Rishma Maini, Josephine Borghi, Natasha Palmer, David Hotchkiss Measuring the motivation of health workers in the DRC
Background: The DRC • Poor provision of basic services • Public sector health workers often not paid by government • Donor...
Programme changes • Red zones – PBF programme was operating (now stopped) • Green zones – workers were never exposed to PBF
Aims • To measure the determinants and outcomes of motivation in health workers in the DRC • To compare any differences be...
Methods DATA COLLECTION • Health worker motivation survey administered in red and green zones, and outside of those zones....
CHARACTERISTIC PROPORTION OF WORKERS (n=430) SEX Male 69% Female 31% AGE <30 years 11.2% 30-44 years 59.5% 45-60 years 26....
Predefined constructs Construct Cronbach’s alpha Determinants Financial 0.6 Management 0.5 Job description 0.6 Workload 0....
Latent factors overall scale Intrinsic Extrinsic Individual factors (Competence and autonomy) Opportunities Job tasks /cha...
OLS regression Dependent variables: Overall motivation score Scores for constructs of motivation Independent variables: He...
OLS regression Mean score for overall motivation weighted by construct and clustered by health facility Explanatory variab...
Strengths and Limitations Strengths Factor analysis widely used method to study the dimensionality of a set of variables H...
Summary Revealed dimensions relevant to motivation of workers in the DRC Intrinsic and extrinsic motivation framework Resu...
Próximos SlideShares
Carregando em…5
×

Measuring the motivation of health workers in the DRC

120 visualizações

Publicada em

Rishma maini

Publicada em: Saúde
0 comentários
0 gostaram
Estatísticas
Notas
no profile picture user

  • Seja o primeiro a comentar

  • Seja a primeira pessoa a gostar disto

Sem downloads
Visualizações
Visualizações totais
120
No SlideShare
0
A partir de incorporações
0
Número de incorporações
0
Ações
Compartilhamentos
0
Downloads
5
Comentários
0
Gostaram
0
Incorporações 0
Nenhuma incorporação

Nenhuma nota no slide
  • However, as with many studies there are a few strengths and limitations. In terms of using factor analysis as a method, I would say that the advantages are that it has been widely used to study the dimensionality of a set of variables. It is a powerful data reduction tool, allowing us to measure the core elements of motivation which are most relevant to workers in the DRC and not focus on any redundant attributes.
    It also allows us to develop more conservative, parsimonious tools for measuring motivation in the future. 62 questions is a lot to ask so being able to reduce this down prevents questionnaire fatigue in participants.
    However, the limitations are that factor analysis can be only as good as the data allows. In this example, we had to rely on less valid and reliable measures such as self-reports, which can introduce bias and be problematic for the factor analysis. There was also the challenge of ensuring we had enough variables to start with before conducting the factor analysis. As we didn’t know much about motivation of worker’s motivation in the DRC, we developed the original tool by integrating previous scales and discussing with partners and experts, however, unfortunately there was no time to do any qualitative work with health workers prior to administering the tool. This could have helped to refine the original items further and ensure that we were not missing anything that could be important. We did manage to conduct some qualitative interviews with workers following the survey, and reassuringly, the themes identified corroborated well with the themes identified through the factor analysis. The qualitative interviews also helped us better to understand why these dimensions of motivation were so important to workers.
    Another challenge was that it was difficult to know which theoretical framework of motivation or predefined constructs to include, as there are many out there in the literature. The Franco et al. framework has been widely used which is why it was chosen, but it may not have been most applicable to DRC. Looking at how our latent constructs grouped, it would seem that they were more suited to intrinsic and extrinsic motivation as opposed to determinants and outcomes.
    Other challenges with this study were that it was not possible to establish causality around the effects of removing PBF on motivation, as this is only a cross-sectional study. However, it is worth pointing out that the decision to withdraw PBF payments in the new programme was highly controversial, and although a before and after study had been proposed, partners had already removed the PBF payments by the time the study started. This is a good example of where real world programmatic decisions can impact on research.
  • So in summary – exploratory factor analysis revealed certain themes relevant to the motivation of health workers in the DRC. Obviously there are several limitations to the study, one of the most important being self-report bias. However, a survey of this kind was thought to be the most practical approach in this setting as data on more objective measures such as working hours and absenteeism is not consistently recorded in the routine information system. In addition, given this is a cross-sectional study, we cannot attribute causality between the withdrawal of performance-based payments and effects on motivation. Nonetheless, there were significant differences in the levels of motivation between the red zone and green zone workers and so this study indicates the potential effects on motivation when donors cease to fund PBF payments to workers.
    Factors could be broadly classified as either individual or external. This suggests a more fitting framework would have been that of intrinsic and extrinsic motivation.
    Nonetheless, in terms of future steps, it will be necessary to do some repeat testing of the tool in order to validate the factor structure perhaps using a confirmatory factor analysis.
     

