Functionality of Quality Improvement Teams in Primary Healthcare Facilities in Tanzania: Findings from Star Rating Assessment ()
1. Introduction
Primary health care (PHC) services are important for people’s health and well- being. The World Health Organization (WHO) has shown that “scaling up PHC interventions could save 60 million lives and increase average life expectancy by 3.7 years by 2030” ( WHO, 2021 ). Quality of services in PHC facilities is an important determinant for the people to utilize the services ( Hokororo et al., 2020 ; Nwokoro et al., 2022 ; Otovwe & Elizabeth, 2017 ). The Coronavirus disease 2019 (COVID-19) has revealed the weakness of PHC systems both in developed and developing countries ( Lauriola et al., 2021 ; Muzammil & Lopes, 2021 ). The Astana Declaration of 2018 has made it clear that strengthening PHC services is essential for achieving universal health coverage (UHC) ( WHO, 2019 ). To achieve UHC in lower- and middle-income countries (LMICs), the health service delivery must be improved ( Masefield et al., 2020 ). PHC is a top priority in the global health agenda following the Astana Declaration in 2018 where many nations reaffirmed their promises ( WHO, 2019 ). The effects of COVID-19 on delivery of PHC services were felt in many countries ( Edelman et al., 2021 ; Phiri et al., 2022 ; Rezapour et al., 2022 ; Yadav et al., 2021 ). This has led to calls for reimagining PHC ( Barış et al., 2022 ); and a need for strengthening PHC across four key levers namely political commitment and leadership, governance and policy, funding and allocation of resources, and engagement of communities and other stakeholders ( Peiris et al., 2021 ). The lever on funding and allocation of resources has received attention in the report of “The Lancet Global Health Commission on financing primary health care”, which “argues that all countries need to both invest more and invest better in PHC by designing their health financing arrangements—mobilising additional pooled public funding, allocating and protecting sufficient funds for PHC, and incentivising providers to maintain the health of the populations they serve ( Hanson & Commissioners, 2022 ). Despite all the efforts, poor quality of care is often a limiting factor ( Kruk et al., 2018 ). This is due to inadequate accountability systems that hinder the efficient operationalization of healthcare service delivery ( Saltman & Ferroussier-Davis, 2000 ; Scott et al., 2014 ). On the other hand, quality of health care cannot be achieved without teamwork consideration; hence, to address gaps in healthcare quality and safety team-based methods to patient care and quality improvement need to be taken into considerations ( Meltzer et al., 2010 ). Teams whose members commit to achieving the highest levels of work performance, take part in decision- making, and encourage creative ideas are likely to achieve their pre-determined objectives ( Strating & Nieboer, 2009 ). In order to build team spirit and strengthen accountability systems in PHC facilities, it is important to strengthen management capacities and leadership skills for in-charges of PHC facilities ( Kim et al., 2021 ).
With a focus on institutionalizing the culture of continuous quality improvement at all levels of health care delivery by improving the working environment, strengthening leadership, commitment, coordination, and relationships among key stakeholders to achieve sustainable quality improvement programs, the Tanzania Health Quality Improvement Framework (TQIF) second edition was developed in 2011 ( MOHSW, 2011 ). The TQIF also stipulated the quality governance structures at health facilities, which is entrusted in three teams, namely: Health Facility Management Team (HFMT); Quality Improvement Team (QIT); and Work Improvement Teams (WITs). The functions of HFMTs, QITs, and WITs as well as reporting lines are elaborated in Figure 1 ( MOHSW, 2011 ). Some studies have documented the beneficial effects of quality improvement teams in Tanzanian healthcare facilities ( Kacholi & Mahomed, 2020 ; Kamiya et al., 2017 ). These studies report that, quality improvement teams have a positive effect on staff job satisfaction and workload, as well as on patient waiting times and records management ( Ishijima et al., 2016 ; Kacholi & Mahomed, 2020 ; Kamiya et al., 2017 ). Notwithstanding the recognized benefits of QI teams and the rise in their acceptance and mainstreaming in Tanzanian hospitals, a sizable portion of these teams have been underperforming the targets ( MoHCDGEC, 2019 ). Several elements, including management and leadership support and the availability of financial resources to support team activities, have been linked to team effectiveness as reported by previous researches ( Savelsbergh et al., 2010 ). Also, another study in Tanzania by Kacholi & Mahomed (2021) in 2021 noted that the QITs in selected Regional Referral Hospitals despite of having good performance, they were perceived to underperform in terms of involvement of health care workers in development and implementation of hospital QI plans, and inadequate patient waiting time reduction ( Kacholi & Mahomed, 2021 ). Yet, modifying the actions of specific employees or institutions to adhere to predetermined criteria is a crucial factor in improving the quality of healthcare delivery ( Kruk et al., 2018 ). This should extend to QITs to foster its activeness and responsiveness provided that community trust and overall perceived quality of care have been shown to be important predictors of care seeking behavior indicating that health facilities management viability is an important improvement target ( Kruk et al., 2009 ).
