Performance-based contracting for hospitals: a mixed methods analysis of impacts on patient outcomes, equity and efficiency in a middle income country
Lead Research Organisation:
American University of Beirut
Department Name: Faculty of Health Sciences
Abstract
The Ministry of Public Health (MoPH) is the largest insurer for hospitalizations in Lebanon, covering 52% of citizens and about 240,000 hospital admissions annually. Typical patients are those who are unable to afford health insurance, are unemployed or self-employed, are older than 64 years, or have a chronic disease (e.g. diabetes, hypertension, cancer). To provide these services, the MoPH contacts with 135 public and private hospitals.
Since 2001 the reimbursement rate paid to hospitals by the ministry was determined by the results of a hospital accreditation process. However, over the past several years evidence has accumulated that this was not an effective way to manage the relationship between the MoPH and hospitals. Importantly, the ministry has imperfect information on the performance of hospitals. In 2014 the MoPH began a transition away from the accreditation-only contracting system, and towards one based on performance, including patient outcomes.
The main purpose of this research is to develop a performance-based contracting (PBC) system between the MoPH and hospitals in Lebanon, and evaluate its impact on patients and the health system. Such contracting means that the ministry would reward hospitals that perform better by paying them a higher base rate per patient. We will investigate what factors may affect hospital performance and how hospitals responded to this intervention.
There has been much work on PBC in health services over the past two decades. However the evidence to support its benefit to patients and cost-effectiveness presents mixed results. One of the main reasons for this is the limited number of strongly designed studies. Recent evidence from England and the United States has also found that positive effects such as reduced readmissions and mortality may be limited to the short-term, and underlined the importance of PBC measurement, context and design. In low/middle-income countries (LMIC) evidence is still more limited, though PBC holds much promise as it may have larger impact on health outcomes given the potential to improve. However this also means it may have larger unintended or negative consequences, and should be designed with great care and close monitoring of impact.
In designing PBC, it is important to determine how performance will be measured and how we would evaluate its impact. In our research, at the patient level we will look at changes in patient readmissions for specific conditions, which could indicate inadequate treatment, hospital-acquired infections, or other causes. We will also look at the proportion of patients admitted to each hospital in terms of their age and presence of chronic diseases, as some hospitals may 'cherry-pick' and avoid patients with more complex conditions. We will also develop a patient satisfaction questionnaire, and use it to measure the satisfaction of patients that would be representative of the hospital they were treated at. At the health system/hospital level we will look at the utilization and cost of different services, as well as how complex are the cases being admitted to each hospital (case-mix). We will compare the results for these performance indicators before and after implementation of PBC, and investigate any changes. We will also interview a sample of hospital managers to understand how hospitals responded to PBC and what changes they may have made to affect their performance, such as better application of clinical guidelines, increased training or incentives to the health workforce.
We will actively share our research findings with stakeholders and the public through various channels including developing knowledge translation materials and events such as seminars and policy roundtables. The knowledge gained will be used to inform future PBC development in Lebanon and similar initiatives in LMICs.
Since 2001 the reimbursement rate paid to hospitals by the ministry was determined by the results of a hospital accreditation process. However, over the past several years evidence has accumulated that this was not an effective way to manage the relationship between the MoPH and hospitals. Importantly, the ministry has imperfect information on the performance of hospitals. In 2014 the MoPH began a transition away from the accreditation-only contracting system, and towards one based on performance, including patient outcomes.
The main purpose of this research is to develop a performance-based contracting (PBC) system between the MoPH and hospitals in Lebanon, and evaluate its impact on patients and the health system. Such contracting means that the ministry would reward hospitals that perform better by paying them a higher base rate per patient. We will investigate what factors may affect hospital performance and how hospitals responded to this intervention.
There has been much work on PBC in health services over the past two decades. However the evidence to support its benefit to patients and cost-effectiveness presents mixed results. One of the main reasons for this is the limited number of strongly designed studies. Recent evidence from England and the United States has also found that positive effects such as reduced readmissions and mortality may be limited to the short-term, and underlined the importance of PBC measurement, context and design. In low/middle-income countries (LMIC) evidence is still more limited, though PBC holds much promise as it may have larger impact on health outcomes given the potential to improve. However this also means it may have larger unintended or negative consequences, and should be designed with great care and close monitoring of impact.
In designing PBC, it is important to determine how performance will be measured and how we would evaluate its impact. In our research, at the patient level we will look at changes in patient readmissions for specific conditions, which could indicate inadequate treatment, hospital-acquired infections, or other causes. We will also look at the proportion of patients admitted to each hospital in terms of their age and presence of chronic diseases, as some hospitals may 'cherry-pick' and avoid patients with more complex conditions. We will also develop a patient satisfaction questionnaire, and use it to measure the satisfaction of patients that would be representative of the hospital they were treated at. At the health system/hospital level we will look at the utilization and cost of different services, as well as how complex are the cases being admitted to each hospital (case-mix). We will compare the results for these performance indicators before and after implementation of PBC, and investigate any changes. We will also interview a sample of hospital managers to understand how hospitals responded to PBC and what changes they may have made to affect their performance, such as better application of clinical guidelines, increased training or incentives to the health workforce.
