Mwanza region is bordering Lake Victoria located in the northern part of Tanzania. According to the 2012 national census, Mwanza region had a population of 2,772,509 in an area of 35,187 square kilometers (2). For 2002–2012, the average annual population growth rate was 3.0%making Mwanza among eight regions with high growth rate. The Tanzania Human Development Report ranks Mwanza region 13th among the 35 regions of Tanzania, with population living in severe poverty (32.8%) and population vulnerable to poverty (19.7%) as compared to the national average of 31.3% and 18.2% respectively (1).
The region is part of the Lake Zone where the maternal mortality rate was 453 deaths per 100,000 live births and under-five mortality rate was 88 deaths per 1,000 live births in the 10-year period preceding the TDHS 2015/16 and, these rates failed to meet Tanzania’s MDG targets (2). The neonatal mortality rate of Mwanza region in 2015 was 29/1,000 live births which remains higher than the national average of 25/1,000 (2).
National data indicates that among pregnant women, only 50.7% attend at least 4 recommended health facility visits for focused ante-natal care (FANC) during the last pregnancy (2). Health facility deliveries in Mwanza region account for 63.6% on average, while large disparities within the region persist with 87% deliveries occurring in facilities in urban areas versus 54.7% in rural areas (2). With the focus on its poor RMNH indicators, Mwanza region is one of five prioritized regions in Tanzania targeted by the Government (2). Understanding the training needs of healthcare workers in Mwanza region forms an important entry point for the IMPACT project in seeking to increase their contribution towards improving RMNH indicators. This created a need for validating the TNAQ in Mwanza
A cross-sectional survey using self-administered questionnaires was conducted. The survey involved RMNH healthcare workers at selected health facilities in all eight districts of Mwanza Region, Tanzania.
Sampling and Eligibility
Healthcare workers were selected from 36 sampled health facilities. The healthcare facilities involved all 7 district hospitals, 12 of 51 health centers and 17 of 292 dispensaries. The sample size was determined by the IMPACT Project’s need. The sampling of health facilities included in the baseline survey was based a) the number of health facilities included in the project (in total 80), and b) an estimated percentage of facilities considered (based on experiences in other similar surveys) of sufficient power to provide precise measurements for the project indicators at health facility levels. Thus, all district hospitals were included owing to their number, and 63% of health centers and 32% of dispensaries within IMPACT Project were considered sufficient to provide the required study power for the purposes of monitoring and evaluation. The relative homogeneity of dispensaries in terms of infrastructure, service provision, and human resources was also a reason for selecting one-third of the IMPACT Project’s target dispensaries.
All adult health workers responsible for RMNH service provision and present in the facility at the time of the survey were eligible to participate. The specific inclusion and exclusion criteria were as follows:
- All health care workers aged 18 years and above
- RMNH staff present at the facility during the time of the survey
- Health care workers who can understand and communicate well in Kiswahili or English language
- Health care workers who were working in RMNH
- Health care workers who were willing to give informed consent.
- Health care workers who were found in RMNH but not usually working in such a unit.
- Healthcare workers unable to answer questions because of physical or/and mental impediments
- Healthcare workers not willing to participate.
Data collection took place in August 2017. The administered TNA questionnaire was designed for providers of RMNH at the primary (dispensary and health center) and secondary (health center and district and designated district hospital) levels. As noted above, the tool was adapted from the Hennessy-Hicks TNAQ instrument (12, 15), which has been psychometrically tested for reliability and validity and adopted by the WHO (15). The Hennessy-Hicks instrument has been similarly adapted to assess the training needs of different health care practitioners in a range of cultural contexts (12, 15, 16). In the adaptation of the TNAQ, the pooled items were obtained from literature review and expert opinion basing on their experience in the field of reproductive health. These pooled items were validated by the health expert panel with expertise in teaching, reproductive health, research and local culture including customs, traditions and local languages (Kiswahili). The adapted TNA tool divides questions into broad categories, allowing for both intra-category and inter-category comparison of training needs.
During data collection, the person in-charge of the selected facility identified the RMNH personnel to whom the questionnaire was given for completion. Participants were requested to assess their own performance and rate the importance of specific RMNH services/activities through a self-administered, confidential paper-based questionnaire. The questionnaire asked participants how each RNMH related activity is important to the successful performance of their work and how well they considered their performance in each activity with each item in the questionnaire rated along a seven-point Likert scale. Participants were also asked to identify areas in which they most wanted to receive additional training and the trainings that they had most recently completed. Research assistants were on hand to answer questions and clarify elements of the questionnaire. The returned forms were checked for completeness and accuracy before leaving each health facility.
Data management, analysis and interpretation
The Statistical Product and Service Solutions (SPSS, version 20.0) was used for data entry and statistical analysis. Data from the questionnaires were reviewed to identify consistencies and differences, coded and quantified. The data were then manually entered into a password-protected database via an entry screen that performed validation checks for accuracy. The missing data was excluded during analysis.
The TNAQ was validated through a three-phase process: Phase 1 consisted of forward and backward translation of the TNA into Kiswahili by a consultative process involving bilingual (English and Kiswahili individuals. Phase 2 examined the content validity of TNAQ. Phase 3 involved field testing to implement the exploratory factor analysis (17) and the confirmatory factor analysis (CFA) (Lane et al., 1997; Ford 1997). Analysis of Moment Structures (AMOS) software embedded in SPSS version 20 was used to confirm the factor structure of the TNAQ from the exploratory analysis.
The data collected by TNA were analyzed by comparing the participants’ rating scores of the importance of the RMNCAH service/activity (A) with the competence in the service/activity (B) as perceived by them. The greater the difference in scores in TNAQ, the greater the training need.
The Exploratory factory analysis which is a statistical method that helps to identify a set of latent constructs underlying a battery of measured variables (18). In order to make the interpretation of factor analysis, the first step is to extract a set of factors (relevant factors) from a data set, then rotation of the remaining factors (19, 20). Rotation makes the output more comprehensible through formulation of the structure so-called "Simple Structure" (20). Two main types of rotations are used; orthogonal; when the new axis are in orthogonal to each other and oblique; when the new rotations are free to take any space. The promax rotation is an alternative non-orthogonal (oblique) rotation method and it is computationally faster than the direct oblimin method. Therefore it is sometimes used for very large datasets.
The reliability of the TNAQ was assessed. Internal consistency was measured by Cronbach’s alpha coefficient with 0.7 indicating acceptance of the instrument. Construct validity was evaluated by EFA and CFA. The EFA was conducted using Principal axis factoring estimation with promax rotation that facilitates the determination of the underlying factor structure of the items and it has advantage of being fast and conceptually simple (21). The factor retention applied the following criteria: (a) eigenvalues greater than 1.0, (b) the percentage of total variance explained, (c) scree plot, and (d) factor loadings above 0.40 were retained. The CFA was determined by using generalized least squares estimation to compare the current and the original 5-factor model of the scale. The model fit was considered acceptable if χ2/df < 2, adjusted goodness-of-fit index (AGFI) > 0.9, comparative fit index > 0.9, goodness-of-fit index (GFI) > 0.9, root mean square error of approximation (RMSEA) < 0.04, and incremental fit index (IFI) > 0.9 (18, 22).