Delivering a Multi-disease Screening Tool to Migrant Populations
- Conditions
- StrongyloidiasisChagas DiseaseFemale Genital Mutilation Type II StatusTuberculosisFemale Genital Mutilation Type III StatusHepatitis BHepatitis CSchistosomiasisFemale Genital Mutilation Type I StatusHiv
- Interventions
- Other: Clinical decision support system for screening of migrants using the ISMiHealth software tool.
- Registration Number
- NCT05868005
- Lead Sponsor
- Barcelona Institute for Global Health
- Brief Summary
Migrants' overall health status may be improved by increasing the detection of certain infectious diseases and other conditions for which effective care is available. This can be achieved through a systematic screening of these conditions using innovative and digital solutions implemented in routine health care.
This study aims to evaluate the implementation of a screening programme for migrants at primary care level in two different settings of Spain (Catalonia and Andalusia) using an innovative digital and user-friendly software tool (ISMiHealth). In Catalonia, the ISMiHealth tool has already been integrated into the Electronic Patient Record (EPR) system (eCAP) as part of a pilot study in 2018; currently, the research team aims to validate the tool in a higher number of primary care centres in this area. Therefore, a pragmatic cluster randomised controlled trial will be conducted with two parallel groups, in which selected centres using the novel software ISMiHealth will be compared to others that follow the current routine practice. On the other hand, in Andalusia a pilot cluster randomised controlled trial will be carried out, where the ISMiHealth tool will be implemented in the EPR system (DIRAYA) to evaluate the preliminary effectiveness of the tool in other settings.
The ISMiHealth software is a clinical decision support system that provides recommendations for primary healthcare professionals on screening for targeted conditions. It currently includes: 7 communicable diseases (Human immunodeficiency virus, Hepatitis B and C virus, Tuberculosis, Chagas diseases, strongyloidiasis and schistosomiasis) and one key health condition (female genital mutilation). Through routinely collected variables (country of birth, age, and sex), the software performs an individualised risk assessment and provides real-time prompts to healthcare professionals on screening for the selected health conditions. In any case, health professionals will be responsible for requesting screening tests and/or referrals to specialists.
- Detailed Description
1. Objectives
General objective: To develop an integrated and primary care based, communicable diseases and female genital mutilation (FGM) screening programme for migrant populations in Spain in order to promote better health and integration.
Specific objectives:
* Aim 1: To develop a Spanish consensus guidance on migrant screening recommendations based on existing European examination/screening programmes and on a consensus of a country-expert transdisciplinary national collaborative network.
* Aim 2: To scale-up and validate a simple and user-friendly innovative digital tool that will be integrated into the local electronic patient record (EPR) system of primary care in two different settings (Catalonia and Andalusia), and that aims to facilitate targeted screening to migrants presenting for a routine appointment. This digital tool will support health professionals to follow best-practice recommendations, while considering patients' individual characteristics (country of birth, age, sex) and therefore specific risk factors.
2. Study design
This study will be developed into two sub-studies in order to accomplish the specific objectives.
2.1 Study 1 related to aim 1. Spanish consensus guidance on migrant screening recommendations.
1. A comparative review and analysis of existing screening programmes at regional, national and international level will be performed.
2. A report with the reviewed results will be developed and presented in the consensus workshops.
3. A consensus workshop with national experts will be carried out by study site to establish which diseases will be included in the final screening algorithm and what will be defined as the screening criteria for each condition. The tentative elements to include are Human immunodeficiency virus (HIV), Hepatitis B virus (HBV), Hepatitis C virus (HCV), Tuberculosis (TB), selected parasitic infections and FGM. For most of the communicable diseases, the screening criteria will be fully aligned with the current European Centre for Disease Prevention and Control (ECDC) screening guidelines and recommendations.
4. To measure the level of agreement with the recommendations, a questionnaire based on a Likert Scale of 5 points (1- Strongly disagree, 2- Disagree, 3- Neutral, 4- Agree and 5- Strongly agree) will be sent to the different experts invited to the workshops.
2.2 Study 2 related to aim 2. Operational implementation of a screening programme using the ISMiHealth software and evaluation of its impact and equity.