We will actively share our research findings with stakeholders and the public through various channels including developing knowledge translation materials and events such as seminars and policy roundtables. The knowledge gained will be used to inform future PBC development in Lebanon and similar initiatives in LMICs.
Technical Summary
The Ministry of Public Health (MoPH) contracts with 135 public and private hospitals to provide hospitalization services to 52% of Lebanon's citizens. Since 2001 the MoPH has used a three-tier reimbursement rate for hospitals based on their results in a hospital accreditation process. Evidence has accumulated over the past several years that this contracting mechanism was not appropriate, and the MoPH has turned to performance-based contracting (PBC) to address its lack of information on services provided (classical principal-agent problem).
The research team will develop a PBC intervention between the MoPH and hospitals, and evaluate its impact at patient and health system/hospital levels. Our principle research question is: What factors determine the impact of performance-based hospital contracting to improve patient outcomes, equity and efficiency?
We will use a mixed-methods approach, with qualitative and quantitative components. The MoPH hospitalization database will be used to measure hospital case-mix, service utilization and cost, and 30-day readmissions for specific conditions/procedures. We will design a patient satisfaction questionnaire using focus group discussions, and apply it to measure patient satisfaction across hospitals. With one year of pre-intervention data available the baseline hospital performance will be calculated and used by the MoPH to set new hospital reimbursement rates. With one year of post-intervention data available we will evaluate PBC impact at patient and health system/hospital levels. We will also interview a sample of hospital managers to investigate changes hospitals undertook as a response to PBC (e.g. staff incentives, trainings).
We will share our findings using stakeholder seminars, policy roundtables, professional networks, policy briefs and other knowledge-translation tools, and use this to use our findings to inform future development of PBC in Lebanon and similar initiatives in low/middle-income countries.
The research team will develop a PBC intervention between the MoPH and hospitals, and evaluate its impact at patient and health system/hospital levels. Our principle research question is: What factors determine the impact of performance-based hospital contracting to improve patient outcomes, equity and efficiency?
We will use a mixed-methods approach, with qualitative and quantitative components. The MoPH hospitalization database will be used to measure hospital case-mix, service utilization and cost, and 30-day readmissions for specific conditions/procedures. We will design a patient satisfaction questionnaire using focus group discussions, and apply it to measure patient satisfaction across hospitals. With one year of pre-intervention data available the baseline hospital performance will be calculated and used by the MoPH to set new hospital reimbursement rates. With one year of post-intervention data available we will evaluate PBC impact at patient and health system/hospital levels. We will also interview a sample of hospital managers to investigate changes hospitals undertook as a response to PBC (e.g. staff incentives, trainings).
We will share our findings using stakeholder seminars, policy roundtables, professional networks, policy briefs and other knowledge-translation tools, and use this to use our findings to inform future development of PBC in Lebanon and similar initiatives in low/middle-income countries.
Planned Impact
This is an implementation-focused initiative, and as such the time-frame for benefits to be realized is within the duration of the project, and will include active engagement relevant to Lebanon and other low/middle-income countries (LMICs).
Who will benefit from this research?
1. Persons covered by the MoPH
2. Ministry of Public Health
3. LMICs interested in PBC
4. Public and private hospitals
5. Healthcare professionals
How will they benefit from this research?
The Ministry of Public Health (MoPH) provides hospitalization coverage for all citizens without health insurance. These are typically the unemployed or self-employed, are older than 64 years, or have a chronic disease (e.g. diabetes, hypertension, cancer). These citizens are the most vulnerable in society and would be the first to benefit from improved hospital performance. The performance-based contracting design creates incentives for hospitals to decrease patient readmissions and out-of-pocket payment, and improve patient satisfaction, all of which have direct effect on patients. Additionally, allowing for measures such as case-mix and co-morbidities to interpret performance results will discourage hospitals from avoiding patients with more complex conditions.
The MoPH is also a direct beneficiary of this initiative. Using PBC as a tool to address the current principal-agent problem will allow the ministry to have greater ability to incentivize improved patient outcomes and health system equity and efficiency. This is also of increased importance given the budgetary limitations on healthcare as well as the socioeconomic conditions affected by the refugee influx from the conflict in neighboring Syria.
An additional benefit of research findings is to inform similar performance-based initiatives that are being considered by the MoPH for its network of primary health centres throughout the country.
A main purpose of this research is to effectively contribute to very limited evidence available on PBC in health services, in particular for low/middle-income countries. This research will inform on what factors affected PBC impact at patient and health system/hospital levels, and provide some insight into how PBC may be designed to sustainably integrate in the healthcare system and incentivize improved performance in limited settings. We will actively share our findings using knowledge-translation events and tools, with particular emphasis on PBC in LMICs, to benefit policymakers and stakeholders who may be considering similar initiatives in their countries.
