side effects (mean ¼ 1.41,

95% CI ¼ 1.22–1.61).

Multiple regression analysis predicting counseling

on oral contraceptives

An analysis of OC counseling predictors ispresented in Table 4. The model explained 34.5%of the variance in COC. Statistically significant

predictors of the amount of OC counselingincluded the reported number of women whoasked for a pharmacist’s help in selecting an OC

without providing a prescription in the pastweek (b ¼ 0.160, P ¼ 0.018), perceived patronwelcome to pharmacist initiated OC counseling(b ¼ 0.315, P ! 0.001) and perceived adequacy

of time available to counsel women on OCs(b ¼ 0.290, P ¼ 0.003).

Correlations

Table 5 shows the correlation matrix with allsimple coefficients of correlation between studyvariables. Pharmacists’ perceptions of women’s

OC knowledge was associated with the theirperceived welcome to pharmacist initiated OCcounseling (r ¼ 0.34; P ! 0.0001), average num-

ber of women who requested an OC withoutproviding a prescription in the past week(r ¼ 0.16; P ¼ 0.03) and average number of

women who asked for pharmacist’s help in select-ing an OC product without providing a prescrip-tion in the past week (r ¼ 0.17; P ¼ 0.03).

Women’s perceived welcome of pharmacistinitiated counseling was associated with the phar-macists’ perceived responsibility for notifyingwomen about OC topics (r ¼ 0.44; P ! 0.0001),

average number of women who asked for pharma-cist’s help in selecting an OC product withoutproviding a prescription in the past week

(r ¼ 0.25; P ¼ 0.001), and the number of womenwho solicited advice from pharmacists on any topicrelated to OCs among the last five women who

obtained their OC from the pharmacist (r ¼ 0.45;P ! 0.0001). Lower perceived welcoming of phar-macist initiated OC counseling was associated


Table 4

Multiple regression analysis predicting Counseling on Oral Contraceptives

Variablesa Ba Robust SE t P b

Average number of women

who asked for

pharmacist’s help in

selecting an OC product

without providing a

prescription in the past

week

0.048 0.020 2.39 0.018 0.160

Perceived Knowledge of

Women

�0.078 0.130 �0.60 0.546 �0.046

Perceived Welcome of

Counseling

0.3716 0.099 3.77 !0.001 0.315

Perceived Responsibility for

Notification

0.060 0.107 0.56 0.577 0.050

Perceived adequacy of time

available to counsel

women on OCs

0.334 0.110 3.03 0.003 0.290

Confidence in their

therapeutic knowledge

regarding OCs

0.026 0.107 0.24 0.811 0.020

Perceived importance of

dispensing OCs on

pharmacy revenue

�0.142 0.110 �1.29 0.200 �0.095

Average number of working

hours in the pharmacy

every day

�0.041 0.0391804 �1.05 0.294 �0.073

Pharmacist’s age �0.001 0.009 �0.06 0.951 �0.005

Pharmacist’s gender �0.156 0.200 �0.78 0.435 �0.052

Pharmacist’s marital status 0.109 0.227 0.48 0.633 0.039

Pharmacy practice degree 0.050 0.272 0.18 0.855 0.012

Constant 1.309 0.454 2.88 0.005 _

B: Unstandardized regression coefficients; Robust SE: Robust standard error; t: t-statistics; P: 2-tailed P-values; Beta

(b): Standardized regression coefficients.

N ¼ 166 pharmacists, F (12, 153) ¼ 9.28, P ! 0.0001, R2 ¼ 0.3453, Root mean squared error ¼ 1.1246.a Gender 0 ¼ Female and 1 ¼ Male; 15: Marital status 0 ¼ Single and 1 ¼ Married; 16: Pharmacy practice degree

0 ¼ Bachelor of Pharmaceutical science and 1 ¼ Pharm.D.


with the pharmacist being male (r ¼ �0.27;

P¼ 0.0005). Perceived responsibility of OC patroneducation was associated with the number ofwomen who solicited advice from pharmacists onany topic related toOCs among the last five women

who obtained their OC from the pharmacist(r ¼ 0.25; P ¼ 0.001).