As part of strengthening PHC services in Tanzania, from 2007 to 2017, Tanzania Mainland implemented the Primary Health Services Development Programme (PHSDP) which aimed at provision of accessible quality health services to all Tanzanians ( MoHSW, 2007 ). Also, in 2014, the Government of Tanzania introduced the implementation of “Big Results Now” initiative in health sector; which resulted into design of Star Rating Assessment (SRA) as one of its interventions aiming at improving performance of PHC facilities. In the financial year 2015/2016 Tanzania Mainland conducted a countrywide assessment of all PHC facilities and assigned a star ranging from 0 to 5 stars based on the status of quality services that were being provided; and in 2017/2018 countrywide second assessment was done ( Gage et al., 2020 ; Yahya & Mohamed, 2018 ). This study is therefore aiming at assessing the functionality of QITs in PHC facilities in Tanzania in the context of SRA for the second assessment that was conducted in fiscal year 2017/2018. Unlike other studies including ( Ishijima et al., 2016 ; Kacholi & Mahomed, 2020 ; Savelsbergh et al., 2010 ), This study is going to be conducted in wider geographical space covering all twenty six (26) Tanzania mainland regions focusing on functionality of quality improvement teams standards for Primary Healthcare facilities in the context of SRA as prescribed in Tanzania Quality Improvement Framework ( MOHSW, 2011 ) and hence more generalizable.
This paper is organized to incorporate introduction that encompasses background and review of related literature, methods encompassing study design, data management and analysis, definition of study variables, and study population, results entailing findings of the study, discussion, conclusion and reccomendations.
2. Methods
2.1. Study Design
This study employed the use of analytical cross-sectional design to ascertain the functionality of QITs in the context of the second SRA conducted during the fiscal year 2017/2018. The second Star Rating Assessment conducted in a financial year 2017/2018 is the most recent data available. The QITs functionality was measured in terms of availability of the team, regular quarterly meetings of the team with recorded meeting minutes, and conducting self-internal assessments on quality of services using tools that meet national standards.
2.2. Data Management and Analysis
Data were extracted from SRA data base for each health facility. They were transposed and manipulated in Microsoft Excel to form a single dataset. The dataset was imported to Stata SE 16 for descriptive and inferential statistical analysis. Data were geospatially visualized with the aid of Quantum Geographical Information Software (QGIS) 3.16 Hannover. Chi square test tests were was employed for comparing proportions, and modified Poisson regression was used to estimate prevalence ratios of the determinant factors for QITs functionality. Comparative analyses were made to ascertain differences in % coverage of health facilities in QITs functionality in reference to 4 performance categories, i.e., less than 20%, between 20% to less than 40%, between 40% to less than 60% and greater than or equal to 60% assigned as poor, weak, good progress and good performance respectively.
This study exhaustively analyzes and compares these components regionwide, in regard of facility characteristics embracing facility levels (District Hospitals, Health Centers and Dispensaries), facility location (Urban and Rural) and other significant variables including Human Resources for Health (HRH) availability, functionality of HFMTs, health facility autonomy, working conditions and client satisfaction. The regional variation documentation highlights the need to understand the broader subnational systems and context which may influence culture and effectiveness at the facility level ( Macarayan et al., 2019 ).