Public and private hospitals contracted with the MoPH will also benefit from this initiative by being reimbursed for hospitalization expenditures based on their performance. The linkage between reimbursement and performance (rather than accreditation) is a considerable improvement which is also supported by hospitals, and would increase mutual trust and transparency between them and the MoPH. It is also important to note that processes by which hospitals improved their performance may provide valuable lessons for uptake by lesser-performing hospitals.
Healthcare professionals may also be beneficiaries of this initiative, in particular physicians and nurses. Hospitals may have different responses to PBC, such as providing incentives or trainings to personnel or improving working conditions.
Who will benefit from this research?
1. Persons covered by the MoPH
2. Ministry of Public Health
3. LMICs interested in PBC
4. Public and private hospitals
5. Healthcare professionals
How will they benefit from this research?
The Ministry of Public Health (MoPH) provides hospitalization coverage for all citizens without health insurance. These are typically the unemployed or self-employed, are older than 64 years, or have a chronic disease (e.g. diabetes, hypertension, cancer). These citizens are the most vulnerable in society and would be the first to benefit from improved hospital performance. The performance-based contracting design creates incentives for hospitals to decrease patient readmissions and out-of-pocket payment, and improve patient satisfaction, all of which have direct effect on patients. Additionally, allowing for measures such as case-mix and co-morbidities to interpret performance results will discourage hospitals from avoiding patients with more complex conditions.
The MoPH is also a direct beneficiary of this initiative. Using PBC as a tool to address the current principal-agent problem will allow the ministry to have greater ability to incentivize improved patient outcomes and health system equity and efficiency. This is also of increased importance given the budgetary limitations on healthcare as well as the socioeconomic conditions affected by the refugee influx from the conflict in neighboring Syria.
An additional benefit of research findings is to inform similar performance-based initiatives that are being considered by the MoPH for its network of primary health centres throughout the country.
A main purpose of this research is to effectively contribute to very limited evidence available on PBC in health services, in particular for low/middle-income countries. This research will inform on what factors affected PBC impact at patient and health system/hospital levels, and provide some insight into how PBC may be designed to sustainably integrate in the healthcare system and incentivize improved performance in limited settings. We will actively share our findings using knowledge-translation events and tools, with particular emphasis on PBC in LMICs, to benefit policymakers and stakeholders who may be considering similar initiatives in their countries.
Public and private hospitals contracted with the MoPH will also benefit from this initiative by being reimbursed for hospitalization expenditures based on their performance. The linkage between reimbursement and performance (rather than accreditation) is a considerable improvement which is also supported by hospitals, and would increase mutual trust and transparency between them and the MoPH. It is also important to note that processes by which hospitals improved their performance may provide valuable lessons for uptake by lesser-performing hospitals.
Healthcare professionals may also be beneficiaries of this initiative, in particular physicians and nurses. Hospitals may have different responses to PBC, such as providing incentives or trainings to personnel or improving working conditions.
Publications
Khalife J, Barakat E
(2018)
Report on patient co-payments under the coverage of the Ministry of Public Health
Barakat E
(2019)
Report on the results of the national patient satisfaction survey
Khalife J
(2020)
Hospital performance and payment: impact of integrating pay-for-performance on healthcare effectiveness in Lebanon
in Wellcome Open Research
Khalife J
(2020)
Hospital performance and payment: impact of integrating pay-for-performance on healthcare effectiveness in Lebanon
in Wellcome Open Research
| Description | The Lebanese Ministry of Public Health (MoPH) first integrated pay-for-performance (P4P) into setting hospital reimbursement tiers in 2014. This was intended to improve effectiveness by decreasing unnecessary hospitalizations as well as improve fairness by including risk-adjustment in the setting of hospital performance scores that determined reimbursement tier. Our project developed the P4P integration and aimed to evaluate the impact of its implementation on hospital performance. We first evaluated the impact of the 2014 intervention, to develop the evaluation framework which would also be used for subsequent evaluations. The project's original intervention intended for 2017 was divided into two interventions, due to the delay of the approval of the then-health minister in issuing the P4P implementation. The first of these was the engagement of hospitals on the new performance scoring model in January-March 2018; the second was the application of reimbursement tiers based on the new model, in May 2019. Sufficient data-points were available for interrupted-time-series analysis of the impact of the first intervention, while the second was interrupted by the October 2019 protests in Lebanon and subsequent impact of COVID-19. In addition to the P4P interventions of 2014 and 2018, the MoPH implemented the introduction of Third Party Administrators (TPAs) at all contracted hospitals in September 2016. The purpose of the TPAs was to improve effectiveness by reducing unnecessary hospitalizations, which was similar to the goal of P4P integration. Although this intervention was independent of our project, the MoPH benefitted from our evaluation framework for preliminary evaluation of TPA impact. Due to the relation with P4P, we also include this intervention in our analysis of P4P integration; thus in effect evaluation the impact of the 2014, 2016 and 2018 interventions. Our findings suggest that