The average number ofwomenwho requested an

OC without providing a prescription in the pastweek was associated with the average number ofwomen who asked for a pharmacist’s help in

selecting an OC product without providing a pre-scription in the past week (r¼ 0.57;P! 0.0001) andthe number of women who solicited advice from

pharmacists on any topic related to OCs amongthe last five women who obtained their OC fromthe pharmacist (r ¼ 0.25; P ¼ 0.0012).

Perceived adequacy of time available to

counsel women on OCs was associated with thenumber of women who solicited advice frompharmacists on any topic related to OCs amongthe last five women who obtained their OC from

the pharmacist (r ¼ 0.30; P ¼ 0.0001) andperceived importance of dispensing OCs for phar-macy revenue (r ¼ 0.31; P ! 0.0001).

Discussion

This study systematically examines pharma-cists’ counseling on oral contraceptives using anadapted theory of planned behavior framework.As

hypothesized, the likelihood of counseling womenon OCs was positively associated with women whoasked for pharmacist’s help in selecting an OC


Table 5

Correlation matrix for study variables (The number of observations used in estimating correlations ranged from 166

to 167)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

1 1.00

2 0.17* 1.00

3 0.49* 0.34* 1.00

4 0.32* 0.07 0.44* 1.00

5 0.28* 0.16* 0.06 0.09 1.00

6 0.10 0.14 0.12 �0.11 0.21* 1.00

7 0.29* 0.17* 0.25* 0.03 0.57* 0.43* 1.00

8 0.49* 0.10 0.45* 0.25* 0.25* 0.06 0.31* 1.00

9 0.46* 0.25* 0.46* 0.48* 0.14 0.07 0.23* 0.30* 1.00

10 0.24* 0.25* 0.33* 0.25* 0.12 0.00 0.07 0.25* 0.47* 1.00

11 0.12 0.08 0.29* 0.26* �0.00 0.05 0.02 0.04 0.31* 0.28* 1.00

12�0.00 0.05 0.05 0.01 0.25* 0.13 0.11 0.11 0.12 0.14 �0.05 1.00

13 0.04 0.14 0.00 0.12 0.12 �0.09 �0.04 �0.03 0.17* 0.23*�0.00 0.18* 1.00

14�0.18*�0.07 �0.27*�0.13 �0.07 �0.06 �0.13 �0.06 �0.13 �0.01 �0.14 0.06 0.26* 1.00

15 0.05 0.13 0.08 0.08 0.12 �0.10 �0.08 �0.03 0.17* 0.17* 0.12 0.16* 0.60* 0.33* 1.00

16�0.02 0.16* 0.02 �0.25* �0.07 0.05 0.01 0.09 �0.09 0.07 �0.01�0.10 �0.05 �0.06 �0.11 1.00

*Significant at P ! 0.05.

1: Counseling on Oral Contraceptives (COC); 2: Perceived Knowledge of Women (PKW); 3: Perceived Welcome of

Counseling (PWC); 4: Perceived Responsibility for Notification (PRN); 5: Average number of women who requested an

OC without providing a prescription in the past week; 6: Average number of women who requested an OCs while

providing a prescription in the past week; 7: Average number of women who asked for pharmacist’s help in selecting

an OC product without providing a prescription in the past week; 8: The number of women who solicited advice

from pharmacists on any topic related to OCs among the last five women who obtained their OC from the pharmacist;

9: Perceived adequacy of time available to counsel women on OCs; 10: Confidence in their therapeutic knowledge

regarding OCs; 11: Perceived importance of dispensing OCs on pharmacy revenue; 12: Average number of working

hours the pharmacy every day; 13: Pharmacist’s age; 14: Gender 0 ¼ Female and 1 ¼ Male; 15: Marital status

0 ¼ Single and 1 ¼ Married; 16: Pharmacy practice degree 0 ¼ Bachelor of Pharmaceutical science and 1 ¼ Pharm.D.


