Effectiveness of Evidence-Based Medicine on Knowledge, Attitudes, and Practices of Family Planning Providers: A Randomized Experiment in Jordan

Last registered on November 03, 2014

Pre-Trial

Trial Information

General Information

Title
Effectiveness of Evidence-Based Medicine on Knowledge, Attitudes, and Practices of Family Planning Providers: A Randomized Experiment in Jordan
RCT ID
AEARCTR-0000539
Initial registration date
November 03, 2014

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published
November 03, 2014, 9:42 AM EST

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Locations

Region

Primary Investigator

Affiliation
Abt Associates Inc.

Other Primary Investigator(s)

PI Affiliation
Abt Associates Inc.
PI Affiliation
Abt Associates Inc.
PI Affiliation
University of Michigan

Additional Trial Information

Status
Completed
Start date
2011-12-05
End date
2013-01-18
Secondary IDs
Abstract
Objective: To evaluate the effectiveness of an Evidence-Based Medicine (EBM) program conducted in Jordan on private providers' knowledge, attitudes, and practices toward an injectable contraceptive, Depot Medroxy Progesterone Acetate (DMPA).
Design: A stratified randomized controlled study by region and gender. Computer generated random assignments allocated treatment and control groups.
Setting: Provider misconceptions and biases about contraceptive methods are major barriers to family planning access for women in low- and middle-income countries, including Jordan. Evidence-Based Medicine programs aim to reduce biases and misconceptions by providing doctors with the most up-to-date scientific evidence on contraceptive methods. Study was conducted December 2011 to January 2013.
Participants: 267 private doctors providing family planning services in Amman and Zarqa regions in Jordan.
Intervention: The treatment group was invited to attend a roundtable seminar on the injectable contraceptive DMPA, and to receive two 15 minute one-on-one educational visits that reinforce the messages from the seminar. The control group received two educational visits on the previous year's material on Combined Oral Contraception pills and no information on DMPA.
Main outcome measures: Providers' knowledge of DMPA, attitudes towards DMPA, self-perceived confidence level, and reported clinical practice.
Results: 267 providers randomized (treatment group n= 135; control group n= 132). 38 providers excluded because they did not complete an endline survey leaving 229 doctors (treatment group n = 117 ; control group n = 112) for intention to treat analysis. No significant impact of the EBM program on knowledge of DMPA's side effects (point estimates -0.02) or on reported clinical practices (point estimates -0.01). Attitude and Confidence Scores of providers in the treatment group are 0.14 and 0.21 standard deviations higher (respectively) compared to the control group, but not statistically significant. Evidence shows positive selection into program participation.
Conclusions: EBM may not be effective as a stand-alone program targeting a family planning method with a high level of provider and consumer bias. Evidence of positive selection into program participation underscores the importance of randomization to avoid overestimating the true effects of interventions.
External Link(s)

Registration Citation

Citation
Chatterji, Minki et al. 2014. "Effectiveness of Evidence-Based Medicine on Knowledge, Attitudes, and Practices of Family Planning Providers: A Randomized Experiment in Jordan." AEA RCT Registry. November 03. https://doi.org/10.1257/rct.539-1.0
Former Citation
Chatterji, Minki et al. 2014. "Effectiveness of Evidence-Based Medicine on Knowledge, Attitudes, and Practices of Family Planning Providers: A Randomized Experiment in Jordan." AEA RCT Registry. November 03. https://www.socialscienceregistry.org/trials/539/history/3025
Sponsors & Partners

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information
Experimental Details

Interventions

Intervention(s)
The intervention is a Depot Medroxy Progesterone Acetate (DMPA) evidence-based medicine program consisting of a two-hour roundtable seminar and two educational visits that would reinforce messages from the seminar. Programmers invited treatment group providers to the DMPA roundtable seminar led by trained private peer providers who presented clinical research findings related to DMPA. The roundtable format of the seminar allowed providers to discuss these findings within the local context, with the overall goal of correcting misconceptions and biases and improving providers' knowledge of evidence-based health benefits of method use. The seminar covered findings on fertility return, anemia, amenorrhea, and other real or perceived side effects of DMPA. The two educational visits, to be scheduled ahead of time with individual doctors, were conducted by a trained health worker. These 15-minute one-on-one sessions, held in individual providers' clinics, were intended to review specific information that was discussed during the seminar. These sessions had three objectives: to enhance the availability of evidence to support providers in counseling about DMPA; to address concerns about its side effects and perceived harm (such as delays in returns to fertility); and to convey information about its potential benefits (such as protective effects against cancer). The seminar and educational visits were implemented between January and June 2012.