    • Measuring the motivation of health workers in the DRC

    1. 1. Rishma Maini, Josephine Borghi, Natasha Palmer, David Hotchkiss Measuring the motivation of health workers in the DRC
    2. 2. Background: The DRC • Poor provision of basic services • Public sector health workers often not paid by government • Donors implementing PBF (performance based financing) programmes for short periods • Little known about health worker motivation in the DRC
    3. 3. Programme changes • Red zones – PBF programme was operating (now stopped) • Green zones – workers were never exposed to PBF
    4. 4. Aims • To measure the determinants and outcomes of motivation in health workers in the DRC • To compare any differences between workers who used to work in a PBF programme (red zones) with health workers who had never worked in a PBF programme (green zones).
    5. 5. Methods DATA COLLECTION • Health worker motivation survey administered in red and green zones, and outside of those zones. • 46 questions on determinants and 16 items on outcomes of motivation (13 constructs) answered using Likert scales ANALYSIS OF QUANTITATIVE DATA • Cronbach’s alpha on pre-defined constructs • Exploratory factor analysis on entire scale • OLS regression model to identify relationships between independent health worker and health facility variables and scores for latent constructs and overall motivation.
    6. 6. CHARACTERISTIC PROPORTION OF WORKERS (n=430) SEX Male 69% Female 31% AGE <30 years 11.2% 30-44 years 59.5% 45-60 years 26.1% >60 years 3.3% POSITION Doctor 0.9% Nurse 89.5% Laboratory worker 1.2% Pharmacy worker 1.4% Traditional birth attendant 2.8% Auxiliaries, medical and nursing assistants 4.2% Sample characteristics
    7. 7. Predefined constructs Construct Cronbach’s alpha Determinants Financial 0.6 Management 0.5 Job description 0.6 Workload 0.7 Training 0.7 Resources 0.6 Work harmony/relationships 0.6 Pride 0.7 Self-efficacy 0.5 Outcomes Timeliness/attendance 0.5 Conscientiousness 0.7 Commitment 0.1 Satisfaction 0.4 Determinants 0.8 Outcomes 0.6 Overall scale 0.8
    8. 8. Latent factors overall scale Intrinsic Extrinsic Individual factors (Competence and autonomy) Opportunities Job tasks /characteristics Working environment /relationships Financial factors Number of items 11 5 8 6 6 % of variance explained 30.3% 23.2% 22.8% 20.6% 20.1% Cronbach’s alpha 0.8 0.7 0.7 0.7 0.5 Cronbach’s alpha full scale 0.8 KMO test 0.8 (middling)
    9. 9. OLS regression Dependent variables: Overall motivation score Scores for constructs of motivation Independent variables: Health worker: Age, gender, health worker position/cadre, education, number of financial dependents, and years worked in position. Health facility: Urban-rural facility, reference or health centre, distance from village, number of services, total personnel, population catchment, and whether in previous PBF-supported zone or not .
    10. 10. OLS regression Mean score for overall motivation weighted by construct and clustered by health facility Explanatory variables β (SE) Full model Reduced model Years in position -0.002 (0.003) Population served 0.000 (0.000) Total personnel -0.006 (0.021) Urban (vs rural) -0.039 (0.069) Number of services 0.031 (0.020) Distance of facility from village -0.004 (0.004) Reference heath centre (vs heath centre) 0.131 (0.084) Nurse (versus other positions) 0.089 (0.064) Age 0.002 (0.002) Male (vs female) 0.058 (0.042) Number of dependents 0.006 (0.004) 0.008 (0.004)* University (vs school education) 0.025 (0.042) PBF removed -0.165 (0.053)** -0.162 (0.047)** Constant 2.546 (0.445)*** 3.458 (0.039)*** Pseudo R2 0.11 0.06 Number observations (n) 348 348 *P≤0.05 **p≤0.01 ***p≤0.001
    11. 11. Strengths and Limitations Strengths Factor analysis widely used method to study the dimensionality of a set of variables High power of data reduction – focus on core elements and not redundant attributes Yielded a parsimonious scale Limitations Factor analysis only as good as the data allows Inclusion of enough questions to start with Multiple motivational frameworks in the literature Could not establish causality around PBF payment removal and motivation
    12. 12. Summary Revealed dimensions relevant to motivation of workers in the DRC Intrinsic and extrinsic motivation framework Results indicate potential consequences of donors exiting from PBF programmes Future research Repeated field testing of needed - confirmatory factor analysis

    ×