all three interventions had an impact on case complexity among hospitals contracted by the MoPH. Overall the impact was an increased casemix index. The greatest level changes followed the 2016 TPA intervention followed by the 2014 P4P intervention. The greatest trend changes followed the 2014 P4P intervention. The 2018 P4P intervention increased casemix index in private but not public hospitals. It is possible that the impact of the 2018 P4P intervention was more limited as the improvement space for decreasing unnecessary hospitalizations may have been mostly exhausted by the 2014 and 2016 interventions. Evaluations of long-term effectiveness of P4P initiatives have often found limited or no improvement. Our findings are consistent with those of other investigations which suggested that short-term response to P4P engagement and implementation may differ from long-term response, particularly in terms of effectiveness. Our findings also suggest that effective hospital regulation can be achieved through the systematic collection and analysis of readily available routine data. Low and middle-income countries that lack casemix adjustment and incentives for improving hospital performance may choose to adopt similar approaches and monitoring systems to measure and improve hospital performance over time. |
| Exploitation Route | We have developed an evaluation framework for interventions that impact hospitalizations in Lebanon that are covered by the main public payer, the Ministry of Public Health. Our P4P model also provides the Ministry with a transparent and evidence-based approach to evaluate hospital performance and improve fairness and effectiveness of contracting with public and private hospitals. Our findings overall indicate improved health system effectiveness as a result of the adoption of pay-for-performance in the short-term. The longer-term impact was however more limited, and our findings are consistent with initiatives in other countries which similarly found more positive impacts of P4P in the short rather than long term. |
| Sectors | Healthcare |
| Description | In May 2019 the Minister of Public Health approved the implementation of the new hospital reimbursement tiers based on the pay-for-performance (P4P) model our project had developed. As a result, 140 hospitals received their performance results and associated reimbursement tier. This is also expected to have downstream implications that benefit patient safety. Based on our evaluation of the 2014 P4P intervention, we expect this to increase healthcare system effectiveness and efficiency, particularly by decreasing unnecessary hospitalizations. Our evaluation of the engagement of hospitals regarding the revised P4P model, which took place January-March 2018, has found an increased case complexity resulting in private but not public hospitals. Further investigations of the longer-term impact of the May 2019 are currently being evaluated, particularly on hospital case complexity and 30-day readmissions. The development of the reimbursement categorization has increased the transparency between the Ministry and hospitals, and is expected to increase barriers for political favoritism or corruption. |
| Sector | Healthcare |
| Impact Types | Policy & public services |
| Description | Assessment of patient out-of-pocket co-payments for hospitalization; presented to the Ministry of Public Health policymakers for consideration of increased risk pooling interventions and financial protection of high-cost co-payments. |
| Geographic Reach | National |
| Policy Influence Type | Contribution to a national consultation/review |
| Description | Calculation of hospital performance scores for 2016 contracting round, using 2015 hospitalization data. The results were subsequently used by the MoPH to review and re-categorize the reimbursement rates of hospitals that improved their performance (based on the 2014 mixed-model scoring). |
| Geographic Reach | National |
| Policy Influence Type | Contribution to a national consultation/review |
| Description | Hospital performance score calculation and reimbursement categorization for the MoPH, using the 2018 performance model. |
| Geographic Reach | National |
| Policy Influence Type | Contribution to a national consultation/review |
| Description | Hospital performance score calculation and reimbursement categorization for the MoPH, using the 2019 performance model and latest available hospitalization data. |
| Geographic Reach | National |
| Policy Influence Type | Contribution to a national consultation/review |
| Impact | In May 2019 the new Minister of Public Health approved to implement our project's main intervention, which was the revision of the reimbursement tiers of hospitals contracted with the Ministry, based on the project's pay-for-performance model. Our evaluation of the 2014 intervention which resulted in improved system effectiveness (decreased unnecessary hospitalizations and improved coding), we anticipate a similar impact to result from the 2019 intervention. This will be formally evaluated throughout 2020 when the required data will become available for post-intervention impact evaluation. The immediate response received from hospitals following this intervention was encouraging, particularly regarding increased transparency and dissemination of performance results through a standardized hospital performance card. This is also expected to decrease the influence of potential favoritism or corruption with regards to hospital reimbursement tier classification. |
| Description | Impact assessment of Third-Party Administrators (TPAs) on hospitalizations at private hospitals under MoPH coverage |
| Geographic Reach | National |
| Policy Influence Type | Contribution to a national consultation/review |
| Impact | Using algorithms developed for indicators to be used in our project, we conducted an assessment of the impact of Third Party Administrators (TPAs) that were contracted by the Ministry of Public Health(MoPH) in September 2016 to improve implementation of hospital admission guidelines, regarding cases being admitted under MoPH coverage at private hospitals. This was also an opportunity to assess the sensitivity and validity of our tools in detecting changes in the policy environment. In addition to the quantitative analysis, we conducted 5 interviews with TPA front-line personnel and managers to increase understanding of their practices. We developed a report for the MoPH with our findings, including recommendations for improved practice. There was a general positive impact from TPAs in decreasing non-indicated hospitalizations and improving coding practices at hospitals. The impact of the report informed discussions held among policymakers and the Minister of Public Health on deciding whether or not to extend contracting with TPAs. |