The control group providers were not invited to a DMPA seminar and were not offered educational visits related to DMPA. However, in order to continue to engage the control group providers during the period of the study (the first six months of 2012), they were offered two repeat educational visits during this same time period; these visits were related to Combined Oral Contraception (COC) pills and a repeat of the previous year's materials. Thus, the treatment and control groups differed in the content as well as the extent of program offerings: the treatment group was offered participation in the DMPA seminar and two DMPA educational visits, and the control group was offered two repeat educational visits on COCs. No information about DMPA was discussed during the COC visits; thus, any detectable differences on knowledge, attitudes, or practice related to DMPA can be attributed to the influence of the DMPA program.
Intervention Start Date
2012-01-15
Intervention End Date
2012-06-29

Primary Outcomes

Primary Outcomes (end points)
Providers' knowledge of DMPA, attitudes towards DMPA, self-perceived confidence level, and reported clinical practice.
Primary Outcomes (explanation)
All outcome variables are defined as scores and constructed from main variables: (1) providers' knowledge of DMPA (for example, correct identification of side effects), represented as their Knowledge Score; (2) providers' attitudes towards DMPA (including willingness to recommend DMPA), represented as their Attitude Score; (3) providers' self-perceived confidence level (including confidence in discussing DMPA with clients), represented as their Confidence Score; and (4) providers' reported clinical practice (such as reported prescription or discussion of DMPA with clients), represented as their Practice Score. Each outcome measure is constructed as explained below.

To measure knowledge and attitudes towards DMPA, the survey respondents were asked to rate specific knowledge and attitudes statements using a five-point Likert scale, ranging from "strongly agree" to "strongly disagree." Examples of these statements are: "Use of DMPA is positively associated with weight gain"; or, "I would have no hesitation recommending DMPA to a healthy woman. "We assign values ranging from -2 to +2 to each of the knowledge and attitude variables, where +2 denotes the most desirable item response (e.g., "I strongly agree" that I would have no hesitation recommending DMPA to a healthy woman), and -2 denotes the least desirable response. We create a Knowledge Score and an Attitude Score for each provider using the simple average of related variables. Following Kling et al. (2007), we standardize the scores by subtracting the mean and dividing by the standard deviation of the variable among the EBM control group, such that the control group's mean is zero and its standard deviation is one.

To measure self-perceived confidence in DMPA, the respondents were asked to provide answers to three questions, using a ten-point scale. Examples of these questions are: "How knowledgeable do you feel about DMPA?" and "How comfortable do you feel prescribing DMPA for your clients?" We form the simple average of these three variables and create a Confidence Score that we also standardize according to Kling et al. (2007). Confidence outcomes were not collected at baseline.

To measure clinical practices related to DMPA, the respondents were asked to report whether they have any DMPA stock at their clinic, as well as the number of times they discussed DMPA with clients or prescribed it to their clients during the month prior to the survey. We form z-scores of each of these three variables, compute a Practice Score using the simple average, and then similarly apply Kling et al. (2007) to standardize the score among the control group.

Secondary Outcomes

Secondary Outcomes (end points)
Secondary Outcomes (explanation)

Experimental Design

Experimental Design
1. Determining the sample
From 2005 to 2010, the USAID-funded predecessor project, Private Sector Project for Women's Health-Jordan, worked with private obstetricians/gynecologists and general practitioners who provide family planning services in Jordan. The project compiled a list of providers that the study team used as the sampling frame for this evaluation. The sample of eligible providers consisted of 267 private health providers in two regions, Amman and Zarqa. The sample did not include doctors practicing in the North and South regions of Jordan because it was logistically difficult to randomize those doctors and provide a widely dispersed treatment group with EBM seminars. About 16 percent of the doctors in the sample are specialists; the rest are general practitioners.

2. Randomization
Using STATA v.12, the sample of 267 providers was randomly assigned to treatment and control groups, stratifying by geographic area, gender, and on whether the providers belonged to the project's network of providers. In total, there were 135 providers in the treatment group and 132 in the control group. Random assignment was done prior to data collection and the results were not communicated to anyone outside of the research study team.

3. Baseline Data Collection
A baseline survey was conducted in December 2011. Survey questionnaires were mailed to all 267 providers and follow-up phone calls were made to providers who did not respond to the mailed questionnaire. The questions asked for specific information on the provider’s knowledge of DMPA's side effects, attitudes towards the method, and clinical practices such as discussing DMPA with clients or prescribing it.

4. DMPA EBM Intervention
Treatment group providers were invited to a two-hour roundtable seminar and received two 15 minute educational visits that would reinforce messages from the seminar.

5. Endline Data Collection
The endline survey conducted in December 2012 collected information similar to the baseline: provider knowledge of DMPA's side effects, attitudes towards the method, and practices. It also collected information such as years of experience, number of patients seen, whether the provider has a dual practice, and the types of family planning methods that the provider has prescribed. Enumerators conducted in-office surveys. Primarily due to the change in the data collection method, the endline response rate was higher than the baseline rate- 87 percent (229 providers).
Experimental Design Details
Randomization Method
Randomization done in office by a computer using Stata prior to data collection. Randomization was stratified by region and gender.
Randomization Unit
Private providers
Was the treatment clustered?
No

Experiment Characteristics

Sample size: planned number of clusters
N/A
Sample size: planned number of observations
Planned 267 private providers
Sample size (or number of clusters) by treatment arms
135 providers in the treatment group and 132 in the control group
Minimum detectable effect size for main outcomes (accounting for sample design and clustering)
IRB

Institutional Review Boards (IRBs)

IRB Name
Abt Associates Inc., Institutional Review Board
IRB Approval Date
2012-01-01
IRB Approval Number
0600

Post-Trial

Post Trial Information

Study Withdrawal

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Intervention

Is the intervention completed?
No
Data Collection Complete
Data Publication

Data Publication

Is public data available?
No

Program Files

Program Files
Reports, Papers & Other Materials

Relevant Paper(s)

Reports & Other Materials