2.3. Power of the Study
Power refers to the probability that a significance test will produce a significant result when the null hypothesis is false ( Hallahan & Rosenthal, 1996 ). Its major determinants are significance level and sample size with a desired level being 0.8 i.e., 80% ( Cohen, 2016 ). To ascertain the extent to which QIT functionality prevails, one sample proportion test was employed. With a sample of 5933 health facilities, the significance level of 0.05 (5%) and a prevalence of 0.225 (22.5%) of health facilities with functional QITs, the estimated statistical power of this study for one sample proportion z test is 1 (100%) as shown by Stata outputs in Figure 2 below.
Source: Analysis from SRA dataset (2017/2018).
Figure 2. Statistical power of the study.
2.4. Goodness of Fit Test for a Poisson Regression Model
We determined an association between binary variable (the facility having functional QIT or not) and independent variables (facility location, facility type, HRH availability, HMT functionality, facility autonomy, working conditions and clients satisfaction) to estimate the predictors of QIT functionality at the PHC facilities. Both Univariate and Multivariate Poisson regression model was used to estimate crude and adjusted prevalence ratios regarding factors influencing QIT functionality. Major findings were reported with Multivariate Poisson regression where Deviance and Pearson goodness-of-fit tests indicated a model being a perfect fit (2480.35, 3335.02, p = 1.00) (refer Figure 3 below). P values less than or equal to 0.005 were considered statistically significant.
Source: Analysis from SRA dataset (2017/2018).
Figure 3. Goodness of fit tests for the model and Poisson regression outputs.
2.5. Definition of Study Variables and Constructs (Table 1)
Table 1. Study variables and their definition.
Source: TQIF (2011-2016), HRHSW Strategic Plan (2014-2019) and Star Rating Assessment Tools (2017/2018).
2.6. Study Population
This paper aims at revealing the prevailing state of QITs functionality in PHC facilities in Tanzania which is located in Eastern part of Africa by considering all 26 administrative regions of Mainland (Figure 4) visited for second SRA in the financial year 2017/2018 with a focus of 5933 PHC facilities that account for 81.4% of visited health facilities. 18.6% of visited PHC facilities were dropped from analysis due to having missing values with consideration of greater than 5% missing variables to most of indicators necessary for this study ( Dong & Peng, 2013 ).
Source: Sketch on geodata from National Bureau of Statistics 2021.
Figure 4. Map of Tanzania showing distribution of visited regions.
3. Results
3.1. Functionality of QITs in Primary Healthcare Facilities
Findings revealed that 1338 out of 5933 (22.5%) of QITs are functional, 27.8% of urban located health facilities have functional QITs which is significantly higher than 21.1% of health facilities that are located in rural areas (p < 0.001). On the other hand, district hospitals/hospitals at district level had 41.3% of health facilities with functional QITs, this is significantly higher compared to 34.7% and 20.8% of health centers and dispensaries with functional QITs respectively. Moreover, 22.6% of public owned health facilities had functional QITs which is almost similar to 22.5% of privately owned (p = 0.966).
Region-wise coverage of the availability of functional QITs geospatially is shown in Figure 5, which shows that no region had desirable coverage of health facilities with functional QITs. Kilimanjaro is the leading and only region with good progress by having 41.1% of health facilities with functional QITs. Kigoma and Songwe regions are noted with the poorest coverage, i.e., 4.6% and 7.3% of functional QITs respectively.
3.2. Availability of Formally Appointed QITs with Terms of References
From the findings, it was revealed that 1802 out of 5911 (69.5%) of QITs have formally appointed members with well-defined terms of references, 72.7% of
Source: Sketch on geodata from National Bureau of Statistics 2021.
Figure 5. Regionwise performance of HFQIT functionality.
urban located health facilities have availability of formally appointed QITs with well-defined terms of references for each appointed member which is significantly higher than 68.6% of health facilities that are located in rural areas (p = 0.004). On the other hand, district hospitals/hospitals at district level had 93.8% of health facilities with availability of formally appointed QITs, this is significantly higher compared to 89.0% and 66.7% of health facilities with the availability of formally appointed QITs in health centers and dispensaries respectively.
Regionwise coverage of availability of formally appointed QITs geospatially is shown in Figure 6, which shows that, with the exception of Songwe, Kigoma, Morogoro, Iringa, and Ruvuma that showed good progress, the remaining regions had good performance coverage, i.e., greater than 60%.