| Description | Provision of evidence to influence MoPH policy/practice regarding utilization review and auditing; identification of limitations in current case coding system; and in dialogue with the Syndicate of Private Hospitals on future development of hospital contracting. |
| Geographic Reach | National |
| Policy Influence Type | Implementation circular/rapid advice/letter to e.g. Ministry of Health |
| Impact | 1. Utilization review and auditing We have supported the development of algorithms and programing scripts to analyze 15 diagnoses admitted under MoPH coverage in private and public hospitals that it is contracted with. We have identified 37 hospitals who were outliers in terms of cost or length of stay indicators, and disseminated this information to the MoPH utilization review function and auditing committee for further examination and/or investigation. This has informed the regulation of hospital services, however specific impacts to hospitals have not yet been specified, but will be subsequently available from the MoPH auditing committee (e.g. cases investigated, reimbursement deductions). 2. Identification of limitations in current case coding system We have examined the structure and approach of coding (using ICD10 and surgical procedure coding) of hospitalization cases, in particular those regarding cases with prostheses. We have identified specific limitations of the current manner of coding such prosthesis cases inside the MoPH hospitalization database, and have discussed this with the MoPH Service of Hospitals, Dispensaries and Medical Professions and the IT Department. Discussions are ongoing on adopting an alternative coding structure for prosthesis cases. 3. Dialogue with the Syndicate of Private Hospitals on future development of hospital contracting The MoPH has used the latest results of hospital performance scores and hospital casemix index (2015 data) that have been calculated by the project team, to inform its discussions with the Syndicate regarding developments of the hospitalization system and of future performance-based contracting. |
| Title | National Patient Satisfaction Survey tool |
| Description | We have developed a patient satisfaction survey tool, using the results of the previously held focus group discussions and review of the questionnaire previously used by the Ministry of Public Health. This tool has been piloted and used for sampling former patients hospitalized under the Ministry of Public Health coverage at various hospitals throughout 2018. |
| Type Of Material | Physiological assessment or outcome measure |
| Year Produced | 2018 |
| Provided To Others? | No |
| Impact | This tool has allowed us to measure patient satisfaction at hospital level, the results of which have been used for the patient satisfaction component of the hospital performance score, which is used to determine the hospital reimbursement tier. The results of this tool will also be used to conduct a pre/post intervention analysis to assess any changes to this component following the introduction of the new contracting model. |
| Title | Casemix index algorithms and programming scripts for hospital all-stay cases |
| Description | We have developed the algorithms for the calculation of hospital casemix index, for each of medical cases, surgical cases and mixed medical-surgical cases. Among medical cases this includes a combination of short-, medium- and long-stay cases each with their own casemix indices and relative weights, based on length-of-stay status. A final combination script provides the overall casemix index for each hospital contracted with the Ministry of Public Health (MoPH). The surgical algorithm includes two optional approaches: one that includes the updated prosthesis codes adopted in 2015, and another that excludes these codes for better comparability with previous years. These algorithms have been expressed in STATA programming scripts, and results have been calculated for all hospitals throughout the period between 2011 and 2015. |
| Type Of Material | Computer model/algorithm |
| Provided To Others? | No |
| Impact | The initial scripts developed calculated annual casemix index; the results were analyzed and discussed by the research team, based on which we developed the scripts further to calculate casemix index by quarter (20 quarters spanning 2011-2015). Analysis of quarterly data revealed clear seasonality in casemix index, and an increasing trend since 2011. Further work is continuing to develop monthly casemix index and analyze impact of previous policy interventions using interrupted time-series analysis. This work allows us to establish a clearer baseline with which to compare our upcoming project intervention, as well as tools and indicators to monitor impact. |
| Title | Focus group discussions with former patients under MoPH coverage |
| Description | We conducted 8 Focus Group Discussions (FGDs) with 42 participants who are Lebanese citizens that had been admitted to hospital under MoPH during the preceding 3 months. This aimed at understanding the patient's perception regarding patient experience and quality of healthcare services, in order to develop the patient satisfaction survey used. |
| Type Of Material | Database/Collection of data |
| Year Produced | 2017 |
| Provided To Others? | No |
| Impact | The results of the FGDs were used to develop the patient satisfaction survey tool, which will populate the associated component of the hospital performance scoring model in future contracting rounds, and allow subsequent impact assessment of the project's intervention. |
| Title | Hospital 30-day readmission algorithms automated by month and by hospital, for 6 selected conditions |