1. Study design and subjects. A pragmatic cluster randomised controlled trial will be conducted in 35 Primary care centers (PCCs) of Catalonia to explore and assess the effectiveness of the software tool ISMiHealth \[see sample size calculation in the Statistical Analysis Plan (SAP)\]. The screening programme will be implemented in a real-life context in the selected PCCs, and the receivers/users of the tool will be all health professionals (general practitioners and nurses) working at the intervention centres. On the other hand, the tool and the outcomes of the study will be evaluated in the migrant patients, the indirect beneficiaries of the tool. Additionally, a pilot clustered randomised controlled trial will be conducted in Andalusia, implementing the screening programme in six PCCs to explore preliminary effectiveness data.
In the site of Catalonia approximately 840 health professionals will participate in the intervention while in Andalusia approximately 140 health professionals will participate.
ISMiHealth set a series of rules that provide real-time prompts to health professionals on screening of infectious diseases and FGM in migrants. Therefore, when an individual comes to the PCC for any reason, the health professional will receive a message in the EPR system, inserted as a pending task assignment, with a recommendation on the conditions that should be considered for screening. For this, the software will perform an individual risk-assessment based on the patient's background characteristics (country of birth, age and sex). The software tool is also able to identify if a person had already a diagnosis of any of the conditions included in the algorithm \[based on the International Classification Diagnosis codes, Ninth version (ICD-9) or Tenth version (ICD-10)\] or if a diagnostic test had been performed for any condition included in the program. In such cases, the automated electronic prompt does not appear for that condition.
Before the intervention, we will hold training sessions with the same screening recommendations for both the intervention and control centres but the latter ones will not initiate the tool/prompts onto the EPR system. In addition, a pilot test will be conducted with 3 health professionals to assure the correct functioning of the tool. At last, the ISMiHealth software will be implemented in the EPR systems of the participating PCCs.
2. Implementation strategy. During the implementation phase, real-time monitoring of the strategy will be performed (by a data manager and a study coordinator from each PCC) in order to identify opportunities for improvement and optimization of the process. The technical monitoring will help identify errors in the implementation of the alerts in the EPR system, to be mitigated in real-time. Quantitative indicators will also be provided by the Information Service of the Primary Care Services (SISAP) to evaluate the outcomes of the implementation.
3. Intervention
The implementation of the screening programme will be facilitated using a simple and user-friendly tool that helps health professionals to follow best-practice recommendations while taking into account patients' individual characteristics: country of birth , age, sex.
In Catalonia, the digital tool has already been integrated into the EPR system (eCAP), a program shared by all PCCs that are part of the Catalan Institute of Health, as part of a pilot study. The three demographic variables (country of birth, age, and sex) are routinely collected by the PCCs' administrative staff and documented in electronic medical records. Once the administrator collects these variables in the computer system, the tool will generate an alert that will be reflected once the patient's medical history is reopened.
In Andalusia, the software tool will be generated from demographic data stored in the Database of Users of the Andalusian Health Service (SAS, as its Spanish acronym). Once the health professional accesses the digital tool, he/she will have to mark in a check-list produced by the tool the conditions that the patient has already been screened for. In both study sites, the data of the patients is not registered nor collected by the tool, it is only used to produce the alert.
4. Data analysis
Since the ISMiHealth tool is in different phases of implementation in each site, and the algorithms are based on different recommendations, there will be two independent data analyses: i) in Catalonia, the health impact of the tool will be evaluated in 35 PCCs; and ii) in Andalusia, the preliminary effectiveness data will be estimated in the six PCCs where the tool will be pilot tested. See SAP attached.
Recruitment & Eligibility
- Status
- COMPLETED
- Sex
- All
- Target Recruitment
- 980
Not provided
Not provided
Study & Design
- Study Type
- INTERVENTIONAL
- Study Design
- PARALLEL
- Arm && Interventions
Group Intervention Description Primary care centres with the ISMiHealth software tool. Clinical decision support system for screening of migrants using the ISMiHealth software tool. Primary care centres implementing the screening programme through the ISMiHealth software (tool-based arm).
