This stage of primary data collection depended on interviewing staff available at facilities; the teams made three attempts to interview the relevant selected providers. We looked at problems of motivation and resilience in the face of poor working conditions to provide a more complete picture of retention. Information about professional and personal characteristics was collected and responses were linked to geographic area, facility level, cadre type, and other such characteristics. Major factors impacting motivation (pay, promotion opportunities, supervision, relations with colleagues, and work conditions) were assessed through the standard Job Descriptive Index (JDI). Similarly, as part of a 360 degree assessment, there was another structured questionnaire for managers. Providers and managers were asked different questions, but with common themes so as to offer the possibility of comparing responses on similar issues, e.g. housing, pay, safety and security, and incentives to allow a comprehensive view of working environments.
A situation analysis methodology was employed to rapidly assess whether the facility being evaluated was in fact capable and equipped to deliver MNCH services. Then, all the specified health provider cadres were randomly selected, interviewed, and data collected on socio‐demographic factors, along with factors affecting retention and motivation.
Data collected during the primary phase was entered using the CSPro 3 data entry programme and analysed using SPSS (Statistical Package for the Social Sciences) version 14. The data collected though IDI and FGD sessions were digitally recorded, and discussions were transcribed and stored in an mp3 audio file format along with respondent face sheets. Data was electronically forwarded on a daily basis to the data manager
who collated and prepared it for analysis. The data manager conducted quality checks on questionnaire completeness, inter‐record checks, and mistakes in data files, and suggested corrections in consultation with the principal investigator.
The analyses were performed in two stages. The first stage used “secondary data” to identify gaps in sanctioned versus filled posts. The second part of the analysis classified answers to key questions about motivation and turnover and retention against some basic demographic features such as region, cadre, years of experience, and employment status etc. Gender differences in responses were analysed based on data related to WMOs and LHVs. Univariate analysis was first carried out to obtain descriptive statistics, and this was then supplemented by advanced inferential bivariate analysis. Significant relationships at the bivariate level, as well as theoretically relevant factors were further analysed using logistic regression at the multivariable level.
Analysis of various tested differences between demographic groups and linear regressions measured the association between motivational determinants and outcomes. For the qualitative data, after a thorough content analysis, codes were given and a matrix was developed and themes identified. Data was sorted based on themes and sub‐themes and subsequently analysed to identify the social and organizational factors influencing motivation of female healthcare providers.
Study team
Eight teams were constituted to carry out the field work. Each study team comprised nine members with sub‐teams of two members each, with one person entering data in the field. The teams consisted of medical doctors, sociologists, and anthropologists. Facility assessments were carried out by medical doctors and the providers were interviewed by trained and experienced enumerators. Quality assurance A number of measures were instituted to ensure the maintenance of the highest quality standards in both data collection and data analysis. These measures are discussed below. Standard operating procedures The Population Council developed standard operating procedures for carrying out all processes to ensure quality assurance. A number of measures were instituted to ensure the maintenance of the highest quality standards in both data collection and data analysis. Determination of roles and responsibilities
Team member and field coordinator roles and responsibilities were clearly identified. Members were provided a written set of roles and responsibilities that they were required to follow.
Training
The interviewers received nine days of training conducted by the Population Council. The training included sessions on the research protocol, ethics, obtaining informed consent, maintaining privacy during the interview process, sampling methodology, and interviewing techniques.
The training focused on how to adhere to the standard operating procedures and familiarized participants with study objectives. Two nine‐day trainings on research methods and data collection tools were conducted for the data collection teams.
A questionnaire manual was prepared for interviewers and each member was provided with a written set of responsibilities and standard quality checklists. Each team had a quality control supervisor for on‐the‐spot data verification during the data collection phase.
Monitoring
The principal investigator, study coordinator, and field coordinators regularly visited randomly selected districts in each region to ensure all protocols were being followed. They randomly selected completed questionnaires during monitoring visits to check for completeness, data accuracy, and to determine re‐ interviewing requirements.
Ethical considerations
Ethical approval was obtained from the Institutional Review Board (IRB) of the Population Council’s headquarters in New York (Annex3A) and by the National Bioethics Committee (NBC) Pakistan(Annex 3B). Informed consent was obtained from all study participants after describing to them in detail the issues related to the study. For the structured questionnaire, the interviewers described the scope and purpose of the questionnaire and its approximate length, and stressed that participation was voluntary. The structured questionnaires were administered in private. All data collected in each phase of the study was kept confidential and anonymous. The structured questionnaires and IDIs were identified by personal identification numbers rather than participant names. Response Rates and Refusals A total of 1296 interviews were conducted out of a sample size of 1372. The number of interviews that took place depended on staff posting at the facility at the time of visit. The response rate was 94%, and refusal rate less than 0.1%
1.10: Study Limitations and Implications for Future Research
One of the most challenging aspects of this study was the fact that motivation is a complex issue and difficult to measure because of the subjectivity of the topic and the lack of standard measurement tools. Furthermore, motivation levels do not remain static but vary depending upon conditions and circumstances affecting an individual’s life.
Another limitation of this study is that interviews were only conducted with current employees and not with providers who have left the system, which means we do not have a complete picture of reasons that contribute to attrition and turnover. It can be assumed that providers who remain within the system have higher motivation levels than those who have left.
The availability and quality of secondary data on HRH was a limitation. Data on sanctioned and filled posts for providers was not available in Balochistan, and was only available according to grades for KP which
made a comparative analysis difficult. Additionally, while managers have given us estimates; proper data on attrition rates and attendance records is lacking, because of which the findings of this study could not be used to make a cause and effect assumption about turnover. The results from this study have now put in place a point of referenceon determinants of motivation and retention for healthcare providers working in the public health system in Pakistan. Future research should use this studyas a measure against which to gauge the impact of the health sector reform efforts that are underway, and to analyse the extent to which the recommendations put forward here have translated into policies and programs. This study design can also be used amongstprivate sector employees of the same cadre in order to create a true comparison between the public and private sector employees.