| Description | Assessment of 30-day all-cause readmissions by month and by hospital, for each of general/any-readmission, stroke, pneumonia, cholecystectomy, cesarean section, and myocardial infarction. This STATA script is currently being used to investigate the impact of the 2018-2019 interventions on readmission, using interrupted time-series analysis. |
| Type Of Material | Computer model/algorithm |
| Year Produced | 2022 |
| Provided To Others? | No |
| Impact | The algorithm developed was used to calculate monthly readmissions for the six selected conditions, for the period between January 2011 to December 2019. This includes the project interventions, whose impact will be evaluated using interrupted time-series analysis. The 30-day readmissions for each of general/any-readmission, stroke, pneumonia and cholecystectomy had been used within the pay-for-performance model; besides the interrupted time-series analysis, these will also be further compared with those that had not been included within the model (cesarean section and myocardial infarction). |
| Title | Hospital 30-day readmission algorithms for 10 conditions |
| Description | These algorithms are used to calculate 30-day readmission results for hospitals for 10 conditions: general admission, myocardial infarction, stroke, pneumonia, coronary artery by-pass graft, caesarean section, cholecystectomy, colo-rectal surgery, hip replacement, and knee replacement. Validation was completed on remaining seven indicators. Results of 2016 were also calculated (in addition to previous 2011-2015 results) at national and hospital levels (135 hospitals) for all conditions. |
| Type Of Material | Computer model/algorithm |
| Year Produced | 2017 |
| Provided To Others? | No |
| Impact | The validation process results were used to make algorithm improvements. The readmission results provided the research team and MoPH the scale of readmissions within each condition, and informed decision-making during the selection process regarding which indicators to include in the performance-based contracting scoring model. The team selected to use 2-year readmission rates rather than 1-year, and including 30-day readmission results for general admissions, pneumonia, stroke and cholecystectomy in the performance model's first intervention (2018). |
| Title | Hospital Performance Scoring Model (2018; project intervention) |
| Description | A composite scoring model was developed to include 6 components for the new hospital performance scores: casemix index, readmissions, patient satisfaction, accreditation, intensive care unit proportion, and elderly cases proportion. Review of the literature and current practices were conducted to inform the methodology used. Hospitals were engaged in the component decision-making process, including the Syndicate of Private Hospitals. Component weights were determined by an analytic hierarchical process and further refined using an iterative process with pre-defined rules for balancing components. Four variations of the model were presented to the MoPH policymakers for selection and implementation, each with certain components varying within a limited range. |
| Type Of Material | Computer model/algorithm |
| Year Produced | 2018 |
| Provided To Others? | No |
| Impact | The design of the new model is intended to incentivize improved casemix, reduced readmissions, improved patient satisfaction, reduced exclusion of elderly patients, and increased investment in needed ICU capacity. The stateholders (hospitals) have been actively engaged and supportive of the new model. The impact of the model (project's intervention) will be assessed in the post-intervention period. |
| Title | Hospital casemix index algorithms automated by month, for all hospitalization cases |
| Description | Assessment of casemix index results by year and quarter (previously developed) proved to be insufficient for analysis of project intervention impact, in particular considering the decision to use interrupted time-series analysis for evaluation. To increase both sensitivity of analysis tools and datapoints available, algorithms were developed to calculate hospital casemix index by month. In addition, all algorithms were fully automated within STATA to reduce possibility of human-error in the calculation process. |
| Type Of Material | Computer model/algorithm |
| Year Produced | 2017 |
| Provided To Others? | No |
| Impact | The results for 2011-2016 were calculated, establishing a baseline for the project's pre-intervention period, as well as allowing assessment of tool sensitivity by analyzing impact of past interventions on the hospitalization system that were not related to the project (third-party administrators intervention in 2016). In addition, the 2016 casemix index results were calculated and used to populate the casemix component of the hospital performance scoring model for the 2018 contracting round (intervention). |
| Title | Mixed-cases casemix algorithm |
| Description | Analysis of previously developed casemix results of mixed (medical-surgical) cases increased understanding of hospital coding practices of such cases and limitations of hospitalization database. A new algorithm was developed to calculate hospital casemix index for mixed cases, to correct the inadequacies of the previous algorithm. |
| Type Of Material | Computer model/algorithm |
| Year Produced | 2017 |
| Provided To Others? | No |
| Impact | Developing a reliable and valid algorithm for mixed cases allowed the project and MoPH to include mixed cases (6% of total cases) in the hospital performance scoring model, covering 100% of all hospitalizations. |
| Title | System-level casemix index calculation algorithms, for each of medical, surgical and mixed hospitalization cases |
| Description | Three algorithms were made for the calculation of casemix indices for each of medical, surgical and mixed hospitalization cases, at system-level, whose formula and approach differed from hospital-level algorithms previously developed. All algorithms were made in STATA, with results calculated for all hospitalizations at monthly and annual intervals from January 2011 to December 2017. These algorithms have been used to analyze the impact of the 2014 contracting intervention and the same approach will be used to evaluate the 2018 knowledge-sharing events and 2019 intervention of changing hospital reimbursement tiers (based on performance). |
| Type Of Material | Computer model/algorithm |