- Primary Outcome Measures
Name Time Method Comparison of the detection rate per month of all aggregated infections (HIV, HBV, HCV, TB, T.cruzi, S.stercoralis and Schistosoma spp. infections) between the intervention and control centres Over at least 5 years until the end of the intervention (1 year). In the Andalusian site, syphilis, latent TB and intestinal parasites will also be included. The monthly detection rate will be based on positive serologies, chest radiographies, the ICD-9 (for Andalusia) or ICD-10 (for Catalonia) of FGM and/or gynaecologist referrals, within the migrant patients who visited their assigned centre during the intervention period. Also, positive tuberculin skin test (TST) and/or Interferon Gamma Release Assay (IGRA) and stool samples will be considered for the Andalusian site. Control and intervention PCCs will be compared before and after the implementation.
- Secondary Outcome Measures
Name Time Method Associations between the screening performance of HBV and/or HCV (yes/no) and the levels of platelets (U/µL) will be analysed using the data collected from the EPR system Through study completion, 1 year. Bivariate and multivariate analysis (Fisher's exact tests, Chi-square tests and/or logistic regressions) will be performed using the screening performance of each individual condition (yes-if the patient has been tested for HBV and/or HCV /no-if the patient was not tested) as the outcome variable and the levels of platelets as the predictor variable.
Number of diagnosed individuals with follow-up visits in the hospital of reference Through study completion, 1 year. The number of migrant patients under follow-up in the hospital after a diagnosis of an infection or FGM will be estimated using the Minimum basic dataset (MBDS).
Comparison of the detection rate per month of each individual condition, the infections and FGM cases, between the intervention and control centres Over at least 5 years until the end of the intervention (1 year). The monthly detection rate will be based on positive serologies, chest radiographies, ICD-9/ICD-10 of FGM and/or gynaecologist referrals, TST/IGRA and stool samples (for Andalusia), within the migrant patients who visited their assigned centre during the intervention period. Control and intervention PCCs will be compared before and after the implementation.
Associations between the screening performance of all aggregated infections (yes/no) and patients' sex (female/male) will be analysed using the data collected from the EPR system Through study completion, 1 year. Bivariate and multivariate analysis (Fisher's exact tests, Chi-square tests and/or logistic regressions) will be performed using the screening performance of the aggregated infections (yes-if the patient has been tested for at least one of the seven infections during the study period/no-if the patient was not tested) as the outcome variable and patients' sex as the predictor variable.
Associations between the screening performance of all aggregated infections (yes/no) and patients' immunosuppression status (yes/no) will be analysed using the data collected from the EPR system Through study completion, 1 year. Bivariate and multivariate analysis (Fisher's exact tests, Chi-square tests and/or logistic regressions) will be performed using the screening performance of the aggregated infections (yes-if the patient has been tested for at least one of the seven infections during the study period/no-if the patient was not tested) as the outcome variable and patients' immunosuppression status (yes- if the patient has an ICD-code diagnosis of an immunosuppressive disorder and/or have a prescription of an immunosuppressive drug/no- if the patient is not immunosuppressed) as the predictor variable.
Associations between the screening performance of each individual condition (yes/no), including FGM, and patients' sex (female/male) will be analysed using the data collected from the EPR system Through study completion, 1 year. Bivariate and multivariate analysis (Fisher's exact tests, Chi-square tests and/or logistic regressions) will be performed using the screening performance of each individual condition (yes- yes-if the patient has been tested for one of the seven infections and/or screened for female genital mutilation/no-if the patient was not tested) as the outcome variable and patients' sex as the predictor variable.
Associations between the screening performance of each individual condition (yes/no), including FGM, and the type of region where the PCC is located in Catalonia, Spain (urban/rural) will be analysed using the data collected in the SISAP database Through study completion, 1 year. Bivariate and multivariate analysis (Fisher's exact tests, Chi-square tests and/or logistic regressions) will be performed using the screening performance of each individual condition (yes-if the patient has been tested for one of the seven infections and/or screened for female genital mutilation/no-if the patient was not tested) as the outcome variable and the type of region of the PCC as the predictor variable.
Associations between the screening performance of all aggregated infections (yes/no) and patients' age (in years) will be analysed using the data collected from the EPR system Through study completion, 1 year. Bivariate and multivariate analysis (Fisher's exact tests, Chi-square tests and/or logistic regressions) will be performed using the screening performance of the aggregated infections (yes-if the patient has been tested for at least one of the seven infections during the study period/no-if the patient was not tested) as the outcome variable and patients' age as the predictor variable.