without providing a prescription, the pharmacist’sperception of the proportion of women who wouldwelcome the pharmacist initiation of a conversa-tion onCOC topics and perceived adequacy of time

available to counsel women on OCs. However, apharmacist’s perception of the proportion ofwomenwho possess adequate knowledge regarding

OCs and a pharmacist’s perception of responsibil-ity for notifying women with information on topicsrelated to OCs were not significantly associated

with counseling on OCs. One, two or all three TPBconstructs have been found to be significant pre-dictors of other types of pharmacist behaviors in

previous studies.18–20

Compared to other topics, pharmacists re-ported they were less likely to counsel women onside effects and how to act when forgetting to take

one or more doses of OCs. Pharmacists in thisstudy had the perception that women in generalwould not prefer to listen to possible side effects

before they start taking OCs. It may also beexplained by a classical case where the principle ofbeneficence and autonomy collide as a pharmacist

would fear that the women would not take herOCs if she is informed about the OC side effects.This may explain the demographic health surveyfinding of women not receiving adequate infor-

mation on side effects from pharmacies, necessaryto make an informed choice and manage sideeffects should they occur.1 Providing information

on how to act when forgetting to take one ormore doses of OCs is time demanding and wouldneed a certain competency in therapeutic knowl-

edge that most Egyptian pharmacists may nothave acquired in their training.

Pharmacists’ perception of having adequate

time to counsel women was associated with coun-seling more women on OCs. This should come asno surprise since pharmacist busyness is commonlycited as a barrier for providing information to

patrons in different studies addressing counselingprovided by community pharmacists. Amount ofinformation provided to patrons was affected by

the level of pharmacy busyness.21–23 Interventionsaddressing pharmacist counseling on OCs andother topics should use strategies that address



organizational factors such as time constraints,complementary to concerns about the training pro-vided to pharmacists and/or patrons.24

The pharmacist’s perception of the proportionof women who would welcome the pharmacistinitiation of a conversation on OCs was thestrongest predictor of counseling on OCs after

controlling for other predictors. Further, thisperception was associated with the number ofwomen who were perceived to have solicited OC

advice from pharmacists. This gives insight onhow pharmacists perceive whether their consultswould be welcomed by women. A woman who is

willing to ask questions is seen as more likely towelcome the pharmacist’s initiating a conversa-tion on a topic related to OCs. While pharmacist’sgender was not a significant predictor of coun-

seling but was associated with perceived patronwelcome of OC counseling. Female pharmacistsfelt that their advice would be better received by

women than their male counterparts did.In Egypt as in many developing countries, most

medicines are routinely supplied without a prescrip-

tion. In some cases where a medication is dispensedwithout a prescription, a pharmacist is assumingprescribing duties.25,26 In this study, pharmacists

were more likely to provide information to womenwho are considering using a contraceptive whenthey prescribed OCs. There is a special need toensure that pharmacists who are serving women

withoutOCprescriptionshave the adequate trainingto carry out this role or to refer women to physicianswho are adequately trained to do so.

Perception of responsibility to counsel on OCswas particularly low for the majority of pharma-cists. It is possible that pharmacists view this as

part of the gynecologist’s responsibility ratherthan theirs. It was alarming to see the perceptionof the proportion of women who possess adequateknowledge was not a significant predictor. Ideally,

a pharmacist would provide counseling to some-one who lacks therapeutic information.

Strengths and limitations

As with all studies, this research had both

strengths and weaknesses. The probability samplewith high response rate reduced nonresponse biasand improved the generalizability of the study. The

study uses scales that employ item specific re-sponses to questions rather than agree/disagreeanswer options which strengthens it further. The

questionnaire was provided to pharmacists in theirnative language improving the comprehension ofquestionnaire items and ultimately improving the

quality of data. Theory based a priori hypothesesdrove the analyses.