| Year Produced | 2018 |
| Provided To Others? | No |
| Impact | Using the results obtained, we have been able to address whether or not casemix index is a tool that is sufficiently sensitive to detect changes in hospital performance due to contracting intervention changes. The results of the interrupted time-series analyses conducted using casemix data has highlighted the hospitalization types and medical/surgical codes which are most likely to be impacted by contracting interventions. We therefore expect that if the subsequent events held (2018 and 2019 intervention) have a significant impact, this should be reflected in casemix changes using our system-level algorithms. Discussion with policymakers at the Ministry of Public Health regarding this has also increased their confidence in the potential intervention impact and analytical approach which has been comprehensive in terms of analyzing all hospitalization types and at system-level. |
| Description | Collaboration of project team at AUB, LU and MoPH |
| Organisation | Lund University |
| Country | Sweden |
| Sector | Academic/University |
| PI Contribution | The Ministry of Public Health has provided in-kind support by committing to the project personnel from each of the Department of Information Technologies, Department of Statistics, and the Service of Hospitals, Dispensaries and Medical Professions. Lund University has provided indirect support by waiving/funding 35% of overhead costs. |
| Collaborator Contribution | Ministry of Public Health has supported the project in the following manner: - Mrs. Rita Freiha (Patient Satisfaction coordinator, MoPH) has been involved on a daily basis in supporting the project team in developing the patient satisfaction activities and supporting development of various indicators to be used. - Dr. Hilda Harb (Head of Department of Statistics, MoPH) and Dr. Jihad Makouk have been involved on a regular basis in supporting the development of various indicators to be used, and in interpretation and analysis of indicator results. - Mrs. Jenny Romanos (Head of IT Department, MoPH) and Mrs. Rabiaa Rachid (IT personnel) have supported the project team with the extraction of all necessary data from the MoPH Hospitalization Database. - Provided office space, facilities and supplies for research coordinator and two part-time research assistants Lund University has provided indirect support by internally funding the 18% faculty overhead and waiving the additional 17% overhead. Collaboration between individuals at MoPH and LU began in 2014 with a part-time doctoral student accepted at Lund University to focus on hospital performance-based contracting. Individuals from AUB were also involved since 2009 in the process leading up to the current project. However collaboration at the institutional level and resource commitment began in 2016 with the launch of this project. |
| Impact | All outputs are the result of this joint collaboration project team. The research team has multi-disciplinary expertise, across health systems and policies, health economics and financing, medicine, epidemiology, medical sociology and qualitative methodology. |
| Start Year | 2016 |
| Description | Collaboration of project team at AUB, LU and MoPH |
| Organisation | Ministry of Public Health |
| Country | Lebanon |
| Sector | Public |
| PI Contribution | The Ministry of Public Health has provided in-kind support by committing to the project personnel from each of the Department of Information Technologies, Department of Statistics, and the Service of Hospitals, Dispensaries and Medical Professions. Lund University has provided indirect support by waiving/funding 35% of overhead costs. |
| Collaborator Contribution | Ministry of Public Health has supported the project in the following manner: - Mrs. Rita Freiha (Patient Satisfaction coordinator, MoPH) has been involved on a daily basis in supporting the project team in developing the patient satisfaction activities and supporting development of various indicators to be used. - Dr. Hilda Harb (Head of Department of Statistics, MoPH) and Dr. Jihad Makouk have been involved on a regular basis in supporting the development of various indicators to be used, and in interpretation and analysis of indicator results. - Mrs. Jenny Romanos (Head of IT Department, MoPH) and Mrs. Rabiaa Rachid (IT personnel) have supported the project team with the extraction of all necessary data from the MoPH Hospitalization Database. - Provided office space, facilities and supplies for research coordinator and two part-time research assistants Lund University has provided indirect support by internally funding the 18% faculty overhead and waiving the additional 17% overhead. Collaboration between individuals at MoPH and LU began in 2014 with a part-time doctoral student accepted at Lund University to focus on hospital performance-based contracting. Individuals from AUB were also involved since 2009 in the process leading up to the current project. However collaboration at the institutional level and resource commitment began in 2016 with the launch of this project. |
| Impact | All outputs are the result of this joint collaboration project team. The research team has multi-disciplinary expertise, across health systems and policies, health economics and financing, medicine, epidemiology, medical sociology and qualitative methodology. |
| Start Year | 2016 |
| Description | A talk or presentation - The Patient-centered Care, A New Healthcare Model |
| Form Of Engagement Activity | A talk or presentation |
| Part Of Official Scheme? | No |
| Geographic Reach | National |
| Primary Audience | Professional Practitioners |
| Results and Impact | The pay-for-performance initiative undertaken by the Ministry of Public Health was shared with audience members within the presentation on "Shifting from Volume-Based to Value-Based Care: Policy and Practice Implications" by Prof.Fadi El-Jardali, on 27th Septmber 2019, at the conference organized by the Bellevue Medical Center, Lebanon. This included the preliminary results of the P4P intervention and indicators used for the developed model for setting hospital reimbursement tiers based on hospital performance. |
| Year(s) Of Engagement Activity | 2019 |
| Description | Engaging Lebanese hospitals with the Ministry of Public Health: linking payment to performance |