Associations between the screening performance of all aggregated infections (yes/no) and the time that patients have been registered in the Spanish National Health System (in years) will be analysed using the data collected from the EPR system Through study completion, 1 year. Bivariate and multivariate analysis (Fisher's exact tests, Chi-square tests and/or logistic regressions) will be performed using the screening performance of the aggregated infections (yes-if the patient has been tested for at least one of the seven infections during the study period/no-if the patient was not tested) as the outcome variable and time registered in the Spanish Health System as the predictor variable.
Associations between the screening performance of each individual condition (yes/no), including FGM, and patients' area of birth (Africa/Latin America/Asia/Eastern Europe) will be analysed using the data collected from the EPR system Through study completion, 1 year. Bivariate and multivariate analysis (Fisher's exact tests, Chi-square tests and/or logistic regressions) will be performed using the screening performance of each individual condition (yes-if the patient has been tested for one of the seven infections and/or screened for female genital mutilation/no-if the patient was not tested) as the outcome variable and patients' area of birth as the predictor variable. The areas of birth will be based in the geographical division of the United Nations Statistical Division.
Number of diagnosed individuals under treatment Through study completion, 1 year. The number of migrant patients under treatment after a diagnosis will be estimated using data from the EPR systems and the MBDS.
Comparison of the number of early HBV and HCV diagnoses Over at least 5 years until the end of the intervention (1 year). The number of early diagnoses of HBV and HCV will be assessed using the levels of transaminases, platelets, bilirubin, and clotting parameters from blood analyses of migrant patients diagnosed with HBV or HCV. The early diagnoses will be compared between the control and intervention PCCs before and after the implementation. If possible, we will estimate the monthly detection rate of early diagnoses and also compare it between the intervention and control PCCs.
Associations between the screening performance of all aggregated infections (yes/no) and if the patient fulfilled the criteria of the screening recommendation (yes/no) will be analysed using the data collected from the EPR system Through study completion, 1 year. Bivariate and multivariate analysis (Fisher's exact tests, Chi-square tests and/or logistic regressions) will be performed using the screening performance of the aggregated infections (yes-if the patient has been tested for at least one of the seven infections during the study period/no-if the patient was not tested) as the outcome variable and fulfilled the criteria of the screening recommendation (yes- if the patient had the criteria of being tested for at least one infection, criteria generated by the algorithm of the ISMiHealth software using the variables country of birth, age and sex/no- if the patient did not meet the criteria) as the predictor variable.
Associations between the screening performance of all aggregated infections (yes/no) and the arm of the study (Intervention/Control) will be analysed using the data collected from the EPR system Through study completion, 1 year. Bivariate and multivariate analysis (Fisher's exact tests, Chi-square tests and/or logistic regressions) will be performed using the screening performance of the aggregated infections (yes-if the patient has been tested for at least one of the seven infections during the study period/no-if the patient was not tested) as the outcome variable and the arm of the study as the predictor variable.
Associations between the screening performance of all aggregated infections (yes/no) and the presence of another condition (yes/no) will be analysed using the data collected from the EPR system Through study completion, 1 year. Bivariate and multivariate analysis (Fisher's exact tests, Chi-square tests and/or logistic regressions) will be performed using the screening performance of the aggregated infections (yes-if the patient has been tested for at least one of the seven infections during the study period/no-if the patient was not tested) as the outcome variable and the presence of another condition (yes- if the patient has an ICD-code of another condition \[co-morbidity\]/no- if the patient does not present another condition) as the predictor variable.
Comparison of the number of screening tests performed for all aggregated infections Over at least 5 years until the end of the intervention (1 year). The number of the screening tests performed for all aggregated infections (serological tests, chest radiographies, TST or IGRA tests and stool samples in the case of the Andalusian site) will be compared between the intervention and control PCCs before and after the implementation.
Comparison of the the number of screening tests performed for each individual condition Over at least 5 years until the end of the intervention (1 year). The number of the screening tests performed for each individual condition will be compared between the intervention and control PCCs before and after the implementation.