3.3. Quarterly QITs Meetings on Regular Schedule
Findings revealed that 3222 out of 5919 (54.3%) of QITs meet on regular schedule quarterly; 56.7% of urban located health facilities meet on regular schedule
Source: Sketch on geodata from National Bureau of Statistics 2021.
Figure 6. Regionwise coverage of health facilities with availability of formally appointed QITs.
quarterly which is insignificantly higher than 53.8% of health facilities that are located in rural areas (p = 0.07). On the other hand, district hospitals/hospitals at district level had 90% of QITs that meet on regular schedule quarterly, this is significantly higher compared to 74.8% and 51.3% in health centers and dispensaries respectively (p < 0.001). Moreover, 55.5% of public owned health facilities have QITs that regularly meet, this is significantly higher than 50.7% of privately owned (p = 0.002).
Regionwise coverage of QITs that meet quarterly on regular schedule is geospatially shown in Figure 7 below.
3.4. Conduct and Report of Self-Internal Assessment on Quality of Services
Findings revealed that 2455 out of 5908 (41.6%) of QITs conduct and report internal self-assessment on quality of services using tools that meet national standards including Star Rating Assessment Tool and Standard Based Management and Recognition for Infection Prevention and Control. It was also revealed that,
Source: Sketch on geodata from National Bureau of Statistics 2021.
Figure 7. Regionwise coverage of health facilities with QITs that meet quarterly.
45.1% of QITs in urban located health facilities conduct and report on internal self-assessment on quality services by using standard tools, this is significantly higher than 40.6% of QITs in health facilities that are located in rural areas (p = 0.04). On the other hand, district hospitals/hospitals at district level had 61.3% of QITs that conduct and report on internal self-assessment, this is significantly higher compared to 74.8% and 51.3% for health centers and dispensaries respectively (p < 0.001). Moreover, public-owned health facilities had 42.0% of QITs that conduct and report on internal self-assessment which is insignificantly higher than 40.0% of privately owned (p = 0.201).
Regionwise coverage of health facilities with QITs that conduct and report self-internal assessments on the quality of services is geospatially shown in Figure 8 below.
3.5. Factors Influencing QITs Functionality
As shown in Table 2 below, urban located health facilities are 13% more likely to
Source: Sketch on geodata from National Bureau of Statistics 2021.
Figure 8. Regionwise coverage of health facilities with QITs that conduct and report internal self-assessment.
Table 2. Factors influencing functionality of QITs.
Source: Star Rating Assessment Findings (2017/2018).
have functional QITs compared to rural located (APR = 1.13, p = 0.06). Health facilities with satisfactory HRH availability are 29% more likely to have functional QITs (APR = 1.29, P < 0.001), health facilities with functional HFMTs are about 3 times likely to have functional QITs (APR = 2.86, P < 0.001), autonomous health facilities are 42% more likely to have functional QITs (APR = 1.42, P < 0.001), health facilities with favorable conditions are 24% more likely to have functional QITs (APR = 1.24,p = 0.001), health facilities with satisfied clients during the day of assessment are 24% more likely to have functional QITs (APR = 1.24, p = 0.002). Moreover, health facility level was not a good predictor of QITs functionality whereby health centers were 35% more likely to have functional QITs (p = 0.21), dispensaries were 20% less likely to have functional QITs compared to district hospitals (p = 0.35).
4. Discussion
As observed from the findings, most of the PHC facilities do not have functional QITs. This may be one among reasons for poor healthcare delivery in Tanzania ( Afnan-Holmes et al., 2015 ; Kruk et al., 2017 ). Findings from other studies reveal facility management including functional QITs as an important consideration to improved healthcare delivery and consequently enhance patient satisfaction ( Gage et al., 2020 ; Larson et al., 2019 ). Poor QITs functionality affects how smoothly WITs activities are implemented. WITs play a critical role in discovering, assessing, and resolving difficulties in work units and departments (i.e., frontlines of care delivery to patients/clients), and report and seek support from QITs ( MOHSW, 2011 ). A study by Kruk and colleagues in 2009 revealed that, with irresponsible facility teams, patient satisfaction cannot be achieved and may consequently lead to limited care seeking resulted by mistrust to health services provided in PHC facilities ( Kruk et al., 2009 ).