| Form Of Engagement Activity | Participation in an activity, workshop or similar |
| Part Of Official Scheme? | No |
| Geographic Reach | National |
| Primary Audience | Industry/Business |
| Results and Impact | A stakeholders' engagement event was held on 18th January at the Riviera Hotel, Beirut, between the Lebanese Ministry of Public Health (MoPH) and public and private hospitals. The event allowed the MoPH to share developments regarding its contracting model, which links hospitals' performance to the reimbursement rates they receive for hospitalizations under MoPH coverage. The contracting model impacts the reimbursement rates of about 135 hospitals, including more than 240,000 hospital admissions annually. The event was held with the participation of the MoPH Director-General, the President of the Syndicate of Private Hospitals, and numerous hospital representatives including general directors, financial directors and quality managers. The event began with the presentation of proposed performance indicators to be adopted in the updated contracting model, including patient outcomes such as readmissions and patient satisfaction. This was followed by a discussion engaging hospitals representatives who expressed support and enthusiasm to implement interventions aiming at enhancing quality of care and patient safety. Participants shared their thoughts on the proposed indicators, and also requested more details on the methods used for certain indicators. The concluding note of the event had the parties agree that this is the strong step towards improving hospitals performance, and there was general agreement on the commitment to improving patient outcomes, equity and efficiency, and continuing the development of the model linking improved hospital performance to higher reimbursement levels. |
| Year(s) Of Engagement Activity | 2018 |
| Description | Faculty of Health Sciences Seminar, AUB |
| Form Of Engagement Activity | A talk or presentation |
| Part Of Official Scheme? | No |
| Geographic Reach | National |
| Primary Audience | Professional Practitioners |
| Results and Impact | The seminar was held at the Faculty of Health Sciences auditorium, American University of Beirut, on 12th February 2020. Participants included FHS and visiting faculty members, medical and nursing practitioners from the AUB Medical Center, graduate and undergraduate students, and other persons involved in the healthcare sector in Lebanon. The purpose of this seminar was to disseminate the research findings of our project's evaluation of the 2014 intervention, and engage participants regarding the Ministry's new pay-for-performance model that was implemented in May 2019 to categorize hospital reimbursement tiers. |
| Year(s) Of Engagement Activity | 2020 |
| Description | Meetings with previous and current Ministers of Public Health |
| Form Of Engagement Activity | A talk or presentation |
| Part Of Official Scheme? | No |
| Geographic Reach | National |
| Primary Audience | Policymakers/politicians |
| Results and Impact | The project team has engaged both the previous Minister of Public Health (April 2018) and the new Minister of Public Health (March 2019) regarding the new hospital performance-based contracting model. The purpose of these events was to share the components of the new model and approach of the project intervention, and gather feedback from the Minister and his advisors. Although the meeting with the previous Minister (2018) was generally positive, the Minister did not sign off on the new hospital reimbursement tiers that resulted from our new model, as he was concerned of political impact to his political party prior to the May 2018 national parliament elections. Therefore, although he had approved of the model, he did not implement the updated reimbursement tiers. Following the elections, due to a caretaker government being in place, the Minister no longer had the legal authority to update reimbursement tiers; thus we had to wait until a new government was formed (January 2019). We engaged the current Minister of Public Health in April 2019, who has been very positive of this initiative and has approved of the new model and verbally committed to the implementation. We are currently awaiting his signature of the Ministerial decree to legalize the updated reimbursement tiers. |
| Year(s) Of Engagement Activity | 2018,2019 |
| Description | Syndicate of Private Hospitals engagement |
| Form Of Engagement Activity | A talk or presentation |
| Part Of Official Scheme? | No |
| Geographic Reach | National |
| Primary Audience | Policymakers/politicians |
| Results and Impact | Several discussions have been held between the project team and the leadership of the Syndicate of Private Hospitals throughout the past year, including on preliminary results of casemix indices and other indicators. This has included an event where the project inception and development was presented to an audience of the Syndicate's leadership, and subsequent debate held. The engagement of the Syndicate is an important factor to support the eventual impact of the project's upcoming intervention. |
| Year(s) Of Engagement Activity | 2016 |
| Description | Technical event engaging quality and financial personnel of Lebanese public and private hospitals, regarding the new model for hospital performance-based contracting. |
| Form Of Engagement Activity | Participation in an activity, workshop or similar |
| Part Of Official Scheme? | No |
| Geographic Reach | National |
| Primary Audience | Professional Practitioners |
| Results and Impact | As a follow-up to the first engagement event held for hospital directors in January 2018, we hosted a technical event at the Ministry of Public Health with about 135 hospital quality and financial personnel and managers participating. The purpose of this event was to share the methodological details of the indicators used, including indicator definitions, rules used in scoring, and features of the hospital performance card. An engaging discussion followed with various technical questions and suggestions being exchanged. The event concluded with noting the next steps to be taken, namely: the selection of indicator weights, calculation of hospital scores and planned future engagements. |
| Year(s) Of Engagement Activity | 2018 |