Limitations of this project merit discussion. Self-

report measures are more likely to result in theoverestimation of counseling, a socially desirablebehavior. In an attempt to reduce this bias, thesurvey was self-administered and no identifying

information was collected from pharmacists. Thestudy did not include a measure that addressesopportunity for privacy which could have been a

predictor of counseling on oral contraceptives.However, it is unlikely that pharmacies vary in thisregard since counseling spaces are rarely provided in

Egyptian pharmacies. Finally, itwas not possible forpharmacists to discriminate between first time re-quests vs refill requests for OCs from patorns. Thisdistinction was not possible since patrons routinely

obtain medications without a prescription fromEgyptian pharmacies, which are not routinelyequipped with electronic patron medical records.

Need for further research

Qualitative research with pharmacists would

be beneficial for a more in-depth understanding ofthe relationship between the different variablesand counseling on OCs. It would be interesting to

see how useful the PWC is in predicting coun-seling on other medication classes. It would alsobe interesting to know what proportion of women

actually welcome a pharmacist initiating a con-versation on OCs. This would be important sinceearlier research has indicated that pharmacistsand patrons do not necessarily follow the same

“relationship script.”27

Conclusion

The Theory of Planned Behavior appears tohelp predict pharmacists’ OC counseling. There isa need to prepare pharmacists who are frequentlyrequested to assist women with the selection of an

oral contraceptive. Interventions that would facil-itate women’s requests for information may bevaluable to increase pharmacists’ counseling on

oral contraceptives. Future qualitative and obser-vational studies are needed to assess complexitiesin counseling patrons on oral contraceptives.

Acknowledgments

The authors would like to thank Dr. AshrafWahdan at the High Institute of Public Health,

Alexandria University for assisting the researchteam with the project logistics including datacollection and entry.



References

1. Ministry of Health and Population [Egypt], El-

Zanaty and Associates [Egypt], and ICF Interna-

tional. Egypt Demographic and Health Survey 2014.

Cairo, Egypt and Rockville, Maryland, USA: Minis-

try of Health and Population and ICF International;

2015. Available at: http://dhsprogram.com/pubs/

pdf/fr302/fr302.pdf. Accessed 10.06.15.

2. Bennett W, Petraitis C, D’Anella A, Marcella S.

Pharmacists’ knowledge and the difficulty of obtain-

ing emergency contraception.Contraception 2003;68:

261–267.

3. Fuentes EC, Azize-Vargas Y. Knowledge, attitudes

and practices in a group of pharmacists in Puerto

Rico regarding emergency contraception. P RHealth

Sci J 2007;26:191–197.

4. BorregoME, Short J, HouseN,GupchupG,NaikR,

Cuellar D. New Mexico pharmacists’ knowledge,

attitudes, and beliefs toward prescribing oral emer-

gency contraception. J Am Pharm Assoc (2003)

2006;46:33–43.

5. Whelan AM, Langille DB, Hurst E. Nova Scotia

pharmacists’ knowledge of, experiences with and

perception of factors interfering with their ability to

provide emergency contraceptive pill consultations.

Int J Pharm Pract 2013;21:314–321.

6. Shaaban OM, Fathalla MM, Shahin AY, Nasr A.

Emergency contraception in the context of marriage in

Upper Egypt. Int J Gynaecol Obstet 2011;112:195–199.

7. Ratanajamit C, Chongsuvivatwong V. Survey of

knowledge and practice on oral contraceptive and

emergency contraceptive pills of drugstore personnel

in Hat Yai, Thailand. Pharmacoepidemiol Drug Saf

2001;10:149–156.

8. Minh PD, Huong DT, Byrkit R, Murray M.

Strengthening pharmacy practice in Vietnam: find-

ings of a training intervention study. Trop Med Int

Health 2013;18:426–434.

9. Hellerstedt WL, Van Riper KK. Emergency contra-

ceptive pills: dispensing practices, knowledge and at-

titudes of South Dakota pharmacists. Perspect Sex

Reprod Health 2005;37:19–24.

10. Sattari M, Mokhtari Z, Jabari H, Mashayekhi SO.

Knowledge, attitude and practice of pharmacists

and health-care workers regarding oral contracep-

tives correct usage, side-effects and contraindica-

tions. East Mediterr Health J 2013;19:547–554.

