PropertyValue
is nif:broaderContext of
nif:broaderContext
is schema:hasPart of
schema:isPartOf
nif:isString
  • Multiple methods were used to conduct this evaluation. As summarized in Table 1, this included a desk review; a national survey of the lay counselors; in-depth interviews at the national, district, and facility levels; focus group discussions with lay counselors; observations of lay counselors carrying out their duties; client exit interviews; and secondary data analyses. This research was approved by the Botswana Health Research and Development Committee. The study was conducted by the International Training and Education Center for Health (I-TECH), which is a collaboration between the University of Washington and University of California, San Francisco under the guidance of a technical working group comprised of healthcare stakeholders. All participants provided signed informed consent. Table data removed from full text. Table identifier and caption: 10.1371/journal.pone.0061601.t001 Summary of data collection methods used in the evaluation of the lay counselor cadre in Botswana. For the national survey, a self-administered questionnaire was distributed to all lay counselors at public health facilities in the country. Of 408 lay counselors, 385 returned their surveys (response rate = 94%). The questionnaire contained items related to demographics, duties, human resources, training, support, and job performance as well as a work-related knowledge test based on material covered during the pre-service training. Seventy-nine in-depth interviews with health workers were conducted with a purposively-selected stratified sample at the national-level (n = 17), district-level (n = 23), and facility-level (n = 39). These were semi-structured interviews. At the national-level, interviews were held with individuals involved in the development, administration, or training of the lay counselor cadre. District-level interviews were conducted in seven districts purposely selected to obtain a mixture of two urban, three semi-urban, and two rural districts with a high number of lay counselors. This represented 29% of all health districts in the country (7/24). Interviews were held with district coordinators overseeing the health programs supported by the lay counselors, as well as with individuals who supervise and support the lay counselors, such as PMTCT focal persons, nurses, social workers, and doctors. Facility-level interviews were conducted at two facilities within each of the seven districts with health workers closely involved with lay counselors. All interviews were conducted face-to-face, with a rapporteur present to take notes. With permission, the interviews were captured with a digital voice recorder and transcribed. Focus group discussions (FGDs) with lay counselors were held in each of the seven districts described above. This was used to compliment the quantitative information collected from the national survey. In total, 76 lay counselors participated in the FGDs, which represented19% of the cadre (76/408). During these sessions, in-depth information was collected about the lay counselor duties, pre-service and in-service training, support, and job performance. Each of the FGDs was captured with a digital voice recorder and transcribed. Direct observations of lay counselors carrying out their duties were conducted by experienced nurse counselors in the same seven districts in which interviews and FGDs took place. Stratified random sampling was used to select three facilities in each of the seven districts to ensure that a hospital, clinic, and health post were included (if there was no health post in a district, a second clinic was selected). Lay counselors at each facility were asked to participate in the study. This encompassed one to three lay counselors per facility. Observations were conducted with 47 clients who consented to be observed. The observational assessment tools and procedures were adapted from those developed by UNAIDS,[42] which have been used in a similar setting. [43] The two nurse counselors who conducted the observations were trained on the tool to ensure consistent administration. They were not affiliated with any of the data collection facilities. Client exit interviews were conducted to assess patient satisfaction with care by the same nurse counselor who observed the counseling sessions. Only clients whose counseling sessions were observed were included in the exit interviews (n = 47). A semi-structured questionnaire to elicit close-ended responses on clients' feelings about the services received that day was used. To help minimize social desirability influences, the nurse counselors who conducted the interviews emphasized that the purpose of this activity was for program improvement. Qualitative data included information from the desk review, key informant interviews, and focus group discussions as well as open-ended questions from the national survey and exit interviews. Transcripts were coded using ATLAS.ti v6.0 software for thematic analysis. Given that little information was available related to the day-to-day activities of the lay counselors or their performance; a grounded theory approach was taken as an inductive strategy for characterizing the performance of the cadre. Quantitative data included data from the national survey as well as data collected as part of the observations, exit interviews, and district-level routine HIV testing and counseling data. The survey data were entered into an Access database using a two-pass data verification process and analyzed using SPSS v15.0 software. Simple univariate analyses were used to describe the data. Paired t-tests were used to compare comfort-level ratings for job duties within and outside of the lay counselors' job description. Pearson correlations were used to compare years of experience as a lay counselor with comfort-rates relative to job duties.
rdf:type