Associations between the screening performance of all aggregated infections (yes/no) and patients' area of birth (Africa/Latin America/Asia/Eastern Europe) will be analysed using the data collected from the EPR system Through study completion, 1 year. Bivariate and multivariate analysis (Fisher's exact tests, Chi-square tests and/or logistic regressions) will be performed using the screening performance of the aggregated infections (yes-if the patient has been tested for at least one of the seven infections during the study period/no-if the patient was not tested) as the outcome variable and patients' area of birth as the predictor variable. The areas of birth will be based in the geographical division of the United Nations Statistical Division.
Associations between the screening performance of each individual condition (yes/no), including FGM, and the time that patients have been registered in the Spanish Health System (in years) will be analysed using the data collected from the EPR system Through study completion, 1 year. Bivariate and multivariate analysis (Fisher's exact tests, Chi-square tests and/or logistic regressions) will be performed using the screening performance of each individual condition (yes-if the patient has been tested for one of the seven infections and/or screened for female genital mutilation/no-if the patient was not tested) as the outcome variable and time registered in the Spanish Health System as the predictor variable.
Associations between the screening performance each individual condition (yes/no), including FGM, and if the patient fulfilled the criteria of the screening recommendation (yes/no) will be analysed using the data collected from the EPR system Through study completion, 1 year. Bivariate and multivariate analysis (Fisher's exact tests, Chi-square tests and/or logistic regressions) will be performed using the screening performance of each individual condition (yes-if the patient has been tested for one of the seven infections and/or screened for female genital mutilation/no-if the patient was not tested) as the outcome variable and fulfilled the criteria of the screening recommendation (yes- if the patient had the criteria of being tested for at least one infection, criteria generated by the algorithm of the ISMiHealth software using the variables country of birth, age and sex/no- if the patient did not meet the criteria) as the predictor variable.
Comparison of the number of early HIV diagnoses Over at least 5 years until the end of the intervention (1 year). The number of early diagnoses of HIV will be assessed using the CD4 cell count of migrant patients diagnosed with an HIV infection. The early diagnoses will be compared between the control and intervention PCCs before and after the implementation. If possible, we will estimate the monthly detection rate of early diagnoses and also compare it between the intervention and control PCCs.
Associations between the screening performance of all aggregated infections (yes/no) and the region where the PCC is located (Terres de l'Ebre/Lleida/Tarragona/Costa de Ponent) will be analysed using the data collected in the SISAP database Through study completion, 1 year. Bivariate and multivariate analysis (Fisher's exact tests, Chi-square tests and/or logistic regressions) will be performed using the screening performance of the aggregated infections (yes-if the patient has been tested for at least one of the seven infections during the study period/no-if the patient was not tested) as the outcome variable and the region of the PCC as the predictor variable.
Associations between the screening performance of all aggregated infections (yes/no) and the type of region where the PCC is located in Catalonia, Spain (urban/rural) will be analysed using the data collected in the SISAP database Through study completion, 1 year. Bivariate and multivariate analysis (Fisher's exact tests, Chi-square tests and/or logistic regressions) will be performed using the screening performance of the aggregated infections (yes-if the patient has been tested for at least one of the seven infections during the study period/no-if the patient was not tested) as the outcome variable and the type of region of the PCC as the predictor variable.
Associations between the screening performance of each individual condition (yes/no), including FGM, and patients' age (in years) will be analysed using the data collected from the EPR system Through study completion, 1 year. Bivariate and multivariate analysis (Fisher's exact tests, Chi-square tests and/or logistic regressions) will be performed using the screening performance of each individual condition (yes-if the patient has been tested for one of the seven infections and/or screened for female genital mutilation/no-if the patient was not tested) as the outcome variable and patients' age as the predictor variable.
Associations between the screening performance of HBV and/or HCV (yes/no) and the levels of bilirubin (mg/dL) will be analysed using the data collected from the EPR system Through study completion, 1 year. Bivariate and multivariate analysis (Fisher's exact tests, Chi-square tests and/or logistic regressions) will be performed using the screening performance of each individual condition (yes-if the patient has been tested for HBV and/or HCV /no-if the patient was not tested) as the outcome variable and the levels of bilirubin as the predictor variable.