11. Sarino LV, Dang KH, Dianat N, et al. Drug interac-

tion between oral contraceptives and St. John’s

Wort: appropriateness of advice received from com-

munity pharmacists and health food store clerks.

J Am Pharm Assoc 2007;47:42–47.

12. Obreli-Neto PR, Pereira LR, Guidoni CM, et al. Use

of simulated patrons to evaluate combined oral con-

traceptive dispensing practices of community phar-

macists. PLoS One 2013;8:e79875.

13. Ajzen I. The theory of planned behavior. In: Van

Lange PAM, Kruglanski AW, Higgins ET, eds.

Handbook of theories of Social psychology, vol. 1. Thou-

sandOaks,CA:SagePublicationsLtd;2012.p.438–459.

14. Ajzen I. The theory of planned behavior. Organ

Behav Hum Dec 1991;50:179–211.

15. FlemingML, Barner JC, Brown CM, ShepherdMD,

Strassels S, Novak S. Using the theory of planned

behavior to examine pharmacists’ intention to utilize

a prescription drug monitoring program database.

Res Social Adm Pharm 2014;10:285–296.

16. Gavaza P, Fleming M, Barner JC. Examination of

psychosocial predictors of Virginia pharmacists’

intention to utilize a prescription drug monitoring

program using the theory of planned behavior. Res

Social Adm Pharm 2014;10:448–458.

17. Amin ME, Chewning B. Predicting pharmacists’

adjustment of medication regimens in Ramadan us-

ing the Theory of Planned Behavior. Res Social

Adm Pharm 2015;11:e1–e15.

18. Herbert KE, Urmie JM, Newland BA, Farris KB.

Prediction of pharmacist intention to provide Medi-

care medication therapy management services using

the theory of planned behavior. Res Social Adm

Pharm 2006;2:299–314.

19. Gavaza P, Brown CM, Lawson KA, Rascati KL,

Wilson JP, Steinhardt M. Examination of pharma-

cists’ intention to report serious adverse drug events

(ADEs) to the FDA using the theory of planned

behavior. Res Social Adm Pharm 2011;7:369–382.

20. Walker A, Watson M, Grimshaw J, Bond C.

Applying the theory of planned behaviour to phar-

macists’ beliefs and intentions about the treatment

of vaginal candidiasis with non-prescription medi-

cines. Fam Pract 2004;21:670–676.

21. Pradel FG, Obeidat NA, Tsoukleris MG. Factors

affecting pharmacists’ pediatric asthma counseling.

J Am Pharm Assoc 2007;47:737–746.

22. Witry MJ, Doucette WR. Community pharmacists,

medicationmonitoring, and the routine nature of refills:

aqualitative study.JAmPharmAssoc2014;54:594–603.

23. Odukoya OK, Chui MA, Pu J. Factors influencing

quality of patron interaction at community phar-

macy drive-through and walk-in counseling areas.

Int J Pharm Pract 2014;22:246–256.

24. Patwardhan PD,AminME, Chewning BA. Interven-

tion research to enhance community pharmacists’

cognitive services: a systematic review. Res Social

Adm Pharm 2014;10:475–493.

25. Goel P, Ross-Degnan D, Berman P, Soumerai S.

Retail pharmacies in developing countries: a

behavior and intervention framework. Soc Sci Med

1996;42:1155–1161.

26. Smith F. The quality of private pharmacy services in

low and middle-income countries: a systematic re-

view. Pharm World Sci 2009;31:351–361.

27. Worley MM, Schommer JC, Brown LM, et al. Phar-

macists’ and patrons’ roles in the pharmacist-patron

relationship: are pharmacists and patrons reading

from the same relationship script? Res Social Adm

Pharm 2007;3:47–69.

http://dhsprogram.com/pubs/pdf/fr302/fr302.pdf

http://dhsprogram.com/pubs/pdf/fr302/fr302.pdf

http://refhub.elsevier.com/S1551-7411(15)00165-5/sref2














































































































Pharmacists' counseling on oral contraceptives: A theory informed analysis

Documents