A study by Clarke-Deelder and colleagues which aimed at developing a measure of user experience of care among caregivers of sick children in rural Tanzania, “found that 69% of variation in user experience was within facilities and 31% was between facilities” ( Clarke-Deelder et al., 2022 ). The findings indicated that in order to improve user experience of care, more efforts need to be within a particular facility. Therefore, given the low functionality of QITs in our study, it indicates that this need to be addressed as one of the measures that can help a facility to improve the quality of services both technical quality and interpersonal aspect of quality of care (user experience of care). Therefore, putting in place strategies for improving the performance of QITs in PHC facilities is an important endeavor for all stakeholders.
With the low functionality of QITs in PHC facilities in Tanzania, further research is needed through strengthening the use of implementation science approach ( Alonge et al., 2019 ; Bauer et al., 2015 ), is needed to document what works in the efforts to improve the functionality of QITs and hence the quality of services in PHC facilities. Researchers and policymakers may need to borrow a leaf from organizational psychology in studying the functionality of QITs (as well as, other governance structures in PHC facilities, which in one way or another, their roles contribute to quality improvement and assurance of services provided in PHC facilities) ( Rosen et al., 2018 ).
5. Conclusion
Institutionalization of healthcare quality improvement is yet to be satisfactorily achieved. It is required that healthcare facilities establish and maintain quality improvement as an integral component of healthcare delivery. Quality activities ought to be incorporated into the structure of health facilities and continuously implemented with the involvement and engagement of all staff of a facility (both technical and support staff). For this goal to be accomplished, the functionality of QITs and other governance structures in PHC facilities is crucial. Based on the findings on predictors for functionality, the HFMTs need to be strengthened in all PHC facilities as a measure to ensure that they can support the QITs to perform better to improve the quality of service delivery.
Authors’ Contributions
· Conception and design: Eliudi S. Eliakimu, Syabo M. Mwaisengela, Erick S. Kinyenje, Joseph C. Hokororo, Chrisogone J. German Radenta P. Bahegwa, Yohanes S. Msigwa, Mbwana M. Degeh, Laura E. Marandu, Lilian D. Lyakurwa, Alpha E. Cholobi, Bushi N. Lugoba, Edwin C. Mkwama, Talhiya A. Yahya, Mohamed A. Mohamed, Omary N. Kipuli and Ruth R. Ngowi
· Acquisition of data: Syabo M. Mwaisengela, Chrisogone J. German, Erick S. Kinyenje, Omary A. Nassoro, Ruth R. Ngowi, Laura E. Marandu, Radenta P. Bahegwa and Yohannes S. Msigwa
· Analysis and interpretation of data: Syabo M. Mwaisengela, Eliudi S. Eliakimu, Joseph C. Hokororo, Radenta P. Bahegwa and Ruth R. Ngowi
· Drafting of the manuscript: Syabo M. Mwaisengela, Eliudi S. Eliakimu, Michael F. Habtu, Chrisogone J. German, Joseph C. Hokororo, Radenta P. Bahegwa and Ruth R. Ngowi
· Critical revision of the manuscript for important intellectual content: Eliudi S. Eliakimu, Syabo M. Mwaisengela, Michael F. Habtu, Erick S. Kinyenje, Joseph C. Hokororo, Chrisogone J. German Radenta P. Bahegwa, Yohanes S. Msigwa, Mbwana M. Degeh, Laura E. Marandu, Lilian D. Lyakurwa, Alpha E. Cholobi, Bushi N. Lugoba, Edwin C. Mkwama, Talhiya A. Yahya, Mohamed A. Mohamed, Omary N. Kipuli and Ruth R. Ngowi
· Statistical analysis: Syabo M. Mwaisengela, Eliudi S. Eliakimu, Chrisogone J. German, Joseph C. Hokororo, Radenta P. Bahegwa and Ruth R. Ngowi
· Obtaining funding: Not applicable/None
· Administrative, technical, or material support: Eliudi S. Eliakimu, Michael F. Habtu, Lilian D. Lyakurwa, Mbwana M. Degeh, Joseph C. Hokororo, Alpha E. Cholobi
· Supervision: Eliudi S. Eliakimu, Joseph C. Hokororo, Mbwana M. Degeh, Chrisogone J. German and Erick S. Kinyenje.