Associations between the screening performance of each individual condition (yes/no), including FGM, and the region where the PCC is located (Terres de l'Ebre/Lleida/Tarragona/Costa de Ponent) will be analysed using data collected in the SISAP database Through study completion, 1 year. Bivariate and multivariate analysis (Fisher's exact tests, Chi-square tests and/or logistic regressions) will be performed using the screening performance of each individual condition (yes-if the patient has been tested for one of the seven infections and/or screened for female genital mutilation/no-if the patient was not tested) as the outcome variable and the region of the PCC as the predictor variable.
Associations between the screening performance of each individual condition (yes/no), including FGM, and the arm of the study (Intervention/Control) will be analysed using the data collected from the EPR system Through study completion, 1 year. Bivariate and multivariate analysis (Fisher's exact tests, Chi-square tests and/or logistic regressions) will be performed using the screening performance of each individual condition (yes-if the patient has been tested for one of the seven infections and/or screened for female genital mutilation/no-if the patient was not tested) as the outcome variable and the arm of the study as the predictor variable.
Associations between the screening performance of each individual condition (yes/no), including FGM, and the presence of another condition (yes/no) will be analysed using the data collected from the EPR system Through study completion, 1 year. Bivariate and multivariate analysis (Fisher's exact tests, Chi-square tests and/or logistic regressions) will be performed using the screening performance of each individual condition (yes-if the patient has been tested for one of the seven infections and/or screened for female genital mutilation/no-if the patient was not tested) as the outcome variable and the presence of another condition (yes- if the patient has an ICD-code of another condition \[co-morbidity\]/no- if the patient does not present another condition) as the predictor variable.
Associations between the screening performance of HIV (yes/no) and the viral load (copies/mL) will be analysed using the data collected from the EPR system Through study completion, 1 year. Bivariate and multivariate analysis (Fisher's exact tests, Chi-square tests and/or logistic regressions) will be performed using the screening performance of each individual condition (yes-if the patient has been tested for HIV /no-if the patient was not tested) as the outcome variable and the viral load as the predictor variable.
Associations between the screening performance of HBV and/or HCV (yes/no) and the levels of transaminases (U/L) will be analysed using the data collected from the EPR system Through study completion, 1 year. Bivariate and multivariate analysis (Fisher's exact tests, Chi-square tests and/or logistic regressions) will be performed using the screening performance of each individual condition (yes-if the patient has been tested for HBV and/or HCV /no-if the patient was not tested) as the outcome variable and the levels of transaminases as the predictor variable.
Associations between the screening performance of HBV and/or HCV (yes/no) and the levels of clotting parameters [such as prothrombin (seconds), fibrinogen (mg/dL), among others] will be analysed using the data collected from the EPR system Through study completion, 1 year. Bivariate and multivariate analysis (Fisher's exact tests, Chi-square tests and/or logistic regressions) will be performed using the screening performance of each individual condition (yes-if the patient has been tested for HBV and/or HCV /no-if the patient was not tested) as the outcome variable and the levels of clotting parameters as the predictor variable.
Associations between the screening performance of each individual condition (yes/no), including FGM, and patients' immunosuppression status (yes/no) will be analysed using the data collected from the EPR system Through study completion, 1 year. Bivariate and multivariate analysis (Fisher's exact tests, Chi-square tests and/or logistic regressions) will be performed using the screening performance of each individual condition (yes-if the patient has been tested for one of the seven infections and/or screened for female genital mutilation/no-if the patient was not tested) as the outcome variable and patients' immunosuppression status (yes- if the patient has an ICD-code diagnosis of an immunosuppressive disorder and/or have a prescription of an immunosuppressive drug/no- if the patient is not immunosuppressed) as the predictor variable.
Associations between the screening performance of HIV (yes/no) and the CD4 cell count (cell/µL) will be analysed using the data collected from the EPR system Through study completion, 1 year. Bivariate and multivariate analysis (Fisher's exact tests, Chi-square tests and/or logistic regressions) will be performed using the screening performance of each individual condition (yes-if the patient has been tested for HIV during the study period/no-if the patient was not tested) as the outcome variable and the CD4 cell count as the predictor variable.
Trial Locations
- Locations (2)
Servicio Andaluz de Salud (SAS) (administration of the participant PCCs in Andalusia)
🇪🇸Almería, Andalusia, Spain
Institut Català de la Salut (ICS) (administration of the participant PCCs in Catalonia)
🇪🇸Tortosa, Catalonia, Spain